Springer Handbook of Aerogels (Springer Handbooks) [1 ed.] 3030273210, 9783030273217

Citation preview

Springer

Handbook Aerogels

oƒ

Aegerter Leventis Koebel Steiner III Editors

12 3

Springer Handbooks

Springer Handbooks maintain the highest standards of references in key areas of the physical and applied sciences for practitioners in industry and academia, as well as graduate students. Designed to be useful and readable desk reference books, but also prepared in various electronic formats, these titles allow fast yet comprehensive review and easy retrieval of essential reliable key information. Springer Handbooks cover methods, general principles, functional relationships and fundamental data and review established applications. All Springer Handbooks are edited and prepared with great care by editors committed to harmonizing the content. All chapters are written by international experts in their field. Indexed by SCOPUS. The books of the series are submitted for indexing to Web of Science.

Michel A. Aegerter • Nicholas Leventis • Matthias Koebel • Stephen A. Steiner III Editors

Springer Handbook of Aerogels With 1370 Figures and 191 Tables

Editors Michel A. Aegerter Bottens, Switzerland

Matthias Koebel Siloxene AG Dübendorf, Switzerland

Nicholas Leventis Department of Chemistry Missouri University of Science and Technology Rolla, MO, USA Stephen A. Steiner III Aerogel Technologies, LLC Boston, MA, USA

ISSN 2522-8692 ISSN 2522-8706 (electronic) Springer Handbooks ISBN 978-3-030-27321-7 ISBN 978-3-030-27322-4 (eBook) https://doi.org/10.1007/978-3-030-27322-4 © Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Foreword

The opportunity to write a foreword to an essential handbook in any field falls to very few people, and it is my honor to introduce to you the Springer Handbook of Aerogels. In the 30 years that I have first observed, then dabbled in, and then finally made a career in the fascinating world of gels and aerogel materials, I often wished for a central resource like a handbook to be close at hand as a reference, a guide, an inspiration. The first edition of the Aerogels Handbook has been recognized over the past 10 years as that essential resource for materials scientists, engineers, designers, and end users. In this fast-paced world where the limits of knowledge are seemingly expanded day by day, it is the perfect time to welcome the Springer Handbook of Aerogels to update our constantly growing community as to the greatest and most impactful advances in the field of aerogel and aerogel-like materials. The Aerogels Handbook was a detailed and pragmatic compilation of the state of the art written by many luminaries in the field. The Springer Handbook of Aerogels builds on that towering achievement, and frankly surpasses it. The many chapters covering the multidisciplinary advances of nanoporous materials science, chemistry, and applications brought in by the best minds in the field form the heart of a highly utilitarian and pragmatic exploration of aerogel science. For example, this handbook is constructed to first whisk the reader through how aerogels are made, how their unique morphologies are characterized, and then followed up by a tour through an incredible palette of diverse composition classes of aerogel materials (inorganic, organic, hybrids). The handbook finishes off the tour of today’s aerogel universe as we know it with a survey of incredibly diverse applications, industrial manufacturers, and even a section for recipes and helpful equipment design contributions. The handbook has been compiled masterfully by the editorial staff to provide the most up-todate findings, but also to honor the fundamentals of the field, often forgotten in many advanced handbooks, that provide the context and the practical knowledge in a single-source volume. The Springer Handbook of Aerogels is an essential volume for any technical library, the advanced practitioners’ shelf, or as an introduction and reference guide for new initiates into the field. All interested parties can now continue to leverage the outstanding contributions provided in these Handbooks into quests for discovery of knowledge or entrepreneurial value.

George L. Gould Chief Technology Officer Aspen Aerogels, Inc. v

Preface

No single definition encompasses the complete scope of materials referred as aerogels. Typically, they are open, porous, and three-dimensional solid-phase networks permeated by a co-continuous gas-filled or vacuous pore network, most commonly formed by removing all swelling agents from a gel without substantial volume reduction or compaction of its solid network and which most commonly exhibits nanoscale features. The first aerogels have been created by Samuel Stephens Kistler, most probably between 1927 and 1929 at the College of the Pacific in Stockton, California, USA. Till the beginning of this century, thousands of papers appeared in the literature. Therefore, in November 2008, when Dr. David Parker, Executive Editor at Springer, asked Prof. Dr. Michel A. Aegerter, Editor-in-Chief of the Journal of the Sol-Gel Science and Technology (JSST), about his interest to edit an aerogels handbook, his answer was of course positive. The book had to summarize the development and the state-of-the-art of these outstanding materials. However, as the task was impossible to realize alone, Dr. Aegerter required the collaboration of two renowned scientists in the field: Prof. Dr. Nicholas Leventis, presently Director of Research, Aspen Aerogels, formerly Curators’ Distinguished Professor of Chemistry, Missouri University of Science & Technology, and Dr. Matthias M. Koebel, Head of the Building Materials Group at the Swiss Federal Laboratories for Materials Testing and Research (EMPA) in Duebendorf (Switzerland). The editorial structure of the handbook was rapidly built-up to present for the first time in a single book the state of the art in the development, processing, and properties of inorganic, organic, and composite aerogels, the most important techniques used for their characterization as well as a multidisciplinary description of their use, their recent applications, and the main products commercialized by companies at this time. This first handbook presented in 932 pages, with 42 chapters written by 41 leading authors, as well as a Glossary, Acronyms, and Abbreviations section and a large Subject Index partly based on the authors’ suggestions. The success of this handbook which appeared in 2011 was quite impressive with more than 250,000 chapter downloads till date. This is the reason why Springer asked recently the editor-in-chief of this first volume to realize a second up-dated edition. The task was of course readily accepted by the former editors, but as the field had been largely developed during the last 9 years, a fourth editor, Dr. Stephen A. Steiner III, President, CEO, and founder of Aerogel Technologies, a leading company working in the field, was asked to help them in proposing a fully new and updated volume. Practically, all former participants accepted to fully rewrite and update their former chapters, and several new authors were contacted in order to contribute to the new recent developments. This second edition presents therefore a completely new presentation within its 67 chapters. It starts with a large and updated introduction summarizing the impressive developments realized from around 1950 till today, followed by several parts summarizing the unit operations: processing steps used in aerogel science, the characterization of these materials, oxidebased aerogels, synthetic polymer aerogels, biopolymer aerogels, organic-inorganic hybrid aerogels, carbon-based aerogels, and emerging aerogels. Also, the various applications, commercial products, recipes, and designs are presented. The book ends with a part dedicated to conclusions and outlook. vii

viii

Preface

This new handbook also contains a Glossary with important terms collected by the production board as well as a Subject Index generated on markups suggested by all the authors. The editors express their most sincere thanks to all authors who revised or wrote new chapters for this second aerogels handbook. We also extend our gratitude to Dr. Judith Hinterberg, Springer Handbook Coordinator, who organized many video meetings with the editors for the organization and the realization of this new Springer Handbook of Aerogels. Our thanks go also to Sara Kate Heukerott, Springer Executive Editor, Journals Physics & Materials Science, Springer New York, who participated actively in the early works of this new handbook. The editors and all the scientists who have participated in this new edition hope that this new handbook will have the same worldwide interest as the first edition and, most important, will contribute efficiently to the development of new types of such important materials. Michel A. Aegerter Nicholas Leventis Matthias M. Koebel Stephen A. Steiner III

Contents

1

The Story of Aerogel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stephen A. Steiner III and Alain C. Pierre

Part I

1

Unit Operations: Processing Steps used in Aerogel Science . . . . . . . . . .

51

2

Overview of the Sol–Gel Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Plinio Innocenzi

53

3

Gel-Phase Processing and Solvent Exchange . . . . . . . . . . . . . . . . . . . . . . . . . . Justin S. Griffin, Ryan T. Nelson, Pavel Gurikov, Irina Smirnova, and Stephen A. Steiner III

71

4

Supercritical Drying of Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Raman Subrahmanyam, Ilka Selmer, Alberto Bueno, Dirk Weinrich, Wibke Lölsberg, Marc Fricke, Sohajl Movahhed, Pavel Gurikov, and Irina Smirnova

93

5

Freeze Drying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 Justin S. Griffin, Massimo F. Bertino, Tyler M. Selden, Sylwia M. Członka, and Stephen A. Steiner III

6

Postprocessing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 Stephen A. Steiner III, Justin S. Griffin, Ryan T. Nelson, Frances I. Hurwitz, and Marcus A. Worsley

Part II

Characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

7

Structural Characterization of Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 Gudrun Reichenauer

8

Mechanical Characterization of Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 Huiyang Luo, Sadeq Malakooti, Habel Gitogo Churu, Nicholas Leventis, and Hongbing Lu

9

Thermal Properties of Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 Hans-Peter Ebert

10

Permeability of Silica Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 Thierry Woignier, Liz Anez, Sylvie Calas-Etienne, Juan Primera, Pascal Etienne, and Jean Phalippou

11

Simulation and Modeling of Aerogels Using Atomistic and Mesoscale Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273 Lev D. Gelb

12

Modeling the Structural, Fractal and Mechanical Properties of Aerogels . . . . 289 Ameya Rege ix

x

Contents

Part III

Oxide-Based Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

13

Silica Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 Alain C. Pierre and Arnaud Rigacci

14

Hydrophobic Silica Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335 Ann M. Anderson and Mary K. Carroll

15

Superhydrophobic and Flexible Aerogels and Xerogels Derived from Organosilane Precursors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367 Kazuyoshi Kanamori, Ana Stojanovic, Gerard M. Pajonk, Digambar Y. Nadargi, A. Venkateswara Rao, Kazuki Nakanishi, and Matthias M. Koebel

16

Sodium Silicate-Based Aerogels by Ambient Pressure Drying . . . . . . . . . . . . . 393 A. Venkateswara Rao, Shanyu Zhao, Gerard M. Pajonk, Uzma K. H. Bangi, A. Parvathy Rao, and Matthias M. Koebel

17

Synthesis of Metal Oxide Aerogels via Epoxide-Assisted Gelation of Metal Salts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419 Theodore F. Baumann, Alexander E. Gash, Joe H. Satcher Jr, Nicholas Leventis, and Stephen A. Steiner III

18

High Temperature Oxide Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437 Frances I. Hurwitz, Haiquan Guo, Richard B. Rogers, Nathaniel Olson, and Anita Garg

19

Zirconia Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 459 Lassaad Ben Hammouda, Imen Mejri, Mohamed Kadri Younes, and Abdelhamid Ghorbel

Part IV

Synthetic Polymer Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 477

20

Phenolic Aerogels and Their Carbonization . . . . . . . . . . . . . . . . . . . . . . . . . . 479 Chariklia Sotiriou-Leventis, Nicholas Leventis, and Sudhir Mulik

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 507 Nicholas Leventis

22

Aerogels from Engineering Polymers: Polyimide and Polyamide Aerogels . . . 567 Mary Ann B. Meador, Stephanie L. Vivod, Baochau Nguyen, Haiquan Guo, and Rocco P. Viggiano

23

ROMP-Derived Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 595 Nicholas Leventis and George L. Gould

Part V

Biopolymer Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 621

24

Cellulose Aerogels: Monoliths, Beads, and Fibers . . . . . . . . . . . . . . . . . . . . . . 623 Lorenz Ratke, Kathirvel Ganesan, and Maria Schestakow

25

Biopolymer-Silica Aerogel Nanocomposites . . . . . . . . . . . . . . . . . . . . . . . . . . . 653 Shanyu Zhao, Wim J. Malfait, Chunhua Jennifer Yao, Xipeng Liu, Matthias M. Koebel, and William M. Risen

26

Polysaccharide (Non-cellulosic) Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 677 Tatiana Budtova

Contents

xi

27

Nanocellulose Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 707 Nathalie Lavoine

28

Potential of Anisotropic Cellulose Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . 727 Sven Plappert and Falk Liebner

Part VI

Organic-Inorganic Hybrid Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . 747

29

Polymer-Crosslinked Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 749 Nicholas Leventis, Chariklia Sotiriou-Leventis, Chandana Mandal, Suraj Donthula, and Hongbing Lu

30

Improving Elastic Properties of Polymer-Reinforced Aerogels . . . . . . . . . . . . 791 Mary Ann B. Meador and Baochau Nguyen

31

Aerogels Containing Metal, Alloy, and Oxide Nanoparticles Embedded into Dielectric Matrices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 809 Anna Corrias, Danilo Loche, and Maria Francesca Casula

32

Tuning the Physical Properties of Aerogels by Spatially Selective Modification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 835 Massimo F. Bertino and Gudrun Reichenauer

33

Aerogels Through Ultrasonically-Assisted Synthesis . . . . . . . . . . . . . . . . . . . . 857 Luis Esquivias, M. Piñero, V. Morales-Flórez, and Nicolás de la Rosa-Fox

34

Clay-Based Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 883 Mingze Sun and David A. Schiraldi

Part VII

Carbon-Based Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 919

35

Preparation and Application of Carbon Aerogels . . . . . . . . . . . . . . . . . . . . . . 921 Jun Shen, Dayong Guan, Xueling Wu, and Kai Zhao

36

Nanocarbons: Diamond, Fullerene, Nanotube, Graphite, and Graphene Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 941 Swetha Chandrasekaran, Patrick G. Campbell, Theodore F. Baumann, and Marcus A. Worsley

37

Direct CVD Synthesis of Carbon Nanotube Aerogels and Textiles . . . . . . . . . . 971 David S. Lashmore and Stephen A. Steiner III

Part VIII

Emerging Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 987

38

Chalcogenide Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 989 Stephanie L. Brock and Hongtao Yu

39

Metal Fluoride and Fluorinated Metal Oxide Aerogels . . . . . . . . . . . . . . . . . . 1011 Tomaž Skapin

40

Nanoparticle-Based Inorganic Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1041 Markus Niederberger

41

Metal Nanoparticle Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1061 Dennis Müller, Dan Wen, Alexander Eychmüller, and Nadja C. Bigall

42

Noble Metal Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1089 F. John Burpo

xii

Contents

43

Nanostructured Metal Foams via Combustion Synthesis . . . . . . . . . . . . . . . . . 1129 Bryce C. Tappan

44

Interpenetrating Phenolic/Oxide Networks and Carbothermal Synthesis of Metallic and Carbide Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1155 Nicholas Leventis, Chariklia Sotiriou-Leventis, and Suraj Donthula

45

Aerogel-Like Metals Produced Through Physical Vapor Deposition . . . . . . . . 1189 Racheli Ron and Adi Salomon

46

Aerogel-Inspired Materials Derived from Industrial Waste . . . . . . . . . . . . . . . 1211 Hai M. Duong

Part IX

Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1239

47

Aerogels and Sol–Gel Composites as Nanostructured Energetic Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1241 Alexander E. Gash, Randall L. Simpson, Joe H. Satcher Jr, and Nicholas Leventis

48

Aerogels for Superinsulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1263 Jannis Wernery, Arnaud Rigacci, Patrick Achard, and Matthias M. Koebel

49

Aerogels as Platforms for Chemical Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . 1289 Mary K. Carroll and Ann M. Anderson

50

Aerogels for Electrochemical Energy Storage Applications . . . . . . . . . . . . . . . 1305 Debra R. Rolison, Megan B. Sassin, and Jeffrey W. Long

51

Transparent Silica Aerogel Blocks for High-Energy Physics Research . . . . . . 1333 Makoto Tabata

52

Aerogels for High-Energy-Density Physics Targets . . . . . . . . . . . . . . . . . . . . . 1353 Christopher E. Hamilton and Thomas Murphy

53

Porous Glasses, Binary Glasses, and Composite Glasses from Aerogels . . . . . 1369 Thierry Woignier, Jerome Reynes, and Jean Phalippou

54

Environmental Applications for Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1383 Thierry Woignier, Osman Karatum, and Desiree L. Plata

55

Aerogels for Pollution Mitigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1399 Bradford A. Bruno, Ann M. Anderson, and Mary K. Carroll

56

Application of Aerogels in Optical Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . 1431 Yaprak Özbakır, Alexandr Jonáš, Alper Kiraz, and Can Erkey

57

Biomedical Applications of Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1455 Wei Yin and David A. Rubenstein

58

In Vivo Biomedical Applications of Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . 1471 Firouzeh Sabri

59

Pharmaceutical Applications of Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1489 Irina Smirnova, Carlos A. García-González, and Pavel Gurikov

60

Applications of Aerogels in Space Exploration . . . . . . . . . . . . . . . . . . . . . . . . 1505 Steven M. Jones, Jeffrey Sakamoto, and Jong-Ah Paik

61

Airborne Ultrasonic Transducer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1525 Hidetomo Nagahara, Masahiko Hashimoto, and Michel A. Aegerter

Contents

xiii

62

Aerogels for Foundry Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1535 Lorenz Ratke, Barbara Milow, and Eva van Klaveren

63

Aer( )sculpture: A Free-Dimensional Space Art . . . . . . . . . . . . . . . . . . . . . . . 1555 Ioannis Michaloudis

Part X 64

The Aerogel Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1583 Richard A. Collins, Shanyu Zhao, Jiaqing Wang, Justin S. Griffin, and Stephen A. Steiner III

Part XI 65

Conclusions and Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1729

Aerogels in the 2020s and Beyond . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1731 Michel A. Aegerter, Nicholas Leventis, Matthias M. Koebel, and Stephen A. Steiner III

Part XIII 67

Recipes and Designs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1641

Recipes and Designs for Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1643 Stephen A. Steiner III, Ann M. Anderson, Stephanie L. Brock, Moriah C. Buckwalter, Mary K. Carroll, Steve De Pooter, Shannan L. Downey, Alexander Eychmüller, Maximilian Georgi, Justin S. Griffin, Michael D. W. Grogan, Pavel Gurikov, Karl Hiekel, Lawrence W. Hrubesh, Kazuyoshi Kanamori, Barbara Milow, Ryan T. Nelson, A. Venkateswara Rao, Marina Schwan, Karunamuni L. Silva, Marcus A. Worsley, and Shanyu Zhao

Part XII 66

Commercial Products and Industry Overview . . . . . . . . . . . . . . . . . . . . . 1581

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1743

Glossary of Aerogel Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1745 Stephen A. Steiner III, Michel A. Aegerter, Matthias M. Koebel, and Nicholas Leventis

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1763

About the Editors

Michel A. Aegerter was born on January 23, 1939, and lives in Bottens, a small city in the French part of Switzerland. He studied physics engineering (1958–1962) at the Swiss Institute of Technology in Lausanne (EPFL, Switzerland) and then got his PhD degree from the University of Neuchâtel in 1966. After a 3-year stay at the University of Utah (Salt Lake City, USA) he became assistant professor at the University of Neuchâtel (1969–1978) and then moved as ordinary professor to the University of São Paolo, Campus of São Carlos (Brazil), till 1996; he has been dean of the postgraduation in sciences and engineering of materials. He then accepted the position of director of the Department of Coating Technology at the Leibniz-Institut fuer Neue Materialien (INM) in Saarbruecken (Germany, 1995–2006). Since then, he became Honorary Professor at the University of Saarland (Saarbruecken). Michel started getting involved in the field known as “the Sol–Gel Process” during a short visit by Professor J. Zarzycky in 1996, beginning to study the development of silica aerogels! Since then, his R&D interests have been dedicated exclusively to this process, in particular to nano compounds, functional coatings, transparent conducting oxide coatings, photoelectrochemical and electrochromic coatings on glasses and plastics prepared by the sol–gel and nanoparticle approaches, devices such as electrochromic windows and dye-sensitized solar cells, and new sol–gel coating technology (e.g., coatings inside tubes, coatings on very thin glasses) as well as modern printing techniques such as ink-jet, gravure, and flexoprint to print large-scale and patterned functional coatings. He has also developed many industrial applications. Michel has published more than 500 international scientific papers and holds 8 patents in the field of color centers, non-oxide glasses, and sol–gel sciences and technology. He is the author or co-author of 14 books. He has been a referee for many scientific journals and has organized many international meetings. Since 2007, he is the editor-in-chief of the Journal of Sol-Gel Science and Technology (JSST).

xv

xvi

About the Editors

Nicholas Leventis received his BS in chemistry from the University of Athens, Greece (1980), and his PhD from Michigan State University in organic chemistry/photochemistry (1985). In 1992, he completed the 1-year graduate program in administration and management of Harvard University. He was a postdoctoral fellow at MIT (1985–1988) and worked for 6 years in the private sector (1988–1994) before joining the Chemistry Department of the Missouri University of Science and Technology (MS&T) as an assistant professor, advancing to associate professor in 1999 and to professor in 2007. In 2010, he was named Curators’ Distinguished Professor. In 1998, he was a senior faculty fellow at the U.S. Naval Research Laboratory, Washington D.C. During an extended leave from MS&T (2002–2006), Dr. Leventis joined the Polymers Branch, Materials Division of the NASA Glenn Research Center, where he established the aerogels research program. Dr. Leventis retired from MS&T in June 2019 and joined Aspen Aerogels as Director, Research. Dr. Leventis has received the Katie F. Young Award from Harvard University (1992), the Arthur K. Doolittle Award of the American Chemical Society (1992), the NASA Exceptional Scientific Achievement Medal (2005), the Society for American Military Engineers Award (2008), and the Nano50TM Award twice, (2005 and 2007). Dr. Leventis has published over 190 research articles in all major areas of chemistry and has been awarded 30 patents. His h-index is 51. Matthias M. Koebel received his PhD from Brown University in 2004. After a postdoctoral stay at UC Berkeley with G.A. Somorjai, focusing on nanocatalysis, he joined Empa back in his home country – Switzerland – in 2006, where he began building a research group in soft chemistry and aerogels. His core activities are linked to process-scale up and lab-to-market transfer of nanomaterials science.

Dr. Stephen A. Steiner III is the president, CEO, and founder of Aerogel Technologies, LLC, a leading manufacturer of aerogels. Steiner holds a PhD in Materials Chemistry and Engineering from the Department of Aeronautics and Astronautics at MIT, an SM in Materials Science and Engineering from MIT, and a BS in Chemistry Course from the University of Wisconsin–Madison. He is an accomplished researcher in the field of nanomaterials, with interests including aerogels and related nanoporous architectures, synthesis of nanocarbons, microgravity materials processing, and design of advanced materials for aerospace applications. Dr. Steiner began his journey into aerogels in 1999 as a 17-yearold basement tinkerer, conducting research that led to his first of numerous patents. In 2008, Dr. Steiner cofounded the open-source initiative Aerogel.org, which today is one of the world’s leading

About the Editors

xvii

online resources about aerogels. Steiner’s expertise in aerogels includes commercialization, ambient-pressure freeze drying, polymer aerogels, nanocarbon aerogels, and nanoporous metals. He has also been featured in numerous television programs and Internet videos, including productions on the BBC, PBS, Discovery Networks, and YouTube’s Veritasium channel. In his spare time, Steiner moonlights as a flight director for the Zero Gravity Corporation, where he has spent over 10 hours weightless in 20-second bursts.

Contributors

Patrick Achard MINES ParisTech., PSL Research University, PERSEE – Center for Processes, Renewable Energies and Energy Systems, Sophia Antipolis, France Michel A. Aegerter JSST, Bottens, Switzerland Ann M. Anderson Department of Mechanical Engineering, Union College, Schenectady, NY, USA Liz Anez Departamento de física, Universidad del Zulia, Maracaibo, Venezuela Uzma K. H. Bangi Department of Physics, PAH Solapur University, Solapur, India Theodore F. Baumann Lawrence Livermore National Laboratory (LLNL), Livermore, CA, USA Lassaad Ben Hammouda Laboratoire de Chimie des Matériaux et Catalyse (LCMC), Département de Chimie, Faculté des Sciences de Tunis, Université Tunis El Manar, Tunis, Tunisia Massimo F. Bertino Department of Physics, Virginia Commonwealth University, Richmond, VA, USA Nadja C. Bigall Institute of Physical Chemistry and Electrochemistry, Leibniz Universität Hannover, Hannover, Germany Stephanie L. Brock Department of Chemistry, Wayne State University, Detroit, MI, USA Bradford A. Bruno Department of Mechanical Engineering, Union College, Schenectady, NY, USA Moriah C. Buckwalter Aerogel Technologies, LLC, Boston, MA, USA Tatiana Budtova MINES ParisTech, PSL Research University, Center for Materials Forming (CEMEF), UMR CNRS 7635, Sophia Antipolis, France Alberto Bueno Aerogel-it GmbH, Osnabrueck, Germany F. John Burpo Chemistry and Life Science, United States Military Academy, West Point, West Point, NY, USA Sylvie Calas-Etienne Laboratoire Charles Coulomb (L2C), Université Montpellier, Montpellier, France Patrick G. Campbell Lawrence Livermore National Laboratory, Livermore, CA, USA Mary K. Carroll Department of Chemistry, Union College, Schenectady, NY, USA Maria Francesca Casula Department of Mechanical, Chemical and Materials Engineering, University of Cagliari, Cagliari, Italy Swetha Chandrasekaran Lawrence Livermore National Laboratory, Livermore, CA, USA xix

xx

Habel Gitogo Churu Department of Mechanical Engineering, LeTourneau University, Longview, TX, USA Richard A. Collins IDTechEx, Cambridge, UK Anna Corrias School of Physical Sciences, Ingram Building, University of Kent, Canterbury, UK Sylwia M. Członka Institute of Polymer and Dye Technology, Faculty of Chemistry, Lodz University of Technology, Lodz, Poland Nicolás de la Rosa-Fox Departamento de Física de la Materia Condensada, Universidad de Cádiz, Cádiz, Spain Steve De Pooter KU Leuven, Leuven, Belgium Suraj Donthula Department of Chemistry, Missouri University of Science and Technology, Rolla, MO, USA Department of Lithography, Intel Corporation, Hillsboro, OR, USA Shannan L. Downey Ocellus, Inc., Livermore, CA, USA Hai M. Duong Department of Mechanical Engineering, National University of Singapore, Singapore, Singapore Hans-Peter Ebert Center for Applied Energy Research e.V., Würzburg, Germany Can Erkey Department of Chemical and Biological Engineering, Koç University, Istanbul, Turkey Luis Esquivias Departamento de Física de la Materia Condensada, Facultad de Física, Instituto de Ciencias de Materiales de Sevilla (CSIC), Universidad de Sevilla, Seville, Spain Pascal Etienne Laboratoire Charles Coulomb (L2C), Université Montpellier, Montpellier, France Alexander Eychmüller Physical Chemistry, Technical University of Dresden, Dresden, Germany Marc Fricke Aerogel-it GmbH, Osnabrueck, Germany Kathirvel Ganesan German Aerospace Center, DLR, Cologne, Germany Carlos A. García-González Departamento de Farmacología, Farmacia y Tecnología Farmaceutica, R+D Pharma Group (GI-1645), Facultad de Farmacia and Health Research Institute of Santiago de Compostela (IDIS), Universidade de Santiago de Compostela, Santiago de Compostela, Spain Anita Garg University of Toledo, Toledo, OH, USA Alexander E. Gash Lawrence Livermore National Laboratory (LLNL), Livermore, CA, USA Lev D. Gelb Department of Materials Science and Engineering, University of Texas at Dallas, Richardson, TX, USA Maximilian Georgi Technical University of Dresden, Dresden, Germany Abdelhamid Ghorbel Laboratoire de Chimie des Matériaux et Catalyse (LCMC), Département de Chimie, Faculté des Sciences de Tunis, Université Tunis El Manar, Tunis, Tunisia George L. Gould Research and Development, Aspen Aerogels, Inc., Northborough, MA, USA

Contributors

Contributors

xxi

Justin S. Griffin Aerogel Technologies, LLC, Boston, MA, USA Michael D. W. Grogan Department of Physics, University of Bath, Bath, UK Dayong Guan Shanghai Key Laboratory of Special Artificial Microstructure Materials and Technology, Pohl Institute of Solid Physics, Tongji University, Shanghai, P.R. China Tianjin NPCA New Materials Technology Co., Ltd., Tianjin, P.R. China Haiquan Guo Ohio Aerospace Institute, NASA Glenn Research Center, Cleveland, OH, USA Pavel Gurikov Laboratory for Development and Modelling of Novel Nanoporous Materials, Hamburg University of Technology, Hamburg, Germany Institute of Thermal Separation Processes, Hamburg University of Technology, Osnabrueck, Germany Christopher E. Hamilton Materials Science & Technology Division, Los Alamos National Laboratory, Los Alamos, NM, USA Masahiko Hashimoto Advanced Technology Research Laboratory, Panasonic Corporation, Kyoto, Japan Karl Hiekel Technical University of Dresden, Dresden, Germany Lawrence W. Hrubesh Ocellus, Inc., Livermore, CA, USA Frances I. Hurwitz NASA Glenn Research Center, Cleveland, OH, USA Plinio Innocenzi Laboratory of Materials Science and Nanotechnology, Department of Biomedical Sciences, University of Sassari, Sassari, Italy Alexandr Jonáš Department of Microphotonics, Institute of Scientific Instruments, The Czech Academy of Sciences, Brno, Czech Republic Steven M. Jones Department of Analytical Chemistry and Material Development, California Institute of Technology, Jet Propulsion Laboratory, Pasadena, CA, USA Kazuyoshi Kanamori Department of Chemistry, Graduate School of Science, Kyoto University, Kyoto, Japan Osman Karatum Florida Gulf Coast University, Fort Myers, FL, USA Alper Kiraz Department of Physics, Koç University, Istanbul, Turkey Department of Electrical and Electronics Engineering, Koç University, Istanbul, Turkey Matthias M. Koebel Building Energy Materials and Components, Laboratory for Building Technologies, Empa, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland David S. Lashmore American Boronite Corporation, Burlington, MA, USA University of New Hampshire at Durham, Durham, NH, USA Nathalie Lavoine Department of Forest Biomaterials, North Carolina State University, Raleigh, NC, USA Nicholas Leventis Department of Chemistry, Missouri University of Science and Technology, Rolla, MO, USA Department of Research and Development, Aspen Aerogels, Inc., Northborough, MA, USA Falk Liebner Institute for Chemistry of Renewable Resources, University of Natural Resources and Life Sciences Vienna, Tulln, Austria Department of Chemistry, University of Aveiro, Campus Universitário de Santiago, Aveiro, Portugal

xxii

Xipeng Liu Northboro Research and Development Center, Northboro, MA, USA Danilo Loche School of Physical Sciences, Ingram Building, University of Kent, Canterbury, UK Wibke Lölsberg BASF SE, Ludwigshafen am Rhein, Germany Jeffrey W. Long U.S. Naval Research Laboratory, Washington, DC, USA Hongbing Lu Department of Mechanical and Aerospace Engineering, The University of Texas at Dallas, Richardson, TX, USA Huiyang Luo Department of Mechanical Engineering, The University of Texas at Dallas, Richardson, TX, USA Karagozian & Case, Glendale, CA, USA Sadeq Malakooti Department of Mechanical Engineering, The University of Texas at Dallas, Richardson, TX, USA Wim J. Malfait Building Energy Materials and Components Laboratory, EMPA, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland Chandana Mandal Department of Chemistry, Missouri University of Science and Technology, Rolla, MO, USA Lithography, Intel Corporation, Hillsboro, OR, USA Mary Ann B. Meador Materials and Structures, NASA Glenn Research Center, Cleveland, OH, USA Imen Mejri Laboratoire de Chimie des Matériaux et Catalyse (LCMC), Département de Chimie, Faculté des Sciences de Tunis, Université Tunis El Manar, Tunis, Tunisia Ioannis Michaloudis Visual Artist and Research Affiliate, Institute of Nanoscience and Nanotechnology, NCSR Demokritos, Acharnes, Greece Barbara Milow Institute of Materials Research, DLR, German Aerospace Center, Cologne, Germany V. Morales-Flórez Departamento de Física de la Materia Condensada, Facultad de Física, Instituto de Ciencias de Materiales de Sevilla (CSIC), University of Seville, Seville, Spain Sohajl Movahhed Covestro Deutschland AG, Leverkusen, Germany Sudhir Mulik The Dow Chemical Company, Collegeville, PA, USA Dennis Müller Institute of Physical Chemistry and Electrochemistry, Leibniz Universität Hannover, Hannover, Germany Thomas Murphy Physics Division, Los Alamos National Laboratory, Los Alamos, NM, USA Digambar Y. Nadargi Building Energy Materials and Components, Laboratory for Building Technologies, Empa, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland School of Physical Sciences, Solapur University, Solapur, Maharashtra, India Hidetomo Nagahara Advanced Technology Research Laboratory, Panasonic Corporation, Kyoto, Japan Kazuki Nakanishi Institute of Materials and Systems for Sustainability, Nagoya University, Nagoya, Japan Ryan T. Nelson Aerogel Technologies, LLC, Boston, MA, USA

Contributors

Contributors

xxiii

Baochau Nguyen Ohio Aerospace Institute, NASA Glenn Research Center, Brookpark, OH, USA Markus Niederberger Laboratory for Multifunctional Materials, Department of Materials, ETH Zurich, Zurich, Switzerland Nathaniel Olson NASA Glenn Research Center, Cleveland, OH, USA Yaprak Özbakır Department of Chemical Engineering, Üsküdar University, Istanbul, Turkey Jong-Ah Paik Department of Thermal Energy Conversion Applications and Systems, California Institute of Technology, Jet Propulsion Laboratory, Pasadena, CA, USA Gerard M. Pajonk Laboratoire des Matériaux et Procédés Catalytiques, Université Claude Bernard Lyon 1, Villeurbanne, France Jean Phalippou Laboratoire Charles Coulomb (L2C), CNRS-Université Montpellier II, Montpellier Cedex 5, France M. Piñero Departamento de Física Aplicada, CASEM, Universidad de Cádiz, Cádiz, Spain Alain C. Pierre Chemistry Department, CNRS, UMR 5256, IRCELYON, Institut de recherches sur la catalyse et l’environnement de Lyon, Université Claude Bernard-Lyon 1, Villeurbanne, France Sven Plappert Institute for Chemistry of Renewable Resources, University of Natural Resources and Life Sciences Vienna, Tulln, Austria Desiree L. Plata Massachusetts Institute of Technology, Cambridge, MA, USA Juan Primera Instituto de Ciencias Básicas, Departamento de Física, Avenida Urbina y Che Guevara, Universidad Técnica de Manabí, Portoviejo Provincia de Manabí, Ecuador A. Parvathy Rao Air Glass Laboratory, Department of Physics, Shivaji University, Kolhapur, Maharashtra, India A. Venkateswara Rao Air Glass Laboratory, Department of Physics, Shivaji University, Kolhapur, Maharashtra, India Lorenz Ratke Institute of Materials Research, DLR, German Aerospace Center, Cologne, Germany Ameya Rege Department of Aerogels and Aerogel Composites, Institute of Materials Research, German Aerospace Center, Cologne, Germany Gudrun Reichenauer Bavarian Center for Applied Energy Research, Würzburg, Germany Jerome Reynes CNRS-Université Montpellier II, Montpellier Cedex 5, France Arnaud Rigacci PERSEE – Center for processes, renewable energies and energy systems, MINES ParisTech., PSL Research University, Sophia Antipolis, France William M. Risen Department of Chemistry, Brown University, Providence, RI, USA Richard B. Rogers NASA Glenn Research Center, Cleveland, OH, USA Debra R. Rolison U.S. Naval Research Laboratory, Washington, DC, USA Racheli Ron Department of Chemistry, Institute of Nanotechnology and Advanced Materials (BINA), Bar-Ilan University, Ramat-Gan, Israel David A. Rubenstein Department of Biomedical Engineering, Stony Brook University, Stony Brook, NY, USA

xxiv

Firouzeh Sabri Department of Physics and Materials Science, University of Memphis, Memphis, TN, USA Jeffrey Sakamoto Mechanical Engineering, Materials Science and Engineering, and Macromolecular Science and Engineering, University of Michigan, Ann Arbor, MI, USA Adi Salomon Department of Chemistry, Institute of Nanotechnology and Advanced Materials (BINA), Bar-Ilan University, Ramat-Gan, Israel Megan B. Sassin U.S. Naval Research Laboratory, Washington, DC, USA Joe H. Satcher Jr Lawrence Livermore National Laboratory (LLNL), Livermore, CA, USA Maria Schestakow German Aerospace Center, DLR, Cologne, Germany David A. Schiraldi Case Western Reserve University, Cleveland, OH, USA Marina Schwan DLR, Cologne, Germany Tyler M. Selden Department of Physics, Virginia Commonwealth University, Richmond, VA, USA Ilka Selmer Institute of Thermal Separation Processes, Hamburg University of Technology, Osnabrueck, Germany Jun Shen Shanghai Key Laboratory of Special Artificial Microstructure Materials and Technology, Pohl Institute of Solid Physics, Tongji University, Shanghai, P.R. China Tianjin NPCA New Materials Technology Co., Ltd., Tianjin, P.R. China Karunamuni L. Silva Wayne State University, Detroit, MI, USA Randall L. Simpson Lawrence Livermore National Laboratory, Livermore, CA, USA Tomaž Skapin Department of Inorganic Chemistry and Technology, Jožef Stefan Institute, Ljubljana, Slovenia Irina Smirnova Institute of Thermal Separation Processes, Hamburg University of Technology, Hamburg, Germany Chariklia Sotiriou-Leventis Department of Chemistry, Missouri University of Science and Technology, Rolla, MO, USA Stephen A. Steiner III Aerogel Technologies, LLC, Boston, MA, USA Ana Stojanovic Building Energy Materials and Components, Laboratory for Building Technologies, Empa, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland Raman Subrahmanyam Aerogel-it GmbH, Osnabrueck, Germany Mingze Sun Columbia University, New York, NY, USA Case Western Reserve University, Cleveland, OH, USA Makoto Tabata Department of Physics, Graduate School of Science, Chiba University, Chiba, Japan Bryce C. Tappan High Explosives Science & Technology Group, Los Alamos National Laboratory, Los Alamos, NM, USA Eva van Klaveren Institute of Materials Research, DLR, German Aerospace Center, Cologne, Germany Rocco P. Viggiano NASA Glenn Research Center, Cleveland, OH, USA

Contributors

Contributors

xxv

Stephanie L. Vivod NASA Glenn Research Center, Cleveland, OH, USA Jiaqing Wang Nanjing Fiberglass Research & Design Institute, Nanjiang, People’s Republic of China Dirk Weinrich Aerogel-it GmbH, Osnabrueck, Germany Dan Wen Center for Nano Energy Materials, School of Materials Science and Engineering, Northwestern Polytechnical University, Xi’an, P. R. China Jannis Wernery Empa – Swiss Federal Laboratories for Materials Science and Technology, Siloxene, Dübendorf, Switzerland Thierry Woignier CNRS, IRD, IMBE, Aix Marseille University, University Avignon, Marseille, France Institut Méditerranéen de Biodiversité et d’Ecologie, IRD UMR 237-Campus Agro Environnemental Caraïbes, Le Lamentin, France Marcus A. Worsley Lawrence Livermore National Laboratory, Livermore, CA, USA Xueling Wu Shanghai Key Laboratory of Special Artificial Microstructure Materials and Technology, Pohl Institute of Solid Physics, Tongji University, Shanghai, P.R. China Tianjin NPCA New Materials Technology Co., Ltd., Tianjin, P.R. China Chunhua Jennifer Yao Firestone Building Products Co. LLC, Indianapolis, USA Wei Yin Department of Biomedical Engineering, Stony Brook University, Stony Brook, NY, USA Mohamed Kadri Younes Laboratoire de Chimie des Matériaux et Catalyse (LCMC), Département de Chimie, Faculté des Sciences de Tunis, Université Tunis El Manar, Tunis, Tunisia Hongtao Yu Department of Chemistry, Wayne State University, Detroit, MI, USA Shanyu Zhao Building Energy Materials and Components Laboratory, EMPA, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland Department of Chemistry, Brown University, Providence, RI, USA Kai Zhao Shanghai Key Laboratory of Special Artificial Microstructure Materials and Technology, Pohl Institute of Solid Physics, Tongji University, Shanghai, P.R. China Tianjin NPCA New Materials Technology Co., Ltd., Tianjin, P.R. China

1

The Story of Aerogel Stephen A. Steiner III and Alain C. Pierre

Contents 1.1 1.1.1 1.1.2 1.1.3 1.1.4

What Are Aerogels? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Porosity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Solid Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 2 4 5 9

1.2 1.2.1 1.2.2 1.2.3 1.2.4 1.2.5 1.2.6

The History of Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Founding Studies by Kistler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Early Commercialization of Aerogels . . . . . . . . . . . . . . . . . . . . . . The Aerogel Renaissance: 1960–1990 . . . . . . . . . . . . . . . . . . . . . . Advances in Processing and Understanding: The 1990s . . . New Possibilities and Commercialization: 2000–2020 . . . . . Coming Together and Pushing Forth . . . . . . . . . . . . . . . . . . . . . . .

10 10 13 14 18 22 28

1.3 Applications of Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 1.3.1 Commercial Applications of Aerogels . . . . . . . . . . . . . . . . . . . . . . 29 1.3.2 Emerging and Specialty Applications of Aerogels . . . . . . . . . 32 1.4

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

Abstract

In this handbook, we explore the diverse class of porous materials called aerogels – what they are, how they are made, how they are characterized, materials properties they can exhibit, their applications, and their increasing role in commerce. In this chapter, we provide the reader with a general introduction to the topic of aerogels and an overview of their history.

S. A. Steiner III (*) Aerogel Technologies, LLC, Boston, MA, USA e-mail: [email protected] A. C. Pierre (*) Chemistry Department, CNRS, UMR 5256, IRCELYON, Institut de recherches sur la catalyse et l’environnement de Lyon, Université Claude Bernard-Lyon 1, Villeurbanne, France e-mail: [email protected]

Keywords

Aerogel · Aerogels · Aerogel-like · Aerogel-inspired · Introduction · Overview · History · Synthesis · Properties · Characterization · Applications · Review

1.1

What Are Aerogels?

No single definition fully encompasses the complete scope of materials that are referred to as aerogels in the scientific literature and commerce. Many researchers in fact disagree on how the term should be defined, or if a definition is needed at all. This said, at a minimum, aerogels are solid-phase, porous materials that possess a high degree of porosity arising from a finely divided pore structure. The exact meanings of “high degree of porosity” and “finely divided pore structure” are areas where researchers may offer alternate viewpoints. Some feel that specific thresholds for pore size and percent porosity should be what define an aerogel [1–3], while others feel such feature thresholds arbitrarily exclude materials that should be considered aerogels. From which precursors and by what process an aerogel may be made are additional areas where researchers differ. Some take the approach that an aerogel is what results when the liquid component of a gel is replaced by air in a way that results in only very moderate shrinkage of the gel’s solid network [4–7], while others feel the term describes a particular type of material architecture irrespective of how it is made or from what it is derived. Some researchers use the term aerogel to specifically refer to the material product that results from supercritically drying a gel [8](a), while others feel aerogels can be made through many different processes [9]. To many, a quintessential aspect of aerogels is nanostructure – that aerogels comprise a nanostructured solid phase and a co-continuous nanoscale porous network. This aspect of aerogels is all the more fascinating when one considers that aerogels can be easily produced as macroscopic forms –

© Springer Nature Switzerland AG 2023 M. A. Aegerter et al. (eds.), Springer Handbook of Aerogels, Springer Handbooks, https://doi.org/10.1007/978-3-030-27322-4_1

1

2

S. A. Steiner III and A. C. Pierre

“nanotechnology you can hold in your hand” as Rolison has frequently remarked [10]. The lack of consensus regarding a definition arises in part from ongoing technical developments that have continually pushed the frontiers of synthesis, characterization, and applications of porous materials and, in the process, challenged preconceived notions about what makes something an aerogel. This said, most people familiar with the field typically associate the term aerogel with materials that exhibit certain structure–property relationships that give rise to characteristic extreme and useful materials properties, such as low bulk density, high specific surface area, and low thermal conductivity (Fig. 1.1a), to name just a few. Although a number of definitions have been proposed in the academic literature [1, 2, 4–7, 9] and on the Internet [3], the complete set of materials presently referred to as aerogels includes materials encompassing an ever-increasing range of substances [11–22], morphologies [23–30], and functionalities [31–46], produced through an increasingly diverse set of synthetic pathways (see ▶ Chaps. 4, ▶ 5, ▶ 6, ▶ 16, ▶ 29, ▶ 37, ▶ 42, ▶ 43 and ▶ 45 for an array of various approaches). Rather than attempt to commit to a single technical definition and risk alienating important contributions and research vectors from the field, herein we shall consider the full breadth of emergent usages of the term aerogel in the literature and in commerce in order to maximally represent the ethos of the field. After all, given the degree to which aerogels have pushed and continue to push the frontiers of materials science, it would seem that the full extent of what aerogels can be and do has not yet been fully revealed.

1.1.1

Porosity

As mentioned above, aerogels are highly porous materials. For most materials historically referred to as aerogels, the material’s porosity constitutes the majority of its volume and primarily consists of void spaces with characteristic dimensions extending from nanometers to microns [47, 48], noting that materials exhibiting less than 50% porosity and/or containing pore sizes outside this range are also sometimes referred to as aerogels [49–51]. This said, most materials called aerogels exhibit porosities greater than 50% and frequently greater than 90% [8]. In fact, for some ultralowdensity aerogels, this value can be upwards of 99.9% or higher [52–54]. The porosity of an aerogel is open-celled in nature [23] and is usually permeated by a gas, for example air, but can also be evacuated. In this regard, aerogels are dry materials, that is, they do not contain substantial amounts of liquid in their pore network, and thus are differentiated from gels, such as hydrogels or alcogels, in which the pore network of the material is fully permeated by a liquid.

It is worth discussing a few points about terminology before proceeding. The field frequently uses terms derived from the word “gel” to mean specific things, for example, hydrogel (or occasionally aquogel or aquagel), referring to gels whose pore fluid primarily comprises water; alcogel, referring to gels whose pore fluid primarily comprises an alcohol; organogel, referring to gels whose pore fluid primarily comprises an organic solvent; solvogel, referring to a gel comprising a pore fluid of an arbitrary solvent (noting that all gels comprise a solvent); cryogel, referring to an aerogel prepared by freeze drying of a gel; lyogel, which can refer to either a gel comprising a pore fluid of an arbitrary solvent (wherein lyo- means solvent) or an aerogel prepared by freeze drying of a gel (because some researchers refer to freeze drying as lyophilization); and xerogel or occasionally ambigel, referring to an aerogel prepared by ambientpressure evaporative drying of a gel performed in a way that does not result in significant shrinkage of the gel’s solid skeleton. The term xerogel in this context is particularly problematic, since its original intended meaning as first introduced by Freundlich was to describe the low-to-noporosity material that results when a gel is evaporatively dried in a way that results in unhindered shrinkage [55]. While IUPAC defines a xerogel as an “open network formed by the removal of all swelling agents from a gel” [56], capillary stresses that arise during drying of a gel can easily cause it to contract down to 25% or less of its initial volume [8](b), a much lower value than typically exhibited by most aerogels [8](c). Accordingly, many papers and patents have used the term xerogel to refer to aerogel materials that have been dried by means other than supercritical drying, despite possessing high porosity and other structure–property relationships typical of aerogels. Because of this, a number of researchers have pushed to eliminate this usage of the term xerogel as it is, at a minimum, confusing and inconsistent, and confounds search queries. The various prefixed terms used to describe gels based on the composition of their pore fluid listed above can also be confusing and imprecise: for example, the term alcogel, which appears frequently in the silica aerogel literature, is often misappropriated to describe gels that do not contain alcohol. Furthermore, the composition of a gel’s pore fluid can change several times as the gel is processed into an aerogel. As such, while the various terms described above may appear in the literature from time to time and it is important to understand their meanings, the reader is encouraged to use them sparingly in future work: instead, it is suggested that liquid-containing gels be simply referred to as gels without an additional prefix, and that the term xerogel only be used to describe the low-to-no porosity dried product of a gel. The field has also largely adopted IUPAC terminology to describe pore sizes, specifically, the terms mesoporous,

1

The Story of Aerogel

3

a

b

Mesopores

c

Micropores

10 nm d

h

e

g

f

i

j

34 kg

k

100 nm l

m

Fig. 1.1 Some representative examples of the diverse class of materials known as aerogels (image credits in parentheses, used with permission). (a) “The Flower,” in which a silica aerogel monolith protects a flower from the heat of a flame (Lawrence Berkeley National Laboratory). (b) Silica aerogel monolith composed of >98% air by volume holding a brick 1000 times its weight (NASA Jet Propulsion Laboratory). (c) Schematic illustration of the porous string-of-pearls microstructure

n

exhibited by many types of aerogels, highlighting constituent mesopores and micropores. (d) Trimethylsiloxy-functionalized hydrophobic silica aerogel monolith (BuyAerogel.com). (e) Flexible organically modified silica prepared from methyltrimethoxysilane (Shivaji University). (f) A variety of colorful lanthanide oxide aerogels prepared through epoxideassisted gelation of metal salts (Lawrence Livermore National Laboratory). (g) Mechanically robust polymer-crosslinked silica aerogel

1

4

meaning contains mean pore widths ranging from 2 to 50 nm; macroporous, meaning contains mean pore widths greater than 50 nm; and the unfortunate and confusing term microporous, meaning contains mean pore widths less than 2 nm. In recent years, the blanket term nanoporous has also arisen, having no formal definition but generally referring to materials that contain pores with feature sizes ranging from ones to hundreds of nanometers, noting that many people generally consider technical terms prefixed with nano (such as nanotechnology) to refer to things having feature sizes of less than 100 nm. In this regard, many researchers refer to aerogels as being nanoporous. The drawback of such a broad term is its lack of clarity regarding characteristic length scales, which, for example, makes it difficult to gauge what type of nanoparticles, macromolecules, and biomolecules can diffuse into the material, whereas the yardstick is obvious for terms like mesoporous and macroporous. Historically, most materials considered aerogels have been describable as mesoporous solids, that is, porous solid-phase materials whose pores primarily range from 2 to 50 nm in diameter [23], noting such mesoporous aerogels frequently contain pores with diameters outside this range as well [48]. Definitions based on pore size alone are problematic, since even classic mesoporous aerogel formulations almost always contain a significant population of macropores. Furthermore, even if the number of mesopores in a material exceeds the number of macropores, a majority of the material’s pore volume may in fact arise from its macroporosity, since pore volume scales cubically with pore diameter. Regardless, aerogels are typically differentiated from other highly porous materials on the basis of pore size, including microporous materials such as zeolites [57] and metalorganic frameworks [58], which primarily contain pores less than 2 nm in diameter, and macroporous engineering materials such as expanded polystyrene and polyurethane foams, which primarily contain pores tens to thousands of microns in diameter [59]. The nanometric structure inherent to many aerogels gives rise to a fascinating phenomenon called the Knudsen effect, wherein the pores of the material are smaller than the mean free path of molecules in air. This results in a dramatic stifling of convective flow through the material, thereby substantially reducing a major component of the thermal conductivity exhibited by traditional porous insulation materials [60, 61] and making many aerogels thermal superinsulators [62, 63] ä Fig. 1.1 (continued) (Missouri University of Science and Technology). (h) Cadmium sulfide quantum dot aerogel with photoluminescent properties (Wayne State University). (i) Ultralightweight graphene aerogel balancing on a dandelion (Zhejiang University). (j) Resorcinol-formaldehyde polymer aerogel and its carbonized derivative (BuyAerogel.com). (k) Metallic silver aerogel-like network made

S. A. Steiner III and A. C. Pierre

(see ▶ Chaps. 10 and ▶ 48). In recent years, however, new material architectures that exhibit average pore diameters on the order of hundreds of nanometers and, in some cases, even microns in diameter, have also emerged, borne from chemistries and synthetic approaches first used to make traditional mesoporous aerogels [45, 49, 50, 64]. Such materials, sometimes referred to as aerogel-like materials, aerogel-inspired materials, or just aerogels, albeit possessing larger pores than what might have traditionally been considered aerogel in the past, still fulfill many of the structure–property relationships typified by traditional aerogels and/or exhibit pore structures orders of magnitude finer than achievable with, for example, foaming techniques used to make traditional macroporous foams [59, 61]. These pore–solid architected cousins to traditional aerogels have accordingly proven to be valuable additions to the porous materials ecosystem (see, for example, ▶ Chaps. 20, ▶ 21, ▶ 37, ▶ 43, ▶ 45 and ▶ 46).

1.1.2

Solid Phase

The porosity of an aerogel is only half of the story, however. What carves out this porosity is the solid-phase component of the aerogel, a sponge-like solid network (also referred to as a skeleton, backbone, scaffold, or framework) that can comprise any of a number of different substances. The specific substance(s) that make up this network accordingly define many of the materials properties of the aerogel, in that an aerogel will take on many of the physical and chemical properties inherent to its constituent solid-phase component (s), including chemical behavior [34], electrical conductivity [32, 65], magnetism [38, 39, 66–68], and spectral properties [35, 36, 69], to name a few – sometimes with property enhancements arising from nanostructuring effects. The solid-phase network of an already existing aerogel may be amorphous [52] or crystalline [24, 70], or somewhere in between, and may be made up of interconnected particles [12, 19], agglomerated grains [71], physisorbed 1D and/or 2D structures [22, 72, 73], or contiguous streams of intertwined polymer molecules [25, 26]. Its constituent particles may be polydisperse [74] or monodisperse [75] in size and/or molecular weight [26], and may comprise one or many substances. Such networks may be initially formed by polymerizing monomers in a liquid via wet-phase colloidal chemistry and allowing the resulting particles to self-assemble into

through physical vapor deposition (Bar-Ilan University). (l) Cryogel ® Z flexible fiber-reinforced silica aerogel composite blanket made by Aspen Aerogels (BuyAerogel.com). (m) Lumira ® LA1000 aerogel particles used in daylighting applications made by Cabot Aerogel (BuyAerogel.com). (n) Assortment of Airloy ® strong polymer aerogel panels made by Aerogel Technologies (Aerogel Technologies)

1

The Story of Aerogel

a three-dimensional network via sol–gel chemistry [8], or, alternatively, by synthesizing discrete particles (e.g., via chemical vapor deposition, plasma synthesis, or solvothermal synthesis) and subsequently assembling the resulting prefabricated particles in a second step [19, 72, 73, 75]. In some cases, the solid-phase network of an already existing aerogel may be chemically converted into an isomorphic aerogel of a different composition, for example, by pyrolyzing an aromatic organic polymer aerogel to produce an amorphous carbon aerogel [65] (Fig. 1.1j), or by carbothermally or magnesiothermally reducing an inorganic oxide aerogel to produce an isomorphic aerogel of its corresponding zerovalent metal/metalloid [69, 71, 76]. In all cases, the solidphase network of an aerogel is what defines the solid-state behavior of the aerogel and is what sculpts and therefore defines the pore structure of the aerogel. Typically, the solid-phase network of an aerogel is highly disordered, that is, lacks long-range periodicity in structure and pore placement, and exhibits fractal-like self-similarity qualities [77]. It is this combination of disorder and selfsimilarity that gives rise to many of the unique and extreme materials properties that can be achieved in an aerogel architecture (Fig. 1.1a and b). Legacy aerogel materials, for example, silica aerogels, typically exhibit morphologies consisting of polydisperse, spheroidal particles interconnected in a three-dimensional mesoporous network described as a string-of-pearls morphology [8](d), [78] (Fig. 1.1c). These string-of-pearls-like strands of spheroidal so-called secondary particles define a large portion of the material’s mesoporosity and in turn are microporous (and potentially mesoporous) in and of themselves, that is, contain internal pores (usually less than 2 nm in diameter) defined by a subnetwork of nanometer-sized constituent particles called primary particles [77] (Fig. 1.1c). This kind of hierarchically porous string-of-pearls morphology is embodied by many colloidally synthesized aerogel materials, including silica aerogels and other metal oxide and metalloid oxide aerogels [12] as well as many synthetic polymer [78] and biopolymer aerogels [79]. In contrast, some metal and metalloid oxides give rise to other lesscommon morphologies, such as networks comprising nanoscale constituent particles that exhibit a leaf-like [24] or a worm-like [29] structure. Still, other types of aerogels, for example, aerogels based on polyureas and polyimides, exhibit fiber-like morphologies comprising branches of entangled polymer molecules free of any visibly appreciable discrete particle subcomponents [25, 80–82], while aerogels based on two-dimensional building blocks such as graphene exhibit sheet-like morphologies [30]. In recent years, aerogels comprising disordered three-dimensional mesoporous networks of crystalline nanoparticles such as quantum dots have enabled production of macroscopic objects that simultaneously exhibit special materials properties only

5

achievable through nanoengineering, for example, quantumconfinement bandgap tuning of semiconductors through tailoring of particle diameter [75]. Accordingly, aerogels of a wide variety of different substances have been prepared, and undoubtedly aerogels of even more compositions will be developed in the future. Table 1.1 lists a non-limiting set of substances from which aerogels have been previously produced.

1.1.3

Synthesis

Most aerogel materials start their lives out as liquid-permeated gels synthesized through sol–gel chemistry; however, aerogels can now be prepared by other ways, for example, by direct assembly of prefabricated nanoscale objects through freeze casting [22, 54], chemical vapor deposition [270] (see also ▶ Chap. 37) and physical vapor deposition techniques [261, 266] (▶ Chap. 45), expansion of polymer beads in supercritical or near-critical fluids [169], and combustion synthesis [262] (▶ Chap. 43). Perhaps in the not-too-distant future, aerogels will be made deterministically at the molecular and/or nano level through additive manufacturing techniques that provide nanometer-scale resolution without the need for liquid-phase colloidal templating. This said, in this section we will focus on sol–gel-derived aerogel materials in order to provide a general overview of the traditional and still most representative process used to make most aerogels. Other synthetic pathways for producing aerogel and aerogel-like materials are described in more depth elsewhere in this handbook (see, for example, ▶ Chaps. 37, ▶ 42, ▶ 43 and ▶ 45). Preparation of aerogels via sol–gel chemistry follows a general synthetic workflow that applies to a wide variety of both inorganic and organic compositions. It is instructive to break the synthesis process down into its typical set of constituent unit operations, which are didactically described in more detail in Part I of this handbook. In the synthesis of a sol–gel-derived aerogel material, these unit operations typically include: (1) sol–gel synthesis of a precursor gel (▶ Chap. 2), wherein a gel of a desired composition is synthesized through wet-phase chemistry; (2) solvent exchange, wherein the pore fluid of the gel precursor is replaced with a pure target solvent through successive diffusion-mediated soakings in baths of the target solvent (▶ Chap. 3); (3) drying, wherein the pore fluid is removed from the gel in a way that prevents the solid-phase structure of the gel from undergoing substantial collapse/densification, resulting in an aerogel (▶ Chaps. 4, ▶ 5, ▶ 14, ▶ 16, ▶ 21 and ▶ 29); and optionally (4) postprocessing, wherein the resulting aerogel is further processed to change its composition, crystallinity, phase, functionality, shape, or other properties (▶ Chap. 6). In the case of sol–gel-derived materials, the material that will eventually become the aerogel is contained within the gel

1

6

S. A. Steiner III and A. C. Pierre

Table 1.1 A selection of different aerogel compositions that have been prepared Metal and Metalloid Oxides Main group metal and metalloid oxides

Transition metal oxides

Lanthanide and actinide metal oxides

Alkaline earth metal oxides Organic Polymers Phenolics Isocyanate-derived polymers Step-growth polymers Radical-mediated and ROMP-derived polymers

Biopolymers Polysaccharides

Proteins, proteoglycans/glycoproteins, and polyglucosamides (see also 1D nanoscale objects) Aromatic matrices Minerals Clays and soils Fulgerites Inorganic Non-oxides Carbon allotropes (see also 1D nanoscale objects and 2D nanoscale objects) Metal and metalloid carbides, nitrides, and carbonitrides (see also 2D nanoscale objects) Metal and metalloid chalcogenides (see also 0D nanoscale objects)

Metal fluorides Metal and metalloid phosphides (see also 0D nanoscale objects) Metals and metalloids (see also 0D nanoscale objects)

Silica [52, 83, 84], alumina [24, 85, 86], borates [87–90], aluminosilicates [70, 91, 92], gallium oxide [93, 94], germanium oxide [95–97], indium oxide [93, 98], tin oxide [98–105] Scandia [106], titania [105, 107–111], vanadia [29, 112, 113], chromia [93, 105, 114, 115], manganese oxide [116–118], iron oxides [38, 93, 119, 120], cobalt oxide [105, 121–123], nickel oxide [105, 124], copper oxide [105, 125–128], zinc oxide [129], yttria [105, 106, 130–133], zirconia [93, 102, 131, 134–136], niobia [93, 137– 139], molybdenum oxide [140], ruthenia [141, 142], lanthanum oxide [106, 143, 144], hafnia [93, 105], tantala [93, 145, 146], tungsten oxide [93, 147, 148] Ceria [144, 149, 150], praseodymium oxide [106, 143, 144], neodymium oxide [106, 143, 144], samarium oxide [106, 144], europium oxide [106, 144], gadolinium oxide [106, 144], terbium oxide [106, 144], dysprosium oxide [106, 144, 151], holmium oxide [106, 144], erbium oxide [106, 143, 144], thulium oxide [106, 144], ytterbium oxide [106, 151], lutetium oxide [106, 144], thoria [152–154], urania [155] Magnesium oxide [156, 157], calcium oxide [158], strontium oxide [159] Resorcinol-formaldehyde [78], melamine-formaldehyde [160], phenol-furfural [161], polybenzoxazines [162, 163] Polyureas [25, 164], polyisocyanurates [25, 42], polyurethanes [49, 164], polyimides [80, 81], polyamides [165] Polyimides [82], polyamides [166], poly(ether ether ketone) [167] Polyacrylonitrile [168], polystyrene [169–171], poly(divinylbenzene) [172, 173], poly (trimethylol propane triacrylate) [174], polyethylene [175–177], polycyclopentadiene [178–182], polynorbornene [183] Terrestrial polysaccharides [79] including agar [184], beta glucans [185, 186], cellulose [27, 28, 187–192], galactomannan [193], hyaluronic acid [194], pectin [195– 197], starch [197–199]; marine polysaccharides such as alginate [184, 197, 199–203], alginic acid [184], carrageenan [184, 200], and chitosan [184, 200, 204–206] Casein [207, 208], chitin [184, 209], collagen [210–213], egg white protein (albumin) [101, 207, 208, 214, 215], gelatin [101, 215–218], silk fibroin [219–222], and whey protein [207, 208, 223]; M13 bacteriophage viruses [224, 225] Lignin [226–229] Clay/polymer nanocomposites [230, 231], allophanic soils [232] (▶ Chap. 54) Type I, II, III, IV, and V fulgerites (product of lightning-melted sand, rock, clay, or other minerals) [233] Amorphous carbon [65], carbon nanotubes [72, 234–236], carbon nanofibers [187], graphene and graphene oxide [54, 73, 237–240], graphite [168], fullerenic carbon [44], diamond [241] Boron nitride [22, 242–244], chromium carbide [105], iron carbide [71, 105], silicon carbide [4, 245], hafnium carbide [105], silicon nitride [246], titanium carbonitride [247] Sulfides, selenides, and tellurides of cadmium, lead, and zinc [17, 75, 248, 249], bismuth telluride [250, 251], silver selenide [252], tungsten disulfide [253], molybdenum disulfide [253, 254], iron sulfide [255], germanium sulfide [16, 256, 257], multicomponent sulfides and selenides [16, 257] Aluminum fluoride [258] (see ▶ Chap. 39) Nickel phosphide [259], indium phosphide [260] Titanium [261], iron [18, 71, 261, 262], cobalt [18, 105, 262], nickel [18, 105, 261, 262], copper [18, 105, 128, 261–264], tin [105]; noble metals including palladium, platinum, silver, and gold [18–20, 261, 265, 266]; silicon [69], aluminum [76, 261], titanium [267] (continued)

1

The Story of Aerogel

7

Table 1.1 (continued) Low-Dimensional Nanoparticles 0D nanoscale objects

1D nanoscale objects 2D nanoscale objects

1 Quantum dots, nanorods, nanotetrapods, and other nanoparticles of metal/metalloid chalcogenides [17, 75, 248–252]; metal/metalloid phosphides [259, 260, 268]; noble metal nanoparticles of palladium, platinum, silver, gold, and others [19, 20, 265]; germanium oxide [97] Carbon nanotubes [72, 234–236], metal nanowires [22, 269], M13 bacteriophage viruses [224, 225] Graphene [54, 73, 237–240], 2D hexagonal boron nitride [22, 243, 244], 2D manganese oxide [118], molybdenum disulfide [22]

precursor from which it is made, namely, the solid porous backbone of the gel. Gels are colloidal systems in which a porous, solid-phase network is permeated by a liquid component. In this regard, the aerogel-to-be exists within its gel precursor, only with liquid permeating its pores instead of a gas or vacuum. Accordingly, to produce an aerogel, this solid skeleton must be somehow separated from the liquid permeating its pores. If the liquid in a gel is merely evaporated, capillary stresses will arise as the resulting liquid–vapor interface recedes into the gel’s solid-phase skeleton. This liquid– vapor interface exhibits a surface tension, and, accordingly, when molecules of liquid within the pores of the gel escape into the surrounding atmosphere (i.e., evaporate), intermolecular attractive forces cause the remaining neighboring molecules to draw inward and fill the void resulting from the escaped molecule. These solvent molecules not only exert intermolecular attractive forces on one another but also on the solid backbone of the gel and thus also tug on and draw inward the very thin struts/particles that make up the gel backbone. As liquid continues to evaporate from the gel, tremendous capillary pressures arise, drawing the gel’s backbone in on itself and causing densification (shrinkage) and collapse of its porous structure. As the backbone densifies, its modulus increases accordingly – in some cases reaching a point where its modulus balances the evaporative capillary stresses and densification halts and in other cases collapsing with unhindered shrinkage until all of the liquid in its pores has evaporated. As the gel’s skeleton collapses inward, solidphase intermolecular interactions between its constituent struts/particles result in stiction and potentially chemical bond formation, or cross-condensation, thereby rendering shrinkage of the network irreversible. The resulting densified, low-porosity solid is typically referred to as a xerogel, and is not an aerogel. As a result, to produce an aerogel from a gel, the gel’s liquid component must be removed in a way that does not result in unhindered irreversible shrinkage of the gel’s solid skeleton. Reducing shrinkage may be achieved by one of several ways, each of which is designed to somehow circumvent or mitigate the capillary stresses associated with

evaporation. One approach is supercritical drying, also referred to as supercritical extraction, critical-point drying, hypercritical drying, and hypercritical venting (see ▶ Chap. 4). In this process, the liquid in a gel is brought past the liquid’s critical point, a thermodynamic transition that occurs at a specific temperature and pressure characteristic of a given substance wherein the liquid loses its surface tension and transforms into a liquid-like gas that can diffuse out of the pores of the gel’s solid skeleton without imparting capillary stress. Past the critical point, the meniscus between liquid and vapor disappears. Thus, at supercritical conditions no liquid–vapor interface can form within the gel upon removal of the gel’s pore fluid, thereby eliminating the threat of capillarity-induced stress. Supercritical fluids possess densities comparable to liquids, exhibit no surface tension, and expand and compress like a gas, allowing them to be removed from a gel without imparting damage to the gel’s solid skeleton. In practice, this process is performed by placing a gel inside a sealed pressure vessel, sometimes called a supercritical dryer, supercritical extractor, criticalpoint dryer, or autoclave, along with an excess volume of the same liquid held within the pores of the gel. The vessel is then heated and the resulting autogenic vapor pressure of the liquid within is allowed to rise until the critical point of the liquid is exceeded. Once past the critical point, the vessel can be quasi-isothermally depressurized to ambient conditions, that is, the fluid in the vessel can be vented to atmosphere while the temperature of the vessel is maintained above the critical temperature of the fluid such that once conditions inside the vessel drop below the critical point of the fluid, there is not enough fluid remaining in the gel to recondense a liquid phase. As most solvents used to synthesize sol–gel-derived gel precursors are organic in nature and thus dangerously flammable and explosive near their critical points, it is common to replace the liquid held within the pores of a gel (also called its pore liquor) with a safer, nonflammable solvent with a milder critical point, namely, liquid carbon dioxide, prior to performing supercritical drying. Alternatively, a flow of supercritical carbon dioxide can be used to directly extract organic solvent from a gel (more about this approach can be

8

S. A. Steiner III and A. C. Pierre

found in ▶ Chap. 4). Today, supercritical drying with carbon dioxide is the gold standard for drying of aerogels on account of its ability to maximally preserve pore structure and its compatibility with the widest range of aerogel compositions. A list of critical points for various fluids of relevance to aerogels is provided in Table 1.2. Various detailed procedures regarding supercritical drying protocols, such as the heating– pressurization path, the atmosphere above the samples during drying, and the duration of each step in the drying process, have been summarized by Pajonk [273]. Such parameters are important to control in order to minimize differential stresses and cracking within gel parts during drying. Further details about supercritical drying and its practical implementation are provided in (▶ Chap. 4). In addition, an illustrated stepby-step guide detailing how to build and operate a supercritical dryer suitable for the production of aerogels can be found in ▶ Chap. 65. Supercritical drying is not the only means by which an aerogel can be dried, however. As long as the pore structure of the gel’s skeleton is generally preserved after removal of its pore liquor, it does not matter how the pore liquor is removed. In this regard, aerogels can in fact be made through ambient-pressure evaporative drying provided that the manner in which the liquid is evaporated from the gel does not result in significant collapse of the gel’s solid skeleton (see ▶ Chaps. 14, ▶ 15, ▶ 16, ▶ 21 and ▶ 29). In order to achieve such a result, a few strategies can be employed. First, the surface tension of the pore liquor can be minimized so that when it is evaporated, the resultant capillary stresses are accordingly minimized, for example, by solvent exchanging the gel into a low-surface-tension solvent such as pentane (see ▶ Chap. 3). Second, stiction and/or cross-condensation between skeletal struts/particles due to shrinkage can be minimized so that any shrinkage that does occur can be reversed. Once evaporation of the solvent is complete, the dried gel network can then spring back to a voluminous state. Such a structural recovery can be achieved through chemical

modification of sticky, cross-condensable surface functional groups that natively line the gel’s backbone; for example, in the case of silica, which natively is heavily hydroxylated, surface hydroxyls can be replaced with bulky nonpolar trimethylsiloxy (see ▶ Chaps. 13, ▶ 16 and ▶ 65) groups through treatment with a reactive hydrophobe [84, 274]. Alternatively, precursors bearing hydrophobic side groups can be used to synthesize the gel, for example, alkyl-functionalized alkoxysilanes (▶ Chap. 14). Finally, the modulus of the gel backbone can be increased so that the gel can better withstand the capillary forces involved in evaporative drying. This can be accomplished by preparing a gel with high weight-percent solids, selecting for a gel morphology free of interparticle necks in the backbone, or tailoring the chemical composition of the backbone [25, 50]. This said, for some gels such as polyurea gels and other high-strength synthetic polymer gels, the modulus of the gel may natively be high enough that the gel skeleton can resist the stresses of solvent evaporation with no additional modification [25]. In recent years, freeze drying and related techniques have become increasingly popular alternatives to supercritical drying and evaporative drying (▶ Chap. 5). In freeze drying, the pore liquor of a gel is frozen and then subsequently sublimed, typically under vacuum. Like supercritical drying, freeze drying circumvents the formation of a liquid–vapor interface, however by going down and around the triple point of the solvent’s phase diagram rather than up and around its critical point. For many materials, freeze drying provides comparable results to supercritical drying using more commonly available laboratory equipment. Freeze drying can even been streamlined with sol–gel synthesis in some cases, for example, as in the case of materials made through freeze casting. However, freeze drying can still result in gel shrinkage and/or fracture and in most implementations requires vacuum. In addition, steps need to be taken to avoid crystallization of the gel’s pore liquor that can damage the gel’s solid skeleton. Regardless, freeze drying has proven to be an

Table 1.2 Critical points for a number of fluids relevant to the production of aerogels Fluid Water Methanol Ethanol Isopropanol Acetone Acetonitrile Neon Xenon Propane Carbon dioxide Freon ® 116

Formula H2O CH3OH C2H5OH C3H7OH C3H6O CH3CN Ne Xe C3H8 CO2 (CF3)2

Tc (
C) 373.99 239.85 240.85 235.85 234.85 271.85 228.75 16.62 96.75 31.05 19.85

Pc (MPa) 22.06 8.1 6.3 4.9 4.8 4.87 2.76 289.77 4.25 7.38 3.4

Tc (K) 647.14 513 514 509 508 545 44.4 289.74 369.9 304.2 293

Pc (atm) 217.63 79.91 62.15 48.34 47.35 48.05 27.23 57.62 41.93 72.81 33.54

Reference [271](a) [271](b) [271](c) [271](d) [271](e) [271](f) [272](a) [272](b) [271](g) [271](h) [271](i)

1

The Story of Aerogel

effective alternative to supercritical drying for preparing a variety of aerogel compositions [54, 203, 275–278]. In this regard, many feel an aerogel should be considered a state function and not a product-by-process, that is, it is not the method used to make a material that defines its aerogelness, but rather the structure–property relationships of the material. After all, one should be able to ascertain what a material is by analysis of the material alone, irrespective of whether or not one possesses knowledge of how it was created.

1.1.4

Properties

Aerogels are scientifically and technologically compelling in part because of their ability to combine a wide array of unusual and extreme materials properties into a single material envelope. Some examples include ultralow bulk density, high internal surface area, high surface-to-volume ratio, nanostructured solid-phase composition, and pore sizes smaller than the mean free path of molecules in air. From these structural features arises one of the most important functional properties inherent to many aerogels, namely, their ultralow thermal conductivity, which makes many aerogels superlative thermal insulators [62, 63, 279]. Likewise, these structural features result in a low speed of sound through aerogels and a high reflectance of acoustic waves, making aerogels superlative soundproofing materials [280– 282]. Furthermore, aerogels comprise a high volume fraction of air dispersed throughout their nanostructured pore network, resulting in a blurring of materials properties at not only the macroscale but at the microscale as well. As a result, in many respects aerogels behave as effectively homogeneous materials that exhibit many physical properties similar to those of air, such as low index of refraction [283] (▶ Chap. 56) and low and frequency-invariant dielectric constant and loss tangent over wide bandwidths [46]. Some aerogels, most notably silica, can also be optically transparent, making them uniquely valuable for applications requiring transparent thermal insulators. Note, however, this transparency is typically accompanied by a bluish cast attributable to Rayleigh scattering when viewed under indirect lighting, and, consequently, a yellowish hue when viewed in front of a light source due to Mie scattering (the large number of polydisperse nanoscale pores and particles in the aerogel, which exhibit different indices of refraction, enhances overall scattering, and since scattering intensity scales with one over wavelength to the fourth power, i.e., I / 1/λ4, blue and violet scatter more heavily than longerwavelength colors, making the aerogel appear blue). Additionally, the fractal-like structure of aerogels results in a

9

highly efficient dispersion of matter throughout threedimensional space, giving aerogels remarkable high massnormalized mechanical properties such as high compressive strength-to-weight and stiffness-to-weight ratios [25, 166, 280, 284–288] (see Fig. 1.1b and g). In the past, this aspect of aerogels was marred by the low intrinsic fracture toughness inherent to oxidic aerogels, meaning that despite possessing high mass-normalized strength and stiffness, most legacy aerogel formulations are also extremely brittle (friable) thereby limiting their ability to be used as loadbearing materials. However, over the past two decades a number of approaches have emerged for preparing aerogels with high fracture toughness and, in some cases, even plastic yielding behavior, making such aerogels compelling candidates for multifunctional lightweight structures [25, 82, 286]. In addition to their interesting intrinsic materials properties, aerogels provide a tremendously flexible architecture for installing multiple functionalities into a single material. One aspect of this versatility lies in that aerogels can be made as macroscopic parts that simultaneously embody properties arising from the nanoscale objects that comprise them, including quantum confinement effects (▶ Chap. 38), enhanced catalytic activity (▶ Chaps. 31 and ▶ 55), and high functional group density (▶ Chap. 14). Nanostructured gold aerogel monoliths, for example, exhibit catalytic behavior that bulk gold does not [265]. In this regard, aerogels represent a powerful platform for the creation of highly tailored application-specific functional materials, with examples including high-energy-density and high-power-density energy-storage devices (▶ Chap. 50), catalytic converters (▶ Chap. 55), and time-delayed drug delivery systems (▶ Chap. 59), to name a few. Because of their ability to be simultaneously highly porous and nanostructured, aerogels take on extreme values of many materials properties. In fact, various aerogels hold world records for exhibiting the most extreme value of numerous materials properties. The Guinness Book of World Records, for example, officially recognizes a silica aerogel with a bulk density of 0.0019 g/cm3 synthesized by scientists at Lawrence Livermore National Laboratory as the world’s lightest solid [53]. Since then, graphene aerogels that exhibit bulk densities as low as a 0.00016 g/cm3 when evacuated have also been reported [54] (see ▶ Chap. 36). Aerogels hold records for several other properties as well. Trimethylsilyl-functionalized silica aerogels have been prepared that exhibit ambient thermal conductivities as low as 10 mW/m-K at ambient temperature [62, 63] – two to three times lower than traditional insulating foams [59] (see ▶ Chap. 48). Electrically conductive amorphous carbon aerogels have been synthesized with surface areas in excess of 3000 m2/g – the equivalent of the area of a football field wrapped up into something the size of an ice cube [51] (see ▶ Chaps. 20 and ▶ 35). High-density polyurea

1

10

S. A. Steiner III and A. C. Pierre

Table 1.3 Select list of materials properties exhibited by various aerogels and representative examples of aerogels possessing such properties Materials Property Ultralow density Ultralow thermal conductivity Superelasticity Mechanical durability High compressive stiffness High specific energy absorption Optical transparency Low index of refraction Low dielectric constant High specific surface area Electrical conductivity Photoluminescence Superparamagnetism High-temperature stability

Cryogenic ductility Hydrophobicity Biodegradability High sound transmission loss Flexibility

Example of Aerogel with Property Silica aerogel [52, 53], graphene aerogel [54] Silica aerogel [62, 63], RF aerogel [78] Carbon nanotube aerogel [72], graphene aerogel [237] Polyurea aerogel [25], polyimide aerogel [82]

Table 1.4 Materials properties of relevance to aerogels and characterization techniques used to measure them. Consult the literature for details regarding methods and standards used to perform these techniques Materials Property Bulk density Thermal conductivity Specific surface area Pore size statistics

Polyamide aerogel [166] Polymer-crosslinked vanadia aerogel [29]

Bulk solid-phase composition and crystallinity

Silica aerogel [289–291] Silica aerogel [283]

Surface chemistry

Polyimide aerogel [46] Hierarchically porous carbon aerogel [51] Carbon aerogel [292] Cadmium sulfide quantum dot aerogel [17] Iron oxide aerogel [38], nickel ferrite aerogel [293] Yttria-stabilized zirconia aerogel [131], aluminosilicate aerogel [70], lanthanum silicate aerogel [143, 294], ceria aerogel [149] (see ▶ Chap. 18) Polymer-crosslinked vanadia aerogel [29] Hydrophobic silica aerogel [84, 274] Cellulose aerogel [28] Polyurea aerogel [282]

Hydrophobicity Morphology and particle size

Skeletal density Compressive/tensile strength and stiffness Electrical conductivity Optical transparency Ionic conductivity Sound transmission loss

Characterization Technique Dimensional analysis of mass and volume Calibrated hot plate (see ▶ Chap. 65), guarded hot plate, heat flow meter Sorptimetry using BET model Sorptimetry using BJH model, mercury intrusion porosimetry, cryoporosimetry Powder X-ray diffraction (XRD), magic-angle-spinning solid-state nuclear magnetic resonance (MAS NMR) X-ray photoelectron spectroscopy (XPS), Auger spectroscopy, energydispersive analysis (EDAX), infrared spectroscopy Contact angle, water uptake Scanning electron microscopy (SEM), transmission electron microscopy (TEM), small-angle X-ray scattering (SAXS), small-angle neutron scattering (SANS) Helium pycnometry Universal testing machine Four-point probe UV–vis spectrophotometry Impedance spectroscopy, terahertz spectroscopy Multi-microphone impedance tube

Polyimide aerogel [82]

aerogels exhibit up to 1000 times higher sound transmission loss over audible frequencies than typical acoustic insulation materials [282]. Silica aerogels have been prepared with lower indices of refraction and dielectric constants than any other material [283, 295, 296]. These are only a few examples of the extreme materials properties that aerogels can exhibit. Table 1.3 summarizes some of the interesting properties that can be embodied by aerogels and examples of aerogel materials that exhibit such properties. A list of materials properties relevant to aerogels along with a summary of techniques commonly used to measure such properties is provided in Table 1.4. These and other aspects of aerogels make them compelling for a wide range of applications, which are discussed throughout this handbook. In the following section, we review the origin of aerogels and how their unique combinations of properties have enabled and continue to enable new technological possibilities.

1.2

The History of Aerogels

1.2.1

Founding Studies by Kistler

The first aerogels were created by Samuel Stephens Kistler (Fig. 1.2) [215], most likely some time between 1923 and 1931, but probably between 1927 and 1929, and most likely at the College of the Pacific in Stockton, California, USA. (It is worth noting that Kistler’s name has been erroneously cited as Steven Kistler in many places, including the preface of the first edition of this handbook; he was never called this, and if anything, he went by Sam.) Kistler was aided by the assistance of Pacific undergraduate student Charles H. Learned [297] and conducted his research with equipment on loan from Pacific professor Dr. J. W. McBain, which consisted of a 75-cc autoclave capable of withstanding at least 300 atm that was heated in an electric oven [101]. Kistler was interested in the physical chemistry of gels and membranes at the time, appreciating their two-phase porous nature and the ability to arbitrarily exchange their pore fluid

1

The Story of Aerogel

Fig. 1.2 Prof. Samuel Kistler, inventor of aerogel. Kistler went on to become Dean of the College of Engineering at the University of Utah in 1952. (Photos courtesy Special Collections, J. Willard Marriott Library, University of Utah)

for another by soaking in a bath of a solvent with which the gel’s pore fluid is miscible [215, 298]. Kistler further appreciated that evaporation of the liquid from a gel generally results in a collapsed, densified solid and, as such, he was interested in testing the hypothesis that the liquid in a gel could be replaced by a gas without disrupting the gel’s solid component. Stemming from work done during pursuit of his master’s degree, Kistler notably also had an interest in supercritical fluids [299]. By placing a gel inside a sealed pressure vessel with an excess amount of solvent and heating the vessel beyond the critical temperature of the solvent inside while simultaneously maintaining the solvent’s autogenic vapor pressure in the vessel, Kistler found that the liquid within a gel could be converted into a gas without ever imparting capillary forces on its solid porous structure. At that point, the gas could be removed through depressurization to leave behind a low-density, dry solid possessing essentially the same volume as the original gel. It is this remaining low-density solid material that he dubbed aerogel. Therein, Kistler invented a process in which the solid porous structure of a gel could be isolated from its liquid component without resulting in significant shrinkage, a process we today call supercritical drying. Kistler had not yet obtained a PhD when he began working at the College of the Pacific, a recently constructed school with limited resources where he taught undergraduate courses, and likely discovered aerogels before he finished earning his degree. Kistler began pursuing his doctoral work at Stanford University in 1927 during his summers off from teaching, which he completed in 2 years. It is not certain what the topic of his PhD thesis was, but more than likely it was at the College of the Pacific where the first aerogels were made, as McBain, the professor from whom he borrowed his research equipment, was located there, and his assistant

11

Charles Learned was enrolled as an undergraduate there in the late 1920s. In 1930, Kistler accepted an International Student Fellowship and spent 1 year in Europe at the Kaiser Wilhelm Institute in Berlin and at the University of Göttingen, and upon his return in 1931, took a position at the University of Illinois. It is unlikely he was able to conduct his research on aerogels in Germany due to the specialized nature of the apparatus involved and none of his papers refer to his work being conducted there. Kistler’s 1932 J. Phys. Chem. paper on aerogels does cite work regarding pore fluid exchange with gelatin done at Göttingen in 1897 [101], suggesting he was thinking about the porous nature of gels during his year abroad in 1930. Deductively, it can be concluded it was during the late 1920s when Kistler and his colleagues made the first aerogels, and that it was at Pacific, not Stanford, where they were made. Kistler performed his supercritical drying technique on a variety of different gels, producing a range of aerogels with distinct chemical compositions in the process, including aerogels of silica, alumina, titania, tungstic oxide, ferric oxide, stannic oxide, chromic oxide, thoria, magnesium hydroxide, nickel hydroxide, nickel tartrate, cellulose, nitrocellulose (collodion), gelatin, agar, egg albumin, and rubber, and felt there was no reason why the list of possible aerogel compositions could not be indefinitely extended [101, 152, 153, 215]. Kistler had initially attempted supercritically drying gels from water, but found that doing so dissolved away the solid component of the gels, as water is extremely corrosive above its critical point. To circumvent this, prior to heating a gel to supercritical conditions, Kistler would exchange a gel’s pore fluid for a solvent that could be supercritically extracted without dissolving its solid component, employing ethanol for inorganic gels, diethyl ether for easily reduced substances, and propane for organic gels [215]. In his first communication on the topic published in Nature in 1931 (Fig. 1.3), Kistler somewhat casually introduces the term aerogel to describe what he had made as a seemingly obvious descriptor of a gel containing air in its pores [215]. In fact, Kistler never really defines what an aerogel is in any of his papers. Kistler published a total of nine papers [101, 152, 153, 215, 300–304] and received six patents [305–310] related to aerogels, and was the first to appreciate many of the key interesting materials properties typified by them. He appreciated that aerogels exhibit low densities arising from a high degree of porosity, producing materials as low as 0.02 g/cc in density. He appreciated their ability to be made transparent, and their opalescent qualities that arise from Rayleigh scattering of visible light. He noted the unusual metallic ringing sound that silica aerogels make when dropped. He recognized that aerogels could be extremely good thermal insulators, measuring the thermal conductivity of silica aerogel powders he had synthesized as being 10% lower than that of still air [215], and in one patent reporting a value of 20.3 mW/m-K at 34
C for a

1

12

S. A. Steiner III and A. C. Pierre

Fig. 1.3 Kistler’s original communication on aerogels, published in Nature in 1931. This one-column-long report ultimately created a worldwide multidisciplinary field

silica aerogel of density 0.18 g/cc [305]. He recognized that aerogels had incredibly high internal surface areas, measuring values of around 300 m2/g [303], and appreciated their remarkable sorptive properties. He also saw the potential of aerogels to serve as highly efficient catalysts, demonstrating the efficacy of

thoria aerogels for converting aliphatic organic acids and esters to ketones [153] and of chromia-containing alumina aerogels for converting alcohols to amines [302]. He was the first to use epoxide-assisted gelation to prepare metal oxide aerogels [302, 311] (▶ Chap. 17), a technique of critical importance to

1

The Story of Aerogel

modern aerogel research that was only recently rediscovered by the aerogel community. He was the first to solvent exchange gels with a liquid that must be manipulated inside a pressure vessel, as is frequently done today with carbon dioxide, and was in fact the first to attempt supercritical drying with carbon dioxide [101]. He envisioned that some gels, such as gels of vulcanized rubber, could be made into aerogels by evaporation of a solvent exhibiting a suitably low surface tension such as liquid nitrogen, suggesting that aerogels could in principle be produced through ambient-pressure drying [101]. He predicted cellulose aerogels, a formulation that has garnered a great deal of attention in recent years, would specifically “prove valuable” [305]. He even patented the first hydrophobic silica aerogels made through silylation with trichloromethylsilane, dimethyldichlorosilane, and trimethylchlorosilane for use as water repellents [310]. Indeed, Kistler not only discovered a new materials platform, but also set the stage for future scientists and engineers to appreciate the multifunctional superpowers of aerogels and their potential to benefit society.

1.2.2

Early Commercialization of Aerogels

Silica aerogels were the first aerogel composition of interest for commercial applications, likely because of their extreme thermal insulating properties, high-temperature stability, high-surface-area properties, low-cost sodium silicate precursor, and relative ease of synthesis. In the late 1930s, Kistler completed a licensing agreement for three of his patents with Monsanto Chemical Co. [307–309], and soon after, Monsanto began the first commercial production of aerogel,

13

producing a line of particulate silica aerogel products sold under the trade name Santocel, manufactured out of a plant in Everett, MA, USA (Fig. 1.4) — not far from where several of today’s leading aerogel companies including Aspen Aerogels, Cabot Corporation, and Aerogel Technologies are located. Much of the commercial development of aerogels at Monsanto was led by an engineer named John F. White, who was the first researcher to introduce infrared opacifiers into silica aerogel to improve high-temperature insulating performance [313], the first to develop aerogel insulation for use in refrigerators [314], and the first to explore application of aerogels for use as performance additives in coatings [315]. Monsanto was actively scaling production of aerogel as early as 1942, where researchers worked to solve problems related to convective heat transfer underlying supercritical drying of silica gels from ethanol at scale [316, 317]. Aerogel materials were apparently already being produced in small quantities before America’s involvement in the Second World War, and after the war in 1946, Monsanto completed construction of a plant with expanded production capacity [312, 318]. Santocel was produced at Monsanto’s colloidal silica plant and was manufactured by pumping water-based sol into 20-ft-tall (6.1 m) autoclaves, followed by addition of ethanol and subsequent heating to perform near-critical-point drying from the resulting ethanol–water mix. Monsanto was able to produce silica aerogel particles with densities as low as ~0.029 g/cc; however, commercial Santocel that was sold into industry is reported to have had a density of ~0.096 g/cc, corresponding to a porosity of 94%. The solid component of Santocel was reported to comprise 90% silica, the balance

Fig. 1.4 Autoclaves used to make Monsanto’s Santocel silica aerogel products at its aerogel plant in Everett, MA, USA, circa 1943. Silica aerogel powder can be seen being removed from Autoclave 3 by a technician in the photograph on the right [312]

1

14

S. A. Steiner III and A. C. Pierre

being primarily sodium sulfate (a byproduct of the sulfuric acid used to gel the sodium silicate) [318]. Santocel reportedly consisted of 2.5- to 3.5-nm particles and exhibited an average pore size of 33 nm with a specific surface area of 600 m2/g [318]. Six grades of Santocel of different densities and particle size distributions were produced, consisting of particles ranging from 10 μm to millimeters in diameter. Santocel was apparently manufactured in volumes of 2000 to 3000 tons per year (1800 to 2700 metric tons per year) [319], with specific known grades including Santocel A, Santocel C, and Santocel CF. Santocel’s primary applications were as thermal insulation, as a matting/flattening additive for paints, and as a thickening agent [318, 320, 321]. Specific known use cases include thermal insulation for an oxygengenerating plant, thermal insulation for a national line of freezers, matting agents for paints, an additive for ink used on cigarette packages, a thickening agent for screwworm salves for sheep, a thickening agent for Napalm bomb jelly [320, 321], and as a physical insecticide (e.g., Dri-Die brand) [322]. Santocel was apparently not cheap [318], but provided unique applications benefits inherent to aerogels, especially for the coatings industry. For example, Santocel provided higher loadings at lower build (viscosity) than achievable with competing matting agents [318]. Monsanto produced Santocel aerogel until around 1970 when the company sold off its colloidal silica division, which included its aerogel plant. Production of Santocel was subsequently discontinued, likely due to high maintenance and labor costs and to increasing competition from other insulation technologies such as polymer foams and fiberglass and other thickening agents such as fumed silica.

1.2.3

The Aerogel Renaissance: 1960–1990

Academic research into aerogels lost momentum for several decades after Kistler’s early work, however interest was rekindled about the same time Monsanto abandoned commercial production of Santocel thanks to the development of safer and more efficient processing techniques, as well as the emergence of a handful of exotic applications that called for unusual materials requirements which aerogels were uniquely poised to address. Simultaneously, the birth of materials science as a discipline in the 1960s [323] provided a unified framework for scientists to understand and characterize aerogels in the context of a material architecture that possesses aspects spanning multiple disciplines. With these developments came insights into how aerogels could be engineered at the length scale where their unique structure– property relationships arise, namely, the nanoscale. One of the first notable improvements in the production of aerogels was the introduction of silicon alkoxides as a substitute for sodium silicate in the preparation of silica aerogels.

Sodium silicate (formula Na2SiO3, also known as sodium metasilicate, or in aqueous solution, as waterglass) was the precursor initially used by Kistler to produce silica aerogels [101]. Sodium silicate is inexpensive and easily obtainable, but is gelled through the addition of acid, which results in the formation of problematic sodium salts that reduce transparency, increase hydrophilicity, and are corrosive to metals in high-temperature applications. Accordingly, silica gels had to be soaked, or solvent exchanged, in multiple water baths in what is a time-consuming and tedious process in order to remove Na+ ions before drying. The world of aerogel synthesis became a lot less salty upon the recognition that metal alkoxides could be used as precursors for forming gels in place of sodium silicate. Ebelmen had previously demonstrated the use of silicon alkoxides (compounds of the form SiOR4 where R is a typically an alkyl group) for synthesizing silica xerogels [324]. Silicon alkoxides hydrolyze when combined with water and catalyst in a cosolvent, resulting in subsequent polycondensation into a silica gel via alcohol condensation and water condensation reactions without forming any solid-phase precipitates [52]. Furthermore, alkoxides are often available in solutions of their parent alcohol, making it easier to directly supercritically extract the pore fluid from the resulting gels. Peri [325] at Standard Oil Co. was the first to apply this approach to silica aerogels in 1966, employing tetraethoxysilane, Si(OEt)4 (also known as tetraethyl orthosilicate or TEOS) as the silica aerogel precursor. This approach provided a significant reduction in process time compared to the previous sodium silicate pathway, reducing the multiple water bath soakings required to remove sodium salts from silica gels and eliminating the tedious water-to-alcohol exchange required prior to supercritical drying. More extensive studies with alkoxide precursors followed from the Teichner group at the Claude Bernard University in Lyon, France, in support of efforts to develop porous media for the storage of liquid oxygen and liquid rocket fuels for the French government [326]. Legend has it [327, 328] that when tasked with producing a large number of samples for this project, Teichner’s PhD student had a nervous breakdown realizing it would take many, many years to complete his PhD given that his first silica aerogel sample, which was prepared from sodium silicate, had taken weeks to synthesize. After a brief rest, the student returned highly motivated to find a more efficient process and developed a method similar to the one used by Peri, this time employing tetramethoxysilane (TMOS, also known as tetramethyl orthosilicate) in methanol solution instead of TEOS in ethanol [329]. With this new approach, the Teichner group could now prepare silica gels in a matter of hours instead of weeks. Water was now a reactant that could be added in controlled proportions with gelation catalysts to enable unprecedented control over the sol–gel process.

1

The Story of Aerogel

Teichner’s group later applied the alkoxide approach to a variety of other metal oxides as well, further extending the range of aerogels that could be synthesized [131, 326] (see ▶ Chap. 65 for a representative example of the synthesis of alumina aerogels from an alkoxide). These new compositions supported development of novel applications for aerogels, including immobilization media for rocket propellants and high-surface-area catalysts and catalyst supports [326]. New mixed-matrix oxides such as NiO–Al2O3 and CuO–Al2O3 aerogels were synthesized and could be chemically reduced to produce catalytically active Ni–Al2O3 and Cu–Al2O3 aerogel nanocomposites containing finely dispersed metal particles [326]. Aerogel nanocomposites have since been extended to numerous other compositions, with applications ranging from catalysts to energy-storage devices to energetic materials (see ▶ Chaps. 31, ▶ 47, ▶ 50 and ▶ 55). The alkoxide approach also enabled scientists to gain new insights into the formation of the nanostructure of aerogels, as the physical chemistry of hydrolysis and condensation of alkoxide precursors could be easily studied (▶ Chap. 7). Among the most outstanding studies in this area, Woignier et al. investigated the structural evolution of gels during hightemperature supercritical extraction from alcohol [330], while Vacher in Montpellier, France [331] and Schaefer et al. at Sandia National Laboratory in Albuquerque, New Mexico, USA [332] revealed the fractal nature of silica aerogel networks using small-angle X-ray scattering (SAXS) and small-angle neutron scattering (SANS). A kinetic growth model was also developed at Sandia, providing insights into how control over the morphology of silica aerogels could be improved [8](e). Concurrently, a new twostep acid–base catalysis process was developed that permitted design of very low-density silica aerogel monoliths [333]. A further benefit of the alkoxide approach is its ability to produce highly transparent monolithic aerogels, a critically enabling feature for an exotic new application of aerogels – the detection of high-energy subatomic particles produced in particle accelerators. When a charged particle such as a proton, electron, or muon passes through a medium at a velocity faster than light is able to pass through that same medium, the particle emits a cone-shaped electromagnetic shockwave of visible light called Cherenkov radiation, analogous to the sonic boom produced by an aircraft when it breaks the sound barrier. The velocity of the particle can then be deduced from the angle between this cone of light and the particle’s flight path. Previously, compressed gases and low-density liquids had been employed as media for detecting subatomic particles by the Cherenkov effect; however, no materials with indices of refraction between liquids and gases were available, thereby limiting the range and accuracy of available Cherenkov detectors [47]. In the late 1970s, particle physicists identified that transparent monolithic silica aerogels produced by the alkoxide method

15

represented an ideal medium for this application. Not only was the optical transparency of silica aerogels made by two-step catalysis (i.e., acid-catalyzed hydrolysis followed by base-catalyzed condensation) suitably high for this application [334] (▶ Chaps. 13 and ▶ 51), but their index of refraction could be tuned to controlled values in the range of 1.007 to 1.24 [283, 335], bridging the index of refraction gap needed for Cherenkov detectors. This combination of properties made silica aerogels uniquely valuable for the identification of electrically charged particles for distinguishing antimatter particles from matter particles in high-energy physics experiments (▶ Chap. 51). Along these lines, Poelz at DESY (Deutches Elektronen Synchrotron, the Germany national accelerator laboratory) [336], and Henning and Hardel at the company Airglass SE in Sjöbo, Sweden, working for CERN (the European particle physics laboratory) [337], produced hundreds of highly transparent, flawless 20 cm  20 cm  2 cm hydrophobic silica aerogel tiles in support of fabricating two large Cherenkov detectors for these facilities. In total, 1700 L of silica aerogel tiles were produced for the TASSO detector at DESY and another 1000 L for the detector at the Intersecting Storage Rings at CERN, manufactured via high-temperature supercritical drying from methanol. Similar tiles were produced for the Ring Imaging Cherenkov (RICH) counters in the HERMES experiment at the German DESY-HERA facility [338] and for threshold-type Cherenkov detectors including the BELLE detector at the KEK B-Factory in Japan [339] and the KEDR detector in Russia [340]. Momentum for using aerogels to improve the energy efficiency of buildings was also ignited in the 1980s, oddly enough thanks to Cherenkov detectors. Modern interest in aerogels for buildings can trace its origins back to the University of Würzburg, Germany, in 1983 [341]. One day on a visit to Würzburg, Prof. Max Scheer of DESY was walking through the hallways of the university’s physics institute holding a strange object in his hand when he spotted his colleague Prof. Jochen Fricke, who had been developing opaque thermal superinsulation for buildings. Scheer’s colleague Günter Poelz at DESY had just finished making the 1700 L of aerogel tiles for the TASSO detector as mentioned above. “I’ve got something for you,” said Scheer to Fricke, piquing Fricke’s curiosity. Scheer then revealed the strange object he was holding in his hand to Fricke – a transparent superinsulator in the form of a silica aerogel that had been produced for a Cherenkov detector. And with that began Fricke’s journey into exploring the structural, thermal, optical, and acoustic properties of aerogels – a journey that eventually grew into numerous worldwide collaborations that would become the foundation for efforts to integrate aerogels into buildings that are still ongoing to this day (see ▶ Chaps. 7, ▶ 8, ▶ 9, ▶ 48 and ▶ 64). Many of Fricke’s students have in fact gone on to make significant

1

16

contributions to the field of aerogels and head many of the world’s leading aerogel research groups in both academia and industry today. Low-density silica aerogels were quickly found by Fricke and others to be not only superlative thermal insulators but also superlative sound insulators [280, 342]. Significant interest in developing aerogels for use as insulation in buildings ensued (▶ Chap. 48), resulting in the development of silica aerogel insulators with good optical transparency suitable for use as window insulation [289–291, 343, 344]. Interest in using silica aerogels for windows has since led to methods that enable production of transparent silica-based aerogels without the need for supercritical drying [345, 346] (▶ Chaps. 15, ▶ 16 and ▶ 64). A substantial amount of effort was directed toward fabrication of transparent monolithic aerogel panels for windows through the European JOULE II and III (HILIT project) programs and through two French ADEME programs (PACTE projects) [279]. In support of this potential use for aerogels, transparent aerogel panels with dimensions of 55 cm  55 cm  2 cm sandwiched between panes of glass were made by Airglass as part of the European EUROSOL program [279]. Examples of transparent silica aerogel cylinders and a tile, as well as monolithic titania gels and their porous structure that illustrate the appeal of aerogels for such applications, are shown in Fig. 1.5. In addition to windows, aerogel insulators were studied for use as insulation for heating and cooling systems and for piping [347]. This included evaluation of aerogel granules, which are much less difficult and much less expensive to produce than monoliths, as insulation for translucent roofs, daylighting panels, and wall insulation. High-temperature supercritical drying proved to be a major obstacle for scaling production of aerogels, as tragically demonstrated by the explosion of the Airglass pilot plant in Sjöbo, Sweden, on 27 Aug 1984 [343]. In addition to producing silica aerogel panels for Cherenkov detectors for CERN, Airglass was contracted by the Swedish Building Research Council and the Swedish Board for Technical Development to develop aerogels for use as transparent superinsulation for windows and solar panels. The Airglass plant featured a 3000-L, 3.4-m-tall pressure vessel capable of supercritically drying one hundred 60 cm  60 cm  2 cm aerogel panels at a time from methanol at temperatures up to 260
C and pressures up to 90 bar. During a production run when the vessel was at a pressure of 86 bar and a temperature of 260
C, a 25-cm section of the gasket used to seal the enormous lid of the vessel unexpectedly failed and blew away, despite being new and carefully inspected. Methanol vapor immediately filled the building and began condensing on the walls inside. Three people were in the building at the time, two of whom escaped while the third entered the control room to initiate an emergency shutdown program on the control computer. Suddenly, a large explosion blew the roof

S. A. Steiner III and A. C. Pierre

a

b

10 mm c

d

50 nm

Fig. 1.5 Oxide gels and aerogels. (a) Silica aerogel pellets obtained from supercritical CO2 drying of TEOS-derived gels produced using a two-step acid–base catalysis first employing H2SO4 followed by NH4OH, produced as part of the French ADEME ISOGEL project. (b) Transparent silica aerogel tile (1-cm thick) synthesized using a two-step acid catalysis employing H2SO4 followed by HF, produced as part of the European HILIT+ project. (c). Titania gels obtained by substoichiometric hydrolysis of titanium tert-butoxide through HNO3 catalysis before supercritical drying, produced in support of the French Carnot MINES project. (d). Transmission electron micrograph of titania aerogel obtained from supercritical CO2 drying of the gels shown in (c). (Images courtesy of Prof. Arnaud Rigacci of the Centre Energétique et Procédés at Mines ParisTech, France)

off the building and an intense fire ensued, and soon after the building was completely destroyed. All three operators were hospitalized with burns, one of whom for a month. It is speculated that a spark from the keyboard of the computer or a static discharge from a vapor jet spraying out of the pressure vessel ignited the explosion. Airglass later rebuilt with a smaller vessel, this time planted in a bunker and controlled with a remote control system. Nonetheless, the event rattled the aerogel community and inspired efforts to switch to manufacturing processes that could be scaled more safely. Beginning around 1983, Hunt et al. at Lawrence Berkeley National Laboratory in Berkeley, CA, USA [289, 328] and concurrently scientists at BASF in Ludwigshafen, Germany [328, 348, 349] started developing means of supercritically drying aerogels from a safer, nonflammable solvent – carbon dioxide. Carbon dioxide represents an intrinsically ignitionproof alternative to organic solvents, offering a milder critical point of 31.1
C at 72.8 atm [271](h) while still providing miscibility with the types of organic solvents typically used to synthesize many aerogel precursors. Notably, Kistler had tried supercritical drying with carbon dioxide in an attempt to

1

The Story of Aerogel

17

make aerogels from swollen rubber, however his attempts were met with little success [101]. Hunt et al. developed numerous types of aerogel materials including silica aerogels for window glazings and insulating window panels using this technique, which has since been widely adopted as the standard approach for supercritically drying aerogels fieldwide. Around the same time, BASF explored supercritical drying using carbon dioxide as a method for producing its sodiumsilicate-derived silica aerogel bead product Basogel, which it marketed as a CFC-free insulation alternative for energyefficient daylighting panels, refrigerators, freezers, and boilers [348–350]. Commercial Basogel, produced at pilot scale from 1993 to 1996, was ultimately made using supercritical drying from isopropanol, as doing so provided the side benefit of imparting hydrophobicity stable to 170
C via isopropxy functionality; only public-facing and academic interactions utilized materials dried from supercritical carbon dioxide [351, 352]. Interestingly, supercritical carbon dioxide drying was used to make cellulose aerogels as early as 1971 [353] and had been used by biologists for the preservation of biological specimens for a number of years prior to being used for aerogels. This said, it was the efforts of Hunt et al. and BASF that led to its widespread adoption in the field of aerogels. See ▶ Chap. 4, for more information about supercritical drying with carbon dioxide and ▶ Chap. 65, for instructions on how to build and operate a supercritical dryer suitable for use with carbon dioxide. Development of new aerogel compositions furthered the frontiers of aerogel science in the 1980s. The list of silicate-

a

based aerogels steadily increased [355], and a large range of simple and binary oxide aerogels were investigated [326], including borate aerogels prepared by Brinker et al. at Sandia National Laboratory and by Woignier et al. [87–89, 356]. Perhaps the most significant new formulation developed in the 1980s, however, was carbon aerogel, which represented the first example of a non-oxide inorganic aerogel [65] (Fig. 1.1j). Pekala et al. at Lawrence Livermore National Laboratory in Livermore, CA, USA, had developed a new type of organic aerogel for a laser project based on the polycondensation of resorcinol–formaldehyde polymer (RF), a synthetic phenolic system [78] (see ▶ Chap. 20). RF aerogels in and of themselves represented an interesting new material system, being composed of an organic polymer and exhibiting thermal conductivities as low as 0.012 W m1 K1. However, Pekala also demonstrated that such RF aerogels, thanks to their high aromatic character, could be pyrolyzed at elevated temperatures under inert atmosphere to produce amorphous carbon aerogels – isomorphic analogs of their parent RF aerogels composed entirely of carbon [65, 357]. Examples of an RF aerogel, a carbon aerogel derived from it, and the fractal structure of the resulting carbon aerogel as observed by TEM are illustrated in Fig. 1.6. See ▶ Chaps. 20, ▶ 35 and ▶ 65 for several example procedures for synthesizing various types of RF and carbon aerogels with a diversity of materials properties. Most notably, carbon aerogels introduced a new property into the mix of functionalities attainable in an aerogel architecture – electrical conductivity. With it came the recognition

d

b

c 2 nm

Fig. 1.6 The journey taken by a resorcinol-formaldehyde polymer (RF) gel to become an amorphous carbon aerogel: (a) RF gel before drying; (b) RF aerogel after supercritical drying from carbon dioxide; (c) carbon aerogel derived by pyrolyzing an RF aerogel at 1050
C under a

100 nm flow of nitrogen gas; (d) transmission electron micrograph of the resulting carbon aerogel. See also [354]. (Images courtesy of S. Berthon-Fabry of the Centre Energétique et Procédés at Mines ParisTech, France)

1

18

S. A. Steiner III and A. C. Pierre

that aerogels could be architected to facilitate transport of electrons, holes, and ions in support of electrochemical applications. This combination of electrical conductivity with ultrahigh internal surface area opened a host of new technological possibilities for aerogels, including desalination of water, functional supports for batteries, electrodes for fuel cells, and high-energy-density electrochemical capacitors [358]. It is this latter application where carbon aerogels have had the greatest impact to date, enabling circuit-boardcomponent-sized capacitors with energy densities orders of magnitude higher than previously achievable that has resulted in what today are called supercapacitors or ultracapacitors. In fact, the first supercapacitors based on carbon aerogels were commercialized not long after their invention by the company CooperBussman (now Easton). See ▶ Chaps. 6 and ▶ 65 for guidance on how to prepare carbon aerogels from polymer aerogel precursors. These and other developments made aerogels sufficiently important to merit discussing in symposia [11, 359–364], and in 1985 the International Symposium on Aerogels (ISA) series was founded by Fricke and colleagues. Reviews on applications of aerogels were also being regularly published around this time [347, 365–369]. The first ISA was held in Würzburg, Germany [11], with subsequent meetings held in 1988 (ISA2, Montpellier, France) [359], 1991 (ISA3, Würzburg) [360], 1994 (ISA4, Berkeley, California, USA) [361], 1997 (ISA5, Montpellier) [362], 2000 (ISA6, Albuquerque, New Mexico, USA) [363], and finally 2003 (ISA7, Alexandria, Virginia, USA) [364].

1.2.4

Advances in Processing and Understanding: The 1990s

A number of notable developments in aerogel processing and characterization emerged in the 1990s. Several important advances originated at Lawrence Livermore National Laboratory in Livermore, CA, USA, during this time. Among them was a new two-step process for producing ultralowdensity silica aerogels developed by Tillotson and Hrubesh [52]. In this technique, an oligomeric silica oil is distilled from an acid-catalyzed alkoxide-derived sol and subsequently diluted with an aprotic solvent (in order to prevent reverse hydrolysis reactions). The diluted sol is then gelled in a second step via base catalysis to form an ultralow-density silica gel [52]. This process integrated two important ideas: (1) the notion of separately optimizing hydrolysis and subsequent condensation reactions in order to produce aerogels with smaller constituent nanoparticles and thus a higher degree of transparency [332, 333], and (2) the notion of separating nanoparticle formation from gelation. Concurrently, Poco and Hrubesh at Lawrence Livermore developed a new rapid supercritical drying process that unified gel

formation and supercritical drying into a single step, avoiding the need for time-consuming solvent exchanges [370, 371]. In this process, a silica sol is poured directly into a slightly-leaky steel mold, which is placed inside an autoclave along with a volume of an organic solvent (such as methanol, acetone, or acetonitrile). The autoclave is then heated above the critical point of the solvent in the autoclave, during which time the sol in the mold undergoes gelation. Upon reaching supercritical conditions, the vessel is depressurized and the supercritical solvent within the gel is removed, resulting in an aerogel. Thanks to this new process, monolithic aerogel parts with complex geometries and ultralow densities could be fabricated starting from a liquid precursor in as little as 4 h. Combined with Livermore’s two-step process for making ultralow-density silica aerogels, the rapid supercritical drying technique enabled production of silica aerogels of unprecedented low density, and in 2004, The Guinness Book of World Records officially recognized a silica aerogel monolith exhibiting a bulk density of only 0.0019 g/cc and a porosity of 99.9% made by scientists at Livermore as the world’s lowest-density solid [53]. This approach was later adopted by the NASA Jet Propulsion Laboratory in Pasadena, CA, USA, to prepare silica aerogel monoliths for insulating electronics on the Mars rovers Pathfinder, Sojourner, Spirit, and Opportunity and to prepare comet-particle collector tiles for the Stardust probe [372, 373] (see ▶ Chap. 60). The rapid supercritical drying process is currently used on a commercial basis for the production of precision silica aerogel monoliths by Ocellus, Inc., Livermore, CA, USA (see ▶ Chap. 64). An innovative variation of this technique employing a hot press and a pressurized mold rather than an autoclave has since been developed by Anderson and Carroll at Union College in Schenectady, NY, USA [374]. A recipe for synthesizing ultralow-density silica aerogels using the two-step Livermore process along with instructions for how to perform high-temperature rapid supercritical extraction using a hot press according to the Union rapid supercritical extraction method can be found in ▶ Chap. 65. One of the most impactful developments made at Livermore in the 1990s was the rebirth of epoxide-assisted gelation of metal salts for the production of metal and metalloid oxide aerogels [93, 143] (▶ Chap. 17). Before alkoxides were introduced for the synthesis of aerogels, simple metal salt precursors could be used to prepare a relatively limited number of different metal oxide gels. While alkoxides enabled additional metal oxide formulations such as alumina, titania, zirconia, hafnia, niobia, and tantala to be accessed, many metal oxide systems are difficult to make via the alkoxide route, as many metal alkoxides are easily hydrolyzed, expensive, and/or difficult to obtain. A partial charge model proposed by Livage et al. [375] highlighted that the positive partial charge δ+ on cations such as Ti, Zr, and Al is

1

The Story of Aerogel

significantly higher than for Si, meaning that hydrolysis reactions of their corresponding metal alkoxides by water (which transforms –OR ligands into –OH groups needed for oxide network formation) occur much faster than for silicon alkoxides, and is even more so the case for other transition metals. Consequently, condensation of the hydrolyzed alkoxide precursors of such metal cations generally leads to dense oxides or hydroxides rather than porous gels. Replacing one or more of the –OR groups on the alkoxide with non-hydrolyzable ligands can mitigate this problem in some instances; however, the limitations of alkoxide chemistry precluded its use for accessing aerogels of many metal oxides. Some time around 1992, Tillotson, Sunderland, and Hrubesh at Lawrence Livermore began developing a more robust technique for preparing metal oxide gels involving gelling common metal salts using epoxides (oxiranes) as gelation agents [143], explored in depth later by Gash et al. in 2001 [93]. Although generally thought of as a recent invention, the technique was actually first reported by Ziese in 1933 [311] and was even used by Kistler in 1938 to prepare a variety of metal oxide aerogels including aerogels of alumina, titania, chromia, and iron oxide [302], however was seemingly forgotten about for nearly 50 years. In this technique, a hydrated metal salt such as a chloride serves as the metal oxide source. An epoxide such as propylene oxide or epichlorohydrin then serves as a proton scavenger, plucking off protons from the aquo ligands that are datively bound to the metal cation. The protonated epoxide then undergoes an irreversible ring-opening reaction, converting the aquo ligand into hydroxyl and decreasing the charge on the metal center in the process. The resulting metalol group (M–OH) can subsequently undergo water condensation with another metalol group in the system to form a metal–oxygen–metal bridge (M–O–M). Since the formation of metalols occurs at a different (and often quite narrow) pH range for each element, the epoxide-assisted route offers exquisite control over the sol–gel transition by slowly and quiescently raising the solution pH until the exact point where gelation can occur is reached. This is in contrast to metal alkoxides, which often undergo rapid, uncontrolled precipitation upon hydrolyzing, resulting in the formation of an oxide, hydroxide, or for certain early transition metals, terminal oxo groups (M ¼ O) instead of metal–oxygen–metal bridges. The rebirth of epoxide-assisted gelation has since led to the opening of the periodic table to dozens of previously inaccessible compositions of inorganic oxide aerogels, as well as improved synthetic routes for metal oxide aerogels previously only accessible with alkoxides. This includes facile methods for synthesizing high-quality transparent alumina [85], titania [107], and other refractory oxide aerogels, as well as robust synthetic routes for historically challenging oxides such as chromia [93], iron oxide [38, 93, 119, 120],

19

nickel oxide [124], copper oxide [125], oxides of the lanthanides [106, 143, 144] (Fig. 1.1f) and actinides [154, 155], main group metal oxides such as tin oxide [99], mixed-matrix oxides such as yttria-stabilized zirconia [131], and mixed-metal oxides [293, 376, 377]. For some elements, different phases of the same oxide can even be accessed simply by adjusting solution-phase process parameters [120]. Metal oxide aerogel nanocomposites (▶ Chap. 31) such as aerogel thermites containing dispersed aluminum nanoparticles [378] can be easily produced as well (▶ Chap. 47). Metal salts and alkoxides can be also be blended to achieve stable multimetal oxide compositions that are inaccessible through the use of epoxide-assisted gelation or alkoxides alone [294, 379]. Indeed, the ability to create single-phase multimetal oxide compositions exhibiting nearly the same stoichiometry as the precursors used to form the gel has proven to be a powerful feature of the epoxide-assisted route. See ▶ Chap. 65 for recipes for preparing a variety of single and multimetal oxides via epoxide-assisted gelation of metal salts. In addition to alkoxides and metal salts, many more precursors were adapted in support of synthesizing aerogels of various compositions and functionalities, including oxyalkoxides OxM(OR)y (used, for example, to synthesize vanadia aerogels [29, 112]) and a long list of XSi(OR)3 precursor molecules, where X is an organic group (see ▶ Chap. 14). This latter class of compounds enabled new, useful characteristics to be introduced into gels and aerogels (Fig. 1.1e), including functional properties such as hydrophobicity and flexibility, the ability to graft substrates via orthogonal reaction mechanisms (e.g., using radical-mediated chemistry to bond to a vinyl side group), addition of a secondary polymerization mechanism, and installation of chemical, catalytic, and sensing functionalities (see ▶ Chaps. 29 and ▶ 49). Consequently, a wide range of hydrophobic inorganic aerogels (▶ Chap. 14) and hybrid organic–inorganic aerogels have since been developed (▶ Chap. 15). See ▶ Chap. 65 for example recipes for synthesizing organically modified silica aerogels from such precursors. Another approach for incorporating functionality into aerogel architectures, referred to as nanogluing, was demonstrated in the 1990s by Rolison et al. at the Naval Research Laboratory (NRL) in Washington, DC, USA. Rolison observed that adding functional guest materials into metal oxide sols at the initial stages of sol preparation frequently results in a near-total encapsulation of the guest material by surface-active colloidal oxide well before gelation occurs. Guest surfaces in the resulting nanocomposite are accordingly obstructed, as are guest–guest ionic, electronic, and thermal communication, thereby precluding many end uses of such guest–host aerogel nanocomposites. By adding solidphase guests to the sol just prior to gelation instead

1

20

(preferably less than a minute before gelation occurs), guest particles can be swept into the gel’s solid phase in a way that leaves their surfaces accessible upon drying. The resulting guest-accessible aerogel nanocomposites were shown to retain the high porosity and surface area expected of the aerogel host while simultaneously adding the properties of the guest. The NRL team showed nanogluing to be a simple and reliable method for incorporating solid guests of a wide range of compositions (polymers, ceramics, metals, superconductors, bioconjugates, carbons), functionalities, and sizes (ranging from a few nanometers to millimeters) into oxide aerogels without the need for prior functionalization or expensive precursors [380–383] (Fig. 1.7). One of the most significant advances in aerogel production in the 1990s was the ability to produce silica aerogels at ambient pressure without supercritical drying, as summarized by Land et al. [384]. Research into subcritical drying of aerogel-like high-surface-area silica powders had already been pursued in the 1950s in industry, with notable early developments from Alexander et al. at Du Pont [385] and from Tyler [386] and Lentz [387] at Dow Corning. In the Du Pont approach, silica gels are strengthened and esterified in an alcohol at elevated temperatures prior to vacuum drying, permitting retention of surface areas in excess of 400 m2/g. In Dow Corning’s approach, silica gels are chemically modified with reactive organosilanes such as hexamethyldisiloxane, trimethylchlorosilane, and 3,3,3-trifluoropropyldichloromethane and subsequently exchanged into a low-surface-tension solvent such as n-heptane prior to

Fig. 1.7 Silica-based nanocomposite aerogels prepared through nanogluing. From left to right: pure silica aerogel; colloidal platinum/silica nanocomposite aerogel (2-nm Pt); colloidal gold/silica nanocomposite aerogel (30-nm Au); carbon black/silica nanocomposite aerogel (Vulcan carbon black, XC-72); poly(methylmethacrylate)/silica nanocomposite aerogel (polymer molecular weight ~15,000, sieved to SiðOC2 H5 Þ4 > Siðn-OC3 H7 Þ4 > Siðn-OC4 H9 Þ4

2.3.3

Exchange Reactions of Alkoxides with Alcohols

An important property of metal alkoxides is the capability of activating exchange reactions with alcohols. This has to be

H3C

CH3 O O

Si

O

M.W. = 208.33 d = 0.94 g cm−3 b.p. = 168°C

O

H3C

CH3 Si(OC2H5)4

Tetrabutyl orthosilicate H3C H3C

O O

Si

O O

Tetrapropyl orthosilicate CH3 M.W. = 320.54 d = 0.899 g mL−1 b.p. = 275°C CH3

Si(OCH3CH2CH2CH2)4 Fig. 2.1 The most common silicon alkoxides and their properties

ð2:1Þ

tetramethyl orthosilicate > tetraethyl orthosilicate > tetra n-propylorthosilicate > tetrabutyl orthosilicate

Tetraethyl orthosilicate M.W. = 152.22 d = 1.03 g cm–3 b.p. = 121°C

H3OC Si OCH3

2.3.2

H3C H3C

O O

Si

O

CH3

O

CH3

Si(OCH3CH2CH2O)4

M.W. = 264.43 d = 0.916 g mL−1 b.p. = 224°C

56

P. Innocenzi

taken into account during sol–gel processing, in fact, the presence of alkoxides with mixed ligands affects the hydrolysis and condensation reactions because they have a different reactivity and solubility. In general, mixing a metal alkoxide with a different alcohol produces almost immediately an exchange reaction with the formation of mixed ligands [16]. An example is the case of tetramethyl orthosilicate (TMOS); the equilibrium reaction with ethanol becomes (Eq. 2.2): SiðOCH3 Þ4 þ C2 H5 OH Ð SiðOCH3 Þ3 ðOC2 H5 Þ þ CH3 OH

2.3.4

ð2:2Þ

Hybrid Materials and Organically Modified Silicon Alkoxides

Another important family of silicon precursors is composed of the so-called organosilanes or organically modified alkoxides which are characterized by the presence of at least one Si-C bond [17]. This specific group of alkoxides has allowed the preparation of hybrid organic-inorganic materials through sol–gel reactions due to the presence of the non-hydrolyzable bond between silicon and carbon [18]. The chemistry and property of hybrids is dependent

on the different geometry, length, rigidity, and functionality of the organic functional groups. The true nature of hybrid organic-inorganic materials synthesized via sol–gel processing is somehow difficult to define with clarity. In general, they are characterized by a direct covalent chemical bond which connects the inorganic and organic species. Incorporation of organic molecules into an oxide matrix, which can be easily done because of the synthesis of the material, is performed at low temperature route through solution processing, gives in most of the cases the formation of composites at molecular level [17] even if they are considered by some authors also a particular type of hybrid. An example is the incorporation of fluorescent dyes [19], such as rhodamine 6G [20] or rhodamine B into a sol–gel matrix. The final material can be considered a composite at the molecular level because the host molecule does not change the chemical-physical properties. The surrounding chemical environment can affect the optical response but rhodamine B can be still clearly identified as a single molecule [21]. On the other hand, if rhodamine B is modified to form a silylated dye (Fig. 2.2), it can be directly used during the sol–gel reactions to form a hybrid where rhodamine B is covalently connected to the silica network and is a network modifier [22].

O N

OH O N

+ +

N

O

H2N

N

NH2 N NH2

NH2 N

NH2

N

O

1

Tris(2-aminoethyl)amine Rhodamine B

+

O O Si O HN

O N

N N

N

CH3

O

H3C

O HN

O

HN

HN Si O O O

O O Si O

NCO

CH3

3-(triethoxysilyl)propyl isocyanate

2 Fig. 2.2 Synthesis of a rhodamine B derivative as precursor of a hybrid functional material. Reaction of rhodamine B with tris(2-aminoethyl)amine gives the intermediate 1, and further reaction with 3-(triethoxysilyl)propyl isocyanate allows forming the hybrid precursor 2

2

Overview of the Sol–Gel Process

Fig. 2.3 Silicon derivatives containing organic groups as modifiers

57

H3C H3C

O

O Si S

CH3 O CH3

CH3

H3C H3C

O

Sii

H3C

O CH3

H3C

O

O Si

2

CH3 O

NH2

Methyltrimethoxysilane Dimethyldimethoxysilane (3-aminopropyl)trimethoxysilane

The identification of a hybrid organic-inorganic sol–gel material is, therefore, not always so straightforward, and a clear definition is difficult to achieve. Materials formed by interpenetrating organic and inorganic networks are another interesting example of hybrid composite. They can be obtained via sol–gel processing by independent polymerization of the organic and inorganic networks [23]. The possibility of forming a homogeneous composite depends on the capability of obtaining comparable rates of organic polymerization and inorganic polycondensation during the synthesis. When the organic polymerization is faster than the formation of an extended inorganic network via sol–gel reactions and vice versa, a heterogeneous structure with phase separation is observed. The organosilanes can be divided into two main groups: in the first one the organic function is a modifier of the oxide network which is formed via polycondensation reactions, the second one is polymerizable and can form an organic network by itself which would be covalently linked to the inorganic backbone [24]. In this last case the formation of the organic and inorganic networks, which are linked through a Si-C bond, is a competing process. The family of organically modified alkoxides includes also some very peculiar precursors such as the silsesquioxanes [25] and the bridged polysilsesquioxanes which form hybrid organicinorganic materials via polymerization of monomers with two or more trialkoxysilyl groups. Because of the presence of organic linear chains and the possibility of secondary bonds between adjacent molecules these precursors can easily form organic crystalline structures.

2.3.5

Modifiers

The most common group of organosilanes is formed by alkoxides that are bonded to one or more organic functional groups that modify the inorganic silica network. The organic species cannot react and do not participate in the sol–gel reaction even if the reactivity and solubility of the alkoxide are changed as a function of the type of functional group. They have the general formula R′Si(OR)3 (with R′ the organic modifying group) but can also substitute two (R′2Si(OR)3) or three (R′3Si-OR) alkoxy functions. A very wide range of precursors with different functional groups such as amine, isocyanate, thiol, amide, and polyether and is now commercially available.

In this group of organosilanes are included some very common compounds that are largely employed for sol–gel hybrid chemistry such as methyltrimethoxysilane (MTMS, CH3-SiO (CH3)3) [26] or (methyltriethoxysilane (MTES)) [27], and coupling agents or surface modifiers, such as 3-aminopropyl triethoxysilane (APTES, H2N(CH)3Si(OC2H5)3) (Fig. 2.3). APTES is somehow a special type of precursor, because of the presence of a primary amine which can also easily react with many other organic species, such as epoxides, to form hybrid material of more complex structure [28]. This group of organosilanes can form a hybrid material by themselves via hydrolytic reactions but in general they are co-reacted with another alkoxide, such as TEOS to obtain the final hybrid material. This requires careful control of the kinetics of the reaction to obtain a homogeneous material without phase separation [29].

2.3.6

Organosilanes with Polymerizable Organic Groups

Another group of organosilanes which are common precursors for hybrid materials is formed by alkoxides modified to have polymerizable functions as the organic group, such as epoxy (3-glycidoxypropyltrimethoxysilane, GPTMS) [30, 31], vinyl (vinyltrimethoxysilane, VTMS), or methacrylate (3-methacryloxypropyltrimethoxysilane (MPTMS)) (Fig. 2.4). An important example is GPTMS which has an epoxy ring that forms upon controlled opening a poly(ethylene oxide) polymeric chain. Such as in the case of interpenetrating organic and inorganic networks the control of the kinetics is very important. In general, higher will be the condensation of the silica network and shorter will be the organic chains because of the smaller room for growth within the gel structure. The hybrids produced using this class of organosilanes may contain, therefore, an organic chain whose extent within the silica matrix depends on the synthesis conditions.

2.3.7

Silsequioxanes

Cage-like structured organosilicon molecules with Si-O-Si bonds and silicon atoms at the tetrahedral vertices are another particular class of hybrid precursors. These compounds are

58

P. Innocenzi

Fig. 2.4 Silicon derivatives modified with polymerizable organic functional groups

H3C H3C

O

O Si

H3C

CH3 O

H3C

Si

CH3 O

O

O

3-glycidoxypropyltrimethoxysilane

Vinyltrimethoxysilane H3C H3C

O

O

O

O

Si

CH3 O

CH3 O

CH2

O 3-methacryloxypropyltrimethoxysilane

R O

R

Si

O

Si O

R Si O

Si

O

R O Si Si

R

O

O

R O O

O Si

O

Si

R

R Fig. 2.5 The structure of a polyhedral oligomeric silsesquioxane (POSS)

characterized by a well-defined structural unit and silsequioxanes may also have a polymeric structure with a ladder-like repeating unit [32] or random or open cage structures [25]. The general formula of silsequioxanes is (RSiO1.5)n with the substituent R ¼ H, alkyl, aryl, or alkoxy. The composition explains the name because every silicon atom is linked in average to one and a half (sesqui) oxygen atoms and to one hydrocarbon group (ane). The functional groups give the property of the silsesquioxanes and can be also hydrolyzed and condensed in the case of alkoxy, chlorosilanes, silanols, and silanolates. It is also possible to synthesize well-defined structures defined as polyhedral oligomeric silsesquioxanes (POSS) (Fig. 2.5).

2.3.8

Bridged Silsesquioxanes

A particular type of precursor for hybrid materials is composed by bridged silsesquioxanes organosilanes (R′O)3SiRSi (OR)3 [33, 34]. They are characterized by the presence of an

organic spacer bridging two or more silicon atoms (Fig. 2.6). If the nature of the organic spacer, R, and the synthesis are carefully designed, hybrid materials with a long-range structural order can be fabricated. The bridged silsesquioxanes organosilanes precursors have a higher number of available siloxane linkages. In the case of two silicon atoms bridged by the organic spacer, they become six instead of four. This property is reflected in a peculiar reactivity during the sol–gel reactions which favor the formation of closed ring structures. The organic spacers produce also favorable conditions to self-organization into a crystalline hybrid structure. This has been observed in bridged polysilsesquioxanes with different types of organic spacers; lamellar crystals [35] but even helical fibers have been observed to form during gelation of a bulk hybrid gel and in mesoporous hybrid silica materials. Bridged silsesquioxanes have been also widely used as precursors to obtain different types of aerogels [36].

2.3.9

Transition Metal Alkoxides

The chemistry of transition metal alkoxides is much different from that of silica; silicon is in fact tetrahedrally coordinated to oxygen, while metals have usually an octahedral coordination [37, 38]. The tetrahedral structure of silica is much more flexible than octahedras and is able to form “polymeric” structures of different types. Transition metals, such as Ti or Zr, are generally more electropositive than silicon and this makes titanium (and the other transition metals) more prone to nucleophilic attacks [39]. Non-silicate metal alkoxides are, therefore, very reactive with water because they are salts of alcohol or acids and react as strong bases. The hydrolysis rate of

2

Overview of the Sol–Gel Process

59

Fig. 2.6 Examples of bridged polysilsesquioxanes: 1,4-bis (triethoxysilyl)benzene and 1,10bis(triethoxysilyl)decane

CH3 O

H3C

O

O

Si O

O

( )8

O

CH3

CH3 Si

2 CH3

O

Si

H3C CH3

O

CH3

O

H3C

O

CH3

Si

H3C

1,10-bis(triethoxysilyl)decane

O

O CH3

1,4-bis(triethoxysilyl)benzene

titanium alkoxide is generally up to 105 times faster than for the corresponding silicon alkoxide. The hydrolysis and condensation reactions must be, therefore, controlled by using complexing ligands such as acetylacetone to inhibit condensation reactions and avoid precipitation. In general, a better control of the reactivity is obtained by forming complexes by replacing the alkoxy groups with diols, β-diketonates, carboxylic acids, amines, or other organic groups such as cyclooctatetraene. The most common titanium alkoxides are: titanium ethoxide, Ti(OCH2CH3)4 (Ti(OEt)4); titanium isopropoxide, Ti(OCH(CH3)2)4 (Ti(OiPr)4); and titanium butoxide, Ti (OCH2CH2CH2CH3)4 (Ti(OBu)4) (Fig. 2.7).

H3C

O

H3C

O

Waterglass

Soluble silicates of alkali metals (sodium, potassium, or lithium), commonly known as waterglass, are another class of precursors for sol–gel processing. They are highly soluble in water which is also used as solvent of the reaction and this is, besides the much lower cost, the difference with silicon alkoxides, which are instead immiscible in water and require in general an alcohol as co-solvent for the reaction. Another difference is that in waterglasses the process is initiated by a pH change while in the alkoxides by the addition of water and the catalyst. The silica glass precursors have the general formula (Eq. 2.3):

O

CH3

O

CH3

CH3 H3C H3C

H3C

2.4

Ti

Titanium (IV) ethoxide M.W. = 228.11 b.p. = 150°C Titanium (IV) isopropoxide M.W. = 284.22 b.p. = 232°C

O− Ti4+ 4

CH3 Titanium (IV) butoxide

O O Ti O O

CH3

M.W. = 340322 b.p. = 206°C

Fig. 2.7 Titanium alkoxides are most commonly employed in sol–gel synthesis :

:

mSiO2 M2 O nH2 O

ð2:3Þ

with M the alkali metal and m the molar ratio which defines the number of silica moles per oxide metal (M2O). The average composition of silicate species in waterglass solutions is M2SiO3 (with M ¼ Na or K); the solutions of waterglass are formed by mixtures of monomeric and oligomeric silicates with negatively charged non-bridging oxygen. In the case of sodium silicate (Na2SiO3), for instance, the hydrolysis reactions are initiated by addition of hydrochloric acid (Eq. 2.4):

60

P. Innocenzi þ

þ



þ

 Si-O-Na þ H3 O Cl ! Si-OH þ Na Cl



ð2:4Þ

and the condensation by reaction of two silanols. The stability of waterglass solutions is reached only in strongly basic conditions when the anionic species repeal each other. On the other hand even if the equilibrium of waterglass solutions depends on several parameters, such as temperature, concentration, and pH, the higher complexity chemistry makes the silicon alkoxides a much more flexible precursor for sol–gel processing.

H Ti4+

+ :O

½TiðOH2 Þ

Hydrolysis

MðORÞn þ H2 O ! ðROÞn1 MOH þ ROH

ð2:5Þ

with R the alkyl group, ROH the alcohol, and M the metal. The hydrolysis reaction in the case of TEOS becomes (Eq. 2.6):

ð2:8Þ

Ti:O

H

H

Ð ½TiðOHÞ3 þ H Ð ½Ti¼O

The beginning of the sol to gel conversion is the hydrolysis of the alkoxide. In general, because silicon alkoxides react slowly in a water-ethanol mixture the addition of a catalyst is necessary to start the reaction. During the hydrolysis the hydroxyl groups (OH) replace the alkoxide species (OR); the process produces the release of an alcohol molecule and the formation of a metal hydroxide, M-OH (Eq. 2.5):

4+

This produces a charge transfer from the oxygen to the metal atom with a contemporary increase of the partial charge of hydrogen; this makes the water molecules coordinated with the metal ions more acidic than those not coordinated. The extent of hydrolysis is therefore depending on water acidity and the entity of the charge transfer up to reaching the equilibria (Eq. 2.9): 4þ

2.5

H

2þ

þ þ

þ 2H

ð2:9Þ

In non-complexing aqueous media, therefore, following Eq. 2.4, three types of different ligands would form [42, 43]: Ti-ðOH2 Þ ðAquoÞ Ti-OH ðHydroxoÞ Ti¼O ðOxoÞ Metal cation charge and pH are the two parameters that regulate the extent of the three domains, aquo, hydroxo, and oxo, as shown in Fig. 2.8: Because condensation goes through the reaction of hydroxo groups with water elimination (Eq. 2.10):  Ti-OH þ OH-Ti ! Ti-O-Ti þH2 O

ð2:10Þ

SiðOC2 H5 Þ4 þ H2 O ! ðC2 H5 OÞ3 SiOH þ C2 H5 OH ð2:6Þ

MðORÞn þ H2 O

ðROÞn1 MOH þ ROH

ð2:7Þ

Alcohol has, therefore, an active role in sol–gel reactions, not only as solvent but a careful choice of the alcohol as a function of the reaction design has to be taken into account. Non-hydrolytic reactions are also possible and several synthesis routes, especially for transition metal alkoxides and mixed precursors, have been proposed [40, 41]. In the case of titanium, but it can be generally extended to other transition metals, the nucleophilic addition of water to the Ti center is the mechanism at the base of the hydrolysis. When titanium is dissolved in water as a salt the T4+ cations are solvated by the water molecules (Eq. 2.8):

Charge (z)

The reaction of alcohol with a hydrolyzed species can, however, change direction forming again a water molecule and an alkoxide ligand (esterification) (Eq. 2.7):

+8 +7 +6 +5 +4 +3 +2 +1

O2− OH −

H2O 0

7 pH

14

Fig. 2.8 The aquo, hydro, and oxo domains as a function of the charge and pH

Overview of the Sol–Gel Process

61

Equation 2.5 has to be shifted to the Ti-OH region by controlling the pH. In the case of titanium alkoxides, and more in general transition metal alkoxides, because they are stronger Lewis acids than silicon, a nucleophilic attack is easier and this results in a higher hydrolysis rate. The condensation goes through the M-OH reaction which can be so fast that almost immediate precipitation can be observed upon addition of water [44].

2.6

The Point of Zero Charge

The hydrolysis reactions do not define a very specific stage of the sol–gel process because as soon as hydrolysis starts, the condensation reactions proceed. The pH of the solution strongly affects the species that are forming, and the final gel would have a different structure. The condensation reactions of silica are directly depending on the catalytic conditions that are employed. The Point of Zero Charge is very helpful to understand this point. The surface potential of hydroxides is given by the balance between H+ and OH ions and the charge of the surface is pH dependent (Eq. 2.11): þ

þ

M-OH þ H ! M-OH2 ðpH < PZCÞ

ð2:11Þ

pH < PZC ! Surface positively charged M-OH þ OH ! M-O þ H2 O ðpH > PZCÞ

ð2:12Þ

pH > PZC ! Surface negatively charged In the two extremes of pH, below 2 and more than 13, the hydrolysis is very fast while condensation is hindered or very slow; the sols tend to stabilize because the particles with the same charge repeal each other. The pH and the PZC affect therefore the reaction rates and the gelation time. In an acidcatalyzed sol (TEOS, HCl with H2O/TEOS ¼ 4), the gelation time as a function of pH shows a response which looks like a Gaussian curve (Fig. 2.9). At pH ¼ 2.2, which corresponds also to the PZT of the system, the sol is quiet stable with the longest gelation time. At higher pH the gel time quickly decreases such as at low pH.

2.7

Gelation time (hours)

2

From a Sol to a Gel: The Condensation

The condensation reactions begin through condensation of reactive –OH to form M-O-M- units by realizing a water or an alcohol molecule. The reaction of two -M-OH groups gives water as by-product (Eq. 2.13) while the reaction of -M-OH with -M-OR releases an alcohol molecule (Eq. 2.14):

2

90 80 70 60 50 40 30 20 10 0

1

2

pH

3

4

5

Fig. 2.9 Gelation time as a function of pH for a silica (TEOS) acid catalyzed (HCl) sol

-M-OH þ -M-OH ! -M-O-M- þ H2 O

ð2:13Þ

-M-OH þ -M-OR ! -M-O-M- þ ROH

ð2:14Þ

The polycondensation reactions allow forming an extended oxide network whose structure and growth depends on a set of synthesis parameters. Water alone cannot activate the hydrolysis and condensation reaction in absence of a catalyst. The choice of the catalyst is very important because the sol structure and the gel time are largely dependent on this selection [45].

2.7.1

Acid-Catalyzed Hydrolysis and Condensation

In an acid-catalyzed sol–gel reaction, the protons that are available in solutions will attack the oxygen atoms of the Si-OR groups to look for electrons. As a consequence the electronic cloud in the Si-O bond shifts from silicon to oxygen with an increase of the silicon atom positive charge. This makes silicon more electrophilic and more reactive to the attack from water in the hydrolysis or from silanols in the condensation reactions. The higher electrophilicity of silicon induced by the protonation has also the effect of changing its reactivity; the unreacted alkoxide (Si-(OR)4) hydrolyzes faster than the partially hydrolyzed (Si(OR)4-x(OH)x) or condensed (-Si-O-Si-) species (Fig. 2.10). This means that the pH of the sol changes with the progress of the hydrolysis and condensation reactions. The silanol groups, in fact, become more acid with the increase of condensation when more

62

P. Innocenzi

Fig. 2.10 Protonation of the different silica species in an acidcatalyzed reaction

Protonation of

Si-OR

H Si OR + H+

H + HOH

Si O+

Si O+ R

R

Protonation of

Si-OH

H Si OR +

H+

Si O+

Si

Si-O-Si bonds are present; this is also reflected in a change of PZC with the increase of condensation.

Base-Catalyzed Hydrolysis and Condensation

OH + ROH

H O+

H

2.7.2

Si

+ Si R

OH

Si O

Si

Deprotonation of water Si OR + HO

−

OH Si OR

–

Si OH + RO

−

Deprotonation of silanols The hydrolysis and condensation reactions, in the case of base-catalyzed sols, are promoted by the hydroxyl ions (OH) which have strong nucleophilicity and are strong enough to attack directly the silicon atom. Silicon in the alkoxide is the atom with the highest positive charge and becomes, therefore, the target of the nucleophilic attack from deprotonated hydroxyls (OH) or silanols (Si-O). In the base-catalyzed reaction OH and Si-O species replace OR (hydrolysis) or Si-OH (condensation), respectively. The associative mechanism involves the formation of a pentacoordinate intermediate (Fig. 2.11). The condensation reaction can happen also for R at the place of H. These reactions in high basic conditions are also reversible via cleavage by OH.

2.8

The Role of Water

The different stages of the sol to gel transition are strongly dependent on the water/alkoxide ratio, r, and the amount of water available to start the hydrolysis affects the kinetic of the polycondensation process. The stoichiometric value of r is 4 means that four molecules of water are necessary for complete hydrolysis of a tetravalent alkoxide M(OR)4, while a ratio of 2 is enough for conversion of M(OR)4 into an oxide. An increase in the amount of water available for hydrolysis should also increase the polycondensation rate. This is not actually the case because increasing the water content while keeping constant the amount of solvent produces a decrease

−

Si OH

−

SiO

Si O Si OH

–

Si O Si HO

−

Fig. 2.11 Protonation of the different silica species in a base-catalyzed reaction

of the silicates concentration. This dilution effect changes the hydrolysis and condensation rate with an increase of the gel time. The gel time changes as a function of the water/alkoxide ratio keeping constant the solvent content. An example of this effect is shown in Fig. 2.12, for a TEOS sol with ethanol used as the solvent. The response changes when water is present in substoichiometric amount, r < 4 (gray area), or higher, r > 4 (white area). As soon as the content of water increases, the gel time decreases, because more water is available for the hydrolysis; after around r ¼ 5, however, the dilution effect is more effective and the gelation time increases quite quickly with the water content. The gel time also increases with the amount of ethanol in the sol; the concentration of the oxide species is very important, more the sol is diluted longer will be the gel time. Another question that arises with the increase of water is that the system could potentially enter in an area of immiscibility in the ternary phase diagram water/TEOS/ethanol; the polycondensation reactions, however, also produce alcohol as a by-product which in most of the cases is enough to homogenize the system.

2

Overview of the Sol–Gel Process

63

1

3

102

2

Relative reaction rate

Gelation time (hours)

103

2

Condensation

Hydrolysis

0 0

7

14

pH

1

Fig. 2.13 Relative hydrolysis and condensation rates as a function of pH for a silicon alkoxide

1 2

4

8

16

H2O/ TEOS molar ratio Fig. 2.12 The gel time as a function of the water/TEOS molar ratio. The three different curves show the change of gel time at different ethanol/TEOS ratios (1, 2, and 3) [46]

The organically modified silicon alkoxides, R′(SiOR)3, have a higher electron density at the silicon atom and Eq. 2.14 can be rewritten: 0

Si-R > Si-OR > Si-OH > Si-O-Si

2.9

Hydrolysis Versus Condensation

During hydrolysis and condensation reactions the silicon alkoxides undergo a transformation through transition states; the electronic density of the silicon atom, as we have seen, depends also on the nature of the substituents and decreases with the progress of the reactions in the following order (Eq. 2.15): Si-OR > Si-OH > Si-O-Si

ð2:15Þ

The decrease in electron density of silicon during acid catalyzed reactions has the consequence that in acidic conditions also the reaction rates of hydrolysis and condensation increase with the electronic density; higher the electronic density (Si-OR), higher is the hydrolysis rate. In acid-catalyzed systems, therefore, the hydrolysis is faster than condensation while in base-catalyzed sols a reverse trend is observed [47]. Besides the differences in the reaction rate the basic and acidic routes produce also a more subtle difference which is the structure of the silica clusters. In acid conditions, because of the higher reactivity of the electrophilic silicon atom with the growth of -Si-O-Sibonds the formation of more chain-like structures is favored. On the other hand, in basic conditions branched and more connected silica structures are obtained.

ð2:16Þ

this means that, in comparison with a silicon alkoxide, they have in acidic conditions a higher reactivity which increases with the number of organic substituent and in the case of R ′ ¼ CH3 (Eq. 2.16): ðCH3 Þ3 -Si-OCH3 > ðCH3 Þ2 -Si-ðOCH3 Þ2 > ðCH3 Þ-Si-ðOCH3 Þ3

ð2:17Þ

The reactivity of organically modified alkoxides depends also on the steric hindrance of the organic substituent groups and increases in the order (Eq. 2.18): MTES > VTES > TEOS

ð2:18Þ

In basic conditions the reaction rates of hydrolyzed or partially hydrolyzed species are higher than the monomeric alkoxide, the opposite for what we have seen for acid catalysis. This is also true for the organically modified alkoxides which in basic conditions react slower than the corresponding silicon alkoxide. The pH of the solution will trigger, therefore, the reaction rates of hydrolysis and condensation which remain competing reactions for all the sol to gel process. The change of hydrolysis and condensation rates as a function of pH can be followed in Fig. 2.13. At pH lower than around 5 the hydrolysis rate is faster than condensation, in accordance with our expectations; the hydrolysis rate

64

P. Innocenzi

decreases also with the increase of pH and reaches a minimum around 7. After this value it increases quite quickly with the alkalinity of the solution. The condensation rate, on the other hand, follows a similar trend and decreases with the increase of pH even if there is a lower reaction rate with respect to hydrolysis up to the value of around 5. After this pH value the condensation rate quickly rises up to around 10, and then decreases again. Why is this trend observed? Because we should always keep in mind that cleavage of silica bonds at higher pH values is quickly rising and condensation and hydrolysis are in competition. Condensation and hydrolysis reactions are also depending on the size of the alkoxy group [48]; the reactivity of silicon alkoxides in fact decreases with the increase of the size of the alkoxy because of the steric hindrance. The reactivity of silicon alkoxides decreases when the size of the alkoxy group increases because of steric hindrance factors. The reaction rate order of silicon alkoxides with different alkoxy groups follows this order (Eq. 2.19): n

SiðOMeÞ4 > SiðOEtÞ4 > Si O PR i

> Si O PR

4

n

4

> SiðOHexÞ4

> Si O Bu

4

ð2:19Þ

To resume in acid conditions linear or weakly branched silica species are preferentially formed; they aggregate through entanglements which eventually cause gelation of the system. In basic conditions gelation occurs via formation of agglomerated silica clusters which condense to form a 3D network. The acid conditions give rise, therefore, to a final material which is denser with respect to the basic route which forms a material with a more porous network because of the free space between the particles. A dense silica material is obtained only after firing at high temperatures but in the gel or xerogel state the structural differences between acid and silica gels are still important.

2.10

The Gel Structure

The extent of hydrolysis and condensation reactions in the case of silicon alkoxides are essentially governed by the pH of the sol. The choice to use an acid or a base catalyst has also a direct effect on the structure of the final gel and in general an acidic route gives a more compact structure and a basic one a more open and porous (Fig. 2.14) [49]. This difference in the structure is due to the change in reactivity of the species which form upon the beginning of hydrolysis and condensation. In general, the acid-catalyzed silicon alkoxides have a higher reactivity with respect to the hydrolyzed or condensed species which favors the formation

of branched structures. In basic sols the reactivity shows a reverse trend and this gives rise to the formation of silica clusters and “spherical” particles. The gel forms, therefore, through the growth of a branching macromolecule (acid conditions) or aggregation of silica clusters. In the last case a porous and less interconnected gel structure would form. Another peculiarity of sol–gel reactions of silica is the formation of many different species from monomers to trimers and larger aggregates upon hydrolysis. Not only linear or branched macromolecules would form; cyclic species typically from threefold to sixfold rings are also commonly observed. The presence of cyclic molecules would also affect the gelation process because they could behave as local thermodynamic sinks reducing the condensation and growth process.

2.11

Modeling the Sol to Gel Transition

To describe the sol to gel transformation three main models have been used: the classic theory of a branching molecule, the percolation theory, and the fractal model. The classic statistical theory (or mean field theory) is based on three main assumptions: the reactivity of all the monomers remains the same during the polycondensation, the formation of cyclic species is not allowed, and the steric hindrance effects are negligible. These assumptions do not work well for inorganic gelling systems because from the very beginning of the process they would form species which have a different reactivity while cyclization is also a common phenomenon especially in silica systems [50]. In general the models fail to be quantitative predictive but are helpful to reach a general understanding of the process. The classic branching model predicts that the system would gel when around one-third of the available bonds have been formed. This value is far to fit with the experimental data because, as we have seen, the assumptions are too restrictive to be applied to the chemistry of sol–gel precursors. The classic theory, however, allows obtaining a general prediction of the evolution of the weight fraction of the different reactive species (Fig. 2.4) [51]. Figure 2.15 shows the change of reaction degree, p, as a function of the weight fraction, wx, of the aggregates formed by x monomeric units (y axis, left) and of the weight fraction of the gel (y axis, right). At the gel point, pc, the reaction degree according to the classic model is 0.33. At the beginning of the reaction most of the reactive species would be the monomers that quickly reduce to form dimers and larger aggregates. As soon as condensation proceeds the weight fraction of the aggregates (x > 2) would also decrease as soon as they become part of the large spamming cluster

2

Overview of the Sol–Gel Process

a

65

2

Acid catalyzed

Gel

Gelation Sol Acid catalyzed hydrolysis

b

Base catalyzed

Gel

Gelation

Sol

Base catalyzed hydrolysis Fig. 2.14 Formation of a gel from an acid (a) or base (b) catalyzed hydrolysis, the change of inorganic structure from a sol to a gel

Fig. 2.15 Weight fraction (wx) of aggregates formed by x monomeric units in a sol of tetrafunctional molecules as a function of reaction degree, p. wg is the weight fraction of the gel. (Redrawn from Ref. [51])

0.40

1.0

0.35

0.9 0.8

x=1

0.30 0.25

0.7 x=2

0.6 W 0.5 g 0.4

Wx 0.20 0.15 x=4

0.10 0.05 0.00 0.0

x = 10 pc 0.1

0.2

0.3

0.4

0.5 p

0.6

0.7

0.8

0.9

1.0

0.3 0.2 0.1 0.0

66

P. Innocenzi

500

100

% Si

M

P

400

50

0

0.5

t / tgel

1.0

0.8

P( p)

300

0.6

200

0.4

P

s, l

D, T 0

1.0

sav(p)

100

1.5

0

0.2

Fig. 2.16 Time evolution (t/tgel) of monomeric (M), dimeric (D), trimeric (T), and higher branched (P) silicon groups. Data obtained by NMR and Raman analysis from a TMOS aging sol

0.4

0.6

0.8

1.0

p (bond fraction) Fig. 2.17 Cluster size, s, spanning length, l, percolation probability, P, as a function of the fraction of the bonds which are formed in the system, p. (Redrawn from Ref. [54])

3.0 2.5 Logη (cP)

which eventually would form the gel. At the gel point not all the molecules have reacted and the liquid phase would be composed by a sol with aggregates of different dimensions. The theoretical data can be compared with the experimental results which show the change of the gelation time (t/tgel) as a function of the content of monomers (M), dimers (D), trimers (T), and highly branched species. The data have been obtained by NMR and Raman spectroscopy analysis. The experimental data qualitatively follow quite well the theoretical previsions as shown in Fig. 2.16. Dimers and trimers increase at expenses of monomers and around the gelation times they decrease to form higher branched, P, species. The similarities with the theoretical model in Fig. 2.15 are very striking. Percolation theory [52] has been also applied to modeling the sol–gel transition in inorganic sol–gel systems [53]. In the percolation model the gel point is reached when a growing cluster is able to span over the whole sample region. The theory has no analytical solution and the gel point can be defined only on a statistical base. Without going too much in detail, we can use the theory to follow the change of the bond fraction, P, as a function of the size of the spanning cluster, s, and spanning length defined as the maximum distance between any bond center in a cluster (Fig. 2.17). The percolation probability, P(p), is the key function of the process and beyond the percolation threshold at pc is correlated with the growth in volume of the network with the increase of the bond fraction. The average values of s and l, sav and lav, show singularities close to the percolation threshold. This model fits quite well with gelling silica systems which are characterized by a divergence of the viscosity close to the gel point (Fig. 2.18) [54]. During gelation the crosslinking reactions produce a constant increase of viscosity and the molecules gradually lose their mobility. In the

0.2

pc

lav

2.0 1.5 1.0 0.5 0.0 0.0

0.3

0.6

0.9

1.2

t / tgel

Fig. 2.18 Viscosity, η, of TMOS sols as a function of relative gelation time t/tgel

proximity of the gel point the system enters in a viscoelastic regime and shows and elastic response to a shear stress. The failure of the classic model of a growing branching system when applied to silica systems is basically due to the restrictive assumptions which do not take into account the specificity of the sol–gel chemistry. The minimum number of bonds necessary to observe gelation predicted from the theory results quite far from the experimental values. If we follow the transition of a TEOS sol to a gel, a much larger fraction of bonds forms at the gel point and around 83% of the available bonds are converted at longer reaction times (Fig. 2.19) [55]. The model of linear chains which randomly grow is not suitable, therefore, to describe the complexity of the

2

Overview of the Sol–Gel Process

67

1.0

2 1500

0.6 tg (min)

Conversion

0.8

0.4 0.2

500

0 0

1000

2000 3000 Time (h)

4000

Fig. 2.19 Hydrolysis and condensation of TEOS: later in the reaction, conversions all slow down and converge to 83%. Gel times: (red line) 5064 h; (blue line) 4392 h; (light blue line) ~1700 h; (yellow line) ~1500 h; (black line) 1020 h; (orange line) 1200 h. (Redrawn from Ref. [55])

0.6

0.6 0.5

Chains

0.4

0.4

Dimers

0.3

0.3

0.2 0.1 0.0

0.2

Rings Branchings

Monomers

0

10

20 Time (h)

30

Mol fraction

0.5 Mol fraction

1000

0.1 0.0 40

Fig. 2.20 Evolution of mole fractions of monomer and different types of oligomers in the R-0.7 sample (derived from Q0, Q1, and Q2 signals in 29 Si NMR) and fraction of branching expressed as the fraction of Q3 in totality of silicon atoms. Qi represents the fraction of silicon sites with different siloxane bonds (i) connected to other silicon atoms

composition of a sol and its transition to a gel. It is necessary to take into account the formation of different species with different reaction rates and the effect due to cyclization. NMR analysis has confirmed the complex interactions of the different species which form in the sol [56]. Monomers and dimers transform with the progress of the hydrolysis and condensation into linear chains and branching structures or give rise to cyclic species (Fig. 2.20). This works as a general rule, however, even small changes in the synthesis conditions can produce a totally different sol and gel structures.

0 0.08

0.10

0.12 ϕ

0.14

Fig. 2.21 Experimental dependence of the gel time tg with the tetramethoxysilane (TMOS) volumic fraction Φ for different container sizes. Circles, squares, and diamonds correspond to container diameters of 10, 14, and 32 mm. The gels were prepared by hydrolysis of TMOS using a 0.05N ammonia-water solution. (Reprinted with permission from Ref. [58])

Using Dynamic Monte Carlo simulations, which consider the effect of the nearest-neighbors and cyclization of silica oligomers in an acid catalyzed sol, the calculated molecular weight distribution has been found to be in good agreement with the experimental values [57]. Extensive cyclization has to be taken into account to predict the distribution of species which give rise to a silica gel. The control of experimental conditions is extremely critical and little differences in the synthesis can affect the sol to gel transition and the gel time. An interesting example is the effect of the size of the sol container which has been found to affect the gelation time [58]. The experimental values of the gelation time are longer with the increase of the container size. This effect has been experimentally observed only in base-catalyzed sols while the gel time remains size independent in neutral conditions when no catalysts are employed in the synthesis. Dependence on the container size has been also found in the case of acid-catalyzed sols [59] (Fig. 2.21).

2.12

Conclusions

In this brief overview the main features of a sol to gel transition in inorganic systems have been outlined. Silica has been taken as the main example because its sol–gel chemistry is well known and can be used to outline the main stages of the process. The sol to gel transition is a

68

continuous process and no thermodynamic variables can be used to identify the formation of a gel as difficult to measure and to define. It is a stochastic process which is randomly governed by the chemistry of the precursor sol. An exact prevision and measure of gelification, which means the exact moment when a growing macromolecule forms an interconnected continuous network spanning within the container, is difficult to obtain. In general, the models applied to sol to gel transitions in inorganic systems fail to give an accurate agreement with the experimental data. The divergence of some properties, such as the viscosity, close to the gel point well supports the general understanding of the process. The measure and definition of the sol–gel transition is still far from being rigorous and, especially in fast evaporating systems, such as thin films, is still elusive to measure.

References 1. Innocenzi, P.: Introduction to Sol to Gel Chemistry. II Edition. Springer Briefs in Materials. Springer (2019). This article has been written reorganizing some parts of the text from this source 2. Yunqi, L., Tongfei, S., Zhaoyan, S., Lijia, A., Qingrong, H.: Investigation of sol-gel transition in pluronic F127/D2O solutions using a combination of small-angle neutron scattering and Monte Carlo simulation. J. Phys. Chem. B. 110, 26424–26429 (2006) 3. PAC, 1972, 31, 577: Manual of symbols and terminology for physicochemical quantities and units, appendix ii: definitions, terminology and symbols in colloid and surface chemistry, p. 606. (1972) 4. PAC, 1972, 31, 577: Manual of symbols and terminology for physicochemical quantities and units, appendix ii: definitions, terminology and symbols in colloid and surface chemistry, p. 605. (1972) 5. PAC, 2007, 79, 1801: Definitions of terms relating to the structure and processing of sols, gels, networks, and inorganic-organic hybrid materials (IUPAC Recommendations, 2007). p. 1806 6. Almdal, K., Dyre, J., Hvidt, S., Kramer, O.: Towards a phenomenological definition of the term ‘Gel’. Polym. Gels Netw. 1, 5– 17 (1993) 7. Henish, H.K.: Crystal Growth in Gels. The Penn State University Press, University Park (1970) 8. Encyclopædia Britannica.: http://www.britannica.com/science/gel 9. James, P.F.: The gel to glass transition: chemical and microstructural evolution. J. Non-Cryst. Solids. 100, 93–114 (1988) 10. PAC, 2007, 79, 1801: Definitions of terms relating to the structure and processing of sols, gels, networks, and inorganic-organic hybrid materials (IUPAC Recommendations 2007). p. 1825 11. PAC, 2007, 79, 1801: Definitions of terms relating to the structure and processing of sols, gels, networks, and inorganic-organic hybrid materials (IUPAC Recommendations 2007). p. 1809 12. Grant, M.C., Russel, W.B.: Volume-fraction dependence of elastic moduli and transition temperatures for colloidal silica gels. Phys. Rev. E. 47, 2606–2614 (1993) 13. Orgaz, F., Rawson, H.: Characterization of various stages of the sol-gel process. J. Non-Cryst. Solids. 82, 57–68 (1986) 14. Brinker, J., Scherer, G.: Sol-Gel Science. Academic (1990) 15. Turova, N.Y., Turevskaya, E.P., Kessler, V.G., Yanovskaya, M.I.: The Chemistry of Metal Alkoxides. Kluwer AP, Dordrecht (2002) 16. Dong, H., Lee, M., Thomas, R.D., Zhang, Z., Reidy, R.F., Mueller, D.W.: Methyltrimethoxysilane sol-gel polymerization in acidic ethanol solutions studied by 29Si NMR spectroscopy. J. Sol-Gel Sci. Technol. 28, 5–14 (2003)

P. Innocenzi 17. Sanchez, M., Boissiere, C., Cassaignon, C., Chaneac, C., Durupthy, O., Faustini, M., Grosso, D., Laberty-Robert, C., Nicole, L., Portehault, D., Ribot, F., Rozes, L., Sassoye, C.: Molecular engineering of functional inorganic and hybrid materials. Chem. Mater. 26, 221–238 (2014) 18. Nicole, L., Laberty-Robert, C., Rozes, L., Sanchez, C.: Hybrid materials science: a promised land for the integrative design of multifunctional materials. Nanoscale. 6, 6267–6292 (2014) 19. Beija, M., Alfonso, C.A.M., Martinho, J.M.G.: Synthesis and applications of rhodamine derivatives as fluorescent probes. Chem. Soc. Rev. 38, 2410–2433 (2009) 20. Avnir, D., Levy, D., Reisfeld, R.: The nature of silica cage as reflected by spectral changes and enhanced photostability of trapped rhodamine 6G. J. Phys. Chem. 88, 5968–5958 (1994) 21. Severin-Vantilt, M.M.E., Oomen, E.W.J.L.: The incorporation of Rhodamine B in silica sol-gel layers. J. Non-Cryst. Solids. 159, 38–48 (1993) 22. Lee, M.H., Lee, S.J., Jung, J.H., Lim, H., Kim, J.S.: Luminophoreimmobilized mesoporous silica for selective Hg2+ sensing. Tetrahedron. 63, 12087–12092 (2007) 23. Jackson, C.L., Bauer, B.J., Nakatani, A.I., Barnes, J.D.: Synthesis of hybrid organicinorganic materials from interpenetrating polymer network chemistry. Chem. Mater. 8, 727–733 (1996) 24. Schottner, G.: Hybrid solgel-derived polymers: applications of multifunctional materials. Chem. Mater. 13, 3422–3435 (2001) 25. Cordes, D.B., Lickiss, P.B., Rataboul, F.: Recent developments in the chemistry of cubic polyhedral Oligosilsesquioxanes. Chem. Rev. 110, 2081–2173 (2010) 26. Alam, T.A., Assink, R.A., Loy, D.A.: Hydrolysis and esterification in organically modified alkoxysilanes: a 29Si NMR investigation of methyltrimethoxysilane. Chem. Mater. 8, 2366–2374 (1996) 27. Rankin, S.E., Macosko, C.W., McCormick, A.V.: Solgel polycondensation kinetic modeling: methylethoxysilanes. AICHE J. 1998(44), 1141 (1998) 28. Innocenzi, P., Kidchob, T., Yoko, T.: Hybrid organic-inorganic sol-gel materials based on epoxy-amine systems. J. Sol-Gel Sci. Technol. 35, 225–235 (2005) 29. Fyfe, C.A., Aroca, P.P.: A kinetic analysis of the initial stages of the sol-gel reactions of methyltriethoxysilane (MTES) and a mixed MTES/Tetraethoxysilane system by high-resolution 29Si NMR spectroscopy. J. Phys. Chem. B. 101, 9504–9509 (1997) 30. Innocenzi, P., Brusatin, G., Guglielmi, M., Bertani, R.: New synthetic route to 3-Glycidoxypropyl trimethoxysilane-based hybrid organic-inorganic materials. Chem. Mater. 11, 1672–1680 (1999) 31. Innocenzi, P., Sassi, A., Brusatin, G., Guglielmi, M., Favretto, D., Bertani, R., Venzo, A., Babonneau, F.: A Novel Synthesis of Sol-Gel Hybrid Materials by a Nonhydrolytic/Hydrolytic Reaction of (3Glycidoxypropyl)trimethoxysilane with TiCl4. Chem. Mater. 13, 3635–3643 (2001) 32. Ayandele, E., Sarkar, B., Alexandridis, P.: Polyhedral oligomeric silsesquioxane(POSS)-containing polymer nanocomposites. Nano. 2, 445–475 (2012) 33. Shea, K.J., Loy, D.A.: A mechanistic investigation of gelation. The solgel polymerization of precursors to bridged polysilsesquioxanes. Acc. Chem. Res. 34, 707–716 (2001) 34. Cerveau, G., Corriu, R.J.P.: Some recent developments of polysilsesquioxane chemistry for material science. Coord. Chem. Rev. 1051, 178–180 (1998) 35. Mehdi, A.: Self-assembly of layered functionalized hybrid materials. A good opportunity for extractive chemistry. J. Mater. Chem. 20, 9281–9286 (2010) 36. Wang, Z., Wang, D., Qia, Z., Guo, J., Deng, H., Zhao, N., Xu, J.: Robust superhydrophobic bridged silsesquioxane aerogels with tunable performances and their applications. ACS Appl. Mater. Interfaces. 7, 2016–2024 (2015)

2

Overview of the Sol–Gel Process

37. Livage, J.: Sol-Gel Synthesis of Transition Metal Oxopolymers. Frontiers of Polymers and Advanced Materials, pp. 659–667. Springer (1994) 38. Livage, J., Henry, M., Sanchez, C.: Sol-gel chemistry of transition metal oxides. Prog. Solid St. Chem. 18, 250–341 (1988) 39. Mehrotra, R.G., Singh, A.: Chemistry of Oxo-alkoxides of metals. Chem. Soc. Rev. 25, 1–13 (1996) 40. Debecker, D.P., Mutin, P.H.: Non-hydrolytic sol-gel routes to heterogeneous catalysts. Chem. Soc. Rev. 41, 3624–3650 (2012) 41. Styskalik, A., Skoda, D., Barnes, C.E., Pinkas, J.: The power of non-hydrolytic sol-gel route: a review. Catalyst. 7, 168 (2017) 42. Rozes, L., Sanchez, C.: Titanium oxo-clusters: precursors for a Lego-like construction of nanostructured hybrid materials. Chem. Soc. Rev. 40, 1006–1030 (2011) 43. Rozes, L., Steunou, N., Fornasieri, G., Sanchez, C.: Titanium-oxo clusters, versatile nanobuilding blocks for the design of advanced hybrid materials. Monatshefte Chem. 137, 501–528 (2006) 44. Cargnello, M., Gordon, T.R., Murray, C.B.: Solution-phase synthesis of titanium dioxide nanoparticles and nanocrystals. Chem. Rev. 114, 9319–9345 (2014) 45. Pope, E.J.A., Mackenzie, J.D.: Sol-gel processing of silica. II. The role of the catalyst. J. Non-Cryst. Solid. 87, 185–198 (1986) 46. Klein, L.C.: Sol-gel processing of silicates. Ann. Rev. Mater. Sci. 15, 227–248 (1985) 47. Brinker, C.J.: Hydrolysis and condensation of silicates: effects on structure. J. Non-Cryst. Solids. 100, 31–50 (1988) 48. Hook, R.: A 29Si NMR study of the sol-gel polymerisation rates of substituted ethoxysilanes. J. Non-Cryst. Solids. 195, 1–15 (1996) 49. Bailey, J.K., Nagase, T., Broberg, S.M., Mecartney, M.L.: Microstructural evolution and rheological behavior during gelation of ceramic sols. J. Non-Cryst. Solids. 109, 198–210 (1989) 50. Bailey, J.K., Macosko, C.W., Mecartney, M.L.: Modeling the gelation of silicon alkoxides. J. Non-Cryst. Solids. 125, 208–223 (1990) 51. Flory, P.J.: Principles of Polymer Chemistry. Cornell University Press, New York (1953) 52. Stauffer, D., Aharony, A.: Introduction to Percolation Theory. Taylor & Francis, London (1992)

69 53. Cohen-Addad, J.P.: Sol or gel-like behaviour of ideal silica-siloxane mixtures: percolation approach. Polymer. 33, 2762 (1992) 54. Zallen, R.: The Physics of Amorphous Solids. Wiley (1998) 55. Ng, L.V., Thompson, P., Sanchez, J., Macosko, C.W., McCormick, A.V.: Formation of Cagelike intermediates from nonrandom cyclization during acid-catalyzed sol-gel polymerization of tetraethyl Orthosilicate. Macromolecules. 28, 6471–6476 (1995) 56. Depla, A., Lesthaeghe, D., van Erp, T.S., Aerts, A., Houthoofd, K., Fan, F., Li, C., Van Speybroeck, V., Waroquier, M., Kirschhock, C. E.A., Martens, J.A.: 29Si NMR and UV-Raman investigation of initial oligomerization reaction pathways in acid-catalyzed silica sol-gel chemistry. J. Phys. Chem. C. 115, 3562–3571 (2011) 57. Sefcik, J., Rankin, S.E.: Monte Carlo simulations of size and structure of gel precursors in silica polycondensation. J. Phys. Chem. B. 107, 52–60 (2003) 58. Anglaret, E., Hasmy, A., Jullien, R.: Effect of container size on gelation time: experiments and simulations. Phys. Rev. Lett. 75, 4049 (1995) 59. Huber, C.J., Butler, R.L., Massari, A.M.: Evolution of ultrafast vibrational dynamics during sol-gel aging. J. Phys. Chem. C. 121, 2933–2939 (2017)

Plinio Innocenzi is a full professor of Materials Science at the University of Sassari and director of the Laboratory of Materials Science and Nanotechnology. He has served as Science Counsellor at the Embassy of Italy to China from 2010 to 2018. His research is focused on sol-gel chemistry, hybrid materials, and self-assembly and he has authored more than 220 scientific articles on this subject.

2

3

Gel-Phase Processing and Solvent Exchange Justin S. Griffin, Ryan T. Nelson, Pavel Gurikov Stephen A. Steiner III

, Irina Smirnova, and

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

3.1

3.2 Molding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.1 General Molding for the Laboratory . . . . . . . . . . . . . . . . . . . . . . . . 3.2.2 Molding of Complex Shapes and Parts with Large Dimensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.3 Demolding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

73 74 75 76

3.3

Aging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

3.4 3.4.1 3.4.2 3.4.3 3.4.4

Solvent Exchange . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Modelling of Solvent Exchange . . . . . . . . . . . . . . . . . . . . . . . . . . . . Effects of Solvent Exchange on Gel Properties . . . . . . . . . . . . Practical Aspects of Solvent Exchange . . . . . . . . . . . . . . . . . . . . . Additional Safety Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.5

Liquid-Phase Functionalization . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

3.6

Solvent Exchange into Liquid Carbon Dioxide . . . . . . . . . . 89

3.7

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

77 77 79 81 88

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

molding. Furthermore, after a gel has been formed, additional processing steps are typically required before its pore fluid can be removed to produce an aerogel. Such steps include aging of the gel to strengthen its solid-phase backbone, diffusion-mediated replacement of its pore fluid with a target solvent to compatibilize the gel for drying, and optional liquid-phase chemical treatment(s), for example, to impart hydrophobicity, introduce a desired functionality, or increase strength. This chapter discusses important unit operations including molding and demolding, aging, and solvent exchange, as well as examples of liquid-phase chemical treatments used to prepare various functional aerogels. Keywords

Solvent exchange · Gel-phase processing · Aging · Functionalization · Polymer crosslinking · Molding · Demolding

Abstract

Several important unit operations involved in aerogel production pertain to the formation and processing of gels. Formation of a gel typically occurs in a mold, and accordingly, important considerations should be given to gel J. S. Griffin (*) · R. T. Nelson (*) · S. A. Steiner III (*) Aerogel Technologies, LLC, Boston, MA, USA e-mail: jsgriffi[email protected]; [email protected]; [email protected] P. Gurikov Laboratory for Development and Modelling of Novel Nanoporous Materials, Hamburg University of Technology, Hamburg, Germany Institute of Thermal Separation Processes, Hamburg University of Technology, Osnabrueck, Germany e-mail: [email protected] I. Smirnova Institute of Thermal Separation Processes, Hamburg University of Technology, Hamburg, Germany e-mail: [email protected]

3.1

Introduction

In a typical process for making an aerogel, a sol–gel-derived gel serves as the precursor that will eventually become the final aerogel. The aerogel is then produced by removing the gel’s pore fluid in a way that enables isolation of the gel’s solid skeleton from its liquid component without resulting in significant loss of volume of the solid skeleton. However, before the pore fluid of a gel can be removed to produce an aerogel, a number of processing steps must typically be performed in order to prepare and compatibilize the gel for the particular drying process to be used. A gel that will be used to make an aerogel but that has not yet been dried is often times referred to as a wet gel, regardless of whether or not it contains water as a component of its pore fluid. Accordingly, herein, the terms “wet gel,” “wet state,” and “wet phase” shall be used to refer to states linked to processing

© Springer Nature Switzerland AG 2023 M. A. Aegerter et al. (eds.), Springer Handbook of Aerogels, Springer Handbooks, https://doi.org/10.1007/978-3-030-27322-4_3

71

72

steps that are done with a gel before it is dried, in contrast to so-called postprocessing steps, which are performed on an aerogel after its pore fluid has been removed and which are discussed in ▶ Chap. 6. Thus the initial step in producing most aerogels is the synthesis of a gel. While the science of the sol–gel process underlying gel synthesis is described in the preceding chapter, there are several important practical aspects of formation of a gel such as molding and, when making aerogels, demolding, that should be considered. After a gel has been initially formed, that is, after gelation of its precursory sol has occurred, the gel’s solid-phase backbone may in fact not yet be finished forming, as reactions may still be occurring in the backbone even though the gel has set. Accordingly, a gel is typically allowed to stand, or age, for a period of time before subsequent processing steps are employed. Gel molding and aging, along with important considerations thereof, are discussed below. Second, after a gel has been sufficiently aged, its pore fluid must be compatibilized for the drying process that will be used to remove said pore fluid from the gel. In his foundational studies on aerogels, Kistler synthesized silica gels from waterglass. Accordingly, his gels contained water as the primary component of their pore fluid. Kistler initially tried to directly supercritically extract this water from the gels but in doing so discovered that supercritical water, which is extremely corrosive, readily disintegrated (peptized) the gels’ silica backbones. To circumvent this, prior to drying Kistler replaced the pore fluid of his silica gels with ethanol, which could then be supercritically extracted from the gels without dissolving their backbones. To do so, Kistler employed a process that is now commonly referred to as solvent exchange. Solvent exchange, also referred to as a pore fluid exchange, washing, or rinsing, is the displacement of the pore fluid of a gel by another solvent. This is typically done by placing the gel in a bath of the solvent that is desired to be the gel’s pore fluid and allowing the gel to soak in said bath for a given period of time. Since gels typically exhibit an open-celled pore network, the solvent held within the gel will spontaneously and readily diffuse into and mix with the surrounding target solvent in the bath until equilibrium has been reached. The resulting concentration of target solvent in the gel can then be determined by rule of mixtures, provided the gel–bath system has reached equilibrium. By replacing the bath solvent with fresh solvent periodically, the concentration of target solvent in the gel can be successively increased until the gel’s pore network contains a suitably high purity of the target solvent. Solvent exchanges are frequently necessary in order to ensure that a gel can be dried in a way that does not result in collapse of its solid skeleton during drying. Typically, this means exchanging the gel into a specific solvent that can be supercritically extracted, evaporated, or sublimed without causing the gel to shrink. In general, a gel can be solvent

J. S. Griffin et al.

exchanged into any solvent with which its pore fluid is miscible. Accordingly, one must consider both the nature of the gel’s initial pore fluid immediately after gelation, i.e., the gel’s mother liquor, as well as the target solvent required for drying in order to determine if a gel can be directly solvent exchanged into the target solvent or whether a solvent exchange into an intermediate transfer solvent i.e., a solvent that is miscible with both the gel’s mother liquor and the target solvent, is first required. More about supercritical drying can be found in ▶ Chap. 4, ▶ Chap. 65, Sect. 3. If an aerogel is to be dried via supercritical drying, its pore fluid must itself either be the solvent that will be made supercritical or be extractable by whatever supercritical fluid is to be used. In the case of direct supercritical extraction, gels are often solvent exchanged into methanol, ethanol, acetone, or acetonitrile, all of which can be removed from the gel via high-temperature supercritical drying. Since organic solvents are typically flammable and potentially explosive near their critical points, supercritical drying is instead frequently performed using a safer, nonflammable solvent, namely, carbon dioxide, in a process often referred to as the Hunt process, low-temperature supercritical drying, or supercritical CO2 drying. As a result, many solvent exchange procedures are designed to transition a gel’s pore fluid from its mother liquor into liquid carbon dioxide or, in cases where aerogel monolithicity is less important, into a solvent that can be removed by carbon dioxide in its supercritical state. One important note is that because carbon dioxide does not exist as a liquid at atmospheric pressure, solvent exchange into liquid carbon dioxide must be performed in a pressure vessel. This is not additionally burdensome if the carbon dioxide is to be supercritically extracted, since the supercritical drying step must also be performed in a pressure vessel. Since supercritical drying from carbon dioxide is currently the most commonly employed method for producing aerogels, procedures in the literature for making aerogels ranging from inorganic oxides to synthetic polymers to biopolymers frequently include a solvent exchange step into an intermediate carbon-dioxide-miscible solvent prior to drying. Common intermediate solvents include methanol, ethanol, acetone, and, occasionally, acetonitrile or amyl acetate for certain sensitive gel systems. For example, in a typical synthesis of a sol–gel-derived silica gel prepared from an alkoxide precursor, the gel will initially contain a mixture of water, alcohol, ammonium hydroxide, and unreacted alkoxide in its pore network immediately after gelation. If supercritical drying from carbon dioxide were attempted on such a gel, water in the gel’s pore network would remain in the resulting solid after the drying process is completed, causing undesireable effects such as opacity, collapse of the porous solid network, and/or cracking. This is because water and carbon dioxide are not miscible and, as a result, the water is not extracted well. By first exchanging the gel’s mother liquor for a solvent that is miscible with carbon dioxide

3

Gel-Phase Processing and Solvent Exchange

such as methanol, however, the gel’s pore fluid can instead be readily extracted by carbon dioxide to give a high-quality, high-transparency, low-density aerogel. In another example that highlights the need for solvent exchange, sodiumsilicate-derived gels often contain high concentrations of sodium salts resulting from the addition of acid used to invoke gelation. Such salts are typically considered undesirable in the final aerogel, as they result in low-transparency, highly hydrophilic aerogels with reduced mesoporosity and are corrosive to things like pipes at high temperatures. As a result, such gels are often soaked in successive water baths, that is, solvent exchanged into water, to dissolve out these salts prior to drying. The water-exchanged gels are then exchanged into another solvent that is miscible with carbon dioxide prior to drying. (Note that while solvent exchanging sodium-silicatederived gels into water, a process first demonstrated by Kistler, illustrates a circumstance where solvent exchange is called for, substantially more efficient processes for making silica aerogels from sodium silicate have since been developed; see ▶ Chaps. 15, ▶ 16, and ▶ 65, Sect. 2.) The need for solvent exchange is not limited to supercritically dried aerogels, however. In the case of ambientpressure drying as employed in the production of hydrophobic silica aerogels via the springback method, solvent exchange of a gel into a dual-function low-surface-tension solvent and hydrophobe, namely, hexamethyldisiloxane, is often employed. This is done prior to evaporative drying of the gel to both minimize capillary stresses that arise when solvent is evaporated from the pores of the gel and to prevent cross-condensation of surface silanols that would otherwise tie the gel network together as it undergoes transient shrinkage. In the case of freeze drying, gels are typically exchanged into a solvent that can be readily sublimed under vacuum conditions, for example, water, tert-butanol, or camphor. As such, solvent exchange is an important unit operation for a wide variety of aerogel synthesis procedures. Important considerations must be made when solvent exchanging a gel. Interactions between the solid-phase backbone of a gel and target solvent can cause the gel to denature, that is, undergo physicochemical changes that result in severe geometric distortions, as is the case when a gelatin gel is exchanged from water into ethanol, for example. Certain combinations of solvents may also result in swelling or contraction of a gel as its pore fluid is exchanged from one solvent to another due to entropy of mixing effects (also known as excess volume phenomena), for example, when a resorcinol-formaldehyde polymer gel is exchanged from water into dimethylformamide. Certain solvents may also pose chemical compatibility problems for containers, o-rings, hoses, pumps, and vessels, including and especially for supercritical dryers, which involve expensive highpressure equipment. Viton o-rings, for example, are swollen

73

by many organic solvents, while plastics such as acrylic and polystyrene are easily embrittled or dissolved. Recyclability, solvent cost, and cost of disposal may also need to be considered, since solvent exchanges often consume large volumes of solvents. Furthermore, parameters such as soaking time, number of baths, and volume of solvent per bath heavily influence process time, total volume of solvent required, cost, and material quality. Accordingly, selection of target solvent and the way a solvent exchange process is executed should be considered carefully. The physics underlying solvent exchange of gels and practical considerations thereof are discussed in more depth below. Finally, in addition to aging and solvent exchange, certain processing steps such as liquid-phase functionalization involve chemical reactions with surface functional groups on a gel’s solid backbone and may need to be performed before a gel undergoes drying, for example, due to a need for a solvent system in order to perform such reactions or a need to infiltrate a reagent throughout the gel’s pore network. Examples include replacing surface silanols of a silica gel with trimethylsiloxy groups via reaction with hexamethyldisilazane, electroless deposition of an electrochemically active oxide layer throughout a carbon aerogel, and infiltration of reactive polyisocyanates into the pore network a metal oxide gel to form a polymer-crosslinked oxide hybrid gel. Such steps are performed via solvent exchange of a gel into a solvent solution containing the reactive reagents required to introduce the desired functionality, which are then typically followed by an additional solvent exchange into a pure solvent to wash out residual unreacted reagents. Such wet-phase functionalization of gels is described in more depth below.

3.2

Molding

Molding refers to the step of imparting a shape unto a gel by allowing a liquid-phase sol to undergo gelation in a container of the desired shape. Molding is an especially important consideration for the production of monolithic parts, but even if the end goal is to produce particles or just a volume of material for scientific analysis, the gel is typically still initially formed in some sort of mold. In general, a gel is molded by pouring a mixture of the reactive reagents, solvents, and/or catalysts required to make the gel (i.e., its precursory sol) into a container (i.e., the mold) and allowing the gel to set. Before deploying a molding strategy, mold materials should be tested for chemical compatibility with the solvents and other chemicals to be used for gel synthesis. Depending on the specifics of the gel formulation being used, common options for molding materials with generally good chemical compatibility include high-density polyethylene (HDPE), polypropylene, polytetrafluoroethylene (PTFE),

3

74

silicone rubbers, stainless steel, aluminum, and glass. While glass can be a suitable mold material for many gel compositions, inorganic oxide gels such as silica are prone to sticking to it. Acrylic, polycarbonate, and polystyrene are generally poor mold materials due to chemical incompatibility with common organic solvents and liquid/supercritical carbon dioxide. Smooth mold surfaces are generally desirable and aid in eventual removal of the gel part from the mold, or demolding. For fragile gel formulations, sharp corners and high-aspectratio inclusions and features may need to be avoided. For such formulations, additional shape control that cannot be achieved directly through molding can be accomplished through post-drying machining steps, discussed in ▶ Chap. 6, or, in some cases, through machining of the wet gel itself. Certain processes, such as high-temperature supercritical drying (i.e., direct extraction of organic solvent from a gel), integrate molding and supercritical extraction into a single process and have additional molding requirements specific to these processes; see ▶ Chap. 65 Sect. 3 for more information. Because gels are primarily composed of liquid by volume, they generate a vapor pressure of the solvent contained within their pores. Vapors generated by gels should be contained not only for safety reasons but also to prevent solvent loss during gelation and to prevent exposed surfaces of the resulting gels from drying out and cracking during aging later, discussed in depth in a following section. Containment of vapors can be achieved by placing gels in a sealed container or by sealing the open face of a mold with a lid or Parafilm. Accordingly, for gels that contain a volatile solvent such as acetone as a component of their pore fluid, it is beneficial to place the gels in a sealed container containing an atmosphere saturated with that solvent’s vapor during gelation and aging (e.g., by adding a small amount of excess solvent identical in composition to the pore fluid of the gels to the bottom of the sealed container). In addition, the temperature during gelation and aging can influence the materials properties of the resulting gels and thus the resulting aerogels. Accordingly, process temperature should be considered in developing a molding strategy. For example, if tolerated by the sol–gel chemistry being used, lowering the process temperature can reduce the rate of solvent evaporation during gelation and aging, thereby yielding a better product. In other configurations, an elevated process temperature may be needed in order for gelation to occur, for example, as is the case of resorcinol-formaldehyde polymer gels. In such cases, it is important to select mold materials (paying particular attention to containers and gaskets) that will not soften, melt, or lose chemical compatibility at the required process temperatures while still maintaining an airtight seal as to not allow solvent to escape when heated.

J. S. Griffin et al.

General safety considerations related to temperature control, vapor containment, and chemical compatibility are discussed at greater length later in the context of solvent exchange.

3.2.1

General Molding for the Laboratory

Plastic syringes serve as the de-facto standard for molding of experimental gel formulations in many laboratories. Syringes are well suited for many formulations but in some cases may contain an elastomeric plunger tip that can become problematic when contacted by certain chemicals, for example, tetramethoxysilane (TMOS) and tetraethoxysilane (TEOS), which tend to swell many elastomers. In these cases, all-polyethylene syringes such as those used for aspirating and dispensing organic solvents (e.g., Norm-Ject® brand) are a useful alternative. Advantages of using syringes for molding of gels include widespread availability, good chemical compatibility, easy-tomanage volumes, ease of demolding (achieved by simply pushing the gel out using the syringe’s plunger), selfcontainment of solvent vapors, and an axisymmetric cylindrical geometry. This latter property offers two distinct advantages for research environments: right-angle cylinders are amenable to measuring materials properties such as bulk density and compressive strength/stiffness, and small cylinders normalize out solvomechanical responses (e.g., warping) associated with flat plates. In a typical syringe-based molding process, the tip end of the syringe is cut off, the syringe’s plunger is moved to the position on the syringe barrel corresponding to the desired gel volume, and sol is poured into the syringe. The open end of the syringe is then sealed with Parafilm or the like. The syringe is then placed vertically in a rack to allow gelation to occur. After the gel has formed and been suitably aged (described below), it can be easily pushed out of the syringe with the plunger in support of furthering processing, preferably while submerged under a bath of solvent where it is neutrally buoyant in order to prevent the gel from breaking as it is pushed out. Other good options for molding basic shapes in a research environment include plastic and glass beakers and small mass-produced plastic containers with accompanying press-fit lids, preferably made of polyethylene or polypropylene (such as hinged-lid polypropylene boxes used to protect video cassettes, two-part plastic lip balm containers, pill organizers, and centrifuge vials). While convenient, such mass-produced plastic containers often have injection molding artifacts, which can serve as nucleation sites for bubble defects in gels or as stress concentrators that can lead to cracking. A diverse range of massproduced containers in various shapes and sizes made of various types of plastic suitable for molding gels can be purchased from commercial container suppliers.

3

Gel-Phase Processing and Solvent Exchange

3.2.2

75

Molding of Complex Shapes and Parts with Large Dimensions

Custom-shape molds can be readily machined from a blank of a stock material such as polypropylene, polyethylene, PTFE, stainless steel, or aluminum. In this context, machining refers to pocketing of a material via subtractive manufacturing to carve the shape of the desired aerogel part into the mold. This may be achieved using a lathe, mill, drill, or other appropriate machining tool. Machining is in fact the only way to make custom molds made from stock shapes of PTFE, which exhibits excellent chemical compatibility. This said, while PTFE has a reputation of being a non-stick plastic, it is usually not a better mold material than other plastics with respect to stiction of gels and aerogels. Surface roughness resulting from machining of mold surfaces may pose problems with respect to bubble nucleation and eventual demolding of gel parts. Smooth mold surfaces and corners with a slight radius of curvature help address these issues. Sanding with successively finer grit sandpaper can help reduce surface roughness. For projects requiring a large number of identical molds, injection molding with polyethylene or polypropylene can be used. Additive manufacturing (i.e., 3D printing), especially with consumer-grade filament printers, generally does not yield suitable molds for aerogel production due to both surface roughness and chemical incompatibility issues. This said, additive manufacturing encompasses a wide range of techniques and materials and is developing at a rapid pace, so likely some variation thereof could be leveraged to produce suitable molds given some development effort. HDPE plastic sheet can easily be made into rectangular molds to produce parts such as tiles and plates (Fig. 3.1). Using 1/1600 -thick (1.5-mm) HDPE sheet, mold edges are first marked and scored using a razor blade. The edges are then bent upward along the scoring lines to form the walls of the mold. Excess material at the corners is cut off, and adjacent walls are welded using a plastic welder or soldering iron. This method can be used to make curved shapes as well by welding a base of the desired 2D cross-section to separately cut walls rather than scoring and bending a single sheet. Since HDPE sheet is readily available in large sizes, molds of this type can easily be made to accommodate gels of large dimensions for relatively low cost. Note that for larger parts (e.g., 30-cm plates and larger), molds should be carefully levelled in order to attain uniform part thickness. Multi-part molds can be a good option for gel materials containing volatile solvents to prevent evaporation or when a meniscus artifact is undesirable. However, multi-part molds can be relatively expensive and cumbersome to produce and require sealing of mold parts, usually with a gasket of some

3

Fig. 3.1 Square-cross-section HDPE molds of dimensions 10 cm  10 cm  5 cm for production of aerogel tiles made by scoring, bending, cutting, and welding of HDPE sheet

type, introducing additional chemical compatibility concerns. Examples of applications where multi-part molds are helpful include production of spherical parts and vertically oriented molds for production of plates of formulations that exhibit out-of-plane warping due to gel density gradients. Molds for complex 3D shapes can be produced from thermoplastic sheet using a basic vacuum-forming setup. Both HDPE and polypropylene can be vacuum-formed into complex geometries. Polystyrene, ABS, and other thermoplastics commonly used for vacuum forming of parts with high-resolution features are not good choices as they have relatively poor chemical compatibility with solvents commonly used in the production of aerogels. Note that vacuum forming is not ideal for making molds for large, flat parts, as flat surfaces tend to warp during the vacuum forming process. For applications requiring complex high-resolution (millimeter-scale) features, silicone molds work particularly well and are easier and less expensive to produce than machined molds. A two-part silicone mold can be used for complex three-dimensional geometries (though care must be taken to prevent or account for leaking along the parting line of the mold). Two-part liquid RTV silicones are readily available and a good option for producing such molds (e.g., Polytek ® PlatSil ® 73-25). Importantly, however, silicones often absorb and are swollen by solvents, which can alter gelation kinetics and target gel densities. This property can also limit the lifetime of such molds. Accordingly, gels produced with silicone molds should generally be demolded prior to further processing.

76

3.2.3

J. S. Griffin et al.

Demolding

Demolding refers to removal of a gel part from the mold in which it was formed. Demolding exposes gel surfaces that are obstructed by the mold in order to decrease the time required for subsequent diffusion-mediated processes such as solvent exchange and drying and prevents gels from warping and/or fracturing during such processes. For example, exposing the surfaces of a gel prior to solvent exchange can be important in many cases to prevent cracking of the gel part due to localized volume changes that may arise when solvents mix. Accordingly, demolding is typically performed after a gel has been suitably aged as described in the following section. In general, demolding involves breaking suction arising from capillary forces between a gel and its mold, ideally without breaking the gel. The surface finish of mold surfaces accordingly greatly influences ease of demolding. Extruded plastics provide much smoother surface finishes than machined plastics and are consequently a better option for producing molds that facilitate easy demolding. PTFE molds often exhibit only moderate demoldability since PTFE molds generally have to be machined to shape and, as a result, end up exhibiting rougher mold surfaces than extruded thermoplastic polyolefin sheets. To assist in demolding of gels from rigid or semi-rigid molds, a mold release agent such as PTFE spray or wax may be used. Mold release agents reduce stiction between the gel and the mold but will generally leave a residue on the surface of the gel, which may be unacceptable for some applications. Shrinkage of a gel part due to syneresis can also aid in demolding in that the gel may recede from mold walls, ideally resulting in detachment from the mold in the process. If shrinkage due to syneresis is too great and the gel has low tensile strength, however, forces arising from stiction between the gel and the mold floor and/or walls as it undergoes large deformations may cause the gel to crack. In the case of silicone and other mold materials that may be swollen by solvents, swelling of the mold can also benefit demolding. In these cases, the mold grows away from the gel rather than the gel shrinking in the mold, resulting in detachment of the mold surfaces from the gel. Mold liners can also be used to facilitate demolding. Examples include thin plastic or rubber sheets, aluminum foil, and household plastic wrap. The gel can then be demolded by simply pulling the liner out of the mold. This method often leaves artifacts in the gel due to creases/waviness in the liner, however, and introduces a consumable into the molding process. Creasing and waviness can be further exacerbated by interactions of the liner with solvents, as even very slight swelling can result in significant undulations in the liner.

3.3

Aging

Gelation can be defined as the first point in time at which a macromolecular network in a sol (which is a colloidal dispersion of nanoparticles in a liquid) extends to the boundaries of the vessel containing the sol. At this point, the sol undergoes a visible transition from flowing in a liquid-like manner to maintaining its shape like a solid, that is, it reaches infinite viscosity. The resultant gel is considered now to have two co-mingled, continuous phases: a solid phase comprising discrete interconnected nanoparticles or intertwined polymer chains, and a liquid phase comprising solvent, unreacted monomers, catalysts, etc. Gelation is by no means the end of the story, however. For example, unreacted monomers in the liquid phase can continue to react and incorporate into the gel’s solid network. Sections of the compliant, semi-mobile solid phase can come into contact with each other, reacting to form larger structures. Neighboring surface functional groups can also cross-condense, tying portions of the network together. In the case of organic polymer gels, polymer chains that make up the gel backbone may experience internal restructuring after gelation, for example, during the formation of a polyimide gel from a poly(amic acid) wherein ongoing imidization can occur after gelation. Such processes that occur after gelation has occurred but before the gel reaches a final equilibrium state are collectively referred to as aging [1]. Aging in silica gels is well characterized and can be attributed to three mechanisms: incorporation of unreacted monomers; contraction of the gel network due to increasing connectivity and resultant expulsion of pore fluid, known as syneresis; and the dissolution and redeposition of material from convex surfaces onto concave surfaces (interparticle necks), known as coarsening or Ostwald ripening [2]. Aging in polymer gels is less extensively studied. Mulik et al. investigated the effect of aging on resorcinolformaldehyde (RF) aerogels, finding that aging time had a significant impact on strength, shrinkage, and bulk density of the final aerogel. In this case, coarsening/Ostwald ripening is not relevant because the polymerization reaction is generally not reversible. However, they found that the mechanisms of incorporation of unreacted monomers, syneresis, and continued reaction between residual functional surface moieties explained the aging and strengthening observed in RF gels well [3]. The purpose of aging is, in the broadest sense, to achieve a gel with stable properties. The properties of an aerogel are highly dependent on the structure of the gel from which it is derived, and therefore, it is important to ensure that the gel is developed fully and repeatably [4]. In the lab, an aging procedure is typically developed empirically and based on gel properties and timing

3

Gel-Phase Processing and Solvent Exchange

convenience (e.g., aging for 24 h or overnight). Although not optimized for processing efficiency, these procedures are designed to produce reliable, repeatable results. Because of the huge variability in the behavior of different gel formulations and chemistries, a single example procedure is not useful. In light of this, a set of considerations for developing an aging procedure are presented below. • Aging Time. Gels are often over-aged to ensure complete reaction. Typical lab procedures call for aging overnight, for 24 h, or some similarly convenient time period. When trying a new formulation, these are good starting points. If gels seem very weak, not fully formed, or develop wavy surfaces upon solvent exchange, aging time prior to solvent exchange can be increased to allow for completion of ongoing reactions. Note that some gels may exhibit significant syneresis upon aging. In some cases, it may be desirable to wait until the gel stops undergoing syneresis before proceeding with solvent exchange to ensure the gel has reached maximum strength, while in other cases, it may be desirable to begin solvent exchange before syneresis has ceased in order to arrest further syneresis when shrinkage of the gel is undesirable. This said, a small degree of syneresis is sometimes desirable since allowing the gel to partially recede from the edges of its container may be helpful in demolding the gel. • Temperature. It may be necessary to heat a gel in order to drive aging reactions to completion in a reasonable amount of time. It is important to consider solvent and gel properties (e.g., boiling point, vapor pressure, flammability, decomposition temperature) when heating. If the gel is not heated during aging, it is important to consider the ambient temperature in the room. Seasonal variation in ambient temperature can cause variability in aerogel properties due to changes in gel aging kinetics. As a result, an isothermal bath set to 20  C may be advantageous for ensuring reliable gel results year round. • Atmosphere. The long standing times and optional heating associated with aging provide ample opportunities for evaporation of solvent from gels, which can cause them to shrink, crack, and/or develop a densified skin on their exterior surfaces. Accordingly, aging should be performed in a sealed, gas-tight container. In systems that use a particularly volatile solvent (e.g., acetone), it may be necessary to add solvent to the aging vessel in order to saturate the vessel’s atmosphere with vapor to inhibit evaporation of solvent from the gel. Alternatively, a gel can be sprayed with a small amount of solvent to keep its exposed surfaces wetted if, for example, it must be exposed to atmosphere when transferred to another container or into a supercritical dryer.

77

• Presence of Moisture. Monomers used in the synthesis of gels used to make aerogels are in many cases reactive with or sensitive to water. Similarly, some common solvents (e.g., acetone, DMF) are prone to readily absorbing water from the atmosphere. The presence of unintended and unmeasured amounts of water in the aging atmosphere and pore fluid should accordingly be avoided, as it can introduce variability. Seasonal variability in humidity can be especially problematic for some gels in this regard. It is thus important to assess the sensitivity of a gel and solvent system to moisture and, in cases where moisture is a concern, to control the humidity of the aging atmosphere by introducing desiccated air or another dry gas into the aging vessel and/or using molecular sieves, calcium sulfate, or another desiccant to absorb moisture.

3.4

Solvent Exchange

3.4.1

Modelling of Solvent Exchange

Solvent exchange is often the most time-consuming step in the production of an aerogel, and as such, it is important to have a fundamental understanding of its underlying process kinetics. The amount of solvent used in solvent exchange is also quite large compared to the volume of gel produced, and the initial acquisition and subsequent recycling and/or disposal of exchange solvent(s) represents a large cost. Therefore, efficient use of solvent is critical. Solvent exchange is typically characterized as a diffusionbased or diffusion-limited process. This means that mass transfer is driven by a species gradient, rather than an external force or pressure gradient. Because of the small pore sizes and low permeability of gels, this is a reasonable assumption in most but not all cases. Modification and exceptions to this are discussed below.

Transport Equations Modelling of the solvent exchange process can be relatively complicated if one considers the fluid mechanics of flow in the bulk solvent outside the gel, the possibility of convective flow within the gel, the effect of the gel network on transport properties, and other factors that may influence the process. The most detailed approach to modelling the solvent exchange process requires solving conservation of mass, species, and momentum simultaneously in the bulk fluid, coupled with a model governed by the same equations in the gel. The resulting model is computationally expensive and not easily generalized to consider changes in gel or bath geometry, flow profiles, etc. However, a few assumptions

3

78

J. S. Griffin et al.

greatly simplify the problem. The following section discusses several simplifying assumptions. Depending on the accuracy required of the model and the specifics of the gel, solvents, and solvent exchange system considered, one can decide which assumptions are acceptable and which may invalidate the model. First, we may assume that the pore fluid is a stationary medium, and that mass transport within the gel occurs only by diffusion. This simplifies the transport equations for the gel domain and allows the system to be modelled by solving the equation for conservation of species within the gel. Second, we may assume that the gel structure does not influence diffusion. As discussed below, it is known that the gel structure, particularly the porosity, can have an impact on the effective diffusion coefficient and therefore the mass transfer kinetics. However, for gels with very low density/high porosity, this effect may be minor. Assuming gel effects are negligible, the governing equation in the gel reduces to @ ðρm1 Þ ¼ ∇  ½D12 ∇ðρm1 Þ @t where ρ is the mixture density, m1 is the mass fraction solute, and D12 is the binary diffusion coefficient of the solute in the solvent. Third, we may assume that the diffusion coefficient and density are constant and concentration-independent, further simplifying the conservation of species equation to @ 2 ðm Þ ¼ D12 ∇ m1 : @t 1 Finally, if the bulk solvent bath is well mixed and large with respect to the gel, we can assume that the solute concentration in the bulk is constant and, ideally, equal to zero, and that the average convective mass transfer coefficient is the same on all surfaces of the gel. This allows us to apply a concentration or flux boundary condition to the gel domain, rather than having to solve the coupled conservation of species equation in the gel and bulk domains simultaneously. If solvent properties are also assumed constant, this reduces the problem to one with an exact solution, which can be solved analogously to transient one-dimensional conduction in a plane wall, cylinder, or sphere [5–7].

Mixture Properties In practice, the properties of a solvent mixture are composition-dependent. In the case of solvent exchange, where the composition of the pore fluid could change from one pure solvent to another pure solvent over the course of the process, understanding how these properties change and affect the process kinetics is important. A couple key properties and their dependence on composition are discussed below.

• Density/Excess Volume. Most liquid mixtures exhibit some degree of non-ideality when mixed together. The practical impact of this on solvent exchange is that combining two equivalent volumes of different solvents will not necessarily result in exactly twice the volume of each solvent on its own. This non-ideality is known as excess volume and can be a positive value, meaning the volume of the mixture is greater than the sum of its components on their own, or a negative value, meaning the volume of the mixture is smaller than the sum of its separated components. The resulting pressure gradients that arise due to this phenomenon are believed to be responsible for some of the deformation observed in weak gels during solvent exchange. • Diffusion Coefficient. The binary diffusion coefficient describes the proportionality between species gradient and diffusive mass flux of the same species. This value is, particularly in liquids, highly composition-dependent. For example, the diffusion coefficient of the acetone– water system changes by a factor of six across the composition range [8]. Measurements of the diffusion of a solute into a solvent at infinite dilution are relatively common and represent the endpoints of the diffusion coefficient vs. composition curve for a binary system. These measurements are typically only valid for dilute solutions, however, and above approximately 10% solute concentration can no longer be used [9]. Commonly, if the infinite-dilution diffusion coefficients can be measured or calculated from semi-empirical correlations (e.g., Wilke-Chang [10] and others), then the intermediate values are calculated using the Vignes correlation [11]. In some cases, experimental values for binary diffusion coefficient for the complete composition range are known in the literature [12, 13].

Porous Media Effects Diffusion in the pore fluid depends not only on the solvent and solute but also on the solid structure of the gel. The volume fraction of the solid phase, as well as its morphology, has an effect on the diffusion coefficient. Behr et al. measured the self-diffusion coefficients of methanol and other solvents in situ in silica alcogels. Comparison of these with selfdiffusion coefficients for free fluid showed that the selfdiffusion coefficient decreases proportionally to decreasing porosity [14]. The morphology of the alcogel (changed by adjusting the pH conditions during gel synthesis) was found to have a smaller but still measurable effect: diffusion in gels with more-branched backbone morphology was lower than in gels with less-branched backbones. These works found that tortuosity (defined as the ratio of the free-fluid diffusion coefficient to the effective self-diffusion coefficient in the porous medium) could be up to 2.5 for the range of conditions studied [15].

3

Gel-Phase Processing and Solvent Exchange

Bulk fluid movement through the gel is countered by the permeability of the porous network and has been shown for silica and organic gels to obey Darcy’s law [16]. Fluid permeability measurements are not widely reported for a wide range of gel formulations and solvents but are available for a few specialized cases [17]. As discussed in the following section, the rate of advective transport is generally small with respect to diffusive transport but not always negligible.

Non-diffusive Mass Transfer While it is typically assumed that mass transport during solvent exchange is purely diffusive, several studies in recent years have investigated the effect of so-called suction and spillage – advective mass transport driven by compositiondependent mixture density – specifically during solvent exchange into carbon dioxide. Essentially, as the solvent outside a gel diffuses into the gel’s pore fluid, the composition of the gel’s pore fluid changes, and as a result, the density of the pore fluid changes. This change in density can cause a bulk velocity to arise in the liquid phase, causing fluid to spill out of or get sucked into the solid network of the gel. Karamanis et al. developed a numerical model of this process and found that the mass flow rate due to densitydriven advection during supercritical drying is about one order of magnitude smaller than the mass flow rate due to diffusion [18]. Bueno et al. experimentally investigated a similar effect as gaseous carbon dioxide diffused into the pore fluid of a gel during the pressurization phase of supercritical drying. They found that the significant change in mixture density that occurs during this phase can be responsible for removal of up to 60% of the pore fluid in certain cases [19]. In addition to contributing to the kinetics of solvent exchange, density-driven pressure and flow can give rise to stresses in the gel network, resulting in change in dimension or, in severe cases, fracture. This is discussed more below.

3.4.2

Effects of Solvent Exchange on Gel Properties

In almost all cases, there is a degree of shrinkage that occurs during aerogel processing. This has been known since Kistler, who observed that it was not possible to produce an aerogel from a hydrogel without some amount of shrinkage and that this shrinkage could be attributed to both exchange from water into solvent and supercritical drying [20]. Shrinkage can be defined as either linear shrinkage, SL, where SL ¼

Li  L f Li

or volumetric shrinkage, SV, where

79

SV ¼

Vi  V f : Vi

Volumetric yield can then be defined as the ratio of the gel volume after exchange to its initial volume, which is related to volumetric shrinkage: Y¼

Vf ¼ 1  SV : Vi

Polymer Concentration A survey of published data on volumetric yield for biopolymer aerogels as defined above was collected and plotted against weight percent polymer in the original sol by Gurikov et al. [21]. Although the data represent a variety of biopolymer chemistries, morphologies, and gel geometries, the general trend shows that low-weight-percent biopolymer gels are much more likely to undergo significant shrinkage. This supports the common-sense rule known by researchers that gels derived from polymer-rich solutions are less likely to shrink than gels derived from dilute solutions. This can be attributed in part to the fact that mechanical strength and stiffness scale nonlinearly with bulk density, and so polymer-rich gels are more resistant to shrinkage arising from capillary stress. The curve fit shown in the figure in Gurikov et al. suggests that it is exponentially more difficult to control shrinkage as polymer concentration decreases. As discussed below, the most commonly used process to mitigate shrinkage during solvent exchange is multi-step or gradient solvent exchange, wherein the composition of the solvent bath progresses from more similar to the pore fluid to more similar to the target solvent. Through the use of several intermediate steps, the severity of concentration gradient (defined herein as the difference in concentration of the solvent in the pore fluid vs. the solvent bath at the beginning of a solvent exchange) to which the gel is subjected is decreased. The logic behind multi-step solvent exchange is simple. If the concentration gradient between the pore fluid and the solvent bath is reduced, the process is closer to equilibrium and therefore less likely to result in spatial differences in properties, thereby reducing stress on the gel network. It should be noted, however, that reducing the concentration gradient necessarily reduces the driving force for diffusion and thereby slows the mass transfer process, and requiring successive steps increases the total duration of solvent exchange. Therefore, the trade-off between shrinkage/volumetric yield and process time must be considered and optimized for each aerogel formulation. It is also noted that for some gel systems such as those based on biopolymers, a gel’s sensitivity to solvent exchange gradient changes throughout the solvent exchange process.

3

80

J. S. Griffin et al.

Therefore, it may be possible to develop a solvent exchange procedure in which the concentration gradient is matched to gel sensitivity in order to maximize driving force for mass transfer while avoiding excessive shrinkage [21].

Solvent–Polymer Interaction Interactions between the solid backbone of a gel and the solvent must also be considered. Figure 3.2 shows a particularly solvent-sensitive aromatic polyamide gel derived from terephthaloyl chloride and 2,2′-bistrifluoromethylbenzidine that was synthesized in NMP after solvent exchanging into several different solvents commonly used in aerogel processing. Depending on the solvent into which they are exchanged, such gels may contract, swell, denature, or dissolve upon solvent exchange. Similarly, Subrahmanyam et al. found that alginate hydrogels solvent exchanged from water into 15 different solvents maintained volumetric yield of more than 10% in only six cases [22]. Recently, Takeshita et al. showed that for certain chitosan gels, the process of exchanging the gels from methanol into CO2 critically impacts not only the volumetric yield but the actual structure of the gel itself. Using SAXS, they observed that the nanofibrous structure of the final aerogel is not

present after gelation or solvent exchange into methanol but is only formed during their final exchange into CO2 [23]. The trace water content of solvents is also a very important concern. Some gels are very sensitive to water content, for example, gels with reactive side groups that need to be preserved for later reaction. Gels that are sensitive to or reactive with water may require special conditions for solvent exchange such as inclusion of desiccating agents like molecular sieves and/or even blanketing with a dry cover gas. Most solvents commonly used in the lab contain water as an impurity and are very likely to uptake more water from the air the more they are handled if steps are not taken to prevent this [24]. Gurikov et al. developed a generalized description of the effect of solvent–polymer interaction on gel shrinkage. Starting from the concepts that there should be a balance between the hydrogen bonding ability of the solvent (which promotes swelling) and the idea that solvents with a low dielectric constant should allow for significant interaction between polymers (which promotes shrinkage), they proposed that the ratio of hydrogen bonding component δh of the Hansen solubility parameter to the dielectric constant ε of the solvent should correlate with shrinkage. They found that

Water

THF

DMSO

Tert-butanol

Ethanol

Acetone

Fig. 3.2 Aromatic polyamide gel derived from terephthaloyl chloride and 2,2′-bistrifluoromethylbenzidine after solvent exchanging from NMP-based mother liquor into various solvents, exhibiting a range of solvent-dependent solvent–backbone interactions including shrinkage,

swelling, dissolution, and denaturing. The mold shown in the photograph labeled tert-butanol provides a reference for the initial gel shape before solvent exchanging. Gel formulation from [38]

3

Gel-Phase Processing and Solvent Exchange

for several biopolymer gels, minimal shrinkage occurred when this ratio was between 0.45 and 0.65 [21].

Influence of Gel Stiffness Shrinkage during solvent exchange is resisted by the stiffness, also known as modulus, of the gel network. By increasing the stiffness of the solid backbone, gels can be made to shrink less during solvent exchange and even survive evaporative drying. In the case of silica aerogels, this can be achieved by aging in a monomer solution via deposition of additional silica over mechanically weak interparticle neck regions between secondary particles in the gel’s solid backbone, or alternatively through introduction of a conformal polymer coating over the skeleton of the gel [25]. Polymer and biopolymer gels, which may not necessarily be modifiable in the same way, can be reinforced through the incorporation of high-stiffness fibrils [26–28]. The use of fillers (chemically non-interacting reinforcement) such as starch [29] or nanofibrillated cellulose [30] has been shown to have a moderate effect on reducing shrinkage. Compositing one polymer network with another has also been shown to effectively reduce shrinkage. An example of this is alginate/chitosan hybrid aerogels [31]. It is hypothesized that hydrogen bonding between components of the two polymers may decrease aggregation during solvent exchange [21]. It is also possible for the solid backbone of the gel to change during and as a result of solvent exchange. Ganesan and Ratke found that κ-carrageenan gels crosslinked with potassium thiocyanate showed significant axial shrinkage after solvent exchange into water and then acetone, but almost no shrinkage when exchanged directly into acetone. In this case, the additional water exchange washed out the crosslinker, leaving an uncrosslinked gel with low stiffness [32]. Similarly, small oligomers and unreacted monomers can also be washed out of gels during solvent exchange before the monomers are able to covalently attach to the gel network. Other Influencing Factors Gel size also has an impact on shrinkage. Veronovski et al. found that gel beads less than a few millimeters in diameter showed much lower shrinkage than larger gels, even in a single-step exchange from water into pure ethanol [29]. It has also been observed that in certain cases, addition of a small amount of particular low-molecular-weight compounds to a gel can reduce shrinkage. Veronovski et al. found that a small amount of nicotinic acid added to a low-weight-percent alginate solution before gelation reduced shrinkage of the resulting gel [29]. Modelling of Gel Shrinkage A theoretical basis for change in gel volume would prove extremely useful in designing solvent exchange protocols in

81

order to maximize volume yield. One proposed theory is the Flory-Huggins model, which considers phase transition to be a balance between repulsive forces that act to swell a polymer and attractive forces that contract the network. At equilibrium, these two forces must be equal. The equation describing this equilibrium condition is given by Tanaka et al. [33] and offers value for qualitative description of shrinkage during solvent exchange. A detailed description of its application to biopolymer aerogels can be found in Gurikov et al. [21].

3.4.3

Practical Aspects of Solvent Exchange

How Much Solvent Is Required? In order to effectively and efficiently exchange a gel from one solvent into another, one must, at a minimum, understand the objectives of the solvent exchange process. What is the composition of the mother liquor? Does the mother liquor contain substances that will not be removed by the drying process to be used? Is the solvent the gel is being exchanged into going to be evaporated, sublimed, or supercritically extracted from the gel? If supercritically drying from carbon dioxide, is the gel of a suitable composition and size that it can tolerate having its pore fluid extracted by a flow of supercritical carbon dioxide, or is the gel large, complex in shape, or otherwise sensitive to cracking and must be fully exchanged into liquid carbon dioxide prior to supercritical extraction? In general, the goal of solvent exchange is to replace the liquid within the pores of a gel with another liquid composition. Ideally, this means that at the end of the solvent exchange process, the pore fluid is composed entirely of, or very close to entirely of, the target solvent system. However, since solvent exchange is inherently a dilutive process governed by the rule of mixtures, it is not practicable to reach a pore fluid concentration of 100% target solvent. As such, one must establish a threshold for pore fluid purity that is considered sufficient. As an example, alginate gels that were solvent exchanged into ethanol were shown to maintain their surface area after supercritical drying from CO2 as long as the pore fluid was more than 91% ethanol by weight before supercritical drying. The same gels if solvent exchanged into DMSO required 98% purity to maintain the same surface area [22]. When exchanging from pure ethanol into CO2 at 323 K, the mixture must be greater than 98.7% CO2 by weight in order to maintain a single-phase mixture [34]. The case of alkoxide-derived silica aerogels can be used as an example. Synthesis of gels from tetramethoxysilane (TMOS) or tetraethoxysilane (TEOS) requires addition of water in order to hydrolyze alkoxide groups into silanols which subsequently condense into siloxane bridges, rereleasing water as a by-product. As a result, a significant amount of water is present in the pore fluid of such gels

3

82

following gelation. If supercritical drying of such a gel with carbon dioxide were attempted without first performing a compatibilizing solvent exchange into a carbon-dioxide-miscible solvent such as methanol, water in the pore fluid would not be removed during drying since water and carbon dioxide are immiscible [21, 22]. Thus water would remain in the gel’s pore network after supercritical drying and cause the remaining product to undergo collapse as the remaining water evaporates. As previously stated, performing a solvent exchange with a solvent that is miscible with carbon dioxide prior to supercritical drying can eliminate this problem, however since solvent exchange is dilutive in nature, some water will always remain in the gel even after multiple solvent exchanges, albeit only in vanishingly small quantities. For some types of gels, especially oxidic gels, this can be further complicated by the fact that water may remain adsorbed on the surface of the gel backbone even after multiple solvent exchanges. This is acceptable, however, since a mixture of an organic solvent such as methanol that contains a small amount of water can freely mix with carbon dioxide i.e., there are single-phase compositions in the threecomponent system. Thus, a threshold exists below which the presence of some residual water in the gel network is not problematic. It should be noted, however, that the preceding discussion assumes that each solvent exchange step is long enough for the system to reach equilibrium, after which no concentration gradient remains within the gel. Before this time, the water content at the middle of the gel is higher than the average value calculated from a simple mixing model. Pore fluid composition during solvent exchange is not typically measured directly but is inferred from another measurement such as expected average solvent bath composition. It is therefore necessary to use a diffusion model to analyze the composition within the gel or to use a conservative estimate for required solvent exchange time.

Equations for Solvent Exchange Process Design Solvent Composition As detailed in the example above, the goal of solvent exchange is often to reduce the amount of a particular species present in the mother liquor. For example, it may be necessary to reduce the concentration of water, unreacted monomers, or catalysts in the pore fluid before supercritical drying. In other cases, it may be necessary to exchange the mother liquor for a completely different solvent system, for example, as in the case of freeze drying, where the synthesis solvent (e.g., a primary alcohol or a ketone) must be exchanged for a solvent that can be frozen and sublimed (e.g., water or tertbutanol). Solvent exchanges dilute the pore fluid within a gel with solvent from the bath surrounding the gel.

J. S. Griffin et al.

Solute Volume Fraction ðφB Þ ¼

Solute Volume : Total Volume

Assume we have a gel of volume Vgel and want to purify its pore fluid via solvent exchange. Assume the gel comprises a pore network that contributes volume Vgel porosity to the volume of the gel and a solid skeleton that contributes volume Vgel skeleton to the gel. The gel’s total volume Vgel is thus V gel ¼ V gel porosity þ V gel skeleton : Initially, the volume of the mother liquor in the as-synthesized gel is equal to the volume of the gel’s porosity, since by definition the mother liquor is the liquid that is contained within the pores of the gel immediately after gelation, i.e., V mother liquor in gel; initial ¼ V gel porosity : For low-weight-percent-solid gels, the volume occupied by the gel skeleton can be ignored and Vgel porosity can be approximated as V gel porosity ¼ V gel : Volume fraction, denoted by φ, refers to the ratio of the volume that a species of interest occupies in a mixture or gel relative to the entire volume of the mixture or gel, i.e., φspecies ¼ V species ∕ V total : The volume fraction of mother liquor in the gel is thus initially φmother liquor in gel; initial ¼ V gel porosity ∕ V gel : Because Vgel can be approximated as Vgel porosity for lowweight-percent-solid gels, the approximation φmother liquor in gel, initial = 1 can also be made for such gels. Now assume that we place the as-synthesized gel in a bath of a target solvent having a volume Vbath and that we let the gel–bath system reach compositional equilibrium. This represents performing a first solvent exchange of the gel. After this first solvent exchange, the volume fraction of residual mother liquor in the gel will be φresidual mother liquor in gel; 1 ¼ φmother liquor in gel; initial

 V gel porosity ∕ V gel porosity þ V bath ¼ φmother liquor in gel; initial  1 ∕ 1 þ V bath ∕ V gel porosity For low-weight-percent-solid gels, this becomes φresidual mother liquor in gel; 1 ¼ 1 ∕ 1 þ V bath ∕ V gel

3

Gel-Phase Processing and Solvent Exchange

83

Assume now the gel is transferred into a new bath of fresh target solvent, again having a volume Vbath, and that the gel– bath system is again allowed to reach equilibrium. This represents performing a second solvent exchange of the gel. After this second solvent exchange,

we consider some critical target volume fraction, φresidual mother liquor in gel, target, that must be achieved, then the amount of solvent required to achieve this is simply V target ¼ V mother liquor in gel; initial 

φresidual mother liquor in gel; 2 ¼ φresidual mother liquor in gel; 1

 V gel porosity ∕ V gel porosity þ V bath ¼ φmother liquor in gel; initial  V gel porosity ∕ V gel porosity þ V bath  V gel porosity ∕ V gel porosity þ V bath ¼ φmother liquor in gel; initial  1 ∕ 1 þ V bath ∕ V gel porosity

2

For low-weight-percent-solid gels, this again reduces to φresidual mother liquor in gel; 2 ¼ 1 ∕ 1 þ V bath ∕ V gel

2

Thus after n such successive solvent exchanges into fresh target solvent, the concentration of mother liquor remaining in the gel will be φresidual mother liquor in gel; n ¼ φmother liquor in gel; initial n

 1 ∕ 1 þ V bath ∕ V gel porosity

which, for low-weight-percent-solid gels, likewise reduces to n

φresidual mother liquor in gel; n ¼ 1 ∕ 1 þ V bath ∕ V gel : Finally, if the goal is to remove a specific contaminant or component of the initial mother liquor, an additional factor representing the volume fraction of said contaminant or component in the initial mother liquor can be included accordingly: φresidual contaminant in gel, n ¼ φcontaminant in mother liquor; initial  φmother liquor in gel; initial  1 ∕ 1 þ V bath ∕ V gel porosity

n

which again for low-weight-percent-solid gels reduces to φresidual contaminant in gel, n ¼ φcontaminant in mother liquor; initial n

 1 ∕ 1 þ V bath ∕ V gel : Total Volume of Solvent Required In the ideal case, the amount of solvent used during solvent exchange is simply the amount required to dilute the mother liquor or other specific contaminant to an acceptable level. If

φmother liquor in gel; initial 1 φresidual mother liquor in gel; target

In practice, the total amount of solvent used may exceed the critical (minimum required) volume for various reasons, the most common being that solvent exchange is typically conducted as a batch process in which gels are soaked in a discrete number of successive pure solvent baths of identical volume. As such, the total volume of solvent used may overshoot the minimum required volume value. Additionally, the geometry of a gel part or solvent bath container may require excess volume of solvent to be added to the bath in order to fully cover the gel. For these reasons, a volume of solvent greater than the mathematical minimum may be required in order to practicably implement a solvent exchange. Minimizing Time vs. Volume of Solvent Used Mass transfer during solvent exchange of gels takes place by diffusive transport within the gel and by coupled convective and diffusive transport outside the gel. Under fixed conditions (temperature, pressure, solvent/solute combination), the maximum solvent exchange rate is achieved when the solute concentration at the gel surface is minimized, thereby maximizing the driving force for diffusion within the gel. To achieve this, the solvent bath should be perfectly mixed and infinitely large with respect to the gel. In practice, this ideal situation is approached by using a very large, wellmixed bath and/or by providing a continuous flow of clean solvent. In a batch solvent exchange process, solute that diffuses out of the gel accumulates in the bath and the rate of diffusive transport within the gel slows as the composition of the pore–fluid and the bath system approaches equilibrium. In a continuous-flow solvent exchange process, the flow rate through the vessel determines the rate at which clean solvent is provided, the amount of mixing in the free volume of the vessel, and the mass transfer coefficient at the gel surface. Low flow rate results in pore fluid accumulation near the gel surface, slowing the solvent exchange process. By defining a critical solvent concentration that must be reached and assuming the solvent bath-to-gel-volume ratio is the same for each exchange for ease of execution, one can easily calculate the number of exchanges that need to be

3

84

J. S. Griffin et al.

Solvent use vs. Bath size

50

9

45

8

40

7

35

6

30

5

25

4

20

3

15

2

0

2

4

8

6

10

2

Volumes solvent used

Number of solvent exchanges

10

length are constants, state that diffusion time scales with the square of characteristic length Lc:

10 12

Vbath/Vgel Fig. 3.3 Given a fixed bath-to-gel-volume ratio, a certain number of solvent exchange steps need to be performed in order for a gel’s pore fluid to reach a critical target solvent. This plot shows the number of exchanges and the total solvent required to reach 99.99 vol % target solvent concentration for a range of bath-to-gel-volume ratios. Plot assumes an ideal gel that does not undergo volumetric or compositional changes upon solvent exchange

performed. Then, from the number of exchanges and the volume ratio, total volume of solvent can be calculated. Figure 3.3 shows this for achieving a critical value of 99.99 vol % target solvent in the gel’s pore fluid. It is apparent from the figure that by increasing the bath-to-gel-volume ratio, it is possible to perform the solvent exchange in fewer steps. However, this comes at the cost of total solvent usage. The same basic trend is evident in continuous-flow solvent exchange as well. For the case of ethanol exchange into supercritical carbon dioxide, Griffin et al. found that over the range investigated, increasing carbon dioxide flow rate decreased solvent exchange time but increased total carbon dioxide consumption [35]. Exchange Time Required vs. Characteristic Gel Dimensions The analytical solution to the problem of transient diffusion within a gel (considered a plane wall) is typically represented by an infinite series. However, if the mass transfer Fourier number is sufficiently large (i.e., Fom > 0.2), the solution can be represented by a single term. Given the mass transfer Fourier number, Fom ¼

DAB t L2c

where DAB is the coefficient of diffusion between species A and B and t is diffusion time, we can, in the case where the Fourier number, diffusion coefficient, and characteristic

t / Lc

where Lc is equal to the part volume divided by its exposed surface area. In the example of a plate (where length and width are large compared to thickness) exposed to solvent on all sides, the characteristic length is one-half of the thickness. More detailed discussion of this derivation can be found in Incropera et al. [5]. In practice, the preceding calculation is used more often to calculate the relative change in solvent exchange or drying time between gels with different dimensions than to calculate the absolute time required. For example, given two gels with characteristic dimension (i.e., thickness) L and 2L, the ratio of time required for each to reach the same state is 2 t2 ð2LÞ 4L2 ¼ 2 ¼ 2 ¼4 t1 L L

In this case, doubling the thickness of the gel increases the time required to reach equilibrium in each solvent exchange by a factor of four, or different terms, cutting the thickness of a gel in half quarters the solvent exchange time. Note this rule of thumb applies to drying as well. This is an effective method for back-of-the-envelope estimation of the time required for solvent exchange. In practice, the actual time required will also depend on factors including bath geometry, spacing of gels in the bath, mixing efficiency, etc. It should be noted that the characteristic length as defined above is based on the exposed external surface area of the gel part. If a gel is confined in a mold during solvent exchange or placed against an impermeable surface (e.g., the bottom of the vessel, or another gel), the exposed surface area is reduced, thereby increasing the characteristic length and thus the required solvent exchange time. It is therefore important to ensure that the gel is exposed to free solvent on all major faces in order to minimize solvent exchange time. Further discussion of rack design to facilitate this can be found in section “Filling, Draining/Decanting, and Mixing.”

Multi-step and Gradient Solvent Exchange As early as Kistler’s original works on aerogels, it was appreciated that it may be necessary for certain gel chemistries, or for certain combinations of pore fluid and target solvent, to perform a more elaborate solvent exchange than simply soaking a gel in the desired solvent. In some cases, the pore fluid, or a component of the pore fluid, may be immiscible with a target solvent. Kistler recognized that complete miscibility is a requirement for solvent exchange. Immiscibility can result in the formation of a

3

Gel-Phase Processing and Solvent Exchange

two-phase mixture, and the interfacial forces at the phase boundary can cause damage to the gel in the same way that a liquid-vapor interface does during evaporative drying. Even in a less extreme situation, immiscibility will result in incomplete exchange [20]. A very common example of this is the standard procedure for preparing alkoxide-derived silica gels for supercritical carbon dioxide drying. Hexane and heptane, which are commonly used low-surface-tension solvents used in evaporative drying of aerogels, are immiscible with methanol and acetonitrile, which are common synthesis solvents for aerogels. In another example, de Vries et al. produced oleogels through stepwise solvent exchange to replace the aqueous pore fluid in a protein hydrogel with sunflower oil by first exchanging the hydrogel into THF or acetone, or a mixture of the transfer solvent and water or oil [36]. Okay et al. found that polyacrylamide hydrogels developed heterogeneous structures when exchanged into acetonerich mixtures of water and acetone. They found that the emergence of a dense polymer region at the surface of the gel then hindered further diffusion and prevented equilibration with the surround solvent bath [37]. Williams et al. encountered this effect in the fabrication of polyamide aerogels. They noted that while polyamide gels derived from terephthaloyl chloride shrank uniformly during solvent exchange and drying (typical of most gels), formulations derived from isophthaloyl chloride developed a densified skin and, in some cases, exhibited cracking, warping, and/ or blistering [38]. One way around such challenges is the use of a multi-step gradient solvent exchange process, where baths of successive solvent blends of gradually increasing concentration of the target solvent are used. This helps to step a gel gradually into the target solvent in a way that circumvents denaturing. A typical procedure for gradient solvent exchange is to vary the composition of the successive solvent exchange baths from more similar in composition to the initial pore fluid to more similar to the target solvent. For example, to solvent exchange a gel from NMP into acetone, one might use the following series of baths: 3:1 NMP:acetone, 1:1 NMP:acetone, 1:3 NMP:acetone, and finally pure acetone. Examples of similar procedures can be found in the literature [39, 40].

Gel and Liquid Handling Temporary Removal and Transfer of Gels When performing a solvent exchange, gels will typically be transiently exposed to atmosphere between solvent baths. On a lab scale, this typically means emptying and refilling a container holding the gels, or alternatively transferring gels from one container of solvent to another. On a larger scale, this means draining and refilling a tank, typically with a pump. During any period when gels are not fully submerged, they should be kept in an enclosed environment in order to

85

allow a vapor pressure to build up around them as to prevent damage due to evaporation from exposed gel surfaces. Often, for a lab-scale batch solvent exchange, intentional maintenance of this vapor atmosphere is not necessary if the time that the gels are exposed to atmosphere is sufficiently short such that a small amount of liquid remains clinging to the exterior surfaces of the gels and container walls so that the gels inside will not dry out. More robust gels that can tolerate modest tensile forces, for example, synthetic polymer gels, are more forgiving in this regard relative to, for example, a low-density silica gel that is prone to immediate cracking as soon as its surfaces begin to dry out. This said, greater care to keep gels from drying out must be exercised when gels are exchanged into readily evaporated solvents such as hexane or acetone than for, say, ethanol or NMP. Spraying gels with the final target solvent while they are exposed to the atmosphere can help buy time before evaporation of solvent from exterior gel surfaces causes damage to the gels. Vapor Management For larger solvent exchanges that cannot be performed in a fume hood, supplementary vapor management becomes necessary for both quality control and safety. Consider a batch solvent exchange performed in a tank: as the tank is drained and filled, or opened and closed to insert or remove gels, there will be vapor release from the tank, which can be a flammability and/or health concern. To manage vapor release when the tank is opened, a ducted exhaust vent can be placed over the tank to collect volatile vapors. Doing so may not be necessary in all cases, but whether or not an exhaust vent is needed should be carefully considered according to expected vapor concentrations and associated health and flammability hazards in the surrounding area. The tank itself should have a gas inlet and exhaust outlet appropriately designed to manage the gasses that will inevitably flow into and out of the tank when it is emptied and filled with solvent. If not controlled, when ambient air is allowed to enter the tank upon emptying of solvent from the tank, the gels will be exposed to humidity and oxygen that can cause variability in product properties and/or pose a flammability hazard. Likewise, upon filling the tank with solvent, the solvent vapor that previously filled the empty tank will be displaced and therefore must be safely vented. Vapor management is also necessary when transferring gels from their aging step to solvent exchange or transferring the gels from solvent exchange to a drying chamber. As such, it is important to carefully consider the risks associated with solvent vapors and how these risks scale with solvent volume, and to observe proper safety protocols and equipment design practices at all times. Inerting of large tanks is also essential to prevent an explosive fuel-air mixture; consult the safety data sheet for the solvent being used and check its lower explosive limit when designing such equipment.

3

86

Solvent Exchange Apparatus and Methodologies This section discusses various considerations underlying the development of solvent exchange protocols and associated apparatus. It should be noted that should a process involve quantities of solvents beyond what would typically be used for benchtop experiments or involve automated handling equipment (e.g., pumps, transfer lines, storage vessels, relays, etc.), careful attention must be paid to mitigating potential hazards that may arise (e.g., flammable vapor buildup, static discharge, leaks/spills, component failure due to material incompatibilities). Accordingly, be sure to understand the risks associated with handling flammable solvents and to take all necessary precautions to ensure that equipment is properly designed and safely implemented. Materials and Chemical Compatibility Small volumes (e.g., lab-scale batches) of gels can easily be solvent exchanged manually using compatible plastic or glass containers. Glass food containers with plastic sealable lids are a common option for solvent exchange, as are gasketed polypropylene plasticware storage containers (e.g., Snapware ® or Lock & Lock ® brand). Note, however, that the gasket seals for such containers are typically made of silicone rubber and therefore should be minimally subjected to solvents, as many solvents will swell or damage silicone. Bench-scale solvent exchange can easily be performed in a fume hood using funnels and pouring from small solvent containers. Small manual pumps can also be used in these cases, as can small electric pumps (e.g., rotary, diaphragm, or peristaltic pumps, covered in further detail later in this section). At this scale, waste volumes can easily be collected for disposal, spills will typically be minor, and vapors are well managed by a fume hood ventilation system. If the volume and dimensions of the gels and solvent exchange containers are such that a solvent exchange can no longer be easily performed manually in a fume hood, safety concerns, materials compatibility, pumping systems, and solvent exchange tank designs become more important. Careful attention to materials compatibility is necessary for solvent exchange, since tubing, tanks, and pumps will be subject to extended continuous exposure to solvents. Materials that are typically compatible with most common organic solvents include 316 stainless steel, high-density polyethylene, and polypropylene (hexane being a notable exception, which is incompatible with polypropylene). Chemical/materials compatibility tables are a good first step to verify compatibility in advance, noting such tables often lack detail about mode of failure, do not list desired operating conditions, or are sometimes just plain wrong. It is therefore recommended that a material is tested for compatibility with the target solvent through exposure testing before being used for any critical components. This can be as simple as soaking a coupon of the material in the solvent of concern (at the

J. S. Griffin et al.

relevant operating temperature) for a period of time and then removing it, measuring any change in mass that results, and inspecting for other changes in properties (e.g., flexibility, color, overall integrity). Elastomeric seal materials are notably challenging to pair with many solvents. Silicone rubber, for instance, tends to swell in many solvents, which may or may not be acceptable depending on the nature of its use. Filling, Draining/Decanting, and Mixing It is often desirable to accelerate the solvent exchange process. Forced convection can help to maximize the species gradient at the surface of the gel such that the time required for the solvent exchange is limited only by the diffusion through the gel material and is not needlessly extended by buildup of the pore fluid adjacent to the gel. Likewise, if there is a density difference between pore fluid and the target solvent, mixing may be needed to make the solvent bath homogenous from top to bottom. Particularly with larger gel parts or complex gel geometries, the buildup of the pore fluid in dead zones (i.e., where natural convection and/or diffusion are impeded by geometry or density gradient) can yield inconsistent solvent exchange throughout the part if mixing is insufficient. Magnetic stir plates, magnetically coupled mixers, or overhead mixers can be used assuming that the geometry of the tank or bath permits. A diagram of an example setup for performing solvent exchange of gels with mixing at the lab scale is shown in Fig. 3.4. In a typical setup, gels are placed on top of an upside-down-U-shaped rack made of a steel (note that aluminum is not a good choice for performing solvent exchanges into methanol as methanol can degrade aluminum over time as indicated by blackening of the methanol) or polypropylene grid inside a sealable polypropylene box. A magnetic stir bead is then placed under the rack and rotated by a magnetic stir plate. This

Gel

Fig. 3.4 Diagram of a setup for performing solvent exchange of gels with mixing at the lab scale. A gel is placed on a steel or polypropylene grid inside a sealable polypropylene box filled with the target solvent. A magnetic stir bar is placed under the rack

3

Gel-Phase Processing and Solvent Exchange

ensures that the solvent diffusing out of the gels and the surrounding bath solvent can be mixed without the magnetic stir bead hitting the gel parts. In some instances, gels may float in the solvent exchange bath if their pore fluid comprises a liquid that is lower in density than the target solvent. In these cases, a polypropylene grid may be placed on top of the gels to weight them down without overly obstructing solvent exchange. Depending on bath geometry, a mix head or stir bar still may not properly mix portions of the tank, however. In these cases, another approach is to induce circulation in the tank using a pump. If further dead zones need to be eliminated within the tank, they can be addressed with multiple outlet nozzles to evenly distribute flow throughout the tank. Note that for some low-strength gels, swelling or shrinkage associated with solvent exchange can cause cracking or warping of the gel part if the solvent exchange process is too abrupt, as discussed in the previous section. The dimensions of available containers and of gel parts may mean that excess solvent volume is needed in the solvent bath in order to cover the parts. In such cases, adding ballast to the bath can help to minimize wasted solvent. For baths that will be drained, the bottom of the tank should be shaped to allow for complete filling and emptying of the tank in order to reduce the amount of contaminated solvent remaining in the tank after each solvent exchange. For multiple gels or large parts, racks may be needed to easily handle and/or stack gels. The racks should be designed such that they allow solvent to easily access all sides of the gels. When well executed, this effectively halves the characteristic diffusion length through the gel relative to the situation where a gel is resting on a flat impermeable surface. Perforated sheet, louvered or expanded metal sheet, and wire shelves all work well as rack materials. Minimizing large areas of contact of the gel with the rack can help to prevent heterogeneous solvent exchange. For example, in the case of a perforated sheet used to support a gel, if the spacing between perforations is greater than the thickness of the gel part, diffusion through the underside of the gel will be hindered, and the characteristic diffusion length is increased relative to if the gel were suspended in the solvent. If the rack is too sparse, pressure at contact points can impart indentations to the gel during solvent exchange. Using a heavy-gauge wire rack layered with a thingauge louvered sheet gives the rack rigidity, supports the gel, and minimizes continuous contact area with the gel. For large batches of materials where a tank is necessary, pumping may be needed to fill and drain the tank. Centrifugal or gear pumps in particular are suitable to move liquid solvents. These pumps are typically not self-priming, and particular attention should be paid to the chemical compatibility of gaskets and seals in them, particularly in moving-seal pumps. A magnetic drive pump (which utilizes a magnetically coupled impeller) has only static seals and thus reduces the chance of leaking from the seal if, for instance, the seal

87

material is slightly swelled by the solvent. Diaphragm pumps are another good option and can be self-priming. These pumps may require more frequent part replacement than a well-chosen centrifugal pump due to the nature of their operation and the stresses experienced by the polymeric diaphragms, however, they can be made from highly chemically compatible materials without elastomeric seals. Peristaltic pumps are challenging for large volumes, since the flexible tubing required can pose a chemical compatibility risk and may fail prematurely if the polymer tubing is weakened by the solvent. Both electrically driven and air-powered pumps are available with different hazard-class ratings depending on the area in which they will be used. Air-driven pumps eliminate electronics from the pumping area and are therefore safer for use in hazardous locations without the substantial added cost associated with specifying a hazardous-location-rated explosion-proof electric motor. Temperature Control Temperature control may be important for controlling the rate of diffusion in solvent exchange and, in turn, managing quality control. An electric lab oven or cold chamber may be used for lab-scale solvent exchanges, but caution should be taken to avoid buildup and ignition of vapor, as these units are typically not explosion-proof. An exception would be a friction-heated oven, which heats by friction of the air in the oven as it is circulated. Any enclosure, as with an oven or heated room, has the added risks of solvent vapor accumulation should there be a small vapor leak in the primary solvent tank. Electric heaters such as flexible silicone heaters can be used to heat surfaces of a solvent tank, but caution should be taken to avoid ignition risk and also the risk of overheating or boiling the solvent. Submersible electric heaters are not commonly used, since there is a greater potential for auto-ignition given that these heaters typically have a high heat density and exposed portions can rapidly overheat. A circulating chiller or heater can be used with a submerged heat exchanger, a pad heat exchanger on the solvent tank, or an enclosure that surrounds the tank. This method can remove potential electric ignition sources from the hazardous area. For large installations, steam or hot water from the facility can be used.

Solvent Recycling If a process is such that solvent exchange is required prior to drying, one must decide whether to recycle or dispose of the used solvent/solvent mixtures, the volume of which is typically many times greater than the volume of the gel itself. For lab-scale operations, single use and disposal may be acceptable, but for larger-scale operations, solvent use can become the dominant cost in the manufacturing process if not recycled effectively. From an early stage in process development, it is important to consider the ease of separation and

3

88

J. S. Griffin et al.

recycling of solvents to be used. Typically, the most economical method of solvent separation is distillation. Solvents with similar boiling points can be quite challenging to separate via distillation. Even trace amounts of unintended solvents can accumulate after many cycles of use if the boiling point of the contaminating solvent is too similar to that of the primary solvent that is being distilled. Fractional distillation increases the ability to separate solvents of similar boiling points but is more complex and expensive than simple distillation. Another pitfall of separation by distillation is azeotrope formation. An azeotrope is a mixture of two solvents that has a set composition and boiling point and is not separable by conventional distillation. Water is an example of a solvent that forms an azeotrope with many common organic solvents, e.g., ethanol. Check available azeotrope tables to be sure that the solvents to be used are separable. Solvent systems with azeotropes or like-boiling-point solvents are not impossible to separate, for example, by using pressure-swing distillation, but may involve added challenges and processing costs. Lab-scale distillation systems are readily commercially available, as are systems capable of distilling 20–60 L batches of materials. These are not necessarily tailored to a particular solvent system, however, so an off-the-shelf fractional distillation system may not be effective if the solvents being separated are too close in boiling point. Larger systems are typically custom built to the customer’s specific needs. Accordingly, attention should be paid to the chemistry employed in gel formation as well as to transfer solvents required for drying when designing a process for making an aerogel in order to avoid a need for fractional distillation if at all possible. This includes considering not only the solvent system used to synthesize the gel but also catalysts, reactive monomers, and by-products of reactive monomers used in gel formation. It is possible for a solvent recycling system to be directly coupled to the solvent exchange vessel, wherein the mixture of target solvent and pore fluid is fed into a recycling system that separates the components and supplies purified target solvent back to the solvent exchange vessel. The use of a Soxhlet extractor to achieve this by distillation has been demonstrated in the literature [41–43]. In the more specific case of solvent exchange into liquid or supercritical CO2, pressure separation can be used to remove alcohol from the drying vessel effluent, and the purified CO2 can be reliquified and pumped back to the drying vessel [44].

3.4.4

Additional Safety Considerations

Solvent exchange often involves large volumes of solvents and vapors that can pose health and/or flammability hazards if not properly controlled. Vapor should be controlled by

ventilation or containment for the sake of both minimizing exposure and reducing explosion/ignition risk. Flammability risks should also be mitigated. Three elements are necessary for combustion: fuel, oxidizer, and ignition source. Efforts to remove at least two of these three elements of the so-called fire triangle should be made when possible. If only one of these elements can be removed, there must be no chance that the third element can exist (e.g., by using ATEX- or NEC-rated Class 1-Division 1 equipment designed for hazardous locations when flammable vapor mixtures may be present in order to entirely prevent an ignition source). Hazardous area class ratings for equipment are based on the nature of the flammability hazard and the likelihood of the hazard occurring. Steps such as blanketing solvent tanks in nitrogen (inerting) and using explosion-proof equipment can be taken to limit risks. In the case of nitrogen blanketing, this can create the additional hazard of suffocation risk. Grounding of equipment is necessary when pumping flammable liquids that are non-conducting, as charge buildup can occur on tubing or tanks due to the friction of the flowing liquid if not properly grounded [45]. Consult chemical engineering and process safety professionals for further guidance about safe design and implementation of such equipment.

3.5

Liquid-Phase Functionalization

In addition to removing contaminants and unwanted solvents from gels, the solvent exchange process is sometimes used to infiltrate a gel with a particular material or reactant that will react with the backbone of the gel or become trapped in the gel’s pore network. One common example is hydrophobization of silica aerogels. In a typical process, a hydrophobe such as hexamethyldisilazane (HMDS or HMDZ) or hexamethyldisiloxane (HMDSO) is diffused into a silica gel, sometimes as a dilution in an organic solvent such as alcohol. The gel is then optionally heated to 60  C, at which point the hydrophobe reacts with surface silanols throughout the gel, replacing them with hydrophobic, noncondensable trimethylsiloxy groups [46]. See ▶ Chap. 65 for a representative recipe of hydrophobization of silica gels. Another example is polymer crosslinking of inorganic oxide gels to improve mechanical strength and stiffness, first demonstrated by Leventis [25]. In this process, a reactive polymer such as a triisocyanate is diluted in a nonreactive solvent, such as acetonitrile, and diffused into the pore network of the oxide gel to form a conformal polymer skin over the bones of the aerogel backbone. Examples of polymer crosslinking via solvent exchange are seen in Katti et al. and Meador et al. [47, 48] wherein diluted isocyanate is diffused into a silica gel that has been functionalized with reactive amino groups on its surface through cogelation of

3

Gel-Phase Processing and Solvent Exchange

TMOS with 3-aminopropyltriethoxysilane (APTES). The isocyanate reacts with both surface amino groups on the silica backbone to form a covalent bond to the oxide network and with water in the pore fluid to form amine groups that subsequently react with unreacted isocyanate groups on neighboring molecules or monomers in solution to form urea linkages. Consequently, a conformal polyurea coating forms over the internal surfaces of the silica gel, resulting in mechanical reinforcement of the otherwise brittle silica aerogel with a 100 increase in compressive strength for only a threefold increase in density relative to the native silica gel. Very recently a similar approach has been applied to alginate gels [49]. See ▶ Chap. 65 for representative recipes for polymer-crosslinking of gels and suitable precursors. Another application of the use of solvent exchange to introduce functionality is seen in Sassin et al. in the production of FeOx-carbon nanofoam materials for use as electrode structures for aqueous electrochemical capacitors. In this process, aerogel-like carbon nanofoams are soaked in aqueous K2FeO4 which then reacts via a redox reaction to form a conformal coating of nanoscale iron oxide over the carbon nanofoam [50].

3.6

Solvent Exchange into Liquid Carbon Dioxide

Supercritical drying of gels using carbon dioxide typically involves replacing the solvent in the pores of the gel with supercritical carbon dioxide [51]. However, in some cases, it may be preferable to separate the solvent exchange and supercritical extraction steps of this process by first solvent exchanging into liquid carbon dioxide and then, in a second step, raising the liquid-phase carbon dioxide inside the gel’s pore network past its critical point [52]. The most common reason to use this approach rather than directly exchanging/ extracting with supercritical-phase carbon dioxide is to reduce the stress induced within the gel due to solvent exchange in support of mitigating cracking. Although supercritical drying avoids problems associated with capillary stresses that arise during evaporative drying, stresses can still arise due to excess volume changes as solvents mix, osmotic pressure [53], and/or flow/pressure caused by changes in pore fluid density. An abrupt change in the density of the pore fluid puts stress on the gel as the fluid tries to flow through its low-permeability gel network [54]. Carbon dioxide is a particularly low-density liquid, and mixtures of carbon dioxide and common solvents undergo significant, often non-monotonic, changes in density as a function of composition. This makes the process of solvent exchange particularly hazardous for delicate gels, as stresses are more likely to be induced. Furthermore, the density of carbon dioxide is strongly dependent on pressure and temperature, which

89

makes it possible for thermal gradients within the autoclave to induce stress on the gel network as well [55]. Exchanging into liquid carbon dioxide prior to supercritical extraction can be an effective way to reduce stresses due to solvent exchange since the diffusion coefficient of liquid carbon dioxide in a solvent is significantly lower than the diffusion coefficient of supercritical carbon dioxide. Doing so slows down the process of diffusive mass transfer that occurs in the gel and reduces the severity of species gradients, thereby reducing density and pressure gradients. A typical process for liquid carbon dioxide solvent exchange entails loading wet gels into a pressure vessel, filling the vessel with organic solvent (ideally the same solvent as the pore fluid of the gels), and then sealing the vessel. Liquid carbon dioxide is then introduced into the pressure vessel, displacing some or all of the excess organic solvent through a drain valve. The carbon dioxide can be supplied by pumping liquefied carbon dioxide into the vessel or by cooling the vessel relative to the carbon dioxide supply such that liquid carbon dioxide siphons from the tank into the pressure vessel via a vapor-pressure differential. This purging process is repeated several times, dwelling in between drains to allow the mixture to equilibrate. The amount of new carbon dioxide that is introduced in each cycle can be varied in order to further modulate the severity of composition-gradient effects resulting from the solvent exchange. This process is analogous to the stepwise solvent exchange discussed elsewhere in this chapter, wherein the composition of the solvent bath is gradually transitioned from pore-fluid-rich to target-solvent-rich, only this time performed in a pressure vessel. The number of exchanges and dwell times between exchanges likewise depends on the gel geometry (characteristic length for diffusion) and the amount of fresh carbon dioxide introduced with each cycle. See ▶ Chap. 65, Sect. 65.3.1 for an example of how to solvent exchange gels into liquid carbon dioxide. The liquid carbon dioxide solvent exchange process is repeated until the effluent from the pressure vessel is solvent-free, indicating that all of the excess solvent and the solvent in the gel has been sufficiently removed and replaced with carbon dioxide. This endpoint can be determined by monitoring the composition of the effluent (e.g., using an IR hydrocarbon detector) or simply by evaluating the collected effluent for residual solvent smell [35]. The effluent can also be collected in order to condense, recover, and weigh any recoverable solvent in support of measuring the total solvent that has been extracted from the vessel. In practice, for a vessel in which carbon dioxide is not recycled with a closed-loop system, it is not atypical to only recover about 80% of the solvent originally placed in the vessel in condensed form by the time the gels are ready for supercritical drying, since much of the solvent is vaporized when carbon dioxide is removed from the vessel.

3

90

3.7

J. S. Griffin et al.

Conclusions

In this chapter, we have reviewed important unit operation pertaining to gel-phase processing including molding, demolding, aging, solvent exchange, and liquid-phase functionalization – what they are, when they are used, how they are performed, and practical considerations. As we have discussed, solvent exchange in particular is an important part of gel processing and thus aerogel production, facilitating effective drying and enabling functionalization of surface chemistry, installation of material functions, and tailoring of material properties. Solvent exchange can also have a significant effect on the structure and properties of the final material. Consideration should be given when designing a solvent exchange procedure to maximize material outcome and minimize volume of solvent required. Finally, liquid-phase functionalization of gels can be achieved through solvent exchange into baths of functional reagents to enable tailoring of the materials properties of derivative aerogels.

References 1. Brinker, C.J., Scherer, G. (eds.): Sol-Gel Science: The Physics and Chemistry of Sol-Gel Processing, 1st edn. Academic Press, San Diego (1990) 2. Scherer, G.W.: Aging and drying of gels. J. Non-Cryst. Solids. 100, 77–92 (1988) 3. Mulik, S., Sotiriou-Leventis, C., Leventis, N.: Time-efficient acidcatalyzed synthesis of resorcinol-formaldehyde aerogels. Chem. Mater. 19(1), 6138–6144 (2007) 4. Lee, J.K., Gould, G.L., Rhine, W.: Polyurea based aerogel for a high performance thermal insulation material. J. Sol-Gel Sci. Technol. 49, 209–220 (2009) 5. Incropera, F.P., Dewitt, D.P., Bergman, T.L., Levine, A.S.: Fundamentals of Heat and Mass Transfer, 6th edn. Wiley, New York (2006) 6. Nellis, G., Klein, S.: Heat Transfer, 1st edn. Cambridge University Press, New York (2009) 7. Bird, R.B., Stewart, W.E., Lightfoot, E.N.: Transport Phenomena, 2nd edn. Wiley, New York (2002) 8. Estrada-Baltazar, A., De Leon-Rodriguez, A., Hall, K.R., RamosEstrada, M., Iglesias-Silva, G.A.: Experimental densities and excess volumes for binary mixtures containing propionic acid, acetone, and water from 283.15 K to 323.15 K at atmospheric pressure. J. Chem. Eng. Data. 48, 1425–1431 (2003) 9. Lee, B.E., Prausnitz, J.M., O’Connell, J.P.: The Properties of Liquids and Gases, 5th edn. McGraw-Hill, New York (2001) 10. Wilke, C.R., Chang, P.: Correlation of diffusion coefficients in dilute solutions. AICHE J. 1, 264–270 (1955) 11. Vignes, A.: Diffusion in binary solutions: variation of diffusion coefficient with composition. Ind. Eng. Chem. Fundam. 5(2), 189–199 (1966) 12. Tyn, M.T., Calus, W.F.: Temperature and concentration dependence of mutual diffusion coefficients of some binary liquid systems. J. Chem. Eng. Data. 20(3), 310–316 (1975) 13. Tyn, M.T., Calus, W.F.: Diffusion coefficients in dilute binary liquid mixtures. Engineering. 20(1), 2–5 (1975) 14. Behr, W., Haase, A., Reichenauer, G., Fricke, J.: Self and transport diffusion of fluids in SiO2 alcogels studied by NMR pulsed gradient

spin echo and NMR imaging. J. Non-Cryst. Solids. 225, 91–95 (1998) 15. Behr, W., Behr, V.C., Reichenauer, G.: Self diffusion coefficients of organic solvents and their binary mixtures with CO2 in silica alcogels at pressures up to 6MPa derived by NMR pulsed gradient spin echo. J. Supercrit. Fluids. 106, 50–56 (2015) 16. Scherer, G.W.: Influence of viscoelasticity and permeability on the stress response of silica gel. Langmuir. 12(5), 1109–1116 (1996) 17. Gross, J., Scherer, G.W.: Dynamic pressurization: novel method for measuring fluid permeability. J. Non-Cryst. Solids. 325, 34–47 (2003) 18. Karamanis, G., Dinh, H., Waisbord, N., Hodes, M.: Effects of suction and spillage on supercritical carbon dioxide-based drying of aerogels. In: Proceedings of the 16th International Heat Transfer Conference (IHTC-16), Beijing, 2018, pp. 5909–5917 19. Bueno, A., Selmer, I., Raman, S.P., Gurikov, P., Loelsberg, W., Weinrich, D., Fricke, M., Smirnova, I.: First evidence of solvent spillage under subcritical conditions in aerogel production. Ind. Eng. Chem. Res. 57(26), 8698–8707 (2018) 20. Kistler, S.S.: Coherent expanded aerogels. J. Phys. Chem. 36, 52–64 (1932) 21. Gurikov, P., Raman, S.P., Griffin, J.S., Steiner III, S.A., Smirnova, I.: 100th Anniversary: solvent exchange in the processing of biopolymer aerogels: current status and open questions. Ind. Eng. Chem. Res. 58(40), 18590–18600 (2019) 22. Subrahmanyam, R., Gurikov, P., Dieringer, P., Sun, M., Smirnova, I.: On the road to biopolymer aerogels—dealing with the solvent. Gels. 1(2), 291–313 (2015) 23. Takeshita, S., Sadeghpour, A., Malfait, W.J., Konishi, A., Otake, K., Yoda, S.: Formation of nanofibrous structure in biopolymer aerogel during supercritical CO2 processing: the case of chitosan aerogel. Biomacromolecules. 20(5), 2051–2057 (2019) 24. Burfield, D.R., Smithers, R.H.: Desiccant efficiency in solvent and reagent drying. 7. Alcohols. J. Org. Chem. 48, 2420–2422 (1983) 25. Leventis, N., Sotiriou-Leventis, C., Zhang, G., Rawashdeh, A.-M. M.: Nanoengineering strong silica aerogels. Nano Lett. 2(9), 957–960 (2002) 26. Heath, L., Thielemans, W.: Cellulose nanowhisker aerogels. Green Chem. 12, 1448–1453 (2010) 27. Heath, L., Zhu, L., Thielemans, W.: Chitin nanowhisker aerogels. ChemSusChem. 6, 1–9 (2013) 28. Pircher, N., Veigel, S., Aigner, N., Nedelec, J.M., Rosenau, T., Liebner, F.: Reinforcement of bacterial cellulose aerogels with biocompatible polymers. Carbohydr. Polym. 2014(13), 505–513 (2014) 29. Veronovski, A., Novak, Z., Knez, Z.: Synthesis and use of organic biodegradable aerogels as drug carriers. J. Biomater. Sci. Polym. Ed. 23, 873–886 (2012) 30. Ghafar, A., Gurikov, P., Raman, S., Parikka, K., Smirnova, I., Mikkonen, K.S.: Mesoporous guar galactomannan based biocomposite aerogels through enzymatic crosslinking. Composites Part A. 94, 93–103 (2017) 31. Conzatti, G., Faucon, D., Castel, M., Ayadi, F., Cavalie, S., Tourrette, A.: Alginate/chitosan polyelectrolyte complexes: a comparative study of the influence of the drying step on physicochemical properties. Carbohydr. Polym. 172(15), 142–151 (2017) 32. Ganesan, K., Ratke, L.: Facile preparation of monolithic κ- carrageenan aerogels. Soft Matter. 10, 3218–3224 (2014) 33. Tanaka, T., Sun, S.T., Nishio, I., Swislow, G., Shah, A.: Phase transitions in ionic gels. Phys. Rev. Lett. 45(20), 1636–1639 (1980) 34. Panagiotopoulos, A.: High pressure phase equilibria: experimental and Monte Carlo simulation studies. PhD thesis, MIT, Cambridge (1982) 35. Griffin, J.S., Mills, D.H., Cleary, M., Nelson, R., Manno, V.P., Hodes, M.: Continuous extraction rate measurements during supercritical CO2 drying of silica alcogel. J. Supercrit. Fluids. 94, 38–47 (2014)

3

Gel-Phase Processing and Solvent Exchange

36. De Vries, A., Hendriks, J., Van Der Linden, E., Scholten, E.: Protein oleogels from protein hydrogels via a stepwise solvent exchange route. Langmuir. 31(51), 13850–13859 (2015) 37. Okay, O., Akkan, U.: Heterogeneities in polyacrylamide gels immersed in acetone-water mixtures. Phys. Rev. Lett. 370, 363–370 (1998) 38. Williams, J., Meador, M.A.B., McCorkle, L., Mueller, C., Wilmoth, N.: Synthesis and properties of step-growth polyamide aerogels cross-linked with triacid chlorides. Chem. Mater. 26, 4163–4171 (2014) 39. Meador, M.A.B., McMillon, E., Sandberg, A., Barrios, E., Wilmoth, N.G., Mueller, C.H., Miranda, F.A.: Dielectric and other properties of polyimide aerogels containing fluorinated blocks. ACS Appl. Mater. Interfaces. 6, 6062–6068 (2014) 40. Chidambareswarapattar, C., Larimore, Z., Sotiriou-leventis, C., Mang, J.T., Leventis, N.: One-step room-temperature synthesis of fibrous polyimide aerogels from anhydrides and isocyanates and conversion to isomorphic carbons. J. Mater. Chem. 20, 9666–9678 (2010) 41. Li, Y., Grishkewich, N., Liu, L., Wang, C., Tam, K.C., Liu, S., Mao, Z., Sui, Z.: Construction of functional cellulose aerogels via atmospheric drying chemically cross-linked and solvent exchanged cellulose nanofibrils. Chem. Eng. J. 366, 531–538 (2019) 42. Talley, S.J., AndersonSchoepe, C.L., Berger, C.J., Leary, K.A., Snyder, S.A., Moore, R.B.: Mechanically robust and superhydrophobic aerogels of poly(ether ether ketone). Polymer. 126, 437–445 (2017) 43. Lázár, I., Fábián, I.: A continuous extraction and pumpless supercritical CO2 drying system for laboratory-scale aerogel production. Gels. 2(4), 26 (2016) 44. Bommel, M.J., de Haan, A.B.: Drying of silica aerogel with supercritical carbon dioxide. J. Non-Cryst. Solids. 186, 78–82 (1995) 45. Crowl, D.A., Louvar, J.F.: Chemical Process Safety: Fundamentals with Applications. PTR Prentice Hall, New Jersey (2011) 46. Yokogawa, H., Yokoyama, M.: Hydrophobic silica aerogels. J. Non-Cryst. Solids. 186, 23–29 (1995) 47. Katti, A., Shimpi, N., Roy, S., Lu, H., Fabrizio, E.F., Dass, A., Capadona, L.A., Leventis, N., Cle, V.: Chemical, physical, and mechanical characterization of isocyanate cross-linked amine-modified silica aerogels. Chem. Mater. 18(2), 285–296 (2006) 48. Meador, M.A.B., Capadona, L.A., McCorkle, L., Papadopoulos, D.S., Leventis, N.: Structure-property relationships in porous 3D nanostructures as a function of preparation conditions: isocyanate cross-linked silica aerogels. Chem. Mater. 19(2), 2247–2260 (2007) 49. Paraskevopoulou, P., Smirnova, I., Athamneh, T., Papastergiou, M., Chriti, D., Mali, G., Čendak, T., Chatzichristidi, M., Raptopoulos, G., Gurikov, P.: Mechanically strong polyurea/polyurethane-crosslinked alginate aerogels. ACS Appl. Polym. Mater. 2(5), 1974–1988 (2020) 50. Sassin, M.B., Mansour, A.N., Pettigrew, K.A., Rolison, D.R., Long, J.W.: Electroless deposition of conformal nanoscale iron oxide on carbon nanoarchitectures for electrochemical charge storage. ACS Nano. 4(8), 4505–4514 (2010) 51. Bommel, M.J., de Haan, A.B.: Drying of silica gels with supercritical carbon dioxide. J. Mater. Sci. 29, 943–948 (1994) 52. Tewari, P.H., Hunt, A.J., Lofftus, K.D.: Ambient-temperature supercritical drying of transparent silica aerogels. Mater. Lett. 3(9–10), 363–367 (1985)

91 53. Wang, S.Y., Wu, N.L.: Tin oxide gel shrinkage during CO2 supercritical drying. J. Non-Cryst. Solids. 224(3), 259–266 (1998) 54. Scherer, G.W.: Stress and strain during supercritical drying. J. Sol-Gel Sci. Technol. 90, 8–19 (2019) 55. Woignier, T., Phalippou, J., Despetis, F., Calas-Etienne, S.: Aerogel processing. In: Klein, L., Aparicio, M., Jitianu, A. (eds.) Handbook of Sol-Gel Science and Technology: Processing, Characterization and Applications, pp. 985–1011. Springer, Cham (2018)

Justin S. Griffin is the Chief Operating Officer of Aerogel Technologies, LLC, in Boston, MA, USA. He holds an MS and BS in mechanical engineering from Tufts University and is co-inventor on numerous patents pertaining to production of mechanically strong aerogels. Griffin is recognized in the aerogel community for his contributions to mass transport phenomena in nanoporous media and manufacturing technologies for production of monolithic aerogels.

Ryan T. Nelson is the Chief Product Officer of Aerogel Technologies, LLC, in Boston, MA, USA. Nelson holds an MS and BS in mechanical engineering from Tufts University where he researched mass transfer kinetics in supercritical CO2 drying of aerogel. Later work with Aerogel Technologies focused on development of polymeric aerogels and methods for manufacture at scale.

3

92

Pavel Gurikov received his PhD from the Mendeleev University of Technology (Russia, 2010). He worked as a lecturer in the Mendeleev University of Technology and Technical University of Communication and Informatics (Russia, 2008–2012). He joined Hamburg University of Technology (Germany) as postdoctoral researcher and group leader in 2012. In 2017, he was appointed as junior professor with a focus on development and modelling of novel nanoporous materials.

Irina Smirnova received her PhD from the Technical University of Berlin (Germany, 2002). She was a postdoctoral researcher and a group leader at the University of Erlangen-Nürnberg (Germany, 2002–2008). Since 2008, she is a full professor in Hamburg University of Technology being head of the Institute of Thermal Separation Processes. Her research interests embrace nanoporous materials, separation processes (including at high pressure), and biorefinery.

J. S. Griffin et al.

Stephen A. Steiner III is the President, CEO, and founder of Aerogel Technologies, LLC, a leading aerogel manufacturer. Steiner holds a PhD in materials chemistry and engineering from MIT’s Department of Aeronautics and Astronautics, an SM in materials science and engineering from MIT, and a BS in chemistry course from the University of Wisconsin—Madison. He is an accomplished nanomaterials researcher, with expertise in aerogels, nanocarbons, and aerospace materials.

4

Supercritical Drying of Aerogels Raman Subrahmanyam, Ilka Selmer, Alberto Bueno, Dirk Weinrich, Wibke Lo¨lsberg, Marc Fricke, Sohajl Movahhed, Pavel Gurikov, and Irina Smirnova

Contents 4.1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

94

4.2

Fundamentals of Supercritical Fluid Drying . . . . . . . . . . .

94

4.3

The Evolution of Supercritical Drying: Inspired by Supercritical Extraction with CO2 . . . . . . . . . . . . . . . . . . . . . .

95

4.4 4.4.1 4.4.2 4.4.3 4.4.4

4.5

Practical Implementation and Mathematical Modeling of Supercritical Drying for Monoliths and Particles . . . 97 Gel Loading and Pressurization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 Minimum Solvent Concentration Required to Start Supercritical Drying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Is It Necessary to Stay in a Supercritical Region? The Phenomenon of Solvent Spillage . . . . . . . . . . . . . . . . . . . . . . . . . . 102 Mathematical Modeling of Supercritical Drying Kinetics for Monolithic and Particulate Gels . . . . . . . . . . . . . . . . . . . . . . . 104

R. Subrahmanyam (*) · A. Bueno · D. Weinrich · M. Fricke Aerogel-it GmbH, Osnabrueck, Germany e-mail: [email protected]; [email protected]; [email protected]; [email protected] I. Selmer Institute of Thermal Separation Processes, Hamburg University of Technology, Osnabrueck, Germany e-mail: [email protected] W. Lölsberg BASF SE, Ludwigshafen am Rhein, Germany e-mail: [email protected] S. Movahhed Covestro Deutschland AG, Leverkusen, Germany e-mail: [email protected] P. Gurikov Laboratory for Development and Modelling of Novel Nanoporous Materials, Hamburg University of Technology, Hamburg, Germany

4.4.5 How to Determine Process Conditions and Drying Time: Experimental and Theoretical Views . . . . . . . . . . . . . . . . . . . . . . 110 4.4.6 Observations on Depressurization Rates . . . . . . . . . . . . . . . . . . 112 Practical Aspects, Scaling Up, and Economic Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5.1 Typical Laboratory-Scale Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5.2 Important Aspects to Consider when Scaling up . . . . . . . . . . 4.5.3 Commercial Viability of Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . 4.6

112 113 115 116

Conclusions and Future Perspectives . . . . . . . . . . . . . . . . . . . 117

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

Abstract

In this chapter, we discuss the process of supercritical drying as it relates to the production of aerogels. What was once a heuristic technique typically performed using overly conservative process rules can now be quantitatively designed and efficiently implemented. Herein, thermodynamic and kinetic considerations of supercritical drying are reviewed along with practical aspects and process-engineering considerations at both the lab and industrial scale. Finally, considerations for process scaleup and further development of aerogel production processes are presented. Keywords

Supercritical CO2 drying · Biopolymers · Thermodynamic phase diagrams · Process modeling · Hydrogel · Solvent spillage · Solvent exchange · Mass transport · Depressurization · Lab scale equipment · Process scale up · Commercial viability

Institute of Thermal Separation Processes, Hamburg University of Technology, Osnabrueck, Germany e-mail: [email protected] I. Smirnova Institute of Thermal Separation Processes, Hamburg University of Technology, Hamburg, Germany e-mail: [email protected] © Springer Nature Switzerland AG 2023 M. A. Aegerter et al. (eds.), Springer Handbook of Aerogels, Springer Handbooks, https://doi.org/10.1007/978-3-030-27322-4_4

93

94

4.1

R. Subrahmanyam et al.

Introduction

4.2

Supercritical drying is a process used to remove the pore liquid from a gel under conditions free from capillary stresses to enable isolation of the gel’s solid component from its liquid component without collapsing its porous structure. Supercritical drying was the enabling invention that led to the development of the first aerogels by Kistler in the late 1920s and remains the gold standard for production of aerogels of all types today. In his early efforts at making aerogels, Kistler attempted to dry gels borne from aqueous media (i.e., hydrogels) by directly supercritically extracting the water from their pores. Because of the incredibly corrosive nature of supercritical water, however, Kistler found his gels were dissolved away rather than dried. To circumvent this, Kistler discovered he could instead solvent exchange gels into an organic solvent such as ethanol, which can be supercritically extracted without dissolving the gel’s solid component. A general scheme depicting Kistler’s approach is shown in Fig. 4.1. More information about the sol–gel process and solvent exchange can be found in Chapter Sol–Gel and ▶ Chap. 3, respectively. Herein, we focus on the final step of this scheme, namely, drying of gels with supercritical fluids. First, the fundamentals of supercritical drying are discussed. Phase diagrams illustrating supercritical drying from organic solvents (so-called high-temperature supercritical drying) and from carbon dioxide (so-called low-temperature supercritical drying) are presented along with a discussion of the range of possible process conditions for these two cases. Next, considerations regarding the overall production of aerogels are discussed in the context of the process steps that have the greatest impact on drying. Subsequently, we present modeling of the supercritical CO2 drying process and discuss how to choose suitable drying conditions and times based on a gel’s solvent composition, shape, and dimensions. Lastly, we address practical aspects of supercritical drying along with considerations for scale-up.

The difference between ambient-pressure drying (e.g., evaporative drying in an oven or via spray drying) and supercritical fluid drying can be appreciated from the phase diagram for a single-component liquid (in this case, the pore liquid of a gel) (Fig. 4.2). During ambient-pressure drying, liquid is evaporated into a gas by increasing its temperature and the resulting gas is continuously removed from the system. Since two phases are involved, forces attributable to interfacial tension (in this case, surface tension) build within the gel matrix. The differential capillary pressure (ΔP) that results at a liquid-gas interface whose meniscus forms an angle θ with parallel capillary walls due to a liquid surface tension (γ) for a given capillary radius r is given by the Young-Laplace equation (Eq. 1). ΔP ¼

Hydrogel

2γ cos θ r

ð1Þ

For a pore diameter of 10–100 nm, the capillary pressure at the liquid-gas interface is enormous, ranging from 40 to 100 MPa (~400–1000 atm) [1]. Forces of this magnitude are too strong for most gel matrices, and therefore ordinary evaporative drying results in pore collapse and densification of the gel into a low-porosity solid called a xerogel [2]. Formation of a two-phase boundary can, however, be circumvented by instead removing solvent from the gel at conditions above the solvent’s critical temperature and pressure (i.e., the solvent’s critical point) (Fig. 4.2b). Doing so allows for the pore structure of the gel to be preserved upon drying, as only a single phase is ever present during the drying process, and thus capillary forces do not arise. This is the fundamental principle behind supercritical drying [3–5] and is referred to as direct supercritical drying.

Drying under CO2 flow

Solvent exchange

Gelation

Sol

Fundamentals of Supercritical Fluid Drying

Organogel

Aerogel

Fig. 4.1 Scheme depicting the typical steps employed in the production of aerogels using supercritical drying starting with an aqueous sol–gel system

4

Supercritical Drying of Aerogels

95

18

PTP

Solid

P

Pcr (L) Liquid

Critical point

(1) Gaseous

Triple point (4) TTP

P (MPa)

s.c. Fluid (2)

(3) T

Tcr

Fig. 4.2 Thermodynamic trajectories for evaporative and supercritical drying

The thermodynamic trajectories for evaporative and supercritical drying are compared in Fig. 4.2. In evaporative drying, the gel’s pore liquid (L1, assumed to be a pure solvent at atmospheric conditions) is heated to its boiling point (LV), vented at a constant temperature until virtually no solvent vapor remains in the gel system (Psolvent ! 0), and then cooled to room temperature (3). Note that air as a second component is not represented in this P-T diagram, and so the pressure corresponding to point (L1) should be understood as being at atmospheric pressure in contact with air. In direct supercritical drying (Fig. 4.2), the gel’s pore liquid (L1, assumed to be a pure solvent), initially in equilibrium with its own vapor phase, is brought to a supercritical state (1) by heating in a closed pressure vessel. Heating causes the pressure inside the closed vessel to increase according to the solvent’s vapor-liquid equilibrium (Trajectory: L1 ! LV! (1)), as the volume of the pressure vessel remains essentially constant. The supercritical fluid is then isothermally expanded back to atmospheric pressure [Trajectory: (1) ! (2)], normally by venting out of the pressure vessel) thereby dropping the partial pressure of solvent in the vessel to essentially zero (Psolvent ! 0). At this point, the system is cooled to room temperature [Trajectory: (2) ! (3)] and the vessel is opened causing solvent vapor in the pores of the resulting aerogel to equilibrate with air. Note that in this case no air is present in the system at any point during supercritical drying. In the case of a hydrogel (i.e., a gel containing water as its pore fluid), process conditions above 373  C (647 K) and 220 bar are required to perform direct supercritical drying. These are not only harsh operating conditions, but as mentioned, water also becomes a powerful solvent under these conditions [6] and disintegrates most gels [7]. Hence, Kistler’s insight to swap water for an organic solvent with a lower chemical activity at its critical point (such as ethanol or acetone) [8] became the basis for production of virtually all aerogels derived from hydrogels (such as biopolymer aerogels). Although Kistler initially employed solvent

16 14 Tewari (1985) Kistler (1932) 12 Approach Approach 10 8 Single phase Single phase CCO2 6 CEtOH Two phase Two phase 4 L L G G 2 0 0 50 100 150 200 250 T (°C)

Fig. 4.3 Pure-component phase diagrams for CO2 and ethanol plotted on the same diagram

exchange to reduce the chemical activity of the supercritical solvent in support of enabling drying rather than dissolution, the high process temperatures (Tc > 200  C) associated with supercritically extracting organic solvents limited the scope of this approach to production of aerogels of stable inorganic oxides for nearly 50 years. In addition, the flammability risks posed by such solvents further limited interest in supercritical drying. Accordingly, this technique is referred to as hightemperature supercritical drying (HTSCD). The advent of supercritical drying employing carbon dioxide (Tc ¼ 31.1  C and Pc ¼ 74 bar) instead of an organic solvent by Hunt et al. in 1983 (first reported in Tewari et al. in 1985) [9] provided a more benign approach for producing aerogels. In this approach, also referred to as low-temperature supercritical drying (LTSCD) or occasionally the Hunt process (Fig. 4.3), the organic pore liquid of a gel is replaced with liquid CO2 which is then extracted by direct supercritical drying above its critical point. This technique was later improved upon by Bommel et al. [10] in 1993, who used supercritical CO2 instead of liquid CO2 to replace the gel’s pore liquid, thus forming the basis for present-day indirect supercritical drying.

4.3

The Evolution of Supercritical Drying: Inspired by Supercritical Extraction with CO2

In their work drying gels with supercritical CO2 [10], Bommel et al. leveraged the fact that in the case of certain binary mixtures such as ethanol-CO2, for each temperature there exists a pressure above which the binary system always exists in the single-phase region. This pressure is called the mixture critical pressure (Pcrit). The mixture critical pressure is thus only dependent on temperature (Fig. 4.4a).

4

96

R. Subrahmanyam et al.

a

Locus of critical points — P-T plane

18

b

Locus of critical points — P-T-x space

16 14 15

10 8 CCO2

6

CEtOH

4

P (MPa)

P (MPa)

12

10 5 250

0 0

200 150

2

18 16

50

100 150 T (°C)

200

60 °C Two phase

10

50 °C 31 °C

8 6 240 °C

T (°C)

60 °C

10 8 6 4

40 °C 35 °C 25 °C Two phase

4 2

2 0

50 1 0

Operating line – P-x plane Single phase

12

180 °C

12 P (MPa)

d

Locus of critical points – P-x plane Single phase 140 °C

14

xCO2 (–)

250

Opera ting li ne

c

0

Pressure P (MPa)

0

100

0.5

0

0.2

0.4 0.6 xCO2 (–)

0.8

1

0

0

0.2

0.4 0.6 xCO2 (–)

0.8

1

Fig. 4.4 CO2 and ethanol mixture critical points on: (a) P-T diagram (composition xCO2 into the plane); (b) P-T-xCO2 three-dimensional diagram; and (c) P-xCO2 diagram (T into the plane). (d) A typical safe

supercritical drying operating line showing pressurization, drying, and depressurization steps

Additionally, mixture composition (xcrit) at the mixture critical pressure is fixed for a given temperature (Fig. 4.4b, c). By replacing a gel’s pore liquid with fresh CO2 above the mixture critical pressure (Fig. 4.4d), the process implemented by Bommel et al. always remains in the single-phase region, enabling production of crack-free silica aerogels [10]. A common mistake made when interpreting binary phase diagrams for indirect supercritical drying (Fig. 4.4a) arises when extrapolating supercritical drying logic applied to single-component phase diagrams (Fig. 4.2) to the binary phase diagram (Fig. 4.4a–c). In the case of a single-component phase diagram, the supercritical drying operation (operating line) can be represented on the P-T diagram itself. In the case of binary mixtures, however, the operating line is given by a P-x diagram (pressure-composition diagram, Fig. 4.4c), which can also represent the displacement of organic solvent

by CO2. The green points (curve) in Fig. 4.4a only indicate the locus of the mixture critical pressures. The third axis is the composition (xco2), which is perpendicular (out of the plane) to the P and T axes (Fig. 4.4b). Therefore, to derive a supercritical drying process using binary phase diagrams (Fig. 4.4d), one needs to first fix the operating temperature and then identify the mixture critical pressure Pcrit on the P-x diagram. With this, any pressure above the mixture critical point can be selected to create an operating line for performing indirect supercritical drying. The methodology of indirect supercritical drying is principally applicable for a wide variety of solvent-CO2 systems, including water [11]. The main problem with a water-CO2 binary mixture is that the mixture’s critical point, i.e., the point where the system exists as a single phase, occurs at very high temperatures (>270  C or 543 K) and pressures

4

Supercritical Drying of Aerogels

97

Methanol, (Ke et al. 2005) Methyl acetate, (Byun et al., 2006) Ethyl acetate, (Byun et al., 2006) Propyl acetate, (Byun et al., 2006) Butyl acetate, (Byun et al., 2006) Pentyl acetate, (Byun et al., 2006) Hexyl acetate, (Byun et al., 2006) Acetic anhydrid, (Calvo and de Loos, 2006) Cyclohexane, (Zhang et al., 2005) Ethylbenzene, (Zhang et al., 2005) Styrene, (Zhang et al., 2005) n-Octane, (Yu et al., 2006) Methyl-t-butyl ether, (Chester and Haynes, 1997) Ethyl acetate, (Chester and Haynes, 1997) Methyl ethyl ketone, (Chester and Haynes, 1997) Dioxane, (Chester and Haynes, 1997) m-Xylene, (Ng et al., 1982) Pentane, (Chen et al., 2003) Acetone, (Chen et al., 2003) Methylcyclohexane, (Nasrifar et al., 2003) Tetrahydropyran, (Nasrifar et al., 2003) Triethylamine, (Ribeiro and Aguiar-Ricardo, 2001) Toluene, (Reaves et al., 1998) Acetone, (Reaves et al., 1998) iso-Propanol, (Reaves et al., 1998) Benzene, (Reaves et al., 1998) Methylene chloride, (Reaves et al., 1998) Methanol, (Yeo et al., 2000) Ethanol, (Yeo et al., 2000) 1-Propanol, (Yeo et al., 2000) 1-Butanol, (Yeo et al., 2000)

20 18 16

Pcrit (MPa)

14 12 10 8 6 4 2

0

50

100

150

200

250

300

350

400

Tcrit (°C) Fig. 4.5 P-T diagram of various CO2-solvent binary systems [12–21]

(>300 bar). For many organic solvent-CO2 systems, the operating conditions required are milder, with maximum operating pressures that do not exceed 18 MPa (180 bar) independent of the operating temperature (Fig. 4.5). However, high-temperature processing conditions are generally not preferred, as most organic solvents are flammable and involve explosion risks. Such concerns can easily be circumvented by leveraging CO2, which is intrinsically nonflammable and enables milder process temperatures (typically 300–350 K). Thus, any gel system comprising an organic solvent for its pore liquid can be transformed into an aerogel employing CO2 by fixing the temperature and replacing the organic solvent with CO2 above the mixture critical pressure. In this regard, CO2 provides optionality for creating different drying profiles suited to the specific nature of the gel material being dried. Most commonly, CO2 is preheated above its critical temperature before being introduced into the system, allowing the entire supercritical drying process to be conducted above the corresponding mixture critical pressure of the system. Alternatively, a gel can be initially solvent exchanged into liquid CO2 and then heated to supercritical conditions for drying [9] provided that the critical pressure of the solvent mixture that results within the pores of the gel (which should be mostly CO2 after solvent exchange but will contain some residual organic solvent) is reached, and that at no time during drying is evaporation of

solvent from the pores of the gel allowed to occur. Later on, we will show the influence of different process conditions on the kinetics of supercritical drying and assess these differences quantitatively through process modeling.

4.4

Practical Implementation and Mathematical Modeling of Supercritical Drying for Monoliths and Particles

Supercritical CO2 drying typically involves the following three steps: (1) loading an organogel (i.e., a gel comprising an organic solvent in its pores) into an autoclave (i.e., a pressure vessel, not to be confused with a clinical autoclave which is rated for substantially less pressure) and pressurizing; (2) drying the gel under supercritical CO2 flow; and finally (3) depressurizing to ambient conditions. The following section addresses some common issues encountered during these steps.

4.4.1

Gel Loading and Pressurization

One of the key questions when loading a gel into a pressure vessel is whether or not to submerge the gel under liquid solvent prior to pressurization. In the methodology of

4

98

0.6

Acetone Ethanol DMSO

0.5 xevap.solv. (%)

Bommel [10], gels are completely submerged under a cold (10–20  C) organic solvent such as ethanol prior to pressurization. The primary reason given by Bommel for doing this (in the case of silica gels) is to minimize solvent evaporation from the gels during loading and pressurization, which can cause cracks to form. However, for gel systems that are less prone to cracking upon handling, such as those based on synthetic polymers or biopolymers, complete submersion under organic solvent may not be necessary. Most solvents used to compatibilize gels for supercritical CO2 drying (most frequently methanol, ethanol, acetone, acetonitrile, amyl acetate, or hexane) are organic in nature and, as such, typically exhibit high vapor pressures, although solvents with lower vapor pressures such as dimethyl sulfoxide (DMSO) can also be used. Because of this, once a gel sample is sealed inside an autoclave and heated, a fraction of the solvent in the gel will evaporate into the vapor phase to maintain vapor pressure equilibrium (calculable using the Antoine equation). Two important parameters that determine the amount of evaporation of solvent from the gel are (1) the temperature of the gel in the pressure vessel and (2) the amount of vacant volume in the vessel. Herein, autoclave loading (LA) is defined as the volume fraction occupied by gel (Vgel) relative to the total volume of the autoclave (VA). The remaining unused autoclave volume is thus termed the vacant volume (Vvac). Aside from cracking, significant solvent evaporation from wet gels is undesirable, as evaporation can cause the three-dimensional pore network of the gel to collapse, resulting in densification of the gel. The evaporation behavior of solvent from a gel as a function of temperature (T) is described below for three commonly used gel solvents: (a) acetone, (b) ethanol, and (c) DMSO. First, the effect of autoclave temperature on the evaporation of these different solvents was analyzed for a fixed autoclave loading (LA ¼ 45%). The results of this calculation are depicted in Fig. 4.6, where the fraction of solvent evaporated from the gel xevap.solv is plotted for different autoclave temperatures. One noticeable takeaway from this plot is that acetone evaporates to a greater extent than ethanol or DMSO. This is to be expected, as acetone has the lowest boiling point (56  C at 1 bar) and is the most volatile of these three solvents. Also, the vapor pressure of a solvent increases with increasing temperature, corresponding to an increase in the amount of solvent that evaporates from the gel. This said, the relative amount of evaporated solvent is ultimately very low for all three solvents: Even at 80  C (total P > 1 bar for acetone), less than 0.6 wt % of the acetone evaporates from the gel. This is because autoclave loading plays a far more significant role in determining the amount of solvent evaporation that takes place than temperature does, as loading determines the amount of vacant volume inside the autoclave, and if the vapor pressure of the solvent inside the autoclave reaches saturation quickly, further evaporation will not take place.

R. Subrahmanyam et al.

0.4 0.3 0.2 0.1 0

0

10

20

30

40 50 T (°C)

60

70

80

Fig. 4.6 Relative solvent evaporation for acetone, ethanol, and DMSO as a function of autoclave temperature for an autoclave loading (LA) of 45%

This effect becomes more appreciable when Fig. 4.7 is considered, which plots the absolute and relative amount of evaporated solvent at different autoclave loadings at 60  C. The absolute amount of evaporated solvent per liter of autoclave volume decreases linearly with increasing loading. This is a consequence of the ideal gas law, as the amount of evaporated solvent (nevap.solv) and vacant volume (Vvac) are proportional (nevap.solv/Vvac ! mevap.solv/Vvac ! mevap.solv/1/LA). The ideal gas law provides only a rough estimate, however, and so for a more accurate calculation, thermodynamic equations of state (EoS) such as Peng-Robinson (PR EoS) or Redlich-Kwong (RK EoS) should be used. The percentage of solvent evaporated from the gel xevap.solv, in contrast, exhibits a hyperbolic behavior (Fig. 4.7b) with respect to the autoclave loading LA. At very low loadings (LA < 5%), more solvent needs to evaporate from the gel to reach vapor pressure saturation due to a larger vacant volume (Vvac > 95%). Likewise, when the amount of solvent is low due to a lower gel loading, more solvent must evaporate from the gel in order to reach vapor pressure saturation. The simultaneous influence of autoclave loading and temperature on the evaporation of acetone from a gel is shown in Fig. 4.8. If a low loading of 5% is chosen, then the autoclave temperature plays a major role in the amount of solvent that evaporates from the gel. In such cases, additional solvent can be added to the autoclave to prevent solvent from evaporating from the gel, noting that the amount required depends on temperature. However, even for the most conservative cases, the y-intercept of Fig. 4.7a represents the maximum solvent required per liter of autoclave volume to prevent evaporation. Consequentially, gels do not need to be completely immersed in solvent in order to perform supercritical drying. Above 15% autoclave loading, the amount of solvent that evaporates from

Supercritical Drying of Aerogels

99

Fig. 4.7 Solvent evaporation for acetone, ethanol, and DMSO as a function of autoclave loading for an autoclave temperature of 50  C: (a) mass of solvent evaporated versus loading; (b) relative solvent evaporation versus loading

mevap.solv. (g/LAutoclave)

a Absolute solvent evaporation 1.8 Acetone Ethanol 1.6 DMSO

90

70

1.2

60

1 0.8

40 30

0.4

20

0.2

10

20

40 LA (%)

60

80

4

50

0.6

0

Acetone Ethanol DMSO

80

1.4

0

Fig. 4.8 Solvent evaporation dependence on autoclave loading and autoclave temperature for acetone. The operating conditions modeled (5% < LA < 80% and 20  C < TA < 80  C) are representative of a typical loading process

b Relative solvent evaporation

xevap.solv. (%)

4

0 10–2

100 log10 (LA (%))

102

10

xevap.solv. (%)

9 10

8

8

7

6

6 5

4

4 2 0 5 14

3 2 340

24 33 43

52 61 71 80 LA (%)

the gel is less than 3%, regardless of operating temperature even for volatile solvents like acetone. Thus, unless the gels in question are fragile and require dampening against CO2 influx during pressurization, solvent addition is not required.

320 300

TA (K)

1 0

Next we will evaluate the possibility of solvent evaporation during pressurization with CO2. After the pressure vessel is sealed, it is then pressurized with CO2. As the pressure increases, the gas phase can no longer be considered an ideal

100

R. Subrahmanyam et al.

Constant T, TA= 50.0 (°C)

a 0.5 mevap.solv. (g/LAutoclave)

Fig. 4.9 Solvent evaporation from a gel during pressurization for ethanol as a function of pressure at different autoclave loadings (a) and temperatures (b). PA is the pressure in the autoclave due to pressurization with CO2

Ideal gas pA =1.0 bar pA =2.0 bar pA =4.0 bar pA =6.0 bar pA =8.0 bar pA =10.0 bar

0.4 0.3 0.2 0.1 20

30

mevap.solv. (g/LAutoclave)

0.5

0 20

60

70

80

70

80

Ideal gas pA =1.0 bar pA =2.0 bar pA =4.0 bar pA =6.0 bar pA =8.0 bar pA =10.0 bar

30

gas, and thus more sophisticated equations of state must be applied. Here we will consider the Peng-Robinson equation of state applied to the binary ethanol-CO2 system (20  C < T < 80  C and 20% < L < 80%) using van der Waal’s mixing rules with a mixing parameter adapted from Adrian et al. [22]. It can be observed that for a given temperature, the evaporated solvent increases slightly with pressure but not significantly (Fig. 4.9a). In contrast (Fig. 4.9b), at a fixed autoclave loading, temperature still plays a significant role in determining the amount of solvent required to saturate the gas phase. The above calculations were made assuming that the liquidphase volume remains constant; however, in actuality this is usually not the case. More on this topic is covered in Sect. 4.3.

4.4.2

50 LA (%)

Constant loading, LA = 20.0%

b 1.5 1

40

Minimum Solvent Concentration Required to Start Supercritical Drying

An important question that arises during solvent exchange of hydrogels (i.e., in changing from water to organic solvent) is the extent of solvent exchange that is required prior to supercritical CO2 drying. Should the water in a hydrogel be

40

50 T (°C)

60

completely exchanged out for an organic solvent (such as ethanol), or is the presence of a small amount of water in the pore liquid still tolerable at the start of supercritical drying? The technical relevance of this question is demonstrated in the following example. Suppose a hydrogel of mass mH is solvent exchanged with an equal mass of organic solvent (mH) with the following assumptions: (1) The mass of the gel backbone is ignored; (2) there is no shrinkage of the gel during solvent exchange; (3) there is no interaction between the gel backbone and the solvent during the solvent exchange; and (4) ρhydrogel ¼ ρsolvent. When equilibrium is reached (stage 1), the final concentration of organic solvent in the gel should be 50 wt %. When the gel (50 wt %) is further exchanged into a second equal mass of fresh solvent (mH), the final solvent concentration in the gel after equilibration should be 75 wt % (stage 2). Table 4.1 shows that for higher and higher solvent concentration targets, the amount of solvent consumed increases drastically. This is an important aspect in considering solvent consumption and recycling in a production setting. Not knowing the concentration at which it is acceptable to start supercritical drying can result in enormous and unnecessary solvent consumption, which affects both operating expenses and

4

Supercritical Drying of Aerogels

101

Table 4.1 Solvent consumed versus solvent concentration in the gel

capital infrastructure. Furthermore, solvent usage influences solvent recovery options. For example, ethanol forms an azeotrope with water at 96 wt %. In the case that concentrations of ethanol higher than 96 wt % inside the gel are deemed necessary, solvent recovery operations such as azeotropic distillation, pressure swing distillation, membrane separation, or molecular sieving techniques need to be considered in addition to simple distillation, which can drive up capital and operating costs. In practice, however, solvent concentrations this high may not be needed. Therefore, determining the amount of water that can be tolerated inside the gel prior to supercritical drying is essential. This said, too high of a concentration of water inside the gel will result in shrinkage and deformation of the gel upon drying. The influence of water content in the pore liquid prior to supercritical drying on the final properties of the resultant aerogel is discussed for the case of alginate gels below. Nitrogen sorptimetry surface area measurements can serve as a sensitive assessment of the influence of various operating conditions on aerogel quality, especially for biopolymer aerogels. A plot of surface area versus the concentration of organic solvent in the gel when supercritical drying was commenced is shown in Fig. 4.10. This plot shows that supercritical CO2 drying of alginate gels containing ethanol as the primary solvent can proceed at ethanol concentrations as low as 91 wt % without causing significant deterioration of the surface area of the resulting aerogel. Concentrations lower than 91 wt % result in a gradual initial drop in surface area until 90 wt %, where approximately 90% of the maximum attainable surface area is still retained. Past that, a further 3% decrease in solvent concentration to 87 wt % results in a steep drop in surface area, resulting in retention of only 5% of the maximum achievable surface area. The response is completely different for DMSOcontaining gels, where solvent concentrations greater than 98 wt % are required prior to commencement of supercritical drying to obtain the maximum achievable surface area. One possible reason for the observed lower tolerable solvent concentration when using DMSO could be due to strong water-matrix interactions (from hydrogen bonding and/or covalent bonding) that inhibit the formation of a liquid

400

100 90 350 Ethanol 80 300 70 250 60 200 50 40 150 DMSO 30 100 20 50 10 0 0 84 86 88 90 92 94 96 98 100 Csolv. at s.c.-drying onset (wt-%)

Relative surface area (%)

Solvent concentration in the gel (wt %) 50 75 87.5 93.75 99.2 99.9 99.99

Specific surface area (m2 / g)

Pure solvent consumed mH 2  mH 3  mH 4  mH 7  mH 10  mH 14  mH

Fig. 4.10 BET surface area from nitrogen sorptimetry of alginate aerogels as a function of solvent concentration in the gel immediately prior to the onset of supercritical CO2 drying [23]. Two different solvents, ethanol and DMSO, were evaluated for solvent exchange to study the effect of solvent composition. Solvent concentration on the x-axis represents the composition of the pore liquid reached after solvent exchange

phase regardless of thorough solvent exchange. This water can then remain adsorbed on the aerogel backbone through subsequent processing and is only removed at higher temperatures. This said, in general the main reason for loss of surface area at lower gel solvent concentrations is due to the lack of a single-phase supercritical transformation. Besides influencing the properties of the resulting aerogels, water content influences the supercritical drying process itself. To consider the effects of the presence of water, we must transition our discussion from a binary solvent-CO2 diagram (Fig. 4.4) to a ternary solvent-waterCO2 diagram (Fig. 4.11). As discussed above, the underlying enabling principle in preparing a gel for supercritical drying is that one solvent can be replaced for another inside the gel without causing pore collapse as long as doing so only involves a single-phase transformation (in the case of a solvent exchange from water to ethanol, for example, the solvent replacement is also a single-phase transformation, i.e., a liquid-phase transformation). Thus, determining the minimum gel solvent concentration required for commencing supercritical drying can be determined from thermodynamic principles by considering the single-phase and two-phase regions of the relevant ternary diagram. Let us consider the net transformation of a hydrogel into an aerogel in which the three components involved are water, ethanol, and CO2. A ternary phase diagram based on published data [24] and the vapor-liquid equilibrium correlation reported by Duarte et al. [25] for this system are shown in Fig. 4.11.

4

102

R. Subrahmanyam et al.

Fig. 4.11 Water-ethanol-CO2 ternary phase diagram at 120 bar (12 MPa) and 60  C [24, 25]

91 wt.% Ethanol 9.0 wt.% Water

0 1 0,1

0,9

0,2

0,8

0,3

0,7

2O

H

0,6

0,5

O H6 C2

Single phase

0,4

0,2

0,5

0,6

0,1

0,4

0,7

0,3 0,9 0,2

Two phase

0,8 0,9 1 0

1

0

0,1

0,1

0,2

0,3

0,4

0,5

0,6

CO2

The change in the composition of the gel’s pore liquid during solvent exchange is represented by the blue operating line. The final concentration achieved in the pore liquid of the gel is 91 wt % ethanol, 9 wt % water, and 0 wt % CO2 (Fig. 4.11). The gel is then contacted with pure CO2 (100 wt % CO2), and supercritical drying is commenced (red operating line). It is observed that the operating line again represents a single-phase transformation (i.e., the line does not cut through the two-phase envelope), thus enabling preservation of the pore structure of the gel upon drying. As a result, supercritical drying can be initiated once the pore liquid has reached an ethanol concentration of 91%. No further solvent exchange is required. This is in agreement with the results presented above (Fig. 4.10). Thus, ternary diagrams for the applicable solvent-water-CO2 system are suitable to evaluate the minimal required solvent concentration from a thermodynamic point of view, and we encourage the aerogel community to use them. For many solvents, the corresponding data are available in the literature. However, it should be noted that the two-phase envelope of the water-ethanol-CO2 ternary system is very sensitive to temperature and pressure: Higher pressures (at constant

0,7

0,8

0,9

0 1 12 MPa, 60 °C

temperature) shrink the two-phase envelope, whereas higher temperatures (at constant pressure) expand the two-phase envelope [25].

4.4.3

Is It Necessary to Stay in a Supercritical Region? The Phenomenon of Solvent Spillage

Recent works investigating supercritical drying have shown the process to be controlled by a combination of convective and diffusive mass transport mechanisms [26–31]. However, one of the biggest challenges in describing the kinetics of supercritical drying is how to capture the initial stages of drying accurately. One reason for this has to do with the gas-expanded liquid (GXL) behavior of CO2-solvent systems [32, 33], wherein the solubility of the gas component (here CO2) in the liquid phase (e.g., acetone, ethanol, etc.) increases with the absolute pressure of the system. This results in a volume expansion of the liquid phase (Fig. 4.12). At the critical pressure of the mixture, which is determined by the system temperature, this volume

4

Supercritical Drying of Aerogels

103

expansion approaches infinity, and above the critical pressure, the system exists as a single phase, i.e., a supercritical fluid. The primary work analyzing the influence of CO2 in the volume expansion of organic solvents was reported by

1000 Acetonitrile

900

1,4-Dioxane

800

Ethyl acetate

Delta V (%)

700

DMF

600 500 400 300 200 100 0 0

1

2

3

4 5 p (MPa)

6

7

8

Fig. 4.12 Relative volume expansion of various solvents due to CO2 versus pressure. (Reproduced with permission from Kordikowski et al [34]. Reprinted from [34], with permission from Elsevier)

V ðT, PoÞ ΔV ¼ V Mixture T, P, xCO2  Ethanol V V Ethanol ðT, PoÞ

ð2Þ

where P0 represents ambient atmospheric pressure. Recently, it has been shown that by taking advantage of gas-expanded liquids, it is possible to remove a large quantity of solvent from a gel during pressurization without affecting the solid structure of the wet gel [35]. This was demonstrated with the following experiment. A gel monolith containing an organic solvent as its pore liquid was placed inside a highpressure view cell (Fig. 4.13 point A) and pressurized to given set points with CO2 (points B, C, and D in Fig. 4.13). As the autoclave was pressurized, CO2 dissolved into the solvent contained within the gel’s solid matrix. As a result, the volume of the gel’s liquid phase (solvent + CO2) increased; however, since the volume of the gel is constant, the expanded liquid phase spilled out of the gel’s solid matrix and collected at the bottom of the vessel, as observed in the case of points B, C, and D in Fig. 4.13. This phenomenon as applied to aerogel drying is referred to as solvent spillage. To maximize the mass transfer driving force, the surface of the gel was kept in contact with the gas phase. The spilled liquid can then be removed using a valve positioned at the bottom of the autoclave without damaging the gel, provided the autoclave pressure is kept constant. In this manner, 40–70% of the solvent in a gel can be extracted before reaching the critical point of the mixture. It should be noted that the remaining solvent contained within the gel

VLE, T=60 °C 12

Conventional CO2 sc-drying ach ppro a e illag t sp n e Solv

10 P (MPa)

Fig. 4.13 The phase diagram for a solvent-CO2 system below the mixture critical point exhibiting solvent spillage due to the formation of a gas-expanded liquid within the pores of the gel

Kordikowski et al. [34]. The relative volume expansion (ΔV/V) of the liquid phase at a given pressure (P) and temperature (T) can be mathematically quantified as follows:

8 6

D C B

4 2 0 0 0.1 0.2 tequilibration=1 h G L G

A 0.4 0.5 0.6 xCO2 (mol/mol)

0.3 G L1 L2

A: 60°C, 0.1 MPa B: 60°C, 5.0 MPa

G L1 L2

0.7

0.8

0.9 G L1 L2

C: 60°C, 8.0 MPa D: 60°C, 10.0 MPa

1

4

104

R. Subrahmanyam et al.

80 60 40 20 0

0

100 200 Time (min)

400 Relative volume expansion ('V/V0, %)

Pressure (bar)

100

100 90 80 70 60 50 40 30 20 10 0 300

Ethanol removed (wt.%) (outlet flow rate, g/min)

Pressure profile Ethanol removed Mixture Pc Outlet f low

120

200

100

0 0

Fig. 4.14 Example drying profile that incorporates solvent spillage to accelerate drying, temperature 308.15 K, gel composition 3-wt-% Ca-alginate. (Reproduced with permission from Bueno et al. [35]. Reprinted with permission from [35]. Copyright (2018) American Chemical Society)

pores can only be removed at increasingly higher pressures or by extracting above the critical pressure of the mixture (Fig. 4.14). Working at a constant subcritical pressure to extract the remaining solvent will result in pore collapse and gel shrinkage as doing so permits formation of a liquidgas interface within the pores of the gel. Solvent spillage represents the dynamic dissolution and mixing of CO2 with solvent in the pores of the gel during the pressurization step of the supercritical drying process, and the solvent that spills out of the gel can be considered a gas-expanded liquid. Since using this approach results in large amounts of solvent being removed within the two-phase region of the CO2-solvent system, it is important that the liquid in the gel be kept continually expanding at all times in order to prevent the liquid-gas interface from receding into the pores of the wet gel. Thus, the solvent spillage phenomenon seamlessly fits into the pressurization step of the supercritical CO2 drying process, as the relative volume expansion of the solvent monotonically increases with increasing pressure as the vessel pressurizes. The relative volume expansion of the liquid phase depends on the mole fraction of CO2 dissolved in it (Fig. 4.15). Thus, it is possible to control the amount of solvent spilled from the gel by controlling the pressure and temperature of the system. Furthermore, controlling the system temperature allows for controlling of the rate at which the liquid phase spills out of the gel. An important outcome from the above observation is that the autoclave pressurization step is now also a solvent removal step! The size of the pressurization

300

0.2

0.4 0.6 xco2, Mole fraction

0.8

1

Fig. 4.15 Volume expansion of the CO2-ethanol binary system for 291.15 K < T < 343.15 K and 0 < P < 250 bar. (Reproduced with permission from Bueno et al. [35]. Reprinted with permission from [35]. Copyright (2018) American Chemical Society)

system required can thus be reduced significantly since pressurization can proceed more slowly and since 40– 70 wt % of the initial solvent in the gel can be removed without continuous flushing of the autoclave. There is also a clear benefit with respect to CO2 consumption, as the amount of CO2 consumed per amount of solvent removed is reduced thanks to pressurization-induced solvent removal starting at subcritical pressures. In principle, any solvent in which CO2 is soluble is a viable candidate for solvent spillage provided the right process conditions exist.

4.4.4

Mathematical Modeling of Supercritical Drying Kinetics for Monolithic and Particulate Gels

The primary aim of using mathematical modeling for supercritical drying is to provide guidelines for how to select optimal drying conditions and to enable easier scaling up of supercritical drying processes. Furthermore, modeling provides a deeper understanding of mass transport limitations for different gel types and sizes at different process conditions. The supercritical drying models that have been developed to date are derived from simple mass transport equations [26, 28–31, 36–40] within the provisions of continuum mechanics to represent a supercritical system in terms of a set of mass, momentum, and energy conservation equations [27]. Due to

Supercritical Drying of Aerogels

105

the partial differential equations involved, most such models have to be solved numerically. To achieve good predictability, such models should be combined with a precise description of the properties of the physical mixture, which adds further complexity. In the following discussion, we describe models of supercritical drying kinetics that have been developed and use them to derive guidelines for how to choose the best conditions for supercritical drying of aerogels. Complete computational details are beyond the scope of this chapter; however, the reader is cordially invited to contact the authors in the case further advice is needed. During the supercritical drying process, the mixture composition within the gel body varies from nearly pure solvent (e.g., ethanol) to nearly pure supercritical CO2 (Fig. 4.16). In addition to mixture composition, the physical properties of the mixture depend highly on system temperature and pressure. In the case of ethanol as the solvent, we analyzed literature data for the CO2-ethanol system at relevant process conditions (T ¼ 305–350 K, P ¼ 8–20 MPa) [41] and note the following: • A good description of the mixture molar density ρm (also called the concentration c) was achieved using appropriate

modifications of the cubic Peng-Robinson Equation of state (PR EoS) for the pure substances (for CO2 the modified PR EoS by Hekayati et al. 2016 [42] and for EtOH a volume-translated PR EoS [43, 44]) combined with the asymmetric mixing rules developed by Stryjek and Vera [45]. • Due to missing experimental data, the binary diffusion coefficient DEtOH ,CO2 was calculated in most studies [27, 29–31, 38] using an empirical equation that takes into consideration the dependence of the mixture composition as proposed by Vignes [46]. This value is based on the corresponding diffusion coefficients at infinite dilution. The best fitting correlation for the diffusion coefficient of ethanol in supercritical CO2 at infinite dilution was Eq. 7 by Magalhães et al. [47] and for the diffusion of supercritical CO2 in ethanol at infinite dilution the correlation developed by Lito et al. [48]. • For the mixture viscosity ηf, the Arrhenius approach, which combines the viscosities of the pure substances and mixture composition, was applied [49]. The viscosity of pure supercritical CO2 can be described precisely by the correlation described by Fenghour et al. [50]. The correlation suggested by Laesecke and Muzny is slightly less precise in the pressure and temperature range investigated

a

b Drying/supercritical extraction

10 Liquid ation

4

Drying/sc. extraction

8

Critical point

6

suriz

6

Pres

Pressure (MPa)

8

10

Liquid + Gas

Depressurization

4

end

X CO2,gel

4

2

2 Gas

Gas

0 0

0.1

0.2

0.3

0.4 0.5 0.6 0.7 Mole fraction of CO2

0.8

0.9

1

0 0.95

0.96 0.97 0.98 0.99 Mole fraction of CO2 end

1

end

XEtOH,gel = 1 – X CO2,gel

Fig. 4.16 (a, b) Typical (p,x,y) diagram for ethanol-CO2 at 323 K (black line) and supercritical drying trajectory (arrows). Mixture critical pressure is marked by a red x. The gray area on the right shows the region from 0.95 to 1 mole fraction CO2 zoomed in. xend EtOH ,gel refers to

the ethanol composition in the gel when the drying process is finished (equal to the maximum on the phase boundary of the corresponding vapor-liquid (VL) line). (Data from [54]. Reprinted from [41], with permission from Elsevier)

4

106

R. Subrahmanyam et al.

but covers the viscosity of CO2 over a wide range of pressures and temperatures with high precision [51]. The viscosity of pure ethanol at elevated pressures can be described by the correlation described by Mamedov [52], which uses the viscosity at atmospheric pressure. For the discussion below, we used the correlation from Gonçalves et al. [53]. From both a cost and efficiency perspective, drying time is one of the most critical aspects of supercritical drying. We define the completion of the supercritical drying/extraction process as reaching the composition of the gel’s pore fluid at which no liquid will be formed upon depressurization. In Fig. 4.16, this slightly temperature-dependent composition (xend EtOH,gel ) is marked (blue cross) on the CO2-ethanol phase diagram at 323 K [41]. Thus, for this system the supercritical extraction time tse is defined as the time required to decrease the solvent mole fraction within the gel to at least the value that ensures conditions will stay within a singlephase region during isothermal depressurization back to ambient pressure (Fig. 4.16). Time required to pressurize and depressurize the system are not included in the defined supercritical extraction time tse. The overall drying time tdrying is thus the sum of the pressurization time (including solvent spillage), the supercritical extraction time tse, and the depressurization time. As models reported in the literature describe the supercritical extraction phase of the supercritical drying process, we will now analyze factors that affect the supercritical extraction time tse. Fig. 4.17 Model of supercritical drying of gel particles in a packed bed (z: axial coordinate of packed bed, z ¼ 0: top of packed bed, z ¼ L: bottom of packed bed, Δz: _ in,CO2 : CO2 slice of packed bed, m inlet flow rate, ṁout,CO2: CO2 outlet flow rate, ṁout,EtOH: ethanol _ volume flow outlet flow rate, V: rate, ρf: fluid density (mixture of CO2 and ethanol), r: radial coordinate of spherical gel particle, r ¼ 0: gel particle center, r ¼ R: gel particle radius). (Reprinted from [41], with permission from Elsevier)

As a representative example, the basic equations for a supercritical drying model considering spherical gel particles in a cylindrical packed bed are presented and analyzed here [41]. The corresponding supercritical drying setup is presented in Fig. 4.17: The mass transport is described by the continuity equation in the domain of spherical gel particles and in the domain of a cylindrical autoclave which interchange mass. Thus, three mass transport steps are described: the diffusional mass transport within the spherical gels Eq. (3), the mass transfer from the gel surface to the bulk fluid Eq. (9), and the mass transport in the bulk fluid from top to bottom of the autoclave Eqs. (5) and (6). To simplify the model, the following assumptions are made: • The entire volume of the autoclave is filled with gel particles. • Mixing of the bulk fluid in the radial direction of the autoclave is ideal. • An isothermal, quasi-stationary process with a constant mass flow of the solvent-CO2 mixture (m_ ) but varying bulk fluid density (ρf, depending on molar composition) and varying interstitial bulk fluid velocity u (depending on bulk fluid density) Eq. (6). The equations were written to model an initial organic pore liquid being extracted by supercritical CO2 (here ethanol, component 2). The solvent concentration within the bulk fluid is dependent on the time t and axial position z.

m⋅ inCO2 z=0 z

⋅ (V ⋅rf )|z

CO2 Diffusion r=0

Δz

EtOH r=R

z=L

⋅ (V ⋅rf )|z+Δ z

m⋅ out,CO2 + m⋅ out,EtOH

r

Transport in boundary layer

Transport in bulk fluid

4

Supercritical Drying of Aerogels

107

The concentration within a gel particle is dependent on the time t, the radial position within the gel r, and the axial position of the gel sphere within the autoclave z. Diffusion through the porous gel network is captured by the porosity εg and tortuosity τg of the gel body Eq. (3). The first part in Eq. (5) considers nonideal flow behavior in the packed bed using a flow- and thus time-dependent dispersion factor DL. Solvent moving from the inside of the gel particles to the bulk fluid acts as source term Eq. (4) in the bulk fluid Eq. (5). In this model, the structure of the flow itself is not taken into account as is done in CFD simulations (where conservation of momentum and energy is needed in addition to conservation of mass) since doing so is computationally intensive for small particles. However, nonideal flow behavior due to solvent dispersion, changes in flow velocity due to changing fluid composition and density over time, and height of the packed bed Eqs. (5) and (6) are considered. The usage of a specific mass transfer coefficient (β) correlation for spherical particles packed in a cylindrical bed includes the specific mass transfer in the packed bed due to the particle packing as well as the specific flow conditions (combined laminar and turbulent flow around the particles), which in this model are considered for spherical gel particles in a cylindrical packed bed. Mass transport in gel particles in radial direction:

u x2,f ðz, tÞ ¼

@ 2 Eg r ∙ ∙DEtOH,CO2 x2,g ðr, z, tÞ @r τg

@x2,g ðr, z, tÞ ∙cg x2,g ðr, z, tÞ ∙ @r

ð3Þ

Extracted ethanol from gel particles (acts as source term in bulk fluid): source2,f x2,g ðr, z, tÞ, x2,f ðz, tÞ ¼  @ ∙ @t

Np Aac ∙ψ∙L

r¼R 2

4∙π∙r ∙Eg ∙c2,g ðr, z, tÞ dr

ð4Þ

r¼0

Mass transport in bulk fluid/autoclave in axial direction: @c2,f ðz, tÞ @ 2 c2,f ðz, tÞ @ c2,f ðz, tÞ∙u x2,f ðz, tÞ  ¼ DL ðtÞ∙ @z @t @z2 þ source2,f x2,g ðr, z, tÞ, x2,f ðz, tÞ ð5Þ

ð6Þ

To solve this set of partial differential Eqs. (3–6), initial and boundary conditions are necessary (Eqs. 7–12): In the following drying process analysis, it is assumed that the pore liquid of the gel particles comprises pure ethanol at the beginning of the supercritical extraction time tse ¼ t ¼ 0. Initial and boundary conditions for mass transport within a gel particle: 8r, 8z, t ¼ 0 r ¼ 0, 8z, 8t r ¼ R, 8z, 8t 

c2,g ðr, z, 0Þ ¼ c2,g,0

ð7Þ

@c2,g ð0, z, tÞ ¼0 @r

ð8Þ

Eg @c2,g ðR, z, tÞ ∙D x ðR, z, tÞ ∙ τg EtOH ,CO2 2,g @r

¼ β x2,f ðz, tÞ, u x2,f ðz, tÞÞ∙ c2,g ðR, z, tÞ  c2,f ðz, tÞ ð9Þ Initial and boundary conditions for mass transport in the bulk fluid of the autoclave: 8z, t ¼ 0 z ¼ 0, 8t

@c2,g ðr, z, tÞ 1 ¼ 2 @t r

m_ Aac ∙ψ∙ρ f x2,f ðz, tÞ

c2,f ðz, 0Þ ¼ c2,f ,0

0¼ u x2,f ð0, tÞ ∙c2,f ð0, tÞ þ DL ðtÞ∙

z ¼ L, 8t

@c2,f ðL, tÞ ¼0 @z

ð10Þ @c2,f ð0, tÞ @z ð11Þ ð12Þ

Most of the kinetic studies performed to date have dealt with the supercritical extraction of silica gel monoliths [10, 26–31, 36, 39, 40, 55–57]. As discussed earlier, if we distinguish between inner mass transport (diffusion within the gel body) and outer mass transport (convection and diffusion outside the gel body including the mass transfer from the gel surface to the bulk fluid), the following can be concluded from the works published so far: • The inner mass transport plus the outer mass transport comprise the overall supercritical extraction kinetics. • The overall supercritical extraction kinetics of monolithic gels depends strongly on the thickness of the gel part. The common explanation for this is that the supercritical extraction process is limited by the diffusion length of the gel body [10, 26, 27, 29–31, 39]. Thus, inner mass transport is the slowest mass transport step and therefore the limiting step in the overall supercritical extraction kinetics. Accordingly, for monoliths it can be concluded that the supercritical drying process can be accelerated by increasing temperature and/or by decreasing in pressure,

4

R. Subrahmanyam et al.

which increases diffusion rates within the gel body (inner mass transport). • Outer mass transport can be accelerated by increasing CO2 flow rate, but such a faster outer mass transport only accelerates the overall supercritical extraction kinetics for large gel monoliths to a certain extent [27, 29, 30]. To summarize, for large gel monoliths, the overall supercritical extraction kinetics are primarily limited by diffusional inner mass transport due to large diffusion lengths and are influenced less by the outer mass transport. The interplay between inner and outer mass transport and process and gel parameters is not fully covered in the literature. For example, it is unclear at what size a gel is limited by outer mass transfer and thus process conditions rather than inner mass transfer because of a sufficiently short diffusion length. As a basis for such an analysis, we introduce the following two dimensionless numbers: the Biot number Bi [41] and the K1mean number [58]. • The Biot number Bi relates the inner mass transfer to the mass transfer coefficient β for a single gel body Eq. (14). In case of a sphere, the radius R is used as the characteristic length as expressed in Eq. (14). Keff is formally defined as the effective diffusion coefficient, which is calculated via the theoretical minimal supercritical extraction time tse, min of a spherical gel body of the radius R Eq. (13). K eff ðP, T Þ ¼ BiðP, T Þ ¼

τg 2 1 ∙ ∙R tse, min Eg

τg ∙β∙R Eg ∙K eff ðP, T Þ

ð13Þ ð14Þ

• K1mean represents the relation between inner mass transfer and outer mass transfer in the bulk fluid Eq. (15). It is defined as the mean value of all K1mean numbers calculated at each time interval k and space interval s. In the case of a cylindrical packed bed of spherical gel particles, the characteristic lengths are the sphere radius R and the length of the packed bed L. U is the superficial velocity, and DEtOH ,CO2 ,g,mean is the mean value of the mixture diffusion coefficient within the gel body at a certain time interval k and autoclave/bulk fluid space interval s. Both values are pressure- and temperature-dependent. K1mean ¼

kend

S

k¼1 s¼1 2

τg Eg ∙DEtOH,CO2 ,g,mean ðP, T, k, sÞ

R ∙ ∙U ðP, T, k, sÞ L

ð15Þ

It should be noted that the real diffusion coefficient DEtOH,CO2 varies with mixture composition (Vignes equation

tse,min (s)

108

1.0E+05 1.0E+04 1.0E+03 1.0E+02 1.0E+01 1.0E+00 1.0E–01 1.0E–02 1.0E–03 0.01

Hg / Wg = 1 Hg / Wg = 0.4 Hg / Wg = 0.125 Hg / Wg = 0.07

1 0.1 Radius gel sphere (mm)

10

Fig. 4.18 Theoretical minimal supercritical extraction time tse, min as function of gel sphere radius and ratio of gel porosity εg to gel tortuosity τg (T ¼ 318 K, P ¼ 12 MPa, xend EtOH,gel ¼ 0.0097). (Reprinted from [41], with permission from Elsevier)

[46]) and therefore changes during the supercritical extraction process, unlike the above-defined effective diffusion coefficient Keff. To determine Keff, tse, min must be calculated (13) as described. The theoretical minimal supercritical extraction time tse, min is the extraction time achieved in the case of outer mass transport with the highest possible gradient (wherein solvent concentration in the bulk fluid stays zero). Fig. 4.18 shows tse, min for different gel sphere sizes. For monolithic spherical gels with a radius of 1 cm, a tse, min of less than 2 h can be achieved. This is close to reported experimental drying/supercritical extraction times [27, 29, 30]. Similarly, much shorter supercritical extraction times, in the range of a few seconds to a few minutes, are all that is required for particles with diameters in the range of microns to millimeters. In addition to size, the porosity and tortuosity of the gel also influence minimal supercritical extraction time [10, 26–31, 38, 39, 55]. High-porosity gels with low tortuosity show faster supercritical extraction times due to less hindered mass transport within the gel network. Likewise, lower pressures and higher temperatures enhance inner diffusional transport and thus decrease tse,min (Table 4.2) or, rather, increase Keff (Table 4.3), Eq. (13). The above-defined Biot number allows for analysis of supercritical extraction time independent of the size and inner structure of the gel particles being examined. Variation of the supercritical extraction time (relative to the theoretical minimal drying time at β ! 1) as a function of Biot number is shown in Fig. 4.19. Two distinct supercritical extraction regimes can be identified from this figure: Unless the Biot number is higher than 65.0 (at T ¼ 318 K and P ¼ 12 MPa), the variation of supercritical extraction time compared to tse,min is negligible (less than 2%) and thus limited by diffusion within the gel matrix. At lower Biot numbers, the extraction time increases exponentially resulting in limited mass transfer from the particle surface to the surrounding fluid. The transition Biot number thus defines the lowest Biot number at which the first regime dominates over the second regime and

4

Supercritical Drying of Aerogels

109

Table 4.2 Theoretical minimal supercritical extraction time tse, min at various temperatures and pressures for spherical gel particles with R ¼ 250 μm, porosity εg¼1, and tortuosity τg¼1. End points: xend EtOH,gel (at T ¼ 313 K) ¼ 0.0078, xend EtOH ,gel (at T ¼ 318 K) ¼ 0.0097, and tse, min (s) Temperature (K)

313 318 323

Pressure (MPa) 9 1.84 1.73 (1.69)a

xend EtOH ,gel (at T ¼ 323 K) ¼ 0.0118). Mixture critical pressures: Pcrit (at T ¼ 313 K) ¼ 7.60 MPa, Pcrit (at T ¼ 318 K) ¼ 8.37 MPa, and Pcrit (at T ¼ 323 K) ¼ 9.09 MPa

10 1.97 1.77 1.65

12 2.13 1.92 1.75

13 2.19 1.98 1.80

17 2.40 2.18 1.98

a

Supercritical extraction at 9 MPa and 323 K cannot be performed completely in the single-phase region above the mixture critical point; thus, the result is placed in parentheses

Table 4.3 Factor Keff at various temperatures and pressures for spherend ical gels. End points: xend EtOH ,gel (at T ¼ 313 K) ¼ 0.0078, xEtOH,gel end (at T ¼ 318 K) ¼ 0.0097, and xEtOH,gel (at T ¼ 323 K) ¼ 0.0118. Mixture Keff (109 m2/s) Temperature (K)

313 318 323

Pressure (MPa) 9 33.9 36.1 (36.9)a

critical pressures: Pcrit (at T ¼ 313 K) ¼ 7.60 MPa, Pcrit (at T ¼ 318 K) ¼ 8.37 MPa, and Pcrit (at T ¼ 323 K) ¼ 9.09 MPa

10 31.8 35.4 37.8

12 29.3 32.5 35.8

13 28.5 31.5 34.7

17 26.0 28.7 31.6

a

Supercritical extraction at 9 MPa and 323 K cannot be performed completely in the single-phase region above the mixture critical point; thus, the result is placed in parentheses

Supercritical extraction time/tse,min (%)

1.0E+04 Limited by internal mass transfer

1.0E+03

Limited by external mass transfer

1.0E+02 1.0E+01 1.0E+00 1 2 4 0 3 1 –0 +0 +0 +0 +0 +0 E E E E E E 0 0 0 0 0 0 1. 1. 1. 1. 1. 1. Biot-number Bi (–)

Fig. 4.19 Supercritical extraction time variation (relative to the theoretical minimal supercritical extraction time at β ! 1) as a function of Biot number (14) for various particle radii (15 μm to 2.5 mm) and εg/τg ratios (0.125 to 1); T ¼ 318 K, P ¼ 12 MPa. (Reprinted from [41], with permission from Elsevier)

thus represents the ideal conditions for attaining a fast extraction using low CO2 amounts (the lowest possible β and therefore velocity and CO2 mass flow at tse, min ∙ 1.02). This transition Biot number depends on the system conditions as shown in Table 4.4. High pressures and low temperatures lead to a decreasing transition Biot number [41].

Table 4.4 Effect of temperature and pressure on the transition Biot end number Bitransition. End points: xend EtOH ,gel (T ¼ 313 K) ¼ 0.0078, xEtOH ,gel end (T ¼ 318 K) ¼ 0.0097, and xEtOH,gel (T ¼ 323 K) ¼ 0.0118) Bitransition () Temperature (K)

313 318 323

Pressure (MPa) 9 12 78.1 66.6 86.5 65.0 – 67.6

17 58.9 61.1 60.1

As shown, the supercritical extraction phase can be described by the Biot number. When the solvent concentration in the bulk fluid remains low, it can be neglected. When the solvent concentration in the bulk fluid is not negligible, for example, due to a high ratio of gel volume to autoclave volume, the properties of the bulk fluid must be considered during analysis. The dimensionless number K1mean Eq. (15) was therefore developed to analyze the supercritical extraction of spherical gel particles in a packed bed where varying solvent concentrations in the bulk fluid are considered Eqs. (4)–(6), Eqs. (10)–(12) [58]. By introducing this number, which relates the outer mass transport in the bulk fluid to the diffusional mass transport within the porous gel spheres, an explicit universal dependence can be observed which is independent of gel sphere radius, gel porosity, gel tortuosity, L/dac-ratio, volume and porosity of the packed bed, CO2 density, temperature and pressure combination, and

4

R. Subrahmanyam et al.

20

20 420 kg/m^3 560 kg/m^3 720 kg/m^3

18 16 14

18 16 14

Limitation in the bulk fluid

12

12

10

10

8

8

6

6 Diffusion limitation

4 2

4 2

0

CO2 consumption excl. pressurization/initial ethanol mass (g/g)

Fig. 4.20 Calculated supercritical extraction time relative to minimum theoretical supercritical extraction time and calculated CO2 consumption are dependent on K1mean (xstart EtOH ,gel ¼1, xstart EtOH,fluid ¼0.05). Removal of solvent due to pressurizationinduced spillage is not included. CO2 for initial pressurization is not included in calculated CO2 consumption relative to initial ethanol mass. (Reprinted from [58], with permission from Elsevier)

Supercritical extraction time/tse,min (–)

110

0 0

2

4

6

8

10

12

14

16

18

K1mean (–)

compositional end point xend EtOH,gel ¼ f ðT Þ. This dependence is shown as black diamonds in Fig. 4.20. Again, distinct regions were observed: When K1mean < 2, supercritical extraction times were limited by mass transport in the bulk fluid, whereas when K1mean > 5, overall mass transport was limited by the inner diffusional mass transport in the gel spheres. For 2 < K1mean < 5, a transition between these two limits was observed [58]. To provide an overview of the implication of these insights, the amount of CO2 consumed per gram of ethanol extracted was calculated for each of the supercritical drying conditions investigated, shown in Fig. 4.20 sorted by the corresponding CO2 density. It can be seen that CO2 consumption per gram of extracted ethanol increases with increasing dimensionless number K1mean. Additionally, CO2 consumption is observed to increase with CO2 density. Figure 4.20 thus serves as a guideline for how to choose ideal process conditions for supercritical drying, discussed in more depth in the following section [58].

4.4.5

How to Determine Process Conditions and Drying Time: Experimental and Theoretical Views

A compilation of experimental drying times (i.e., supercritical extraction times) reported in the literature from 1994 to 2018 as a function of the smallest gel thickness reportedly dried within such times is shown in Fig. 4.21. We selected only those studies in which supercritical drying conditions were controlled, dimensions of the gels were reported, and the end point of the process was detected. Although there is

clear scatter in the experimental drying times reported, the average value was found to be in the range of 3–3.5 h. This shows that (contrary to a popular belief) supercritical drying is a fast process, at least for gel thicknesses up to 15 mm (which covers most practical needs). The solid line in Fig. 4.21 represents the minimal supercritical extraction time tse, min calculated by the above model when an infinite mass transfer coefficient is assumed. Thus, this line can be interpreted as the theoretical minimal time required for supercritical drying, i.e., when no mass transfer hindrance in the fluid phase is present. Inspection of Fig. 4.21 suggests that manipulation of mass transfer in the bulk phase could potentially reduce supercritical drying time, which can be achieved through rational design of the geometry of the autoclave and flow structure. The dimensionless numbers presented in the previous section (Biot number and K1mean number) can thus be used to determine optimal supercritical extraction conditions for a given gel size in support of achieving a fast supercritical extraction process involving low CO2 consumption. The Biot number can be used in the case of low solvent concentrations in the bulk fluid (i.e., a high concentration gradient between the gel surface and the bulk fluid). Procedure for optimizing supercritical extraction of individual spherical gel monoliths containing ethanol as the pore liquid using a CO2-rich bulk phase: 1. Determine the gel particle radius R, gel porosity εg, and gel tortuosity τg. 2. Select a high Keff value from Table 4.3 and determine the supercritical extraction temperature and pressure. A preferably high temperature and low pressure should be

Supercritical Drying of Aerogels

Fig. 4.21 Experimental supercritical drying times as a function of the smallest gel thickness dried (data from [10, 26–31, 38–40, 55]) compared with estimated theoretical minimal supercritical extraction time (calculated for a gel body in form of a sphere at 17 MPa and 313 K (lowest Keff from Table 4.3) with a porosity-to-tortuosity ratio of 0.3)

111

9 Experimental drying/supercritical extraction time (h)

4

Experimental sc. extraction/drying times

8 7

Min. theoretical sc. extraction time

6 5

4

4 3 2 1 0 0

5

10

15

20

Smallest gel thickness (mm)

chosen for monolithic gels due to a slightly faster inner mass transport and higher tolerable final ethanol mole fraction within the gel (xend EtOH,gel ¼ f ðT Þ, Sect. 4.4). 3. Determine the corresponding CO2 density. 4. Calculate β from Eq. (16) using Bitransition from Table 4.4 to determine the optimal process window where inner mass transport is as fast as mass transfer from the gel surface to the bulk fluid: β¼

Eg ðP, T Þ∙K eff ðP, T Þ ∙Bi τg ∙R transition

ð16Þ

5. Estimate the mass flow needed to achieve the calculated β using the CO2 density and known mass transfer correlations from the literature in accordance with the geometry of the gel configuration and autoclave setup used. For supercritical extraction of gels in the form of a cylinder or plate/tile, the gel particle radius R in the first step of the procedure above can be approximated by half of the smallest dimension of the gel part. Note that due to geometrical reasons, the theoretical minimal supercritical extraction time tse, min will be lengthened approximately twofold for the case of a cylinder and approximately fourfold for the case of a plate. Thus, the selected Keff value should be multiplied by approximately 1/2 for the case of a cylinder and by approximately 1/4 for the case of a plate to account for the slower extraction rate. Thus, corresponding lower mass flows and

longer extraction times (compared to the extraction of a gel sphere) are recommended for such geometries. In the case of small gel particles (for instance, in a packed bed), the above approach is still valid, although some modifications are suggested as described below. Procedure for optimizing supercritical extraction of ethanol from spherical gel particles in a packed bed configuration using a cylindrical autoclave [58]: 1. Determine the gel particle radius R, gel porosity εg, and gel tortuosity τg. 2. Calculate the initial ethanol mass mEtOH, start (volume of gel particles times gel porosity times ethanol density plus the amount of excess ethanol in the autoclave (should be small as discussed in Sect. 4.1)). 3. Estimate the approximate theoretical minimal supercritical extraction time tse, min for a single gel particle from Fig. 4.18. 4. Choose a K1mean value from Fig. 4.20 that corresponds to an acceptable combination of the ratio supercritical extraction time to tse, min and CO2 consumption per mEtOH, start. For example: (a) In the case of particles with diameters in the range of millimeters to centimeters, K1mean ¼ 10. At this K1mean value, the overall supercritical extraction time is 1.5 times tse, min – in the range of few hours – and involves a CO2 consumption of 8 g CO2/g EtOH (at a CO2 density of 420 kg/m3).

112

R. Subrahmanyam et al.

(b) In the case of microparticles, K1mean ¼ 0.5. At this K1mean, the overall supercritical extraction time is seven times tse, min – in the range of several seconds – requiring a relatively low CO2 consumption (for example, 1.9 g/g at CO2 density of 420 kg/m3). 5. Choose a CO2 density corresponding to the fixed CO2 consumption. Take the temperature/pressure combination of the chosen CO2 density with the highest

m_ 

CO2 consumption per initial ethanol mass  mEtOH ,start f sc:extraction time per min:theoretical sc:extraction time  tse, min corr

f corr ðK1mean  2Þ ¼ 1:5  0:113  K1mean f corr ð2 < K1mean < 40Þ ¼ 1:23  0:054  ln ðK1mean Þ f corr ðK1mean  40Þ ¼ 1:025 ð18Þ 7. For a packed bed in a cylindrical autoclave, use an L/dac ratio of 2.6 or higher.

4.4.6

temperature due to a slightly faster mass transport and higher tolerable final ethanol mole fraction within the gel (xend EtOH,gel ¼ f ðT Þ, Sect. 3.4). 6. Estimate the outlet mass flow using Eqs. (17) and (18):

Observations on Depressurization Rates

After supercritical drying is complete, the autoclave is typically depressurized at a constant temperature, i.e., isothermally. As a result, the depressurization rate plays an important role. The most important contributions regarding depressurization rates were provided by Woignier [59], who proposed a variabledepressurization-rate protocol for obtaining crack-free silica aerogel monoliths. Total depressurization times reported in literature vary from 48 min [59] to 10 h [7]. Our experience shows that, typically, a fast depressurization from 120 bar to 1 bar over the course of 20–30 min can be used for monolithic samples up to 20 mm in thickness. Many aerogel compositions including silica, polyurethanes, and biopolymers such as alginates can be depressurized within 30 min without any significant change in bulk density or surface area. In the case of silica, additional care should be exercised during depressurization for the reasons previously discussed by Woignier [59]; however, for nonbrittle systems such as polyurethanes and certain biopolymers, depressurization times longer than 30 min are not required. For aerogel particles, depressurization times can be as fast as 10–15 min. Depressurizing from 120 to 1 bar in under 10 min will cause the resulting aerogel samples to undergo significant cooling and, for polyurethane-type and alginate monoliths, will also result in breakage of monoliths. For alginate aerogel particles, this can further result in shrinkage, as CO2 can freeze inside the pores and sublime out of the pores in an

ð17Þ

uncontrolled fashion. Operating at higher autoclave temperatures (60–70  C) is one potential strategy to compensate for this problem when using fast depressurization rates. As a rule of thumb, depressurization rate should be pegged to autoclave temperature: Depressurization should only proceed as fast as can be performed such that that the autoclave temperature (and thus the aerogel temperature) does not drop below 40  C during depressurization. Some of these considerations have been previously recognized by commercial producers of aerogels. For example, in 1999 Aspen Aerogels filed a patent application claiming a reduction in the time required for supercritical drying using the following approaches: (1) raising the initial temperature of the system above the critical point of the supercritical fluid, i.e., CO2, to save heating time; (2) introducing and extracting with CO2 in the supercritical state instead of a liquid CO2 to accelerate diffusion; and (3) oscillating the pressure of the system to invoke increased mass flow in and out of the gel pores via suction and spillage (WO 0128675 from Aspen Systems Inc, 2000) [60].

4.5

Practical Aspects, Scaling Up, and Economic Considerations

Thus far, we have discussed various physical phenomena encountered during supercritical drying. In this section, we address practical aspects related to the implementation of supercritical drying. For those interested in purchasing or constructing their own equipment for performing supercritical CO2 drying, we present a laboratory setup suited for drying of aerogels based on our experience. This said, it should be appreciated that many equipment configurations are possible and that different system configurations may be called for depending on the type, form factor, and size of aerogel materials you are interested in making. Additional designs and procedures for a low-cost basic supercritical dryer and a high-temperature supercritical drying system based on a hot press are described in ▶ Chap. 65.

Supercritical Drying of Aerogels

4.5.1

113

Typical Laboratory-Scale Setup

A schematic for a typical lab-scale supercritical drying setup is depicted in Fig. 4.22. The setup comprises a pressurization section (consisting of a condenser/cryostat for fresh and recycled CO2, a pump, and a heater), a main autoclave (rated for 120–130 bar) equipped with accompanying heating jacket, and a pressure separator (55–60 bar) also equipped with a heating jacket (Fig. 4.22). Typical operating conditions for the autoclave are 35–65  C and 80–120 bar depending on the material being dried. Table 4.5 provides a cost breakdown for such a lab-scale supercritical drying setup. CO2 is a weak acid, and so to avoid potential corrosion issues, all pressure vessels, tubing, fittings, and valves are made of stainless steel (ASTM 316 L, grade 1.4404), duplex steel (ASTM 318 LN, grade 1.4462), or higher (ASTM 316 Ti, grade 1.4571). We recommend not compromising on heating jackets for the pressure vessels. While for smaller autoclaves (50–500 cc) electrical heating might suffice, for larger vessels we recommend oil or waterheating jackets. If an experiment requires independent temperature control over CO2 preheating, the primary autoclave, and the solvent separator, then three heaters are required. This said, in most cases one heater is sufficient to heat the CO2 stream and the two pressure vessels. Such a single-jacket configuration can provide a 10% savings in equipment cost. The above supercritical drying setup connects to a CO2 cylinder. A standard CO2 cylinder holds 37.5 kg of CO2,

55–60 bar

which exists as a liquid under its own vapor pressure inside the cylinder. CO2 cylinders can come equipped with a dip tube inside to draw from the liquid phase or without a dip tube to draw from the gas phase. We recommend selecting the dip tube (liquid-phase draw) option. We have observed that drawing CO2 from the gas phase cools down the cylinder and causes suction problems with the CO2 pump, especially in for long continuous CO2 flows. Note that for a basic supercritical dryer such as the one described in ▶ Chap. 65, a dip tube is required, as such systems do not include a pump. At lab scale (e.g., for autoclave volumes up to 5 L), we use a maximum of 15 kg CO2 per run for drying gel samples. For a 5-L autoclave, CO2 consumption is distributed over time (1–5 h) depending on sample thickness (0.5 mm to 20 mm). The flow can be either continuous or intermittent. For an intermittent flow profile, we recommend using more CO2 at the beginning (>50% CO2 in first time quartile) and less toward the end ( 1/‘M, the scattering intensity approaches a K′  q4 decay (with K′ ¼ b/‘M4, 7.10). The meaning of K′ is equivalent to K, however characterizing the structure on the micropore scale rather than the backbone particles. K and K′, respectively, can easily be determined when plotting (dΣ/dΩ)coh (q)  q4 rather than (dΣ/dΩ)coh (q) (Fig. 7.15). In this representation, K′ is masked by the overlapping background. Independent of the model used to describe the scattering pattern and the underlying structure of the sample, the so-called invariant Q, defined as Q¼

dΣ dΩ

2

q dq,

ð7:11Þ

coh

is a characteristic parameter of the system investigated. Due to the general properties of the correlation function it turns out that Q is containing the integral information about the volume fraction of the two phases (e.g., pores and solid, or liquid and solid in case of a gel) present: 2

2

Q ¼ 2π ϕ  ð1  ϕÞ  ðΔρSLD Þ ,

ð7:12aÞ

with ϕ and 1 ϕ the volume fraction and ΔρSLD the difference of the scattering length densities of the two phases (see also Table 7.2) Eq. (7.12a) simplifies to 2

Q ¼ 2π ϕ  ð1  ϕÞ  ðρS  CN,X Þ

2

ð7:12bÞ

when one of the phases are (empty) pores. Here ϕ is the porosity of the sample, ρs is the mass density of the solid phase, and CN, CX are the factors to be applied in case of neutron and X-ray scattering, respectively, for relating the mass to the respective scattering length density (Table 7.2). Hence (7.12b) allows quantifying the total porosity. Figure 7.15 visualizes the change of the integral as a function of the integration limit q′. Note that the numbers are given on a double-log scale. To extract the total porosity it is necessary that the experimental q range covers all features of the

7

162

G. Reichenauer

b 102

104 (d6/d:)q4 (10–28 cm–5)

3

(d6/d:) (cm–1)

10

102 101 100 10–1 0.1

1 Scattering vector (nm–1)

10

(d6/d:)q2 (a.u.)

a

q’ (d6/d:)q2 dq 0 101

100

0.1

1 Scattering vector (nm–1)

10

Fig. 7.15 (a) Experimentally derived differential scattering cross-section and cross-section times q4 for a carbon aerogel. The dashed line indicates the position of K′ that is masked in the original dataset by the

background B [see (7.10)]. (b) Kernel of the integral Eq. (7.11) and the integral itself as a function of the integration limit q′ (dashed line)

scattering curve that are significantly contributing to the integral value. Therefore, it is required that the crossover of the scattering intensity to a plateau at low q values is included in the experimental q-range. If the invariant can be calculated from the experimental data it can be used to normalize them; this is particularly valuable when the scattering data cannot be given on an absolute scale (e.g., for non-monolithic pieces of aerogel). The Porod constant K can be used to determine the specific surface area S/m either by applying the invariant Q or directly from the experimentally derived scattering crosssection; when no micropores are present the following relations hold for the evaluation of an aerogel or xerogel:

present [as it is the case for microporous samples (see Fig. 7.15)], both can be evaluated according to (7.13a, 7.13b). Hereby care has to be taken to correctly match the Porod constants with the respective densities of the solid phase and the surface area that can be extracted (In case of carbon aerogels the pairs are: K, density of the interconnected backbone particles ⇨ particle surface, K′, ρcarbon ⇨ surface of particles and micropores.). The correlation length ‘M in (7.10) is the geometrical average over the chord length, i.e., the mean extensions of the solid ‘S and the void phase ‘p:

S S 1 K ϕ ¼  ¼ π with m V 0 ð1  ϕÞ  ρS Q ρS K ¼ q!1 lim I ðqÞ  q

4

ð7:13aÞ

or 4

2

K ¼ lim ðdΣ=dΩÞcoh  q ¼ 2π ðΔρSLD Þ  q!1

S 2 S ¼ 2π ðρS  CN,X Þ : V0 V0

ð7:13bÞ

Here, V0 is the total volume of the sample, ρS is the mass density of the solid phase, ϕ is the porosity of the sample, ΔρSLD is the difference in coherent scattering length density between the solid and the pore phase, and CN, CX are the factors to be applied in case of neutron and X-ray scattering, respectively (Table 7.2). When two Porod regimes are

1 1 1 ¼ þ and ‘S ¼ ‘M =ϕ, ‘P ¼ ‘M =ð1  ϕÞ: ð7:14Þ ‘M ‘S ‘P Furthermore the relation ‘S,P ¼ 4 ΦS,P V0/S holds for any statistical system with two phases present [24, 96]. ‘p can be correlated to the width of the micropores in the carbon aerogel backbone. In addition, ‘M is related to the ratio of the invariant Q and the Porod constant K by ‘M ¼

4 Q  : π K

ð7:15Þ

When evaluating Q for the micropore term in (7.10) only, the density of the backbone particles and thus the micropore volume can be estimated by solving for ϕ in (7.12a, 7.12b) (It has to be emphasized that due to the large uncertainties of the background contribution at high q values, the accuracy is on the order of 10–20% only.). The integral for the respective term in (7.10) yields

Structural Characterization of Aerogels

Q¼

163

π b  : 4 ‘3M

ð7:16Þ

In addition to the mean chord or correlation length ‘M, a chord length distribution can be calculated from the scattering curve. This approach was used by Cohaut et al. to analyze the impact of the type and concentration of catalyst as well as the carbonization temperature on the pore structure of carbon aerogels [36]. Another option for the analysis of scattering data are reconstruction concepts; in these approaches different models are applied and parameters within these models are varied until the simulation of the scattering curves for the modeled structures corresponds to the experimentally derived scattering data [37, 38]. This approach provides three-dimensional visualization of the aerogel structure (Fig. 7.16) that can be a

b

z (nm)

400 0 400

y (nm)

00

x (nm)

z (nm)

400

400

0 400

400 y (nm)

00

x (nm)

Fig. 7.16 Reconstruction of a carbon aerogel macrostructure from SAXS data for two different samples with the same total porosity of 85% but different particle sizes [about 150 nm (a) and 60 nm (b)]. (Courtesy: Cedric Gommes, University of Liege)

used for modeling of other properties. However, since the scattering data do not contain unequivocal information the result of the reconstruction depends on the assumptions and the strategy used. Approaches for the Evaluation of Scattering Data. Three cases have to be distinguished when evaluating an experimental set of scattering data of an aerogel (Fig. 7.17): (a) The scattering data are not normalized and do not contain any clear structure or crossover in their scattering pattern within the experimentally q-range covered. (b) The scattering data are not normalized, however contain clear crossovers in their scattering pattern within the experimentally covered q-range; in particular the scattering intensity varies only weakly with the q value in the low q-range of the spectrum. (c) The scattering data are normalized, i.e., they are given in terms of a differential scattering cross-section on an absolute scale. In case (a), only the slope of the scattering curve in the double log plot can be determined. If the slope is steeper than 3 the part of the scattering curve shown is likely to be related to the range where the scattering is mainly sensitive to the interface between the two phases, i.e., the solid and void phase present. In Fig. 7.17, the slope of 3.4 in the double-log plot indicates a rough surface area. The fact that no crossover to a regime with a different slope is visible indicates that structural entities are larger than about 2/qmin (for spherical entities, here 2/0.8 nm1 ¼ 2.5 nm) are present.

c 104

103

103

103

Intensity (a.u.)

b 104

Intensity (a.u.)

a 104

102 101 100

q–3.4 1

2 3 4 5 678 q (nm–1)

dV/d: (cm2/(g sterad))

7

102 101 100 0.1

Fig. 7.17 Illustration of the three different cases of experimental information available from a scattering experiment (example: SAXS curve for a silica aerogel). (a) Limited experimental q-range with no distinct features, no absolute calibration; the dashed line is indicating a slope

102 101 100

1 q (nm–1)

0.1

1 q (nm–1)

of 3.4 in the low q-range of the experimental data. (b) Experimental range covering crossover to almost constant scattering intensities at low q values, (c) Scattering information available as differential crosssection given on an absolute scale

7

164

100

Intensity (a.u.)

In case (b), the slope in different regimes and the q values of the crossover between these regimes can be determined (see Figs. 7.13, 7.14). As the scattering curves level off at low q values, also the invariant Q can be determined and used to calibrate the scattering intensity. In case (c), all information is present in the scattering curve and the individual parts can be analyzed on an absolute scale even without determining the invariant Q. Scattering provides even more options than already discussed. For example, Ehrburger et al. [39] as well as Fairen-Jimenez and coworkers [40] extended the q-range from the classical SAXS regime to 10–2 nm1 by applying ultra small angle X-ray scattering (USAXS) to study carbon aerogels; at the same time they included scattering in the wide angle regime, thus providing the database to analyze the structural feature in the range from a few Å to about 100 nm. In particular, Fairen-Jimenez et al. [40] revealed in detail evaluation approaches of SAXS curves suited for carbon aerogels. Emmerling et al. [25], Kjems et al. [28] as well as Reim et al. [41] showed that it is possible to combine SAS and light scattering curves thus extending the experimental q-range to values as low as 10–4 nm1 thus including information on structures in the micron range. Emmerling et al. [25] and later on Li et al. [42] used the concept of the fractal dimension as determined from SAXS to correlate the light transmittance and structural properties of silica aerogels. Often an additional increase in scattering intensities at q values 1 yields a type II shape (see Fig. 7.33). For porous silica, Iler [109] provides an empirical relationship between the C parameter and the number density of OH groups per

b

400

300

Vads / Vads (0.4)

Vads (cm3 STP/g)

1.5

200

100

0 0.0

0.3

0.6

0.9

1.2

1.5

Statistical thickness t (nm) Fig. 7.35 t-plot (a) and αs-plot (b) of the same isotherm of a carbon aerogel. The t-plot results in an intercept (dashed line) at zero layer thickness of an adsorbed volume of 150 cm3(STP)/g representing the specific micropore volume and a slope corresponding to a monolayer

1.0

0.5

0.0 0.0

0.2

0.4

0.6

0.8

1.0

Relative pressure p/p0 volume of 35.9 cm3 (STP)/g (equal to a specific external surface area of about 155 m2/g). The dashed–dotted lines indicate the series of tangent used for the MP method

7

Structural Characterization of Aerogels

177

nm2. C values well above 100 are relatively insensitive to the curvature of the isotherm; very large C values are also a characteristic of microporous samples. Equation (7.29) in its linearized representation, i.e., p=p0 1 C1 þ ¼ V ads  ð1  p=p0 Þ V ml,BET  C V ml,BET  C  ðp=p0 Þ

ð7:31Þ

is called the BET equation. In the corresponding plot, the intercept a′ with the ordinate provides 1/(Vml,BET  C) while the slope b′ is equal to (C  1)/(Vml,BET  C). Hence 0

0

V ml,BET ¼ 1= a þ b ; C ¼ SBET ¼

b0 þ 1 and a0

V ml,BET  Sgas  N A V mol,STP

ð7:32Þ

with Sgas the area taken by a single adsorbate molecule (see footnote 11), NA the Avogadro constant, and Vmol,STP the molar volume at STP (¼22.414 103 cm3 (STP)/mol). Since the result of the fit of the BET equation to the isotherm strongly depends on the range used for the fit, it is important to follow recommended rules for its selection. For aerogels that do not contain a significant amount of micropores the range in rel. pressure to be used for the evaluation according to (7.31) is about 0.05–0.25. For highly microporous aerogels, this range will result in negative values of C; for such samples the evaluation range is usually shifted to lower relative pressures until the C value become positive [83]. A good strategy for the selection of the range to use for the fit is to plot Vads  (1  p/p0) and determine the rel. pressure ( p/p0)max at which this product reaches its maximum (Fig. 7.36). For the fit only relative pressures below ( p/p0)max should be applied [83]. Meso- and Macropores. The simplest approach to get an estimate for a characteristic pore size dav in the aerogel under investigation is to determine the specific surface area by gas adsorption and the specific total pore volume Vpore from the macroscopic and the skeleton density of the aerogel [see (7.21)] and to apply the relationship dav ¼

4  V pore : SBET

ð7:33Þ

This value is the so-called hydraulic pore diameter, a characteristic quantity to describe systems with noncylindrical pores [110]. To determine the average meso- or macropore size in case of microporous sample, the specific total pore volume Vpore has to be replaced by (Vpore – Vmic) and the external surface area has to be used instead of the BET surface area.

The most widely used approach to extract a pore size distribution from a gas adsorption isotherm is the BJH model [111], based on the capillary condensation in mesopores. According to the Kelvin equation the radius rM of the meniscus formed by a completely wetting liquid upon condensation in a pore is related to the relative pressure by (When the adsorbate is only partially wetting a factor of cos (θC), with θC the contact angle, has to be included on the right side of the equation.) rM ¼

2γ  V L,mol : RG  T  ln ðp=p0 Þ

ð7:34Þ

Here γ and VL,mol are the surface tension and the molar volume of the condensed liquid (For liquid nitrogen at 77 K, the values for the surface tension and the molar volume are 0.00885 J/m2 and 34 cm3/mol.). As prior to mesopore filling a surface layer of adsorbate tL is already present, the pore size d is given in case of a cylindrical or spherical pore by d ¼ 2  ðjr M j þ tL Þ:

ð7:35Þ

It has to be pointed out that in contrast to a spherical pore, the radius of the meniscus in a cylindrical pore is upon adsorption twice as large as the one effective upon desorption (For adsorption in cylindrical pores therefore d ¼ 2  (|rM|/ 2 + tL) holds.). As a consequence, a hysteresis is expected for geometrical reasons. However, other sources for a hysteresis having a more basic thermodynamic background have to be kept in mind (see also simulated isotherms for aerogel-like porous structures [88]). Combining the volume adsorbed in a relative pressure interval (corresponding to a given average value of p/p0) with the corresponding pore size (7.35) provides a pore volume distribution. Despite the fact that the shape of the pore defined as void in between the interconnected network is far from spherical or cylindrical pore geometry, Salazar and Gelb showed that comparison of an evaluation of an aerogel isotherm with the BJH model and a more fundamental approach used in their theoretical studies gives very similar pore size information [88]. While the classical BJH approach yields very similar results compared to DFT for large mesopores, BJH will underestimate the pore size for values 0.25 g/cm3 [69]) and even cracked materials so-called xerogels as discussed in the preceding chapter (2), (Fig. 13.4). As shown by Phalippou et al. [70], the irreversible densification during evaporation is coming from the condensation of remaining reactive silica species (see Eqs. 13.3 and 13.4). When submitted to capillary stresses, initially far distant hydroxyl and/or alkoxy groups can come close enough to one another to react and generate new siloxane bonds thus leading to irreversible shrinkage (Fig. 13.5), because of the inherent flexibility of the silica chains. Supercritical drying on the other hand permits to eliminate capillary stresses. Hence this process produces monolithic silica aerogels of rather large dimensions (Fig. 13.6), if required by a targeted application [71] (Fig. 13.7). Supercritical drying can be performed (i) in organic solvents in their supercritical state (generally alcohol as the pore liquid and consequently above 260  C if ethanol is used) according to a so-called HOT process [18, 72, 73] or (ii) in supercritical CO2 at a temperature slightly above the critical temperature of CO2 ~ 31  C according to a so-called COLD process. Application of the COLD process to silica gels was initially investigated by Tewari et al. [74]. For this purpose, the liquid which impregnated the wet gels had to be exchanged with CO2, either in the normal liquid state [26] or directly in the supercritical state [75]. Indeed, the interdiffusion of CO2 with methanol or ethanol, and with it the exchange, is slow and is significantly accelerated when CO2 is in its supercritical state [76, 77]. However, although perfect monolithic silica aerogels can be elaborated by both HOT and COLD supercritical drying routes, the supercritical way could remain too timeconsuming to be widely exploited on an industrial scale to produce this type of samples. Indeed, the low gel permeability results in rather slow CO2 washing and vessel depressurization steps [79, 80], in particular for thick gel plates. To speed up the CO2 washing, simple molecular diffusion must be assisted by forced convection, for example, by integrating

13

Silica Aerogels

t0

313

13

t1 t2

Fig. 13.4 Appearance of cracks during simple evaporative drying under ambient conditions, i.e., without any solvent exchange nor chemical modification of the pores’ surface (phenomenon here observed with mesoporous silica wet gels impregnated with ethylacetoacetate at times Fig. 13.5 Comparison of the shrinkage behavior, from the wet (a and b) to the dry (c and d) states, occurring during evaporative drying of native, e.g., untreated (left hand-side) and sylilated (right hand-side) silica gels. (Courtesy of Rigacci A., (MINES ParisTech, PERSEE, Sophia Antipolis, France))

t = 0 min

T = 25°C N2 flow = 50 sccm

t = 94 min

T = 25°C N2 flow = 50 sccm

compression-decompression cycles into the process [81]. However, if an accelerated depressurization is required, gels must be significantly strengthened prior to drying. If not, they will experience cracks even at low depressurization rates (Fig. 13.8).

t 1 ¼ 5 min and t2 ¼ 20 min, respectively, after the beginning of evaporation t0). (Courtesy of Rigacci A. (MINES ParisTech, PERSEE, Sophia Antipolis, France))

2 mm

t = 0 min

2 mm

T = 25°C N2 flow = 50 sccm

t = 146 min

T = 25°C N2 flow = 50 sccm

To accelerate the supercritical drying process, a “rapid supercritical extraction method” in which the silica sol or the precursors were directly gelled inside the container under HOT supercritical conditions, was investigated in the mid-1990s by Poco et al. [82] and subsequently by Gross

314

Fig. 13.6 Monolithic silica aerogels obtained after supercritical CO2 extraction. (Courtesy of Rigacci A. (MINES ParisTech, PERSEE, Sophia Antipolis, France))

Fig. 13.7 Large monolithic silica aerogel monoliths integrated in demonstration glazing. Original unpublished photograph similar to photographs published in ref. [78]. (Courtesy of K.I. Jensen and J.M. Schultz, DTU, Lyngby, Copenhagen)

et al. [83], Scherer et al. [84], and Gauthier et al. [85]. Even though successful in the case of small samples, this technique does not yet permit to elaborate large crack-free, low-density monolithic silica aerogels. Currently, to try solving the fluid exchange difficulties associated with the standard supercritical CO2 routes, one of the major challenges concerns the direct synthesis of the silica gel in supercritical CO2 by a water-free process [43, 86, 87]. Overall, the cost of production of silica aerogels remains high and their service life remains uncertain. According to Koebel et al. [13] it is expected, however, that the unit m3 cost of aerogel should decrease to £500 by 2050. This could make aerogel the market first super insulation material due to its outstanding thermal, visual and acoustical performance [14]. The latter authors also consider that the use of solvent

A. C. Pierre and A. Rigacci

Fig. 13.8 Typical depressurization crack (perpendicular to the largest surface) experimented by the silica gel during supercritical drying (illustrated here on a 1 cm-thick wet silica tile having a liquid permeability between 5 and 10 nm2, dried with supercritical CO2 at 313 K and 90 bars, and submitted to an autoclave depressurization of 0.15 bar/ min). (Courtesy of Rigacci A. (MINES ParisTech, PERSEE, Sophia Antipolis, France))

extraction by supercritical CO2 should permit to decrease the cost of drying to $ 2 m2 for a 1 cm thick aerogel. Besides, they indicate that regarding the full life cycle of the material, for a given insulation efficiency, silica aerogel is second best regarding the overall energy consumption, after glass wool which is ~10% less energy demanding. To decrease the present high production cost of silica aerogels by supercritical drying, some specific subcritical drying in non-aqueous solvents [88] and ambient pressure drying methods [1] were developed. In this case, the capillary stresses depend on surface tension and viscosity of the pore liquid, solid/liquid interface tension (wetting characteristics), drying rate and wet gel permeability. In order to reduce their negative impact on the drying, Drying Control Chemical Additives (DCCA) such polyethylene glycol (PEG) [25, 89], polyvinyl alcohol (PVA) [90, 91], glycerol [92], or surfactants [93] have been used. They interfere with the hydrolysis products of the respective Si precursors and permit to control the pore size and pore volume as well as their distribution. In typical ambient pressure drying processes, one of the key points relies on the introduction of incondensable species in the system via the sylilation of the silica gel [68] or the use of specific Si precursor like MTMS [30] in order to promote a so-called spring-back effect when the solvent front retreats and capillary stresses are released [64]. To conclude, after some trial and error ambient pressure drying was applied with great success to the synthesis of silica aerogels from alkoxides [94–96], as well as from waterglass [16, 68] and is today the most promising manufacturing technique for SiO2 aerogels. Densities below 0.1 g/cm3, for a total specific pore volume sometimes larger than that of CO2-dried samples could be obtained, something which would have been unimaginable 30 years ago. Besides,

13

Silica Aerogels

lightweight monolithic silica aerogels can sometimes be obtained without surface modification nor mechanical reinforcement, via “forced ageing,” by finally tuning the sol formulation. C. Scherdel and G. Reichenauer demonstrated with a two-step TEOS-based synthesis route that using low-water concentration in the first step and a high amount of ammonia in the second one permits to obtain highly porous ambient-dried silica aerogels, mainly macroporous with apparent densities as low as 0.15 g/cm3 [97]. Other techniques are worth to mention. The first one is the pinhole drying method [98, 99]. Limitation of the drying rate by limiting the solvent evaporation through a very small hole, provides more time for the wet gel to slowly adjust by plastic flow to capillary contraction, which in turn permits to create an earlier and denser siloxane Si-O-Si crosslinking between neighbor gel branches. A second one is a novel ambient drying process in subcritical conditions, based on dielectric heating [100]. The energy transfer provided by electromagnetic radiations can significantly reduce the drying time as well as the energy consumption. Some blanket-type silica-based aerogels with excellent thermal insulation properties have been already synthesized this way [101].

13.1.4 Synthesis Flexibility Besides these synthesis and processing methods, it must be emphasized that the flexibility of sol–gel processes permits to enlarge the selection of silica aerogel-based materials which is currently accessible. Bulk architecture can be tailored by templating techniques [102]. The gel chemistry can be modified by grafting, either during [103] (Fig. 13.9) or after gelation [104]. Composites and nanocomposites can be elaborated by impregnation of macroporous materials and structures such as foams and honeycombs [105] or fibrous networks [106], by dispersion of particles [107], powders [108], fibers [109], nanofibers [110], nanotubes [111], or polymers [112]; or by synthesis of mixed silica-based oxides [113–115]. Hereafter are some illustrations taken from works currently performed at MINES ParisTech respectively dealing with (i) entrapment (Figs. 13.10 and 13.11) and (ii) fibers use. In collaboration with the group of Nadya Pesce da Silveira (UFRGS, Porto Alegre, Brazil) two synthesis techniques for the entrapment of side chain liquid crystal (SCLC) polyacrylate in silica aerogel were compared: direct infiltration of the SCLC in monolithic silica aerogel and photopolymerisation of the SCLC after infiltration of its monomer. In both cases, the SCLC smectic ordering was not destroyed by confinement in the aerogel, but photopolymerization of the monomer increased the stability of this smectic phase [116]. Forced impregnation of a silica-based sol in an ultraporous cellulosic matrix

315

13

4a

4b

O

Si O O Si Si O Si O O Si O O Si O Si Si Si Si OH O O O O OH O Si O Si Si Si Si O Si O O HO O OH Si Si HO Si O HO O O O N O Si HO Si O O Si O NH Si HO O NH O Si Si O OH O Si O Si O O O O O Si Si Si Si Si

Fig. 13.9 Monolithic fluorescent silica aerogels obtained by reaction of silyl-functionalized benzazoles dyes with polyethoxydisiloxane in isopropanol under HF catalysis. (Reprinted from ref. [103], by permission from the Royal Society of Chemistry (Great Britain). Courtesy of Stefani W (UFRGS, Porto Alegre, Brazil) and Rigacci A. (MINES ParisTech, PERSEE, Sophia Antipolis, France))

(so-called Aerocellulose) [117] was also studied and permitted to create some finely textured interpenetrated organic-inorganic composites with rather low thermal conductivity and significantly better mechanical properties than native silica aerogels [118, 119]. These composites could be made displaying a contact behavior with water ranging from hydrophilic to very hydrophobic, and a porosity ranging from largely mesoporous to macroporous [119]. We also focused on the use of short cellulosic fibers (Fig. 13.12) to strengthen superinsulating silica aerogels [120, 121]. When used above their percolation concentration and well-dispersed, they permitted to obtain perfect monolithic samples by ambient pressure-drying (Fig. 13.13) without special needs (e.g., washing with alcanes) except silylation of the pore walls. Whatever the drying mode (with supercritical CO2 or by simple evaporation of the liquid phase in gentle temperature conditions), the mechanical properties of these composites were notably increased compared to their pure silica aerogels counterparts as measured by 3-points bending characterization (Fig. 13.14). The flexural modulus and maximum stress increased by a factor ~ 4. A remarkable result is

316

A. C. Pierre and A. Rigacci

1μm

1μm

Fig. 13.10 SEM micrographs of tritylcellulose (a) and tritylcellulose-silica composite aerogel (b). (Courtesy of Demilecamps A and Budtova T, MINES ParisTech, CEMEF, Sophia Antipolis, France and Rigacci A (MINES ParisTech, PERSEE, Sophia Antipolis, France))

10 9

σ (MPa)

8 7 6 5 4

0.1

a

b c

0.0 0%

2%

4%

3 2 1 0

100 μm 0%

20%

40% 60% ε (%)

80%

100%

Fig. 13.11 Stress-strain uniaxial compression curves of cellulose-silica composite aerogel (a), Aerocellulose (b), and silica aerogel (c). (Courtesy of Demilecamps A and Budtova T, (MINES ParisTech, CEMEF, Sophia Antipolis, France) and Rigacci A (MINES ParisTech, PERSEE, Sophia Antipolis, France). Adapted from Fig. 13.6 in ref [118])

moreover that the extremely low thermal conductivity achieved with pure silica aerogel was maintained, despite the presence of fibers (0.017  0.001 W m1 K1). Organic silica hybrids [122] can also be made either by many techniques such as use of organically modified silica precursors (i.e., ORMOSILS) (Fig. 13.15), co-gelation [123, 124], crosslinking after grafting [125], crosslinking with silanes [126, 127], or reaction with functionalized particles [128]. Recent works performed by Nakanishi, Kanamori, and co-workers [127] are really crucial since they permit – with the use of a single organically modified silica-based precursor (CH2CH(Si(CH3)O2/2))n, without any specific

Fig. 13.12 Tencel ® fibers used for synthesis of silica aerogels-based composites. Original unpublished micrograph similar to micrographs in Fig. 13.6 from ref. [121]. (Courtesy of LENZING AG, Austria, Budtova T, MINES ParisTech, CEMEF, Sophia Antipolis, France and Rigacci A. (MINES ParisTech, PERSEE, Sophia Antipolis, France))

washing step and following an ambient-pressure evaporative drying route – to synthesize monolithic and flexible, superinsulating, and transparent, silica-based material via a scalable process. Here-after are quoted some illustrations taken from works currently performed at MINES ParisTech dealing with cogelation. In a one pot synthesis route performed with sodium silicate as the silica precursor, dispersed silica μm-sized particles were formed inside an ultraporous cellulose network [129] (Fig. 13.16). As a second example, hybrids were made (in collaboration with the group of Matthias Koebel at the Swiss Federal Laboratories for Materials Testing and Research “EMPA,” Zurich, Switzerland) by cogelation of silicic acid and pectin. These

13

Silica Aerogels

samples showed drastically improved mechanical properties, due to the disappearance of necks between colloidal silica particles [130]. Finally, some lightweight and flexible organic-inorganic benzoxazine aerogels were prepared as blankets with a one-pot sol–gel synthesis method by crosslinking resorcinol-formaldehyde (RF) network with ((3-Aminopropyl)triethoxysilane) (APTES) and MTMS (or MTES) as silica precursors [131, 132] (Fig.13.17). After the drying stage, a wide panel of post-treatments can also be applied to increase the huge application potential of silica aerogels-based materials. For examples, chemical modifications by grafting in solution after re-impregnation [133] or in a gaseous atmosphere [134], skeleton coating by chemical vapor infiltration [135], and chemical vapor deposition

317

[136], impregnation of the bulk porosity with reactive species (Fig. 13.18), embedding in polymers (Fig. 13.19) or hydraulic binders for external thermal insulation of building envelopes [137] (Fig. 13.20), mechanical engineering by milling, cutting, laser micromachining [138], thermal processing such as sintering [139], etc. can be performed to target specific applications.

13.2

Main Properties and Applications of Silica Aerogels

The main properties of silica aerogels are addressed in the three following subsections regarding their texture, their chemical characteristics, and their physical properties which can be described altogether with their related applications.

13.2.1 Texture

Fig. 13.13 Example of silica-based ambient-dried superinsulating and superhydrophobic (θ ¼ 140) aerogel composite synthesized with Tencel ® fibers. (Courtesy of Budtova T (MINES ParisTech, CEMEF, Sophia Antipolis, France) and Rigacci A. (MINES ParisTech, PERSEE, Sophia Antipolis, France). From Fig. 13.8 in ref. [120], with permission from Springer) Fig. 13.14 Three-points bending test of ambient-dried superinsulating and superhydrophobic (θ ¼ 140) aerogel composite synthesized with Tencel ® fibers. (Courtesy of Budtova T. (MINES ParisTech, CEMEF, Sophia Antipolis, France) and Rigacci A. (MINES ParisTech, PERSEE, Sophia Antipolis, France))

Silica aerogels are amorphous materials. They have a skeletal density, as measured by Helium pycnometry [141] ~ 2 g/ cm3, close to that of amorphous silica (2.2 g/cm3). They typically have a pore volume in a range from 85 to 99.8% of their whole monolith volume [12, 142, 143]. Some ultraporous and ultralight silica aerogels can be synthesized by two-step process and a density as low as 0.003 g/cm3 has been reported [49]: these are the lightest silica aerogels which can be found in the literature. Silica aerogels are usually largely mesoporous (i.e., pore size from 2 to 50 nm), with interconnected pore sizes typically ranging from 5 to 100 nm and an average pore diameter between 20 and 40 nm [1]. Micropores (i.e., pore size 150 resulted. Another chemical commonly used as a hydrophobizing co-precursor is trimethyl-ethoxysilane (TMES). TMES is an

A. M. Anderson and M. K. Carroll

organosilane with three methyl groups and one methoxy group. It can undergo only one condensation reaction to connect into the Si–O–Si backbone of the sol–gel matrix, leaving the Si(CH3)3 moiety unreacted in the matrix. Rao and coworkers [74, 75] investigated the use of TMOS with TMES. They found that as they increased the amount of co-precursor the aerogels became more hydrophobic and less transparent (similar to the MTMS results). Rao and coworkers [90, 91] used a TEOS-based recipe with TMES. The mixtures required more time to gel but were more hydrophobic than those made with TMOS. Rao et al. [92] studied TEOS with PhTES as a co-precursor and made hydrophobic aerogels. Hegde et al. [59] used a two-step acid/base catalysis process with TEOS and TMES that decreased the gelation time and resulted in higher hydrophobicity (contact angles up to 155 ). Štandeker et al. [93] studied the effects of MTMS and TMES on hydrophobicity using CSCE for toxic organic compound clean-up applications and achieved contact angles of 42–173 for MTMS/TMOS molar ratios of 0.5 to 5 and contact angles of 100 to 180 for TMES/TMOS molar ratios of 0.5 to 5. SanzMoral et al. [48] prepared aerogels from TMOS and then performed surface functionalization using TMES in supercritical carbon dioxide. They showed that the originally hydrophilic aerogels could be rendered hydrophobic. Several investigators have used fluorinated organosilane co-precursors. Hrubesh, Coronado, and Satcher [94] and Reynolds, Coronado, and Hrubesh [95, 96] incorporated FPTMS into silica aerogels using the co-precursor method and demonstrated that this type of aerogel can be used as an adsorbent for organic compounds in water. They measured contact angles of 150 . Zhou et al. [13] used a perfluoroalkylsilane (PFAS) co-precursor. The addition of the PFAS increased hydrophobicity (contact angle of 145 ), density, and gelation time. Lin et al. [97, 98] produced fluoroalkylsilane-modified silica aerogel tubular membranes for use in CO2 absorption with contact angles as high as 140 . Duan et al. [99] studied the use of dimethoxy-methyl (3,3,3trifluoropropyl) silane (SiF3) to hydrophobize silica aerogels by using the SiF3 as a co-precursor added either pre- or postgelation and measured contact angles of 154 on the aerogels that were treated post-gelation. A variety of other precursors/co-precursors have been reported in the literature. These include TriEOS (Pauthe [100]) and TAM (Rao and Wagh [58]). Hegde and Rao [101] used a two-step process with TEOS/HDTMS, which decreased the gelation time and resulted in contact angles as high as 152 , although they found that the use of HDTMS decreased optical transmission. Yun et al. [102] prepared bridged silsesquioxane aerogels (contact angle 142 ) that are highly flexible. Their approach employed a click reaction between vinyltriethoxysilane (VTES) and 2,2′-(ethylenedioxy) diethanethiol, with subsequent sol–gel reactions (acid/base catalyzed), followed by drying under ambient pressure.

14

Hydrophobic Silica Aerogels

Given the number of different variables (processing method, precursor, catalyst type, solvent and water molar ratios), it can be difficult to compare the effects of co-precursor type on aerogel hydrophobicity. However, a number of papers in the literature provide direct comparison studies on the performance of different co-precursors. In general, it can be shown that, up to a limit, co-precursors with more alkyl/aryl groups will result in more hydrophobic aerogels. Rao et al. [70] studied a range of co-precursors with zero to three functional groups in combination with TMOS. They saw an increase in contact angle from 95 for aerogels made with MTMS (a monoalkyl organosilane) to 135 for those made with hexamethyldisilazane (both Si atoms are trialkyl substituted). Rao and Kalesh [103] looked at TEOS with a variety of different co-precursors including TMES, MTMS, PhTES, MTES, ETES, and DMCS. They found that the use of MTMS and MTES co-precursors resulted in monolithic transparent aerogels. PhTES, TMES, and ETES aerogels were monolithic but opaque, whereas the DMCS aerogels were cracked and opaque. The contact angle increased with number of alkyl or aryl groups covering the aerogel surface. Bhagat and Rao [104] looked at TEOS plus a range of co-precursors and found that as the chain length of alkyl groups present in the co-precursor increased from methyl to ethyl to propyl the contact angle increased. However, Anderson et al. [65] synthesized aerogels with MTMS, PTMS, and ETMS using a rapid supercritical extraction technique and saw no appreciable difference in contact angle for the different co-precursors at similar TMOS volume ratios. These RSCE fabricated aerogels also showed higher levels of hydrophobicity (compared to those in the literature) for low levels of precursor material.

14.4.2 Review of Silylation and Other Derivatization Methods Silylation has been performed using a variety of precursor materials including TEOS, sodium silicate, and PEDS. Due to the resulting surface passivation these techniques are generally able to use ambient drying. TMCS is probably the most widely used silylation agent but studies have been performed to look at the effects of other agents (see Table 14.3). Several studies have been performed using TEOS-based precursors with TMCS-based silylation. Jeong et al. [105] provide a direct comparison of TEOS-based, ambient-dried aerogel powders with and without silylation and show an increase in contact angle from near 0 to 158 when silylation was used. Wei et al. [106] present a technique that utilizes multiple treatment steps with TMCS and results in low-density (0.069 g/cm3), hydrophobic (contact angle 143 ) monolithic aerogels. Other researchers have used TEOS with HMDZ as a silylating agent. Rao et al. [107] report on the use of a

353

TEOS precursor with a two-step acid then base catalysis process (to speed up gelation) and HMDZ (in hexane) as a silylating agent with ambient drying. They were able to make super-hydrophobic surfaces with high transparency and report the best quality aerogel had a contact angle of 160 and optical transmission of 90% at 700 nm. However, the process involves a number of solvent exchanges and takes about a week to complete. The work was extended in Rao et al. [108] to examine the effect of different solvents (hexane, cyclohexane, heptane, benzene, toluene, and xylene) in the solvent exchange steps. They obtained highly hydrophobic aerogels (contact angle ¼ 172 ) when xylene was used; however, the level of hydrophobicity decreased over time (down to a contact angle of ~140 120 days later). Roig et al. [109] used PFAS in an aftertreatment process through solvent exchange with both CSCE and ambient drying process and produced aerogels with contact angles of 150 . Gurav et al. [29] looked at reducing the number of steps required to make ambient-dried TEOS-based aerogels and produced aerogels in a four-day process with contact angles as high as 150 . Rao and Kalesh [103] compared TEOSbased hydrophobic aerogels made using the co-precursor and silylation methods and concluded that aerogels made using co-precursors were more hydrophobic but less transparent than those made using the silylation techniques. Mahadik et al. [28] studied the effect of different silylating reagents on the surface free energy of ambient-dried aerogels fabricated from TEOS. They found that increasing the concentration of the silylating reagent increased contact angle and decreased surface free energy. Mahadik et al. [110] used mono-, di-, and tri-functional silylation reagents (trimethylchlorosilane, dimethyldichlorosilane, and methyltrimethoxysilane) to hydrophobize silica aerogels prepared via ambient drying and found that the mono-functional silylated silica aerogels resulted in better physico-chemical properties. Lermontov et al. [66] investigated the use of fluorinated solvents during sol–gel preparation and processing to yield aerogels. When they prepared TEOS gels in a solvent mixture that included hexafluoroacetone, hydrophobic materials resulted with contact angles of 94 and 140 , respectively, for CSCE and ASCE processing. When a fluorinated alcohol, hexafluoroisopropanol, was used as the supercritical fluid for TEOS-based aerogels, the resulting aerogels were hydrophobic (142 contact angle) with an F/Si molar ratio of 1.6. They proposed chemical pathways to explain how F from these solvents could be incorporated into the silica aerogel backbone.

14.4.3 Review of Work with Sodium Silicate Precursors Recent efforts to make aerogels from a lower-cost precursor, sodium metasilicate (frequently called “sodium silicate” or “waterglass”), have resulted in hydrophobic aerogels.

14

354

Schwertfeger and Schmidt [17] synthesized aerogels at ambient pressure using sodium silicate with water-soluble silylation agents (HMDSO and TMCS). Use of these compounds not only provides the necessary surface modification but also fills the pores with solvents more appropriate for ambient drying. The method was designed to avoid the lengthy solvent exchange step. Lee et al. [18] adapted Schwertfeger’s technique using solvent exchange and modification with an IPA/TMCS/n-Hexane solution. The IPA reacts with TMCS to yield isopropoxytrimethylsilane as a silylating agent. Other work with sodium silicate precursors was aimed at decreasing processing time (Shi et al. [111], Hwang et al. [112], Rao et al. [113]), studying the effects of preparation method (Rao et al. [114]), studying the effects of silylating agents (Rao et al. [115]), and studying the effects of solvents (Rao and coworkers [116, 117]). Bangi et al. [19] developed a sodium silicate processing route using tartaric acid instead of an ion exchange process. They found that increasing the number of washings with water improved the transparency of the gels. Their best aerogel has low density (0.084 g/mL), good optical transmission (50%), and high contact angle (146 ). Shewale et al. [118, 119] studied the effects of silylating agent type (HMDZ and TMCS) on the physical properties of silicate-based aerogels. Acetic acid was added to a sodium silicate solution to form and age the wet gel, which was then subjected to a vapor passing process to remove the sodium ions. The process involved several lengthy solvent-exchange steps and results showed that TMCS yielded lower density, more hydrophobic but less transparent aerogels than HMDZ. A number of authors have investigated the use of different materials to form a sodium silicate solution from which they make aerogels through ambient pressure drying with surface modification by TMCS as a silylating agent. Examples include rice husk ash (Halim et al. [120]), wheat husk ash (Liu et al. [121]), fly ash acid sludge (Cheng et al. [27]), gold mine treatment tailings slurry (Mermer et al. [122]), coal ganque (Zhu et al. [123]), diatomite (Wang et al. [124]), and clay (Hu et al. [125]. In general, the starting material is processed to form a powder (through heating, grinding, etc.) and then mixed with sodium hydroxide. The mixture is then heated stirred and filtered to prepare sodium silicate.

14.4.4 Review of Hydrophobic Silica-Containing Composite Aerogels There have been a variety of approaches to preparing composite or hybrid hydrophobic aerogels. Hydrophobic aerogel monoliths can be made by combining a silica core with a cross-linked polymer shell. Ilhan et al. [126] used 3-aminopropyl-triethoxy-silane (APTES) as a co-precursor

A. M. Anderson and M. K. Carroll

with TMOS in order to prepare an amine-derivatized silica sol–gel, then reacted the amine groups with a chlorinated styrene derivative, followed by addition (free-radical) polymerization with either styrene or 2,3,4,5,6-pentafluorostyrene (PFS). The fluorinated aerogels had higher hydrophobicity, with contact angles of 151 . To create stronger TMOS-based silica aerogels, Leventis et al. [127] crosslinked the wet silica gels with poly(hexamethylene diisocyanate) prior to CSCE processing. The resulting materials exhibited mass increases of only 4–5% when exposed to water vapor for 3 days, and withstood direct contact with liquids without collapse. Jung et al. [52] employed an RSCE method to prepare translucent monolithic composite aerogels using TEOS and poly(methyl methacrylate) (MMA), achieving contact angles of 161 with an average optical transmittance of 47%. Boday et al. [128] prepared hexylene- and phenylene-bridged polysilsesquioxane aerogels, and then used chemical vapor deposition (CVD) of methyl cyanoacrylate to enhance the flexural strength of these materials. The resulting composite materials had up to 65% higher densities and showed moderate hydrophobicity (contact angles of 92 and 115 ) with lower water content compared to silica aerogels. Seraji et al. [129] prepared superhydrophobic phenol-formaldehyde/silica hybrid aerogels with contact angle 157 . Although cellulose-based aerogels are inherently hydrophilic, a number of silica-based treatment approaches have been used to make them hydrophobic. Cervin et al. [35] demonstrated the fabrication of nanocellulose aerogels by freeze drying. The aerogels were rendered superhydrophobic (CA 150 ) by using a vapor deposition method to coat the surface with octyltrichlorosilane. He et al. [130] fabricated hydrophobic aerogels using micro-fibrillated cellulose (extracted from cotton) dispersed into silica sols (prepared from MTMS and DMSO which provide the hydrophobicity). The gels were supercritically dried (CO2) and demonstrated hydrophobicity (CA 115–125 ). Sai et al. [36] prepared bacterial cellulose–silica composite aerogels by washing sugar from the foodstuff Nata-de-coco to yield a bacterial cellulose (BC) hydrogel, then adding a sodium silicate precursor and freeze drying. Hydrophobization was achieved by then combining the dried BC-silica aerogel with an MTMS sol and again freeze drying. The composite aerogels had contact angles of 145 . Hayase et al. [131] prepared polymethylsilsesquioxane (PMSQ)–cellulose nanofiber composite aerogels that had high contact angles (~150 ), which they attributed to the inherent nature of the methyl-group-containing PMSQ network. Lin et al. [132] coated an alumina membrane support with an MTMS-based precursor solution, yielding PMSQ aerogel membranes with contact angles >120 after aging, solvent exchanges, and drying at 25  C. Demilecamps et al. [14] developed a method to functionalize a cellulose network with triphenylmethyl

14

Hydrophobic Silica Aerogels

chloride before immersing in a PEDs-based sol. After gelation the gel was then soaked in HMDZ and supercritically dried (CO2). By hydrophobizing both phases the authors produced hydrophobic cellulose-based aerogels with CA 133 .

14.4.5 Effect of Processing Parameters on Hydrophobicity A study by Husing et al. [69] presents a comparison of aerogels dried using CSCE to those dried using ASCE. The aerogels were synthesized using TMOS with an MTMS co-precursor. They found that the concentration of Si–OH on the surface of the ASCE aerogel was negligible (making them hydrophobic) whereas the CSCE aerogels had non-negligible concentrations. However, Tillotson et al. [133] performed a similar comparison for aerogels made using TMOS with an FPTMS co-precursor and found that the water absorbing properties of the CSCE dried aerogels were similar to those of ASCE dried aerogels. Yang, L., et al. [134] performed a four-factor, three level set of orthogonal experiments to look at effects of processing parameters on the hydrophobicity of ambient-dried TEOSbased aerogels that used TMCS and n-hexane as a silylating agent. They considered the effects of: (a) the length of the hydrolysis period (12–48 h); (b) the ethanol/water ratio in the aging liquid (1–9); (c) the length of the TMCS modification period (24–72 h); and (d) the concentration of TMCS modifier (5–15%). They found that the concentration of TMCS significantly affected the hydrophobicity. The hydrolysis period and ethanol/water ratio in the aging liquid had a moderate influence on hydrophobicity and the length of the modification period had the smallest effect.

14.4.6 Effect of Heat Treatment on Hydrophobicity As the use of hydrophobic aerogels has expanded, the effect of heat treatment on the hydrophobic nature of aerogels has been more extensively studied. The effect of heat treatment on hydrophobicity can be assessed by direct measurement of contact angle (as described in Sect. 14.2) or through indirect measurement of mass change (thermogravimetric analysis, TGA) or heat flux (differential scanning calorimetry, DSC) as a sample is heated. TGA and DSC are related techniques. In a TGA experiment, the mass of a sample is monitored as the temperature is increased, typically at a constant rate. Observed losses of sample mass are then attributed to physical processes (e.g., loss of adsorbed solvent) and/or chemical

355

processes (e.g., thermal decomposition). In a DSC experiment, the differential amount of heat required to raise the temperature of the sample and a reference (standard) is monitored as a function of temperature. Exothermic and endothermic processes are shown as peaks in the DSC curve, and these are attributed to physical and/or chemical transitions undergone by the sample. In work in our lab (Smith [135]) we have investigated the effect of heat treatment on aerogels fabricated using MTMS and TMOS and processed using RSCE. A mixture of 17 mL of TMOS/MTMS (100:0, 25:75, 50:50 and 75:25 by volume), 55 mL of methanol, 7.2 mL of water, and 0.27 mL of 1.5 M ammonia was prepared and then poured into a metal mold and processed using RSCE as in Anderson et al. [65]. A Kruss DSA 100 was used to measure contact angles with 2.5-μL drops of deionized water. Five measurements were made at different locations on five samples from each batch. The samples were then exposed (in an air environment) to temperatures ranging from 50 to 500  C until they lost hydrophobicity. Each sample was placed in an oven on an aluminum weighing boat, exposed to a set temperature for 1 h and then allowed to cool. The contact angle measurements were repeated after exposure to each temperature. Figure 14.12 shows water droplets on a 25% MTMS aerogel before and after heat treatment. Figure 14.13 plots contact angle versus exposure temperature for the aerogels prepared from 25%, 50%, and 75% MTMS. We found that the 25% MTMS aerogels lost hydrophobicity after being heated above 200  C, the 50% MTMS aerogels lost hydrophobicity after being heated to 300  C, and 75% MTMS aerogels lost hydrophobicity after being heated to ~500  C (not shown). Furthermore, the heat treatment study found that at high temperature aerogels crack and begin to fall apart. For both 50% and 75% MTMS aerogels this happened above 300  C. The 25% MTMS aerogels were not heated to high enough temperatures to see this effect. Table 14.4 presents a summary of studies from the literature that studied the effects of heat treatment on hydrophobicity. The table indicates the precursor type, the processing method, the chemicals used to render the aerogel hydrophobic with resulting contact angle (CA), the method for measuring hydrophobicity, and the temperature at which the aerogel lost hydrophobicity. Most of these heat treatment studies were undertaken in an air environment. The table is organized by precursor type (SS, TEOS, TMOS/MTMS). In general the aerogels made by ambient pressure drying from sodium silicate and TMCS lose hydrophobicity between 395  C and 411  C. Liu et al. [136] show that this temperature corresponds to a loss of the
CH3 peaks in FTIR. The aerogels made from TEOS appear have a slightly lower temperature limit (ca. 350  C) whereas the aerogels fabricated with MTMS and TMOS have a higher temperature limit (ca. 500  C).

14

356

A. M. Anderson and M. K. Carroll

Contact angle (q)

Fig. 14.12 Beads of 2.5 μL deionized water on an aerogel prepared from 25% MTMS, 75% TMOS: (a) as prepared, CA ¼ 155 ; (b) after heat treatment to 50  C, CA ¼ 152 ; (c) after heat treatment to 200  C (CA 350  C 350  C 300  C 393  C 400  C 425  C 450  C 500  C 500  C 500  C 550  C

–

152

TG

>450  C

–

160

CA

498  C

107 152 147 178 160 156 147 152 161 170

outperformed granulated activated carbon (GAC) on a gram per gram comparison and was 30–130 times more effective. Standecker et al. [93] synthesized TMOS and MTMS- or TMES-based hydrophobic aerogels for the cleanup of toxic organic compounds in water. They showed that the hydrophobic aerogels have 15 to 400 times the capacity of GAC and that this capability persists for more than 20 adsorption/desorption cycles. Rao et al. [144] used MTMS-based aerogels in a study the absorption and desorption of organic liquids (alkanes, aromatics, alcohols and oils) and measured high uptake levels (10–21 times the aerogel mass). Coleman et al. [145] reported on the development of an aerogel-GAC composite to clean groundwater and remove uranium showing that this composite material exhibited superior performance compared to the use of GAC alone. Cui et al. [146] demonstrated that TEOS/MTES aerogels could be used to remove nitrobenzene from wastewater while Loche et al. [147] used graphene-reinforced silicabased aerogels in a water remediation application and found that the graphene (even in as small amounts as 0.1 wt%) enhanced the hydrophobicity. Other interesting uses of hydrophobic aerogels in environmental clean-up and protection applications include He et al. [148] who developed an MTMS-based polyurethane sponge-reinforced silica aerogel and demonstrated the use of a novel pumping system to increase the desorption efficiency and Rao et al. [76] who studied the use of hydrophobic

14

358

aerogels for surface protection (i.e., sculptures or statues) from corrosion due to interaction with the environment. More recently, Lin et al. [97, 98, 149] have investigated the use of hydrophobic silica aerogels for CO2 capture and demonstrated that they could achieve stable CO2 absorption on the membranes.

14.5.2 Biological Applications Aerogels have been used in biological applications (see review by Stergar and Maver [150]). Hydrophobicity is important in some applications such as in the development of drug delivery systems and for enzyme encapsulation. These applications take advantages of the nano-sized structure and high surface area as well as hydrophobic nature. Smirnova et al. [8, 63] and Gorle et al. [151] have studied the feasibility of silica aerogels as platforms for drug delivery systems. They synthesized aerogels from TMOS using CSCE with a methoxylation process to hydrophobize the gels and demonstrated that the adsorption and release of the drugs studied can be controlled by the density and hydrophobicity of the aerogel materials. Giray et al. [152] used silica aerogel as part of a composite material made up of aerogel coated with polyethylene glycol hydrogel for application as a drug delivery system and found that the aerogel hydrophobicity could be used to tune drug loading and release. MurilloCremaes et al. [153] demonstrated the use of silica aerogel as a drug delivery system for a hydrophobic moisturesensitive drug and showed that the aerogel system better prevented hydrolyzation and displayed faster release kinetics when compared to a polymeric system. Aerogels have also been used to encapsulate enzymes (see, e.g., Buisson et al. [154] or Orcaire et al. [155]). El Rassy and coworkers [78, 156, 157] reported that the hydrophobicity of the enzyme support system affects the success of such systems. Gao et al. [158] report on methods to enhance the immobilization of lipase on methyl-modified silica aerogels (HMDS).

14.5.3 Hydrophobic Surfaces The development of hydrophobic and superhydrophobic surfaces is another area of application for hydrophobic aerogels. Li et al. [159] provide a review of super-hydrophobic surfaces including a description of how they are made and where they are used. They describe the properties of superhydrophobic surfaces (self-cleaning and anti-sticking) as having important applications for antifouling and anti-sticking materials, selfcleaning windshields, stain-resistant textiles, and even the manufacture of waterproof and fire-retardant clothing. Several investigators have looked at employing aerogel materials to make superhydrophobic surfaces. In this case the aerogels provide both a nanoscale surface roughness and a chemical

A. M. Anderson and M. K. Carroll

hydrophobicity. For example, Doshi et al. [6] demonstrated that superhydrophobic surfaces can be made from a TMOSbased recipe using FPTMS as a co-precursor to yield an organo-silica aerogel film. Using UV/ozone treatments they were able to control the surface hydrophobicity to yield contact angles from 10 to 160 .

Anti-Fouling/Anti-Bacterial Surfaces A number of studies have focused on the use of hydrophobic silica aerogels in anti-fouling/anti-bacterial surfaces. Akbarzade et al. [7] investigated the use of hydrophobic silica aerogel in epoxy coatings to improve corrosion resistance. Silica aerogel was added to an epoxy resin at weight percents from 0% to 2%. The resulting coating was applied to steel sheets and cured for a week. The effectiveness of the coatings was studied using salt spray and electrochemical impedance spectroscopy. They found that the addition of the aerogel significantly improved the anticorrosive properties. Oh and coworkers [160, 161] investigated the use of hydrophobic silica aerogels as a bacterial anti-adhesion food contact surface. They fabricated TEOS-based films using CSCE and treated them with TMCS, which resulted in surfaces with water contact angles of 132–134 and demonstrated significant inhibition of bacteria due to the bacterial anti-adhesion properties. They attribute this result to the aerogel’s nanoporous nature and the porosity-induced reduction in the van der Waals forces between the aerogel and the bacteria. Drag Reduction Surfaces Rao et al. [162] and Hegde et al. [59] demonstrated the transport of liquids on superhydrophobic aerogel surfaces. The aerogels were prepared using MTMS or TEOS/TMES through ASCE drying. Using aerogel powdered surfaces they were able to make “marbles” of aerogel-coated liquid droplets (~14 microliters) which could be easily rolled on any surface, facilitating the transport of small quantities of liquid with potential applications in micro- and nanofluidic devices. Truesdell et al. [163] used aerogel material to fabricate a superhydrophobic surface. Their surface was constructed of a polydimethylsilane-patterned surface, coated with a layer of gold and then dip-coated in an aerogel material to form a superhydrophobic surface. Through measurements of the force and velocity field near the surface they showed that the drag on the surface could be reduced. Samaha et al. [164] coated surfaces with hydrophobic silica particles to achieve a combination of hydrophobicity and surface roughness. They then studied the effects of elevated pressure on the drag reduction produced by these surfaces in underwater applications. They measured drag reduction of 18–22% and found that a terminal pressure exists above which drag reduction is not possible. Work in our lab (Rodriguez et al. [87]) demonstrated drag reduction up to 30% for a surface coated with hydrophobic (TMOS/MTMS based) aerogels on a rotating disk.

14

Hydrophobic Silica Aerogels

Figure 14.15 shows an image of the bottom of the disk coated with (a) hydrophobic aerogel and (b) hydrophobic xerogel. The aerogel coated surface is able to maintain a large air bubble on its surface thus leading to drag reduction. Although the aerogel and the xerogel have similar contact angles (ca. 160 ), their pore structures are different. The aerogels average pore diameter is 15 nm compared to that of 4 nm for the xerogels. We conclude from these experiments that the morphology of the xerogels and aerogels, in addition to the chemical hydrophobicity, plays a significant role in the drag reduction. Kim et al. [165] coated a channel with superhydrophobic aerogel material using an ambient-dried MTMS-based recipe. SEM revealed that the entire inner surface of the channel was coated with a highly porous network and sessile drop tests on a flat surface resulted in a contact angle of 160 . They measured the pressure drop associated with droplet-based flow through the channel and determined that the aerogel coating reduced friction resistance to motion (by a factor of three) as compared to droplet flow in a smooth hydrophobic channel (CA 90). They attribute this to a reduction of the interfacial force at the liquid-solid-gas interfacial line.

a

b

359

Other interesting applications that take advantage of the hydrophobic properties of silica aerogels include Cherenkov detectors (Tabata et al. [140]), cosmic dust collectors (Tabata et al. [166]), waveguides (Ozbakir et al. [167]), and energy storage (Huang et al. [168]).

14.6

Conclusions and Future Directions

As we have illustrated in this chapter, there is a wealth of methods for making hydrophobic silica aerogels. The sol–gel matrix can be rendered hydrophobic through manipulation of the nano-structured hierarchical morphology, by inclusion in the initial sol–gel precursor mixture of co-precursors that contain nonpolar side chains, or through a variety of post-gelation reactions (organic modification, silylation). Processing of the wet sol–gels via ambient techniques, freeze drying, or supercritical extraction of the solvent yields hydrophobic aerogels. Dry aerogels can be rendered hydrophobic through chemical vapor phase treatment. The co-precursor method of making hydrophobic aerogels is relatively easy to employ. Studies have shown that silica aerogels become more hydrophobic with the addition of increasing amounts of co-precursor; however this leads to longer gel times and reduced transparency. The use of silylation yields hydrophobic aerogels that can be dried under ambient pressure conditions. These aerogels are generally more transparent but the processing can be more complex. Although silica aerogels are relatively fragile, the combination of hydrophobicity with other properties (high surface area, low density, low thermal conductivity, optical translucency, and so forth) makes these materials attractive for a variety of application areas of current interest, including chemical spill clean-up, drug delivery, anti-fouling/anti-bacterial surfaces, drag reducing surfaces, and transparent insulation. Acknowledgments The authors thank the following for photographs used in this chapter: Emily Green, Jason Melville, Caleb Wattley, Justin Rodriguez, Matthew Roizin-Prior, and Myung Joo Lee. Our own work with hydrophobic aerogels has been funded by grants from the National Science Foundation (NSF MRI CTS-0216153, NSF RUI CHE-0514527, NSF MRI CMMI-0722842, and NSF RUI CHE-0847901) and the American Chemical Society’s Petroleum Research Fund (ACS PRF 39796-B10). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.

References Fig. 14.15 Photographic images of a rotating viscometer spindle bottom surface in water. (a) A large bubble is captured on a surface coated with a 10% MTMS aerogel, (b) small bubbles formed on a surface coated with a 75% MTMS xerogel. (This figure is reprinted from Rodriguez et al. [87])

1. Plata, D.L., Briones, Y.J., Wolfe, R.L., Carroll, M.K., Bakrania, S. D., Mandel, S.G., Anderson, A.M.: Aerogel-platform optical sensors for oxygen gas. J. Non-Cryst. Solids. 350, 326–335 (2004) 2. Fesmire, J.: Aerogel insulation systems for space launch applications. Cryogenics. 46(2–3), 111–117 (2006)

14

360 3. Gougas, A., Ilie, D., Ilie, S., Pojidaev, V.: Behavior of hydrophobic aerogel used as a Cherenkov medium. Nucl. Instrum. Methods Phys. Res., Sect. A. 421(1–2), 249–255 (1999) 4. Jelle, B.P., Hynd, A., Gustavsen, A., Arasteh, D., Goudey, H., Hart, R.: Fenestration of today and tomorrow: a state-of-the-art review and future research opportunities. Sol. Energy Mater. Sol. Cells. 96, 1–28 (2012) 5. Bhuiya, M.M.H., Anderson, A.M., Carroll, M.K., Bruno, B.A., Ventrella, J.L., Silberman, B., Keramati, B.: Preparation of monolithic silica aerogel for fenestration applications: scaling up, reducing cycle time, and improving performance. Ind. Eng. Chem. Res. 55(25), 6971–6981 (2016) 6. Doshi, D.A., Shah, P.B., Singh, S., Branson, E.D., Malanoski, A.P., Watkins, E.B.: Investigating the interface of superhydrophobic surfaces in contact with water. Langmuir. 21(17), 7805–7811 (2005) 7. Akbarzade, K., Shishesaz, M.R., Danaee, I., Zarei, D.: The effect of nanoporous silica aerogel on corrosion protection properties of epoxy coatings on carbon steel. Prot. Met. Phys. Chem. Surf. 53(2), 279–286 (2017) 8. Smirnova, I., Mamic, J., Arlt, W.: Adsorption of drugs on silica aerogels. Langmuir. 19(20), 8521–8525 (2003) 9. Karatum, O., Steiner, S.A., Griffin, J.S., Shi, W., Plata, D.L.: Flexible, mechanically durable aerogel composites for oil capture and recovery. ACS Appl. Mater. Interfaces. 8(1), 215–224 (2016) 10. Yokogawa, H., Yokoyama, M.: Hydrophobic silica aerogel. J. Non-Cryst. Solids. 186, 23–29 (1995) 11. Miner, M.R., Hosticka, B., Norris, P.M.: The effects of ambient humidity on the mechanical properties and surface chemistry of hygroscopic silica aerogel. J. Non-Cryst. Solids. 350(7), 285– 289 (2004) 12. Zhang, Z., Shen, J., Ni, X., Wu, G., Zhou, B., Yang, M., Gu, X., Qian, M., Wu, Y.: Hydrophobic silica aerogels strengthened with nonwoven fibers. J. Macromol. Sci. Part A Pure Appl. Chem. 43(11), 1663–1670 (2006) 13. Zhou, B., Shen, J., Wu, Y., Wu, G., Ni, X.: Hydrophobic silica aerogels derived from polyethoxy-disiloxane and perfluoroalkylsilane. Mater. Sci. Eng. C. 27(5–8), 1291–1294 (2007) 14. Demilecamps, A., Alves, M., Rigacci, A., Reichenauer, G., Budtova, T.: Nanostructured interpenetrated organic-inorganic aerogels with thermal superinsulating properties. J. Non Cryst. Solids. 452, 259–265 (2016) 15. Malfait, W.J., Zhao, S., Verel, R., Iswar, S., Rentsch, D., Fener, R., Zhang, Y., Milow, B., Koebel, M.M.: Surface chemistry of hydrophobic silica aerogels. Chem. Mater. 27(19), 6737–6745 (2015) 16. Wong, J.C.H., Kaymak, H., Brunner, S., Koebel, M.M.: Mechanical properties of monolithic silica aerogels made from polyethoxydisiloxanes. Microporous Mesoporous Mater. 183, 23–29 (2014) 17. Schwertfeger, F., Frank, D., Schmidt, M.: Hydrophobic waterglass based aerogels without solvent exchange or supercritical drying. J. Non-Cryst. Solids. 225(1), 24–29 (1998) 18. Lee, C., Kim, G., Hyun, S.: Synthesis of silica aerogels from waterglass via new modified ambient drying. J. Mater. Sci. 37(11), 2237–2241 (2002) 19. Bangi, U.K.H., Rao, A.V., Rao, A.P.: A new route for preparation of sodium-silicate-based hydrophobic silica aerogels via ambientpressure drying. Sci. Technol. Adv. Mater. 9(3), 035006 (2008) 20. Bisson, A., Rigacci, A., Lecomte, D., Rodier, E., Achard, P.: Drying of silica gels to obtain aerogels: phenomenology and basic techniques. Dry. Technol. 21(4), 593–628 (2003) 21. Prakash, S.S., Brinker, C.J., Hurd, A.J.: Silica aerogel films at ambient pressure. J. Non-Cryst. Solids. 190(3), 264–275 (1995) 22. Prakash, S.S., Brinker, C.J., Hurd, A.J., Rao, S.M.: Silica aerogel films prepared at ambient pressure by using surface derivatization to induce reversible drying shrinkage. Nature. 374(6521), 439– 443 (1995)

A. M. Anderson and M. K. Carroll 23. Haereid, S., Einarsrud, A.: Mechanical strengthening of TMOSbased alcogels by aging in silane solutions. J. Sol-Gel Sci. Technol. 3, 1992–1204 (1994) 24. Bhagat, S.D., Oh, C.S., Kim, Y.H., Ahn, Y.S., Yeo, J.G.: Methyltrimethoxysilane based monolithic silica aerogels via ambient pressure drying. Microporous Mesoporous Mater. 100(1–3), 350–355 (2007) 25. Leventis, N., Palczer, A., McCorkle, L., Zhang, G., SotiriouLeventis, C.: Nanoengineered silica-polymer composite aerogels with no need for supercritical fluid drying. J. Sol-Gel Sci. Technol. 35, 99–105 (2005) 26. He, S., Li, Z., Shi, X., Yang, H., Gong, L., Cheng, X.: Rapid synthesis of sodium silicate based hydrophobic silica aerogel granules with large surface area. Adv. Powder Technol. 26(2), 537– 541 (2015) 27. Cheng, Y., Xia, M., Luo, F., Li, N., Guo, C., Wei, C.: Effect of surface modification on physical properties of silica aerogels derived from fly ash acid sludge. Colloids Surf. Physicochem. Eng. Aspects. 490, 200–206 (2016) 28. Mahadik, D.B., Rao, A.V., Rao, A.P., Wagh, P.B., Ingale, S.V., Gupta, S.C.: Effect of concentration of trimethylchlorosilane (TMCS) and hexamethyldisilazane (HMDZ) silylating agents on surface free energy of silica aerogels. J. Colloid Interface Sci. 356(1), 298–302 (2011) 29. Gurav, J.L., Rao, A.V., Bangi, U.K.H.: Hydrophobic and low density silica aerogels dried at ambient pressure using TEOS precursor. J. Alloys Compd. 471(1–2), 296–302 (2009) 30. Liu, G., Zhou, B., Ni, X., Shen, J., Wu, G., Du, A., Zu, G.: Influence of thermal process on microstructural and physical properties of ambient pressure dried hydrophobic silica aerogel monoliths. J. Sol-Gel Sci. Technol. 62(2), 126–133 (2012) 31. Perissinotto, A.P., Awano, C.M., De Vicente, F.S., Donatti, D.A., Mesquita, A., Da Silva, L.F., Vollet, D.R.: Structure and diffuseboundary in hydrophobic and sodium dodecyl sulfate-modified silica aerogels. Microporous Mesoporous Mater. 223, 196–202 (2016) 32. Rao, A.P., Rao, A.V.: Improvement in optical transmission of the ambient pressure dried hydrophobic nanostructured silica aerogels with mixed silylating agents. J. Non-Cryst. Solids. 355(45–47), 2260–2271 (2009) 33. Pajonk, G.M., Repellin-Lacroix, M., Abouarnadasse, S., Chaouki, J., Klavana, D.: From sol-gel to aerogels and cryogels. J. Non-Cryst. Solids. 121, 66–67 (1990) 34. Kalinin, S., Kheifets, L., Mamchik, A., Knot’ko, A., Vertigel, A.: Influence of the drying technique on the structure of silica gels. J. Sol-Gel Sci. Technol. 15(1), 31–35 (1999) 35. Cervin, N.T., Aulin, C., Larsson, P.T., Wågberg, L.: Ultra porous nanocellulose aerogels as separation medium for mixtures of oil/water liquids. Cellulose. 19(2), 401–410 (2012) 36. Sai, H., Xing, L., Xiang, J., Cui, L., Jiao, J., Zhao, C., Li, Z., Li, F., Zhang, T.: Flexible aerogels with interpenetrating network structure of bacterial cellulose-silica composite from sodium silicate precursor via freeze drying process. RSC Adv. 4(57), 30453– 30461 (2014) 37. Yun, L., Zhao, J., Kang, X., Du, Y., Yuan, X., Hou, X.: Preparation and properties of monolithic and hydrophobic gelatin–silica composite aerogels for oil absorption. J. Sol-Gel Sci. Technol. 83(1), 197–206 (2017) 38. Ma, Q., Liu, Y., Dong, Z., Wang, J., Hou, X.: Hydrophobic and nanoporous chitosan-silica composite aerogels for oil absorption. J. Appl. Polym. Sci. 132(15), 41770 (2015) 39. Pan, Y., He, S., Gong, L., Cheng, X., Li, C., Li, Z., Liu, Z., Zhang, H.: Low thermal-conductivity and high thermal stable silica aerogel based on MTMS/water-glass co-precursor prepared by freeze drying. Mater. Des. 113, 246 (2017) 40. Scherer, G.W.: Freezing gels. J. Non-Cryst. Solids. 155(1), 1–25 (1993)

14

Hydrophobic Silica Aerogels

41. Kistler, S.S.: Coherent expanded aerogels. J. Phys. Chem. 13, 52– 64 (1932) 42. Phalippou, J., Woignier, T., Prassas, M.: Glasses from aerogels. J. Mater. Sci. 25(7), 3111–3117 (1990) 43. Danilyuk, A.F., Gorodetskaya, T.A., Barannik, G.B., Lyakhova, V. F.: Supercritical extraction as a method for modifying the structure of supports and catalysts. React. Kinet. Catal. Lett. 63(1), 193– 199 (1998) 44. Pajonk, G.M., Rao, A.V., Sawant, B.M., Parvathy, N.N.: Dependence of monolithicity and physical properties of TMOS silica aerogels on gel aging and drying conditions. J. Non-Cryst. Solids. 209(1), 40–50 (1997) 45. Lermontov, S.A., Sipyagina, N.A., Malkova, A.N., Baranchikov, A.E., Erov, K.E., Petukhov, D.I., Ivanov, V.K.: Methyltrimethoxysilane-based elastic aerogels: effects of the supercritical medium on structure-sensitive properties. Russ. J. Inorg. Chem. 60(4), 488–492 (2015) 46. Tewari, P.H., Hunt, A.J., Lofftus, K.: Ambient-temperature supercritical drying of transparent silica aerogels. Mater. Lett. 3(9), 363–367 (1985) 47. Van Bommel, M.J., de Haan, A.B.: Drying of silica aerogel with supercritical carbon dioxide. J. Non-Cryst. Solids. 186, 78–82 (1995) 48. Sanz-Moral, L.M., Rueda, M., Nieto, A., Novak, Z., Knez, Ž., Martín, Á.: Gradual hydrophobic surface functionalization of dry silica aerogels by reaction with silane precursors dissolved in supercritical carbon dioxide. J. Supercrit. Fluids. 84, 74–79 (2013) 49. Poco, J.F., Coronado, P.R., Pekala, R.W., Hrubesh, L.W.: A rapid supercritical extraction process for the production of silica aerogels. Mater. Res. Soc. Symp. 431, 297–302 (1996) 50. Scherer, G.W., Gross, J., Hrubesh, L.W., Coronado, P.R.: Optimization of the rapid supercritical extraction process for aerogels. J. Non-Cryst. Solids. 311(3), 259–272 (2002) 51. Mahadik, D.B., Lee, Y.K., Chavan, N.K., Mahadik, S.A., Park, H.: Monolithic and shrinkage-free hydrophobic silica aerogels via new rapid supercritical extraction process. J. Supercrit. Fluids. 107, 84– 91 (2016) 52. Jung, H., Lee, Y.K., Parale, V.G., Cho, H.H., Mahadik, D.B., Park, H.: Hydrophobic silica composite aerogels using poly(methyl methacrylate) by rapid supercritical extraction process. J. Sol-Gel Sci. Technol. 83(3), 692–697 (2017) 53. Gauthier, B.M., Bakrania, S.D., Anderson, A.M., Carroll, M.K.: A fast supercritical extraction technique for aerogel fabrication. J. Non-Cryst. Solids. 350, 238–243 (2004) 54. Roth, T.B., Anderson, A.M., Carroll, M.K.: Analysis of a rapid supercritical extraction aerogel fabrication process: prediction of thermodynamic conditions during processing. J. Non-Cryst. Solids. 354(31), 3685–3693 (2008) 55. Anderson, A.M., Wattley, C.W., Carroll, M.K.: Silica aerogels prepared via rapid supercritical extraction: effect of process variables on aerogel properties. J. Non-Cryst. Solids. 355(2), 101– 108 (2009) 56. Carroll, M.K., Anderson, A.M., Gorka, C.A.: Preparing silica aerogel monoliths via a rapid supercritical extraction method. J. Vis. Exp. JoVE. 84, 51421 (2014) 57. Lee, C., Kim, C.J.: Maximizing the giant liquid slip on superhydrophobic microstructures by nano-structureing their sidewalls. Langmuir. 25, 12812–12818 (2009) 58. Rao, A.V., Wagh, P.: Preparation and characterization of hydrophobic silica aerogels. Mater. Chem. Phys. 53(1), 13–18 (1998) 59. Hegde, N.D., Hirashima, H., Rao, A.V.: Two step sol-gel processing of TEOS based hydrophobic silica aerogels using trimethylethoxysilane as a co-precursor. J. Porous. Mater. 14(2), 165–171 (2007)

361 60. Leventis, N., Elder, I.A., Rolison, D.R., Anderson, M.L., Merzbacher, C.I.: Durable modification of silica aerogel monoliths with fluorescent 2,7-diazapyrenium moieties sensing oxygen near the speed of open-air diffusion. Chem. Mater. 11, 2837–2845 (1999) 61. Leventis, N., Rawashdeh, A.-M.M., Elder, I.A., Yang, J., Dass, A., Sotiriou-Leventis, C.: Synthesis and characterization of Ru(II) Tris (1,10-phenanthroline)-electron acceptor dyads incorporating the 4-benzoyl-N-methylpyridinium cation or N-benzyl-N′-methylviologen improving the dynamic range, sensitivity and response time of sol-gel based optical oxygen sensors. Chem. Mater. 16, 1493–1506 (2004) 62. Lee, K.H., Kim, S.Z., Yoo, K.I.P.: Low-density, hydrophobic aerogels. J. Non-Cryst. Solids. 186, 18–22 (1995) 63. Smirnova, I., Suttiruengwong, S., Arlt, W.: Feasibility study of hydrophilic and hydrophobic silica aerogels as drug delivery systems. J. Non-Cryst. Solids. 350, 54–60 (2004) 64. Özbakır, Y., Jonáš, A., Kiraz, A., Erkey, C.: Total internal reflection-based optofluidic waveguides fabricated in aerogels. J. Sol-Gel Sci. Technol. 84, 522–534 (2017) 65. Anderson, A.M., Carroll, M.K., Green, E.C., Melville, J.M., Bono, M.S.: Hydrophobic silica aerogels prepared via rapid supercritical extraction. J. Sol-Gel Sci. Technol. 53(2), 199–207 (2010) 66. Lermontov, S.A., Malkova, A.N., Sipyagina, N.A., Baranchikov, A.E., Petukhov, D.I., Ivanov, V.K.: Hydrophobicity/hydrophilicity control for SiO2-based aerogels: the role of a supercritical solvent. Russ. J. Inorg. Chem. 60(10), 1169–1172 (2015) 67. Schwertfeger, F., Glaubitt, W., Schubert, U.: Hydrophobic aerogels from Si (OMe)/MeSi (OMe) mixtures. J. Non-Cryst. Solids. 145, 85–89 (1992) 68. Schwertfeger, F., Hüsing, N., Schubert, U.: Influence of the nature of organic groups on the properties of organically modified silica aerogels. J. Sol-Gel Sci. Technol. 2(1), 103–108 (1994) 69. Hüsing, N., Schwertfeger, F., Tappert, W., Schubert, U.: Influence of supercritical drying fluid on structure and properties of organically modified silica aerogels. J. Non-Cryst. Solids. 186, 37–43 (1995) 70. Rao, A.V., Kulkarni, M.M., Amalnerkar, D., Seth, T.: Surface chemical modification of silica aerogels using various alkyl-alkoxy/chloro silanes. Appl. Surf. Sci. 206(1–4), 262–270 (2003) 71. Rao, A.V., Haranath, D.: Effect of methyltrimethoxysilane as a synthesis component on the hydrophobicity and some physical properties of silica aerogels. Microporous Mesoporous Mater. 30(2–3), 267–273 (1999) 72. Wagh, P.B., Ingale, S.V., Gupta, S.C.: Comparison of hydrophobicity studies of silica aerogels using contact angle measurements with water drop method and adsorbed water content measurements made by Karl Fischer’s titration method. J. Sol-Gel Sci. Technol. 55(1), 73–78 (2010) 73. Ingale, S.V., Wagh, P.B., Tripathi, A.K., Kamble, V.S., Kumar, R., Gupta, S.C.: Physico-chemical properties of silica aerogels prepared from TMOS/MTMS mixtures. J. Porous. Mater. 18(5), 567– 572 (2011) 74. Rao, A.V., Kulkarni, M.M.: Hydrophobic properties of TMOS/ TMES-based silica aerogels. Mater. Res. Bull. 37, 1667–1677 (2002) 75. Rao, A.V., Kulkarni, M., Pajonk, G., Amalnerkar, D., Seth, T.: Synthesis and characterization of hydrophobic silica aerogels using trimethylethoxysilane as a co-precursor. J. Sol-Gel Sci. Technol. 27(2), 103–109 (2003) 76. Rao, A.V., Hegde, N.D., Shewale, P.M.: Imperviousness of the hydrophobic silica aerogels against various solvents and acids. Appl. Surf. Sci. 253(9), 4137–4141 (2007) 77. Martín, L., Ossó, J.O., Ricart, S., Roig, A., García, O., Sastre, R.: Organo-modified silica aerogels and implications for material hydrophobicity and mechanical properties. J. Mater. Chem. 18(2), 207–213 (2008)

14

362 78. El Rassy, H., Pierre, A.: NMR and IR spectroscopy of silica aerogels with different hydrophobic characteristics. J. Non-Cryst. Solids. 351(19–20), 1603–1610 (2005) 79. Kurahashi, M., Kanamori, K., Takeda, K., Kaji, H., Nakanishi, K.: Role of block copolymer surfactant on the pore formation in methylsilsesquioxane aerogel systems. RSC Adv. 2(18), 7166– 7173 (2012) 80. Mahadik, D.B., Venkateswara Rao, A., Parale, V.G., Kavale, M.S., Wagh, P.B., Ingale, S.V., Gupta, S.C.: Effect of surface composition and roughness on the apparent surface free energy of silica aerogel materials. Appl. Phys. Lett. 99(10), 104104 (2011) 81. Gutzov, S., Danchova, N., Karakashev, S.I., Khristov, M., Ivanova, J., Ulbikas, J.: Preparation and thermal properties of chemically prepared nanoporous silica aerogels. J. Sol-Gel Sci. Technol. 70(3), 511–516 (2014) 82. Mazrouei-Sebdani, Z., Javazmi, L., Khoddami, A., ShamsGhahfarokhi, F., Low, T.: Fabrication of a silica aerogel and examination of its hydrophobic properties via contact angle and 3M water repellency tests. In: IOP Conference Series: Materials Science and Engineering, vol. 204, no. 1, p. 012014. IOP Publishing (2017) 83. Rao, A.V., Pajonk, G.M.: Effect of methyltrimethoxysilane as a co-precursor on the optical properties of silica aerogels. J. Non-Cryst. Solids. 285(1–3), 202–209 (2001) 84. Rao, A.V., Pajonk, G., Haranath, D.: Synthesis of hydrophobic aerogels for transparent window insulation applications. Mater. Sci. Technol. 17(3), 343–348 (2001) 85. Wagh, P.B., Ingale, S.V.: Comparison of some physico-chemical properties of hydrophilic and hydrophobic silica aerogels. Ceram. Int. 28(10), 43–50 (2002) 86. Rao, A.V., Kulkarni, M.M.: Effect of glycerol additive on physical properties of hydrophobic silica aerogels. Mater. Chem. Phys. 77(3), 819–825 (2002) 87. Rodriguez, J.E., Anderson, A.M., Carroll, M.K.: Hydrophobicity and drag reduction properties of surfaces coated with silica aerogels and xerogels. J. Sol-Gel Sci. Technol. 71(3), 490–500 (2014) 88. Fei, T., Chen, H., Lin, J.: Transparent superhydrophobic films possessing high thermal stability and improved moisture resistance from the deposition of MTMS-based aerogels. Colloids Surf. Physicochem. Eng. Aspects. 443, 255–264 (2014) 89. Yun, S., Guo, T., Zhang, J., He, L., Li, Y., Li, H., Zhu, X., Gao, Y.: Facile synthesis of large-sized monolithic methyltrimethoxysilanebased silica aerogel via ambient pressure drying. J. Sol-Gel Sci. Technol. 83(1), 53–63 (2017) 90. Rao, A.V., Kalesh, R.R.: Comparative studies of the physical and hydrophobic properties of TEOS based silica aerogels using different co-precursors. Sci. Technol. Adv. Mater. 4(6), 509–515 (2003) 91. Rao, A.V., Kalesh, R.R., Amalnerkar, D.P., Seth, T.: Synthesis and characterization of hydrophobic TMES/TEOS based silica aerogels. J. Porous. Mater. 10(1), 23–29 (2003) 92. Rao, A.V., Kalesh, R., Pajonk, G.: Hydrophobicity and physical properties of TEOS based silica aerogels using phenyltriethoxysilane as a synthesis component. J. Mater. Sci. 38(21), 4407–4413 (2003) 93. Štandeker, S., Novak, Z., Knez, Ž.: Adsorption of toxic organic compounds from water with hydrophobic silica aerogels. J. Colloid Interface Sci. 310(2), 362–368 (2007) 94. Hrubesh, L.W., Coronado, P.R., Satcher, J.H.: Solvent removal from water with hydrophobic aerogels. J. Non-Cryst. Solids. 285(1–3), 328–332 (2001) 95. Reynolds, J.G., Coronado, P.R., Hrubesh, L.W.: Hydrophobic aerogels for oil-spill cleanup-intrinsic absorbing properties. Energy Sources. 23, 831–834 (2001) 96. Reynolds, J.G., Coronado, P.R., Hrubesh, L.W.: Hydrophobic aerogels for oil-spill cleanup-synthesis and characterization. J. Non-Cryst. Solids. 292, 127–137 (2001) 97. Lin, Y., Chen, C., Tung, K., Wei, T., Lu, S., Chang, K.: Mesoporous fluorocarbon-modified silica aerogel membranes enabling

A. M. Anderson and M. K. Carroll long-term continuous CO2 capture with large absorption flux enhancements. ChemSusChem. 6(3), 437–442 (2013) 98. Lin, Y., Chang, J., Ye, Q., Tung, K.: Hydrophobic fluorocarbonmodified silica aerogel tubular membranes with excellent CO2 recovery ability in membrane contactors. Appl. Energy. 154, 21–25 (2015) 99. Duan, Y., Jana, S.C., Lama, B., Espe, M.P.: Hydrophobic silica aerogels by silylation. J. Non-Cryst. Solids. 437, 26–33 (2016) 100. Pauthe, M.: Hydrophobic silica CO2 aerogels. J. Non-Cryst. Solids. 155, 110–114 (1993) 101. Hegde, N.D., Rao, A.V.: Organic modification of TEOS based silica aerogels using hexadecyltrimethoxysilane as a hydrophobic reagent. Appl. Surf. Sci. 253(3), 1566–1572 (2006) 102. Yun, S., Luo, H., Gao, Y.: Low-density, hydrophobic, highly flexible ambient-pressure-dried monolithic bridged silsesquioxane aerogels. J. Mater. Chem. A. 3(7), 3390–3398 (2015) 103. Rao, A.V., Kalesh, R.R.: Organic surface modification of TEOS based silica aerogels synthesized by co-precursor and derivatization methods. J. Sol-Gel Sci. Technol. 30(3), 141–147 (2004) 104. Bhagat, S.D., Rao, A.V.: Surface chemical modification of TEOS based silica aerogels synthesized by two step (acid–base) sol–gel process. Appl. Surf. Sci. 252(12), 4289–4297 (2006) 105. Jeong, A.Y., Koo, S.M., Kim, D.P.: Characterization of hydrophobic SiO2 powders prepared by surface modification on wet gel. J. Sol-Gel Sci. Technol. 19(1), 483–487 (2000) 106. Wei, T., Chang, T., Lu, S., Chang, Y.: Preparation of monolithic silica aerogel of low thermal conductivity by ambient pressure drying. J. Am. Ceram. Soc. 90(7), 2003–2007 (2007) 107. Rao, A.P., Pajonk, G., Rao, A.V.: Effect of preparation conditions on the physical and hydrophobic properties of two step processed ambient pressure dried silica aerogels. J. Mater. Sci. 40(13), 3481–3489 (2005) 108. Rao, A.P., Rao, A.V., Pajonk, G.: Hydrophobic and physical properties of the two step processed ambient pressure dried silica aerogels with various exchanging solvents. J. Sol-Gel Sci. Technol. 36(3), 285–292 (2005) 109. Roig, A., Molins, E., Rodríguez, E., Martínez, S., Moreno-Mañas, M., Vallribera, A.: Superhydrophobic silica aerogels by fluorination at the gel stage. Chem. Commun. 2004(20), 2316–2317 (2004) 110. Mahadik, S.A., Pedraza, F., Parale, V.G., Park, H.H.: Organically modified silica aerogel with different functional silylating agents and effect on their physico-chemical properties. J. Non-Cryst. Solids. 453, 164–171 (2016) 111. Shi, F., Wang, L., Liu, J.: Synthesis and characterization of silica aerogels by a novel fast ambient pressure drying process. Mater. Lett. 60(29–30), 3718–3722 (2006) 112. Hwang, S., Jung, H., Hyun, S., Ahn, Y.: Effective preparation of crack-free silica aerogels via ambient drying. J. Sol-Gel Sci. Technol. 41(2), 139–146 (2007) 113. Rao, A.P., Rao, A.V., Bangi, U.K.H.: Low thermalconductive, transparent and hydrophobic ambient pressure dried silica aerogels with various preparation conditions using sodium silicate solutions. J. Sol-Gel Sci. Technol. 47(1), 85–94 (2008) 114. Rao, A.P., Rao, A.V., Pajonk, G., Shewale, P.M.: Effect of solvent exchanging process on the preparation of the hydrophobic silica aerogels by ambient pressure drying method using sodium silicate precursor. J. Mater. Sci. 42(20), 8418–8425 (2007) 115. Rao, A.P., Rao, A.V., Pajonk, G.: Hydrophobic and physical properties of the ambient pressure dried silica aerogels with sodium silicate precursor using various surface modification agents. Appl. Surf. Sci. 253(14), 6032–6040 (2007) 116. Rao, A.P., Rao, A.V., Gurav, J.L.: Effect of protic solvents on the physical properties of the ambient pressure dried hydrophobic silica aerogels using sodium silicate precursor. J. Porous. Mater. 15(5), 507–512 (2008) 117. Rao, A.P., Rao, A.V.: Microstructural and physical properties of the ambient pressure dried hydrophobic silica aerogels with various solvent mixtures. J. Non-Cryst. Solids. 354(1), 10–18 (2008)

14

Hydrophobic Silica Aerogels

118. Shewale, P.M., Rao, A.V., Gurav, J.L., Rao, A.P.: Synthesis and characterization of low density and hydrophobic silica aerogels dried at ambient pressure using sodium silicate precursor. J. Porous. Mater. 16(1), 101–108 (2009) 119. Shewale, P.M., Rao, A.V., Rao, A.P.: Effect of different trimethyl silylating agents on the hydrophobic and physical properties of silica aerogels. Appl. Surf. Sci. 254(21), 6902–6907 (2008) 120. Halim, A., Alif, Z., Yajid, M., Azizi, M., Hamdan, H.: Synthesis and characterization of rice husk ash derived-silica aerogel beads prepared by ambient pressure drying. In: Key Engineering Materials, vol. 694, pp. 106–110. Trans Tech Publications (2016) 121. Liu, S., Wei, Q., Cui, S., Nie, Z., Du, M., Li, Q.: Hydrophobic silica aerogel derived from wheat husk ash by ambient pressure drying. J. Sol-Gel Sci. Technol. 78(1), 60–67 (2016) 122. Mermer, N.K., Yilmaz, M.S., Ozdemir, O.D., Piskin, M.B.: The synthesis of silica-based aerogel from gold mine waste for thermal insulation. J. Therm. Anal. Calorim. 129(3), 1807–1812 (2017) 123. Zhu, J.M., Cai, H.W., Dang, N.N.: Study of synthesis of SiO2 aerogel from coal gangue at an ambient pressure. In: Advanced Materials Research, vol. 997, pp. 614–617. Trans Tech Publications (2014) 124. Wang, B., Ma, H., Song, K.: Preparation and characterization of silica aerogels from diatomite via ambient pressure drying. Russ. J. Phys. Chem. A. 88(7), 1196–1201 (2014) 125. Hu, W., Li, M., Chen, W., Zhang, N., Li, B., Wang, M., Zhao, Z.: Preparation of hydrophobic silica aerogel with kaolin dried at ambient pressure. Colloids Surf. A Physicochem. Eng. Asp. 501, 83–91 (2016) 126. Ilhan, U.F., Fabrizio, E.F., McCorkle, L., Scheiman, D.A., Dass, A., Palczer, A., Meador, M.B., Johnston, J.C., Leventis, N.: Hydrophobic monolithic aerogels by nanocasting polystyrene on aminemodified silica. J. Mater. Chem. 16(29), 3046–3054 (2006) 127. Leventis, N., Sotiriou-Leventis, C., Zhang, G., Rawashdeh, A.-M. M.: Nano-engineering strong silica aerogels. Nano Lett. 2, 957– 960 (2002) 128. Boday, D.J., Stover, R.J., Muriithi, B., Loy, D.A.: Strong, low density, hexylene- and phenylene-bridged polysilsesquioxane aerogel-polycyanoacrylate composites. J. Mater. Sci. 46(19), 6371–6377 (2011) 129. Seraji, M.M., Sameri, G., Davarpanah, J., Bahramian, A.R.: The effect of high temperature sol-gel polymerization parameters on the microstructure and properties of hydrophobic phenol-formaldehyde/ silica hybrid aerogels. J. Colloid Interface Sci. 493, 103–110 (2017) 130. He, F., Chao, S., Gao, Y., He, X., Li, M.: Fabrication of hydrophobic silica–cellulose aerogels by using dimethyl sulfoxide (DMSO) as solvent. Mater. Lett. 137, 167–169 (2014) 131. Hayase, G., Kanamori, K., Nakanishi, K.: New flexible aerogels and xerogels derived from methyltrimethoxysilane/ dimethyldimethoxysilane co-precursors. J. Mater. Chem. 21(43), 17077–17079 (2011) 132. Lin, Y.F., Ko, C.C., Chen, C.H., Tung, K.L., Chang, K.S., Chung, T.W.: Sol–gel preparation of polymethylsilsesquioxane aerogel membranes for CO2 absorption fluxes in membrane contactors. Appl. Energy. 129, 25–31 (2014) 133. Tillotson, T.M., Foster, K.G., Reynolds, J.G.: Fluorine-induced hydrophobicity in silica aerogels. J. Non-Cryst. Solids. 350, 202–208 (2004) 134. Yang, L., Wu, H., Chen, Q., Zhou, X.: Effects of preparation parameters on SiO2 aerogels by single-factor and orthogonal experiments. Mater Res Innov. 19(Suppl 2), S2–S90 (2015) 135. Smith, L.S.: The Effects of High Temperatures on the Physical Properties of Silica Aerogels. Summer Report. Union College (2011) 136. Liu, H., Sha, W., Cooper, A.T., Fan, M.: Preparation and characterization of a novel silica aerogel as adsorbent for toxic organic compounds. Colloids Surf. Physicochem. Eng. Aspects. 347(1–3), 38–44 (2009)

363 137. Pierre, A.C., Pajonk, G.M.: Chemistry of aerogels and their applications. Chem. Rev. 102, 4243–4265 (2002) 138. Akimov, Y.K.: Fields of application of aerogels (review). Instrum. Exp. Tech. 46(3), 287–299 (2003) 139. Baetens, R., Jelle, B.P., Gustavsen, A.: Aerogel insulation for building applications: a state-of-the-art review. Energ. Buildings. 43(4), 761–769 (2011) 140. Tabata, M., Adachi, I., Hatakeyama, Y., Kawai, H., Morita, T., Sumiyoshi, T.: Large-area silica aerogel for use as Cherenkov radiators with high refractive index, developed by supercritical carbon dioxide drying. J. Supercrit. Fluids. 110, 183–192 (2016) 141. Adebajo, M., Frost, R., Kloprogge, J., Carmody, O., Kokot, S.: Porous materials for oil spill cleanup: a review of synthesis and absorbing properties. J. Porous. Mater. 10(3), 159–170 (2003) 142. Maleki, H.: Recent advances in aerogels for environmental remediation applications: a review. Chem. Eng. J. 300, 98–118 (2016) 143. Wang, D., McLaughlin, E., Pfeffer, R., Lin, Y.S.: Adsorption of oils from pure liquid and oil–water emulsion on hydrophobic silica aerogels. Sep. Purif. Technol. 99, 28–35 (2012) 144. Rao, A.V., Hegde, N.D., Hirashima, H.: Absorption and desorption of organic liquids in elastic superhydrophobic silica aerogels. J. Colloid Interface Sci. 305(1), 124–132 (2007) 145. Coleman, S.J., Coronado, P.R., Maxwell, R.S., Reynolds, J.G.: Granulated activated carbon modified with hydrophobic silica aerogel-potential composite materials for the removal of uranium from aqueous solutions. Environ. Sci. Technol. 37(10), 2286–2290 (2003) 146. Cui, S., Liu, X., Liu, Y., Shen, X., Lin, B., Han, G., Wu, Z.: Adsorption properties of nitrobenzene in wastewater with silica aerogels. Sci. China Technol. Sci. 53(9), 2367–2371 (2010) 147. Loche, D., Malfatti, L., Carboni, D., Alzari, V., Mariani, A., Casula, M.F.: Incorporation of graphene into silica-based aerogels and application for water remediation. RSC Adv. 6(71), 66516– 66523 (2016) 148. He, S., Cheng, X., Li, Z., Shi, X., Yang, H., Zhang, H.: Green and facile synthesis of sponge-reinforced silica aerogel and its pumping application for oil absorption. J. Mater. Sci. 51(3), 1292–1301 (2016) 149. Lin, Y., Ko, C., Chen, C., Tung, K., Chang, K.: Reusable methyltrimethoxysilane-based mesoporous water-repellent silica aerogel membranes for CO2 capture. RSC Adv. 4(3), 1456–1459 (2014) 150. Stergar, J., Maver, U.: Review of aerogel-based materials in biomedical applications. J. Sol-Gel Sci. Technol. 77(3), 738–752 (2016) 151. Gorle, B.S.K., Smirnova, I., McHugh, M.A.: Adsorption and thermal release of highly volatile compounds in silica aerogels. J. Supercrit. Fluids. 48(1), 85–92 (2009) 152. Giray, S., Bal, T., Kartal, A.M., Kizilel, S., Erkey, C.: Controlled drug delivery through a novel PEG hydrogel encapsulated silica aerogel system. J. Biomed. Mater. Res. – Part A. 100A(5), 1307– 1315 (2012) 153. Murillo-Cremaes, N., López-Periago, A.M., Saurina, J., Roig, A., Domingo, C.: Nanostructured silica-based drug delivery vehicles for hydrophobic and moisture sensitive drugs. J. Supercrit. Fluids. 73, 34–42 (2013) 154. Buisson, P., Hernandez, C., Pierre, M., Pierre, A.C.: Encapsulation of lipases in aerogels. J. Non-Cryst. Solids. 285(1), 295–302 (2001) 155. Orçaire, O., Buisson, P., Pierre, A.C.: Application of silica aerogel encapsulated lipases in the synthesis of biodiesel by transesterification reactions. J. Mol. Catal. B Enzym. 42(3–4), 106–113 (2006) 156. El Rassy, H., Buisson, P., Bouali, B., Perrard, A., Pierre, A.C.: Surface characterization of silica aerogels with different proportions of hydrophobic groups, dried by the CO2 supercritical method. Langmuir. 19(2), 358–363 (2003)

14

364 157. El Rassy, H., Maury, S., Buisson, P., Pierre, A.: Hydrophobic silica aerogel–lipase biocatalysts possible interactions between the enzyme and the gel. J. Non-Cryst. Solids. 350, 23–30 (2004) 158. Gao, S., Wang, Y., Wang, W., Luo, G., Dai, Y.: Enhancing performance of lipase immobilized on methyl-modified silica aerogels at the adsorption and catalysis processes: effect of cosolvents. J. Mol. Catal. B. 62(3–4), 218–224 (2010) 159. Li, X.M., Reinhoudt, D., Crego-Calama, M.: What do we need for a superhydrophobic surface? A review on the recent progress in the preparation of superhydrophobic surfaces. Chem. Soc. Rev. 36, 1350–1368 (2009) 160. Oh, J.K., Perez, K., Kohli, N., Kara, V., Li, J., Min, Y., Akbulut, M.: Hydrophobically-modified silica aerogels: novel food-contact surfaces with bacterial anti-adhesion properties. Food Control. 52, 132–141 (2015) 161. Oh, J.K., Kohli, N., Zhang, Y., Min, Y., Jayaraman, A., CisnerosZevallos, L., Akbulut, M.: Nanoporous aerogel as a bacteria repelling hygienic material for healthcare environment. Nanotechnology. 27(8), 085705 (2016) 162. Rao, A.V., Kulkarni, M.M., Bhagat, S.D.: Transport of liquids using hydrophobic aerogels. J. Colloid Interface Sci. 285, 413–418 (2005) 163. Truesdell, R., Mammoli, A., Vorobieff, P., Van Swol, F., Brinker, C.J.: Drag reduction on a patterned superhydrophobic surface. Phys. Rev. Lett. 97(4), 044504 (2006) 164. Samaha, M.A., Tafreshi, H.V., Gad-el-Hak, M.: Effects of hydrostatic pressure on the drag reduction of submerged aerogel-particle coatings. Colloids Surf. A Physicochem. Eng. Asp. 399, 62–70 (2012) 165. Kim, A., Kim, H., Lee, C., Kim, J.: Effective three-dimensional superhydrophobic aerogel-coated channel for high efficiency water-droplet transport. Appl. Phys. Lett. 104(8), 081601 (2014) 166. Tabata, M., Kawai, H., Yano, H., Imai, E., Hashimoto, H., Yokobori, S.I., Yamagishi, A.: Ultralow-density double-layer silica aerogel fabrication for the intact capture of cosmic dust in low-Earth orbits. J. Sol-Gel Sci. Technol. 77(2), 325–334 (2016) 167. Özbakır, Y., Jonas, A., Kiraz, A., Erkey, C.: Aerogels for optofluidic waveguides. Micromachines. 8(4), 98 (2017) 168. Huang, X., Liu, Z., Xia, W., Zou, R., Han, R.P.S.: Alkylated phase change composites for thermal energy storage based on surfacemodified silica aerogels. J. Mater. Chem. A. 3(5), 1935–1940 (2015) 169. Parale, V.G., Mahadik, D.B., Kavale, M.S., Rao, A.V., Patil, R.A., Ma, Y., Mullens, S., Vhatkar, R.S.: Effect of aluminium and copper acetylacetonate on physico-chemical properties of tetraethoxysilane based silica aerogels. J. Porous. Mater. 20(3), 563– 570 (2013) 170. Kim, C.E., Yoon, J.S., Hwang, H.J.: Synthesis of nanoporous silica aerogel by ambient pressure drying. J. Sol-Gel Sci. Technol. 49(1), 47 (2009) 171. Bangi, U.K., Rao, A.P., Hirashima, H., Rao, A.V.: Physicochemical properties of ambiently dried sodium silicate based aerogels catalyzed with various acids. J. Sol-Gel Sci. Technol. 50(1), 87–97 (2009) 172. Liu, M.L., Yang, D.A., Qu, Y.F.: Effect of different chemical additives and heat-treatment on ambient pressure dried silica aerogels. J. Exp. Nanosci. 5(1), 83–91 (2010)

A. M. Anderson and M. K. Carroll 173. Bangi, U.K., Dhere, S.L., Rao, A.V.: Influence of various processing parameters on water-glass-based atmospheric pressure dried aerogels for liquid marble purpose. J. Mater. Sci. 45(11), 2944–2951 (2010) 174. Cui, S., Liu, Y., Fan, M.H., Cooper, A.T., Lin, B., Liu, X., Han, G. F., Shen, X.D.: Temperature dependent microstructure of MTES modified hydrophobic silica aerogels. Mater. Lett. 65(4), 606– 609 (2011) 175. Hong, S.K., Yoon, M.Y., Hwang, H.J.: Fabrication of spherical silica aerogel granules from water glass by ambient pressure drying. J. Am. Ceram. Soc. 94(10), 3198–3201 (2011) 176. Budunoglu, H., Yildirim, A., Guler, M.O., Bayindir, M.: Highly transparent, flexible, and thermally stable superhydrophobic ORMOSIL aerogel thin films. ACS Appl. Mater. Interfaces. 3(2), 539–545 (2011) 177. Yu, Y., Wu, X., Guo, D., Fang, J.: Preparation of flexible, hydrophobic, and oleophilic silica aerogels based on a methyltriethoxysilane precursor. J. Mater. Sci. 49(22), 7715– 7722 (2014) 178. He, S., Cheng, X.D., Bi, H.J., Yang, H.: Synthesis and effect of DCCA on the ambient pressure dried, TEOS-based silica aerogel. In: Advanced Materials Research, vol. 1044, pp. 119–123. Trans Tech Publications (2014) 179. He, S., Huang, D., Bi, H., Li, Z., Yang, H., Cheng, X.: Synthesis and characterization of silica aerogels dried under ambient pressure bed on water glass. J. Non-Cryst. Solids. 410, 58–64 (2015) 180. Yu, H., Liang, X., Wang, J., Wang, M., Yang, S.: Preparation and characterization of hydrophobic silica aerogel sphere products by co-precursor method. Solid State Sci. 48, 155–162 (2015) 181. Ren, H., Zhu, J., Bi, Y., Xu, Y., Zhang, L.: One-step fabrication of transparent hydrophobic silica aerogels via in situ surface modification in drying process. J. Sol-Gel Sci. Technol. 80(3), 635– 641 (2016) 182. Li, M., Jiang, H., Xu, D., Hai, O., Zheng, W.: Low density and hydrophobic silica aerogels dried under ambient pressure using a new co-precursor method. J. Non-Cryst. Solids. 452, 187–193 (2016) 183. Morales-Florez, V., Piñero, M., Braza, V., del Mar Mesa, M., Esquivias, L., de la Rosa-Fox, N.: Absorption capacity, kinetics and mechanical behaviour in dry and wet states of hydrophobic DEDMS/TEOS-based silica aerogels. J. Sol-Gel Sci. Technol. 81(2), 600–610 (2017) 184. Aoki, Y., Shimizu, T., Kanamori, K., Maeno, A., Kaji, H., Nakanishi, K.: Low-density, transparent aerogels and xerogels based on hexylene-bridged polysilsesquioxane with bendability. J. Sol-Gel Sci. Technol. 81(1), 42–51 (2017) 185. Rao, A.P., Rao, A.V.: Modifying the surface energy and hydrophobicity of the low-density silica aerogels through the use of combinations of surface-modification agents. J. Mater. Sci. 45(1), 51– 63 (2010) 186. Wang, H., Liu, S.M., Zeng, L.K.: Preparation and properties of hydrophobic silica aerogels by ambient pressure drying method. In: Applied Mechanics and Materials, vol. 71, pp. 1040–1043. Trans Tech Publications (2011) 187. Pisal, A.A., Rao, A.V.: Development of hydrophobic and optically transparent monolithic silica aerogels for window panel applications. J. Porous. Mater. 24(3), 685–695 (2017)

14

Hydrophobic Silica Aerogels

365

14

Ann M. Anderson received her PhD from Stanford University in 1990. She is the Agnes S. MacDonald Professor of Mechanical Engineering at Union College where she teaches courses in the thermal fluid sciences and co-directs the Aerogel Fabrication Characterization and Applications Lab. She has expertise in aerogel manufacturing, fluids mechanics, and heat transfer.

Mary K. Carroll, the Dwane W. Crichton Professor of Chemistry at Union College, earned her PhD from Indiana University Bloomington in 1991. At Union, she teaches courses in introductory and analytical chemistry, mentors undergraduate research students, and co-directs the Aerogel Lab. Her current research activities focus on fabrication and characterization of aerogel materials for applications including chemical sensing and catalysis.

15

Superhydrophobic and Flexible Aerogels and Xerogels Derived from Organosilane Precursors Kazuyoshi Kanamori , Ana Stojanovic, Gerard M. Pajonk, Digambar Y. Nadargi , A. Venkateswara Rao, Kazuki Nakanishi Matthias M. Koebel

, and

Contents 15.1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368

15.3.2

15.2 15.2.1

Synthetic Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Organotrialkoxysilanes: Intrinsic Challenges Given by the Molecular Precursor System . . . . . . . . . . . . . . . . . . . . . . . . . . Aerogels Derived from Co-precursor Chemistries with Tetraalkoxysilanes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pure Organotrialkoxysilane-Based Aerogels . . . . . . . . . . . . . Aerogels from Organo-bridged Alkoxysilanes . . . . . . . . . . . Conversion and Chemical Modification of Surface Groups to Extend Material Functionality . . . . . . . . . . . . . . . .

15.3.3

15.2.2 15.2.3 15.2.4 15.2.5 15.3 15.3.1

369 369 372 372 374 375

Mechanical and Other Physical Properties of Aerogels from Organoalkoxysilanes . . . . . . . . . . . . . . . . . . . 376 Properties of Aerogels from MTMS-Based Systems . . . . 376

15.4 15.4.1 15.4.2 15.4.3 15.5 15.5.1 15.5.2

Properties of Aerogels from Organo-Bridged Bisalkoxysilanes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378 Properties of Aerogels Based on Polysiloxane Networks Extended Through Organic Polymerization Reactions (Doubly Crosslinked Systems) . . . . . . . . . . . . . . . . . . . . . . . . . . . 380 Hydrophobic Properties and Environmental Remediation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hydrophobic Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Environmental Remediation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Drag Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

380 380 382 385

Economic Considerations and Commercialization Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386 Economic Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386 Commercialization Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388 K. Kanamori Department of Chemistry, Graduate School of Science, Kyoto University, Kyoto, Japan e-mail: [email protected] A. Stojanovic · M. M. Koebel (*) Building Energy Materials and Components, Laboratory for Building Technologies, Empa, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland e-mail: [email protected]; [email protected]; [email protected] G. M. Pajonk Laboratoire des Matériaux et Procédés Catalytiques, Université Claude Bernard Lyon 1, Villeurbanne, France D. Y. Nadargi Building Energy Materials and Components, Laboratory for Building Technologies, Empa, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland School of Physical Sciences, Solapur University, Solapur, Maharashtra, India A. V. Rao Air Glass Laboratory, Department of Physics, Shivaji University, Kolhapur, Maharashtra, India K. Nakanishi Institute of Materials and Systems for Sustainability, Nagoya University, Nagoya, Japan e-mail: [email protected]

Abstract

The field of organosilane-derived aerogels is a rather young area of aerogel research. Inherently, organosilane precursors offer the advantage that they already contain at least one direct Si-C bond in the silane precursor, thus imparting specific chemical functionality. In particular, if used exclusively to prepare organo-modified polysiloxane aerogels, hydrophobicity originating from the organic substituent group(s) can bypass the need for a second post-modification (hydrophobization) step. Pioneered by early works, recent years have brought tremendous improvements in terms of methodology; combination with organic, hybrid (polymerization-based) chemistries, as well as ambient-drying protocols. Modern organosilane-based preparation techniques are costcompetitive with conventional silica aerogel synthetic routes but offer a tremendous added value potential through the use of suitable “secondary functionalization” protocols, particularly obvious in many examples in the form of improved mechanical properties (bendability, machinability, higher toughness, and e-modulus) but also

© Springer Nature Switzerland AG 2023 M. A. Aegerter et al. (eds.), Springer Handbook of Aerogels, Springer Handbooks, https://doi.org/10.1007/978-3-030-27322-4_15

367

368

K. Kanamori et al.

improved resistance against organic solvents. This chapter gives an overview over this trendsetting and commercially promising field of aerogels, a discussion on potential applications, as well as an outlook and brief analysis of the key challenges ahead. Keywords

Organic-inorganic hybrid aerogels · Hydrophobicity · Mechanical flexibility · Functionality · Ambient pressure drying · Commercialization

15.1

Introduction

The usage of chemical function-bearing organosilanes is widespread in both organic chemistry [1] and polymer chemistry [2, 3] and has gradually made its impact on the material science as well [4]. Organoalkoxysilanes R′4
nSi(OR)n (where R′ is the organic substitutional group(s); R is a linear or branched alkyl group(s) with typically 1–4 carbons, preferably methyl or ethyl; and n ¼ 1–3) are a unique class of organic silicon compounds that carry hydrolysable alkoxy groups Si-OR which can undergo typical sol–gel hydrolysis and polycondensation processes. They can interact and/or chemically react with both inorganic and organic substrates as well as with themselves and other silanes by interdependent hydrolysis/condensation reactions and form a variety of hybrid organic-inorganic structures or covalently attach to hydroxyl groups at the surface of metal oxide substrates. The functionalization of silica surface with alkylorganosilane or other organofunctional silanes in order to enhance the hydrophobicity of silica is very well known in literature [4]. When reacting with the silica surface, organosilane compound can decrease the surface energy or can be used as active sites for further modification. The advantage of replacing alkoxy by alkyl groups in silica aerogel synthesis is twofold. First, the number of possible covalent Si–O–Si linkages responsible for the silica network formation is reduced with increasing R′ substitution. Second, alkyl groups introduce hydrophobicity. During the sol-forming stage, such groups will have a tendency to orient toward one another which results in hydrophobic domains within the microstructure of the network. This, however, does not mean that hydrophobicity of a gel network can be increased by using siloxanes with more alkyl functional groups: with increasing R′/Si ratio, the number of bonding centers available for building polysiloxane network decreases. Trifunctional alkoxysilane

compounds such as methyltrimethoxysilane (MTMS) and methyltriethoxysilane (MTES) offer a reasonable compromise between hydrophobicity and stability of network structure. In combination with the reduced silica network connectivity, superhydrophobicity leads to mechanical properties that are completely different from those of native silica aerogels. Hydrophobicity is an important property of aerogels as it gives structural stability against humidity. Superhydrophobic silica aerogels are broadening the applications of aerogels in fields such as adsorbents for different nonpolar liquids, heavy metals and vapor CO2, self-cleaning surfaces, and drag reduction. Some of these applications will be discussed in detail in the following sections. In the early days of hydrophobic silica aerogel material development, specifically in the early 1990s, there was a significant academic interest in developing alternative strategies to render the resulting gels and aerogels hydrophobic. As a result, co-precursor methods, where various organofunctional silanes of the general formula R′4
nSi (OR)n were hydrolyzed and gelled together with tetraalkoxysilanes (tetramethoxysilane TMOS or tetraethoxysilane TEOS), were introduced, and later on these methods were extended to waterglass (WG)-based systems. Mechanistically, co-precursors condensate competitively with the silanol (Si–OH) and alkoxy (Si–OR) groups, to form a mixed network. The hydrophobicity of the aerogels using a co-precursor method is limited due to the fact that the co-precursor/precursor molar ratio cannot be increased above a certain value (e.g., for MTMS/TMOS it is around 4), because either there is cracking and monolithic aerogels cannot be obtained or the synthesis becomes troublesome. Due to the abovementioned limitations of co-precursor methods, in the last 15 years, there have been numerous studies reported using alkylorganosilanes as single precursor for the synthesis of aerogels and macroporous gels with foamlike morphologies. Generally speaking, gelation of MTMS and MTES is much more difficult than that of TMOS and TEOS due to an extensive cyclization under acidic conditions, premature phase separation over a broad pH range, and fewer functional groups for crosslinking. Loy et al. concluded that it was not possible to prepare monolithic polymethylsilsesquioxane (PMSQ) gels except at extremely high- or low-pH conditions, regardless of monomer or water concentration [5]. The pH of the sol–gel polymerizations of alkoxysilanes is generally recognized as one of the most important reaction and processing parameters, affecting the porosity, density, strength, transparency, and chemical structure of the resulting gels [6].

15

Superhydrophobic and Flexible Aerogels and Xerogels Derived from Organosilane Precursors

15.2

Synthetic Strategies

15.2.1 Organotrialkoxysilanes: Intrinsic Challenges Given by the Molecular Precursor System Here in this chapter, all the typical alkoxysilane precursor structures and typical compound names are summarized in Tables 15.1, 15.2, and 15.3, and the resulting networks and properties of aerogels are summarized in Table 15.4. Networks from organotrialkoxysilanes or organo-bridged bis (trialkoxysilanes) are termed as polyorganosilsesquioxane and organo-bridged polysilsesquioxane, respectively, since these polymer networks are built on silsesquioxane (SiO3/2) units. In addition, since the reaction mechanisms depending on acid- or base-catalyzed conditions are crucial in these organoalkoxysilane systems as discussed later, here we define the process codes as follows: AB2, two-step acid-base process consisting of acid-catalyzed hydrolysis and base-catalyzed condensation; A1, one-step

369

acid process of acid-catalyzed hydrolysis and condensation; B1, one-step base process of base-catalyzed hydrolysis and condensation. As mentioned in the Introduction, the amphiphilic nature of organoalkoxysilanes, R′4
nSi(OR)n (n ¼ 1–3), generally leads to hydrophobic polycondensates for n ¼ 2 and 3 that tend to segregate from aqueous solutions by phase separation, whereas organosilanes with n ¼ 1 only feature a single alkoxy group available for siloxane bond formation and thus are normally used as an end-capping hydrophobe during post-modification of gels. Understanding the nontrivial structure formation of such gels is essential to be able to control and design material properties: the colloidal aggregation of the condensates at nanometer scale usually competes with the coarsening of a phase-separating superstructure at micrometer scale (Fig. 15.1). If gelation occurs sufficiently before the macroscopic phase separation, transparent gels can be formed. Yet, if the progress of phase separation is faster, opaque macroporous gels or precipitates will be obtained. In addition, a high tendency for the

Table 15.1 Organotrialkoxysilane precursors (trifunctional, T series) Chemical structure

Compound name Trimethoxysilane (HTMS)

Ref. [5]

(RO)3Si

Methyltrimethoxysilane (MTMS)

[5, 11–13, 18, 20–22, 24]

(RO)3Si

Ethyltrimethoxysilane (ETMS)

[5, 27]

(RO)3Si

Vinyltrimethoxysilane (VTMS)

[5, 24, 25, 27, 35, 42]

Chloromethyltrimethoxysilane (ClMTMS)

[5, 28]

4-Chloromethylphenyltrimethoxysilane (ClMPhTMS)

[5]

3-Mercaptopropyltrimethoxysilane (MPTMS)

[8, 9, 24, 35, 36]

2-Diphenylphosphinoethyltrimethoxysilane (PhPTMS)

[8, 9]

3-Chloropropyltrimethoxysilane (ClPTMS)

[8, 9, 24]

3-Glycidoxypropyltrimethoxysilane (GLYTMS)

[9]

3-Methacryloxypropyltrimethoxysilane (MATMS)

[9, 24]

(RO)3Si

(RO)3Si

H

Cl

(RO)3Si CH2Cl (RO)3Si (RO)3Si

SH PPh2

(RO)3Si

Cl

(RO)3Si

O

(RO)3Si

O

O

O (RO)3Si

NCO

3-Isocyanatopropyltriethoxysilane (ISOTMS)

[9]

(RO)3Si

H N

3-Carbamatopropyltrimethoxysilane (CBTMS)

[9]

OCH3 O

(RO)3Si

N+ Cl–

N-trimethoxysilylpropyl-N,N,N-trimethylammonium chloride (TMAC)

[17]

(RO)3Si

NH2

3-Aminopropyltrimethoxysilane (APTMS)

[37]

Allyltrimethoxysilane (ATMS)

[42]

(RO)3Si

15

370

K. Kanamori et al.

Table 15.2 Organodialkoxysilane precursors (difunctional, D series) and organomonoalkoxysilane (monofunctional, M series) Chemical structure

(RO)2Si (RO)2Si (RO)2Si (RO)2Si

Compound name Dimethyldimethoxysilane (DMDMS) Dimethyldiethoxysilane (DMDES)

Ref. [21, 22, 25]

Vinylmethyldimethoxysilane (VMDMS)

[21–23, 41–43]

Phenylmethyldimethoxysilane (PMDMS)

[22]

3,3,3-Trifluoropropyldimethoxysilane (TFDMS)

[22]

Mercaptopropylmethyldimethoxysilane (MPMDMS)

[22]

Allylmethyldimethoxysilane (AMDMS)

[42]

Vinyldimethylmethoxysilane (VDMMS)

[43]

CF3 SH

(RO)2Si (RO)2Si (RO)Si

Table 15.3 Organo-bridged alkoxysilane precursors (bis-trifunctional, BT and bis-difunctional BD series) Chemical structure

(RO)3Si

Si(OR)3

(RO)3Si

Si(OR)3

(RO)3Si

Si(OR)3

(RO)3Si

Si(OR)3

(RO)3Si

Si(OR)3

(RO)3Si

Compound name 1,2-Bis(trimethoxysilyl)ethane (BTME)

Ref. [30]

1,6-Bis(trimethoxysilyl)hexane (BTMH)

[30–33]

1,8-Bis(trimethoxysilyl)octane (BTMO)

[30]

1,10-Bis(trimethoxysilyl)decane (BTMD)

[30]

1,2-Bis(trimethoxysilyl)tetradecane (BTMTD)

[30]

1,4-Bis(trimethoxysilyl)benzene (BTMB)

[31, 32]

Bis[3-(trimethoxysilyl)propyl]amine (BTMPA)

[34]

1,6-Bis(methyldiethoxysilyl)hexane (BMDEH)

[38]

1,4-Bis(methyldiethoxysilyl)benzene (BMDEB)

[38]

1,2-Bis(methyldiethoxysilyl)ethane (BMDE-ethy)

[39, 40]

1,2-Bis(methyldiethoxysilyl)ethene (BMDE-ethe)

[40]

Si(OR)3 (RO)3Si (RO)2Si

H N

Si(OR)3 Si(OR)2

(RO)2Si Si(OR)2 (RO)2Si

Si(OR)2

(RO)2Si

Si(OR)2

15

Superhydrophobic and Flexible Aerogels and Xerogels Derived from Organosilane Precursors

371

Table 15.4 Various polyorganosiloxane aerogel/xerogel materials and their typical structural/physical properties Network Co-network Co-network Co-network Co-network Co-network Polymethylsilsesquioxane (PMSQ) CH3SiO3/2 PMSQ PMSQ-CNF composite Co-network Co-network Co-network Polyethylsilsesquioxane (PESQ) CH3CH2SiO3/2 Polyvinylsilsesquioxane (PVSQ) CH2¼CHSiO3/2 Polychloromethylsilsesquioxane (PClMSQ) ClCH2SiO3/2 Hexylene-bridged polysilsesquioxane (HBPSQ) O3/2Si-(CH2)6-SiO3/2 Phenylene-bridged polysilsesquioxane (PBPSQ) O3/2Si-Ph-SiO3/2 HBPSQ Alkylthioether-bridged polysilsesquioxane O3/2Si
(CH2)3
S
(CH2)2
SiO3/2 Ethylene-bridged polymethylsiloxane (Ethy-BPMS) O2/2(CH3)Si–(CH2)2–Si(CH3)O2/2 Ethenylene-bridged polymethylsiloxane (Ethe-BPMS) O2/2(CH3)Si-CH¼CH-Si(CH3)O2/2 Polyvinylpolymethylsiloxane (PVPMS) [CH2CH(Si(CH3)O2/2)]n

*1

ρb*3/ g cm
3 0.287

SBET*4/ m2 g
1 638

Transparency +

B1

0.251

875

+

NA

8

B1

0.238

624



NA

8

B1

0.260

582



NA

9

B1

0.055

1140



NA

11

*2

*5

λ*6/ mW m
1 K
1 Ref. NA 9

Precursor 90TMOS, 10MPTMS 90TMOS, 10ClPTMS 90TMOS, 10PhPTMS 90TMOS, 10MATMS 71TMOS, 29MTMS MTMS

Catalysts B1

AB2

0.04–0.1

NA




NA

12

MTMS MTMS 60MTMS, 40DMDMS 80MTMS, 20VTMS 80MTMS, 20ClPTMS ETMS

AB2 AB2 AB2

0.04–0.45 0.02–0.19 0.12

575–622 525–732 NA

+ 


>14.9 >15.3 30–40

13 18 21,22

AB2

0.056

376




NA

24

AB2

0.061

1




NA

24

AB2

0.24–0.57

201–383



NA

27

VTMS

AB2

0.14–0.17

334–541

+

>15.3

27

ClMTMS

A1

0.17–0.28

NA

+

NA

28

BTMH

B1

0.093

NA




NA

31

BTMP

B1

0.097

808

+

NA

31

BTMH Syn. from VTMS and MPTMS

B1 A1

0.13–0.22 0.06–0.085

874–924 338–363

+


NA 47.1–56.5

33 35

BMDE-ethy

AB2

0.15–0.51

599–797

+

NA

39

BMDE-ethe

AB2

0.053–0.15

946

+

NA

40

VMDMS

B1

0.16–0.31

903–950



>15.0

41,42

*1

Acronyms of precursors are defined in Tables 15.1, 15.2, and 15.3. *2Types of sol–gel system classified by catalysts employed for hydrolysis and polycondensation: AB2 stands for two-step acid-base, A1 for one-step acid, B1 for one-step base processes. *3Bulk density. *4BET surface area. *5 Qualitative denotations as transparent (+), translucent (), and opaque (
). *6Thermal conductivity at ambient condition

formation of cyclic polysiloxane building block species, typified by polyhedral oligomeric silsesquioxanes (POSS), which somehow contradicts the formation of highly crosslinked, three-dimensional siloxane networks, renders the formation of stable chemical networks from alkoxysilane

precursors even more challenging. These two difficulties therefore have been a major barrier to the formation of homogeneous mesoporous structures. However, by carefully designing the reaction conditions of the sol–gel system such as water-to-silicon ratio, pH, and type and concentration of

15

372

K. Kanamori et al.

Phase separation tendency Phase separation Pm scale Colloidal aggregation nm scale With effect of phase separation (dotted line) Fig. 15.1 Colloidal aggregation during gelation competes with macroscopic phase separation caused by the increasing immiscibility gap between aqueous solution and hydrophobic condensates derived from organoalkoxysilanes. Gels with a wide pore size range can be obtained by controlling the relative timing of these two competing phenomena

additives, phase separation and cyclization can be masked; thus organoalkoxysilane precursors can be successfully used in the preparation of porous monolithic materials ranging from transparent aerogels to opaque macroporous foamlike gels [7]. Early on, Loy et al. discussed the possibility of forming monolithic networks in acid-, base-catalyzed, and neutral aqueous systems from various organotrialkoxysilanes, R′Si(OR)3, in which those with small substituent groups R′ such as hydrido (HTMS), methyl (MTMS), ethyl (ETMS), vinyl (VTMS) chloromethyl (ClMTMS), and 4-chloromethylphenyl (ClMPhTMS) were reported to form stable monolithic gels in the parameter range covered in their investigation [5] (Table 15.1). Methoxysilanes generally react faster than the corresponding ethoxysilanes. Many of these gels were opaque due to the macroscopic phase separation, inducing coarsened pore structures with structural features typically at micrometer scale. Organotrialkoxysilanes with a longer/bulkier substituent group such as propyl, n-butyl, n-hexyl, and phenyl did not form stable monolithic gels. It was also demonstrated to be difficult to form crosslinked networks with a well-defined pore structure at mesoscopic scale, which is an essential requirement for transparent, superinsulating aerogels. As a valid strategy, tetraalkoxysilanes (TMOS or TEOS) can be employed as co-precursor to enhance the formation of monolithic gels in many cases.

15.2.2 Aerogels Derived from Co-precursor Chemistries with Tetraalkoxysilanes Hüsing et al. investigated gel preparations using a single step, base-catalyzed (B1) process in methanol from precursor mixtures containing TMOS and R′Si(OR)3, where R′ represents

functional groups such as mercaptopropyl (MPTMS), diphenylphosphinoethyl (PhTMS), chloropropyl (ClPTMS), glycidoxypropyl (GLYTMS), methacryloxypropyl (MATMS), isocyanatopropyl (ISOTMS), and carbamatopropyl (CBTMS) [8, 9] (Table 15.1). Bulk density and surface area of typical aerogels were rather high, 0.24–0.27 g cm
3 and 624–875 m2 g
1, respectively (see also Table 15.4). The authors confirmed that the organotrialkoxysilanes used in the starting mixtures were quantitatively included in the resulting gel network when [TMOS]/[R′Si(OR)3] is 9/1 or lower. In addition, from the molar ratio [TMOS]/[R′Si(OR)3] between 9/1 and 6/4, there is a general trend that gelation time and shrinkage became longer/larger with an increasing mole fraction of organotrialkoxysilane. It was also found that the Brunauer-Emmett-Teller (BET)-specific surface area decreased with increasing fraction of organotrialkoxysilane – a fact that can be solely attributed to the immiscibility-induced coarsening of the resulting structures. In this early study, it was concluded that the combination of reduced crosslinking density with more hydrophobic chemistry and more pronounced steric hindrance negatively affects the formation of homogeneous, mesoporous network structures that are so critical for optical transparency and thermal insulation performance. Similarly, Rao et al. investigated hydrophobic coprecursor aerogel preparations using MTMS and TMOS precursor mixtures using a B1 process in methanol, with a precursor ratio [MTMS]/[TMOS] ranging from 0 to 1.85 (~65/35) (Table 15.4) [10, 11]. Transparent, less hydrophobic aerogels with a homogeneous structure at mesoscopic scale were obtained only for [MTMS]/[TMOS] 1000 m2 g
1) for aerogels and xerogels from the phenylene-bridged precursor prepared via a B1 process. Kanamori et al. used similar methylated precursors with shorter ethylene and ethenylene bridges from BMDE-ethy [39] and BMDE-ethe [40] (Table 15.3), respectively, to obtain transparent, hydrophobic aerogels in the density range of 0.13 g cm
3 or higher. Similar to the VTMS system described above, an AB2 process in the surfactant solvent polyoxyethylene 2-ethylhexyl ether was employed to obtain these aerogels. Wet gels dried at ambient conditions without a prior hydrophobization step yielded good aerogels, whose mechanical properties were influenced largely by the structure of organic bridge, which is discussed in more detail in this chapter. Aerogels Based on Polysiloxane Networks and Extended Hydrocarbon Chains (Doubly Crosslinked Systems) The above examples of organo-bridged systems imply the diverse roles of organic bridges in the formation of networks and the resultant properties of the aerogels. An especially interesting example of organo-bridged polysiloxane networks with extended hydrocarbon chains (denoted as “doubly crosslinked network”), aerogels with enhanced bending flexibility have been prepared [41–43] (Table 15.4). The precursor VMDMS (Table 15.2) was firstly polymerized in the presence of a radical initiator, which gave poly(alkoxysilyl)ated hydrocarbon polymers with Mw of 5000–9000. The polymer then

375

went through hydrolysis and polycondensation via a B1 process in benzyl alcohol as a solvent. The resulting aerogels showed transparency, and their densities and the BET-specific surface areas were in the 0.16–0.31 g cm
3 and 900– 950 m2 g
1 ranges, respectively [41]. As discussed in the following section, it is noteworthy that these aerogels showed very high compressive and bending flexibility, which allowed APD even without any solvent exchange. The same authors also extended the precursors to VTMS, ATMS, and AMDMS (Tables 15.1 and 15.2), and materials with similar properties were obtained [42]. Aerogels from trifunctional VTMS and ATMS were more transparent but also more hydrophilic than those from DMDMS and AMDMS due to the remaining silanol groups. Other unique properties such as thermal conductivity, solvent resistance, and machinability are discussed further in Sect. 15.3.3.

15.2.5 Conversion and Chemical Modification of Surface Groups to Extend Material Functionality So far, the chemical functionality of organoalkoxysilanederived aerogels and related porous materials was determined entirely by the functional silane precursor and/or co-precursor(s). Yet, chemical modification of specific organofunctional groups by organic chemical reactions, primarily substitution reactions, can greatly extend the application fields. Hüsing et al. [24] reported preparation and post-modification of porous materials obtained by AB2 process using a co-precursor systems containing MTMS and organofunctional trialkoxysilane component, with different reactive groups, namely, R′ ¼ vinyl, chloropropyl, mercaptopropyl, and methacryloxypropyl. To avoid phase separation, the molar ratios between CTAB, MTMS, and other trialkoxysilanes were adjusted. Raman spectra of prepared samples clearly confirmed the presence of specific organic functional groups. Monolithic aerogels show low bulk densities between 0.04 g cm
3 and 0.06 g cm
3 and high porosities between 95% and 97%. In the second half of that study, accessibility of functional groups was probed by specific chemical reactions (see Fig. 15.2): the double bond of vinyl and methacryl groups was converted to a geminal diol by redox reactions with potassium permanganate (Fig. 15.2a), which turned from the purple (permanganate) solution color to yellow-brownish as a result of its conversion to MnO2. The images in Fig. 15.2b show chloropropylmodified gel specimens before (left) and after (right) nucleophilic substitution followed by a click reaction which was monitored by the disappearance of the characteristic azido band in FT-IR spectra and directly confirmed by a color change of the monolith from yellow to blue (see Fig. 15.2b (vi)). Last but not least, the presence of mercapto groups was visualized using Ellman’s reaction which is commonly used

15

376

K. Kanamori et al.

a

I

III

II

b

IV

c

V

VI

VII

VIII

Fig. 15.2 Demonstration of functional group accessibility: (a) potassium permanganate solution (i) without a monolith, (ii) with alcogel, (iii) with a pure MTMS alcogel, and (iv) with VTMS silane; (b) (v) CP20 alcogel and (vi) CP20 alcogel after click reaction; (c) (vii) MP20 alcogel

in methanol and (viii) MP20 alcogel immersed in Ellman’s reactant/ buffer solution. (This figure is reprinted from Husing et al. with permission [24])

for the determination of the thiol group content in protein samples [44]. Figure 15.2c (vii–viii) shows a mercaptomodified monolith in methanol (left hand side) and in Ellman’s reactant/buffer solution (right-hand side), where the conversion to the yellow Ellman product is clearly evident. Clearly, the chemical functionality of alkylorganosilane gels can be controlled in multiple ways. Though still quite a new class of organic-inorganic hybrid materials, the freedom for controlling material properties is likely to produce additional functional porous materials with improved chemical and physical properties.

APD processes, which would lower the production cost and increase market uptake rates of aerogels in general. Over the course of an APD cycle, which relies on controlled solvent evaporation from the wet gel, compressive deformations (shrinkage) occur in the gel, which is linked to the surface tension of the drying liquid and the wettability of the inner gel surfaces (surface chemistry). The drying gel will be prone to cracking if the mechanical strength is lower than the stresses developed during deformations exerted by the capillary action. In addition, even if the gel is strong enough, irreversible shrinkage tends to remain in the network structure after completion of drying, particularly when a significant surface coverage of condensable groups such as silanols are easily accessible within the network structure. Condensation of residual silanol groups in the polysiloxane network results in additional covalent siloxane bond formation during drying, and collapse of the gel in the initial drying stage may lead to a high-density xerogel with low mesoporosity. Conversely, lower-density aerogels are by far easier to be obtained if the wet gel does not contain a significant amount of reactive surface groups, which is generally the case when using hydrophobic organoalkoxysilane precursors over standard ones (TEOS, TMOS, and WG). A schematic exhibiting the APD process with successful

15.3

Mechanical and Other Physical Properties of Aerogels from Organoalkoxysilanes

15.3.1 Properties of Aerogels from MTMS-Based Systems Mechanical properties, especially strength and elastic modulus under compression and bending loads at both gel and aerogel levels, are critically important for devising improved

15

Superhydrophobic and Flexible Aerogels and Xerogels Derived from Organosilane Precursors

Before drying

377

a Organogel (e.g. n-hexane)

PMSQ aerogel by SCD

15 During drying Temporal shrinkage

b

Spring-back After drying APD aerogel

PMSQ xerogel by APD

Fig. 15.3 A schematic showing successful APD process via temporal shrinkage and springback. Gels with a low concentration of polar groups are necessary to obtain APD aerogels

Fig. 15.4 (a) A typical compression and springback behavior observed in a PMSQ aerogel (~0.22 g cm
3) during a uniaxial compressiondecompression cycle. The maximum strain was 80%. (b) An example of an APD-dried transparent PMSQ aerogel (~0.11 g cm
3) with a size of 250  250  10 mm3

temporal shrinkage-reexpansion (springback) is shown in Fig. 15.3. In addition, making gels more resilient against bending further supports APD drying and greatly improves handling and usability. This class of advanced porous materials are exposed to various types of external forces and deformations (i) in the course of their production (mostly drying) and (ii) during a targeted application life cycle, and if the materials can endure bending that is decomposed into compression and tensile stresses, processability and applicability will be further enhanced. PMSQ aerogels prepared by Kanamori et al. from MTMS were transparent due to a well-defined mesoporous structure and absence of macropores [13]. During the AB2 process in the presence of an adequate surfactant, the hydrophobic PMSQ networks are kept miscible in the aqueous solution throughout the entire gelation and aging processes. The resultant PMSQ wet gels and aerogels dried by means of SCD were highly transparent and showed high strength and flexibility against compression (Fig. 15.4a): they can withstand up to 80% strain during uniaxial compression and are able to recover their original size and shape when the load is removed. At the same time, there was a limited improvement of bendability of these transparent aerogels compared to classical silica aerogels. The improved flexibility against compression can be attributed to the lower crosslinking density as compared to the classical silica analogs and the presence of hydrophobic methyl groups that repel each other

when compressed and hence increase the efficiency of the “springback” effect. Generally, the AB2 process yields colloidal networks with few remaining silanol groups, which prevents irreversible shrinkage upon compression. During APD, almost complete springback without cracking is expected, since the surface tension of the drying solvent causes similar mechanical loading/unloading conditions during APD. Through some targeted optimization, Kanamori et al. were able to optimize the APD process to produce near identical properties of APD aerogels when compared to SCD reference materials, as well as crack-free monoliths of large sizes: currently, APD-dried PMSQ aerogels with low density (0.10–0.20 g cm
3) and low thermal conductivity (>13 mW m
1 K
1), in the form of large monolithic tiles >300  300  10 mm3, can be prepared (Fig. 15.4b shows one example with 250  250 mm2). The industrial commercialization of this material as transparent thermal insulators is currently under development. For a range of technical applications, a further improvement of the mechanical properties of MTMS-based and PMSQ aerogels is certainly desirable or even imperative. As discussed earlier, the bending flexibility was partly improved by employing CNFs as a physical reinforcing agent [18]. Good surface chemical compatibility between PMSQ and CNF allow for a good dispersion of CNFs throughout the PMSQ network. Since CNF fibers serve as support, higher bendability can be observed in low-density composites of 0.020 g cm
3, but the same reinforcement was not present in

378

higher-density composites of, e.g., 0.14 g cm
3. The molecular-scale structure and physical properties were not disturbed by the presence of CNFs at low weight percent loadings (130 ), polysilsesquioxane analogs from VTMS or ATMS were much less hydrophobic (150 and water uptake below 1% is expected.

15.4.2 Environmental Remediation Early potential applications of aerogel materials include oil spill cleanup and trapping of other environmental and biotoxins from ambient air or liquids. A more detailed overview on this topic can be found in the chapter: Aerogels for Environmental Applications. In this section, we shall peruse the state of the art in alkylorganosilane-based aerogels for hydrocarbon uptake and gas sorption/capture (volatile organic compounds (VOCs) and CO2) by several relevant examples.

Hydrocarbon Uptake Typical aerogels have high surface area and hence are good at selective trapping of contaminants occurring in small concentrations in gases or liquids. Here, the adsorption of volatile, gas-phase contaminants proceeds through monolayer to multilayer adsorption steps and eventually peaks by capillary condensation [54]. However, the oil spill cleanup scenario is a different story. The contaminant is a single solid/liquid phase which is directly brought into contact with the porous absorber material, and the uptake, governed by capillary uptake, is almost

K. Kanamori et al.

instantaneous at least for nonpolar liquid mixtures with low to medium viscosity. Here, the combination of superhydrophobic surface chemistry with large accessible pore volume is the main criterion for good trapping properties. In other words, for oil spill cleanup, the mesoporous structure of typical aerogels is not needed and macroporous superhydrophobic materials are well suited. From an economic perspective, macroporous materials may be superior because strictly mesoporous aerogels are generally more difficult and more expensive to manufacture. In addition to a high oil sorption capacity, an ideal sorbent material should have low water uptake (i.e., high selectivity for carbon versus aqueous phase) as well as excellent reusability. In the following section, the performance of elastic superhydrophobic alkylalkoxysilane-based aerogels as porous traps for oils, organic liquids, and heavy metals will be discussed. In an early study, Hrubesh, Coronado, and Satcher synthesized hydrophobic aerogels containing CF3(CH)2 group (CF3-aerogels) using a B1 method [55]. The same group then further investigated the adsorption capacity of CF3-aerogels for different pure hydrocarbon test solvents and demonstrated significantly higher uptake than granulated activated carbon (GAC) for all of them [56]. An oil spill scenario was simulated by using oil mixtures and salt-water mixtures; the oil was cleanly separated by the CF3-aerogels regardless of the concentration of CF3(CH)2 functional group. Those results indicated that the ultimate absorbing capacity of a 30% CF3-aerogel in powder form is roughly 14 times their weight. To verify practical usability of such material, Coronado et al. [57] developed a deployable system by incorporating CF3-aerogel into commercially available solid support materials such as fiberglass, alumina, alumina tiles, cotton wool, and vitreous carbon foam. The resulting composites demonstrated very high absorption capacities for hydrocarbons and further showed good selectivity in absorbing oil in the presence of water, thus providing a promising platform for oil spill recovery. Similar results were reported by Rao, Park et al. [58], who used APD silica aerogels hydrophobized with HMDZ and evaluated hydrocarbon uptake through soaking aerogel samples in a large selection of organic liquids. Those hydrophobic silica aerogels were shown to take up to 12 times their own mass in organic liquids. Standeker et al. determined the adsorption capacity of superhydrophobic aerogels, which were made by the co-precursor method from TMOS/MTMS (WCA ¼ 173 ) and TEOS/MTES (WCA ¼ 180 ) systems, for various organic compounds from aqueous mixtures [59] were used. For the sorption capacity measurement, aerogel samples were pulverized into small grains with particle size below 0.25 mm. For the benchmark test with chlorobenzene, absorption capacities up to 370 and 472 mg g
1 for the TMOS/MTMS or TEOS/MTES aerogels were found, respectively, whereas benzene uptake levels were approximately 300 mg g
1 for both sorbent materials. Regeneration of the

15

Superhydrophobic and Flexible Aerogels and Xerogels Derived from Organosilane Precursors

liquid-saturated material could be achieved by purging with hot inert gas at 100  C, releasing the organic compounds almost quantitatively and allowing capture in a condenser. The hydrocarbon uptake properties of those aerogels remained unchanged after 20 repetitive adsorption/desorption cycles. In summary, silica aerogels obtained by standard post-hydrophobization and co-precursor method with alkylalkoxysilanes marked the early hydrocarbon uptake studies and showed superior performance and reusability over classical sorbents. Similar performance was observed with superhydrophobic macroporous materials from pure alkylalkoxysilane: Rao et al. investigated the adsorption properties of the elastic, superhydrophobic materials prepared using only MTMS precursors [53]. In their uptake studies [60], four different alkanes, three aromatic compounds, four alcohols, and three different oils were tested. All MTMS-based materials showed excellent absorption properties and were able to be reused at least three times without noticeable structural changes in the case of alkanes, aromatics, and alcohols, which was supported by TEM measurements. In that regard it was further observed that the materials shrank considerably during the desorption step but then sprang back. Oil absorption on the other hand led to partially irreversible shrinkage with less springback and significant densification. Perdigoto et al. [61] investigated the removal of aromatic solvents (benzene and toluene) as contaminants in water by MTMS-based macroporous materials. The contaminant concentrations were selected to simulate real cases and were 200 mg L
1 for both benzene and toluene, and the samples were benchmarked against commercial hydrophobic silica aerogel granules (Cabot). The materials tested in this study showed better absorption capacity than that of the commercial aerogel. The authors also attempted to improve the ability to regenerate the MTMS-based sorbents by subjecting them to a gas-phase oxidative treatment with ozone at the end of the desorption step. During the first three initial cycles, the adsorption capacity remained nearly constant when subjected to ozone but then dropped sharply after the fourth cycle, also evidenced by the breaking up of the material and dispersing in the solution. It is believed that the strong oxidizer ozone also degrades the methyl groups of the MTMS-derived network. After the discovery of MTMS/DMDMS “marshmallow gels,” Kanamori’s group further tested those new materials as potential oil/water separation media, using n-hexane as a model organic compound for uptake and release studies (Fig. 15.9b). Over ten repetition cycles, marshmallow gels showed stable performance, absorbing n-hexane roughly 6.2 times of their dry weight, without showing any signs of structural change or performance loss after ten absorption/ squeezing cycles. The gels could also absorb other organic liquids and still be dried by squeezing out, even in the case of

383

high-density organic liquids (such as chloroform) and viscous oils (Fig. 15.9d). The superhydrophobicity is explained as a combination of the surface roughness, caused by the macroporous structure presumably formed by spinodal decomposition and by the numerous methyl groups present on their surface. The 29Si solid-state NMR spectra showed that residual hydroxy groups are virtually negligible in the structure implying that the network formation is completed within a few hours (Fig. 15.9e). The absorption properties of material could be somewhat influenced by changing the substituent groups in the precursors. Furthermore, the flexibility and pore size of the materials are tuned by changing the sol composition, via control over precursor amounts and relative ratios of organosilanes, urea, surfactant, and water [21, 22]. The marshmallow gels not only show compression/ reexpansion properties similar to that of PMSQ gels but also are very soft and easy to bend. It is the combination of larger pores, superhydrophobic surface chemistry, and soft/flexible mechanics that makes those materials rank among the best absorbers for organic liquids and oils. Further tailoring of properties of “marshmallow” gels targeted the introduction of additional oleophobic (perfluoroalkyl) groups: vinyl-modified marshmallow gels prepared from VTMS and VMDMS were treated with perfluoroalkyl thiol and attached by means of thiol-ene-type reactions. This modification led to a further decrease in surface free energy through fluorocarbon chemistry – the resulting materials repel a variety of liquids ranging from completely polar (water) to nonpolar hydrocarbons (n-octadecane). Contact angle greater than 150 for both water and organic liquids, such as ethylene glycol, formamide, diiodomethane, 1-bromonaphthalene, and n-hexadecane, is reported [23].

Volatile Organic Compound (VOC) Vapor Uptake Superhydrophobic silica aerogels also have a high affinity for VOCs in the gas phase and hence are good candidates for recovering organic vapors from industrial flow streams. Benzene, toluene, ethylbenzene, and xylene (BTEX) are examples of the popular, ecotoxic pollutants that need to be prevented from escaping into the atmosphere. Knez et al. tested adsorption properties of two superhydrophobic aerogels based on MTMS/TMOS and trimethylethoxysilane TMES/TMOS [62]. Continuous adsorption measurements showed that silica aerogels are excellent adsorbents of BTEX vapors from waste gas stream. Compared to the commonly used adsorbents, such as GAC and silica gel, aerogels exhibit capacities that exceed those of both commonly used adsorbents (GAC by a factor of 3 and silica gel by a factor of 2, respectively). Furthermore, they can be reused at least 14 times without loss of efficiency, and BTEX vapors can be recovered as liquids.

15

384

D2

D1

50

0

ppm

T3

T2

–50

–100

Mineral oil

e

1600 1400 1200 1000 800 600 400 200 0

n-Octane

d

Tetradecane

c

1 2 3 4 5 6 7 8 9 10 Cycle

n-Octanol Diethylether

Combinations Me / Me = of R1 / R2 Vinyl / Vinyl 3-Mercaptopropyl / 3-Mercaptopropyl Me / Ph + Me (1:1) Me / 3,3,3-Trif luoropropyl + Me (1:1)

Petroleum ether

+ Si OMe MeO OMe 3 : 2 Me OMe

Si

Chloroform

R2

700 600 500 400 300 200 100 0

Kerosene

R1 MeO

RT

80 °C 8h

n-Hexane Cyclohexane

RT 1h

b

Washing & drying

TMB

Mixing Gelation in one-pot & aging

Toluene

Urea + CTAC in aq. HOAc (5 mM)

Weight gain (%)

a

K. Kanamori et al.

Fig. 15.9 (a) Facile synthesis of marshmallow-like gels derived from di- and trifunctional alkoxysilanes as co-precursors. (b) Weight gain during n-hexane absorption/drying cycles and absorption capacities for various organic liquids and oils, as indicated by weight gain. (c) The shape of the MTMS-DMDMS gel produced in a 2.5 L scale and their

simulated 3D microstructure. (d) The gel can be used to absorb organic liquids and then be squeezed out by hand. (e) 29Si solid-state NMR spectrum of the MTMS-DMDMS gel. (This figure is reprinted from Hayase et al. with permission [22])

CO2 Capture Amino-functionalized silica aerogels have been suggested as CO2 sorbents because of their large surface area and the relative ease to graft functional amine groups onto silica with high surface area [63–66]. These materials generally possess good CO2 cycling capacities and exceptional stability under humid conditions [64]. Perhaps the simplest and most straightforward approach is to impregnate commercially available silica aerogel granules with tetraethylenepentamine (TEPA), resulting in moderate cyclic CO2 capacities toward pure CO2. These materials have not been tested for atmospheric CO2 capture (~400 ppm) but are expected to underperform in such scenario [67, 68]. Several other groups employed bottom-up methods, introducing amino groups through functional organosilanes such as

aminopropyltrimethoxysilane (APTMS/APTES) during gel formation (co-precursor method) or by post-modification of as-prepared alcogels [67–70]. These resulting materials possess quite high CO2 sorption capacities which are retained over many cycles, even at atmospheric CO2 concentrations. However, due to the mostly hydrophilic surface chemistry of these tetraalkoxysilane-derived materials, the resulting amine-modified aerogels had high water uptake and poor selectivity for CO2 over H2O [69]. Aiming to develop an intrinsically more hydrophobic sorbent with low water uptake, researchers from Aspen Aerogels developed amino-functionalized aerogels starting from MTMS and APTMS/APTES co-precursors [71]. They reported roughly 1.8 mmol g
1 CO2 sorption capacity in a humid CO2 (15%)

15

Superhydrophobic and Flexible Aerogels and Xerogels Derived from Organosilane Precursors

atmosphere at 40  C with outstanding cycling stability with no noticeable performance loss over 2000 cycles. As we have shown by a few selected examples, the interaction between the hydrophobic surface of the aerogel or xerogel and the adsorbate molecule is different for each organic compound. Adsorption isotherms describe how adsorbates interact with adsorbent materials and are fundamental to improve our understanding of underlying processes and optimize the performance of adsorbents. Hence, the correlation of equilibrium data using isotherm equations is important to the practical design and operation of adsorption systems [61]. It is clear that for many applications beyond thermal insulation, not only the physical and structural properties of aerogels but also particularly surface chemistry and hydrophobicity are determinant factors that can be deliberately tailored for particular environmentally related tasks and that specific chemical functionality is needed in order to achieve high selectivities for trapping and removing target molecules.

15.4.3 Drag Reduction Despite their still limited mechanical strength and poor wear resistance – the two factors which effectively hinder the use of aerogels and related materials in tribological applications – Rao and coworkers were the first to demonstrate astonishing dynamic mechanical properties of superhydrophobic aerogel surfaces: aerogel-like materials were prepared using MTMS [72] or TEOS/MTES [73] by high temperature SCD or by APD, respectively. After grinding the materials to powders, small water droplets (~14 μL) were coated with the powder, which created the so-called liquid marbles showing very low interaction and drag resistance on many surfaces. Figure 15.10a shows the device used for measuring the sliding velocity at a given inclination angle by a direct measurement of the transit time between two points with optical light sensors. A nearly perfect liquid marble is shown in Fig. 15.10b. At moderate tilt angles of 52 , a terminal velocity of up to 1.44 m s
1 was measured for the droplet coated with the most hydrophobic material, reasonably close to the free fall velocity of about 1.52 m s
1 measured over the same distance. These results clearly revealed that water droplets coated with superhydrophobic porous powders could effectively suppress friction with different material surfaces and also greatly facilitate the transport of small quantities of liquids in confined channels, opening up potential applications in micro- and nanofluidic devices. Truesdell et al. [74] used aerogel-coated, superhydrophobic patterned surfaces as a model system for

385

a 1 3

1

4

t1 2

t2 φ

15 2

b

Fig. 15.10 (a) Schematic setup for water droplet velocity measurements: (1) light-emitting diodes (LED), (2) photoconductive detectors, (3) inclined plane platform, (4) water droplet (size, 2.8 mm (0.2 mm)). (b) Liquid marble (2.8 mm (0.2 mm) size) floating on a water surface. (This figure is reprinted from Rao et al. with permission [72])

surface/fluid flow and slip in a rotating torque cylinder setup. They concluded that patterning of surfaces and controlling surface roughness greatly influence the behavior of the fluid near the solid-phase boundary – a fact that can be exploited for the design of low drag systems. Samaha et al. [75] coated surfaces with hydrophobic silica particles and studied the effects of elevated pressure on the drag reduction produced by these surfaces in underwater applications. They measured drag reduction of 18–22% and also found that a threshold pressure exists, above which this superhydrophobic drag reduction effect ceases. Recently, Rodriguez et al. [76] demonstrated drag reduction up to 30% for surfaces coated with hydrophobic (TMOS/ MTMS-based) aerogel films on a rotating disk. Figure 15.11 shows an image of the bottom of the disk coated with (a) hydrophobic SCD aerogel and (b) hydrophobic APD aerogel. These aerogel-coated surfaces with high enough hydrophobicity are able to stabilize air bubbles, thus effectively suppressing drag. Although these aerogels have similar WCA (ca. 160 ), their pore structures are quite different. In the case of the APD aerogels, only those with high hydrophobicity (prepared from 75% MTMS co-precursor) produced significant drag reduction, while SCD aerogels with as low as 10% MTMS content showed good drag reduction.

386

K. Kanamori et al.

microfluidic applications. It remains to be seen whether the scientific community is going to adopt this approach in the coming years.

15.5

Economic Considerations and Commercialization Outlook

It has been almost 15 years since the start of the second wave of aerogel commercialization. Today, classical hydrophobic silica aerogels are well established in selected markets and applications, and other technologies are now transitioning from lab to market. The most prominent new classes of aerogels on the verge of commercialization with a multiton perspective are polymer aerogels, biopolymer aerogels, and alkylorganosilane-derived aerogels and related porous materials.

15.5.1 Economic Considerations

Fig. 15.11 Photographs of a rotating viscometer spindle bottom surface in water. (a) A large bubble is captured on a surface coated with a 10% MTMS aerogel; (b) small bubbles formed on a surface coated with a 75% MTMS xerogel. (This figure is reprinted from Rodriguez et al. with permission [76])

These experiments indicate that the intrinsic structure and position of the hydrophobic groups in addition to the hydrophobe content play a significant role in determining drag reduction properties. Kim et al. coated the inside of a microfluidic capillary channel with a thin film of a superhydrophobic aerogel material by flowing a dilute slurry of an MTMS-based gel and APD [77]. The inner surface of the channel was entirely coated with a highly mesoporous aerogel network, and sessile drop tests on a flat surface gave a WCA of 160 . The authors also measured the pressure drop associated with droplet-based flow through the channel and determined that the aerogel coating reduced friction resistance to motion by a factor of three compared to a self-assembled monolayer (SAM) reference with a WCA of approximately 90 . This is the first technical demonstration of an aerogel-like coating for

Alkylorganosilane aerogels and related porous materials certainly are not yet cost-competitive with conventional silica aerogels given the very early stage of their commercialization. However, there is a significant potential for future development. As far as the different chemistries are concerned, simple alkylorganosilane-derived materials (MTMS, MTES, and MTMS/DMDMS or MTES/DMDES) will certainly be the first to become commercially available. Doubly crosslinked systems (e.g., including also VTES or methacrylate chemistry) are likely to follow second. In the following, we are discussing basic process-relevant aspects and parameters necessary for upscaling. Our comparison is based on classical WG and TEOS precursor reference systems. Looking at raw material cost based on spring 2019 pricing, the direct comparison shows that WG-based aerogels set the lowest possible reference point (roughly 2–3 USD/kg of dried aerogel), significantly lower than even TEOS-based aerogels (4–5 USD/kg dried aerogel). Aerogels from MTMSand DMDMS-based pure organosilane systems have significantly higher raw material cost on the order of 6–8 USD/kg, while the use of more expensive bridged and other functional silanes would clearly raise raw material prices well into the 10–20 USD/kg range. Certainly, many of the specialty silanes used today in academic studies are not economical candidates for mass production. When looking at any process engineering parameters, both time and solvent freight are the two key aspects in analysis of sol–gel material production. Gelation (including aging), solvent exchange (including post-hydrophobization), and APD are highly optimized in today’s industrial processes. For both WG- and TEOS-based silica aerogels, cycle times

15

Superhydrophobic and Flexible Aerogels and Xerogels Derived from Organosilane Precursors

below half a day from start to finish have been demonstrated at pilot and industrial scales [78]. Clearly, a direct comparison to organosilane-derived porous materials is difficult because transferring to industry is still in its infancy stage and the community’s knowledge is more or less limited to research formulations. Yet, there are some underlying fundamentals that suggest significantly longer processing times for organosilane aerogels. First the hydrolysis and condensation kinetics (and hence gelation/aging) are quite a bit slower than that for the classical silica systems; however, there is no need for a separate post-hydrophobization treatment. The PMSQ aerogels are typically dried by APD from hydrocarbons such as hexane or heptane, which means that solvent exchange into alkanes is necessary. Both of these steps are quite timeconsuming (typical process times are regularly in the range of several days) for high-quality crack-free monoliths and highly dependent on the specimen thickness. For macroporous foam materials on the other hand, there is no need for solvent exchanges, and drying is not really an issue, as typical pores are tens to hundreds of microns in diameter and hence they are very flexible and can cope with drying stresses from accumulating surface tension forces easily. In addition macroporous foam materials can be prepared by a simple squeezing out the syneresis liquid and soaking in a washing solution. At an industrial scale, washing and drying of such materials are very fast processes compatible with continuous roll to roll manufacturing. Solvent consumption and ease of reuse in WG- and TEOSbased systems are rather optimized and typically employ water, alcohol, and nonpolar hydrocarbon mixtures. The separation of the process liquids after condensing the evaporated liquids during APD is relatively straightforward, because the technical separation and purification of those systems are standard in chemical engineering: water/ethanol/hydrocarbon mixtures are commonly separated by addition of water and decanting off the immiscible hydrocarbon phase, followed by concentrating the ethanol through a continuous distillation setup. In the case of the organosilane chemistry, the separation of these standard solvent systems is complicated by the large contents of solid gelation initiators (urea) and surfactant additives (CTAB, PEOPPO block copolymers). Surfactants in particular have a tendency for foaming which greatly disturbs solvent separation, requiring the addition of anti-foaming agents which can then be concentrated again in the process solvent stream. Clearly, there are possible technical solutions to overcome those challenges such as the use of surface-active solvents such as polyoxyethylene 2-ethylhexyl ether instead of solid surfactants. However, generally the separation of water/alcohol/ hydrocarbon mixtures with high surface-active component loadings is a non-negligible, even possibly deal-breaking, technical challenge. In summary, the main challenges in transferring organosilane-derived porous materials from laboratory to

387

industrial scale are linked to (i) raw material cost, (ii) process times linked to solvent exchange and drying for APD monolithic aerogels, and (iii) recycling and complete reuse of organic solvents used. In the case of MTMSDMDMS porous foam materials (“marshmallow” gels), the synthesis can be completed within less than a day, and transfer to the industrial scale seems rather straightforward. However, one needs to keep in mind that these macroporous materials – at approximately 30–35 mW m
1 K
1 – do not have the same low thermal conductivities associated with mesoporous silica or PMSQ aerogel materials (13– 15 mW m
1 K
1) and hence must target different applications. Clearly, the number of possible markets which justify a comparably high price for the superflexible superhydrophobic gels is limited. Mesoporous PMSQ aerogels are likely to find new niche markets; however, the production cost is likely not going to undercut WG- or TEOS-based standard silica aerogels in large volume applications.

15.5.2 Commercialization Outlook Alkylorganosilane-derived aerogels and related porous materials are in their early stage of commercialization with a few players already offering sampling quantities of materials: Active Aerogels in Portugal is producing MTMS-based aerogels and other porous materials based on know-how from the University of Coimbra, and Tiem factory Inc. in Japan is commercializing large-area monolithic highly transparent PMSQ aerogel panels based on technology developed by the team of Kanamori at Kyoto University. The key questions that will decide over market success are cost, added value over conventional materials, and existing aerogels and smart product development/integration efforts. For large volumes, organosilane-based macroporous materials are likely to find their primary application in cryogenic technology because of their excellent flexibility, even at sub-liquid nitrogen temperatures, and in environmental solutions (spill remediation, VOC trapping). The PMSQ aerogels have the great potential to take over the market for highquality transparent monolithic panels with the great advantage of durable hydrophobic properties (by comparison with the pure silica aerogels via high-temperature SCD previously commercialized by Airglass in Sweden). In the field of low-volume high-specialization applications, alkylorganosilanes have a great potential for commercialization, because raw material cost, process efficiency, and solvent recuperation aspects are completely secondary. Here the added value in material properties at a small scale by far outweigh the additional cost, meaning that the margins are probably 2–3 orders of magnitude higher than for commodity materials. Typical applications are likely going to include microelectronics, microelectromechanical systems (MEMS), chemical sensors,

15

388

biotechnology, microsystem thermal packaging, microfluidics, miniaturized 3D-printed devices, and many more. It is where the entire space of chemical flexibility and extended material properties can fully develop and blossom: novel unprecedented materials and system properties achieved by the combination of multifunctional silane linkers, polymerization, and ligand chemistries can produce breakthroughs in designing the interfacial material properties and compatibilization. We conclude this chapter with the remark that the field of alkylorganosilane-based aerogels and related materials is one with an exciting perspective for improving over classical silica aerogels in many ways and creating new science and intellectual property on its path.

References 1. Larson, G.L.: Recent synthetic applications of organosilanes. In: Patai, S. (ed.) Organic Silicon Compounds, vol. 1, pp. 763–808. Wiley, New York (1989) 2. Antonucci, J.M., Dickens, S.H., Fowler, B.O., Xu, H.H.K., McDonough, W.G.: Chemistry of silanes: interfaces in dental polymers and composites. J. Res. Natl. Inst. Stand. Technol. 110, 541–558 (2005) 3. Thames, S.F., Panjnani, K.G.: Organosilane polymer chemistry: a review. J. Inorg. Organomet. Polym. 6, 59–94 (1996) 4. Li, L., Li, B., Dong, J., Zhang, J.: Roles of silanes and silicones in forming superhydrophobic and superoleophobic materials. J. Mater. Chem. A. 4, 13677–13725 (2016) 5. Loy, D.A., Baugher, B.M., Baugher, C.R., Schneider, D.A., Rahimian, K.: Substituent effects on the sol-gel chemistry of organotrialkoxysilanes. Chem. Mater. 12, 3624–3632 (2000) 6. Fidalgo, A., Rosa, M.E., Ilharco, L.M.: Chemical control of highly porous silica xerogels: physical properties and morphology. Chem. Mater. 15, 2186–2192 (2003) 7. Kanamori, K., Nakanishi, K.: Controlled pore formation in organotrialkoxysilane-derived hybrids: from aerogels to hierarchically porous monoliths. Chem. Soc. Rev. 40, 754–770 (2011) 8. Hüsing, N.: Schubert: Organofunctional silica aerogels. J. Sol-Gel Sci. Technol. 8, 807–812 (1997) 9. Hüsing, N., Schubert, U., Misof, K., Fratzl, P.: Formation and structure of porous gel networks from Si(OMe)4 in the presence of A(CH2)nSi(OR)3 (A ¼ functional group). Chem. Mater. 10, 3024–3032 (1998) 10. Venkateswara Rao, A., Haranath, D.: Effect of methyltrimethoxysilane as a synthesis component on the hydrophobicity and some physical properties of silica aerogels. Microporous Mesoporous Mater. 30, 267–273 (1999) 11. Venkateswara Rao, A., Pajonk, G.M.: Effect of methyltrimethoxysilane as a co-precursor on the optical properties of silica aerogels. J. Non-Cryst. Solids. 285, 202–209 (2001) 12. Venkateswara Rao, A., Bhagat, S.D., Hirashima, H., Pajonk, G.M.: Synthesis of flexible silica aerogels using methyltrimethoxysilane (MTMS) precursor. J. Colloid Interface Sci. 300, 279–285 (2006) 13. Kanamori, K., Aizawa, M., Nakanishi, K., Hanada, T.: New transparent methylsilsesquioxane aerogels and xerogels with improved mechanical properties. Adv. Mater. 19, 1589–1593 (2007) 14. Kurahashi, M., Kanamori, K., Takeda, K., Kaji, H., Nakanishi, K.: Role of block copolymer surfactant on the pore formation in methylsilsesquioxane aerogel systems. RSC Adv. 2, 7166–7173 (2012)

K. Kanamori et al. 15. Hayase, G., Kanamori, K., Nakanishi, K.: Structure and properties of polymethylsilsesquioxane aerogels synthesized with surfactant nhexadecyltrimethylammonium chloride. Microporous Mesoporous Mater. 158, 247–252 (2012) 16. Hayase, G., Kanamori, G.K., Maeno, A., Kaji, H., Nakanishi, K.: Dynamic spring-back behavior in evaporative drying of polymethylsilsesquioxane monolithic gels for low-density transparent thermal superinsulators. J. Non-Cryst. Solids. 434, 115–119 (2016) 17. Hayase, G., Nagayama, S., Nonomura, K., Kanamori, K., Maeno, A., Kaji, H., Nakanishi, K.: Fabrication of hydrophobic polymethylsilsesquioxane aerogels by a surfactant-free method using alkoxysilane with ionic group. J. Asian Ceramic Soc. 5, 104– 108 (2017) 18. Hayase, G., Kanamori, G.K., Abe, K., Yano, H., Maeno, A., Kaji, H., Nakanishi, K.: Polymethylsilsesquioxane-cellulose nanofiber biocomposite aerogels with high thermal insulation, bendability, and superhydrophobicity. ACS Appl. Mater. Interfaces. 6, 9466– 9471 (2014) 19. Saito, T., Hirota, M., Tamura, N., Kimura, S., Fukuzumi, H., Heux, L., Isogai, A.: Individualization of nano-sized plant cellulose fibrils achieved by direct surface carboxylation using TEMPO catalyst. Biomacromolecules. 10, 1992–1996 (2009) 20. Liu, Q., Frazier, A.W., Zhao, X., De La Cruz, J.A., Hess, A.J., Yang, R., Smalyukh, I.I.: Flexible transparent aerogels as window retrofitting films and optical elements with tunable birefringence. Nano Energy. 48, 266–274 (2018) 21. Hayase, G., Kanamori, K., Nakanishi, K.: New flexible aerogels and xerogels derived from methyltrimethoxysilane/dimethyldimethoxysilane co-precursors. J. Mater. Chem. 21, 17077– 17079 (2011) 22. Hayase, G., Kanamori, K., Fukuchi, M., Kaji, H., Nakanishi, K.: Facile synthesis of marshmallow-like macroporous gels usable under harsh conditions for the separation of oil and water. Angew. Chem. Int. Ed. 52, 1986–1989 (2013) 23. Hayase, G., Kanamori, K., Hasegawa, G., Maeno, A., Kaji, H., Nakanishi, K.: A superamphiphobic macroporous silicone monolith with marshmallow-like flexibility. Angew. Chem. Int. Ed. 52, 10788–10791 (2013) 24. Ehgartner, C.R., Grandl, S., Feinle, A., Hüsing, N.: Flexible organofunctional aerogels. Dalton Trans. 46, 8809–8817 (2017) 25. Li, L., Li, B., Wu, L., Zhao, X., Zhang, J.: Magnetic, superhydrophobic and durable silicone sponges and their applications in removal of organic pollutants from water. Chem. Commun. 50, 7831–7833 (2014) 26. Nadargi, D., Gurav, J., Marioni, M.A., Romer, S., Matam, S., Koebel, M.M.: Methyltrimethoxysilane (MTMS)-based silica–iron oxide superhydrophobic nanocomposites. J. Colloid Interface Sci. 459, 123–126 (2015) 27. Shimizu, T., Kanamori, K., Maeno, A., Kaji, H., Doherty, C.M., Falcaro, P., Nakanishi, K.: Transparent, highly insulating polyethyland polyvinylsilsesquioxane aerogels: mechanical improvements by vulcanization for ambient pressure drying. Chem. Mater. 28, 6860–6868 (2016) 28. Kimura, T., Shimizu, T., Kanamori, K., Maeno, A., Kaji, H., Nakanishi, K.: Aerogels from chloromethyltrimethoxysilane and their functionalizations. Langmuir. 33, 13841–13848 (2017) 29. Shea, K.J., Loy, D.A.: Bridged polysilsesquioxanes. Molecularengineered hybrid organic-inorganic materials. Chem. Mater. 13, 3306–3319 (2001) 30. Loy, D.A., Jamison, G.M., Baugher, B.M., Russick, E.M.: Alkylene-bridged polysilsesquioxane aerogels: highly porous hybrid organic-inorganic materials. J. Non-Cryst. Solids. 186, 44–53 (1995) 31. Boday, D.J., Stover, R.J., Muriithi, B., Loy, D.A.: Mechanical properties of hexylene- and phenylene-bridged polysilsesquioxane aerogels and xerogels. J. Sol-Gel Sci. Technol. 61, 144–150 (2012)

15

Superhydrophobic and Flexible Aerogels and Xerogels Derived from Organosilane Precursors

32. Boday, D.J., Stover, R.J., Muriithi, B., Loy, D.A.: Strong, low density, hexylene- and phenylene-bridged polysilsesquioxane aerogelpolycyanoacrylate composites. J. Mater. Sci. 46, 6371–6377 (2011) 33. Aoki, Y., Shimizu, T., Kanamori, K., Maeno, A., Kaji, H., Nakanishi, K.: Low-density, transparent aerogels and xerogels based on hexylene-bridged polysilsesquioxane with bendability. J. Sol-Gel Sci. Technol. 81, 42–51 (2017) 34. Kanamori, K., Ueoka, R., Kakegawa, T., Shimizu, T., Nakanishi, K.: Hybrid silicone aerogels toward unusual flexibility, functionality and extended applications. J. Sol-Gel Sci. Technol. 89, 166– 175 (2019) 35. Wang, Z., Dai, Z., Wu, J., Zhao, N., Xu, J.: Vacuum-dried robust bridged silsesquioxane aerogels. Adv. Mater. 25, 4494–4497 (2013) 36. Zou, F., Yue, P., Zheng, X., Tang, D., Fu, W., Li, Z.: Robust and superhydrophobic thiourethane bridged polysilsesquioxane aerogels as potential thermal insulation materials. J. Mater. Chem. A. 4, 10801–10805 (2016) 37. Chen, D., Gao, H., Jin, Z., Wang, J., Dong, W., Huang, X., Wang, G.: Vacuum-dried synthesis of low-density hydrophobic monolithic bridged silsesquioxane aerogels for oil/water separation: effects of acid catalyst and its excellent flexibility. ACS Appl. Nano Mater. 1, 933–939 (2018) 38. Loy, D.A., Jamison, G.M., Baugher, B.M., Myers, S.A., Assink, R. A., Shea, K.J.: Sol-gel synthesis of hybrid organic-inorganic materials. Hexylene- and phenylene-bridged polysiloxanes. Chem. Mater. 8, 656–663 (1996) 39. Shimizu, T., Kanamori, K., Maeno, A., Kaji, H., Nakanishi, K.: Transparent ethylene-bridged polymethylsiloxane aerogels and xerogels with improved bending flexibility. Langmuir. 32, 13427–13434 (2016) 40. Shimizu, T., Kanamori, K., Maeno, A., Kaji, H., Doherty, C.M., Nakanishi, K.: Transparent ethenylene-bridged polymethylsiloxane aerogels: mechanical flexibility and strength and availability for addition reaction. Langmuir. 33, 4543–4550 (2017) 41. Zu, G., Shimizu, T., Kanamori, K., Zhu, Y., Maeno, A., Kaji, H., Shen, J., Nakanishi, K.: Transparent, superflexible doubly crosslinked polyvinylpolymethylsiloxane aerogel superinsulators via ambient pressure drying. ACS Nano. 12, 521–532 (2018) 42. Zu, G., Kanamori, K., Shimizu, T., Zhu, Y., Maeno, A., Kaji, H., Nakanishi, K., Shen, J.: Versatile double-cross-linking approach to transparent, machinable, supercompressible, highly bendable aerogel thermal superinsulators. Chem. Mater. 30, 2759– 2770 (2018) 43. Zu, G., Kanamori, K., Maeno, A., Kaji, H., Nakanishi, K.: Superflexible multifunctional polyvinylpolydimethylsiloxane-based aerogels as efficient absorbents, thermal superinsulators, and strain sensors. Angew. Chem. Int. Ed. 57, 9722–9727 (2018) 44. Bravo-Osuna, I., Teutonico, D., Arpicco, S., Vauthier, C., Ponchel, G.: Characterization of chitosan thiolation and application to thiol quantification onto nanoparticle surface. Int. J. Pharm. 340, 173–181 (2007) 45. Bhushan, B., Jung, Y.C.: Natural and biomimetic artificial surfaces for superhydrophobicity, self-cleaning, low adhesion, and drag reduction. Prog. Mater. Sci. 56, 1–108 (2011) 46. Shirtcliffe, N.J., McHale, G., Atherton, S., Newton, M.I.: An introduction to superhydrophobicity. Adv. Colloid Interf. Sci. 161, 124–138 (2010) 47. Celia, E., Darmanin, T., Taffin de Givenchy, E., Amigoni, S., Guittard, F.: Recent advances in designing superhydrophobic surfaces. J. Colloid Interface Sci. 402, 1–18 (2013) 48. Quéré, D.: Non-sticking drops. Rep. Prog. Phys. 68, 2495–2532 (2005) 49. Rao, A.V., Kulkarni, M.M.: Hydrophobic properties of TMOS/ TMES-based silica aerogels. Mater. Res. Bull. 37, 1667–1677 (2002) 50. Rao, A.V., Kulkarni, M.M., Pajonk, G.M., Amalnerkar, D.P., Seth, T.: Synthesis and characterization of hydrophobic silica aerogels

389

using trimethylethoxysilane as a co-precursor. J. Sol-Gel Sci. Technol. 27, 103–109 (2003) 51. Nadargi, D.Y., Rao, A.V.: Methyltriethoxysilane: new precursor for synthesizing silica aerogels. J. Alloys Compd. 467, 397–404 (2009) 52. Zhong, L., Chen, X., Song, H., Guo, K., Hu, Z.: Highly flexible silica aerogels derived from methyltriethoxysilane and polydimethylsiloxane. New J. Chem. 39, 7832–7838 (2015) 53. Venkateswara Rao, A., Kulkarni, M.M., Amalnerkar, D.P., Seth, T.: Superhydrophobic silica aerogels based on methyltrimethoxysilane precursor. J. Non-Cryst. Solids. 330, 187–195 (2003) 54. Gregg, S.J., Sing, K.S.W.: Adsorption, Surface Area and Porosity, 2nd edn. Academic, London (1982) 55. Hrubesh, L.W., Coronado, P.R., Satcher, J.H.: Solvent removal from water with hydrophobic aerogels. J. Non-Cryst. Solids. 285, 328–332 (2001) 56. Reynolds, J.G., Coronado, P.R., Hrubesh, L.W.: Hydrophobic aerogels for oil-spill cleanup? Intrinsic absorbing properties. Energy Sources. 23, 831–843 (2001) 57. P. R. Coronado, L. W. Hrubesh, J. G. Reynolds: Metod of oil spill recovery using hydrophobic sol-gels and aerogels. US2002/ 0185444 A1 58. Gurav, J.L., Rao, A.V., Nadargi, D.Y., Park, H.-H.: Ambient pressure dried TEOS-based silica aerogels: good absorbents of organic liquids. J. Mater. Sci. 45, 503–510 (2010) 59. Štandeker, S., Novak, Z., Knez, Ž.: Adsorption of toxic organic compounds from water with hydrophobic silica aerogels. J. Colloid Interface Sci. 310, 362–368 (2007) 60. Venkateswara Rao, A., Hegde, N.D., Hirashima, H.: Absorption and desorption of organic liquids in elastic superhydrophobic silica aerogels. J. Colloid Interface Sci. 305, 124–132 (2007) 61. Perdigoto, M.L.N., Martins, R.C., Rocha, N., Quina, M.J., GandoFerreira, L., Patrício, R., Durãesa, L.: Application of hydrophobic silica based aerogels and xerogels for removal of toxic organic compounds from aqueous solutions. J. Colloid Interface Sci. 380, 134–140 (2012) 62. Štandeker, S., Novak, Z., Knez, Ž.: Removal of BTEX vapours from waste gas streams using silica aerogels of different hydrophobicity. J. Hazard. Mater. 165, 1114–1118 (2009) 63. Belmabkhout, Y., Serna-Guerrero, R., Sayari, A.: Amine-bearing mesoporous silica for CO2 removal from dry and humid air. Chem. Eng. Sci. 65, 3695–3698 (2010) 64. Sayari, A., Belmabkhout, Y.: Stabilization of amine-containing CO2 adsorbents: dramatic effect of water vapor. J. Am. Chem. Soc. 132, 6312–6314 (2010) 65. Chen, C., Zhang, S., Row, K.H., Ahn, W.-S.: Amine–silica composites for CO2 capture: a short review. J. Energy Chem. 26, 868–880 (2017) 66. Maleki, H.: Recent advances in aerogels for environmental remediation applications: a review. Chem. Eng. J. 300, 98–118 (2016) 67. Linneen, N., Pfeffer, R., Lin, Y.S.: CO2 capture using particulate silica aerogel immobilized with tetraethylenepentamine. Microporous Mesoporous Mater. 176, 123–131 (2013) 68. Linneen, N.N., Pfeffer, R., Lin, Y.S.: CO2 adsorption performance for amine grafted particulate silica aerogels. Chem. Eng. J. 254, 190–197 (2014) 69. Wörmeyer, K., Smirnova, I.: Adsorption of CO2, moisture and ethanol at low partial pressure using aminofunctionalised silica aerogels. Chem. Eng. J. 225, 350–357 (2013) 70. Cui, S., Cheng, W., Shen, X., Fan, M., (Ted) Russell, A., Wu, Z., Yi, X.: Mesoporous amine-modified SiO2 aerogel: a potential CO2 sorbent. Energy Environ. Sci. 4, 2070–2074 (2011) 71. Begag, R., Krutka, H., Dong, W., Mihalcik, D., Rhine, W., Gould, G., Baldic, J., Nahass, P.: Superhydrophobic amine functionalized aerogels as sorbents for CO2 capture. Greenh. Gases Sci. Technol. 3, 30–39 (2013)

15

390 72. Venkateswara Rao, A., Kulkarni, M.M., Bhagat, S.D.: Transport of liquids using superhydrophobic aerogels. J. Colloid Interface Sci. 285, 413–418 (2005) 73. Hegde, N.D., Hirashima, H., Rao, A.V.: Two step sol-gel processing of TEOS based hydrophobic silica aerogels using trimethylethoxysilane as a co-precursor. J. Porous. Mater. 14, 165–171 (2007) 74. Truesdell, R., Mammoli, A., Vorobieff, P., van Swol, F., Brinker, C. J.: Drag reduction on a patterned superhydrophobic surface. Phys. Rev. Lett. 97, 044504 (2006) 75. Samaha, M.A., Vahedi Tafreshi, H., Gad-el-Hak, M.: Effects of hydrostatic pressure on the drag reduction of submerged aerogelparticle coatings. Colloids Surf. A Physicochem. Eng. Asp. 399, 62–70 (2012) 76. Rodriguez, J.E., Anderson, A.M., Carroll, M.K.: Hydrophobicity and drag reduction properties of surfaces coated with silica aerogels and xerogels. J. Sol-Gel Sci. Technol. 71, 490–500 (2014) 77. Kim, A., Kim, H., Lee, C., Kim, J.: Effective three-dimensional superhydrophobic aerogel-coated channel for high efficiency water-droplet transport. Appl. Phys. Lett. 104, 081601 (2014) 78. Huber, L., Zhao, S., Malfait, W.J., Vares, S., Koebel, M.M.: Fast and minimal-solvent production of super insulating silica aerogel granulate. Angew. Chem. Int. Ed. 55(17), 4753–4756 (2017)

Kazuyoshi Kanamori received his PhD from the Department of Materials Chemistry, Graduate School of Engineering, Kyoto University (Japan) in 2005. After 2 years as a postdoc, he became assistant professor in the present affiliation in 2007. His research interests include liquid-phase synthesis and characterization and application of porous materials – especially aerogels. Chemistry of organic-inorganic hybrids is his major area of concern.

Ana Stojanovic received her PhD from University of Zurich in 2012. After PhD she joined Empa Switzerland in building energy materials group, first as postdoc then as scientist. Her main activities were linked to process-scale-up and lab-to-market transfer of aerogel-based materials. She is currently working for a Swiss company which produces silica gel mostly for HPLC applications.

K. Kanamori et al.

Gerard M. Pajonk was born in Lyon in 1939. He obtained a degree in chemical engineering at Ecole Superieure de Chime Industrielle de Lyon (ESCIL) in 1963 and a PhD at the same university in heterogeneous catalysis with Prof. S.J. Teichner in 1970. Then he took a postdoc degree at Stanford University, California with Prof. M. Boudart. He made research and teaching work both at Montreal, New York, Stanford and the Technion of Haïfa and Lyon. He taught more than 30 PhD students in sol–gel chemistry and catalysis in France and abroad, and published at least 250 scientific papers in many chemistry journals and holds many patents. He specialized in materials sol–gel preparation, physico-chemical characterizations, and supercritical drying processes (xero- and aerogels) and cryogels.

Dr. Digambar Y. Nadargi is a research associate at Solapur University, India. Prior to RA position, he worked as a chief technical officer, at Keey Aerogel France, and postdoctoral fellow at Empa, Swiss Federal Research Laboratories, Switzerland. He did his PhD in hydrophobic properties of silica aerogels, at Shivaji University, India. His research interests are nanomaterials, silica/non-silica aerogels, thin films, gas sensors, and photocatalysis.

A. Venkateswara Rao received his doctorate from Sardar Patel University with Professor A.R. Patel in 1980 and became lecturer in physics at Shivaji University. After a postdoctorate at Ecole Polytechnique in Palaiseau, he did a sabbatical as Rhône-Alpes professor at Lyon-I University in 1992 and later at Poitiers University in 2005. His research interests are flexible aerogels and ambient pressure drying aerogel preparation.

15

Superhydrophobic and Flexible Aerogels and Xerogels Derived from Organosilane Precursors

391

15

Kazuki Nakanishi received his PhD from Kyoto University in 1991. From 1986 to 2018, he worked at Kyoto University as assistant/associate professor, where he launched first organic-inorganic hybrid aerogels project with Drs. Kazuyoshi Kanamori and Mamoru Aizawa. His current position is full professor at Nagoya University and project-specific professor at Institute for Integrated Cell-Material Sciences (iCeMS), Kyoto University.

Matthias M. Koebel received his PhD from Brown University in 2004. After a postdoctoral stay at UC Berkeley with G.A. Somorjai focusing on nanocatalysis, he joined Empa back in his home country – Switzerland – in 2006 where he began building a research group in soft chemistry and aerogels. His core activities are linked to process-scale-up and lab-to-market transfer of nanomaterials science.

Sodium Silicate-Based Aerogels by Ambient Pressure Drying

16

A. Venkateswara Rao, Shanyu Zhao , Gerard M. Pajonk, Uzma K. H. Bangi, A. Parvathy Rao, and Matthias M. Koebel

Contents

Abstract

16.1 16.1.1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393 Sodium Silicate Chemistry and Technical Relevance . . . 394

16.2 16.2.1

Comparison of Commonly Used Synthetic Methods . . Evolution of Different Sodium Silicate-Based APD Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary of Commonly Used Synthetic Methods . . . . . . . Waterglass-Based Aerogel Composites . . . . . . . . . . . . . . . . . .

16.2.2 16.2.3

396 396 403 404

16.3.6

Materials Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Effect of the Silica Concentration in the Sol . . . . . . . . . . . . . Effect of Sol pH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Effect of Aging (ta) Period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Effect of Solvent Exchange . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Silylating Agents and the Hydrophobization Treatment Duration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . General Comments About Parameter Optimizations . . . .

16.4

Applications and Commercialization . . . . . . . . . . . . . . . . . . 411

16.5

Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411

16.3 16.3.1 16.3.2 16.3.3 16.3.4 16.3.5

405 406 407 408 408 409 410

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412

Silica aerogels are the most commercially relevant aerogel materials. By volume, supercritically dried blanket composites are still leading global sales by a substantial amount; however, particle-based aerogels such as granulate and powder are cost-competitive alternatives. This chapter summarizes the last three decades of research and industrial process development in the field of “low-cost,” sodium silicate-based aerogel preparation by means of ambient pressure drying, illustrating key developments and milestones in both academic research and process engineering fields. Key process steps such as gelation, aging, hydrophobization, and APD drying are analyzed in detail in the context of feasibility, simplicity, product quality, and scalability. The chapter finishes with a brief discussion of key process parameters and their effect on the physical properties of the obtained aerogel materials as well as a current snapshot of the most promising applications for particle-based aerogels. Keywords

A. V. Rao · A. P. Rao Air Glass Laboratory, Department of Physics, Shivaji University, Kolhapur, Maharashtra, India S. Zhao (*) Building Energy Materials and Components Laboratory, EMPA, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland Department of Chemistry, Brown University, Providence, RI, USA e-mail: [email protected] G. M. Pajonk Laboratoire des Matériaux et Procédés Catalytiques, Université Claude Bernard Lyon 1, Villeurbanne, France U. K. H. Bangi Department of Physics, PAH Solapur University, Solapur, India M. M. Koebel (*) Building Energy Materials and Components, Laboratory for Building Technologies, Empa, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland e-mail: [email protected]; [email protected]

Sodium silicate · Waterglass · Ambient pressure drying · Hydrophobization · Parameter optimizations · Commercialization

16.1

Introduction

Commercially available aerogel materials today are primarily of the silica variety. The importance of this particular chemistry rests on two main pillars, namely the availability of inexpensive silica precursors and compatible silane hydrophobes as well as their early scientific discovery and academic relevance. Silica aerogels were first prepared by S. S. Kistler at the College of the Pacific, in 1931, using sodium silicate as a precursor and supercritical drying [1]. However, the more elaborate and time-consuming supercritical drying procedure, coupled with the discovery of simpler preparation routes for porous silicates,

© Springer Nature Switzerland AG 2023 M. A. Aegerter et al. (eds.), Springer Handbook of Aerogels, Springer Handbooks, https://doi.org/10.1007/978-3-030-27322-4_16

393

394

A. V. Rao et al.

caused people to lose interest in aerogels for roughly three decades. In 1968, a team of researchers led by Prof. S. J. Teichner at the University of Lyon, France, pioneered the alkoxide-based chemistry for silica aerogels using tetraethoxysilane (TEOS) and tetramethoxysilane (TMOS) to prepare sols and gels. Aerogels could now be prepared within by high temperature supercritical drying from their respective alcohol pore fluids in only 1 day which was significantly faster than Kistler’s original method [2]. The major simplification was the omission of an additional solvent exchange step when using alcohol as a main solvent. In 1995, Prakash et al. reported the first preparation of ambient pressure-dried sodium silicate-based films by surface chemical modification of wet silica films prior to drying [3, 4]. Soon thereafter, Deshpande [5] and Schwertfeger [6, 7] were able to produce aerogels from a waterglass precursor through Na+ ion exchange followed by surface chemical modification and an ambient pressure drying (APD) protocol. This exciting period marked the birth of the two archetypal silica aerogel preparative techniques, namely alcohol-based alkoxide and aqueous sodium silicate chemistries. These are the technological foundation of today’s two major aerogel production companies Aspen Aerogels and Cabot. Both systems have their advantages: the alkoxide precursor route may appear more costly when comparing only raw materials cost, but processing is simplified because the pore fluid is already an organic solvent, which in theory allows a solvent exchange-free process. Though sodium silicate is the cheaper precursor, the use of an aqueous gel medium requires more cumbersome solvent exchange and drying process technology. Following the cornerstone works by Schwertfeger and Fig. 16.1 Histogram of the published history of the ambient pressure dried (APD) waterglassbased silica aerogels

14

Desphande, the early 2000s sparked a strong academic interest in ambient pressure dried sodium silicate aerogels [6, 8]. A large number of early studies described the effect of specific synthetic parameters on the final material properties while simultaneously targeting process simplifications as shown in Fig. 16.1. After 2012 and with the publishing of the first edition of the Aerogel handbook, researchers began to expand their studies into three main areas, namely i) more efficient “one-pot” synthesis routes geared towards process improvement and commercialization [9–12], ii) the development of aerogel composites with properties tailored towards specific applications [13–22], and iii) efforts to source the raw materials for aerogel preparation from silica-rich bio- or industrial wastes through suitable recycling protocols [23, 24]. The commercial significance of silica aerogels derived from low-cost sodium silicate precursors was recognized early on by Kistler, yet even now – a decade and a half into the second attempt to aerogel commercialization – there is still room for improvement and the need to develop a faster and cheaper production method is bigger than ever. In this chapter we are describing the chemistry and preparation of water-soluble silicate-based aerogel materials, their technical relevance as well as their applications in industry.

16.1.1 Sodium Silicate Chemistry and Technical Relevance Sodium silicate is the common name of hydrated alkali silicate compounds with the formula of Na2O ∙ n SiO2; it is well known as waterglass or liquid glass. Traditionally,

APD waterglass composites APD waterglass based aerogels Introducingaerogel material

10 8 6 4 2 0

19 3 19 2 9 19 5 9 20 8 0 20 0 0 20 1 0 20 2 0 20 3 04 20 0 20 6 0 20 7 0 20 8 0 20 9 1 20 0 1 20 1 1 20 2 1 20 3 1 20 4 1 20 5 1 20 6 1 20 7 18

Number of publications

12

16

Sodium Silicate-Based Aerogels by Ambient Pressure Drying

sodium silicate is used in the area of adhesives, passive fire protection, anticorrosion paints, ceramics technology, detergent and binder auxiliaries for textile and lumber processing and of course also as a precursor for the preparation of silica

Statistical relevance (%)

50 40 30 20 10 0 0.5

1

1.5 2 2.5 3 3.5 SiO2/Na2O molar ratio

4

4.5

Fig. 16.2 Histogram of SiO2:Na2O molar ratio used in the literature for silica aerogel synthesis

Fig. 16.3 Three-component phase diagram for the system SiO2–Na2O–H2O showing the range of compounds, anhydrates, and solutions stable at room temperature, together with commercial products [25]. (Modified from Copyright 2002, Elsevier B.V)

395

gels. The applications of sodium silicates rely significantly on its composition, which follows the SiO2-Na2O-H2O ternary phase diagram, shown in Fig. 16.3 [25]. A well-known member is sodium meta-silicate, Na2SiO3, which is an anhydrate with 50% SiO2 and Na2O. As far as aerogel synthesis is concerned, high sodium ion contents are undesirable in terms of final materials properties because sodium ions are undesirable guest species in the final aerogel. Hence a solution with a low molar ratio (e.g., SiO2:Na2O ¼ 1:1), consisting primarily of monomers (SiO44) and dimers (e.g., Si2O52), is less suitable for aerogel preparation. Sodium silicate solutions with a molar ratio above 3:1 (Fig. 16.2), and thus a higher proportion of polysilicate species, are more commonly used for aerogel synthesis [26]. Commercially available sodium silicate solution typically have SiO2: Na2O molar ratio is in the range of 1.5:1– 3.5:1(region A in Fig. 16.3). Chemically speaking, these silicate hydrates are between true solutions (region C, such as Na2SiO3), and colloidal suspensions (region B), which consist of a mixture of polymeric and oligomeric silicates (depending on the molar ratio of SiO2:Na2O) and will be largely protonated owing to the weakness of silicic acid [25]. The solution pH also depends on SiO2/Na2O molar ratio. A saturated waterglass solution with relatively high

H2O Na+ion removal Dilution Colloids and gels

Solutions C

Commercial sodium silicate solutions

A B

50

50 Hydrous glasses

D

NaOH

Na2O

Na2SiO3 50

Na2Si2O5 Na2Si4O9

Soluble powders

SiO2

16

396

A. V. Rao et al.

Table 16.1 Sodium ion removal and gel surface modification in the synthesis process to target a good performance silica aerogel Action Removal of Na+ ions

Steps applied in the synthesis IRa Ion exchange resin [6, 35]

Hydrophobization

HAb Mixed with precursor [37]

IRc IRd Na+ removal by washing [36]

HAc HAd Single-step [38–40]

HAe Two-step solvent exchange [41]

IRe Ion removal during solvent exchange / hydrophobization [37] – Gas phase modification of dry aerogel/xerogela

a

A hydrophilic xerogel or aerogel sample can be hydrophobized by organosilanes after drying [42], although in conjunction with APD does not yield low-density aerogels but rather high density xerogels

SiO2/Na2O ratio is a viscous liquid with a density around 1.4 g cm3 and a pH value around 12.5. Commercially, waterglass is synthesized by the solidstate reaction of quartz sand with sodium hydroxide, sodium carbonate, and/or sodium sulfate at elevated temperature, which leads to the solid form of sodium silicate or cullet. In a second step, cullet is dissolved into aqueous sodium hydroxide, the stoichiometry between cullet and base defining the SiO2:Na2O molar ratio of the waterglass solution. Given the wide abundance and inexpensive nature of the reactants, waterglass is among the least expensive commercial, soluble sources for silica. Besides, it is easy to handle and does not have the flammability hazards of their metal organic silicon alkoxide counterparts such as TEOS or TMOS. It is also chemically long-term stable under standard conditions of use and storage. Consequently, waterglass is an ideal precursor for silica gel and aerogel manufacture at industrial scale. In order to make aerogels more competitive with other conventional insulation materials, cost reduction is a major challenge for industrial manufacturers and possible newcomers to this market. There are two main cost drivers for aerogel materials, namely i) raw materials cost and ii) the process cost which is basically the sum of discounted capital investments into the manufacturing line (CAPEX) and the cost of operation (OPEX). Process cost depends largely on the different process steps, where a large part of industrial R&D is focusing. For silica aerogels, raw materials typically account for roughly half the total manufacturing cost. The lion’s share goes to the silica gel precursor whereas hydrophobization agents typically come second. Note that for an industrial process to be viable, all washing and exchange solvents must be completely reused and that losses of solvent through volatile organic carbon (VOC) emissions also need to be addressed. It is only through completely closed process design that cost-effective manufacturing of aerogels is even theoretically conceivable. Currently, the Cabot Nanogel product and JIOS powder products (▶ Chap. 64) are manufactured from waterglass precursors in a process RIa/HAe and IRe/HAb process according to the definition defined in

Table 16.1 and the graphical representation shown below in Fig. 16.4 [9].

16.2

Comparison of Commonly Used Synthetic Methods

16.2.1 Evolution of Different Sodium SilicateBased APD Aerogels At this point, it seems worth mentioning that the chemistry of silicates in aqueous solution is rather complex. In some published work, the alkoxysilane sol–gel chemistry language had been used to describe the formation of hydrogels from waterglass or ion exchanged waterglass (silicic acid oligomers). In waterglass chemistry for example there is no hydrolysis step and hence the use of this term is technically incorrect. Despite the obvious similarities between waterglass and alkoxide-based silica gel systems, the limitation of such analogies and the respective mechanistic differences made must always be kept in mind. The main challenges in the synthesis of aerogels from sodium silicate are linked to the removal of sodium ions on the one hand and the introduction of the hydrophobe on the other hand. Both actions can be done at various stages of the synthesis process, triggered correspondingly at the sol, hydrogel, aged hydrogel, or organogel state. Common strategies for sodium ion removal (IR) and hydrophobe addition (HA), respectively, are IRa, IRc, IRd, IRe and HAb, HAc, HAd, and HAe, respectively, which will be elucidated in more detail in the following discussion. As previously mentioned, sodium silicate is an ideally suited and commonly used precursor for sol–gel materials. In order to better understand its chemistry, let us discuss the different process steps for sodium silicate-based APD aerogels. Figure 16.4 shows the process steps i) through v) from sodium silicate to the final aerogel material, which are: (i) Preparation of a silicate sol by dilution of a sodium silicate solution and optional ion exchange (ii) Gelation of the gel, typically initiated by neutralization (or base addition in the case of an ion exchanged sol)

16

Sodium Silicate-Based Aerogels by Ambient Pressure Drying

State

a Precursor

b Sol

397

d Aged gel

c Gel

e Organogel

Hydrophobe choices:

Hydrophobic grafting unit: Trimethylsilyl group (TMS)

TMS

H TMS TMS Cl TMS N TMS O

(HMDSO) (TMCS)

Na2O∙nSiO2 Na+ TMS Depleted resin

f Aerogel/xerogel

(HMDZ)

Options for hydrophobe addition (HA) HAb HAc HAd

HAe

16

Protonated resin IRc

SiO2∙nH2O IRa

IRe

IRd

Options for sodium ion removel (IR) Steps

Sol preparation

Gelation

i)

ii)

Aging

Solvent exchange

iii)

iv)

Drying v)

Fig. 16.4 Schematic showing the generally accepted pathways to APD silica aerogel preparation from sodium silicate through gelation to aging, solvent exchange, and drying

(iii) Aging of the gel to improve mechanical strength needed for APD (iv) Solvent exchange into an organic solvent system suitable for APD (v) Ambient pressure drying of the gel by evaporation of the pore liquid Let us begin with the discussion of the original method described by S. S. Kistler in his pioneering work in 1931 [27]. His early works describe gel formation (step ii) from sodium silicate (specific gravity 1.15) adjusted to equivalent SiO2 concentration between 8 and 24 weight percent (step i). Following aging (step iii), the aqueous pore fluid was exchanged to ethanol or methanol (step iv) and the gel was dried by high temperature supercritical drying directly from ethanol (step v). The process simplification which allows for effective APD was developed by Desphande and Schwertfeger and consists primarily in combining the solvent exchange together with a surface modification treatment to render the gel surfaces hydrophobic (HAe). HAe is probably the most common hydrophobe addition mode in modern waterglass-APD aerogel synthesis today. Yet, from the time of their discovery and first patents, substantial developments have been made to further simplify and improve process efficiency, which eventually helped enable the recent industrialization of aerogels. Of particular relevance for scalable processes is the reduction of the cycle time. Hence, the combination of aging (including optional ion removal by

washing), solvent exchange, and surface modification, and drying – steps iii) through v) – are where development efforts are focusing and there is most room for improvements. Yet, also the specific conditions used to initiate the sol–gel transition have a strong influence on the final materials. In the following, we shall compare the most commonly used synthetic methods step by step and elaborate on their differences, with a view on key developments toward reducing process time and improving efficiency.

Neutralization and Sodium Ion Removal In alkoxysilane chemistry (▶ Chap. 13), the gelation process is initiated by the hydrolysis of alkoxy groups Si-OR, which leads to the formation of condensable silanol groups. By definition, hydrolysis is a chemical reaction in which chemical bonds in a molecule are broken and a water molecule enters to become a part of the final product, typically resulting in the formation of -OH functional groups. In the waterglass case, no actual hydrolysis takes place, but the gelation of basic sodium silicate is triggered by simple acid base chemistry, that is, through partial neutralization of Si-O Na+ ion pairs according to

Si

Na+ O−

+ HX

+ Na+ X−

Si OH

398

A. V. Rao et al.

where HX is a Bronsted acid and X its corresponding base. Typical acid sources used to initiate gelation by neutralization are mineral acids or low-molecular organic carboxylic acids such as acetic, oxalic, or citric acid. Alternatively, sodium ions can be removed by ion exchange at the sol-stage (IRa) that is before initiating the gelation. Ion exchange of Na+ by H+ converts the sodium silicate into a silicic acid solution and causes the pH of the sol to drop from around 12.5 to values between 1 and 2.5. Such a sodium-free silicic acid sol are then gelled by neutralization with small amounts of base (typically ammonia or sodium hydroxide). Whether one starts from sodium silicate sols or from ion exchanged silicic acid sols, the partial neutralization initiates condensation reactions, whose gelation kinetics are fastest at intermediate pH. Mechanistically, both the formation of free Si–OH covalent bonds and the overcoming of electrostatic repulsion interactions between the oligomeric silica species are prerequisites for aggregation of colloidal silicate particles and gel formation. In this context, let us focus on the different sodium ion removal process options: When it comes to the final aerogel properties, the covalent Si-O-Si bonding character of the silicate network is quite essential for the final material properties. It is found that even relatively low residual Na+ levels cause large macropore fractions and the resulting aerogels turn white and structurally unstable. In other words, a suitable sodium ion removal strategy is the foundation of any sodium silicate-based aerogel synthesis route. Sodium-free gels are always obtained when using an ion exchanged silicic acid as a precursor (IRa, Fig. 16.4). This is the most straightforward synthetic method used for the preparation of monoliths and high quality “academic” samples [6, 28, 29]. However, the ion exchange process adds significant process complexity and production cost and is a bottleneck for the large-scale commercial production of silica aerogels. Alternatively gels

Concentration of solute

a

Cmax

Na Si/O tetrahedron

can be prepared from sodium silicate and the sodium ions removed later on by washing with water or steam [30–33] after gelation (IRc) or aging (IRd) of the hydrogels. In yet another variation, the so-called single-step solvent exchange surface modification or combined solvent exchange and hydrophobization (CSH) method IRe, sodium ions are extracted together with the aqueous phase during solvent exchange and hydrophobization [9–11, 34]. But why is sodium ion removal so crucial for obtaining high-quality materials? Aside from experimental observations that residual Na+ negatively affect pore size distributions and mechanical strength, there is only little published work on this topic. Yet, from a conceptual standpoint there are at least three factors that affect the mesostructure of a silica gel network: First, at the molecular level, visualized in Fig. 16.5a, the presence of sodium ions reduces the fraction of covalent bonds formed and consequently also the overall framework connectivity. Pauly et al. reported [43] that the covalent, “non-ionic” assembly pathways typically lead to more stable silica gel mesostructures in comparison to electrostatically assembled analogues. For a SiO2 to Na2O ratio 3:1, at least one oxygen in each silicate tetrahedron is negatively charged and the average connectivity between Q2 and Q3. This is considerably lower than for a sodiumfree network where practically observed average connectivities are between Q3 and Q4. As it is known, for example in organosilane-derived aerogels (▶ Chap. 15), a lower connectivity (e.g., T2/T3 in the case of methyltriethoxysilane (MTES) or methyltrimethoxysilane (MTMS) derived aerogels) generally leads to reduced mechanical strength and higher macroporosity. Secondly, at the primary particle level (Fig. 16.5a, b), the nucleation and growth of small colloids is affected by both

b

c

Cmin

Cs

Pre-nucleation

Nucleation

Growth

10 nm

Time Fig. 16.5 Schematic showing (a) the general silica formation from sodium silicate through oligomers to primary particles, (b) TEM image showing the primary particle, and (c) SEM image showing the secondary particle network

16

Sodium Silicate-Based Aerogels by Ambient Pressure Drying

the reduced Si-O-Si bonding order and the ionic interactions of Na+ cations [44], which can influence structural features and mesoporosity of the gels. Thirdly, interactions of colloids are also relevant at secondary particle level and again affected by surface charge and size of the primary particles [45], as well as the ionic strength and nature of the solvents. Although the presence of sodium ions will undoubtedly alter the gel structure at molecular, primary, and secondary particle levels, the dynamic nature of silicate bonded networks seems to be rather forgiving. The fact that virtually identical aerogels are obtained from ion-exchanged sols and non-ion exchanged ones with subsequent Na+ removal suggests that the gel structure for the two systems seems rather comparable in the end. This means that the structure forming effects of Na+ are either rather insignificant or completely reversible, that is, that removal of Na+ ions after gelation makes the gels converge toward the same particle network structure with primarily covalent bonding character.

Condensation (Gelation) The chemistry of aqueous silicates and silicic acid systems is quite complex: repulsion and ion pairing are known to govern intermolecular interactions in alkali silicate solutions [46]. More than 20 different oligomer species of waterglass containing 2–8 silicon atoms have been identified by means of 29Si nuclear magnetic resonance (NMR) spectroscopy [47]. Alkaline silicates tend to form ring and cage-like structures. The condensation kinetics depend strongly on the pH value since they are governed by electrostatic interactions of charged molecular species and clusters: At pH >10, that is, where sodium silicate is the prevalent species, condensation reactions are slow, because of negative charge repulsion, lack of Si-OH species for condensation and increased competition with the backward reaction (dissolution). In addition, the condensation of sodium silicate species with themselves or a partially neutralized silicic acid/silicate molecule is disfavored, instead of this, the dimerization of two molecules of ortho-silicic acid has substantially smaller kinetic barrier, because H4SiO4 is electroneutral (the isoelectric point of bulk silica ranges from 2 to 4) compared to silicate, there is significantly less electrostatic repulsion, and the reaction is also favored from an entropy point of view (formation of H2O as a reaction product). This is why a waterglass solution is rather stable with long shelf life and does not gel unless partially neutralized. This is certainly a simplified view and the complete picture is far more complex, but the example of the dimer formation should serve an illustrative purpose and help the general discussion. At pH 95%. The choice of pore liquid affects not only the aging process (higher water content impacts kinetics of dissolution and reprecipitation of silica from and onto the gel network) but also the surface chemistry and solvent exchange step itself. As described earlier, solvent exchange, surface modification, and optional Na+ ion removal (Fig. 16.4 iv, HAe, IRe) can be done as three different treatments separately, much in the same way as ion exchange can be done during sol preparation (HAb). In general, a process with separate steps is more

a

time-consuming and less desirable, at least if the same performance can be achieved. This is also where researchers are being creative and are finding new ways to combine and simplify process steps, eventually without taking a noticeable reduction in performance. Some of the more recent studies [37–40, 54, 55] combine solvent exchange and surface modification into a single process step, typically by a treatment with a solution containing an alcohol, a saturated linear hydrocarbon, and a hydrophobization agent, for example, IPA/n-hexane/TMCS [38]. For the preparation of aerogel powders from sodium silicate, solvent exchange and surface modification are even directly combined with gelation and aging steps, that is, n-hexane/HNO3/HMDZ [9, 56, 57]. The hydrophobization of silica gel surfaces is well documented and understood and has been described already earlier [58]. A schematic of the surface modification of silica with trimethylsilyl (TMS) groups is shown in Fig. 16.8b:

300 293

# Samples in bin

250 200 150 100

75

62 1

6

1

MTMS

APTES None

HMDZ

TEOS

HMDSO

TMCS

1

MPTMS

9 18 0

IPTMS

50

b

HO HO O Si

HO O HO

Si

OH OH

O

Si Si

O O

O

O Si

Si Si OH OH

O O

OH OH

O

O Si

H3C

CH3

+ 4 CH3 Si

Si

OH OH OH

OH

CH3

R

CH3 Si

H 3C

OH OH

Si

CH3

O O Si O Si Si OH CH3 H 3C O O HO O O O Si Si OH HO O O O CH3 Si O O H3C Si Si O Si CH3 Si O Si OH CH3 OH H3C OH OH CH3

R: OH, Cl, OSi(CH3), NSi(CH3) Fig. 16.8 (a) Histograms of agents used for silica aerogel modification, (b) hydrophobization of a surface silanol group in a silica gel with trimethylsilane (TMS) [154]. (Modified from Copyright 2018, Elsevier B.V)

16

402

A. V. Rao et al.

Si-OH

Projection H

0 2 4 6 8 29Si

c

0 –50 –100 chemical shift (ppm)

TMS WG/SE/TMCS

Q3 Q4

Quantitative 29Si

Projection 29Si

–2 0 2 4 6

Q3

8 Q4

Si-OEt Q3

Internal 0.357 nm Fig. 16.9 (continued)

1H

–2

chemical shift (ppm)

Projection 29Si

1

Quantitative 1H

Si-OH TMS

Quantitative 29Si

29Si

chemical shift (ppm)

Q4

WG/TMCS

Q3 Q4

1H

TMS

C H

TMS

Projection 1H

a

Si O

b

Quantitative 1H

Drying The APD drying mechanism of silica aerogels in general is rather well understood. In a first step, pore fluid is evaporating from the macroscopic gel surfaces. This leads to a shrinkage of the gel and building up of internal strain – similar to charging a spring – up to the point where it can no longer contract further. At this point, further evaporation of pore fluid causes a gradual emptying of the mesopores. It is during this stage where gas, liquid, and solid phase coexist within the gel and stresses on the gel are maximal. As the pores are nearing the point of

complete emptying, the internal strain competes against capillary forces, eventually allowing the gel to retrieve most of its original pore volume and dimensions. This reversible shrinkage is also called spring-back effect [4]. The exact extent of the maximum deformation state of the surface-modified gel structure depends on a number of factors, namely the solvent type, vapor pressure, surface tension, etc. Together with the drying

Ethoxy CH2 CH3 TMS

Silanol groups on the gel surface serve as active sites for covalent grafting (endcapping) with TMS groups through reaction with suitable silane coupling agents (TMCS, HMDSO, or HMDZ). However, if gels are modified with a large excess of hydrophobe, the resulting gels and aerogels should display rather similar chemistries despite their different precursors (waterglass versus alkoxides), hydrophobization agents (HMDZ, TMCS, HMDSO), as well as gelation and APD drying solvents (water, ethanol/heptane). Solid-state MAS NMR spectroscopy has proven a valuable tool to quantify the hydrophobe content of the gel for different treatments, TMS contents are typically in the range of 22–27 wt.% [59]. Another surprising fact is that the exchange kinetics of hydroxyl bearing species (water / alcohol) seems rather fast under the typical conditions used for solvent exchange (room temperature – 70  C, atmospheric pressure). Consequently, for pure hydrogels exchanged with alcoholcontaining solvent mixtures over longer periods of time, replacement of silanol by alkoxy groups was observed (Fig. 16.9a), and identified/quantified by means of 1H29Si heteronuclear solid-state NMR (Fig. 16.9b, c) [59].

0 –50 –100 chemical shift (ppm)

Fig. 16.9 (a) Schematic illustration of a silica nanoparticle: the calculations assume that the surface layer consists of Q3 (with either hydroxyl or ethoxy groups) and Q4 attached to TMS groups; the inside of the particle consists exclusively of Q4 with four siloxane neighbors, (b) 1 H29Si heteronuclear solid-state NMR spectra of sodium silicatebased silica aerogels together with their 1H and 29Si projections and quantitative 1H and 29Si spectra, the aerogels prepared from TMCS/nheptane modification without or with ethanol solvent [59]. (Copyright 2015, American Chemical Society)

16

Sodium Silicate-Based Aerogels by Ambient Pressure Drying

rate and this will influence the structure and final form (monolith, granule, powder) of the products. Note that high surface coverage of the inner gel surface with hydrophobic groups to suppress condensation reactions in the shrunk state is an essential prerequisite for APD. The reversible shrinkage mechanism that we discussed above apply to both alkoxide-based and sodium silicatebased SiO2 aerogels. In the cases of the inorganic aerogels, for example, zeolite [60, 61] or alumina [62], if their inner surfaces are modified, they can also be dried from ambient pressure atmosphere. Hence, in the next section, the uniqueness of drying routes for sodium silicate aerogel shall be discussed in more details. In the literature, four forms of APD waterglass aerogel materials have been reported, that is, monolithic, granulate, bead/sphere, and powder. Monolithic samples are normally dried slowly and in two stages: first, the majority of the pore fluid is evaporated at close to ambient temperatures (25–75  C, Fig. 16.10a) or in a nearly closed chamber with small pinholes [41], the step is typically lasting over 10 h (Fig. 16.10b). For granulate and powder materials, the first predrying step is typically done over 60  C or completely omitted. The second step aims to cross the springback regime as quickly as possible and is done at higher temperatures between 150  C and 200  C. These higher drying temperatures increase the rate of removal of solvents by improved heat and mass transfer (convection), which generally helps maximize the spring-back by preventing an unnecessarily long remaining of the gel in the shrunk state where condensation reactions are competing against spring-back. Considering much faster heat and mass transfer in a small size

403

sample, the overall drying time of powder and granulate is much shorter than monolithic samples, for example, 30– 60 min drying reported by Bhagat [31, 37] and Joung [9]. Overall, regarding to the drying efficiency of the sodium silicate aerogels, powder and granulate are much higher as compared to monolithic samples. More recently, Shiller et al. have reported an alternative APD mechanism for sodium silicate-based gels by means of a pore gas generated through a chemical generation which is then used to displace the pore fluid [63].

16.2.2 Summary of Commonly Used Synthetic Methods Multistep Solvent Exchange and Hydrophobization One of the classical waterglass-based routes goes through a multistep solvent exchange: In this approach, the pore fluid is first exchanged from water to alcohol or acetone, and then, in a second step into an organic hydrocarbon mixture, also containing a reactive hydrophobe. Typical final solvent and hydrophobe systems are linear hydrocarbons or isomer mixtures thereof such as heptane of hexane and trimethylchlorosilane (TMCS), respectively. As mentioned previously, the overall procedure is time consuming, yet the prior removal of water helps to accelerate the hydrophobization reaction. With most of the pore water and alcohol removed from the system, the hydrophobization agent TMCS [54] reacts only with surface silanol groups and is not consumed by side reactions (16.1) and (16.2).

b

a

90

180

120

80

Sphere/bead Granulate Monolith

70

100 80 60

60 40 30 20

20

10

0

0 Initial drying temperature (°C)

Powder Sphere/bead Granulate Monolith

50

40

25 40 50 60 70 75 80 90 100 120 142 170 180 190 200

# Samples in bin

140

Powder # Samples in bin

160

0.5 1 2

4

6 10 13 15 22 26 28 49 53 Drying time (hrs)

Fig. 16.10 Histograms of (a) initial drying temperature, and (b) total drying time for the preparation of four different forms of aerogel materials, powder: tens to several hundred μm; sphere/bead: several hundred μm to 1–2 mm; granulate: several mm; monolith: over several cm

16

404

A. V. Rao et al.

R–OH Cl H 3C Si CH3 H 3C R=H, Alkyl

OR H 3C Si CH2 H 3C

HCl ð16:1Þ

2

OR Si H3C CH3

H 3C

CH3 O Si Si CH3 H3C CH CH3 3

H 3C

ROR ð16:2Þ

O CH3 Si Si CH H3C CH3 CH3 2

H 3C

2H

2

OH Si H3C CH 3

H 3C

ð16:3Þ

Combined Solvent Exchange and Hydrophobization (CSH) The more industrially relevant CSH method combines solvent exchange and hydrophobization into a single step. To do so, the hydrogel is immersed into a mixture of an organic solvent/hydrophobe mixture, typically (hexane or heptane)/ TMCS [38–40]. To facilitate the transfer from the organic into the aqueous phase, amphiphilic molecules such as isopropanol (IPA), ethanol, or methanol are typically added to serve as phase transfer catalysts. For system with very fast sol–gel and exchange kinetics, primarily aerogel powder slurries, alkaline hydrophobes (hexamethyldisilazane, HMDZ) can even be added together with a hydrocarbon solvent together with the sol and the two phase mixture gelled by neutralization [37]. In such a minimal process all necessary process steps gelation/aging/solvent exchange/surface modification/sodium ion removal (steps c, d, e, i, and j) can be combined into a single one, in which case the overall sol-to hydrophobic gel process time can be significantly reduced to just a few hours [9–12, 56, 57]. Note that in the case of CSH routes, the equilibration of pore liquids inside a gel is controlled by diffusion of solvent and/or reagent like in any solvent exchange. The treatment time necessary to ensure a complete exchange and chemical reaction increases with the square of the smallest dimension of a given gel body. For this reason, it is generally far more economical to produce smaller particles or granular samples rather than larger monoliths. Certainly, the largest process time savings can be achieved already when replacing three consecutive exchanges with a single step (CSH) process. This is the reason why industrial CSH is the only viable option for the preparation of sodium silicate-based aerogels.

On the other hand, there is the question of the higher hydrophobe consumption in CSH. At 4–8 US$/kg, hydrophobization agents such as HMDZ or TMCS are the most expensive components in the aerogel synthesis and losses are to be avoided at all cost. Nevertheless if used in aqueous or mixed aqueous/alcohol media, they are known to hydrolyze rather quickly to form different hydrolysates (TMS-OH, TMS-OR) as well as the silylether TMS-dimer hexamethyldisiloxane (HMDSO) as shown in (16.1) and (16.2). In the case of TMCS, during the exchange of the pore water, the water and alcohol solvent both consume TMCS. If large amounts of pore water are present, the formation of silanols is the preferred reaction [(16.1), R ¼ H]. The hydrolysates TMS-OH and TMS-OR and the alcohol are amphiphilic molecules which promote the phase transfer and hence promote solvent exchange from the aqueous to organic medium while actively hydrophobizing the inner gel surfaces. Depending on the partition function in the organic and aqueous phases, respectively, a significant fraction of the hydrophobe typically ends up in the aqueous mixed phase. Since a large excess (>10 based on the total number of the gel’s accessible silanol groups) of hydrophobe must be used to guarantee sufficient hydrophobization to allow for APD, hydrolysis and silylether formation must be well understood and the side reaction products recirculated at industrial scale. Note that the silylether HMDSO can also be hydrolyzed under acidic conditions, thus allowing for continuous regeneration of reactive trimethylsilanol species in the reaction mixture (reaction 16.3), which can then go on to react with free silanol groups on the gel surface at the hydrocarbon/water solvent interface. Despite the more challenging handling of hydrophobe streams in both aqueous and organic phases, CSH is always the method of choice in an industrial process because of the much higher space yield for a given size installation.

16.2.3 Waterglass-Based Aerogel Composites As with many other sol–gel systems, sodium silicate gels and aerogels can be readily combined with selected inorganic or organic materials to produce aerogel composites. This is a research topic of great popularity, as it holds the promise to overcome the mechanical limitations (friability, poor tensile strength, dustiness) of pure silica aerogels. For sodium silicate-based gels, formulations are generally limited to water-soluble guest compounds as shown in Table 16.2. Aside from mechanical reinforcement, the addition of new specific chemical function is also desirable, often allowing for synergistic enhancement of some of silica aerogels outstanding thermal properties or introduction of further functional properties for use in sensing, detection, catalysis, and sorption applications.

16

Sodium Silicate-Based Aerogels by Ambient Pressure Drying

405

Table 16.2 Summary of ambient pressure dried waterglass-based composites Components IE WGa

Applications Thermal insulation

Ref. Kim (2003) [64]

Structure-directing

Naoki(2007) [65]

IE WG IE WG IE WG NIE WG IE WG

Fe3O4 MWCNT TiOCl2 Fe3O4 NP

Paramagnetic Reinforcing Photocatalytic Magnetic

Catalysis, separation, microelectronics, drug delivery Catalysis, separation, microelectronics, drug delivery Magnetic Catalyst support, thermal insulation VOC degradation Chemical and biochemical catalysis

IE WG

CuO

Catalytic

Hydrogen production

NIE WG NIE WG IE WG

Fly ash Glass fiber PET nanofiber

Reinforcing Reinforcing Reinforcing

Thermal insulation Fire resistance Thermal and acoustic insulation

IE WG

MTES, silica fiber Mullite fibers Graphene oxide Fly ash Polyurethane

Reinforcing

Thermal insulation

Lee (2012, 2013) [13, 14] Wang (2012) [15] Kim (2013) [16] Amirkhani(2015, 2016) [18] Amiri (2015, 2016) [20] [67] Li (2015) [68] Motahari (2015) [21] Mazrouei-Sebdani (2015) [69] Shao (2015) [70]

Reinforcing Reinforcing/ pore structure improving Reinforcing Reinforcing

Thermal insulation Sensors, adsorbents, insulation

Liu (2016) [71] Dervin (2017) [72]

Thermal insulation Thermal insulation

Huang (2018) [22] Nazeran (2017) [73]

TiO2 Expanded perlite

Opacifier Reinforcing

Thermal insulation Thermal insulation

Vedenin (2018) [74] Wang (2018) [75]

NIE WGa

IE WG IE WG NIE WG Aerogel granule IE WG NIE WG a

Functions Reinforcing

Polyvinyl butyral Poly(acrylic acid) Gelatin

Structure-directing

Setyawan (2012) [66]

NIE WG and IE WG stand for the non-ion exchanged and ion exchanged waterglass, respectively

16.3

Materials Properties

In the following, we are going to discuss selected materials properties of APD sodium silicate-based aerogels as published in the literature. To do so we have analyzed the key works (91 publications) starting from the introduction of APD method by Schwertfeger in 1998 until today. Today, the APD process yields aerogels that offer properties virtually identical to supercritically dried ones. This is well reflected in the reported density, surface area, and thermal conductivity histograms as shown in Fig. 16.11: published densities range from 0.05 to 0.25 g cm3 and surface areas from 300 to 800 m2 g1. Again, these values compare rather well with alkoxide-based APD aerogels but also with supercritically dried waterglass and alkoxide-based silica aerogels. The corresponding histograms of APD waterglass aerogels for density and surface area are given in Fig. 16.11a, b, respectively. Surprisingly, the thermal conductivity data (Fig. 16.11c) from these collected studies is rather confusing, yet even misleading. First of all, the spread of published values is rather large but more strikingly, the published numbers are systematically much higher than one would expect. This can be attributed to three main reasons –

listed here in order of relevance as perceived by the authors – namely i) poor accuracy of thermal conductivity measurement values and use of unsuitable measurement methods for thermal characterization of aerogel specimens, particularly in the form of complex shapes (granules, powder) obtained by the APD process, ii) general lack of thermal characterization equipment resulting in a comparably low fraction of specimens being characterized in terms of thermal conductivity compared to density and nitrogen sorption which also may lead to a statistical misrepresentation of high density/high thermal conductivity samples (which are easier to characterize), and iii) a large spread of materials prepared, some of the published samples exhibiting higher density and thus also higher thermal conductivity values. Although the different experimental parameters tested during process and method development activities clearly lead to dissimilar materials properties and performance, the fact that a majority of specimens are in the density range between 0.075 and 0.175 and published surface area values are also in the right range, one would indeed expect a majority population also to have thermal conductivities in the 20 mW/(m k) range, as there is a direct correlation between density/mesoporosity and thermal conductivity of silica aerogel materials. Hence the generally much too high reported thermal conductivity values

16

406

A. V. Rao et al.

# Samples in bin

b

# Samples in bin

c

30 20 10 0

18 16

0

0.05 0.1 0.15 0.2 0.25 0.3 0.35 Density (g/cm3)

40 20 0

14 12 10 8 6 4 2

0

200

400 600 SBET (m2/g)

800

1000

15 10 5 0

Samples in bin (%)

# Samples in bin

a

0 0

5 10 15 SiO2 concentration (wt.%)

20

Fig. 16.12 Histogram of the SiO2 concentration used in the literature for sodium silicate-based aerogel synthesis

0

50 100 150 Thermal conductivity (mW/(m.K))

Fig. 16.11 Histograms of (a) envelope density, (b) specific surface area, and (c) thermal conductivity for sodium silica aerogels [1, 6, 11–23, 28, 30–32, 34, 36–39, 45, 54, 65, 66, 68–72, 75–158]

presented in the histogram in Fig. 16.11 are likely to originate from difficulties and lack of experience related to the thermal characterization. In the following let us further analyze the influence of different synthetic parameters on the final materials properties.

16.3.1 Effect of the Silica Concentration in the Sol In the literature, we find different definitions for the silica precursor content used in silica gel preparations, for example, the sodium silicate concentration [87], the SiO2 concentration [6], the sodium silicate to H2O molar ratio [159], or the specific density or specific gravity of sodium silicate solution used [28]. For a direct comparison with alkoxidebased aerogel works, the SiO2 concentration is probably the most relevant definition for total silicate loading in the sol. To facilitate the following discussion, silica contents from different studies have been converted to SiO2 concentration, with a histogram of typical sol concentrations given in Fig. 16.12. Similar to the alkoxide-based aerogel [160], the silica concentration in the sol is the most central parameter for the synthesis of silica aerogels. It determines the amount of silica per added sol volume and hence also the “packing density” of silica units in the gel network. So a variation of the precursor concentration under optimized

processing conditions will lead to a range of materials with controllable density. Note that the processing conditions themselves can also change the properties and be valid only for a certain range of gel densities and at lower gel density values cause the collapse of the gel structure. As mentioned, key process parameters are, for example, gelation pH, aging procedure [161], the final surface coverage of the gel surface with TMS groups after the hydrophobization treatment [162] as well as residual sodium ions present in the silicate network structure [30]. Despite the dependence of the final materials properties on the entire preparation history, the following discussion is a simplified one in terms of the single variable “SiO2 concentration in the starting sol.” An early study by Shewale et al. [159] varied the silica mass concentration from 1 to 4 wt.% (H2O/Na2SiO3 molar ratio from 83.3 to 333.3). Acetic acid was used for neutralization, and the amount was proportionally adjusted to the silicate concentration. Gels were then aged for 2 h at 50  C in water, followed by an ion removal step by water vapor extraction. Gels were then solvent exchanged into MeOH, and hydrophobized using a MeOH:TMCS:n-hexane modification solution (volume ratio 1:1:1) for 24 h at 50  C. Hydrophobized gels were then dried under ambient pressure for (24 h, 1 h, and 1 h at room temperature, 50  C, and 200  C, respectively). It is observed that with decreasing silica concentration in the sol, the gelation time tg increases from 1 min to 6 h. This is due to the fact that with increasing dilution (H2O/SiO2 molar ratio), the colloidal particle collision frequency decreases (the mean free path of colloids increases) and thus the rate of formation of the threedimensional “aggregated particle” network – the gel. Correspondingly, the bulk density of the final aerogel generally

16

Sodium Silicate-Based Aerogels by Ambient Pressure Drying

increases with the waterglass sol concentration. At very low silica concentrations, however, this trend is broken because in these highly dilute systems, the gel is so fragile that it cannot withstand capillary pressure and collapses during ambient pressure drying. Both the dependence of the final APD aerogel density on the silica concentration in the sol and the need for a minimal density and strength of the gel needed to withstand APD stresses are also conceptually valid for other gel or aerogel systems.

16.3.2 Effect of Sol pH As explained earlier, gelation in waterglass-based gel systems is driven by condensation reactions and hence gelation kinetics depend on the pH of the hydrosol. In silicate systems, condensation rates are known to be fastest at moderate pH values. The addition of an acid to a sodium silicate or a base to a silicic acid solution leads to neutralization which then explains the acceleration in condensation and network formation. Rao et al. [90] studied the influence of the pH of the silica hydrosol on gelation time, porosity density, and optical transparency of the final product as shown in Table 16.3. The synthesis protocol started from a sodium silicate solution which was diluted to 1 wt.% SiO2 (H2O/SiO2 molar ratio 370 [90]) and ion exchanged to form a silicic acid solution with a pH value around 2. The pH of the sol was then adjusted to values from 3 to 8 by addition of 1 M NaOH solution. Wet gels were aged and processed by a consecutive washing, exchange (methanol), and hydrophobization (TMCS/n-hexane) process and dried at ambient pressure. The gelation time tg decreased with increasing pH within the studied range. This is consistent with the highest condensation kinetics in silica hydrosols typically occurring at a pH value around 9. At a pH of 4 or lower, the colloidal silica building blocks still carry significant positive surface charges [163, 164], resulting in columbic interparticle repulsion. Two other factors that further hinder gelation at low pH are i) the strong competing effect of H+ as a catalyst, which shifts the equilibrium from the condensation reaction toward the side of

Table 16.3 Effect of the gelation pH on the gelation time tg and aerogel properties pH 3 4 5 6 7 8

tg (min) 32 12 5 2.5 1.6 1.2

Density (g/cm3) 0.25 0.15 0.1 0.05 0.06 0.065

Porosity (%) 82 92 95 97.5 97 97

Optical Transparent Transparent Less transparent Semitransparent Opaque Opaque

407

hydrolysis of siloxane bonds and ii) the neutralization of deprotonated silanols (Si–O), which is a much better nucleophile than free silanols. We also note that at low pH, a change in pH therefore has the largest effect on the gelation time. Under identical processing (solvent exchange, hydrophobization, APD) conditions, it was also found that density decreases and porosity increases with increasing pH. Best performance of the aerogels was found at pH values from 5 to 8, with porosities above 95% and envelope densities of 0.1 g/cm3 or lower. The transparency follows the density trend and can be readily explained by a pore size argument (higher density materials feature smaller mean pore sizes which also means that the fraction of scattering centers with feature sizes on the order of half the wavelength of visible light decreases substantially). These findings are consistent with the classic formation model for silica gel originally formulated by Scherer and Brinker [165]: the gelation of acidic silica hydrosols is dominated by hydrolysis (the rate of hydrolysis kH is much larger than the rate of condensation kC) and the gelation mechanism is understood to proceed through aggregation of charged colloids (Cluster-Cluster model). Corresponding gels are weak and show a more porous primary particle structure with very high -OH functional group coverage (Fig. 16.13b). When the gelation occurs in basic media, the condensation rate kC is much higher than the hydrolysis rate and dominates the process. In this case, gelation is believed to proceed through the condensation of small molecular oligomers onto existing clusters – forming some sort of a glue, which binds cluster together. This mechanism (Monomer-Cluster model) leads to a much coarser, more aggregated primary particle structure. Such gels form more quickly and they are much stronger than acid-derived analogues, but feature also lower porosity as well as a lower -OH surface coverage. Most importantly, the microstructural gelation model helps to explain the results summarized in Table 16.3: Low pH hydrogels formed through the Cluster-Cluster aggregation method are initially more porous but also much weaker mechanically. Under identical aging conditions, the low-pH gels are therefore significantly weaker before being solvent exchanged and hydrophobized. It is primarily the difference in gel mechanics that explains why the low pH aerogels end up with high density and lower porosity, as they cannot withstand the stresses and collapse significantly during APD. To reach the general conclusion that low-pH gels cannot yield satisfactory APD aerogels with low densities would be incorrect. However, for low-pH hydrogels to be converted into good quality aerogels, the aging and hydrophobization treatments must be addressed separately.

16

408

A. V. Rao et al.

a Monomer-cluster model - base catalysis

kC >kH

Dense collidal particles

b Cluster-cluster model - acid catalysis

kH >kC

Polymer - like, wealy cross-linked network

Fig. 16.13 Structural model of silica-based aerogels synthesized with different catalytic conditions [165]. (Copyright 1990, Elsevier B.V)

16.3.3 Effect of Aging (ta) Period Gel aging is of particular relevance for waterglass APD materials as discussed above. Most typical aging processes involve keeping the gel in its mother liquor for some period of time, preferably at intermediate temperatures in the 50  C to 70  C range. Alternative approaches involve aging in dilute silane monomer solutions (TEOS/TMOS) [161]; however, the substantial amount of newly added silicate leads to a significant increase in the gel and aerogel (typically >0.2 g/cm3) density. The main function of the aging treatment is to strengthen interparticle necks and thereby reduces the risk of structural collapse during the subsequent processing steps, particularly APD [81, 166]. Sarawade et al. [81] studied the aging influence on the silica hydrogels synthesized from an 8%w sodium silicate solution which was ion exchanged and then gelled by addition of 1 M NH4OH. Fresh gels were aged for various periods of time from 6 to 24 h in 6 h intervals at four different temperatures (25  C, 40  C, 60  C, 80  C). It is observed that with an increase in the aging period, the density of aerogel decreases first up to 16 h aging and then increases again toward 24 h. This trend is more pronounced at higher temperature which may suggest that evaporation of water may have led to a partial drying and densification (particularly 80  C aged samples). After combined solvent exchange

and hydrophobization CSH with MeOH/n-hexane/TMCS, the gels were predried at room temperature and then dried for 2 h and 1 h, respectively, at 80  C and 200  C, respectively. The densities of all aerogel samples made in this way were in the range of 0.1 to 0.07 g/cm3. Mechanistically, polycondensation of silica species still continues after the gelation with the speciation population being dominated by Q2 at the gelation point, Qn denoting the individual siloxane bonding order of any given silicon atom with n bridging Si–O–Si bonds. 29Si NMR is a well-established tool to study gel aging and can be used to follow the increase in average bonding order, commensurate with an increase of Q3 and Q4 speciation with aging time. In analogy to crosslinking reactions in polymer chemistry, an increase in Qn speciation increases both strength and stiffness of the gel network. Aging under pH neutral to slightly basic conditions also causes a coarsening of the gel network leading to the growth of thicker inter-particle necks due to an Ostwald ripening effect. The observed APD densities are the result of an interplay between i) strengthening of the gel network which prevents structural collapse during APD – and thus maintains the low densities defined by the silica concentration in the hydrosol – and ii) a partial contraction (shrinkage) during the drying itself, due to increasing connectivity at the molecular level (average Qn value) [167]. The latter then explains the higher shrinkage and hence higher aerogel density, which are accompanied by longer aging [30, 159]. Notably, the exact aging conditions must be optimized for each particular system and also depend strongly on the sol/aging liquid pH, solvent system, etc.

16.3.4 Effect of Solvent Exchange As hydrophobization agents (TMCS, HMDSO, HMDZ) are insoluble in water the pore fluid of a hydrogel needs to be exchanged into an organic solvent system before APD as discussed and illustrated in Sect. 16.2. When switching from a protic to an aprotic (hydrocarbon) medium, alcohols are ideal intermediary exchange solvents, as their bifunctional nature (polar hydroxyl group/nonpolar hydrocarbon chain) promotes miscibility of water and organic phase. What seems at first surprising is the fact that the choice of the alcohol exchange solvent in a two-step, non-CSH solvent exchange has a tremendous effect on the pore structure and also on the final materials properties: Rao et al. [168] investigated the influence of different low-molecular alcohols during solvent exchange on the silicic acid (ion exchanged WG) derived aerogels. Aged hydrogels were first exchanged with water, then with a given intermediary alcohol solvent, and finally modified with a mixture of

16

Sodium Silicate-Based Aerogels by Ambient Pressure Drying

409

Table 16.4 Effect of the type of phase transfer catalyst cosolvent used to exchange the pore fluid from aqueous to organic medium in CSH model studies Solvent/formula Methanol/CH3OH Ethanol/C2H5OH n-Propanol/C3H7OH Isopropanol/(CH3)2CHOH n-Butanol/C4H9OH Isobutanol/(CH3)2CH2CH2OH n-Hexanol/C6H13OH

Volume shrinkage (%) 37 55 57.5 30.7 62.5 61 65

Density (g/cm3) 0.10 0.155 0.16 0.07 0.17 0.165 0.19

alcohol, TMCS, and hexane. The results are summarized in Table 16.4. The data suggests that the density and volumetric shrinkage of APD aerogels are influenced by the types of alcohols as solvent or cosolvent used in a first solvent exchange step. Note that in this study, one is looking at the combined effect of the alcohol as both first exchange solvent and APD drying cosolvent. Isopropanol and methanol seem to produce the best aerogels overall based on density. The chain length of the hydrocarbon end influences both vapor pressure and the surface tension and wetting behavior of the solvent/gel interface which is particularly relevant during APD. Generally, low vapor pressure and low surface tension solvents decrease the surface tension in the gel and produce less shrinkage. With increasing chain length of the alcohol, the hydrophobic part becomes dominant over the hydrophilic one, which reduces the exchange efficiency with water. In conclusion, lower vapor pressure solvents and compact hydrocarbon groups (-CH3, -CH(CH3)2) are better suited for APD drying as they allow for a more efficient removal of the pore water from the hydrogel. Long chain alcohols are non-ideal drying cosolvents because of their poor performance in extracting pore water coupled with significantly higher boiling point. In CSH, methanol, ethanol, and isopropanol are the most commonly used cosolvent systems which are in full agreement with these findings.

16.3.5 Silylating Agents and the Hydrophobization Treatment Duration Regardless of their sol precursor, the preparation history or the type of hydrophobe used, a successful surface modification with an excess of hydrophobization agent, and optimized synthesis process will always display remarkably similar physical properties and near-identical TMS contents in the 22–27 wt.% range [59]. From an industrial manufacturing standpoint, it is essential to know the minimum amount of silylating agent needed per equivalent of SiO2 for each preparation method.

a

Porosity (%) 94.9 92.1 91.8 96.6 91.3 91.5 90.0

b

Mean pore diameter (nm)/FWHM (nm) 11.5/4.5 10/7.0 12/8.0 14/5.0 7.5/5.0 15/7.0 –

c

Fig. 16.14 Image of hydrogels during the initial methanol exchange (a) and after a longer immersion period in the heptane/HMDZ hydrophobization mixture (b). A contact angle above 140 indicates a hydrophobic surface of the APD aerogel (c) [85]. (Modified from Copyright 2008, IOPScience)

To study the effect of the concentration and exposure time of the silylating agent on the physical properties of aerogels, an early model study employing the HMDZ/n-hexane with a consecutive solvent exchange was reported by Shewale et al. [169]: sodium silicate hydrogels were obtained by neutralization with acetic acid and Na+ ions removed by water washing. Aged gels were first exchanged into methanol, then into n-hexane, and finally into a hydrophobization mixture consisting of HMDZ/n-hexane. During the initial solvent exchange in the water/methanol solvent system, the gels sank to the bottom of the beaker (Fig. 16.14, left) [85]. Yet, after being left immersed in the hydrophobization mixture, the modified gel began to float atop the organic phase (Fig. 16.14, middle). Note that during CSH in a two-phase aqueous/organic solvent mixture, a transfer of the hydrophobized from the bottom aqueous to the top organic phase is always observed. In the case of the multistep solvent exchange example [29, 35, 170], a variation of the HMDZ/SiO2 molar ratio from 1.8 to 6.9 shows a noticeable effect in the APD aerogel properties: from 1.8 to 3.5, the final aerogel densities are in the desired range and show a systematic decrease from 0.117 down to 0.095 g/cm3 with increasing molar ratio and the optical transparency from 70% to 60% as shown in Fig. 16.15. The systematic decrease observed in this range is attributed to a more complete hydrophobization at higher HMDZ/SiO2 molar ratios, resulting in higher TMS surface

16

410

A. V. Rao et al.

0.21

Density (g/cc)

0.19

80

b 0.125

70

0.12

60

0.17

50

0.15

40

0.13

30

0.11 0.09

20

0.07

10

0.05

0 1

2

5 3 4 6 7 HMDS / Na2SiO3 molar ratio

8

75 Bulk density (g/cc) % Of optical transmission (visual)

0.115

70

0.11 65

0.105 0.1

60

0.095 0.09

55

0.085 0.08 0

5

% Of optical transmission (visual)

Bulk density (g/cc) % Of optical transmission

Bulk density (g/cc)

0.23

Optical transmission (%T700 nm)

a

50 10 15 20 25 30 35 40 Silylation period (hours)

Fig. 16.15 Effect of (a) HMDZ/SiO2 molar ratio and (b) silylation period on bulk density and optical transmission of aerogels [85]. (Copyright 2008, IOPScience)

coverage and thus better protection of the silica surfaces from direct silanol contacting which minimizes volume shrinkage during drying. For HMDZ contents above five equivalents, the aerogel properties rapidly deteriorate, yielding a density >0.2 g/cm3 and transparency below 20% for a molar ratio of 7. This trend can be explained by the solvent systems phase diagram which suggests phase separation into micelle-like HMDZ-rich subphases. This effectively lowers the surface chemical compatibility of the hydrophobe phase with the native silica surface and decreases in the surface modification efficiency. Aside from the hydrophobe amount, the duration of the silylation treatment also plays a significant role in the surface modification treatment. In the same study, the HMDZ/ SiO2 molar ratio was kept fixed at 3.6, and the hydrophobization time varied from 6 to 36 h in 6 h intervals while keeping all other synthesis parameters the same. From 6 to 30 h, the bulk density of the resulting APD aerogels declined monotonically from 0.12 g/cm3 to a minimum value of about 0.085 g/cm3 and then increased again further toward 36 h. In other words, the product of the diffusion of the hydrophobe mixture into the most inner pores of the gel body combined and the kinetics of the surface modification reaction is quite slow. Other silylation agents such as TMCS are more reactive and are expected to react more quickly [171]. Furthermore, other parameters are known to accelerate the hydrophobization reaction speed, namely a reduction in gel body size (which addresses the diffusion component), an increase in temperature (which addresses both diffusion and chemical reaction kinetics) as well as the choice of the hydrophobe system itself (which primarily influences the chemical reaction kinetics). In the literature,

HMDZ and TMCS are the most commonly used hydrophobes used for the preparation of silica aerogels as shown in Fig. 16.8a. They react reasonably fast with surface silanol groups and can be easily handled at laboratory scale to yield good performance materials. When targeting economically scalable process technologies geared toward industrialization [56, 63, 172], alternative silane and hydrophobization chemistries are growing fields of development efforts worldwide.

16.3.6 General Comments About Parameter Optimizations So far, we have analyzed a number of parameters of the whole fabrication process of aerogels from silicates without going into too much detail. The preparation of sodium silicate aerogels by APD contains a total of 2–9 main steps (depending on whether or how ion removed, solvent exchanged, and hydrophobization or CSH are used), each of which comprises at least one main process variable. Hence, it is extremely challenging to optimize all steps involved in the preparation, or in other words, the devil is in the details. Another fact often neglected is that the ideal parameter combination for one particular step, such as for example the hydrophobization, may not be generally valid and depends on the previous history, that is, on the choice of preparation steps, methods, and the set of parameters used before. Just like in human psychology, everything is connected to everything else. Furthermore, industrial scaleup of potential attractive routes cannot always be viewed as a simple linear combination of its individual steps. Particularly

16

Sodium Silicate-Based Aerogels by Ambient Pressure Drying

the mass balance and the enrichment of different components in the various process liquid streams is to be well understood when attempting the design of feasible production schemes. Yet, the rewards can be promising: when done properly, scale-up studies coupled with systematic, fundamental research efforts have the potential to coin next generation aerogels with low thermal conductivity, high optical transparency, improved mechanical properties, and reduced dustiness accessible to the masses. This is what we are all waiting for, so please innovate!

16.4

Applications and Commercialization

When it comes to technical applications of aerogel materials, silica has long been the material of choice. Monsanto set up the first commercial production of Kistler’s original waterglassbased aerogels in the form of granulate and powders under the trade name Santocel in Massachusetts during WWII. This first generation of products quickly found its way into the classical insulation application, primarily for cryogenic transport and storage vessels. Other applications used aerogel powder as an additive and rheology modifier in coatings, medical, and defense applications. Santocel was also used in silicone rubber formulations and in neat form as a “physically acting” insecticide – being a strong desiccant, it essentially disrupted the water balance in insects upon contact. But what happened after Santocel was discontinued in the late 1960s? Clearly, around that time interest in aerogels faded due to lacking competitiveness as alternative insulation materials and micro/mesoporous pyrogenic and precipitated silicas became commercially available at much lower cost. Though tremendous academic research efforts and discoveries were made in silica aerogels during the 1980s and 1990s, it took until the early 2000s for Cabot and Aspen Aerogels to decide to attempt commercialization of silica aerogel-based materials and products for the second time. Cabot is the only large-scale manufacturer of aerogel granulates that is using a sodium silicate-based APD process. The technology was acquired from the Hoechst team in the late 1990s with its pioneering works largely led by Schwertfeger. Cabot’s product is mostly larger granulate with smaller particle fractions (powders) available as a second tier product during the production. Newcomer companies such as Svenska and Jios aerogel have been focusing on smaller particles and powders exclusively. All three companies are using non-ion exchanged waterglass chemistry but a different hydrophobization and drying protocols. For a more detailed discussion on commercial manufacturers and products, the reader is referred to the industrial applications chapters (▶ Chap. 64). Despite the inherent advantage of sodium silicate-based aerogels in granulate and powder forms linked to the small

411

particle size and associated rapid exchange and process times, these materials are to be thought of as semifinished products or even raw materials and require substantial product development and system integration efforts to become usable for end customers. Over the past decade a number of product developments have been made in collaboration with abovementioned aerogel manufacturers and system integrators. A few selected examples of such products include: • Daylighting applications in both window and GFRP (glass fiber reinforced polymer) structural cavities filled with aerogel granulate • Interior insulation/finishing ETICS system “StoTherm In Aevero” developed, produced, and marketed by Sto AG • Various aerogel particle-based blanket products (e.g., Cabot “Aeroclad,” Jios/Armacell “Armagel”) with application focusing on high-temperature insulation • Aerogel-based insulating plasters and other paste-like minerally bonded wet applied mortar products such as the “Fixit 222” insulation render developed by Fixit AG and Empa (Swiss Federal Laboratories for Material Science and Technology) Given the general potential of sodium silicate aerogels as a low-cost raw materials source, the development and commercialization of additional products in different markets is essential to the survival of this fascinating class of materials. Today’s optimized manufacturing technology could allow a significant reduction in price levels from roughly 2000–2500 $/m3 (at a target density of 100 kg/m3) to less than half through improvements and economy of scale, however the main factor holding back commercial development is the lack of good products and the still prohibitively high cost for market adoption in the building and construction sector. In 2018, particle-based sodium silicate powder and granulate markets account for a total annual market volume between 25 M and 30 M US$. A recent market study predicts a rapid drop of aerogel powder and granulate prices beyond the year 2021 to less than half of today’s levels [173]. If this prediction ends up coming true, this would open doors for many new applications of silica aerogel materials, components and systems in an extended range of industries.

16.5

Summary

Sodium silicate aerogels are approaching their first century anniversary and are now finally quite well established, both academically and industrially. This is due to a large extent to the technical revival in the 1990s which largely coined modern production processes. From an academic perspective, the

16

412

different process steps for making sodium silicate aerogels in all its colorful variations are better understood than any other aerogel synthesis procedure. Furthermore, they offer a tremendous market potential as a result of their straightforward and reliable ambient pressure drying protocols. In terms of academic progress, the room for improvement and innovation is fairly limited which is why many researchers are now moving toward hybrid gel systems (e.g., with biopolymers, see also ▶ Chap. 25). Today’s commercial products have now been on the market for over 10 years and successfully applied to various different use cases. In the coming years, granulate and powder aerogels will still remain an important product morphologies, given the cheap and abundant raw material, fast and continuous production options, and resulting low production cost. However, SiO2 aerogel granulates and powders are still semifinished products. In that sense, large-scale implementation primarily requires more product development efforts. Since the first APD aerogel developed by Deshpande and Schwertfeger in the mid-1990s, there were no real breakthroughs in the field anymore. Although the current adaptation of CSH-APD processes is faster and less costly, it is offset however by a partial loss of thermal and optical transparency performance. Furthermore, new competition is arising from recent progress in the fields of polymethylsiloxane [172] (see also ▶ Chap. 15) and polymer [174, 175] aerogels. In conclusion, the proliferation of sodium silicate-based aerogel materials and technologies in real-life applications will depend on the development of derived products as well as competing technologies.

References 1. Kistler, S.S.: Coherent expanded-aerogels. J. Phys. Chem. 36(1), 52–64 (1932) 2. Nicolaon, G., Teichner, S.: On a new process of preparation of silica xerogels and aerogels and their textural properties. Bull. Soc. Chim. Fr. 5, 1900 (1968) 3. Smith, D.M., Deshpande, R., Brinke, C.J.: Preparation of low-density aerogels at ambient pressure. MRS Online Proc. Libr. Arch. 271, 567–572 (1992) 4. Prakash, S.S., Brinker, C.J., Hurd, A.J., Rao, S.M.: Silica aerogel films prepared at ambient pressure by using surface derivatization to induce reversible drying shrinkage. Nature. 374(6521), 439– 443 (1995) 5. Deshpande, R., Smith, D.M., Brinker, C.J.: Preparation of high porosity xerogels by chemical surface modification. US Patent 5,565,142, 1996 6. Schwertfeger, F., Frank, D., Schmidt, M.: Hydrophobic waterglass based aerogels without solvent exchange or supercritical drying. J. Non-Cryst. Solids. 225, 24–29 (1998) 7. Schwertfeger, F.: Process for producing organically modified aerogel. Patent WO1998005591 A1, 1998, Patents

A. V. Rao et al. 8. Deshpande, R., Smith, D.M., Brinker, C.J.: Preparation of high porosity xerogels by chemical surface modification. Google Patents, 1996 9. Joung, Y.C., Roe, M.J., Yoo, Y.J., Park, J.C., Choi, H.J., Kim, M. W.: Method of preparing silica aerogel powder. US Patent 8,961,919, 2015 10. Lee, K.J., Kim, Y.H., Lee, J.K., Hwang, H.J.: Fast synthesis of spherical silica aerogel powders by emulsion polymerization from water glass. ChemistrySelect. 3(4), 1257–1261 (2018) 11. Júlio, M.D.F., Ilharco, L.M.: Superhydrophobic hybrid aerogel powders from waterglass with distinctive applications. Microporous Mesoporous Mater. 199, 29–39 (2014) 12. Rida, M.A., Harb, F.: Synthesis and characterization of amorphous silica nanoparticles from aqueous silicates using cationic surfactants. J. Metals Mater. Miner. 24(1), 37–42 (2014) 13. Sung Lee, J., Hong, S.K., Hur, N.J., Seo, W.-S., Hwang, H.J.: Fabrication of spherical silica aerogel/magnetite nanocomposite particles. Mater. Lett. 112, 153–157 (2013) 14. Lee, J.S., Lee, E.J., Hwang, H.J.: Synthesis of Fe3O4-coated silica aerogel nanocomposites. Trans. Nonferrous Metals Soc. China. 22, s702–s706 (2012) 15. Wang, B., Song, K., Han, Y., Zhang, T.: Synthesis and characterization of multi-walled carbon nanotube doped silica aerogels. J. Wuhan Univ. Technol. Mater. Sci. Ed. 27(3), 512–515 (2012) 16. Kim, Y.N., Shao, G.N., Jeon, S.J., Imran, S.M., Sarawade, P.B., Kim, H.T.: Sol–gel synthesis of sodium silicate and titanium oxychloride based TiO2–SiO2 aerogels and their photocatalytic property under UV irradiation. Chem. Eng. J. 231, 502–511 (2013) 17. Schäfer, H., Milow, B., Ratke, L.: Synthesis of inorganic aerogels via rapid gelation using chloride precursors. RSC Adv. 3(35), 15263 (2013) 18. Amirkhani, L., Moghaddas, J., Jafarizadeh, H.: Effect of surface modification agent and calcination process on the preparation of the hydrophobic magnetic silica aerogels by ambient pressure drying method. Bulg. Chem. Commun. 47, 82–88 (2015) 19. Amirkhani, L., Moghaddas, J., Jafarizadeh-Malmiri, H.: Candida rugosa lipase immobilization on magnetic silica aerogel nanodispersion. RSC Adv. 6(15), 12676–12687 (2016) 20. Yousefi Amiri, T., Moghaddas, J.: Cogeled copper–silica aerogel as a catalyst in hydrogen production from methanol steam reforming. Int. J. Hydrog. Energy. 40(3), 1472–1480 (2015) 21. Motahari, S., Abolghasemi, A.: Silica aerogel–glass fiber composites as fire shield for steel frame structures. J. Mater. Civ. Eng. 27(10), 04015008 (2015) 22. Huang, Y., Gong, L., Pan, Y., Li, C., Zhou, T., Cheng, X.: Facile construction of the aerogel/geopolymer composite with ultra-low thermal conductivity and high mechanical performance. RSC Adv. 8(5), 2350–2356 (2018) 23. Kow, K.-W., Yusoff, R., Aziz, A.R.A., Abdullah, E.C.: From bamboo leaf to aerogel: preparation of water glass as a precursor. J. Non-Cryst. Solids. 386, 76–84 (2014) 24. Shi, F., Liu, J.-X., Song, K., Wang, Z.-Y.: Cost-effective synthesis of silica aerogels from fly ash via ambient pressure drying. J. Non-Cryst. Solids. 356(43), 2241–2246 25. Garcıa, N.J., Ingram, M.D., Bazán, J.C.: Ion transport in hydrated sodium silicates (water glasses) of varying water content. Solid State Ionics. 146(1–2), 113–122 (2002) 26. Garcia-Lodeiro, I., Palomo, A., Fernández-Jiménez, A.: Crucial insights on the mix design of alkali-activated cement-based binders. In: Handbook of Alkali-Activated Cements, Mortars and Concretes, pp. 49–73. Elsevier (2015) 27. Kistler, S.S.: Coherent expanded aerogels and jellies. Nature. 127, 741–741 (1931)

16

Sodium Silicate-Based Aerogels by Ambient Pressure Drying

28. Parvathy Rao, A., Rao, A.V., Pajonk, G.M., Shewale, P.M.: Effect of solvent exchanging process on the preparation of the hydrophobic silica aerogels by ambient pressure drying method using sodium silicate precursor. J. Mater. Sci. 42(20), 8418–8425 (2007) 29. Einarsrud, M.-A., Nilsen, E.: Strengthening of water glass and colloidal sol based silica gels by aging in TEOS. J. Non-Cryst. Solids. 226(1–2), 122–128 (1998) 30. Gurav, J.L., Rao, A.V., Rao, A.P., Nadargi, D.Y., Bhagat, S.D.: Physical properties of sodium silicate based silica aerogels prepared by single step sol–gel process dried at ambient pressure. J. Alloys Compd. 476(1–2), 397–402 (2009) 31. Bhagat, S.D., Park, K.-T., Kim, Y.-H., Kim, J.-S., Han, J.-H.: A continuous production process for silica aerogel powders based on sodium silicate by fluidized bed drying of wet-gel slurry. Solid State Sci. 10(9), 1113–1116 (2008) 32. Li, T., Wang, T.: Preparation of silica aerogel from rice hull ash by drying at atmospheric pressure. Mater. Chem. Phys. 112(2), 398– 401 (2008) 33. Shewale, P.M., Rao, A.V., Rao, A.P.: Effect of different trimethyl silylating agents on the hydrophobic and physical properties of silica aerogels. Appl. Surf. Sci. 254(21), 6902–6907 (2008) 34. Rajanna, S.K., Kumar, D., Vinjamur, M., Mukhopadhyay, M.: Silica aerogel microparticles from rice husk ash for drug delivery. Ind. Eng. Chem. Res. 54(3), 949–956 (2015) 35. Rao, A.P., Rao, A.V., Pajonk, G.M.: Hydrophobic and physical properties of the ambient pressure dried silica aerogels with sodium silicate precursor using various surface modification agents. Appl. Surf. Sci. 253(14), 6032–6040 (2007) 36. Sarawade, P.B., Quang, D.V., Hilonga, A., Jeon, S.J., Kim, H.T.: Synthesis and characterization of micrometer-sized silica aerogel nanoporous beads. Mater. Lett. 81, 37–40 (2012) 37. Bhagat, S.D., Kim, Y.-H., Suh, K.-H., Ahn, Y.-S., Yeo, J.-G., Han, J.-H.: Superhydrophobic silica aerogel powders with simultaneous surface modification, solvent exchange and sodium ion removal from hydrogels. Microporous Mesoporous Mater. 112(1–3), 504– 509 (2008) 38. Lee, C., Kim, G., Hyun, S.: Synthesis of silica aerogels from waterglass via new modified ambient drying. J. Mater. Sci. 37(11), 2237–2241 (2002) 39. Hwang, S.W., Kim, T.Y., Hyun, S.H.: Optimization of instantaneous solvent exchange/surface modification process for ambient synthesis of monolithic silica aerogels. J. Colloid Interface Sci. 322(1), 224–230 (2008) 40. Hwang, S.-W., Jung, H.-H., Hyun, S.-H., Ahn, Y.-S.: Effective preparation of crack-free silica aerogels via ambient drying. J. Sol-Gel Sci. Technol. 41(2), 139–146 (2007) 41. Venkateswara Rao, A., Nilsen, E., Einarsrud, M.A.: Effect of precursors, methylation agents and solvents on the physicochemical properties of silica aerogels prepared by atmospheric pressure drying method. J. Non-Cryst. Solids. 296(3), 165–171 (2001) 42. Michael, G., Kasack, V., Nowak, R.: Hydrophobic silica. US Patent 7,282,236, 2007 43. Pauly, T.R., Petkov, V., Liu, Y., Billinge, S.J., Pinnavaia, T.J.: Role of framework sodium versus local framework structure in determining the hydrothermal stability of MCM-41 mesostructures. J. Am. Chem. Soc. 124(1), 97–103 (2002) 44. Yang, L., Wang, Y., Luo, G., Dai, Y.: A new ‘pH-induced rapid colloid aggregation’ method to prepare micrometer-sized spheres of mesostructured silica in water-in-oil emulsion. Microporous Mesoporous Mater. 94(1–3), 269–276 (2006) 45. Kobayashi, M., Juillerat, F., Galletto, P., Bowen, P., Borkovec, M.: Aggregation and charging of colloidal silica particles: effect of particle size. Langmuir. 21(13), 5761–5769 (2005) 46. Provis, J.L., Duxson, P., Lukey, G.C., Separovic, F., Kriven, W.M., Van Deventer, J.S.: Modeling speciation in highly concentrated

413 alkaline silicate solutions. Ind. Eng. Chem. Res. 44(23), 8899– 8908 (2005) 47. Swaddle, T.W., Salerno, J., Tregloan, P.A.: Aqueous aluminates, silicates, and aluminosilicates. Chem. Soc. Rev. 23(5), 319– 325 (1994) 48. Belton, D.J., Deschaume, O., Perry, C.C.: An overview of the fundamentals of the chemistry of silica with relevance to biosilicification and technological advances. FEBS J. 279(10), 1710–1720 (2012) 49. Icopini, G.A., Brantley, S.L., Heaney, P.J.: Kinetics of silica oligomerization and nanocolloid formation as a function of pH and ionic strength at 25 C. Geochim. Cosmochim. Acta. 69(2), 293– 303 (2005) 50. West, J.K., Hench, L.L.: Molecular orbital models of silica rings and their vibrational spectra. J. Am. Ceram. Soc. 78(4), 1093– 1096 (1995) 51. Einarsrud, M.-A., Britt Kirkedelen, M., Nilsen, E., Mortensen, K., Samseth, J.: Structural development of silica gels aged in TEOS. J. Non-Cryst. Solids. 231(1–2), 10–16 (1998) 52. Hæreid, S., Anderson, J., Einarsrud, M., Hua, D., Smith, D.: Thermal and temporal aging of TMOS-based aerogel precursors in water. J. Non-Cryst. Solids. 185(3), 221–226 (1995) 53. Smitha, S., Shajesh, P., Aravind, P.R., Kumar, S.R., Pillai, P.K., Warrier, K.G.K.: Effect of aging time and concentration of aging solution on the porosity characteristics of subcritically dried silica aerogels. Microporous Mesoporous Mater. 91(1–3), 286– 292 (2006) 54. Hwang, S.-W., Kim, T.-Y., Hyun, S.-H.: Effect of surface modification conditions on the synthesis of mesoporous crack-free silica aerogel monoliths from waterglass via ambient-drying. Microporous Mesoporous Mater. 130(1–3), 295–302 (2010) 55. Shi, F., Wang, L., Liu, J.: Synthesis and characterization of silica aerogels by a novel fast ambient pressure drying process. Mater. Lett. 60(29–30), 3718–3722 (2006) 56. Jantke, D., Hindelang, K., Weidner, R.: Economic process for the manufacture of organically modified aerogels, or lyophilizates. Patent DE201510211812, 2016 57. Hindelang, K., Gottschalk-Gaudig, T., Jantke, D., Weidner, R.: A process for producing organically modified aerogels. Patent DE201510207939, 2016 58. Prakash, S.S., Brinker, J.C., Hurd, A.J., Rao, S.M.: Silica aerogel films prepared at ambient pressure by using surface derivatization to induce reversible drying shrinkage. Nature. 375, 431– 431 (1995) 59. Malfait, W.J., Zhao, S., Verel, R., Iswar, S., Rentsch, D., Fener, R., Zhang, Y., Milow, B., Koebel, M.M.: Surface chemistry of hydrophobic silica aerogels. Chem. Mater. 27(19), 6737–6745 (2015) 60. Adebajo, M.O., Frost, R.L., Kloprogge, J.T., Carmody, O., Kokot, S.: Porous materials for oil spill cleanup: a review of synthesis and absorbing properties. J. Porous. Mater. 10(3), 159– 170 (2003) 61. Sakthivel, T., Reid, D.L., Goldstein, I., Hench, L., Seal, S.: Hydrophobic high surface area zeolites derived from fly ash for oil spill remediation. Environ. Sci. Technol. 47(11), 5843–5850 (2013) 62. Wu, L., Huang, Y., Wang, Z., Liu, L., Xu, H.: Fabrication of hydrophobic alumina aerogel monoliths by surface modification and ambient pressure drying. Appl. Surf. Sci. 256(20), 5973– 5977 (2010) 63. Han, X., Hassan, K.T., Harvey, A., Kulijer, D., Oila, A., Hunt, M. R., Šiller, L.: Bioinspired synthesis of monolithic and layered aerogels. Adv. Mater. 30, 1706294 (2018) 64. Kim, G., Hyun, S.: Effect of mixing on thermal and mechanical properties of aerogel-PVB composites. J. Mater. Sci. 38(9), 1961– 1966 (2003)

16

414 65. Shinozaki, N., Takahashi, R., Sato, S., Sodesawa, T.: Strength and elasticity of bimodal porous silica prepared from water glass. J. Sol-Gel Sci. Technol. 43(3), 275–282 (2007) 66. Setyawan, H., Balgis, R.: Mesoporous silicas prepared from sodium silicate using gelatin templating. Asia Pac. J. Chem. Eng. 7(3), 448–454 (2012) 67. Amiri, T.Y., Moghaddas, J., Khajeh, S.R.: Silica aerogel-supported copper catalyst prepared via ambient pressure drying process. J. Sol-Gel Sci. Technol. 77(3), 627–635 (2016) 68. Li, Y., Cheng, X., Cao, W., Gong, L., Zhang, R., Zhang, H.: Fabrication of adiabatic foam at low temperature with sodium silicate as raw material. Mater. Des. 88, 1008–1014 (2015) 69. Mazrouei-Sebdani, Z., Khoddami, A., Hadadzadeh, H., Zarrebini, M.: Synthesis and performance evaluation of the aerogel-filled PET nanofiber assemblies prepared by electro-spinning. RSC Adv. 5(17), 12830–12842 (2015) 70. Shao, Z., He, X., Niu, Z., Huang, T., Cheng, X., Zhang, Y.: Ambient pressure dried shape-controllable sodium silicate based composite silica aerogel monoliths. Mater. Chem. Phys. 162, 346– 353 (2015) 71. Liu, G., Liu, Y.: Rapid synthesis of super insulation silica aerogel composites strengthened with mullite fibers. In: International Conference on Civil, Transportation and Environment (ICCTE 2016), pp. 1300–1304. Atlantis Press (2016) 72. Dervin, S., Lang, Y., Perova, T., Hinder, S.H., Pillai, S.C.: Graphene oxide reinforced high surface area silica aerogels. J. Non-Cryst. Solids. 465, 31–38 (2017) 73. Nazeran, N., Moghaddas, J.: Synthesis and characterization of silica aerogel reinforced rigid polyurethane foam for thermal insulation application. J. Non-Cryst. Solids. 461, 1–11 (2017) 74. Vedenin, A., Vityaz, P., Ivanova, I., Mazalov, Y.A., Pustovgar, A., Sudnik, L.: Experimental investigation of thermal insulating aerogel composites of hydrothermal reactor for biomass-to-hydrogen conversion. Int. J. Hydrog. Energy. 43(14), 6899–6903 (2018) 75. Wang, L., Li, Z., Jing, Q., Liu, P.: Synthesis of composite insulation materials – expanded perlite filled with silica aerogel. J. Porous. Mater. 25(2), 373–382 (2018) 76. Heinrich, T., Klett, U., Fricke, J.: Aerogels – nanoporous materials part I: sol-gel process and drying of gels. J. Porous. Mater. 1(1), 7– 17 (1995) 77. Yoldas, B., Annen, M., Bostaph, J.: Chemical engineering of aerogel morphology formed under nonsupercritical conditions for thermal insulation. Chem. Mater. 12(8), 2475–2484 (2000) 78. Knoblich, B., Gerber, T.: The arrangement of fractal clusters dependent on the pH value in silica gels from sodium silicate solutions. J. Non-Cryst. Solids. 296(1–2), 81–87 (2001) 79. Marzouk, S., Rachdi, F., Fourati, M., Bouaziz, J.: Synthesis and grafting of silica aerogels. Colloids Surf. A Physicochem. Eng. Asp. 234(1–3), 109–116 (2004) 80. Venkateswara Rao, A., Parvathy Rao, A., Kulkarni, M.M.: Influence of gel aging and Na2SiO3/H2O molar ratio on monolithicity and physical properties of water–glass-based aerogels dried at atmospheric pressure. J. Non-Cryst. Solids. 350, 224–229 (2004) 81. Sarawade, P.B., Kim, J.-K., Park, J.-K., Kim, H.-K.: Influence of solvent exchange on the physical properties of sodium silicate based aerogel prepared at ambient pressure. Aerosol Air Qual. Res. 6(1), 93–105 (2006) 82. Bhagat, S.D., Kim, Y.-H., Moon, M.-J., Ahn, Y.-S., Yeo, J.-G.: A cost-effective and fast synthesis of nanoporous SiO2 aerogel powders using water-glass via ambient pressure drying route. Solid State Sci. 9(7), 628–635 (2007) 83. Chun-Xi, Z.: Recent advances in waterglass sand technologies. China Foundry. 4, 13–17 (2007) 84. Shewale, P.M., Rao, A.V., Gurav, J.L., Rao, A.P.: Synthesis and characterization of low density and hydrophobic silica aerogels

A. V. Rao et al. dried at ambient pressure using sodium silicate precursor. J. Porous. Mater. 16(1), 101–108 (2007) 85. Bangi, U.K., Venkateswara Rao, A., Parvathy Rao, A.: A new route for preparation of sodium-silicate-based hydrophobic silica aerogels via ambient-pressure drying. Sci. Technol. Adv. Mater. 9(3), 035006 (2008) 86. Chandradass, J., Kang, S., Bae, D.-s.: Synthesis of silica aerogel blanket by ambient drying method using water glass based precursor and glass wool modified by alumina sol. J. Non-Cryst. Solids. 354(34), 4115–4119 (2008) 87. Kim, C.E., Yoon, J.S., Hwang, H.J.: Synthesis of nanoporous silica aerogel by ambient pressure drying. J. Sol-Gel Sci. Technol. 49(1), 47–52 (2008) 88. Kim, T., Hwang, S., Hyun, S.: Development of a continuous manufacturing process for silica sols via the ion-exchange of a waterglass. Ind. Eng. Chem. Res. 47(18), 6941–6948 (2008) 89. Parvathy Rao, A., Rao, A.V.: Microstructural and physical properties of the ambient pressure dried hydrophobic silica aerogels with various solvent mixtures. J. Non-Cryst. Solids. 354(1), 10– 18 (2008) 90. Rao, A.P., Rao, A.V., Bangi, U.K.H.: Low thermalconductive, transparent and hydrophobic ambient pressure dried silica aerogels with various preparation conditions using sodium silicate solutions. J. Sol-Gel Sci. Technol. 47(1), 85–94 (2008) 91. Aravind, P.R., Shajesh, P., Soraru, G.D., Warrier, K.G.K.: Ambient pressure drying: a successful approach for the preparation of silica and silica based mixed oxide aerogels. J. Sol-Gel Sci. Technol. 54(1), 105–117 (2010) 92. Gao, G.-M., Liu, D.-R., Zou, H.-F., Zou, L.-C., Gan, S.-C.: Preparation of silica aerogel from oil shale ash by fluidized bed drying. Powder Technol. 197(3), 283–287 (2010) 93. Rani, T.S., Subha, M.C.S., Venkata Reddy, G., Kim, Y.-H., Ahn, Y.-S.: Synthesis of water-glass-based silica aerogel powder via with and without squeezing of hydrogels. J. Appl. Polym. Sci. 115(3), 1675–1679 (2010) 94. Rao, A.V., Bangi, U.K.H., Kavale, M.S., Imai, H., Hirashima, H.: Reduction in the processing time of doped sodium silicate based ambient pressure dried aerogels using shaker. Microporous Mesoporous Mater. 134(1–3), 93–99 (2010) 95. Sarawade, P.B., Kim, J.-K., Hilonga, A., Kim, H.T.: Influence of aging conditions on textural properties of water-glass-based silica aerogels prepared at ambient pressure. Korean J. Chem. Eng. 27(4), 1301–1309 (2010) 96. Sarawade, P.B., Kim, J.-K., Hilonga, A., Kim, H.T.: Production of low-density sodium silicate-based hydrophobic silica aerogel beads by a novel fast gelation process and ambient pressure drying process. Solid State Sci. 12(5), 911–918 (2010) 97. Sinkó, K.: Influence of chemical conditions on the nanoporous structure of silicate aerogels. Materials. 3(1), 704–740 (2010) 98. Venkateswara Rao, A., Ganbavle, V.V., Bangi, U.K.H., Dhere, S. L.: Influence of preparation conditions on nanoporous structure and optical transmission of sodium silicate based ambient pressure dried aerogels employing shaking. J. Porous. Mater. 18(6), 751– 759 (2010) 99. Yuan, M.-R., Lu, J.-T., Kong, G.: Effect of SiO2:Na2O molar ratio of sodium silicate on the corrosion resistance of silicate conversion coatings. Surf. Coat. Technol. 204(8), 1229–1235 (2010) 100. Rao, A.V., Pajonk, G.M., Bangi, U.K.H., Rao, A.P., Koebel, M.M.: Sodium silicate based aerogels via ambient pressure drying. In: Aerogels Handbook, pp. 103–124. Springer (2011) 101. Sarawade, P.B., Kim, J.-K., Hilonga, A., Quang, D.V., Jeon, S.J., Kim, H.T.: Synthesis of sodium silicate-based hydrophilic silica aerogel beads with superior properties: effect of heat-treatment. J. Non-Cryst. Solids. 357(10), 2156–2162 (2011) 102. Sarawade, P.B., Kim, J.-K., Hilonga, A., Quang, D.V., Kim, H.T.: Synthesis of hydrophilic and hydrophobic xerogels with superior

16

Sodium Silicate-Based Aerogels by Ambient Pressure Drying

properties using sodium silicate. Microporous Mesoporous Mater. 139(1–3), 138–147 (2011) 103. Ding, Y., Yin, G., Liao, X., Huang, Z., Chen, X., Yao, Y.: Key role of sodium silicate modulus in synthesis of mesoporous silica SBA-15 rods with controllable lengths and diameters. Mater. Lett. 75, 45–47 (2012) 104. Essien, E.R., Olaniyi, O.A., Adams, L.A., Shaibu, R.O.: Sol-gelderived porous silica: economic synthesis and characterization. J. Miner. Mater. Charact. Eng. 11(10), 976 (2012) 105. Hong, S.K., Yoon, M.Y., Hwang, H.J.: Synthesis of spherical silica aerogel powder by emulsion polymerization technique. J. Ceram. Process. Res. 13, s145–s148 (2012) 106. Jung, I.-K., Gurav, J.L., Ha, T.-J., Choi, S.G., Baek, S., Park, H.-H.: The properties of silica aerogels hybridized with SiO2 nanoparticles by ambient pressure drying. Ceram. Int. 38, S105– S108 (2012) 107. Liu, G., Zhou, B., Ni, X., Shen, J., Wu, G., Du, A., Zu, G.: Influence of thermal process on microstructural and physical properties of ambient pressure dried hydrophobic silica aerogel monoliths. J. Sol-Gel Sci. Technol. 62(2), 126–133 (2012) 108. Pouretedal, H., Kazemi, M.: Characterization of modified silica aerogel using sodium silicate precursor and its application as adsorbent of Cu 2+, Cd 2+, and Pb 2+ ions. Int. J. Ind. Chem. 3(1), 20 (2012) 109. Tadjarodi, A., Haghverdi, M., Mohammadi, V.: Preparation and characterization of nano-porous silica aerogel from rice husk ash by drying at atmospheric pressure. Mater. Res. Bull. 47(9), 2584– 2589 (2012) 110. Yamagata, C., Elias, D.R., Paiva, M.R.S., Misso, A.M., Mello, S.R.H.: Influence of the precursor concentration on the characteristics of silica powder obtained from Na2SiO3 by a facile low temperature synthesis process. J. Mater. Sci. Eng. A. 2(8A), 429 (2012) 111. Dourbash, A., Motahari, S., Omranpour, H.: Effect of water content on properties of one-step catalyzed silica aerogels via ambient pressure drying. J. Non-Cryst. Solids. 405, 135–140 (2014) 112. Gao, T., Sandberg, L.I.C., Jelle, B.P.: Nano insulation materials: synthesis and life cycle assessment. Procedia CIRP. 15, 490–495 (2014) 113. He, P., Gao, X.-d., Li, X.-m., Jiang, Z.-w., Yang, Z.-h., Wang, C.-l., Gu, Z.-y.: Highly transparent silica aerogel thick films with hierarchical porosity from water glass via ambient pressure drying. Mater. Chem. Phys. 147(1–2), 65–74 (2014) 114. Nazriati, N., Setyawan, H., Affandi, S., Yuwana, M., Winardi, S.: Using bagasse ash as a silica source when preparing silica aerogels via ambient pressure drying. J. Non-Cryst. Solids. 400, 6– 11 (2014) 115. Rizkiyanto, D., Putri, W.E., Setyawan, H., Affandi, S.: Preparation of thermal insulator of silica from waterglass using electrophorectic deposition. AIP Conf. Proc. 1586, 101–103 (2014) 116. Shi, H.-X., Cui, J.-T., Shen, H.-M., Wu, H.-K.: Preparation of silica aerogel and its adsorption performance to organic molecule. Adv. Mater. Sci. Eng. 2014, 1–8 (2014) 117. Ui, S.-W., Choi, I.-S., Choi, S.-C.: Synthesis of high surface area mesoporous silica powder using anionic surfactant. ISRN Mater. Sci. 2014, 1–6 (2014) 118. Yun, S., Luo, H., Gao, Y.: Superhydrophobic silica aerogel microspheres from methyltrimethoxysilane: rapid synthesis via ambient pressure drying and excellent absorption properties. RSC Adv. 4(9), 4535–4542 (2014) 119. Cui, S., Yu, S.-w., Lin, B.-l., Shen, X.-d., Gu, D.: Preparation of SiO2 aerogel from rice husk ash. RSC Adv. 5(81), 65818– 65826 (2015) 120. He, S., Li, Z., Shi, X., Yang, H., Gong, L., Cheng, X.: Rapid synthesis of sodium silicate based hydrophobic silica aerogel

415 granules with large surface area. Adv. Powder Technol. 26(2), 537–541 (2015) 121. Laskowski, J., Milow, B., Ratke, L.: The effect of embedding highly insulating granular aerogel in cellulosic aerogel. J. Supercrit. Fluids. 106, 93–99 (2015) 122. Liu, S.-W., Wei, Q., Cui, S.-P., Nie, Z.-R., Du, M.-H., Li, Q.-Y.: Hydrophobic silica aerogel derived from wheat husk ash by ambient pressure drying. J. Sol-Gel Sci. Technol. 78(1), 60–67 (2015) 123. Mahesh, S., Joshi, S.C.: Thermal conductivity variations with composition of gelatin-silica aerogel-sodium dodecyl sulfate with functionalized multi-walled carbon nanotube doping in their composites. Int. J. Heat Mass Transf. 87, 606–615 (2015) 124. Minju, N., Abhilash, P., Nair, B.N., Mohamed, A.P., Ananthakumar, S.: Amine impregnated porous silica gel sorbents synthesized from water–glass precursors for CO2 capturing. Chem. Eng. J. 269, 335–342 (2015) 125. Mohammadi, A., Moghaddas, J.: Synthesis, adsorption and regeneration of nanoporous silica aerogel and silica aerogel-activated carbon composites. Chem. Eng. Res. Des. 94, 475–484 (2015) 126. Wa, L., Fengyun, L., Fanlu, Z., Mengjing, C., Qiang, C., Jue, H., Weijun, Z., Mingwei, M.: Preparation of silica aerogels using CTAB/SDS as template and their efficient adsorption. Appl. Surf. Sci. 353, 1031–1036 (2015) 127. Yu, H., Liang, X., Wang, J., Wang, M., Yang, S.: Preparation and characterization of hydrophobic silica aerogel sphere products by co-precursor method. Solid State Sci. 48, 155–162 (2015) 128. Zong, S., Wei, W., Jiang, Z., Yan, Z., Zhu, J., Xie, J.: Characterization and comparison of uniform hydrophilic/hydrophobic transparent silica aerogel beads: skeleton strength and surface modification. RSC Adv. 5(68), 55579–55587 (2015) 129. Cheng, Y., Xia, M., Luo, F., Li, N., Guo, C., Wei, C.: Effect of surface modification on physical properties of silica aerogels derived from fly ash acid sludge. Colloids Surf. A Physicochem. Eng. Asp. 490, 200–206 (2016) 130. Cui, S., Yu, S., Lin, B., Shen, X., Zhang, X., Gu, D.: Preparation of amine-modified SiO2 aerogel from rice husk ash for CO2 adsorption. J. Porous. Mater. 24(2), 455–461 (2016) 131. Du, M.-H., Wei, Q., Nie, Z.-R., Cui, S.-P., Liu, S.-W., Li, Q.-Y.: A rapid and low solvent/silylation agent-consumed synthesis, pore structure and property of silica aerogels from dislodged sludge. J. Sol-Gel Sci. Technol. 81(2), 427–435 (2016) 132. Firoozmandan, M., Moghaddas, J., Yasrebi, N.: Performance of water glass-based silica aerogel for adsorption of phenol from aqueous solution. J. Sol-Gel Sci. Technol. 79(1), 67–75 (2016) 133. García-Torres, B.A., Aguilar-Elguezabal, A., Román-Aguirre, M., Álvarez-Contreras, L.: Synthesis of silica aerogels microspheres prepared by ink jet printing and dried at ambient pressure without surface hydrophobization. Mater. Chem. Phys. 172, 32–38 (2016) 134. Maleki, H., Duraes, L., Garcia-Gonzalez, C.A., Del Gaudio, P., Portugal, A., Mahmoudi, M.: Synthesis and biomedical applications of aerogels: possibilities and challenges. Adv. Colloid Interf. Sci. 236, 1–27 (2016) 135. Mazrouei-Sebdani, Z., Khoddami, A., Hadadzadeh, H., Zarrebini, M., Karimi, A., Shams-Ghahfarokhi, F.: The effect of the nanostructured aerogel powder on the structural parameters, water repellency, and water vapor/air permeability of a fibrous polyester material. Mater. Chem. Phys. 177, 99–111 (2016) 136. Morsi, R.M.M., Basha, M.A.F., Morsi, M.M.: Synthesis and physical characterization of amorphous silicates in the system SiO2Na2O– RO (R¼Zn, Pb or Cd). J. Non-Cryst. Solids. 439, 57– 66 (2016) 137. Pisal, A.A., Rao, A.V.: Comparative studies on the physical properties of TEOS, TMOS and Na2SiO3 based silica aerogels by ambient pressure drying method. J. Porous. Mater. 23(6), 1547– 1556 (2016)

16

416 138. Sorour, M.H., Hani, H.A., Al-Bazedi, G.A., El-Rafei, A.M.: Hydrophobic silica aerogels for oil spills clean-up, synthesis, characterization and preliminary performance evaluation. J. Porous. Mater. 23(5), 1401–1409 (2016) 139. Yang, H., Xu, H., Frank, J.K., Xu, G., Huan, W., Ni, C., Yang, Y.: Influences of SiO2/Na2O molar ratio on aging and chemical modification of water glass. Open J. Inorg. Chem. 06(02), 125– 134 (2016) 140. Yücel, S., Karakuzu, B., Terzioğlu, P., Temel, T.M.: Influence of gel aging time on the properties of various silica aerogels synthesized from water glass. Mater. Sci. Forum. 866, 176–180 (2016) 141. Abdul Halim, Z.A., Yajid, M.A.M., Idris, M.H., Hamdan, H.: Effects of silica aerogel particle sizes on the thermal–mechanical properties of silica aerogel – unsaturated polyester composites. Plast. Rubber Compos. 46(4), 184–192 (2017) 142. Gu, B., Ko, D., Ahn, S., Choon Hyun, D., Lee, H.-K., Kim, J.: Synthesis of high purity silica from low cost water glass via sol–gel process and soxhlet extraction. J. Sol-Gel Sci. Technol. 82(3), 675– 681 (2017) 143. Karamahmut Mermer, N., Yilmaz, M.S., Ozdemir, O.D., Piskin, M. B.: The synthesis of silica-based aerogel from gold mine waste for thermal insulation. J. Therm. Anal. Calorim. 129(3), 1807– 1812 (2017) 144. Li, L., Wang, Z., Huang, J., Ji, S., Mei, Y., Wang, Y., Fang, Y.: Comparison of silica leaching behaviors from the acid-leached residue of catalytic gasification and combustion. Energy Fuel. 31(10), 10745–10751 (2017) 145. Mazrouei-Sebdani, Z., Javazmi, L., Khoddami, A., ShamsGhahfarokhi, F., Low, T.: Fabrication of a silica aerogel and examination of its hydrophobic properties via contact angle and 3M water repellency tests. IOP Conf. Ser.: Mater. Sci. Eng. 204, 012014 (2017) 146. Nah, H.-Y., Parale, V.G., Jung, H.-N.-R., Lee, K.-Y., Lim, C.-H., Ku, Y.S., Park, H.-H.: Role of oxalic acid in structural formation of sodium silicate-based silica aerogel by ambient pressure drying. J. Sol-Gel Sci. Technol. 85(2), 302–310 (2017) 147. Pan, Y., He, S., Cheng, X., Li, Z., Li, C., Huang, Y., Gong, L.: A fast synthesis of silica aerogel powders-based on water glass via ambient drying. J. Sol-Gel Sci. Technol. 82(2), 594–601 (2017) 148. Pan, Y., He, S., Gong, L., Cheng, X., Li, C., Li, Z., Liu, Z., Zhang, H.: Low thermal-conductivity and high thermal stable silica aerogel based on MTMS/water-glass co-precursor prepared by freeze drying. Mater. Des. 113, 246–253 (2017) 149. Temel, T.M., İkizler, B.K., Terzioğlu, P., Yücel, S., Elalmış, Y.B.: The effect of process variables on the properties of nanoporous silica aerogels: an approach to prepare silica aerogels from biosilica. J. Sol-Gel Sci. Technol. 84(1), 51–59 (2017) 150. Yun, S., Guo, T., Zhang, J., He, L., Li, Y., Li, H., Zhu, X., Gao, Y.: Facile synthesis of large-sized monolithic methyltrimethoxysilanebased silica aerogel via ambient pressure drying. J. Sol-Gel Sci. Technol. 83(1), 53–63 (2017) 151. Feng, Q., Chen, K., Ma, D., Lin, H., Liu, Z., Qin, S., Luo, Y.: Synthesis of high specific surface area silica aerogel from rice husk ash via ambient pressure drying. Colloids Surf. A Physicochem. Eng. Asp. 539, 399–406 (2018) 152. Gao, H., Zhang, Z., Shi, Z., Zhang, J., Zhi, M., Hong, Z.: Synthesis of high-temperature resistant monolithic zirconia-based aerogel via facile water glass assisted sol–gel method. J. Sol-Gel Sci. Technol. 85(3), 567–573 (2018) 153. Jia, G., Li, Z., Liu, P., Jing, Q.: Preparation and characterization of aerogel/expanded perlite composite as building thermal insulation material. J. Non-Cryst. Solids. 482, 192–202 (2018) 154. Li, Z., Cheng, X., Gong, L., Liu, Q., Li, S.: Enhanced flame retardancy of hydrophobic silica aerogels by using sodium silicate as precursor and phosphoric acid as catalyst. J. Non-Cryst. Solids. 481, 267–275 (2018)

A. V. Rao et al. 155. Wu, X., Fan, M., Shen, X., Cui, S., Tan, G.: Silica aerogels formed from soluble silicates and methyl trimethoxysilane (MTMS) using CO 2 gas as a gelation agent. Ceram. Int. 44(1), 821–829 (2018) 156. Zhou, T., Cheng, X., Pan, Y., Li, C., Gong, L., Zhang, H.: Mechanical performance and thermal stability of glass fiber reinforced silica aerogel composites based on co-precursor method by freeze drying. Appl. Surf. Sci. 437, 321–328 (2018) 157. Kistler, S.S., Caldwell, A.G.: Thermal conductivity of silica aerogel. Ind. Eng. Chem. 26(6), 658–662 (1934) 158. Fricke, J.: Aerogels – highly tenuous solids with fascinating properties. J. Non-Cryst. Solids. 100(1–3), 169–173 (1988) 159. Shewale, P.M., Rao, A.V., Gurav, J.L., Rao, A.P.: Synthesis and characterization of low density and hydrophobic silica aerogels dried at ambient pressure using sodium silicate precursor. J. Porous. Mater. 16(1), 101–108 (2009) 160. Wong, J.C.H., Kaymak, H., Brunner, S., Koebel, M.M.: Mechanical properties of monolithic silica aerogels made from polyethoxydisiloxanes. Microporous Mesoporous Mater. 183, 23–29 (2014) 161. Scherer, G.W., Hæeid, S., Nilsen, E., Einarsrud, M.-A.: Shrinkage of silica gels aged in TEOS. J. Non-Cryst. Solids. 202(1–2), 42– 52 (1996) 162. Malfait, W.J., Verel, R., Koebel, M.M.: Hydrophobization of silica aerogels: insights from quantitative solid-state NMR spectroscopy. J. Phys. Chem. C. 118(44), 25545–25554 (2014) 163. Gerber, T., Himmel, B., Hübert, C.: WAXS and SAXS investigation of structure formation of gels from sodium water glass. J. Non-Cryst. Solids. 175(2–3), 160–168 (1994) 164. Knoblich, B., Gerber, T.: Aggregation in SiO2 sols from sodium silicate solutions. J. Non-Cryst. Solids. 283(1–3), 109–113 (2001) 165. Brinker, C.J., Scherer, G.W.: Sol-Gel Science. The Physics and Chemistry of Sol-Gel Processing. Elseiver (1990) 166. Iswar, S., Malfait, W.J., Balog, S., Winnefeld, F., Lattuada, M., Koebel, M.M.: Effect of aging on silica aerogel properties. Microporous Mesoporous Mater. 241, 293–302 (2017) 167. Brinter, C., Scherer, G.: The physics and chemistry of sol-gel processing. In: Sol-Gel Science, p. 373. Academic, New York (1990) 168. Rao, A.P., Rao, A.V., Gurav, J.: Effect of protic solvents on the physical properties of the ambient pressure dried hydrophobic silica aerogels using sodium silicate precursor. J. Porous. Mater. 15(5), 507–512 (2008) 169. Shewale, P.M., Rao, A.V., Rao, A.P., Bhagat, S.: Synthesis of transparent silica aerogels with low density and better hydrophobicity by controlled sol–gel route and subsequent atmospheric pressure drying. J. Sol-Gel Sci. Technol. 49(3), 285–292 (2009) 170. Lee, S., Cha, Y.C., Hwang, H.J., Moon, J.-W., Han, I.S.: The effect of pH on the physicochemical properties of silica aerogels prepared by an ambient pressure drying method. Mater. Lett. 61(14–15), 3130–3133 (2007) 171. Li-Jiu, W., Shan-Yu, Z., Mei, Y.: Structural characteristics and thermal conductivity of ambient pressure dried silica aerogels with one-step solvent exchange/surface modification. Mater. Chem. Phys. 113(1), 485–490 (2009) 172. Zu, G., Shimizu, T., Kanamori, K., Zhu, Y., Maeno, A., Kaji, H., Shen, J., Nakanishi, K.: Transparent, superflexible doubly crosslinked polyvinylpolymethylsiloxane aerogel superinsulators via ambient pressure drying. ACS Nano. 12(1), 521–532 (2018) 173. Collins, R.: Aerogels 2017–2027: Technologies, Markets and Players. 2017. https://www.idtechex.com/research/reports/ aerogels-2017-2027-technologies-markets-and-players000507.asp 174. Takeshita, S., Yoda, S.: Translucent, hydrophobic, and mechanically tough aerogels constructed from trimethylsilylated chitosan nanofibers. Nanoscale. 9(34), 12311–12315 (2017) 175. Nguyen, B.N., Meador, M.A.B., Scheiman, D., McCorkle, L.: Polyimide aerogels using triisocyanate as cross-linker. ACS Appl. Mater. Interfaces. 9(32), 27313–27321 (2017)

16

Sodium Silicate-Based Aerogels by Ambient Pressure Drying

417

16 A. Venkateswara Rao received his doctorate from Sardar Patel University with Professor A.R.Patel in 1980 and became lecturer in Physics at Shivaji University. After a postdoctorate at Ecole Polytechnique in Palaiseau, he did a sabbatical as Rhône-Alpes Professor at Lyon-I University in 1992 and later at Poitiers University in 2005. His research interests are flexible aerogels and ambient-pressure drying aerogel preparation.

Shanyu Zhao is a staff scientist in the Building Energy Materials and Components Laboratory at the Swiss Federal Laboratory for Materials Science and Technology (EMPA). He received his PhD in materials science from the Dalian University of Technology and then studied as a visiting researcher in the Chemistry Department at Brown University. His research interests range from sol–gel chemistry and biopolymers to aerogels and nanocomposites.

Uzma K. H. Bangi received her doctorate from Shivaji University, under the guidance of Prof. A.V. Rao in 2011. She worked as the Brain Korean – 21 Fellow in Yonsei University and is the recipient of Young Scientist Fellowship and Women Scientist – A Fellowship, honored by Department of Science and Technology, Govt. of India. Her research interests are low-cost aerogels and coatings.

A. Parvathy Rao received her Master’s degree in Chemistry from Shri Venkateswara University, Tirupati, India, in 1977, and her PhD degree in Chemistry from Shivaji University, Kolhapur, India. She carried out extensive research work in the fields of nano-semiconductors, sol–gel processing, and aerogel materials. She was given research positions and support from various Indian government agencies and worked as a lecturer at Shivaji University.

Matthias M. Koebel received his PhD from Brown University in 2004. After a postdoctoral stay at UC Berkeley with G.A. Somorjai focusing on nanocatalysis, he joined Empa back in his home country – Switzerland – in 2006 where he began building a research group in soft chemistry and aerogels. His core activities are linked to process-scaleup and lab-to-market transfer of nanomaterials science.

Synthesis of Metal Oxide Aerogels via Epoxide-Assisted Gelation of Metal Salts

17

Theodore F. Baumann, Alexander E. Gash, Joe H. Satcher Jr, Nicholas Leventis, and Stephen A. Steiner III

Contents 17.1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419

17.2 17.2.1 17.2.2 17.2.3 17.2.4

Mechanisms of Epoxide-Assisted Gelation . . . . . . . . . . . . Sol Formation and Gelation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hydrolysis and Condensation of Metal Ions . . . . . . . . . . . . . Epoxide-Assisted Gelation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Post-Gelation Considerations: Syneresis and Residual Organics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Health and Safety Considerations when Selecting an Epoxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

17.2.5

420 421 421 422 426 427

17.3 17.3.1 17.3.2

Aerogels Produced by Epoxide-Assisted Gelation . . . . 427 Metal Oxide Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 428 Mixed Metal Oxide and Nanocomposite Aerogels . . . . . . 430

17.4

Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432

approach, the epoxide acts as a proton scavenger, which drives hydrolysis and condensation of hydrated metal species in the sol–gel reaction. This process is generalizable and applicable to the synthesis of a wide range of metal and metalloid oxide aerogels, xerogels, and nanocomposites. In addition, modification of synthetic parameters allows for control over the structure and properties of the sol–gel product. The method is particularly amenable to the synthesis of multicomponent and nanocomposite sol–gel systems with intimately mixed nanostructures. This chapter describes both the reaction mechanisms associated with epoxide-assisted gelation and an overview of materials that have been prepared using this technique. Keywords

Abstract

Over the past two decades, the diversity of metal and metalloid oxide materials prepared using sol–gel techniques has increased significantly. This transformation can be attributed in part to the development of the technique known as epoxide-assisted gelation. The process utilizes organic epoxides as co-reactants for the sol–gel polymerization of simple inorganic metal salts in aqueous or alcoholic media. In this T. F. Baumann (*) · A. E. Gash (*) · J. H. Satcher Jr (*) Lawrence Livermore National Laboratory (LLNL), Livermore, CA, USA e-mail: [email protected]; [email protected]; [email protected] N. Leventis (*) Department of Chemistry, Missouri University of Science and Technology, Rolla, MO, USA Department of Research and Development, Aspen Aerogels, Inc., Northborough, MA, USA e-mail: [email protected] S. A. Steiner III (*) Aerogel Technologies, LLC, Boston, MA, USA e-mail: [email protected]

Epoxide · Assisted · Proton scavenger · Sol–gel · Gelation · Metal salt · Metal oxide · Metalloid oxide · Rare earth · Lanthanide · Actinide · Mixed metal · Multicomponent · Aerogels

17.1

Introduction

Aerogels [1, 2] are a special class of highly porous open-cell foams that exhibit many interesting properties, such as low bulk densities, continuous porosities, nanoscale features, and high surface areas. These unique properties are derived from the aerogel microstructure, which typically consists of threedimensional networks of interconnected nanometer-sized particles. Because of their unusual chemical and textural properties, aerogels have been investigated for a wide variety of applications, including catalysis, sorption, insulation, energy storage, and even cosmic dust collection (see Part IX of this handbook as well as ▶ Chap. 64 for additional examples). To further expand the utility of these materials, efforts have focused on the development of new synthetic processes that can be used to tailor both the composition and

© Springer Nature Switzerland AG 2023 M. A. Aegerter et al. (eds.), Springer Handbook of Aerogels, Springer Handbooks, https://doi.org/10.1007/978-3-030-27322-4_17

419

420

structure of aerogels for different applications. Aerogels are typically prepared using sol–gel chemistry, a process that involves the transformation of molecular precursors into highly crosslinked inorganic or organic gels that can then be dried using special techniques to preserve the tenuous solid network of the gel. For organic and carbon aerogels (see ▶ Chaps. 20, ▶ 21, ▶ 23, and ▶ 35), this transformation involves the polymerization of multifunctional organic monomers into three-dimensional polymer networks [3]. For inorganic aerogels, especially metal and metalloid oxide aerogels, the sol–gel process begins with the hydrolysis and condensation of reactive metal/metalloid-based precursors. For many years, the most commonly used precursors for the synthesis of metal and metalloid oxide aerogels were alkoxides (M (OR)x), due mainly to their reactivity under a variety of conditions [4]. Metal and metalloid oxide aerogels derived from alkoxides are usually synthesized through the hydrolysis of monomeric alkoxide precursors in an alcohol, typically catalyzed by a mineral acid (e.g., HCl(aq)) or a base (e.g., NH4OH(aq)). This process has been instrumental in the synthesis of various main group and transition metal oxide aerogels, including silica, titania, and vanadia (see ▶ Chaps. 13 and ▶ 65). For the majority of elements, however, alkoxide compounds can be expensive, difficult to obtain, or very unstable, precluding their use in the preparation of many types of oxide aerogels. In the mid-1990s, an alternative and highly generalizable approach for the fabrication of metal oxide aerogels was developed that greatly expanded the compositional range that is accessible for oxide aerogels. This process, now commonly referred to as epoxide-assisted gelation, utilizes epoxides to invoke sol–gel polymerization of simple inorganic metal salts in aqueous or alcoholic media [5]. (Note that epoxide-assisted gelation is sometimes referred to as “epoxide-initiated gelation,” however “epoxide-assisted” is a more accurate term because the epoxide is a reagent included in a stoichiometric amount and is consumed during the gelation process.) In this approach, an epoxide acts as a proton (acid) scavenger that drives the sol–gel process by converting aquo ligands (i.e., H2O) that are coordinatively bound to metal/metalloid ions into hydroxo ligands, which in turn react with one another resulting in network-forming oxide bridges. The slow and uniform increase in the pH of the sol–gel solution leads to the formation of an extended metal/metalloid oxide network structure, while the protonated epoxide undergoes an irreversible ring-opening reaction and generally remains in solution. Epoxide-assisted gelation offers many advantages over alkoxide-based synthesis in the preparation of metal oxide aerogels. First, it utilizes simple metal salts (e.g., metal nitrates or halides) as the metal source for the sol– gel reaction, eliminating the need for expensive precursors such as metal alkoxides. As a result, the method allows for

T. F. Baumann et al.

the preparation of many main group, transition metal, and lanthanide/actinide metal oxide aerogels that were previously impractical to synthesize using traditional alkoxidebased sol–gel chemistry. In addition, the process is highly flexible and allows for control over the microstructure of the gel network through modification of synthetic parameters. For some metal oxides, such as those of iron and tungsten, even the phase of the resulting oxide can be controlled. Furthermore, because of the mild reaction conditions employed, the technique is also amenable to the fabrication of mixed metal oxides and nanocomposites. As a historical note of interest, despite its relatively recent adoption by the aerogel community, epoxide-assisted gelation was actually first reported by Ziese in 1933 [6] and was even adopted by Kistler as early as 1938 for the synthesis of a variety of oxide aerogels including alumina, chromia, titania, and iron oxide [7]. Curiously, this approach to sol–gel synthesis of metal oxides was somehow forgotten for nearly 50 years, only to be reintroduced in 1993 by Itoh et al. for the sol–gel preparation of aluminosilicate gels [8] and then again in 1994 by Tillotson et al. at Lawrence Livermore National Laboratory in support of sol–gel synthesis of lanthanide and lanthanide-silicate aerogels [9]. This said, the technique has been substantially broadened and further characterized since the days of Ziese and Kistler and can now be used to synthesize a virtually limitless set of single- and multicomponent oxide aerogels. This chapter provides an overview of epoxide-assisted gelation, including the mechanisms associated with gel formation via this technique, synthetic parameters that influence these mechanisms, and a survey of different aerogel compositions that have been prepared using it.

17.2

Mechanisms of Epoxide-Assisted Gelation

In general, the sol–gel preparation of metal and metalloid oxide aerogels involves three basic steps as illustrated in Fig. 17.1: (1) formation of a stable colloid (i.e., a sol), (2) gelation, and (3) drying. Wet gels of metal oxides are prepared through the hydrolysis and condensation of inorganic precursors, leading to the formation of a threedimensional inorganic oxide network. To preserve the original gel structure, the solvent that resides within the pores of the solid network is removed using a technique such as supercritical drying or other network-preserving drying techniques, resulting in a metal/metalloid oxide aerogel (see Part I and ▶ Chap. 65). Although the technique used to dry an aerogel can have a significant impact on the properties of the resulting material, these details are outside of the scope of this chapter and will not be discussed here. Rather, this section will focus on the underlying mechanisms of

17

Synthesis of Metal Oxide Aerogels via Epoxide-Assisted Gelation of Metal Salts

Solution of precursors Hydrolysis Condensation Liquid Solid

Sol

Gelation

Wet gel

Supercritical f luid processing

Evaporative drying

Gas

421

molecular monomers, a process that involves the nucleation and growth of particles in solution. The size and properties of these particles depend on the relative rates of these two processes. Sol formation is favored when the rate of nucleation is high and the rate of particle/crystal growth is low. The generation of sols also requires controlled conditions such that the resulting sol is stable toward agglomeration and precipitation. Several factors, such as polarity of the solvent, ionic strength of the reaction medium, and temperature can be used to manipulate both the formation and stability of the sol. Gelation is the process whereby a free-flowing sol is converted into a material comprising a three-dimensional solid network throughout the solvent medium. The point of gelation is typically identified by an abrupt rise in viscosity and an elastic response to stress. This sol–gel transition can be envoked several ways, such as a change in the ionic strength of the solution, through the removal of the reaction solvent other conditions that result in network bond formation. For preparation of metal/metalloid oxide aerogels, however, gelation is most conveniently induced through a change in the pH of the reaction solution. Under controlled conditions, the pH change reduces the electrostatic barrier to agglomeration and promotes intercluster crosslinking, leading to the formation of the three-dimensional gel network.

17.2.2 Hydrolysis and Condensation of Metal Ions

Solid

Aerogel

Xerogel

Fig. 17.1 Preparation of aerogels and xerogels via sol–gel chemistry

epoxide-assisted gelation responsible for driving the formation of the porous oxide network structure.

17.2.1 Sol Formation and Gelation A colloid is defined as a dispersion of finely divided particles in a homogeneous medium [10]. By convention, colloidal particles are smaller than 500 nm in size and consist of on the order of 103–109 atoms. Because of their small size, these particles are small enough that they remain suspended indefinitely due to Brownian motion, a random walk resulting from momentum imparted by collision with molecules of the suspending medium. Such a dispersion of colloidal particles in a liquid medium and can be reacted to form a gel is termed a sol. Sols can be prepared through condensation of

The sol–gel chemistry of metal salts is more complex than that of metal alkoxides because of the numerous molecular species that can be formed depending on the oxidation state of the metal, the pH of the reaction solution, and the concentration of the reactants. Since the sol–gel polymerization of inorganic salts varies widely among the different metal ions, this section will only present a general summary of the topic. For a detailed description of the mechanism of condensation and gelation, the reader is referred to a literature review on the sol–gel chemistry of transition metal oxides [11] and to ▶ Chap. 2. The aqueous chemistry of metal cations is quite complex due to the occurrence of hydrolysis reactions that convert ions into new ionic species or precipitates. When dissolved in pure water, a cation Mz+ becomes solvated by the surrounding water molecules. For transition metal ions, charge transfer occurs from the filled 3a1 nonbonding orbital of the water molecule to an empty d orbital of the transition metal. This interaction causes the positive partial charge on the hydrogen to increase, making the coordinated water molecule more acidic. Depending on the magnitude of the charge transfer, in noncomplexing aqueous media, hydrolysis equilibria in which three types of ligands are present are established: aquo (OH2), hydroxo (OH), and oxo (¼O) ligands (see Fig. 17.2).

17

422

B

B

Fig. 17.3 Protonation and ring opening of an epoxide in the presence of a Brønsted acid, HA. In the second reaction, the role of the Brønsted acid is played by the hydrated metal ion (e.g., [Fe(H2O)6]3+)

[M(OH2)]z+

[M(OH)](z–1)+ + H+

B

B

Fig. 17.2 Hydrolysis equilibria for hydrated metal ions

T. F. Baumann et al.

[M=O](z–2)+ + 2H+

H O

O

⊕

+ HA

OH

A– Ringopening

A

H O

O + Fe(H2O)63+

The degree of hydrolysis depends on several factors, including charge density and electronegativity of the metal, the coordination number of the aquo complex, and the pH of the reaction solution. For example, under similar conditions, low-valent cations (z < 4) tend to yield aquo, hydroxo, or aquo-hydroxo complexes, whereas high-valent cations (z > 5) tend to form oxo or oxo-hydroxo complexes. Tetravalent metals can form any of the possible complexes depending on the pH of the solution. Condensation of these solvated metal ions can proceed through two possible mechanisms. Olation is the process by which one or more hydroxy bridges are formed between two metal centers. For coordinatively saturated hydroxo-aquo precursors, olation occurs through a nucleophilic substitution mechanism, wherein the hydroxo ligand on one metal complex serves as the nucleophile, and water on another metal complex serves as the leaving group. Oxolation is the process in which an oxo bridge is formed between two metal centers. In the case where the metal is coordinatively unsaturated, oxolation occurs rapidly via nucleophilic addition leading to edge- or surface-shared polyhedra. For coordinatively saturated metals, oxolation proceeds by a two-step substitution reaction between oxyhydroxy precursors via nucleophilic addition followed by water elimination to form an M–O–M bond. These charged precursors, however, cannot condense indefinitely to form a solid phase of metal oxide. As electrondonating water molecules are eliminated from the metal centers during the substitution reactions, the hydroxo ligands become less nucleophilic, and condensation stops. Depending on the nature of the metal and reaction conditions, condensation is typically limited to the formation of dimers

⊕

+ Fe(H2O)5OH2+

and tetramers. In order to obtain condensed species (i.e., sols and gels), a change in the reaction conditions is required, such as a change in temperature or solution pH. These variables control the growth and aggregation of the metal oxide species throughout the transition from the sol state to the gel state.

17.2.3 Epoxide-Assisted Gelation In the epoxide-assisted gelation process, organic epoxides are used to invoke the hydrolysis and condensation processes described above, which lead to the formation of a threedimensional gel network. Essentially, the epoxides serve as proton scavengers in the sol–gel reaction, analogous to their use in organic synthesis [12]. This process involves the protonation of the epoxide oxygen by an acid followed by the opening of the epoxide ring through nucleophilic attack by the conjugate base, as shown in Fig. 17.3. For example, introduction of an epoxide to a solution containing the aquo ion [Fe(H2O)6]3+ (pKa ≈ 3) leads to the protonation of the epoxide ring and formation of the aquo-hydroxo Fe3+ complex. The iron complex can now undergo substitution and addition reactions to form condensed iron oxide species. Meanwhile, in the presence of a suitable nucleophile (e.g., the counterion of the metal salt), the protonated epoxide undergoes irreversible ring-opening reactions. The net effect of this process is an overall increase in the pH of the reaction solution, as evidenced by the plot of measured pH versus time following addition of epoxide to an aqueous solution of ferric chloride (Fig. 17.4). Outside of epoxide-assisted

17

Synthesis of Metal Oxide Aerogels via Epoxide-Assisted Gelation of Metal Salts

a

6 FeCl3•6H2O salt 5

423

O

b

O

O

O

O

O O

pH

4

CH2Cl

3

Fig. 17.5 The molecular structures of (a) 1,2-epoxides (top row, left to right, ethylene oxide, propylene oxide, 1,2-epoxybutane; bottom row, left to right, cyclohexene oxide, epichlorohydrin) and (b) 1,3-cyclic ethers (left to right, trimethylene oxide aka oxetane and 2,2-dimethyloxetane)

2 Fe(NO3)3•9H2O salt 1 0

0

10

20

30

40

50

60

Time (min) Fig. 17.4 Plot of pH as a function of time for aqueous solutions of Fe (III) salts following addition of propylene oxide

gelation, the addition of bases (e.g., OH, CO2 3 , or NH3) to aqueous solutions of metal ions has been a popular route to form condensed metal oxide species [11]. The problem with using such an approach in the synthesis of aerogels is that the condensed metal oxide species are almost instantly precipitated from the solution as a result of the rapid reaction of the base with the solvated metal ions. With the epoxide method, the solvated metal complex and the epoxide mix to produce a homogeneous solution before a significant increase in pH occurs. The relatively slow and uniform increase in the reaction pH allows for controlled olation and oxolation reactions to occur, leading to the formation of a stable metal oxide sol and eventually a metal oxide gel network. With the epoxide-assisted gelation approach, synthetic variables such as the choice of epoxide, the anion of the metal salt, and the solvents used in the reaction can all have a profound impact on network formation and thus the properties of the resultant aerogel. For example, the size of the cyclic ether ring (three or four member) as well as ring substituents will affect the cyclic ethers’s reactivity with the hydrated metal ions in the sol–gel reaction. The rate at which the pH of the sol–gel reaction changes will in turn influence the nucleation of the condensed phase and growth of the network structure. These effects can be illustrated through examination of Fe(III) oxide aerogels prepared using epoxide-assisted gelation [13]. Iron oxide aerogels can be synthesized through the addition of a cyclic ether to an ethanolic or aqueous solution of an Fe3+ salt, such as FeCl36H2O or Fe(NO3)39H2O. The addition of epoxides,

such as propylene oxide (Fig. 17.5a), to Fe3+ sol–gel solutions leads to the formation of gels consisting of poorly crystallized or amorphous iron oxyhydroxide phases, such as ferrihydrite (Fig. 17.6a). By comparison, the use of 1,3-cyclic ethers, such as trimethylene oxide or 2,2-dimethyloxetane (see Fig. 17.5b), under the same reaction conditions leads to the formation of the akaganeite phase (β-FeOOH) (Fig. 17.6b). The structural differences observed in these systems can be explained by the different reactivities of the two cyclic ether types, which are attributable to structural differences between the three- versus four-member ring systems. Using geometric arguments, the larger cyclic ether ring of the oxetanes is less strained and, therefore, less reactive than that of the epoxides. As a result, reactions associated with ring opening of oxetanes are likely to occur at slower rates than those of epoxides. This trend can be seen in the gelation times observed for Fe(III) oxide systems prepared with both three- and four-member ring cyclic ethers (Table 17.1). The gelation times for the oxetanes are markedly longer than those for the epoxides, reflecting the disparate reactivity of these reagents. In the case discussed above, the slower rise in the pH with the oxetanes allows for the growth of the highly reticulated akaganeite network (Fig. 17.6b). Interestingly, the elastic moduli (i.e., stiffness) of the akaganeite aerogels are over an order of magnitude greater than those of iron oxide aerogels prepared using the epoxides. The reactivity of epoxides also depends on the number and type of substituents on the epoxide ring. The presence of functional groups on the α-carbon atoms of the epoxide can strongly influence the ring-opening process through both steric and electronic effects. As a result, gelation times can vary significantly for different epoxide derivatives, depending on the ring substituents (Table 17.1). The anion of the metal salt can also have a profound impact on the structure and properties of aerogels prepared by epoxide-assisted gelation. Previous work on the sol–gel chemistry of transition metal oxides has shown that anions

17

424

T. F. Baumann et al. Table 17.1 Summary of gel times (tgel) for iron(III) oxide gels prepared with different epoxides and oxetanesa

a

Type Epoxides

Oxetanes

20 nm

Epoxide Butadiene monoxide Cyclohexene oxide Cis-2,3-Epoxybutane Propylene oxide 1,2-Epoxybutane 1,2-Epoxypentane (2,3-Epoxypropyl)benzene Glycidol Epifluorohydrin Epichlorohydrin Epibromohydrin Trimethylene oxide 3,3-Dimethyloxetane

tgel (minutes) 0.33 0.45 0.72 1.5 2.5 4.8 27 62 82 85 109 480 1,400

Reaction conditions: Fe(NO3)39H2O [0.37 M], ethanol, molar ratio of cyclic ether/Fe ¼ 11

a

b

20 nm

Fig. 17.6 Transmission electron micrographs (TEM) of iron oxide aerogels prepared from FeCl3 using different epoxides as gelation promoters: (a) propylene oxide (phase obtained, ferrihydrite) and (b) trimethylene oxide (phase obtained, akaganeite)

play both a chemical and physical role in the homogeneous precipitation of metal oxides [11]. Some anions can be strongly coordinated to the metal cations, thus changing the reactivity of the solvated metal species toward hydrolysis and condensation. In addition, once the colloidal particles have formed, the anions can also modify the aggregation process through changes to the double-layer composition and the ionic strength of the solution. In general, the interaction of the anion with the metal center is determined by the electronegativity of the anion relative to the ligated water molecules.

The influence of anions on network formation can be seen in the case of alumina aerogels prepared by epoxide-assisted gelation [14]. The synthesis of these materials involves the addition of propylene oxide to ethanolic solutions of hydrated aluminum salts, either AlCl36H2O or Al (NO3)39H2O. Characterization of the different materials showed that aerogels prepared from hydrated AlCl3 possessed microstructures containing highly reticulated networks of pseudoboehmite (AlOOH) leaflets or sheets, 2–5 nm wide and of varying lengths, whereas aerogels prepared from hydrated Al(NO3)3 were amorphous with microstructures comprised of interconnected spherical particles (the so-called string-of-pearls morphology) with diameters in the 5–15 nm range (Fig. 17.7). This difference in microstructure results in distinct physical and mechanical properties for each type of aerogel. In particular, the alumina aerogels with the nanoleaflet microstructure are optically transparent and significantly stiffer than those with the spheroidal-particle network [15]. On the basis of these observations, the chloride and nitrate anions appear to have different effects on the formation of the alumina network. In inorganic chemistry, the chloride anion exhibits complexing behavior, and, therefore, one would expect that Cl would be involved in the formation of the alumina particles. The nitrate anion, on the other hand, is a weak, or noncomplexing, anion and may not be involved in the hydrolysis and condensation reactions. The different complexing behavior of nitrate versus chloride has been invoked to explain historic difficulties in obtaining cobalt oxide aerogels [16]. It has been observed that only cobalt oxide sols from Co(NO3)2.6H2O yield robust gels, noting, however, that gelation times are very long (more than 10 days) [17]. Using the cobalt salt [Co(H2O)6]Cl2 as the cobalt source, which is pink, it was noted [16] that solutions

17

Synthesis of Metal Oxide Aerogels via Epoxide-Assisted Gelation of Metal Salts

a

b

50 nm

50 nm

Fig. 17.7 Transmission electron micrographs (TEM) of alumina aerogels prepared using propylene oxide with (a) hydrated Al(NO3)3 and (b) hydrated AlCl3

of this salt in N,N′-dimethylformamide (DMF) are blue, suggesting that octahedral [Co(H2O)6]2+ is in equilibrium with blue tetrahedral [CoCl4]2 (Fig. 17.8, reaction 1) [18]. The position of this equilibrium was assessed by titrating a DMF solution of [Co(H2O)6](NO3)2 (0.43 M, equal to the concentration typically used to make cobalt oxide gels) with aqueous HCl. According to Fig. 17.9, the absorption spectrum of [Co(H2O)6]Cl2 is very similar to that of [Co (H2O)6](NO3)2 after addition of 4 mol equivalents of HCl. Therefore, epoxide-assisted gelation of [Co(H2O)6]Cl2 (employing epichlorohydrin as the epoxide, see Fig. 17.8,

425

reaction 2) becomes convolved with reaction 1, whose equilibrium tends toward [CoCl4]2. Prior TEM investigations have shown that cobalt oxide nanoparticles do form in DMF sols however as a precipitate, presumably according to reaction 3 of Fig. 17.8 [19]. This implies that reaction 1 of Fig. 17.8 occurs also at the surface of the formed cobalt oxide particles interfering with interparticle Co–O–Co bridging. Indeed, thermogravimetric analysis of such cobalt oxide nanoparticles has shown two mass losses in the 200–400  C range analogous to a superposition of Co(OH)2 and Co2(OH)3Cl [17, 20]. The first mass loss at ~220  C is attributed to dehydroxylation of hydroxo-capped cobalt oxide nanoparticles, while the second mass loss at ~300  C is assigned to loss of chlorine from chloro-capped particles. Overall, the presence of a significant amount of surface –Cl caps appears to reduce the propensity of cobalt oxide nanoparticles to undergo interparticle Co–O–Co bridging. In addition to interactions with the solvated metal ions, the anion of the metal salt can also affect the formation of condensed phases through interactions with the ring-opening reaction of the epoxide. The ring-opening process involves two steps: (1) protonation of the epoxide oxygen by an acidic aquo ion and (2) subsequent attack of carbons in the epoxide ring by a suitable nucleophile (Fig. 17.3). Therefore, the rate at which protons are irreversibly consumed in this reaction is determined by the relative nucleophilicity of the species present. In many cases, the anion of the metal salt can serve as the nucleophile for the ring-opening reaction, but its efficacy in this process depends on a number of factors, including the chemical nature of the anion as well as the reaction solvent. For example, iron oxide gels can be prepared through the addition of propylene oxide to an aqueous solution of FeCl36H2O, while gelation does not occur under the same conditions when Fe(NO3)39H2O is used as the precursor [5]. This result can be understood in terms of the nucleophilic character of the anions in each reaction mixture. When Fe(NO3)39H2O is used as the precursor, the potential nucleophiles for the ring-opening reaction are the nitrate ion and water. Under these conditions, water is a better nucleophile than the nitrate ion. As a result, water preferentially attacks the ring carbons of the protonated epoxide, and ring opening consequently produces 1,2-propanediol. This reaction, however, regenerates a proton into solution, and thus the pH of the solution does not change appreciably (Fig. 17.10a). By contrast, when FeCl36H2O is utilized as the precursor, the potential nucleophiles in the reaction mixture are chloride ion and water. In this case, chloride is the better nucleophile, and the protonated epoxide is ring-opened by the chloride ion to yield 1-chloro-2-propanol (Fig. 17.10b). No protons are regenerated in this reaction, leading to an increase in the pH of the reaction mixture and the formation of the iron oxide network. These processes have been verified both by characterization of the ring-opened products using nuclear magnetic

17

426

T. F. Baumann et al.

Fig. 17.8 Relevant chemical processes involved in epichlorohydrin-assisted condensation of [Co(H2O)6]2+ in DMF in the presence of chloride anions

[CoCl4]2–

Cl –

(1)

[Co(H2O)6]2+

DMF, H2O O Cl

[Co(H2O)6]2+

[Co(H2O)5OH]+

(2)

OH Cl

[Co(H2O)5OH]+

HCl : Co(NO3)2.6H2O

2

% Absorbance

Co(NO3)2.6H2O CoCl2.6H2O 1

OH or Cl

[Co(H2O)5OH]+ – H2O

4:1 2:1

1:1 1:2 1:4

0 300

400

500 600 Wavelength (nm)

700

800

Fig. 17.9 Spectrophotometric titration of a [Co(H2O)6](NO3)2 solution in DMF (0.43 M, red line) with HCl (black dashed lines, the corresponding ratios denote the molar ratio of HCl to [Co(H2O)6] (NO3)2). As the concentration of HCl increases, the intensity of the absorption at 523 nm decreases (red arrow pointing down) and the intensity at 675 nm increases. The blue line shows the spectrum of [Co(H2O)6]Cl2 in DMF at the same concentration (0.43 M). Reprinted with permission from [16], copyright 2019, American Chemical Society

resonance (NMR) spectroscopy and by measurement of the solution pH for each of these anions. As shown in Fig. 17.4, only a slight rise in pH occurs over time upon addition of epoxide to an aqueous solution of Fe(NO3)3. Gelation can still proceed in this system through addition of additional nucleophilic anions, such as chloride or bromide, to the aqueous iron nitrate solution. In contrast, when ethanol is used as the reaction solvent, iron oxide gels can be prepared directly from Fe(NO3)3, as the nucleophilicity of the nitrate ion increases in nonaqueous media. In this case, the attack of the protonated epoxide ring by the nitrate ion leads to the formation of the nitrate ester of the alcohol (i.e., 1- and 2-nitroxy-2-propanol) and a net consumption of protons from

[(H2O)5Co–O–Co(H2O)5]2+

CoOx

the reaction solution. With regard to reaction solvent, it is also important to note the importance of water in the epoxideassisted gelation method. The formation of condensed metal oxide species from metal salts necessitates the presence of an oxygen source (i.e., water) in the reaction medium. In our experience, successful gel synthesis using the epoxide method requires the addition of water, either as the reaction solvent (or co-solvent) or through the use of hydrated metal salts. The aforementioned examples underscore the importance of considering each of these reaction variables when applying this process to the preparation of metal oxide aerogels. Therefore, a basic understanding of how each of these variables influences structure formation can be a powerful tool for tailoring the structure and properties of the aerogel.

17.2.4 Post-Gelation Considerations: Syneresis and Residual Organics It is not uncommon for gels prepared via epoxide-assisted gelation to undergo a high degree of syneresis post-gelation. Alumina gels synthesized from Al(NO3)39H2O and propylene oxide using water as the solvent system, for example, can undergo approximately 50% linear shrinkage within 24– 48 hours following gelation. The degree of syneresis that a gel will undergo varies from procedure to procedure and strongly depends on factors including the anion of the metal salt, concentration of metal ions, solvent system, amount of water, and epoxide used. Aging the gel in a chamber containing a solvent-saturated atmosphere ensures the gel can fully strengthen until syneresis ceases (see ▶ Chap. 3). For applications where shrinkage is undesirable or not welltolerated, syneresis of gels can be arrested to some extent by transferring the gel into a nonaqueous solvent bath soon after the gel has formed, noting that doing so may reduce the strength of the final dried aerogel or xerogel as it arrests aging and network strengthening.

17

Synthesis of Metal Oxide Aerogels via Epoxide-Assisted Gelation of Metal Salts

Fig. 17.10 Different epoxide ring-opening mechanisms: (a) via nucleophilic attack by H2O in the form of an aquo ligand or solvent water and (b) via nucleophilic attack of a counter anion of the hydrated metal salt (e.g., Cl)

a

427

H O

OH

OH

⊕ H

H2O b

+ H+ HO

O ⊕ H

1,2-Propanediol

H O

OH

⊕ Cl

It is also not uncommon for gels prepared through the epoxide approach to retain some amount of residual organics all the way through drying into an aerogel or xerogel, often in a higher amount than for analogous alkoxide-derived materials. Most notable among these residual organics are products resulting from ring opening of the epoxide, which can remain physisorbed on the dry material’s surface, entrapped in the dry material’s microporosity, and/or chemically bound to the dry material’s oxide network. This can be mitigated in part through careful design of a post-gelation solvent exchange protocol (see ▶ Chap. 3) that takes into consideration the solubility of the various ring-opened products that result, keeping in mind that such products may bear groups originating from the anion of the metal salt, e.g., chloro or nitro, and thus may have poor solubility in common exchange solvents such as methanol or ethanol. Residual organics can be further removed through postprocessing of the dried aerogel or xerogel material, for example, by annealing in air or oxygen or by vacuum drying (see ▶ Chap. 6). Additional guidance related to the practical implementation of epoxide-assisted gelation can be found in ▶ Chap. 65.

17.2.5 Health and Safety Considerations when Selecting an Epoxide Despite the tremendous synthetic flexibility and ease of the epoxide-assisted method, health and safety concerns surrounding the use of certain epoxides may complicate the ability to implement some epoxide-assisted procedures in some regions. For example, several epoxides commonly employed in epoxide-assisted gelation including propylene oxide and epichlorohydrin are known or suspected carcinogens, and some institutions impose restrictions on their use.

Cl 1-chloro-2-propanol

In the European Union, for example, REACH (Registration, Evaluation, Authorization and Restriction of Chemicals) legislation has been enacted that identifies propylene oxide as a substance of very high concern (SVHC) [21], which has limited the ability to use it in academic and industrial environments in Europe. In response to this, researchers at Atomic Weapons Establishment (AWE) surveyed a variety of epoxides with an eye toward identifying safer reagents for use in epoxide-assisted gelation [22]. They showed cyclic ethers such as 1,2-epoxybutane, trimethylene oxide, and cyclohexene oxide, which are not identified as SVHC, are effective alternatives to propylene oxide for the production of iron(III) oxide aerogels. Consideration must be paid to process parameters, however, including the amount of epoxide used, the solvent system, the solvent-to-water ratio, and the anion of the metal salt in order to obtain comparable materials. Furthermore, gelation time and textural properties of the resulting materials may be impacted depending on the epoxide used. This said, while the aforementioned non-SVHC epoxides and other epoxides may not necessarily be molefor-mole drop-in substitutes for propylene oxide or other epoxides of concern, safer alternatives do exist (see Table 17.1) and are perfectly suitable for synthesis of aerogels and other sol–gel-derived materials via epoxideassisted gelation.

17.3

Aerogels Produced by Epoxide-Assisted Gelation

The epoxide-assisted approach has been used to prepare a wide variety of metal oxide, mixed metal oxide, and composite aerogel materials as illustrated in Fig. 17.11 and summarized in Table 17.2. This section provides a general overview of the different aerogel materials that have been

17

428 Fig. 17.11 Monolithic aerogels prepared via the epoxide-assisted gelation method: (a) alumina; (b) chromia; (c) β-FeOOH (akaganeite); (d) Ni(II) oxidebased aerogel; (e) tungsten oxide; (f) tin oxide; (g) various rare earth oxide aerogels; and (h) corresponding polyurea-crosslinked nanocomposite aerogels of the oxides shown in frame G (see ▶ Chap. 29). (Frames g and h reprinted with permission from [23], copyright 2007, American Chemical Society)

T. F. Baumann et al.

a

b

c

d

e

f

g

h

prepared using epoxide-assisted gelation reported to date. For more information about procedure specifics and resulting materials properties, please refer to the respective original published report.

17.3.1 Metal Oxide Aerogels A number of oxide aerogel materials of main group elements have been prepared using the epoxide approach (see also ▶ Chap. 44). As described in the previous section, the addition of propylene oxide to ethanolic or aqueous solutions of

hydrated Al3+ salts yields alumina gels that can be supercritically dried using CO2 to produce alumina aerogels [8, 14, 57–60]. Because of their thermal stability and high surface area, alumina aerogels of this type have been pursued for use as high-temperature catalyst supports. Epoxide-assisted gelation has also been applied to the synthesis of tin oxide aerogels [26]. These materials were prepared through treatment of an aqueous solution of SnCl45H2O with propylene oxide. Structural characterization showed that the resulting tin oxide aerogels consisted of interconnected networks of crystalline SnO2 nanoparticles. Because of their high surface area, tin oxide aerogels produced by epoxide-assisted

17

Synthesis of Metal Oxide Aerogels via Epoxide-Assisted Gelation of Metal Salts

Table 17.2 Summary of inorganic oxide aerogels that have been synthesized using epoxide-assisted gelation of metal salts Type Main group metal and metalloid oxides

Transition metal oxides

Lanthanide oxides

Actinide oxides

Oxide Alumina Gallium oxide Indium oxide Tin oxide Scandia Titania Vanadia Chromia Manganese oxide Iron oxide Cobalt oxide Nickel oxide Copper oxide Zinc oxide Yttria Zirconia Niobia Ruthenia Lanthanum oxide Hafnia Tantala Tungsten oxide Ceria Praseodymia Neodymia Samaria Europia Gadolinia Terbia Dysprosia Holmia Erbia Thulia Ytterbia Lutetia Thoria Urania

References [7, 14] [24] [24, 25] [19, 25–28] [23] [7, 19, 29–31] [32] [7, 19, 24] [33] [5, 7, 13, 19, 24, 34] [16, 19, 35, 36] [19, 37] [19, 38–41] [42] [19, 23, 43–45] [24, 43, 44, 46] [24] [47] [23] [19, 24] [24, 48] [24, 49] [50, 51] [9, 23, 52] [9, 23, 52] [23, 51] [23, 45, 51] [23, 51, 52] [23, 51] [19, 23, 51, 53] [23, 51] [9, 23, 52] [23, 51] [23, 53, 54] [23, 51] [55] [56]

gelation are of particular interest for sensor design. Gallium and indium oxide aerogels have also been prepared using the epoxide approach [24]. For example, the treatment of an ethanolic solution containing GaCl36H2O or InCl36H2O with propylene oxide produces opaque white gels that can then be dried by supercritical extraction. For transition metal oxide aerogels, the most extensively studied composition prepared by the epoxide method is iron (III) oxide. As described in the previous section, the structure

429

and properties of the iron oxide network can be controlled through modification of synthetic parameters. The versatility of this approach has allowed for the synthesis of iron oxide materials for a variety of applications, including energetic materials [61–64], magnetic nanostructures [65–68], and catalysis [69, 70]. The epoxide method has also been applied to the preparation of other transition metal oxide aerogels. Amorphous chromia aerogels can be prepared through the addition of epoxides to ethanolic solutions of Cr3+ salts [19, 24]. High-surface-area ruthenia aerogels, of interest for energy storage applications, have been synthesized through the reaction of propylene oxide or 1,2-epoxybutane with hydrated ruthenium chloride [47]. Titania aerogels prepared by the epoxide method [29] have been investigated as highsurface-area photocatalysts [71]. The syntheses of metal oxide aerogels from other early transition elements, including zirconium, hafnium, niobium, tantalum, and tungsten, have also been reported (see ▶ Chap. 44) [19, 24]. The chloride salts of these elements can be reacted with propylene oxide in ethanol to yield the corresponding gel. Synthesis of vanadia aerogels using vanadium oxychloride as the metal source employing epichlorohydrin as the epoxide has been reported as well [32]. Vanadia aerogels made this way were shown to be practically identical to vanadia aerogels obtained via the alkoxide method with respect to their unusual wormlike nanomorphology, crosslinkability by isocyanates, and mechanical properties (see ▶ Chap. 29). Although the epoxide approach works quite well for preparing oxide aerogels from higher-oxidation-state ions of the early- and mid-row transition metals, the process has proven problematic for the synthesis of monolithic aerogels derived from divalent ions of the late 3D transition metals, such as Ni2+, Cu2+, and Zn2+. This issue is likely related to the low relative acidities of these hydrated cations. The acidity of aquo ions depends on charge, with M4+ cations typically being more acidic than M3+ cations and M2+ cations being only weakly acidic. For example, the NiðH2 OÞ2þ 6 species (pKa ~ 6.2–10.2) is a much weaker acid than the corresponding Fe3+ species (pKa ~ 3.0). Due to the lower relative acidities of these aquo ions, the processes that drive epoxide-assisted gelation are significantly slowed or are not favorable for the formation of a stable gel network. This effect is qualitatively observed in the relatively faster rates of gel formation for the higher-oxidation-state metal ions as compared to those of the lower-oxidation-state species under comparable experimental conditions. Nevertheless, recent work has shown that through modification of the reaction conditions, the epoxide approach can indeed be used for the synthesis of these metal oxide aerogels. The use of 2-propanol or DMF instead of water or ethanol in the epoxide reaction has been shown to allow for the synthesis of Cu2+- and Zn2+-based aerogels [19, 34, 38, 42]. The as-prepared aerogels can then be converted into their

17

430

T. F. Baumann et al.

respective crystalline metal oxide phases through calcination at elevated temperatures (see ▶ Chap. 6). Cupric oxide aerogel nanocomposites containing a resorcinolformaldehyde interpenetrating network that behave as energetic materials have been produced through epoxide-assisted gelation [39] (see ▶ Chaps. 44 and ▶ 47). Similarly, monolithic Ni2+- and Co2+-based aerogels have been prepared through the reaction of the respective hydrated chloride salt with propylene oxide [17, 35, 37]. The epoxide-assisted gelation approach has also been applied to the synthesis of lanthanide and actinide oxide aerogels. For example, nanocrystalline cerium oxide aerogels have been prepared by reacting Ce(III) salts with propylene oxide [50]. Because of their high surface areas and electronic conductivities, these materials are of interest as catalysts in solid oxide fuel cells. Aerogel materials have also been prepared from the chloride salts of other lanthanide elements (Fig. 17.11) [9, 19, 23, 51–54]. Oxide aerogels derived from actinide elements, such as thorium and uranium, have also been reported [55, 56]. Materials of this type have received attention for their use in the production of nuclear fuel materials. As an example, low-density uranium oxide (UO3) aerogels can be prepared through the addition of propylene oxide to an ethanolic solution of uranyl nitrate [UO2(NO3)26H2O].

Fig. 17.12 Transmission electron micrographs (TEM) of an iron oxide/silica interpenetrating aerogel framework (atomic ratio Fe/Si ¼ 2) prepared using TMOS as the silica precursor and trimethylene oxide as the gelation promoter. (a) Bright-field image; (b) Fe element map; (c) Si element map; and (d) O element map

a

17.3.2 Mixed Metal Oxide and Nanocomposite Aerogels The epoxide-assisted gelation process has played an important role in the fabrication of aerogel nanocomposites. Nanocomposites are formed by the combination of two or more distinct phases into a new material, typically with the goal of attaining enhanced physical, chemical, or mechanical properties relative to those of the individual components. Sol–gel chemistry allows for the assembly of different phases at the nanometer scale and thus provides the ability to engineer nanocomposite aerogel architectures for a variety of applications. Aerogel nanocomposites can be formed in three different ways: (a) by modifying the chemical composition of the skeletal framework of the aerogel, (b) by introducing a second phase (either discrete or continuous) into the porous space of an aerogel matrix (filling the space either partially or completely), and (c) by conformally coating the skeletal framework of the aerogel matrix with a second phase. This section will concentrate on nanocomposites of types a and b described above; an in-depth review of nanocomposite aerogels of type c, including those produced via the epoxide-assisted method (such as the materials shown in Fig. 17.11h), can be found in ▶ Chap. 29.

b

50 nm c

d

17

Synthesis of Metal Oxide Aerogels via Epoxide-Assisted Gelation of Metal Salts

One of the advantages of epoxide-assisted gelation is that it provides a versatile and relatively straightforward route to the preparation of binary or ternary oxides. Similar to singlecomponent systems, mixed (also called mixed matrix) metal oxide gels can be readily prepared through the addition of epoxides to solutions containing two or more metal salts. When synthesizing such mixed metal oxide aerogels, hydrolysis and condensation reactions can yield a variety of different network architectures within the aerogel framework. For example, the condensed phase can be comprised of separate interpenetrating networks of the two metal oxides, M1–O– M1– and –M2–O–M2–, or mixed phases of the two materials, M1–O–M2–. Alternatively, one of the metal oxides can exist as discrete entities (i.e., nanoparticles) supported by the primary oxide structure. In general, the composition and bonding motif of the gel structure are primarily functions of the reaction stoichiometry of the inorganic precursors and the relative rates of hydrolysis of the metal ions. Numerous examples of mixed metal oxide aerogels prepared by epoxide-assisted gelation have been reported in the literature. Solid-state electrolyte materials, such as yttria-stabilized zirconia (YSZ), have been prepared using this approach [43, 44] (see also ▶ Chap. 65). Strontium-doped lanthanum manganite (LSM) materials for use as cathodes in solid oxide fuel cells have also been fabricated using the epoxide method [72]. Mixed metal oxides with the spinel structure (AB2O4), such as ZnFe2O4 and CoAl2O4, have been synthesized using propylene oxide as the gelation agent [73, 74]. Using a similar approach, novel alumina catalysts containing copper and zinc oxide have been synthesized and investigated as catalysts for methanol reforming [75]. Examples of interpenetrating aerogel networks have also been reported. For example, the epoxide method has been utilized in the design of interpenetrating iron oxide and silicon dioxide frameworks for use as energetic materials [76–78]. Examination of these materials using element-specific TEM clearly illustrates the intimate mixing of the distinct chemical components at the nanometer scale that can be achieved using this approach (Fig. 17.12). A similar strategy was utilized in the synthesis of interpenetrating metal oxide/carbonizable polymer aerogel networks that can then be converted into porous metal or carbide aerogels through carbothermal reduction [19, 34, 39, 41] (see ▶ Chap. 44). Much like in the synthesis of mixed metal oxide systems, aerogel nanocomposites can be readily prepared through the addition of a dispersed phase or phases to the sol–gel reaction mixture prior to gelation [79]. Because of the mild reaction conditions and the flexibility of the process, epoxide-assisted gelation allows for the incorporation of a wide variety of dispersed phases into an aerogel network. One area of research where this approach has been successfully demonstrated has been the design of energetic nanocomposites [49, 80] (see ▶ Chap. 47). Using epoxide-assisted gelation, energetic formulations have been prepared in which solid fuel

431

particles, such as aluminum metal, are dispersed within an oxidizing framework, such as iron oxide [61]. Systems such as these are synthesized through the gelation of hydrated iron (III) salt solutions in ethanol containing a suspension of nanometric aluminum powder. As the fibrous iron(III) oxide network forms, the insoluble aluminum phase is encapsulated and immobilized, leading to uniform dispersion of the aluminum particles within the oxide framework. The intimate mixing of the iron oxide and aluminum metal phases within the nanocomposite is shown in Fig. 17.13. Mixtures of this

17

a

50 nm b O Fe C

50 nm

Fig. 17.13 Transmission electron micrographs (TEM) of an iron(III) oxide aerogel/nanometric aluminum composite prepared using the epoxide-assisted gelation method: (a) bright-field image and (b) colorcoded element maps

432

T. F. Baumann et al.

type are known as thermites and undergo solid-state oxidation/reduction reactions that release a significant amount of energy at rates much higher than those observed for systems prepared through the mixing of micrometer-sized component phases [19, 77, 78]. The approach used for the fabrication of these energetic nanocomposites is also applicable to the incorporation of other nanometric dispersed phases, such as carbon nanotubes or catalytic metal particles, into the aerogel framework [36]. In another approach to energetic aerogel nanocomposites, metallic cobalt aerogels obtained via carbothermal reduction of epoxide-synthesized cobalt oxide have served as a host matrix for LiClO4. In this approach, LiClO4 is introduced into the metallic cobalt aerogel via melt infiltration, fully filling its porosity. The resulting Co/LiClO4 nanocomposite was then ignited with an electric resistor, reaching a measured temperature of over 1,500  C [16].

17.4

Summary

Epoxide-assisted gelation is a general and straightforward technique for the synthesis of a variety of metal oxide aerogels, xerogels, and nanocomposites. The process utilizes organic epoxides to aid in the hydrolysis and condensation of hydrated metal species provided from common metal salts. The epoxide approach serves as a complementary process to traditional alkoxide-based sol–gel chemistry and greatly expands the compositional range accessible by these unique materials. Through judicious selection of epoxide, metal salt, and reaction solvent, this synthetic methodology can be used to tailor the important structural characteristics of a sol–gelderived oxide material such as chemical composition, network morphology, and crystallinity that strongly influence the bulk physical properties of the resulting aerogel. In addition, this epoxide-assisted method is amenable to the synthesis of multicomponent and composite sol–gel systems with intimately mixed nano- and/or microstructures. Acknowledgments Portions of the work described herein were performed under the auspices of the US Department of Energy by Lawrence Livermore National Laboratory under contract DE-AC52-07NA27344. Other portions of the work described herein were funded by the Army Research Office (W911NF-14-1-0369, W911NF-12-2-0029, W911NF-10-1-0476) and the National Science Foundation (1530603, 0907291, 0809562, and 0653919) via awards to the Missouri University of Science & Technology (N.L).

References 1. Hüsing, N., Schubert, U.: Aerogels—airy materials: chemistry, structure, and properties. Angew. Chem. Int. Ed. 37, 22–47 (1998) 2. Pierre, A.C., Pajonk, G.M.: Chemistry of aerogels and their applications. Chem. Rev. 102, 4243–4265 (2002)

3. Pekala, R.W.: Organic aerogels from the polycondensation of resorcinol with formaldehyde. J. Mater. Sci. 24, 3221–3227 (1989) 4. Brinker, C.J., Scherer, G.W.: Sol–Gel Science: The Physics and Chemistry of sol–Gel Processing. Academic (1990) 5. Gash, A.E., Tillotson, T.M., Satcher, J.H., Poco, J.F., Hrubesh, L.W., Simpson, R.L.: Use of epoxides in the sol-gel synthesis of porous iron oxide monoliths from Fe(III) salts. Chem. Mater. 13, 999–1007 (2001) 6. Ziese, W.: Die Reaktion von Äthylenoxyd mit Lösungen von Erdund Schwermetallhalogeniden. Ein neur Weg zur Gewinnung von Solen und reversiblen Gelen von Metalloxydhydraten. Berichte der deutschen chemischen Gesellschaft. 66, 1965–1972 (1933) 7. Kearby, K., Kistler, S.S., Swann Jr., S.: Aerogel catalysts. Conversion of alcohol to amines. J. Ind. Eng. Chem. 30, 1082–1086 (1938) 8. Itoh, H., Tabata, T., Kokitsu, M., Okazaki, N., Imizu, Y., Tada, A.: Preparation of SiO2-Al2O3 gels from tetraethoxysilane and aluminum chloride – a new sol-gel method using propylene oxide as a gelation promoter. J. Ceram. Soc. Jpn. 101, 1081–1083 (1993) 9. Tillotson, T.M., Sunderland, W.E., Thomas, I.M., Hrubesh, L.W.: Synthesis of lanthanide and lanthanide-silicate aerogels. J. Sol-Gel Sci. Technol. 1, 241 (1994) 10. Jirgensons, B., Straumanis, M.E.: A Short Textbook of Colloid Chemistry. The MacMillan Company (1962) 11. Livage, J., Henry, M., Sanchez, C.: Sol-gel chemistry of transition metal oxides. Prog. Solid St. Chem. 18, 259 (1988) 12. Dobinson, B., Hoffman, W., Stark, B.P.: The Determination of Epoxide Groups. Permagon Press, Oxford (1969) 13. Gash, A.E., Satcher, J.H., Simpson, R.L.: Strong akaganeite aerogel monolith using epoxides: synthesis and characterization. Chem. Mater. 15, 3268–3275 (2003) 14. Baumann, T.F., Gash, A.E., Chinn, S.C., Sawvel, A.M., Maxwell, R. S., Satcher, J.H.: Synthesis of high-surface-area alumina aerogels without the use of alkoxide precursors. Chem. Mater. 17, 395–401 (2005) 15. Kucheyev, S.O., Baumann, T.F., Cox, C.A., Wang, Y.M., Satcher, J. H., Hamza, A.V., Bradby, J.E.: Nanoengineering mechanically robust aerogels via control of foam morphology. Appl. Phys. Lett. 89, 041911 (2006) 16. Rewatkar, P.M., Soni, R.U., Sotiriou-Leventis, C., Leventis, N.: A cobalt sunrise: thermites based on LiClO4-filled Co(0) aerogels prepared from polymer-crosslinked cobaltia xerogel powders. ACS Appl. Mater. Interfaces. 11, 22668–22676 (2019) 17. Peterson, G.R., Hung-Low, F., Gumeci, C., Bassett, W.P., Korzeniewski, C., Hope-Weeks, L.J.: Preparationmorphology-performance relationships in cobalt aerogels as supercapacitors. ACS Appl. Mater. Interfaces. 6, 1796–1803 (2014) 18. Greenwood, N.N., Earnshaw, A.: Chemistry of the elements, p. 1131. Butterworth-Heinemann, Woburn (1998) 19. Leventis, N., Chandrasekaran, N., Sadekar, A.G., Mulik, S., Sotiriou-Leventis, C.: The effect of compactness on the carbothermal conversion of interpenetrating metal oxide/resorcinol-formaldehyde nanoparticle networks to porous metals and carbides. J. Mater. Chem. 20, 7456–7471 (2010) 20. Garciamartinez, O., Millan, P., Rojas, R.M., Torralvo, M.J.: Cobalt basic salts as inorganic precursors of cobalt oxides and cobalt metal: thermal behavior dependence on experimental conditions. J. Mater. Sci. 23, 1334–1350 (1988) 21. Health and Safety Executive – REACH regulations. http://www.hse. gov.uk/reach. 12 Apr 2016 22. Arrufat, A.V., Budziszewska, M., Lopez, C., Nguyen, A., Sitek, J., Jones, P., Shaw, C., Hayes, I., Cairns, G., Leighton, G.: REACH compliant epoxides used in the synthesis of Fe(III)-based aerogel monoliths for target fabrication. High Power Laser Sci. Eng. 5(e24), 6 (2017) 23. Leventis, N., Vassilaras, P., Fabrizio, E.F., Dass, A.: Polymer nanoencapsulated rare earth aerogels: chemically complex but

17

Synthesis of Metal Oxide Aerogels via Epoxide-Assisted Gelation of Metal Salts

stoichiometrically similar core-shell superstructures with skeletal properties of pure compounds. J. Mater. Chem. 17(15), 1502–1508 (2007) 24. Gash, A.E., Tillotson, T.M., Satcher, J.H., Hrubesh, L.W., Simpson, R.L.: New sol-gel synthetic route to transition and main-group metal oxide aerogels using inorganic salt precursors. J. Non-Cryst. Solids. 285, 22–28 (2001) 25. Davis, M., Zhang, K., Wangb, S., Hope-Weeks, L.J.: Enhanced electrical conductivity in mesoporous 3D indium-tin oxide materials. J. Mater. Chem. 22, 20163 (2012) 26. Baumann, T.F., Kucheyev, S.O., Gash, A.E., Satcher, J.H.: Facile synthesis of a crystalline, high-surface-area SnO2 aerogel. Adv. Mater. 17, 1546–1548 (2005) 27. Da Cunha, C.R., Toffolo, G.H., Dos Santos, C.E.I., Pezzi, R.P.: Structural, optical and chemical characterizations of sol-gel grown tin oxide aerogels. J. Non-Cryst. Solids. 380, 48–52 (2013) 28. Correa Baena, J.P., Agrios, A.G.: Transparent conducting aerogels of antimony-doped tin oxide. ACS Appl. Mater. Interfaces. 6(21), 19127–19134 (2014) 29. Kucheyev, S.O., van Buuren, T., Baumann, T.F., Satcher, J.H., Willey, T.M., Meulenberg, R.W., Felter, T.E., Poco, J.F., Gammon, S.A., Terminello, L.J.: Electronic structure of titania aerogels from soft x-ray absorption spectroscopy. Phys. Rev. B. 69, 245102 (2004) 30. Kucheyev, S.O., Baumann, T.F., Wang, Y.M., van Buuren, T., Satcher, J.H.: Synthesis and electronic structure of low-density monoliths of nanoporous nanocrystalline anatase TiO2. J. Electron Spectrosc. Relat. Phenom. 144-147, 609–612 (2005) 31. Hong, I.: VOCs degradation performance of TiO2 aerogel photocatalyst prepared in SCF drying. J. Ind. Eng. Chem. (Seoul). 12(6), 918–925 (2006) 32. Fears, T.M., Sotiriou-Leventis, C., Winiarz, J.G., Leventis, N.: Economical synthesis of vanadia aerogels via epoxide-assisted gelation of VOCl3. J. Sol-Gel Sci. Technol. 77(1), 244–256 (2016) 33. Davis, M., Guemeci, C., Kiel, C., Hope-Weeks, L.J.: Preparation of porous manganese oxide nanomaterials by one-pot synthetic sol-gel method. J. Sol-Gel Sci. Technol. 58(2), 535–538 (2011) 34. Leventis, N., Chandrasekaran, N., Sotiriou-Leventis, C., Mumtaz, A.: Smelting in the age of nano: iron aerogels. J. Mater. Chem. 19, 63–65 (2009) 35. Wei, T., Chen, C., Chang, K., Lu, S., Hu, C.: Cobalt oxide aerogels of ideal supercapacitive properties prepared with an epoxide synthetic route. Chem. Mater. 21, 3228–3233 (2009) 36. Gill, S.K., Shobe, A.M., Hope-Weeks, L.J.: Synthesis of cobalt oxide aerogels and nanocomposite systems containing single-walled carbon nanotubes. Scanning. 31(3), 132–138 (2009) 37. Gash, A.E., Satcher, J.H., Simpson, R.L.: Monolithic nickel(II)based aerogels using an organic epoxide: the importance of the counterion. J. Non-Cryst. Solids. 350, 145–151 (2004) 38. Sisk, C.N., Hope-Weeks, L.J.: Copper(II) aerogels via 1,2-epoxide gelation. J. Mater. Chem. 18, 2607–2610 (2008) 39. Leventis, N., Chandrasekaran, N., Sadekar, A.G., Sotiriou-Leventis, C., Lu, H.: One-pot synthesis of interpenetrating inorganic/organic networks of CuO/resorcinol-formaldehyde aerogels: nanostructured energetic materials. J. Am. Chem. Soc. 131, 4576–4577 (2009) 40. Bi, Y., Ren, H., Zhang, L.: Synthesis of a low-density copper oxide monolithic aerogel using inorganic salt precursor. Adv. Mater. Res. 217–218(Pt. 2, High Performance Structures and Materials Engineering), 1165–1169 (2011) 41. Du, A., Zhou, B., Shen, J., Xiao, S., Zhang, Z., Liu, C., Zhang, M.: Monolithic copper oxide aerogel via dispersed inorganic sol-gel method. J. Non-Cryst. Solids. 355, 175–181 (2009) 42. Gao, Y.P., Sisk, C.N., Hope-Weeks, L.J.: A sol-gel route to synthesize monolithic zinc oxide aerogels. Chem. Mater. 19, 6007–6011 (2007) 43. Chervin, C.N., Clapsaddle, B.J., Chiu, H.W., Gash, A.E., Satcher, J. H., Kauzlarich, S.M.: Aerogel synthesis of yttria-stabilized zirconia by a non-alkoxide sol-gel route. Chem. Mater. 17, 3345–3351 (2005)

433

44. Chervin, C.N., Clapsaddle, B.J., Chiu, H.W., Gash, A.E., Satcher, J. H., Kauzlarich, S.M.: Role of cyclic ether and solvent in a non-alkoxide sol-gel synthesis of yttria-stabilized zirconia nanoparticles. Chem. Mater. 18, 4865–4874 (2006) 45. Eid, J., Pierre, A.C., Baret, G.: Preparation and characterization of transparent Eu doped Y2O3 aerogel monoliths, for application in luminescence. J. Non-Cryst. Solids. 351(3), 218–227 (2005) 46. Helmut, S., Susanne, B., Barbara, M., Sachar, I., Martin, S., Lorenz, R.: Zirconia-based aerogels via hydrolysis of salts and alkoxides: the influence of the synthesis procedures on the properties of the aerogels. Chem. Asian J. 8(9), 2211–2219 (2013) 47. Suh, D.J., Park, T., Kim, W., Hong, I.: Synthesis of high-surfacearea ruthenium oxide aerogels by non-alkoxide sol-gel route. J. Power Sources. 117, 1–6 (2003) 48. Ren, H., Zhang, L., Shang, C., Wang, X., Bi, Y.: Synthesis of a low-density tantalum oxide tile-like aerogel monolithic. J. Sol-Gel Sci. Technol. 53(2), 307–311 (2010) 49. Prentice, D., Pantoya, M.L., Gash, A.E.: Combustion wave speeds of sol-gel-synthesized tungsten trioxide and nano-aluminum: the effect of impurities on flame propagation. Energy Fuel. 20, 2370 (2006) 50. Laberty-Robert, C., Long, J.W., Lucas, E.M., Pettigrew, K.A., Stroud, R.M., Doescher, M.S., Rolison, D.R.: Sol-gel derived ceria nanoarchitectures: synthesis, characterization and electrical properties. Chem. Mater. 18, 50–58 (2006) 51. Clapsaddle, B.J., Neumann, B., Wittstock, A., Sprehn, D.W., Gash, A.E., Satcher Jr., J.H., Simpson, R.L., Baeumer, M.: Sol-gel methodology for preparation of lanthanide-oxide aerogel: preparation and characterization. J. Sol-Gel Sci. Technol. 64(2), 381–389 (2012) 52. Zhang, H.D., Li, B., Zheng, Q.X., Jiang, M.H., Tao, X.T.: Synthesis and characterization of monolithic Gd2O3 aerogels. J. Non-Cryst. Solids. 354, 4089–4093 (2008) 53. Bang, A., Sadekar, A.G., Buback, B., Curtin, B., Acar, S., Kolasinac, D., Yin, W., Rubenstein, D.A., Lu, H., Leventis, N., SotiriouLeventis, C.: Evaluation of Dysprosia aerogels as drug delivery systems: a comparative study with random and ordered mesoporous Silicas. ACS Appl. Mater. Interfaces. 6, 4891–4902 (2014) 54. Zhang, Z., Guo, D., Yang, X., Zhang, J.: Calcination of ytterbia aerogels leads to ferromagnetic nanoporous ytterbium oxide networks. Mater. Lett. 261, 126866 (2020) 55. Reibold, R.A., Poco, J.F., Baumann, T.F., Simpson, R.L., Satcher, J. H.: Synthesis and characterization of a nanocrystalline thoria aerogel. J. Non-Cryst. Solids. 341, 35–39 (2004) 56. Reibold, R.A., Poco, J.F., Baumann, T.F., Simpson, R.L., Satcher, J. H.: Synthesis and characterization of a low-density urania (UO3) aerogel. J. Non-Cryst. Solids. 319, 241–246 (2003) 57. Gan, L., Xu, Z., Feng, Y., Chen, L.: Synthesis of alumina aerogels by ambient drying method and control of their structures. J. Porous. Mater. 12, 317–321 (2005) 58. Tokudome, Y., Fujita, K., Nakanishi, K., Miura, K., Hirao, K.: Synthesis of monolithic Al2O3 with well-defined macropores and mesostructured skeletons via the sol-gel process accompanied by phase separation. Chem. Mater. 19, 3393–3398 (2007) 59. Hund, J.F., McElfresh, J., Frederick, C.A., Nikroo, A., Greenwood, A.L., Luo, W.: Fabrication and characterization of aluminum oxide aerogel backlighter targets. Fusion Sci. Tech. 51, 701–704 (2007) 60. Tokudome, Y., Nakanishi, K., Hanada, T.: Effect of La addition on thermal microstructural evolution of macroporous alumina monolith prepared from ionic precursors. J. Ceram. Soc. Jpn. 117, 351–355 (2009) 61. Tillotson, T.M., Gash, A.E., Simpson, R.L., Hrubesh, L.W., Satcher, J.H., Poco, J.F.: Nanostructured energetic materials using sol-gel methodologies. J. Non-Cryst. Solids. 285, 338–345 (2001) 62. Prakash, A., McCormick, A.V., Zachariah, M.R.: Aero-sol-gel synthesis of nanoporous iron oxide particles: a potential oxidizer for nanoenergetic materials. Chem. Mater. 16, 1466–1471 (2004)

17

434 63. Leventis, N., Donthula, S., Mandal, C., Ding, M.S., SotiriouLeventis, C.: Explosive versus thermite behavior in iron (0) aerogels infiltrated with perchlorates. Chem. Mater. 27, 8126–8137 (2015) 64. Mahadik-Khanolkar, S., Donthula, S., Bang, A., Wisner, C., Sotiriou-Leventis, C., Leventis, N.: Polybenzoxazine aerogels. 2. Interpenetrating networks with iron oxide and the Carbothermal synthesis of highly porous monolithic pure iron(0) aerogels as energetic materials. Chem. Mater. 26, 1318–1331 (2014) 65. Long, J.W., Logan, M.S., Rhodes, C.P., Carpenter, E.E., Stroud, R. M., Rolison, D.R.: Nanocrystalline iron oxide aerogels as mesoporous magnetic architectures. J. Am. Chem. Soc. 126, 16879–16889 (2004) 66. Park, C., Magana, D., Stiegman, A.E.: High-quality Fe and γ-Fe2O3 magnetic thin films from an epoxide-catalyzed sol-gel process. Chem. Mater. 19, 677–683 (2007) 67. Carpenter, E.E., Long, J.W., Rolison, D.R., Logan, M.S., Pettigrew, K., Stroud, R.M., Kuhn, L.T., Hansen, B.R., Mørup, S.: Magnetic and Mössbauer spectroscopy studies of nanocrystalline iron oxide aerogels. J. Appl. Phys. 99, 08N711 (2006) 68. Cui, H., Ren, W.: Low temperature and size controlled synthesis of monodispersed γ-Fe2O3 nanoparticles by an apoxide assisted sol-gel route. J. Sol-Gel Sci. Technol. 47, 81–84 (2008) 69. Bali, S., Huggins, F.E., Huffman, G.P., Ernst, R.D., Pugmire, R.J., Eyring, E.M.: Iron aerogel and xerogel catalysts for Fischer-Tropsch synthesis of diesel fuel. Energy Fuel. 23, 14–18 (2009) 70. Bali, S., Turpin, G.C., Ernst, R.D., Pugmire, R.J., Singh, V., Seehra, M.S., Eyring, E.M.: Water gas shift catalysis using iron aerogels doped with palladium by the gas-phase incorporation method. Energy Fuel. 22, 1439–1443 (2008) 71. Chen, L., Zhu, J., Liu, Y., Cao, Y., Li, H., He, H., Dai, W., Fan, K.: Photocatalytic activity of epoxide sol-gel derived titania transformed into nanocrystalline aerogel powders by supercritical drying. J. Mol. Catal. A. 255, 260–268 (2006) 72. Chervin, C.N., Clapsaddle, B.J., Chiu, H.W., Gash, A.E., Satcher, J. H., Kauzlarich, S.M.: A non-alkoxide sol-gel method for the preparation of homogeneous nanocrystalline powders of La0.85Sr0.15MnO3. Chem. Mater. 18, 1928–1937 (2006) 73. Brown, P., Hope-Weeks, L.J.: The synthesis and characterization of zinc ferrite aerogels prepared by epoxide addition. J. Sol-Gel Sci. Technol. 51, 238–243 (2009) 74. Cui, H., Zayat, M., Levy, D.: Sol-gel synthesis of nanoscaled spinels using propylene oxide as a gelation agent. J. Sol-Gel Sci. Technol. 35, 175–181 (2005) 75. Guo, Y., Meyer-Zaika, W., Muhler, M., Vukojevic, S., Epple, M.: Cu/Zn/Al xerogels and aerogels prepared by a sol-gel reaction as catalysts for methanol synthesis. Eur. J. Inorg. Chem. 23, 4774–4781 (2006) 76. Clapsaddle, B.J., Gash, A.E., Satcher, J.H., Simpson, R.L.: Silicon oxide in an iron(III) oxide matrix: the sol-gel synthesis and characterization of Fe-Si mixed oxide nanocomposites that contain iron oxide as a major phase. J. Non-Cryst. Solids. 331, 190–201 (2003) 77. Clapsaddle, B.J., Sprehn, D.W., Gash, A.E., Satcher, J.H., Simpson, R.L.: A versatile sol-gel synthesis route to metal-silicon mixed oxide nanocomposites that contain metal oxides as the major phase. J. Non-Cryst. Solids. 350, 173–181 (2004) 78. Zhao, L., Clapsaddle, B.J., Satcher, J.H., Schaefer, D.W., Shea, K.J.: Integrated chemical systems: the simultaneous formation of hybrid nanocomposites of iron oxide and organo silsesquioxane. Chem. Mater. 17, 1358–1366 (2005)

T. F. Baumann et al. 79. Morris, C.A., Anderson, M.L., Stroud, R.M., Merzbacher, C.I., Rolison, D.R.: Silica sol as a nanoglue: flexible synthesis of composite aerogels. Science. 284, 622 (1999) 80. Plantier, K.B., Pantoya, M.L., Gash, A.E.: Combustion wave speeds of nanocomposite Al/Fe2O3: the effects of Fe2O3 particle synthesis techniques. Combust. Flame. 140, 299 (2005)

Theodore Baumann received his BS (1990) in chemistry at Middlebury College and PhD (1996) in inorganic chemistry at Northwestern University. After a postdoctoral appointment at the University of Michigan, he joined Lawrence Livermore National Laboratory in 1997 as a postdoctoral researcher. Recently, he was the group leader of the Advanced Materials Synthesis group before his current position as a director of the Physical and Life Sciences Postdoc Program. Dr. Baumann’s research interests are inorganic and organic synthesis in areas related to catalysis and materials and polymer chemistry. Current projects include the synthesis of novel organic and carbon aerogel systems as well as new inorganic sol–gel materials for a variety of applications.

Nicholas Leventis received his Ph.D. from Michigan State University in organic chemistry in 1985, and was a postdoctoral associate at MIT (1985–1988) focusing on electrochemistry. His subsequent independent career started in the private sector (Igen, Inc.), and eventually spanned all three sectors of the economy: industry, academia, and the Government (NASA). In 2019 he retired from the Missouri University of Science and Technology as a Curators’ Distinguished Professor of Chemistry, and he is currently the Director of Research at Aspen Aerogels in Northborough, MA. His aerogel work includes polymercrosslinked aerogels, polymer aerogels from most classes of organic polymers, interpenetrating organic-inorganic aerogels, and metallic, ceramic, and carbon aerogels. Professor Leventis has received numerous awards, most notably the NASA Exceptional Scientific Achievement Medal (2005). He has over 200 scientific publications and over 30 patents. His published work has received over 10,000 citations. His H-index is currently 54.

17

Synthesis of Metal Oxide Aerogels via Epoxide-Assisted Gelation of Metal Salts

Dr. Stephen A. Steiner III is the President, CEO, and founder of Aerogel Technologies, LLC, a leading aerogel manufacturer. Steiner holds a Ph.D. in Materials Chemistry and Engineering from MIT’s Department of Aeronautics and Astronautics, an SM in Materials Science and Engineering from MIT, and a BS in Chemistry Course from the University of Wisconsin–Madison. He is an accomplished nanomaterials researcher, with expertise in aerogels, nanocarbons, and aerospace materials.

435

17

High Temperature Oxide Aerogels

18

Frances I. Hurwitz, Haiquan Guo, Richard B. Rogers, Nathaniel Olson, and Anita Garg

Contents 18.1 18.1.1 18.1.2

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437 Considerations in System Selection . . . . . . . . . . . . . . . . . . . . . . 438 Synthesis Routes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438

18.2 18.2.1 18.2.2 18.2.3

Alumina and Alumina Silicate Aerogels . . . . . . . . . . . . . . . Synthesis from AlCl3 and Epoxide . . . . . . . . . . . . . . . . . . . . . . . Synthesis from Boehmite Nanocrystalline Powders . . . . . Phase Changes on Thermal Exposure of Boehmite-Derived Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

438 439 439 442

18.3

Alumina-Titania Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445

18.4

Alumina-Silica-Zirconia Aerogels . . . . . . . . . . . . . . . . . . . . . . 445

18.5

Yttria-Alumina Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446

18.6 18.6.1 18.6.2

Yttria-Stabilized Zirconia (YSZ) Aerogels [60] . . . . . . . 448 YSZ Aerogel Synthesis from ZrOCl2, YCl3.6H2O, and Propylene Oxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449 YSZ Aerogel Synthesis Using Citric Acid Templating . . 451

18.7

Aerogel Composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 454

18.8

Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455

Abstract

Numerous applications from insulation to catalytic supports and fuel cells can benefit from lightweight, high surface area, mesoporous materials which maintain their mesoporous structure to temperatures of 600 to 1200
C. F. I. Hurwitz (*) · R. B. Rogers · N. Olson NASA Glenn Research Center, Cleveland, OH, USA e-mail: [email protected]; [email protected]; [email protected] H. Guo Ohio Aerospace Institute, NASA Glenn Research Center, Cleveland, OH, USA e-mail: [email protected] A. Garg University of Toledo, Toledo, OH, USA e-mail: [email protected]

Polymeric aerogels are limited to temperatures of nominally 400
C due to thermal degradation of organic groups. Silica aerogels begin to densify by 700
C. A number of aerogel systems show stability at higher temperatures, including alumina, alumina silicates, yttriumdoped alumina, and zirconia and yttria-stabilized zirconia aerogels. Within a given chemical composition, the morphology and textural stability of a mesoporous structure is dependent upon the synthesis method used. Other important considerations in choosing an aerogel composition include the time at temperature required for a given application, phase transformations inherent in a given system, and approaches to phase stabilization such as introduction of dopants into the backbone structure. Reinforcement of the aerogel through a composite approach also is addressed. Keywords

Alumina · Aluminosilicate · Yttria stabilized zirconia · Titania-doped alumina · Yttria-doped alumina · Thermal stability

18.1

Introduction

Aerogels are noted for their extremely light weight and mesoporous structure [1, 2]. They are formed by polymerization around a solvent, which is then removed in a manner that averts collapse of the fine-sized pores, replacing the solvent with air [2, 3]. Their structure renders them exceptionally good insulators, as solid conduction must follow a long and tortuous path following the pore walls which may take the form of needle-like struts, and their very small pore sizes inhibit gas convection. Their thermal conductivity may be decreased even further by doping with thermal radiation opacifiers which can vary infrared reflectance and/or absorbance. Aerogels can be formed using a wide range of

© Springer Nature Switzerland AG 2023 M. A. Aegerter et al. (eds.), Springer Handbook of Aerogels, Springer Handbooks, https://doi.org/10.1007/978-3-030-27322-4_18

437

438

elements. The most widely studied system has been that of silica [4] and carbon [5] aerogels. There also has been an increasing focus on polymeric aerogels, which is addressed in other chapters in this Handbook. Polymeric aerogels are limited in upper use temperature by shrinkage and thermal decomposition, generally below 400
C. Silica aerogels begin to sinter at 650–700
C, at which point their porous structure begins to collapse, resulting in an increase in thermal conductivity. Carbon aerogels [5] are stable to much higher temperatures, but only in nonoxidizing environments. There is growing interest in developing aerogels for use at higher temperatures and for defining the time/temperature stability for different aerogel chemistries.

F. I. Hurwitz et al.

15], These methods can include addition of rare earth dopants [16]. However, many of these approaches tend to produce porous solids or rod-like structures rather than high surface area aerogels. These oxides are reported to have high homogeneity, suppressing their phase separation and crystallization on thermal exposure [17, 18]. A large focus of this work is in the synthesis of catalysts [19]. In some non-hydrolytic syntheses porosity can be increased by burning out organic groups [20]. An exception seems to by the non-hydrolytic synthesis of silica, which has been shown to produce high surface area materials [21].

18.2 18.1.1 Considerations in System Selection There are a number of candidate inorganic chemistries for high temperature application, including yttria-stabilized zirconia (YSZ), yttria-alumina (YAG), alumina, alumina silicate, and ternary alumina-silica-zirconia. Most of these exhibit low thermal conductivity and have been used successfully in thermal barrier coatings. In all cases, the aerogel texture varies with the synthesis approach and drying conditions. All of these systems will undergo phase transformations with time and temperature, accompanied by shrinkage, changes in surface area and pore size distribution, and crystalline phase transformation. The driving considerations in selecting a system are use temperature and time (does the application require seconds, minutes, hours or years at temperature) and chemical compatibility with other materials.

18.1.2 Synthesis Routes Many of the synthetic approaches react soluble precursors, with interaction at the molecular level; other approaches start with small-sized crystalline powders, and form the aerogel backbone by a self-assembly of crystallites. The majority of the systems discussed here use a sol–gel route in which an alkoxide or halide precursor is hydrolyzed, typically with an acid or base catalyst, and then undergoes condensation to form a hydrogel or alcogel [1, 6, 7], which then may be dried using supercritical or ambient approaches [1]. Epoxides have been used to achieve high surface area gels in a variety of systems using chloride salt precursors [8– 11]. Addition of epoxides to achieve gelation can be exothermic, but can be controlled by cooling the sol prior to adding the epoxide. Gels with epoxide tend to be stronger than those using other methods, but incorporate organic groups, leading to significant shrinkages on heat treatment. Non-hydrolytic sol–gel routes, in which oxide bridges are formed using donors other than water, also are possible [12–

Alumina and Alumina Silicate Aerogels

The structure of alumina aerogels depends strongly on the precursor, synthesis method, and drying conditions, some of which utilize high-temperature supercritical drying [22– 25]. Yoldas and coworkers [26–29] developed alumina aerogels utilizing aluminum alkoxide precursors which can undergo chemical polymerization, and which result in gels that differ in texture from those formed by the interaction of metallic salts to form colloidal suspensions. By controlling hydrolysis temperature they were able to form either Bayerite or amorphous sols [27]. The alkoxide-derived gels typically are considered to form pseudoboehmite, which has a higher proportion of bound water and less long-range order than crystalline Boehmite. The method leads to incorporation of organic groups, giving rise to large shrinkage on pyrolysis. These gels transform to α-alumina, losing their porous structure, at temperatures above 1150
C [30]. In aluminum sec butoxide-derived gels, a series of transitional aluminas are formed on thermal exposure, with eventual formation of α-alumina, accompanied by a loss of surface area [31, 32]. The phase transformations follow those observed with heat treatment of Boehmite and Bayerite [33]. Mizushima and Hori [34] compared the thermal phase stability of alumina aerogels of sec butoxide derived alumina aerogels that were ambiently dried, supercritically dried using CO2, and dried in an autoclave directly from ethanol, and showed earlier transformation to θ-alumina, but later transformation to α in the autoclave dried material. The higher density xerogels transformed to α the most rapidly. Incorporating silica into the alumina network can produce alumina silicates with varying structures depending upon the alumina precursor [24]. Reaction with silica results in incorporation of tetrahedral alumina groups, with the degree of tetrahedral incorporation related to the water/Si mol ratio [35]. The extent of alumina incorporation varies with choice of precursors [24]. By doping the alumina network structure, the transformation to α-alumina can be suppressed. Mizushima and Hori [31] reacted aluminum sec butoxide with tetraethoxyorthosilicate (TEOS), and were able to

High Temperature Oxide Aerogels

439

maintain a surface area of >100 m2/g after exposure at 1200
C for 5 h. Horiuchi et al. [36] showed similar results using aluminum tri-isopropoxide and TEOS. Popa et al. [37] used the Yoldas alkoxide method [29] to synthesize pseudoboehmite-silica gels at an Al:Si ratio of 15.6:1 and dried them by either ambient and supercritical carbon dioxide (CO2) techniques to produce either xerogels or higher surface area aerogels. The authors demonstrated an increased incorporation of silicon into the alumina lattice for the supercritically dried, higher surface area aerogel, and that this silicon incorporation suppressed phase transformation to α-alumina on heating to 1200
C for short times. Studies of alumina silicate aerogels by Roy and Komarneni and coworkers have shown variation in sintering behavior based on the presence of diphasic or single phase systems [38–41].

Work in our laboratory [42] has compared the thermal stability of alumina and alumnosilicate aerogels synthesized using the epoxide method of Clapsaddle et al. [10], reacting AlCl3.6H2O and TEOS and propylene oxide (PO), with a colloidal synthesis using Boehmite powders of various crystallite sizes and TEOS to compare textural properties and phase stability on thermal exposure. All aerogels were synthesized in absolute ethanol, and supercritically dried using CO2.

+ HCl

CH3

H+ O ClCH3

12 14 16 18 20 22 24 26 1.0

1.5 2.0 2.5 3.0 3.5 4.0 4.5 Log(Aging, days) Leverage, P 650 m2/g and maintained high surface areas after thermal exposure at 600
C, making these systems candidates for intermediate temperature applications. However, surface area was reduced to 75 m2/g, less than the nanocrystalline Boehmite all alumina aerogel, after 18 min at 1100
C. It also was hard to maintain formulated stoichiometry in this system.

455

1Si:1Zr aerogels performed slightly better, again exhibiting surface areas of 400 m2/g at 600
C and 120 m2/g at 1100
C 18-min exposure, at which point they transformed to 60% tetragonal zirconia and 38% tridymite (SiO2), making this formulation of interest for intermediate temperature usage. The yttrium-doped alumina aerogels synthesized using the epoxide method also were characterized by high surface areas of 700–800 m2/g as supercritically dried, and maintaining surface areas of 500–570 m2/g after 10 h at 600
C. However, at 1100
C the surface areas were well below those of the nanocrystalline Boehmite alumina silicate aerogels, making the latter better candidates at very high temperatures. There was a stark contrast in the composition of YSZ aerogels synthesized using the epoxide approach as compared with those prepared using citric acid templating. In the former method, a range of yttrium doping levels was achieved, and the formulated levels closely matched the measured yttrium incorporation. At 10% Y doping levels formation of monoclinic ZrO2 was suppressed completely in favor of tetragonal and monoclinic phases, but no YSZ phase was observed. By controlling the ratio of water to metals, aerogels were produced which maintained small particle sizes after 1100
C exposure. The citric acid templated YSZ, in contrast, incorporated much lesser amounts of yttrium, but did lead to formation of a YSZ phase at 1100
C. More of the YSZ phase was seen with YCl3 compared with Y(NO3)3, and particles remain relatively small after thermal exposure. The templating method avoids the use of possibly toxic epoxides, but produces a very different aerogel morphology. Depending on the application, use temperature, and anticipated time at temperature, there are many approaches to attaining high surface area aerogels for high temperature use. The nanocrystalline Boehmite-derived alumina silicate system offers the highest surface areas for longer term exposures at temperatures to 1200
C. Acknowledgments The author wishes to thank Dereck Johnson and the many student interns who have contributed to our understanding of high temperature aerogels.

References 1. Pierre, A.C., Pajonk, G.M.: Chemistry of aerogels and their applications. Chem. Rev. 102, 4243–4265 (2002) 2. Fricke, J., Emmerling, A.: Aerogels. J. Am. Ceram. Soc. 75, 2027–2036 (1992)

18

456 3. Scherer, G.W.: Theory of drying. J. Am. Ceram. Soc. 73, 3–14 (1990) 4. Pajonk, G.M.: Transparent silica aerogels. J. Non-Cryst. Solids. 225, 307–314 (1998) 5. Antonietti, M., Fechler, N., Fellinger, T.-P.: Carbon aerogels and monoliths: control of porosity and nanoarchitecture via sol–gel routes. Chem. Mater. 26, 196–210 (2014) 6. Brinker, C.J.: Hydrolysis and condensation of silicates: effects on structure. J. Non-Cryst. Solids. 100, 31–50 (1988) 7. Hench, L., West, J.: The sol-gel process. J. Chem. Rev. 90, 33–72 (1990) 8. Gash, A.E., Tillotson, T.M., Satcher Jr., J.H., Hrubesh, L.W., Simpson, R.L.: New sol-gel route to transition and main-group metal oxide aerogels using inorganic salt precursors. J. Non-Cryst. Solids. 285, 22–28 (2001) 9. Baumann, T.F., Gash, A.E., Chinn, S.C., Sawvel, A.M., Maxwell, R.S., Satcher, J.J.H.: Synthesis of high-surface-area alumina aerogels without the use of alkoxide precursors. Chem. Mater. 17, 395–401 (2005) 10. Clapsaddle, B.J., Sprehn, D.W., Gash, A.E., Satcher, J.J.H., Simpson, R.L.: A versatile sol-gel synthesis route to metal-silicon mixed oxide nanocomposites that contain metal oxides as the major phase. J. Non-Cryst. Solids. 350, 173–181 (2004) 11. Itoh, H., Tabata, T., Kokitsu, M., Okazaki, N., Imizu, Y., Tada, A.: Preparation of SiO2-Al2O3 gels from tetraethoxysilane and aluminum-chloride – a new sol-gel method using propylene oxide as a gelation promoter. J. Ceram. Soc. Jpn. 101, 1081–1083 (1993) 12. Acosta, S., Corriu, R.J.P., Leclerq, D., Lefèvre, P., Mutin, P.H., Vioux, A.: Preparation of alumina aerogels by non-hydrolytic sol-gel processing method. J. Non-Cryst. Solids. 71, 234–242 (1994) 13. Corriu, R., Leclercq, D., Lefèvre, P., Mutin, P.H., Vioux, A.: Preparation of monolithic binary oxide gels by a nonhydrolytic sol-gel process. Chem. Mater. 4, 961–963 (1992) 14. Mutin, P.H., Vioux, A.: Nonhydrolytic processing of oxide-based materials: simple routes to control homogeneity, morphology, and nanostructure. Chem. Mater. 21, 582–596 (2009) 15. Vioux, A.: Nonhydrolytic sol-gel routes to oxides. Chem. Mater. 9, 2292–2299 (1997) 16. Nassar, E.J., Avilia, L.R., Pereira, P.F.S., Mello, C., de Lima, O.J., Ciuffi, K.J., Carlos, L.D.: Eu(III) incorporation in sol–gel aluminum–yttrium matrix by non-hydrolytic route. J. Lumin. 111, 159–166 (2005) 17. Andrianainarivelo, M., Corriu, R., Leclercq, D., Mutin, P.H., Vioux, A.: Mixed oxides Si02-Zr02 and Si02-Ti02 by a non-hydrolytic sol-gel route. J. Mater. Chem. 6, 1665–1671 (1996) 18. Andrianainarivelo, M., Corriu, R., Leclercq, D., Mutin, P.H., Vioux, A.: Nonhydrolytic sol-gel process: aluminum titanate gels. Chem. Mater. 9, 1098–1102 (1997) 19. Debecker, D.P., Hulea, V., Mutin, P.H.: Mesoporous mixed oxide catalysts via non-hydrolytic sol–gel: a review. Appl. Catal. A Gen. 451, 192–206 (2013) 20. Hay, J.N., Porter, D., Raval, H.M.: A versatile route to organically modified silicas and porous silicas via the non-hydrolytic sol-gel process. J. Mater. Chem. 10, 1811–1818 (2000) 21. Bourget, L., Corriu, R., Leclercq, D., Mutin, P.H., Vioux, A.: Non-hydrolytic sol-gel routes to silica. J. Non-Cryst. Soids. 242, 81–91 (1998) 22. Pierre, A.C., Elaloui, E., Pajonk, G.M.: Comparison of the structure and porous texture of alumina gels synthesized by different methods. Langmuir. 14, 66–73 (1998) 23. MacKenzie, J.D., Bescher, E.P.: Chemical routes in the synthesis of nanomaterials using the sol-gel process. Acc. Chem. Res. 40, 810–818 (2007) 24. Sinkó, K., Hsing, N., Goerigk, G., Peterlik, H.: Nanostructure of gel-derived aluminosilicate materials. Langmuir. 24, 949–956 (2008) 25. Yoldas, B.E.: Hydrolytic polycondenstaion of Si(OC2H5)4 and effect of reaction parameters. J. Non-Cryst. Solids. 83, 375–390 (1986)

F. I. Hurwitz et al. 26. Partlow, D., Yoldas, B.E.: Colloidal versus polymer gels and monolithic transformation in glass-forming systems. J. Non-Cryst. Solids. 46, 153–161 (1981) 27. Yoldas, B.E.: Hydroylsis of aluminum alkoxides and bayerite conversion. J. Appl. Chem. Biotechnol. 23, 803–809 (1973) 28. Yoldas, B.E.: Alumina gels from porous transparent Al2O3. J. Mater. Sci. 10, 1856–1860 (1975) 29. Yoldas, B.: Alumina sol preparation from alkoxides. Am. Ceram. Soc. Bull. 54, 289–290 (1975) 30. Yoldas, B.: A transparent porous alumina. Am. Ceram. Soc. Bull. 54, 286–288 (1975) 31. Mizushima, Y., Hori, M.: Preparation of heat-resistant alumina aerogels. J. Mater. Res. 8, 2993–2999 (1993) 32. Kayser, S., Shter, G., de Hazan, Y., Cohen, Y., Grader, G.S.: Heat treatment of alumina aerogels. Chem. Mater. 9, 2464–2467 (1997) 33. Pecharromán, C., Sobrados, I., Iglesias, J.E., González-Correño, T., Sanz, J.: Thermal evolution of transitional aluminas followed by NMR and IR spectroscopies. J. Phys. Chem. B. 103, 6160–6170 (1999) 34. Mizushima, Y., Hori, M.: Properties of alumina aerogels prepared under different conditions. J. Non-Cryst. Solids. 167, 1–8 (1994) 35. Sinkó, K., Mezei, R., Rohonczy, J., Fratzl, P.: Gel structures containing Al(III). Langmuir. 15, 6631–6636 (1999) 36. Horiuchi, T., Osaki, T., Sugiyama, T., Suzuki, K., Mori, T.: Maintenance of large surface arrea of alumina heated at elevated temperatures above 1300
C by preparing silica-containing pseudoboehmite aerogel. J. Non-Cryst. Solids. 291, 187–198 (2001) 37. Popa, A.F., Rossignol, S., Kappenstein, C.: Influence of drying technique on silicon insertion into C-alumina and consequences for the homogeneity and thermal stability of silica–alumina aquagels. J. Mater. Chem. 12, 2866–2868 (2002) 38. Hoffman, D.W., Roy, R., Komarneni, S.: Diphasic xerogels, a new class of materials: phases in the system Al2O3-SiO2. J. Am. Ceram. Soc. 67, 468–471 (1984) 39. Komarneni, S., Suwa, Y., Roy, R.: Application of compositionally diphasic gels for enhanced densification: the system Al2O3-SiO2. J. Am. Ceram. Soc. 69, C-155–C-156 (1986) 40. Komarneni, S., Rutiser, C.: Single-phase and diphasic aerogels and xerogels of mullite: preparation and characterization. J. Eur. Ceram. Soc. 16, 143–147 (1996) 41. Komarneni, S., Roy, R., Selvaraj, U., Malla, P.B., Brevel, E.: Nanocomposite aerogels: the SiO2–Al2O3 system. J. Mater. Res. 8, 3163–3167 (1993) 42. Hurwitz, F.I., Gallagher, M., Olin, T.C., Shave, M.K., Ittes, M.I., Olafson, K.N., Fields, M.G., Rogers, R.B., Guo, H.: Optimization of alumina and aluminosilicate aerogel structure for high temperature performance. Int. J. Appl. Glas. Sci. 5, 276–286 (2014) 43. Komarneni, S., Roy, R.: Solid state 27Al and 29Si magic-angle spinning NMR of aluminosilicate gels. J. Am. Ceram. Soc. 69, C42–C44 (1986) 44. Aravind, P.R., Munkundan, P., Pillai, O.K., Warrier, K.G.K.: Mesoporous silica-alumina aerogels with high thermal pore stability through hybrid sol-gel route followed by subcritical drying. Microporous Mesoporous Mater. 96, 14–20 (2004) 45. Klug, F.J., Prochazka, S., Doremus, H.: Alumina-silica phase diagram in mullite region. J. Am. Ceram. Soc. 70, 750–759 (1987) 46. Brühne, S., Gottleib, S., Assmus, W., Alig, E., Schmidt, M.U.: Atomic structure analysis of nanocrystalline boehmite AlO(OH). Cryst. Growth Des. 8, 489–493 (2008) 47. Padmaja, P., Warrier, K.G.K., Padmanabhan, M., Wunderlich, W., Berry, F.J., Mortimer, M., Creamer, N.J.: Structural aspects and porosity features of nano-size high surface area alumina-silica mixed oxide catalyst generated through hybrid sol-gel route. Mater. Chem. Phys. 95, 56–61 (2006) 48. Hurwitz, F.I., Rogers, R.B., Guo, H., Yu, K., Domanowski, J., Schmid, E., Fields, M.G.: The role of phase changes in maintaining

18

High Temperature Oxide Aerogels

pore structure on thermal exposure of aluminosilicate aerogels. MRS Commun. 7, 642–650 (2017) 49. Wefers, K., Misra, C.: Oxides and Hydroxides of Aluminum. Alcoa Research Laboratories (1987) 50. Boumaza, A., Favaro, L., Lédion, J., Sattonnay, G., Brubach, J.B., Berthet, P., Huntz, A.M.: Transition alumina phases induced by heat treatment of boehmite: an x-ray diffraction and infrared spectroscopy study. J. Solid State Chem. 182, 1171–1176 (2009) 51. Krodikis, X., Raybaud, P., Gobichon, A.-E., Rebours, B., Euzen, P., Toulhoat, H.: Theoretical study of the dehydration process of boehmite to γ-alumina. J. Phys. Chem. B. 105, 5121–5130 (2001) 52. Pakharukova, V.P., Shalygin, A.S., Gerasimov, E.Y., Tsybulya, S.V., Martyanov, O.N.: Structure and morphology evolution of silicamodified pseudoboehmite aerogels during heat treatment. J. Solid State Chem. 233, 294–302 (2016) 53. Ananthakumar, S., Juyansankar, M., Warrier, K.G.K.: Microstructural, mechanical and thermal characterisation of sol–gel-derived aluminium titanate–mullite ceramic composites. Acta Mater. 54, 2965–2973 (2006) 54. Hurwitz, F.I., Guo, H., Rogers, R.B., Sheets, E.J., Miller, D.R., Newlin, K.N., Shave, M.K., Palczer, A.R., Cox, M.T.: Influence of Ti addition on boehmite-derived aluminum silicate aerogels: structure and properties. J. Sol-Gel Sci. Technol. 64, 756–764 (2012) 55. Popa, M., Calderón-Moreno, J.M., Popescu, L., Kakihana, M., Torecillas, R.: Crystallization of gel-derived and quenched glasses in the ternary oxide Al2O3–ZrO2–SiO2 system. J. Non-Cryst. Solids. 297, 290–300 (2002) 56. Popa, M., Kakihana, M., Yoshimura, M., Calderón-Moreno, J.M.: Zircon formation from amorphous powder and melt in the silica-rich region of the alumina–silica–zirconia system. J. Non-Cryst. Solids. 352, 5663–5669 (2006) 57. Fryer, J.R., Hutchison, J.L., Paterson, R.: Study of the hydrolysis products of zirconyl chloride. J. Colloid Interface Sci. 34, 238–248 (1970) 58. Al-Yassir, N., Le Van Mao, R.: Thermal stability of alumina aerogel doped with yttrium oxide, used as a catalyst support for the thermocatalytic cracking (Tcc) process: an investigation of its textural and structural properties. Appl. Catal. A Gen. 317, 275–283 (2007) 59. Ponthieu, E., Grimblot, J., Elaloui, E., Pajonk, G.M.: Synthesis and characterization of pure and yttrium-containing alumina aerogels. J. Mater. Chem. 3, 287–293 (1993) 60. Hurwitz, F.I., Olson, N., Guo, H., Rogers, R.B., Phan, D.: YttriaStabilized Zirconia Aerogels for High Temperature Applications: The Role of Synthesis Approaches on Pore Structure after Thermal Exposure, Manuscript in preparation, (2018) 61. Jones, R.I., Mess, D.: Improved tetragonal phase stability at 1400
C with Scandia, yttria-stabilized zirconia. Surf. Coat. Technol. 86–87, 94–101 (1996) 62. Miller, R.A., Smialek, J.L., Garlick, R.G.: Phase stability in plasmasprayed, partially stabilized zirconia-yttria. In: Heuer, A.H., Hobbs, L.W. (eds.) Science Nd Technology of Zirconia, Advances in Ceramics, vol. 3, pp. 241–253. The American Ceramic Society, Columbus (1981) 63. Chervin, C.N., Clapsaddle, B.J., Chiu, H.W., Gash, A.E., Satcher, J.J. H., Kauzlarich, S.M.: Aerogel synthesis of yttria-stabilized zirconia by a non-alkoxide sol-gel route. Chem. Mater. 17, 3345–3351 (2005) 64. Chervin, C.N., Clapsaddle, B.J., Chiu, H.W., Gash, A.E., Satcher, J. J.H., Kauzlarich, S.M.: Role of cyclic ether and solvent in a non-alkoxide sol-gel synthesis of yttria-stabilized zirconia nanoparticles. Chem. Mater. 18, 4865–4874 (2006) 65. Barnardo, T., Hoydalsvik, K., Winter, R., Martin, C.M., Clark, G.F.: In situ double anomalous small-angle x-ray scattering of the sintering and calcination of sol-gel prepared yttria-stabilized-zirconia ceramics. J. Phys. Chem. C. 113, 10021–10028 (2009)

457 66. Chao, X., Yuan, W., Shi, Q., Zhu, Z.: Improvement of thermal stability of zirconia aerogel by addition of yttrium. J. Sol-Gel Sci. Technol. 80, 667–674 (2016) 67. Lieb, E.W., Vainio, U., Pasquarelli, R.M., Kus, J., Czaschke, C., Walter, N., Janssen, R.M., Schreyer, A., Weller, H., Vossmeyer, T.: Synthesis and thermal stability of zirconia and yttria-stabilized zirconia microspheres. J. Colloid Interface Sci. 448, 582–592 (2015) 68. Matsui, K., Ohgai, M.: Formation mechanism of hydrous zirconia particles produced by hydrolysis of ZrOCl2 solutions: iv, effects of ZrOCl2 concentration and reaction temperature. J. Am. Ceram. Soc. 85, 545–553 (2002) 69. Shi, Z., Gao, H., Wang, X., Li, C., Wang, W., Hong, Z., Zhi, M.: One-step synthesis of monolithic micro-nana yttria stabilized ZrO2Al2O3 composite aerogel. Microporous Mesoporous Mater. 259, 26–32 (2018) 70. Wagh, P.B., Rao, A.V., Haranath, D.: Influence of molar ratios of precursor, solvent and water on physical properties of citric acid catalyzed teos silica aerogels. Mater. Chem. Phys. 53, 41–47 (1998) 71. Takahashi, R., Sato, S., Sodesawa, T., Kawakitz, M., Ogura, K.: High surface-area silica with controlled pore size prepared from nanocomposite of silica and citric acid. J. Phys. Chem. B. 104, 12184–12191 (2000) 72. Chen, B., Wang, X., Zhang, S., Wei, C., Zhang, L.: Monolithic Zno aerogel synthesized through dispersed inorganic sol–gel method using citric acid as template. J. Porous. Mater. 21, 1035–1039 (2014) 73. Davar, F., Hassankhani, A., Loghman-Estarki, M.R.: Controllable synthesis of metastable tetragonal zirconia nanocrystals using citric acid assisted sol–gel method. Ceram. Int. 39, 2933–2941 (2013) 74. Liu, Q., Wang, A., Wang, X., Zhang, T.: Mesoporous C-alumina synthesized by hydro-carboxylic acid as structure-directing agent. Microporous Mesoporous Mater. 92, 10–21 (2006) 75. Liu, G., Yolang, G., Li, S., Zhang, W., Jia, M.: Preparation of titaniasilica mixed oxides by a sol-gel route in the presence of citric acid. J. Phys. Chem. C. 113, 9345–9351 (2009) 76. Singh, K.A., Pathak, I.C., Roy, S.K.: Effect of citric acid on the synthesis of nano-crystalline yttria stabilized zirconia powders by nitrate–citrate process. Ceram. Int. 33, 1463 (2007) 77. Zhang, Z., Gao, Q., Liu, Y., Zhou, C., Zhi, M., Hong, Z., Zhang, F., Liu, B.: A facile citric acid assisted sol–gel method for preparing monolithic yttria-stabilized zirconia aerogel. RSC Adv. 5, 84280 (2015)

Frances I. Hurwitz received her Ph.D. in Macromolceular Science from Case Western Reserve University in 1979. She has worked at NASA Glenn Research Center on a variety of materials, including aerogels and aerogel composites, polymer-derived ceramics, polymer matrix and ceramic matrix composites, and carbon/carbon composites. Her recent focus has been on the development of thermal protection systems.

18

458

Haiquan Guo received her PhD in chemistry from City University of New York. She is a senior scientist working at Ohio Aerospace Institute in Cleveland, Ohio, U.S.A. She works on developing both inorganic and polymer aerogels.

Nathaniel Olson received his B.S. in chemical engineering from The Ohio State University in 2018. He completed an internship at NASA Glenn Research Center, mentored by Dr. Frances Hurwitz. Nathaniel is pursuing a PhD in materials science and engineering at the University of Illinois at Urbana-Champaign and is a NASA Space Technology Research Fellow.

F. I. Hurwitz et al.

Anita Garg has over 25 years of experience in the field of material characterization and structure-property correlation using conventional, analytical, and high-resolution scanning and transmission electron microscopy. She has published more than 150 technical papers related to advanced structural and functional materials in international journals and conference proceedings.

19

Zirconia Aerogels Lassaad Ben Hammouda, Imen Mejri, Mohamed Kadri Younes, and Abdelhamid Ghorbel

Contents 19.1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 459

19.2

Zirconia Aerogel Synthesis: Recent Developments . . . 460

19.3 19.3.1 19.3.2 19.3.3 19.3.4 19.3.5

Factors Affecting the Textural and the Structural Properties of Zirconia Aerogels . . . . . . . . . . . . . . . . . . . . . . . . Zirconium Precursor Type and Concentration . . . . . . . . . . . Addition of an Acid and Its Concentration . . . . . . . . . . . . . . Hydrolysis Ratio (H2O/Zr) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gel Aging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Supercritical Drying Temperature . . . . . . . . . . . . . . . . . . . . . . . .

462 462 463 463 463 463

19.4 19.4.1 19.4.2 19.4.3 19.4.4

Main Applications of Modified Zirconia Aerogels . . . . Zirconia Aerogels and Catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . Zirconia Aerogels and Ceramics . . . . . . . . . . . . . . . . . . . . . . . . . Electrochemical Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zirconia Aerogels and Optics . . . . . . . . . . . . . . . . . . . . . . . . . . . .

464 464 471 471 472

19.5

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 472

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 473

Abstract

Zirconia aerogels are very promising materials in industry with exceptional properties related to their high porosity, important surface area, and thermal stability in addition to their low density and thermal conductivity. All of these properties explain the wide potential applications of these aerogels such as catalysis, ceramics, solid oxide fuel cells (SOFCs), optics, thermal barrier coatings (TBC), and many other high-temperature applications. For several years, the preparation of zirconium aerogels was carried out using zirconium alkoxide precursors or inorganic salts through a sol–gel process followed by a L. Ben Hammouda (*) · I. Mejri (*) · M. K. Younes (*) · A. Ghorbel (*) Laboratoire de Chimie des Matériaux et Catalyse (LCMC), Département de Chimie, Faculté des Sciences de Tunis, Université Tunis El Manar, Tunis, Tunisia e-mail: [email protected]; [email protected]; [email protected]; [email protected]

supercritical drying (SCD) method in order to preserve the porous network of the final product. However, it has been reported that the synthesis procedures relying on inorganic salts as starting materials are generally nontoxic, inexpensive, more practical, and easier to develop and allow at the same time, after supercritical drying of the corresponding gels, to better aerogels with respect to surface area and nanoporous structure. Furthermore, the SCD process presents a major handicap when it is set up on an industrial scale, related to the use of a high-pressure autoclave for the drying. Therefore, this has prompted many researchers to develop a new method to produce “aerogels” by ambient pressure drying (APD). During the last 20 years, many research groups interested by zirconia aerogels and their potential applications confirmed that the tetragonal crystal phase of ZrO2 seems to be the most interesting phase. However, it has been reported that the textural and structural properties of zirconia aerogels depend on several preparation parameters such as the zirconium precursor and its concentration, the type of solvent, the use of complexing or silylating agents, the acid concentration, the hydrolysis ratio in the case of sol–gel process, the gel aging time, and the SCD temperature. Keywords

Zirconia · Aerogel · Sol–gel · Supercritical drying · Catalysis · Coating

19.1

Introduction

Zirconia is a material having very interesting redox, acidic, thermal, and mechanical properties. These features allow zirconia to be widely used for various applications such as catalysis, thermal barrier coatings (TBC), thermal protection system, oxygen sensors, electronics, optics, ceramics, and fuel cells [1–11].

© Springer Nature Switzerland AG 2023 M. A. Aegerter et al. (eds.), Springer Handbook of Aerogels, Springer Handbooks, https://doi.org/10.1007/978-3-030-27322-4_19

459

460

L. Ben Hammouda et al.

In order to improve the performances (physicochemical characteristics in general as reactivity for catalysis or mechanical properties for ceramics) of prepared materials containing zirconium oxide, the most important characteristics requested are high surface areas and stable zirconia nanoparticles. Many researches are interested in the development of zirconia with promising textural and structural properties (high surface area, developed porosity, well crystallized, and so on). Up to now, ZrO2 has been successfully prepared by several techniques including precipitation, sol–gel routes, electrolyzing, and hydrothermal process. The zirconium precursors (chloride, oxychloride, nitrate, and alkoxides) have been used. Nevertheless, the majority of researchers agrees that materials issued from drying the liquid components of a gel through supercritical fluid have unique properties, such as high porosity and large surface area. Several works reported that using sol–gel routes for the synthesis of zirconia allows a better control of texture, composition, and structural features. In fact, many authors [12, 13] indicated that zirconia aerogels obtained by a hightemperature supercritical drying (SCD) exhibit developed surface areas, large pore volumes, and a temperature-stable tetragonal phase. The physicochemical properties of zirconia aerogel are affected by many preparation parameters. The choice of the nature of the zirconium precursor (alkoxide or inorganic salt) and its concentration affects the final solid. Indeed, for certain concentrations, the final solid has a better specific surface and a well-developed porosity. The process of hydrolysis and condensation plays a crucial role in obtaining a promising solid. Condensation is governed by the choice of acid precursor and its concentration. The absence of the acid or its presence with very low concentrations leads to precipitates. For high concentrations the gel obtained is very soft, transparent, and very fluid. The hydrolysis step is also very important. Theoretically, at a very low rate of hydrolysis, gel formation does not occur. The solvent evacuation mode is a fairly important parameter. When the evacuation of the solvent was performed under its supercritical conditions, there is no surface tension or capillary pressure. The absence of surface tension leads to solid with a small shrinkage of the pore volume and a high surface area. Uncalcined aerogels are well crystallized and present the tetragonal phase of the zirconia.

19.2

Zirconia Aerogel Synthesis: Recent Developments

Zirconia aerogels present many significant properties related to their high porosity, important surface area, and thermal stability in addition to their low density and thermal

conductivity. All of these properties explain the wide potential applications of these aerogels such as catalysis, ceramics, solid oxide fuel cells (SOFC), optics, coating, and thermal insulators [12, 14–17]. For several years, the preparation of zirconium aerogels was carried out using zirconium alkoxide precursors or inorganic salts through a sol–gel process followed by a supercritical drying (SCD) method in order to preserve the porous network of the final product. However, it has been reported that this process presents a major handicap when it is set up on an industrial scale [18], related to the use of a highpressure autoclave for the drying. Therefore, this has prompted many researchers to develop a new method to produce “aerogels” by ambient pressure drying (APD) [19]. Woignier et al. [20] reported the use of the ambient pressure drying (APD) techniques in order to synthesize monolithic dried gels. Indeed, this new process is based on the control of the drying step either by increasing the gel mechanical properties and permeability or by decreasing the capillary stress gradient. This last goal can be achieved by the drop of the surface energy or by narrowing the pore size distribution. The main advantage of the ambient pressure method is that it does not require dangerous high-pressure equipment. However, several steps of solvent exchange and chemical reaction are necessary. Moreover, these same authors indicated that the freeze-drying is not appropriate to obtain monolithic dried gels. Recently, Bangi et al. [19] revealed the synthesis of nanostructured zirconia powders qualified as aerogels by ambient pressure drying (APD) using zirconium n-propoxide (Zr (OPr)4) as a precursor, n-propanol as a solvent, nitric acid as a catalyst and inhibitor, hexane as an aprotic solvent, and hexamethyldisilazane (HMDS) as a silylating agent. Deionized water with a H2O/Zr molar ratio kept constant at 4 was used for hydrolysis and to prepare a dilute solution of nitric acid (HNO3) (1 mM). The as-synthesized zirconia aerogel powders were obtained after drying of wet gel at 50
C and 200
C in an oven at ambient pressure for 1 h. Obtained results showed that zirconia aerogel powders thermally treated up to 600
C have a large surface area (more than 153 m2/g) and maintained their nanoporous structure. However, when the heat treatment exceeds 600
C, an important decrease in surface area induced by an accelerated agglomeration of zirconia aerogel powders has been indicated. In 2014, zirconia monolithic aerogels had been successfully prepared by Zhong et al. [21] via the epoxide addition method by adjusting the solvent and acid content. Zirconium oxychloride octahydrate (ZrOCl2; 8H2O), propylene oxide (PO), and HNO3 were used. Two beakers were prepared: one with ZrOCl2, ethanol, H2O, and HNO3, the other with ethanol and PO. The total molar ratio of ZrOCl2, H2O, HNO3, and PO was kept constant at 1:10:1:6. The two

19

Zirconia Aerogels

solutions were mixed and vigorously stirred until the vortex created by the stirring disappeared owing to gelation. Furthermore, a water bath adjusted at 20
C was needed in the process of hybrid. After 24-h aging, the alcogels were washed with ethanol every 24 h and renewed 2–3 times in order to make sure the solvents in the gel network were fully replaced by ethanol. The resultant alcogels were then supercritically dried (SCD) in an autoclave. In fact, after pressurizing the autoclave to 2.5 MPa with nitrogen, the temperature and the pressure were adjusted at 538 K and 7.2 MPa, respectively. This supercritical state was kept for 1 h before a slow evacuation at atmospheric pressure. It is important to note that the appearance of the obtained gel depends on the acid to zirconium and water to zirconium molar ratios. So, opaque gels are always obtained at lower molar ratios, and translucent gels are obtained if the acid and water contents increase. These same authors have pointed out that the gel time is prolonged to 25 min if the HNO3/ZrOCl2 molar ratio is fixed at 1, leading not only to the formation of a stable gel with an extended network but also to the possibility of preparation of aerogel composite materials. Furthermore, they indicated a significant drop in the strength of gel structure leading to the appearance of cracks and shrinkage during supercritical drying process when the acid and water contents decrease. Additionally, the acid and water contents were found to be critical factors in the morphology and microporous structure of the resulting aerogels. Fixing the HNO3/ ZrOCl2 and H2O/ZrOCl2 molar ratios at, respectively, 1 and 10 allows the preparation of zirconia aerogel monolith exhibiting a well-developed mesoporous structure and a high specific surface area exceeding 450m2.g1. However, increasing the acid contents leads to a drastic decrease of the specific surface area of the obtained aerogels. This result is attributed to a decrease in crosslinking due to excessive consumption of epoxy by nitric acid, resulting in changes in the gel network during the aerogel drying process. It is important to note that the obtained aerogels were mainly amorphous phase and contained a spot of crystal. The XRD and Raman results suggested that tetragonal zirconia (ZrO2) were obtained after calcination at 750
C for 2 h and revealed that a tetragonal phase growth of the zirconia occurred with increasing the calcination temperature. In fact, the as-prepared zirconia calcined at 850
C or higher temperature were a mixture of monoclinic and tetragonal phases. Zhao et al. [22] reported that zirconia aerogels with high surface areas were successfully prepared by a combined electrolysis/sol–gel method, followed by supercritical extraction (S-aerogel) or freeze-drying (F-aerogel). This process is based on the electrolysis reaction during 120 h, at 25
C and an electrode voltage fixed at 5 V of 30 mL of 0.3 M ZrOCl2; 8H2O aqueous solution, which turns gradually to a transparent sol when the concentration of Cl anions decreased to 0.01 M. The wet gel obtained after the addition, under

461

stirring, of 30 mL of isopropyl alcohol was aged for 24 h. Finally, several drops of 1,2-epoxipropane were added into the gel to deplete the remaining Cl anions, and then the gel was aged in a closed container for an additional 24 h. To obtain the S-aerogel sample, the wet gel was first immersed during 120 h in absolute ethanol in order to exchange the solvent in the network. It is important to note that the absolute ethanol was renewed several times to ensure maximum exchange. The extraction of the ethanol from the gel was then made with liquid CO2 for 72 h in a supercritical extractor where the temperature and pressure were fixed at 45
C and 10 MPa, respectively. Obtained S-aerogel appears as a transparent monolith characterized by a mesoporous structure with an average pore diameter of 9.7 nm and a high specific surface area (640 m2/g). This solid exhibited a mixture of m-ZrO2 and t-ZrO2 phases after calcination at 500
C. Otherwise, in order to obtain the freeze-dried aerogel F-aerogel, the wet gel without exchanging with ethanol was put into a flask and quickly frozen with liquid N2 and then freeze-dried for 12 h at 50 to 60
C and a pressure less than 10 Pa. In that case, F-aerogel was characterized by a microporous structure with mean pore diameter of 0.6 nm, a specific surface area of 400 m2/g, and a single t-ZrO2 phase after calcination at 500
C. In their study Zhao et al. indicated also the total absence of any effects of the Y2O3 contents on the crystalline structure of the calcined zirconia aerogels, since yttria-stabilized zirconia aerogels (YSZ) showed similar properties including particle size, microstructure, pore size, and surface area as well as the phase structure of the calcined samples of pure zirconia aerogels. Schäfer et al. [23] reported that they managed to synthesize high-temperature-stable zirconia aerogels and yttriumdoped zirconia aerogels (YSZ) using zirconium n-propoxide (Zr(OPr)4) or zirconium n-butoxide (Zr(OBu)4) as metal precursors. However, they reveal that the synthesis procedures relying on inorganic salts as starting materials are generally nontoxic, inexpensive, more practical, and easier to develop. Furthermore, they indicated that the use of the inorganic salts as metal precursors leads, after supercritical drying (SCD) of the corresponding gels, to better aerogels with respect to surface area and nanoporous structure. In fact, zirconia aerogels prepared using ZrCl4 showed larger BET surface areas, situated between 360 and 450 m2.g1, and an average pore diameter smaller than 10 nm. On the other hand, aerogels obtained from Zr(OBu)4 or Zr(OPr)4 as precursors are characterized by significant lower BET surface values (200–400 m2.g1) and a pore size distribution ranged from 5 to 200 nm, depending on the reaction conditions and the auxiliary agent. Uncracked zirconia monolithic aerogels were successfully prepared by Li et al. [24] using a sol–gel process followed by

19

462

L. Ben Hammouda et al.

high-temperature supercritical drying (SCD). Their experimental process consists first of all in dissolving 3 ml of zirconium n-butoxide in 11.6 ml of ethanol (EtOH). 0.414 ml of HNO3 was then added dropwise under continuous stirring. In another beaker, 0.286 ml of H2O and a definite amount of PO were dissolved in 11.6 ml of EtOH. After stirring for 10 min, the contents of the two beakers were quickly mixed leading to the formation of a transparent sol. After aging for 24 h at room temperature, the wet gel obtained was placed in an autoclave, which was partially filled with ethanol and pressurized to 3 MPa with N2. The
temperature was raised to 253
C at a rate of 2 /min, while the rising pressure was controlled at 7.5 MPa. After 30 min, the autoclave was depressurized slowly during 3 h and cooled to room temperature. Using this process, opaque intact monolithic aerogels with varied molar ratio of PO/Zr were obtained. Their results showed that increasing the molar ratio of PO/Zr from 0 to 2.5 leads to an increase of both particle size and connectivity of network of ZrO2 aerogels, while average pore radius increased from 6.6 nm to 15.1 nm then decreased to 8.5 nm and pore volume increased from 1.2 cc/g to 1.87 cc/g then decreased to 1.44 cc/g.

19.3

Factors Affecting the Textural and the Structural Properties of Zirconia Aerogels

19.3.1 Zirconium Precursor Type and Concentration ZrO2 aerogels are synthesized using two different preparation routes: alkoxide [25–27] and non-alkoxide [22, 28, 29] ways. The zirconium alkoxide precursors used are zirconium n-propoxide [26, 30, 31], zirconium iso-propoxide (Zr(OiPr)4) [32], and zirconium n-butoxide [33]. The uncalcined solid obtained have an important surface area [34]. Aerogel zirconia nanomaterials synthesized by a one-step sol–gel route under supercritical CO2 drying using zirconium alkoxides and acetic acid were reported by Sui et al. [35]. Obtained materials exhibited high surface areas (up to 399 m2.g1) and porosity, while the calcined materials demonstrated tetragonal and/or monoclinic nanocrystallites. Either a translucent or opaque monolith was obtained. Investigation by electron microscopy showed that the translucent monolithic ZrO2 exhibited a well-defined mesoporous structure, while the opaque monolith, formed using added alcohol as a co-solvent, was composed of loosely compacted nanospherical particles with a diameter of 20 nm. Furthermore, Tyagi et al. [36] have synthesized nanostructured zirconia aerogels using zirconium n-propoxide and n-butoxide. The evacuation of solvent was performed under supercritical conditions of the solvent. Obtained materials developed the

tetragonal phase before calcinations. The formation of monoclinic phase starts at 500
C. Further, the high concentration of zirconium precursor promotes the stabilization of tetragonal phase at higher calcination temperatures. These solids are mesoporous with rather a uniform pore size distribution. Moreover, when zirconium n-butoxide was used as a precursor, the obtained aerogel shows a relatively developed specific surface area which can reach 465 m2.g1 before calcination. The non-alkoxide precursors used are zirconium salts such as ZrOCl2 [22, 37], ZrCl4 [38], and ZrO(NO3)2 [39]. Zhao et al. [22] prepared zirconia aerogel by electrolyzing zirconium oxychloride solutions at ambient temperature. The evacuation of solvent was carried out using the supercritical conditions of CO2 or freeze-drying. The solids obtained using the supercritical conditions of CO2 exhibit an important surface area around 640 m2/g. The aerogel was mesoporous and transparent monolith with an average pore size of 9.7 nm. Besides, sample prepared by the freeze-drying process was an opaque white powder with a microporous structure (pore size of 0.59 nm). Zhang et al. [39] reported the preparation of nanoporous ZrO2 aerogels using zirconyl nitrate (ZrO (NO3)25H2O) as precursor and hydrothermal method followed by supercritical drying technique. Obtained aerogels are nanoporous with an average pore diameter in the interval between 5 < Dp < 60 nm. This preparation method leads to developed specific surface areas reaching 916.5 m2/g and homogeneous pore size distributions. Many papers deal with using epoxides as proton scavenger from inorganic salts in polar protic solvents to prepare the aerogels. Zhong et al. [21] prepared ZrO2 aerogels using zirconium oxychloride octahydrate, propylene oxide (PO), and different amounts of water and nitric acid. The resultant alcogels were then supercritically dried by ethanol in an autoclave. Obtained results showed that the acid and water contents had an important impact on the morphology of the resulting aerogels. At a molar ratio of acid (HNO3/ZrOCl2) of 1.0 and hydrolysis ratio (H2O/ZrOCl2) of 10, the translucent monolithic ZrO2 aerogels were obtained, and they exhibited a well-developed mesoporous structure and a high specific surface area of 454 m2.g1. At higher acid amount, the aerogels were cracked and broke apart as well as their specific surface area decreased. This is attributed to excessive consumption of epoxy by nitric acid which decreases the crosslinking, resulting in the network of gels being damaged during the aerogel drying process. The concentration of the zirconium precursor has an important impact on the physical and chemical properties of the resulting solid [24]. The concentrations of the precursors are variable. The main research results show that when the concentration is between 0.1 M and 0.5 M, the concentrations of acids are lower and the obtained solids developed are well crystallized and exhibit important surface areas and higher porosities. In fact, higher precursor concentrations favored

19

Zirconia Aerogels

olation reactions, and more dense aerogels with smaller pores were formed. A high precursor concentration required large acid concentration to obtain clear and rigid gel leading after SCD at 275
C (using ethanol as solvent) to aerogels with important surface areas which drastically decrease after calcination. In addition, the higher the precursor concentration, the lower the average pore radius and the smaller the pore volume.

19.3.2 Addition of an Acid and Its Concentration To control the process of the gelation which is governed by the hydrolysis and condensation reactions, the use of an acid is crucial. The absence of acids or a not sufficient amount leads immediately to precipitation. The precipitates usually do not possess desirable porous structure and surface areas [30]. It has been reported that acetic acid affects the rate of the hydrolysis [40], while nitric acid affects the rate of condensation [30–41]. The acid contents were found to have a crucial impact on the morphology and the microporous structure of the resulting aerogels. In a certain concentration range, quick formation of a rigid polymeric gel containing some precipitate particulates is observed. Further, the increase of the acid amount results in the increase of the gelation time and the formation of clear rigid and then clear soft alcogels. Obviously, if the acid concentration is sufficiently high, condensation can be completely avoided. As an example, the effect of the nitric acid amount has been treated by D. J. Suh and T. Park [31]. Zirconia aerogel is prepared using zirconium n-propoxide, the hydrolysis ratio is fixed to 4, and the amount of nitric acid is varied. The evacuation of the solvent is carried out under supercritical drying with carbon dioxide. The obtained results show that the amount of nitric acid played an important role in determining the gel morphology. Lower acid content leads to precipitate; when the amount of acid is between 0.7 and 1, the sol–gel reaction could give translucent gels which contained precipitates within the gel network. For higher acid content (between 1 and 2.5), transparent polymeric gels could be obtained. However, when an excessive amount of acid was used, the sol–gel solution formed very soft and transparent gel and exhibited fluidity. All the solids issued from different amounts of acid exhibit, after calcination at 773 K for 2 h, surface areas neighboring 100 m2/g. These aerogels are mesoporous with relatively narrow pore size distributions.

19.3.3 Hydrolysis Ratio (H2O/Zr) The hydrolysis ratio defined by H2O/Zr report is an important parameter which influences the hydrolysis and condensation mechanisms and consequently the resulting compounds from

463

these reactions. This ratio affects not only the size of the molecular species but also their structures, textures, and physical properties. Theoretically, 4 molecules of water should be required for each Zr(OR)4 molecule. If the hydrolysis ratio is less than 1, the condensation reaction is mainly governed by the alcoxolation. Gelation and precipitation cannot take place because the lack of water. When hydrolysis ratio is between 1 and 4, condensation is governed by two competitive reactions: alcoxolation and oxolation. For higher amount, more then 4, gels or precipitates occur. Vicarini et al. [42] study the effect of the hydrolysis ratio; their results indicated that in the range of conditions investigated a hydrolysis level of 4–8 resulted in most favorable morphological properties. The uncalcined zirconia aerogels were X-ray amorphous and mesoporous to macroporous having surface areas between 317 and 400 m2.g1. Ko et al. [43] reported that hydrolysis molar ratio (H2O/Zr ¼ 2) appears to yield the highest surface areas after low temperature SCD followed by calcination at 500
C. However, with the higher ratios of water, higher concentrations of acid were required to prevent precipitation. Moreover, Benedetti et al. [44] prepared aerogel zirconia using an excess of water (hydrolysis ratio ¼ 20). The resulting untreated zirconia aerogel contained mainly tetragonal ZrO2 and a specific surface area of 350 m2.g-l. An average size about 5 nm was found using the SAXS fractal approach. This result was to be close to the value determined by the X-ray diffraction linebroadening analysis as well as by transmission electron microscopy (TEM).

19.3.4 Gel Aging Alcogel aging is known to be an important parameter for sol–gel synthesis process [44]. For zirconia aerogels, long aging time has been recommended [45]. It has been noted that prolonging aging time enhances slightly final surface area from about 95 up to 111 m2/g, but it had no effect on pore structure [31]. Recently, it has been reported that aging the wet gel in TEOS is efficient to obtain crack-free aerogels. This is due to the fact that SiO2 from partially hydrolyzed TEOS could improve the connection between nanoparticles [46].

19.3.5 Supercritical Drying Temperature Zirconia alcogels can be dried using conventional drying at ambient temperature in an oven or using supercritical drying (SCD) process. Two supercritical drying methods exist: hightemperature (using the supercritical temperature and pressure of the solvent) and low-temperature methods (using supercritical conditions of liquid CO2). Many researchers investigated the effect of the SCD drying on the physicochemical

19

464

L. Ben Hammouda et al.

properties of the final solid. They reported that the SCD temperature ameliorates the textural and the structural properties of zirconia aerogels. For example, J. Fenech et al. [47] prepared xerogel and aerogel zirconia stabilized by yttria. The obtained results showed that before calcination aerogel is well crystallized, whereas xerogel is amorphous. After calcination at 950
C, the two solids exhibit the tetragonal phase. While the xerogel exhibits a densified structure of agglomerates constituted by 50 nm crystallites, the aerogel presents a more porous structure with crystallites having an average size of about 26 nm.

19.4

Main Applications of Modified Zirconia Aerogels

Porous zirconia is a very promising material in industry, with exceptional electrical, thermal, and mechanical properties allowing its use in fuel cells, transparent optical devices, thermal barrier coatings (TBC), electrochemical capacitor electrodes, oxygen sensors, advanced ceramics, and many other high-temperature applications [7–11]. In fact, these unusual properties and widespread applications of ZrO2 are the origin of an expansion of the study interested in this material during the last few years. Otherwise, in order to improve the physicochemical properties of prepared materials containing ZrO2, the most requested characteristics are high surface area and thermal stability of ZrO2 nanoparticles. So, it has been reported that monolithic ZrO2 aerogel with low density has a bright future in the field of thermal insulation material because of its high surface area and excellent thermal stability [46]. During the last 20 years, many research groups interested by zirconia aerogels and their potential applications confirmed that the tetragonal crystal phase of ZrO2 seems to be the most interesting phase. However, below 1175
C, the stable phase of zirconia is the monoclinic phase. Thus the tetragonal phase can only exist at a room temperature through stabilization. This is normally achieved by adding a dopant such as Y2O3 or by modifying the surface with sulfuric or phosphoric acid. Furthermore, Skovgaard et al. [48] reported that it is possible to obtain metastable tetragonal ZrO2 at room temperature, by suitable control of the processing parameters and thereby crystal grain size knowing that below a certain critical crystal size zirconia can adopt the metastable tetragonal phase.

19.4.1 Zirconia Aerogels and Catalysis Zirconia aerogel is known to possess several properties required for applications as a catalyst or catalyst support. Its surface is known for its acidic and basic sites as well as for its

oxidizing and reducing chemical properties [49]. In addition, zirconia aerogel exhibits adequate structural and textural characteristics. These properties confer a good dispersion of a catalytically active phase, which enhances catalytic activity, thermal stability, and resistance to poisoning. The most reported catalytic applications of zirconia aerogels are related to the modification of their surfaces by metals such as platinum, copper, cerium, nickel, and chromium or oxoanions such as sulfate, phosphate, and tungstate anions. Zirconia aerogels doped with metals have been investigated for different catalytic applications. Pajonk et al. [50] showed that zirconia aerogels are promising supports for active metallic centers and are also by themselves interesting catalysts for reactions involving hydrogen, toward isomerization and/or hydrogenation of alkenes at low temperature. In fact, it was reported that catalyst activation at 430
C in vacuum leads to the best results in cis-trans isomerization in the temperature range of 80–200
C. It was also shown that, at temperatures below or equal to 150
C, the maximum of selectivity is reached. Selective poisoning experiments by NH3 or CO2 were carried out in order to identify the catalytic sites needed for the isomerization of n-butene. Kalies et al. [51] studied the hydrogenation of formate species on a Pt-zirconia aerogel catalyst. Zirconia was synthesized by the hydrolysis of a solution of zirconium isopropoxide in isopropanol. The solvent was evacuated by SCD. Zirconia was further impregnated with a solution of hexachloroplatinic acid in isopropanol to obtain a 0.5 wt% loading of Pt. The Pt-impregnated zirconia was then dried under supercritical isopropanol conditions. All solids were X-ray amorphous. The reaction was monitored by FTIR and showed that the adsorption of CO (10% CO/He) on ZrO2 and 0.5% Pt/ZrO2 at 646 K lead to the progressive formation of formate species indicating that the presence of Pt does not influence the formation of the formate group, its structure, and its stability in an inert gas. However, the produced formate species were hydrogenated by H2 into methoxy groups only when Pt was present on zirconia. These species were further converted into methane through a reverse spillover mechanism [51]. On pure zirconia, the formate species were not hydrogenated because of the inability of zirconia to dissociate molecular hydrogen. Zirconia aerogel has also been used as support for metal such as copper. For example, Cu/ZrO2 aerogels were prepared using sol–gel methods by controlled precipitation at 295 K from zirconyl nitrate and cupric nitrate precursors and then tested in the hydrogenation reaction of CO toward methanol by Sun et al. [52]. These aerogels exhibited high surface areas (up to 250 m2.g1) even after reduction in H2 at 573 K. X-ray diffraction and transmission electron microscopy showed that the fresh aerogels were amorphous and composed of clusters of particles less than 5 nm in size. A fraction of these primary particles grew to about 10 nm after

19

Zirconia Aerogels

24 h of the catalyzed reaction. The fresh aerogels were very active and stable in this methanol synthesis reaction. A relatively strong interaction of the support with the Cu species is evidenced by TPR, and this appears to correlate well with the obtained high activity. Hence, this activity could not only be simply attributed to the high surface area of such aerogel samples but also to the appropriate interaction between Cu and zirconia. Doped ZrO2 aerogels with rhodium or yttrium are also very active in methane oxidation [53]. Nanoscale iron-doped zirconia solid solution aerogels were prepared by Chen et al. [15] via a simple ethanol thermal route using zirconyl nitrate and iron nitrate as starting materials, followed by a SCD process. The results showed that the obtained iron-doped solid solutions crystallized in the zirconia metastable tetragonal phase, which exhibits excellent dispersion and high solubility with iron oxide. Furthermore, when the Fe/(Fe + Zr) ratio, noted x, is lower than 0.10, all of the Fe3+ ions can be incorporated into ZrO2, by substituting Zr4+, to form Zr1-xFexOy solid solutions. Moreover, for the first time, an additional hydroxyl group band that is not present in pure ZrO2 is observed by DRIFT for the Zr (Fe)O2 solid solution. These Zr1-xFexOy solid solutions are excellent catalysts for the solvent-free aerobic oxidation of n-hexadecane using air as oxidant under ambient conditions. The Zr0.8Fe0.2Oy solid solution catalyst demonstrated the best catalytic properties in this reaction. In fact, the conversion of n-hexadecane reached 36.2%, with a selectivity of 48 and 24% toward ketones and alcohols, respectively. It is also important to note that this catalyst can be recycled five times without significant loss of activity. Fischer-Tropsch synthesis reaction, over cobalt supported on ZrO2-SiO2 aerogels, was investigated by Wang et al. [54]. The results showed that, under favorable conditions allowing the formation of long-chain hydrocarbons, cobaltcatalyzed Fischer-Tropsch reaction appears to be structure sensitive. Moreover, it was reported that support influences significantly the catalytic behavior. In fact, when cobalt was supported on zirconia-coated silica aerogel, heavy products were obtained from syngas (CO + H2). In this case, C5+ yield could reach 150 g Nm3(CO + H2) under the optimal conditions (T ¼ 293 K; P ¼ 2.0 MPa; gas hourly space velocity (GHSV) ¼ 500 h1). However, if cobalt catalyst was supported on ZrO2-SiO2 mixed aerogel, it was shown to produce middle distillate, and the yield of C5-C20 products for this catalyst was about 120 g Nm3 (CO + H2). Interesting acidic properties of zirconia aerogels incite researchers to use them in many reactions. Surface acidity of these materials was enhanced by doping with anions such as sulfate, phosphate, and tungstate. The sulfate groups were the most used dopants of zirconia aerogels in order to reach very high acid catalysts. These groups were introduced in the zirconia aerogel via different methods. It was found that the addition of sulfate groups retards the crystallization of

465

zirconia and stabilizes tetragonal phase when sulfated zirconia aerogels were treated at high temperature. Furthermore, sulfate groups migrate to aerogel surface, and the strong ZrO-S bond combined with hydroxyl groups Zr-OH creates strong Brönsted acidity [55]. Phosphate ions were also used to increase acidity on the surface of zirconia aerogel. Boyse et al. [56] prepared zirconia-phosphate aerogels by two methods: a one-step sol–gel synthesis followed by a supercritical drying (SCD) and an incipient wetness impregnation synthesis of a calcined zirconia aerogel. They reported that zirconia-phosphate aerogels possess Brönsted acid sites, and the phosphate species were claimed to be responsible of their generation. The nature of surface acidity of zirconia aerogels and sulfated zirconia aerogels was studied by many research groups. In fact, sulfated zirconia aerogels, with definite atomic ratio S/Zr and hydrolysis ratio H2O/Zr, showed that the Kelvin probe value of pure zirconia aerogels is around 200 mV. However, this value increases up to 1200 mV for sulfate-doped catalysts. The modification of the work function is probably due to the charge transfer from the zirconium to the oxygen species, responsible for the increase of Lewis acidity [57]. Correlation between X-ray photoelectron spectroscopy (XPS), surface potential measurements, and isopropanol dehydration reaction results showed that when zirconia was doped by sulfate groups, its surface becomes more acidic in terms of Lewis acidity and that the oxygen species in the sample exist in many types, one of which is related to solid acidity [58]. This type of oxygen species, probably of the hydroxyl groups, is different from the oxygen species of zirconia network and the oxygen of sulfate groups. Consequently, acidity of sulfated zirconia is mainly due to the strong Lewis acid sites on the surface, which can convert to Brönsted acidity by chemisorption of water or reactant. Sulfated zirconia aerogel was essentially obtained by solgel method. However, sulfate groups were introduced from different precursors. In their work, Ward et al. [30] added sulfuric acid to zirconium n-propoxide precursor before hydrolyzing with water. Obtained gels were then dried under supercritical conditions. During the drying step, the solvent is exchanged with supercritical CO2 or evacuated at high temperature, where the solvent is converted into the supercritical state by heating in an autoclave. Bedilo et al. [59] synthesized sulfated zirconia aerogel by a sol–gel method followed by high-temperature SCD. Sulfur was introduced during the gel formation step. Resulting samples exhibited high surface areas and pore volumes. Impregnation of these zirconia aerogels with (NH4)2SO4 led to high catalytic activity in the n-butane isomerization reaction, compared to conventional precipitated zirconia. Impregnated aerogels retain more surface sulfates after calcination. The most active samples contain a close monolayer surface

19

466

L. Ben Hammouda et al.

coverage with sulfate groups. Over monolayer surface, sulfates hindered the isomerization activity, whereas bulk sulfates exerted little effect. Studies focusing on the use of sulfated zirconia have shown that reaction taking place at the strong acid sites also needs the metallic phase such as platinum for burning off coke produced over sulfated zirconia, stabilizing the reaction intermediates of the isomerization. Ghorbel et al. [12] reported that aerogel and xerogel sulfated zirconia exhibit different structural, textural, and catalytic properties at various calcination temperatures. In fact, they prepared different sulfated zirconia samples as follows: zirconium n-propoxide, dissolved in propanol, was sulfated with concentrated H2SO4 to obtain a molar ratio of S/Zr ¼ 0.5. Water was then slowly added dropwise to obtain a gel with a hydrolysis ratio (H2O/Zr) of 3. For the next step, the wet gel was dried either by simple evaporation in an oven to give a xerogel called XZS0.5H3 or under supercritical conditions of the solvent (P ¼ 51 bar and temperature ¼ 263.6
C) to give an aerogel named AZS0.5H3. The resulting solids were then calcined under oxygen at different temperatures in the range 300–700
C with a heating rate of 3
C min1 during 3 h. Results obtained (Fig. 19.1) showed that after calcination at 560
C the aerogel develops only the ZrO2 tetragonal phase, whereas the xerogel XZS0.5H3 contains both the monoclinic and the tetragonal phases. Heating at higher temperature causes the transition of the tetragonal phase into the monoclinic one for all the samples by loss of sulfur, but the tetragonal phase remains significantly more stable in the aerogel.

Characterization by XPS indicates that the loss of sulfur at higher temperature is easier for the xerogel. The ability of the aerogel AZS0.5H3 to retain sulfur at higher temperature explains its better stability and confers its good catalytic performances in the n-hexane isomerization reaction. The mixed oxide including sulfated zirconia and silica or alumina was also studied. Mesoporous silica-supported nanocrystalline sulfated zirconia catalysts were prepared via the sol–gel process using an in situ sulfation and dried in an autoclave under supercritical conditions of the solvent (T ¼ 265
C, P ¼ 51 bar) by Akkari et al. [60]. The influence of the S/Zr molar ratio on the properties of the prepared catalysts has been studied. The synthesized solids were characterized using XRD, N2 physisorption, TG-DTA/MS, sulfur chemical analysis, and adsorption-desorption of pyridine and tested in the gas-phase acid-catalyzed isomerization reaction of n-hexane. It has been noted that the gelation process is highly affected by the sulfate loading. Two gelation mechanisms were evidenced depending on the S/Zr molar ratio. The first one was observed when 0.15 < S/Zr < 0.5 and was characterized by a relatively high gelation rate. This mechanism favors the formation of two types of mesopores and a low percentage of retained sulfur. The second gelation mechanism occurs for higher S/Zr ratios: 0.5 < S/Zr < 1.2. In that case, slower gelation rates are observed leading to materials with reduced BET surface area but higher amount of retained sulfur (Fig. 19.2). Appreciable catalytic properties were observed for the sample prepared with the highest S/Zr ratio which presents the smallest size of sulfated zirconia crystallites and shows both Brönsted and Lewis acid sites on its surface.

T M Intensity (a.u.)

M

T

T

M

f e d T

T:Tetragonal M: Monoclinic

T T

M

M

M

c b a

15

25

35

45

55

65 2q (degree)

Fig. 19.1 XRD patterns of the aerogel and the xerogel calcined at different temperatures: (a) AZS0.5H3, (b) AZS0.5H3–560, (c) AZS0.5H3–700, (d) XZS0.5H3, (e) XZS0.5H3–560, (f) XZS0.5H3–700. (Reprinted from Mejri et al. [12], with permission from Elsevier)

Zirconia Aerogels

1.4 1.3 1.2 1.1 1.0 0.9

140 120 100 80 60 40 20 0

T gel (min)

Pore volume (cm3/g)

1.5

467

0

0.15 0.30 0.45 0.60 0.90 1.20 S(Zr)

Fig. 19.2 Effect of S(Zr) molar ratio on the pore volume and gelation time. (Reprinted from Akkari et al. [60], with permission from Elsevier)

Mesoporous aerogels based on sulfated zirconia doped with chromium have been synthesized by sol–gel method and dried in supercritical conditions of the solvent by Ghorbel et al. [61]. The characterization results revealed that ZrO2 tetragonal phase is stabilized by both sulfate groups and metal at high temperature. The physisorption of N2 showed that the sol–gel procedure coupled to the drying under supercritical conditions of the solvent provides to the zirconia aerogels a developed porosity. The important role of chromium metal phase relies on the improvement of n-hexane isomerization reaction by burning off coke produced over sulfated zirconia (Fig. 19.3). The calcination temperature plays an important role in the stabilization of a particular chromium oxidation state. This state seems to be responsible of the improvement of the catalytic performances of sulfated zirconia aerogels doped with chromium [61]. In order to enhance stability and acidity, zirconia aerogels were also doped with tungsten. Boyse et al. [62] reported that zirconia-tungstate aerogels prepared by one-step synthesis required a more elevated activation temperature in order to expel tungstate from the bulk of zirconia before dispersion on the surface. The variation of the activation temperature allowed the study of the transition between a catalytically inactive and active material and consequently identification of the active species. Activity in n-butane isomerization coincided with the presence of a larger and stronger population of Brönsted acid sites on the surface. The preparation method affected the activation behavior of zirconia-tungstate materials but not the active species in n-butane isomerization reaction. The same researcher [63] found that the presence of silica, both dispersed in zirconia and segregated on the surface of zirconia, retarded sintering caused by heat treatment, thereby increasing the surface area of the aerogels. On poorly mixed zirconia-silica, tungstate effectively dispersed onto the surface regardless of the existence of silica-rich patches. The

AZSCr673 Isomerization activity (nmol.g–1.s–1)

19

16 14 12 10 8 6 4 2 0

0

AZSCr9773

40 20 Time (min)

AZS833

60

Fig. 19.3 Activity of AZS (aerogel sulfated zirconia) and AZSCr (aerogel sulfated zirconia doped with chromium) in n-hexane isomerization reaction [61]

ensuing reduction of activity during n-butane isomerization reaction, compared to zirconia-tungstate, established that tungstate groups on the patches did not enhance active sites. On well-mixed zirconia-silica, the presence of siloxane species in the support further reduced catalytic activity by extending the influence of silica in zirconia-tungstate. The lower catalytic activities of these materials correlated with the presence of weaker Brönsted acid sites than that on the surface of zirconia-tungstate. This work demonstrates that silica as a dopant is effective in increasing the surface area of zirconia-tungstate but not in enhancing n-butane isomerization activity. Sulfated zirconia aerogels were also doped with cerium. In order to optimize the catalytic performances of these materials, prepared by the sol–gel method, Mejri et al. [64] studied the impact of the solvent evacuation mode on their properties. The xerogel solids (noted XZC) obtained by ordinary gel drying and calcined at different temperatures exhibit very low surface areas. On the contrary, the aerogels (noted AZC) obtained by solvent evacuation under supercritical conditions show a more developed surface areas. Both aerogels and xerogels exhibit the tetragonal phase of zirconia and/or the zirconium-cerium solid solution phase. Aerogels presented more developed superficial loading of Ce4+ and higher acidity, which explain their good activity in n-hexane isomerization reaction in the whole temperature range investigated (Table 19.1). Ferino et al. [49] studied the catalytic performances of prepared zirconia xerogels and aerogels using zirconium n-propoxide as precursor, in the dehydration of methylpentan-2-ol reaction. They found that xerogel gives tetragonal zirconia upon calcination, during which a mesoporous system is formed. The crystal phase depends on the presence of oxygen during the cooling step in the case of the aerogel, whose texture is partially retained upon calcination.

19

468

L. Ben Hammouda et al.

Both kinds of catalysts have well-balanced concentrations of acid and base sites, but the acid sites are weaker in comparison with the basic ones. At 603 K the initial conversion of 4-methylpentan-2-ol over the calcined xerogel and aerogel was 45% and 63%, respectively. However, the selectivity to 4-methylpent-1-ene is 77% for both samples. Furthermore, these authors noted that the aerogel catalyst was quite stable during operation, whereas change in activity and selectivity were observed for the xerogel sample. Physicochemical and catalytic properties of sulfated zirconia aerogels doped with cerium were compared as a function of both the Ce/Zr molar ratio and the calcination temperature [65]. The aerogel with ratio Ce/Zr ¼ 0.4 calcined at different temperatures from 560 to 700
C exhibits the best catalytic activity in n-hexane isomerization in the reaction temperature range 170–220
C (Table 19.2) related to its more developed surface area and its important total acidity probably due to the presence of relevant sulfate groups. Other derived zirconia aerogel catalysts were also studied and tested in many reactions. Aerogel catalysts based on chromium supported by sulfated and unsulfated zirconia have been synthesized, in one step, by sol–gel method and dried in hypercritical solvent conditions [66]. Comparative study of their catalytic properties shows that dispersed Cr3+ seems to be the active species in the n-hexane aromatization reaction. However the acidity generated by sulfate groups acts as coke eliminator of the layers deposed on the surface mainly when catalyst is calcined at high temperature. Catalysts with different content of tungstophosphoric acid (TPA) on zirconia were prepared by suspending zirconia aerogel in an ethanol solution of TPA, removing the solvent

Table 19.1 Activity and selectivity of aerogel and xerogel sulfated zirconia doped with cerium in n-hexane isomerization reaction [64] Activity (108 mol g1 s1)

Samples AZC-560 AZC-650 XZC-560 XZC-650

Selectivity (%) Reaction temperature (
C) 170 200 220 88 96 98 96 99 99 0 0 0 0 90 91

Reaction temperature (
C) 170 200 220 7 50 113 3 45 142 0 0 0 0 1 9

via evaporation and then calcining the powder at 750
C. Catalytic performances were examined for the polymerization of tetrahydrofuran [67]. Corresponding results indicated that the strong interaction between TPA and zirconia retarded both the crystallization of zirconia and the destruction of the Keggin unit of TPA. In these materials, the major tungsten species were found to be zirconia-anchored heteropolytungstates and Zr-containing pseudo-heteropolyanions produced by the chemical bonding of Zr4+ with the WOx fragments from TPA decomposition as well as some amount of WO3. Prepared catalysts showed both Brönsted and Lewis acidity, and the catalyst with 20% TPA loading had the highest total acidity and catalytic activity because of the monolayer coverage of the active species. Under the reaction conditions of 40
C for 20 h, the most active catalyst with 20% TPA gave a high polymer yield of 30.9%. Aerogel and xerogel WO3/ZrO2 catalysts were prepared by sol–gel technique using two different modes of solvent drying: extraction in supercritical conditions (aerogel) and evaporation under vacuum at room temperature, respectively [68]. Two reactions of industrial interest were investigated under mild conditions, acylation of veratrole with acetic anhydride and acylation of anisole with benzoic anhydride. Results show that the solvent extraction strongly influences metal reducibility, surface area, pore organization, W/Zr surface density, and metallic interactions. The aerogel catalyst shows the best catalytic results for both conversion and yield. The supercritical drying solvent plays a central role especially in the recycling by proper air activation, and it attains the complete restoration of the catalytic activity even after three runs. Kamoun et al. [69] prepared aerogel and xerogel sulfated zirconia doped with nickel (AZSN and XZSN). These materials exhibit different textural, structural, and catalytic properties at various calcination temperatures. Obtained aerogel by drying under supercritical conditions of solvent exhibit a developed specific surface area and stabilize zirconia tetragonal phase before heating and even at high calcination temperature. However, xerogels obtained by ordinary drying in an oven are amorphous and have a low surface area and weak porosity. XPS spectroscopy shows that the nickel in aerogels is more reducible than in xerogels. Furthermore, aerogels exhibit higher activity than xerogels in the n-hexane isomerization reaction (Table 19.3).

Table 19.2 n-Hexane activity (108 mol g1 s1) [65] Temperature (
C) 560 600 650 700

AZC0.2 170 5.5 23.5 0.9 0.6

200 33.8 49.8 19.5 10

220 113 138.0 93.9 36

AZC0.4 170 6.9 57.9 13.6 6

200 49.4 66.7 103.4 11

220 125.9 148.9 210.1 41

AZC0.5 170 0.5 17.9 0.48 0.5

200 9.5 25.0 8.42 0.6

220 48.3 106.0 37.6 2.6

19

Zirconia Aerogels

469

Table 19.3 Catalytic properties of aerogel (AZSN) and xerogel (XZSN) catalysts [69]

Temperature (K) AZSN-773 AZSN-873 AZSN-923 XZSN-773 XZSN-873 XZSN-923

n-Hexane activity (108 mol g1 s1) 443 473 493 5.39 7.59 23.01 1.29 3.00 4.89 0.88 1.26 2.86 – – – – – – – – –

Products Selectivity (%) 443 473 493 88 86 86 87 86 85 88 87 85 – – – – – – – – –

Table 19.4 Textural properties of the aerogel AZrH and xerogel XZrH catalysts [70] Catalysts AZrH XZrH

2

SBET (m /g) 252 193

Dp (Å) 138 61

Dp: average pore diameter

Zirconia doped by heteropolytungstic acid (HPW) have been synthesized by sol–gel method using two drying techniques of the solvent evacuation [70]. Aerogel which is prepared by drying solvent in its hypercritical conditions exhibits a higher surface area and a higher average pore diameter compared to xerogel (Table 19.4). XRD results show that the aerogel develops ZrO2 tetragonal phase, whereas xerogel is amorphous. The thermal analysis studies show that the aerogel thermal stability is better than the xerogel one. The catalytic behavior of the two kinds of catalysts shows that the products of n-hexane isomerization depend on the acidity and the presence of carbide species. In the other works [71], this same research team compared characteristics and reactivity of aerogel zirconia catalyst doped with sulfates (AZrS) or heteropolytungstic acid (AZrH) in n-hexane isomerization in the temperature range 150–220
C [5]. Compared with sulfates, using HPW as dopant gives an active catalyst at 220
C and is selective toward isomers having higher individual octane number (ION), i.e., 2,2-DMB and 2,3-DMB (Table 19.5). The catalyst doped with sulfur deactivates at T > 200
C since agglomerates of zirconia particles favor the coke formation. Furthermore, the use of HPW as a precursor leads to an agglomeration of the tungsten and a formation of WO3 oxide on the surface of zirconia aerogel. This oxide appears to be responsible of a better selectivity towards isomers when the n-hexane isoimerization reaction was conducted at 220
C. Sulfated aerogel and xerogel zirconia catalysts doped with chromium are prepared by sol–gel method [72]. The effects of drying mode and calcination temperatures are reflected in the characteristics of the final solids. The addition of chromium to aerogel sulfated zirconia not only provides an important development of the specific surface area from

Table 19.5 n-Hexane isomerization over AZrS and AZrH catalysts [71]

Selectivity (%)/T(
C) 2,2-DMB 2,3-DMB 2-MP

Catalysts AZrS 150 175

200

220

AZrH 150 175

200

220

5 17 45

0 24 53

0 0 0

0 0 0

0 0 0

24 32 0

0 19 50

0 0 0

2,2-DMB, 2,2-dimethylbutane; 2,3-DMB, 2,3-dimethylbutane; 2-MP, 2-methylpentane

Table 19.6 Textural properties of ZSN and AlZSN catalysts [73] Catalysts ZSN-773 ZSN-873 ZSN-923 AlZSN-773 AlZSN-873 AlZSN-923

SBET(m2 g1) 101 72 76 105 153 158

Dp (Å) 211 262 191 149 128 127

Dp: average pore diameter

100 to 213 m2 g1 but also stabilizes a pure metastable ZrO2 tetragonal phase in a large calcination temperature domain. In addition, the aerogels develop active sulfate groups and preserve them on the surface even at high temperature. Those characteristics contribute to the improvement of catalytic performances of sulfated zirconia in n-hexane conversion. However, in the case of xerogels, the increase of temperature treatment leads to the loss of sulfate active groups and porosity destruction which make them completely inactive in spite of the development of ZrO2 tetragonal phase. In the continuity of their works, Kamoun et al. [73] prepared 2 a series of Ni/ZrO2-SO2 4 (ZSN) and Ni/ZrO2-Al2O3-SO4 (AlZSN) catalysts in one step by the sol–gel method and dried in hypercritical conditions of the solvent. Textural analysis reveals the mesoporosity of all the aerogel catalysts. Moreover, the addition of alumina to nickel sulfated zirconia at high calcination temperature increased twice the specific surface area, from 72 to 158 m2.g1 (Table 19.6). XRD patterns show that nickel-promoted sulfated zirconia calcined at different temperature develops the tetragonal and monoclinic ZrO2 phases, whereas the nickel sulfated zirconia alumina exhibits only the ZrO2 tetragonal phase (Fig. 19.4). The addition of aluminum to nickel sulfated zirconia induces significant changes in symmetry of nickel by the migration of Ni ions from octahedral to tetrahedral coordination. XPS spectroscopy shows that the nickel in Ni/ZrO2SO2 catalysts is more reducible than those in Ni/ZrO24 Al2O3–SO2 4 . Nickel sulfated zirconia catalyst exhibits higher activity than nickel sulfated zirconia alumina, in the n-hexane isomerization reaction (Fig. 19.5).

19

470

L. Ben Hammouda et al.

A high-surface-area zirconia aerogel catalysts were also successfully used by Yunsu et al. [74] in the ketonization of hexanoic acid to 6-undecanone. High conversion, 72.3%, and high selectivity, 92.6%, were obtained. The optimization of the porous structures of zirconia led to possibly improved mass transfer with approximately 35 nm of pores, enhancing the catalytic activity of zirconia. Catalytic performances in esterification of stearic acid with methanol showed that aerogel sulfated zirconia solid acid catalysts exhibited 88% yield of methyl stearate at 60
C after 7 h and significant higher reaction rate

T

M

Intensity (a.u.)

d c

b a 5

15

25

35 45 2q (degree)

55

65

Fig. 19.4 XRD patterns of catalysts calcined at different temperatures: (a) ZSN, 773 K; (b) ZSN, 923 K; (c) AlZSN, 773 K; and (d) AlZSN, 923 K [73]

(18.94 mmol h1 g1) and TOF (8.55 h1) than the xerogel one (12.94 mmol h1 g1 and 3.92 h1, respectively) [75]. Aerogel sulfated zirconia catalyst also exhibited higher activity than other heterogeneous acid catalysts such as ion exchange resins, Nafion, and acid clay under the similar reaction conditions; and it showed comparable activity with conventional Brönsted (H2SO4) and Lewis (ZrOCl2) acids indicating its potential to replace the homogeneous acid catalysts. Aerogel and xerogel Ni-Sr-Al2O3-ZrO2 catalysts were prepared by a one-step epoxide-driven sol–gel method and a subsequent supercritical CO2 drying method [76]. Aerogel catalyst retained higher surface area and larger pore volume than the xerogel one and also showed strong interaction between active metal and support, which led to small nickel particles after the reduction. H2temperature-programmed desorption (H2-TPD) results revealed that aerogel catalyst with smaller nickel particle size had higher nickel dispersion and higher nickel surface area. Performance and stability of the catalyst in the ethanol steam-reforming reaction were enhanced with the employment of supercritical drying. Enhanced physicochemical property and high nickel surface area of aerogel catalyst led to excellent catalytic activity and stability in the hydrogen production by steam reforming of ethanol. Recently works of Ben Nsir et al. [77] show that the evacuation of the solvent at its supercritical conditions in the preparation of catalyst based on the aerogel zirconiadoped phosphate ion with nP/nZr ratio equal to 0.05 (AZrP0.05) promotes the development of both tetragonal phase of zirconia and pore size but slows that of the phases related to the ZrP species (Fig. 19.6). These characteristics do not seem to favor reactivity in the esterification of acetic acid by ethanol.

Fig. 19.5 Activity versus time on stream at 443 K [73]

Isomerisation activity (nmol.g–1.s–1)

60

ZSN

AlZSN

50 40 30 20 10 0

5

22

39

56

73

90 107 124 141 158 175 192 209 223 Time (min)

19

Zirconia Aerogels

700

471

t-ZrO2 and c-ZrO2

600 Intensity (a.u.)

XZrP0.05 AZrP0.05

t-ZrP

500

m-ZrP

400 300

Table 19.7 Effect of heat treatment temperature on phase formation of nYSZ aerogels [37] Heat treatment temperature (
C) 400 600 800 1000 1200

Tetragonal phase (%) 0YSZ 4YSZ 100 100 53 100 30 100 4 95 0 67

6YSZ 100 100 100 100 97

8YSZ 100 100 100 100 100

200 100 0

0

10

20

30 40 50 2q (degree)

60

70

Fig. 19.6 XRD patterns of the catalysts XZrP0.05 and AZrP0.05 [77]

19.4.2 Zirconia Aerogels and Ceramics Chao et al. [37] reported that the sol–gel method was successfully applied to the synthesis of crack-free, monolithic yttrium-stabilized zirconia aerogels (YSZ) which still keep the whole block without collapse, even after heat treatment at 1000
C for 2 h. At the same time, they reported that doping ZrO2 with 8 wt% Y2O3 (8YSZ) stabilized the tetragonal phase even after heat treatment at 1200
C for 2 h. However, it should be noted that the tetragonal phase begins to turn into monoclinic, in the case of pure zirconia aerogel (0YSZ), at about 600
C. Moreover, the average crystallite size decreases with the increase in yttrium content at high temperature. Obtained results showed clearly that the addition of yttrium improves considerably the thermal stability of zirconia aerogel. Indeed, the pure zirconia aerogel exhibits a densified structure characterized by a very low specific surface area (SBET ¼ 8 m2.g1) and a relatively large crystallite size (40 nm), while the 8YSZ aerogel presents a more porous structure with specific surface area and elementary crystallite size estimated at SBET ¼ 28 m2. g1and 25 nm, respectively. In their study, these same authors have confirmed that the monolithic YSZ aerogel can be successfully prepared by adding formamide to control the micellar morphology and size of colloid since the addition of formamide affects the skeleton structure of the gel. The obtained YSZ aerogel shrinks by about 20%, and no cracking occurs after heat treatment at 1000
C for 2 h. However, this is not the case of the pure zirconia aerogel which crack into many small pieces when it is heat-treated at the same temperature. On the other hand and compared with pure zirconia aerogel, the

particle size of the monolithic 8YSZ aerogel is smaller and homogeneously distributed. This result confirms once again that the addition of yttrium prevents the growth of the aerogel grains. Moreover, the improvement of the stability of tetragonal phase by increasing the Y2O3 content (Table 19.7) may be explained by an increase of the concentration of the oxygen vacancy. It is also important to note that obtained monolithic YSZ aerogels exhibit higher porosity compared to similar ceramics made by classical techniques. Zirconia is one of the bio-inert materials which has attracted the most attention because of its stability in the human body without releasing Zr species or provoking a severe tissue response [78]. Since the 1960s, research related to the exceptional characteristics of zirconia has steadily progressed. The first use of zirconium oxide for medical purposes was made in 1969 in orthopedic as a new material for hip head replacement instead of titanium or alumina prostheses [79]. Moreover, it has been reported that based on its excellent physical strength, thermal stability, chemical inertness, low toxicity, enhanced esthetic, and high biocompatibility zirconia ceramic may be used in a wide range of promising clinical applications [80–82]. Recently, aerogel-based slurries with yttrium-stabilized zirconia as a secondary ceramic were prepared and spray-dried according to modified Taguchi experimental design in order to appreciate the effect of both the slurry formulation and drying conditions such as the solid content, the ratio of yttriumstabilized zirconia aerogel added, the amount of dispersant and binder, inlet temperature, atomization pressure, and feeding rate on the median particle size of the resulting spray-dried powder.

19.4.3 Electrochemical Applications Zhou et al. [83] indicated that, because of its high stability and its ionic transference close to the unity, YSZ represents the material which exhibit the most potential to be used as an electrolyte in solid oxide fuel cells (SOFCs). Recently, graphene/ZrO2 composite aerogels were prepared using a simple and inexpensive synthesis method by Guo et al. [84]. In this process the graphene sheets were

19

472

decorated with ZrO2 nanoparticles. The obtained composite aerogels were characterized by a low density, a mesoporous structure with an average pore diameter estimated at 13 nm, and a large specific surface area situated between 380 and 490 m2 g1. On the other hand, the graphene/ZrO2 composite aerogels reveal a high thermal stability accompanied with a low thermal conductivity at room temperature and an interesting electrochemical performances. The synthesis of the graphene/ZrO2 composite aerogels was made using a fixed amount of ZrOCl2;8H2O which was added, respectively, to a graphene oxide (GO)/DMF solution to form a uniform mixture. A fixed amount of epichlorohydrin (PPO) was then added under stirring. Finally and after aging the prepared gel during 3 days, the graphene/ ZrO2 composite aerogels were obtained following a solvent exchange with ethanol, supercritical fluid drying with CO2, and then carbonization under 500
C for 2 h in an Ar atmosphere. It has been verified that the presence of reduced graphene oxide in the graphene/ZrO2 composite aerogels promotes electronic interaction with ZrO2, which enhances the conductivity of the composite aerogels. To further study the electrochemical energy storage of the composite aerogels, electrochemical investigations have been conducted. It has been noted that for the composite aerogel with a mass ratio of ZrO2 to graphene of 0.7, even when the scan rate increases to 100 mV. s1, the cyclic voltammograms of the composite aerogel basically remain as rectangle shapes with some deviation at lower potential, implying a quick charge propagation capability of both double-layer capacitance and pseudo-capacitance. Meanwhile, the rate performances of the composite aerogels were also evaluated by galvanostatic charge/discharge under an enhanced current density. Otherwise, the composite aerogels show good rate capability, since the capacitance of the composite aerogel decreases slowly with an increase of the current density. In fact, when the current density is 0.5 A.g1, the specific capacitance of the composite aerogel is 117 F.g1. However, when the current density increases to as high as 10 A.g1, the specific capacitance of the composite aerogel still remains at 52 F.g1. Furthermore, the electrochemical impedance spectroscopy (EIS) plots of the cell based on the composite aerogel show a typical electric double-layer capacitive behavior. The intercept of the composite aerogel is at 0.471 U.cm2, indicating that the resistance of the cell based on the composite aerogel is not very high. The cycle performance of the capacitor based on the graphene/ZrO2 composite aerogel is relatively stable. However, for the composite aerogels with the higher mass ratios of ZrO2 to graphene, the electrochemical energy storage performances are not the same as that of the composite aerogel with a mass ratio of ZrO2 to graphene of 0.7.

L. Ben Hammouda et al.

19.4.4 Zirconia Aerogels and Optics Among the metal oxides, zirconia is an essential material in optics due to its excellent optical properties, such as a high refractive index, large optical bandgap, low optical loss, and high transparency in the visible and near-infrared regions. These properties explain the use of ZrO2 for high-refractive mirrors, broadband interference filters, and active electrooptical devices [85–87]. Additionally, zirconium oxide is the most frequently used material for antireflection coatings in optical industries. It has been shown that the particle size and shape of nanomaterials have a great impact on their characteristics. So, exploring suitable methods to prepare zirconia and controlling its morphology and particle size are important and necessary.

19.5

Conclusion

Zirconia aerogels have attracted special attention from many research groups viewing their exceptional physicochemical properties as well as their various fields of application. Typically, zirconia aerogels were obtained by removing the solvent by a supercritical drying using one of two different methods: high-temperature supercritical drying or low-temperature extraction with supercritical CO2. This process offers a major advantage, since it considerably reduces the surface tension as well as the capillary pressure leading thus to solids with a small shrinkage of the pore volume. Nevertheless, despite that several research groups reported that zirconia aerogels obtained by high-temperature supercritical drying exhibit developed surface areas, large pore volumes, and a temperature-stable tetragonal phase, this process seems to be expensive enough, relatively risky when it is set up on an industrial scale, and also difficult to control. It has been also reported that zirconia aerogels with high surface areas were successfully prepared by a combined electrolysis/sol–gel method, followed by supercritical extraction or freeze-drying or by ambient pressure drying. Moreover, it is important to point out that in the interest of continuous improvement of the physicochemical properties, several works dedicated to the zirconia aerogels prove that these properties are closely related and highly sensitive to any modification of the different preparation parameters. It was confirmed first that the nature of the zirconium precursor and its concentration affect the obtained solid. Indeed, the main research results show that when the precursor concentration is between 0.1 M and 0.5 M, the final solid has a better specific surface area and a well-developed porosity. Furthermore, most of the works dedicated to the zirconia aerogels use the sol–gel process for their synthesis and indicated that the control of the hydrolysis and condensation steps is crucial in order to obtain a promising solid. The hydrolysis ratio defined by H2O/Zr report

19

Zirconia Aerogels

is an important parameter which influences the hydrolysis and condensation mechanisms and consequently not only the size of the molecular species resulting from these reactions but also their structures, textures, and physical properties. In the same time, condensation seems to be governed by the choice of the acid precursor and its concentration which have a crucial impact on the morphology and the microporous structure of the resulting aerogels. In a certain concentration range, quick formation of a rigid polymeric gel containing some precipitate particulates is observed. On the other hand, long alcogel aging time has been recommended for the synthesis of zirconia aerogels using sol–gel process. In fact, prolonging aging time enhances slightly the final surface areas of obtained solids without otherwise affecting the porous distribution. Recently, the use of epoxides as proton scavenger from zirconium inorganic salts in polar and protic solvents has been advanced as a new path to prepare the aerogels. This preparation technique leads to monolithic gels, which vary from opaque to translucent according to the hydrolysis ratio. Furthermore, most of the works devoted to zirconia aerogels reveal that the desired tetragonal phase can only exist at a room temperature through stabilization. This is normally achieved by adding a dopant such as yttrium or by modifying the surface with sulfuric or phosphoric acid. Zirconia aerogels reveal several properties required for applications as catalysts or catalyst support. In fact, they exhibit adequate structural and textural characteristics and their surfaces are known for their acidic and basic sites as well as for their oxidizing and reducing chemical properties. These properties confer a good dispersion of a catalytically active phase, which enhances catalytic activity, thermal stability, and resistance to poisoning. We must also underline that zirconia aerogels are promising supports for active metallic centers such as platinum, copper, nickel, yttrium, iron, rhodium, cerium, and chromium and/or oxoanions such as sulfate, phosphate, and tungstate anions, allowing the development of efficient catalysts for several reactions such as hydrogenation, isomerization, oxidation, dehydration, acylation, ketonization, esterification, and Fischer-Tropsch synthesis reaction.

References 1. Yamaguchi, T.: Application of ZrO2 as a catalyst and a catalyst support. Catal. Today. 20, 199–217 (1994) 2. Mercera, P.D.L., Van Ommen, J.G., Doesburg, E.B.M., Burggraaf, A.J., Ross, J.R.H.: Zirconia as a support for catalysts: evolution of the texture and structure on calcination in air. Appl. Catal. 57, 127–148 (1990) 3. Morterra, C., Cerrato, G., DiCiero, S., Signoretto, M., Pinna, F., Strukul, G.: Platinum-promoted and unpromoted sulfated zirconia catalysts prepared by a one-step aerogel procedure: 1. Physicochemical and morphological characterization. J. Catal. 165, 172–183 (1997)

473 4. Signoretto, M., Oliva, L., Pinna, F., Strukul, G.: Synthesis of sulfated-zirconia aerogel: effect of the chemical modification of precursor on catalyst porosity. J. Non-Cryst. Solids. 290, 145–152 (2001) 5. Cimino, S., Pirone, R., Lisi, L.: Zirconia supported LaMnO3 monoliths for the catalytic combustion of methane. Appl Catal B. 35, 243–254 (2002) 6. Bahamonde, A., Campuzano, S., Yates, M., Salerno, P., Mendioroz, S.: Influence of zirconia raw materials on the development of DeNOx monolithic catalysts. Appl Catal B. 44, 333–346 (2003) 7. Yamahara, K., Sholklapper, T.Z., Jacobson, C.P., Visco, S.J., Jonghe, L.C.: Ionic conductivity of stabilized zirconia networks in composite SOFC electrodes. Solid State Ionics. 176, 1359–1364 (2005) 8. Blaese, D., Garcia, D.E., Guglielmi, P., Hotza, D., Fredel, M.C., Janssen, R.: ZrO2 fiber-matrix interfaces in alumina fiber-reinforced model composites. J. Eur. Ceram. Soc. 35, 1593–1598 (2015) 9. Somiya, S., Yamamoto, N., Yanagida, H.: Advances in Ceramics, Science and Technology of Zirconia III 24. American Ceramic Society, Tokyo (1988) 10. Ji, Y., Kilner, J.A., Carolan, M.F.: Electrical properties and oxygen diffusion in yttria-stabilised zirconia (YSZ)-La0.8Sr0.2MnO3d (LSM) composites. Solid State Ionics. 176, 937–943 (2005) 11. Rodrigues, C.M.S., Labrincha, J.A., Marques, F.M.B.: Postmortem characterization of one yttria stabilized zirconia (YSZ)-based oxygen sensor. Solid State Ionics. 136–137, 671–675 (2000) 12. Mejri, I., Younes, M.K., Ghorbel, A., Eloy, P., Gaigneaux, E.: Comparative study of the sulfur loss in the xerogel and aerogel sulfated zirconia calcined at different temperatures: effect on n-hexane isomerization. Stud. Surf. Sci. Catal. 162, 953–960 (2006) 13. Ward, D.A., Ko, E.I.: One-step synthesis and characterization of zirconia-sulfate aerogels as solid superacids. J. Catal. 150, 18–33 (1994) 14. Pajonk, G.M.: Aerogel catalysts. Appl. Catal. 72, 217–266 (1991) 15. Chen, L., Hu, J., Richards, M.R.: Catalytic properties of nanoscale iron-doped zirconia solid-solution aerogels. Chem. Phys. Chem. 9, 1069–1078 (2008) 16. Chervin, C.N., Clapsaddle, B.J., Chiu, H.W., Gash, A.E., Satcher Jr., J.H., Kauzlarich, S.M.: Role of cyclic ether and solvent in a non-alkoxide sol-gel synthesis of yttria-stabilized zirconia nanoparticles. Chem. Mater. 18, 4865–4874 (2006) 17. Conroy, J.F.T., Hosticka, B., Davis, S.C., Smith, A.N., Norris, P.M.: Microscale thermal relaxation during acoustic propagation in aerogel and other porous media. Microscale Thermophys. Eng. 3, 199–215 (1999) 18. Singhal, A., Toth, L.M., Lin, J.S., Affholter, K.: Zirconium (IV) tetramer/octamer hydrolysis equilibrium in aqueous hydrochloric acid solution. J. Am. Chem. Soc. 118, 11529–11534 (1996) 19. Bangi, U.K.H., Jung, H.N.R., Park, C.S., Mahadik, D.B., Park, H. H.: Effect of thermal treatment on the textural properties and thermal stability of surface modified zirconia aerogel powders. Int. J. Nanotechnol. 13, 452–462 (2016) 20. Woignier, T., Phalippou, J., Despetis, F., Calas-Etienne, S.: Handbook of Sol-Gel Science and Technology. (2016). https://doi.org/10. 1007/978-3-319-19454-7_27-1 21. Zhong, L., Chen, X., Song, H., Guo, K., Hu, Z.: Synthesis of monolithic zirconia aerogel via a nitric acid assisted epoxide addition method. RSC Adv. 60, 31666–31671 (2014) 22. Zhao, Z., Chen, D., Jiao, X.: Zirconia aerogels with high surface area derived from sols prepared by electrolyzing zirconium oxychloride solution: Comparison of aerogels prepared by freeze-drying and supercritical CO2(l) extraction. J. Phys. Chem. C. 111, 18738–18743 (2007) 23. Schäfer, H., Brandt, S., Milow, B., Ichilmann, S., Steinhart, M., Ratke, L.: Zirconia-based aerogels via hydrolysis of salts and Alkoxides: the

19

474 influence of the synthesis procedures on the properties of the aerogels. Chem. Asian J. (2013). https://doi.org/10.1002/asia.201300488 24. Li, X., Jiao, Y., Ji, H., Sun, X.: The effect of propylene oxide on microstructure of zirconia monolithic aerogel. Integr. Ferroelectrics. 146, 122–126 (2013) 25. Bedilo, A.F., Klabunde, K.J.: Synthesis of high surface area zirconia aerogels using high temperature supercritical drying. Nanostruct. Mater. 8, 119–135 (1997) 26. Stöcker, C., Schneider, M., Baiker, A.: Zirconia aerogels and xerogels: influence of solvent and acid on structural properties. J. Porous. Mater. 2, 171–183 (1995) 27. Stöcker, C., Schneider, M., Baiker, A.: Zirconia aerogels: effect of acid used in the sol-gel process on morphological properties. J. Porous. Mater. 2, 325–330 (1995) 28. Hu, J., Cao, Y., Deng, J.: A simple alcohothermal synthetic route to high surface area zirconia aerogel. Chem. Lett. 30, 398–399 (2001) 29. Wu, Z.G., Zhao, Y.X., Xu, L.P., Liu, D.S.: Preparation of zirconia aerogel by heating of alcohol-aqueous salt solution. J. Non-Cryst. Solids. 330, 274–277 (2003) 30. Ward, D.A., Ko, E.I.: Synthesis and structural transformation of zirconia aerogels. Chem. Mater. 5, 956–969 (1993) 31. Suh, D.J., Park, T.J.: Sol-gel strategies for pore size control of highsurface-area transition-metal oxide aerogels. Chem. Mater. 8, 509–513 (1996) 32. Teichner, S.J., Nicolaon, G.A., Vicarini, M.A., Gardes, G.E.E.: Inorganic oxide aerogels. Adv. Colloid Interf. Sci. 5, 245–2731 (1976) 33. Stöcker, C., Baiker, A.: Zirconia aerogels: effect of acid-to-alkoxide ratio, alcoholic solvent and supercritical drying method on structural properties. J. Non-Cryst. Solids. 223, 165–178 (1998) 34. Biswas, M., Ojha, P.K., Jayasingh, E.M., Prasad, C.D.: Synthesis of nanocrystalline yttria stabilized zirconia for SOFC. Nanomater. Nanotechnol. 1, 55–58 (2011) 35. Sui, R., Rizkalla, A.S., Charpe, P.A.: Direct synthesis of zirconia aerogel nanoarchitecture in supercritical CO2. Langmuir. 22, 4390–4396 (2006) 36. Tyagi, B., Sidhpuria, K., Shaik, B., Jasra, R.V.: Synthesis and characterization of nanocrystalline mesoporous zirconia using supercritical drying. J. Nanosc. Nanotech. 6, 1584–1159 (2006) 37. Chao, X., Yuan, W., Shi, Q., Zhu, Z.: Improvement of thermal stability of zirconia aerogel by addition of yttrium. J. Sol-Gel Sci. Techn. 80, 667–674 (2016) 38. Chervin, C.N., Clapsaddle, B.J., Chiu, H.W., Gash, A.E., Satcher Jr., J.H., Kauzlarich, S.M.: Aerogel synthesis of yttria-stabilized zirconia by a non-alkoxide sol-gel route. Chem. Mater. 17, 3345–3351 (2005) 39. Zhang, H.X., He, X.D., Li, Y., Hong, C.Q.: Preparation of nanoporous zirconia aerogel. J. Aeronau. Mater. 26, 337–338 (2006) 40. Vesteghem, H., Jaccon, T., Lecomte, A.: Synthesis and microstructure of zirconia aerogel. Rev. Phys. Appl. 50(C4), 59–64 (1989) 41. Ayral, A., Assih, T., Abenoza, M., Phalippou, J., Lecomte, A., Dauger, A.: Zirconia by the gel route. J. Mater. Sci. 25, 1268–1274 (1990) 42. Vicarini, M.A., Nicolaon, G.A., Teichner, S.J.: Textural and structural properties of titania and zirconia aerogel, prepared by hydrolyzing corresponding alcogolates in homogeneous or heterogenous. Bull. Soc. Chim. Fr. 107, 1651 (1970) 43. Ko, E.I.: Aerogels as catalysts and catalysts supports. Chemtech. 23(4), 31 (1993) 44. Benedetti, A., Fagherazzi, G., Riello, P., Zeng, Y.W., Pinna, F., Signoretto, M.: Fractal properties of a partially crystalline zirconium oxide aerogel. J. Appl. Crystalogr. 26, 717–720 (1993) 45. Schneider, M., Baiker, A.: Aerogels in catalysis. Catal. Reviews. 37, 515–556 (1995)

L. Ben Hammouda et al. 46. Wang, Q., Li, X., Fen, W.: Synthesis of crack-free monolithic ZrO2 aerogel modified by SiO2. J. Porous. Mater. 21, 127–130 (2014) 47. Fenech, J., Viazzi, C., Bonino, J., Ansart, F., Barnabé, A.: Morphology and structure of YSZ powders: comparison between xerogel and aerogel. Ceram. Int. 35, 3427–3433 (2009) 48. Skovgaard, M., Almdal, K., Van Lelieveld, A.: Stabilization of metastable tetragonal zirconia nanocrystallites by surface modification. J. Mater. Sci. 46, 1824–1829 (2011) 49. Ferino, I., Casula, M.F., Corrias, A., Cutrufello, M.G., Monaci, R., Paschina, G.: 4 Methylpentan-2-ol dehydration over zirconia catalysts prepared by sol-gel. Phys. Chem. Chem. Phys. 2, 1847–1854 (2000) 50. Pajonk, G.M., El Tanany, A.: Isomerization and hydrogenation of butene-1 on a zirconia aerogel catalyst. React. Kine. Catal. Letters. 47, 167–175 (1992) 51. Kalies, H., Pinto, N., Pajonk, G.M., Bianchi, D.: Hydrogenation of formate species formed by CO chemisorption on a zirconia aerogel in the presence of platinum. Appl. Catal. A. 202, 197–205 (2000) 52. Sun, Y., Sermon, P.A.: Cu-doped ZrO2 aerogel: a novel catalyst for CO hydrogenation to CH3OH. Topic. Catal. 1, 145–151 (1994) 53. Sermon, P.A., Self, V.A., Sun, Y.: Doped-ZrO2 aerogels: catalysts of controlled structure and properties. J. Sol-Gel Sci. Techn. 8, 851–856 (1997) 54. Zhang, Y., Xiang, H., Zhong, B., Wang, Q.: ZrO2-SiO2 aerogel supported cobalt catalysts for the synthesis of long-chain hydrocarbon from syngas. Petro. Sci. Techn. 17, 981–998 (1999) 55. Orlovic, A.M., Janackovic, D.T., Skala, D.U.: Aerogels in Catalysis New Developments in Catalysis Research. Nova Science Publishers, Hauppauge (2005) 56. Boyse, R.A.: Preparation and characterization of zirconia-phosphate aerogels. Catal. Letters. 38, 225–230 (1996) 57. Younes, M.K., Ghorbel, A., Rives, A., Hubaut, R.: Study of acidity of aerogels ZrO2-SO2 4 by isopropanol dehydration reaction, surface potential and X-ray photoelectron spectroscopy. J. Sol-Gel Sci. Techn. 19, 817–819 (2000) 58. Younes, M.K., Ghorbel, A., Rives, A., Hubaut, R.: Acidity of sulphated zirconia aerogels: correlation between XPS studies, surface potential measurements and catalytic activity in isopropanol dehydration reaction. J. Sol-Gel Sci. Techn. 32, 349–352 (2004) 59. Bedilo, A.F., Klabunde, K.J.: Synthesis of catalytically active sulfated zirconia aerogels. J. Catal. 176, 448–458 (1998) 60. Akkari, R., Ghorbel, A., Essayem, N., Figueras, F.: Synthesis and characterization of mesoporous silica supported nano-crystalline sulfated zirconia catalysts prepared by a sol-gel process: effect of the S/Zr molar ratio. Appl. Catal. A. 328, 43–51 (2007) 61. Raissi, S., Younes, M.K., Ghorbel, A.: Synthesis and characterization of aerogel sulphated zirconia doped with chromium: n-hexane isomerization. J. Porous. Mater. 17, 275–281 (2010) 62. Boyse, R.A.: Crystallization behaviour of tungstate on zirconia and its relationship to acidic properties: I. Effect of preparation parameters. J. Catal. 171, 191–207 (1997) 63. Boyse, R.A., Ko, E.I.: Crystallization behaviour of tungstate on zirconia and its relationship to acidic properties: II. Effect of silica. J. Catal. 179, 100–110 (1998) 64. Mejri, I., Younes, M.K., Ghorbel, A., Eloy, P., Gaigneaux, E.M.: Effect of the evacuation mode of solvent on the textural, structural and catalytic properties of sulfated zirconia doped with cerium. Stud. Surf. Sci. Catal. 174, 493–496 (2008) 65. Mejri, I., Younes, M.K., Ghorbel, A., Eloy, P., Gaigneaux, E.M.: Characterization and reactivity of aerogel sulfated zirconia-ceria catalyst for n-hexane isomerization. J. Porous. Mater. 17, 545–551 (2010)

19

Zirconia Aerogels

66. Raissi, S., Younes, M.K., Ghorbel, A., Garin, F.: Effect of sulphate groups on catalytic properties of chromium supported by zirconia in the n-hexane aromatization. J. Sol-Gel Sci. Techn. 53, 412–417 (2010) 67. Zhu, Q., Liang, L.-P., Jia, Z.-Q., Gao, C.-G., Zhao, Y.-X.: Polyoxotungstate-modified zirconia aerogel as a solid catalyst for tetrahydrofuran polymerization. J. Acta. Phys. Chem. Sin. 27(491–498) (2011) 68. Signoretto, M., Ghedini, E., Menegazzo, F., Cerrato, G., Crocellà, V., Letizia, C.B.: Aerogel and xerogel WO3/ZrO2 samples for fine chemicals production. Microporous Mesoporous Mater. 165, 134–141 (2013) 69. Kamoun, N., Younes, M.K., Ghorbel, A., Mamede, A.S., Rives, A.: Effect the solvent evacuation mode on the catalytic properties of nickel-modified sulfated zirconia catalysts: n-hexane isomerization. React. Kinet. Mech. Catal. 11, 199–213 (2014) 70. Chakhari, S., Younes, M.K., Rives, A., Ghorbel, A.: Textural, structural and catalytic properties of zirconia doped by heteropolytungstic acid: a comparative study between aerogel and xerogel catalysts. J. Sol-Gel Sci. Techn. 69, 378–385 (2014) 71. Chakhari, S., Younes, M.K., Rives, A., Ghorbel, A.: Effect of the doping agent nature on the characteristic and catalytic properties of aerogel zirconia catalysts doped with sulfate groups or heteropolytungstic acid. J. Mater. Res. 72, 35–42 (2015) 72. Raissi, S., Kamoun, N., Younes, M.K., Ghorbel, A.: Effect of drying conditions on the textural, structural and catalytic properties of Cr/ZrO2-SO2 4 : n-hexane conversion. React. Kinet. Mech. Catal. 115, 499–512 (2015) 73. Kamoun, N., Younes, M.K., Ghorbel, A., Mamede, A.S., Rives, A.: Comparative study of aerogels nanostructured catalysts: Ni/ZrO22 SO2 4 and Ni/ZrO2-Al2O3-SO4 . Ionics. 21, 221–229 (2015) 74. Lee, Y., Choi, J.W., Suh, D.J., Ha, J.-M., Lee, C.-H.: Ketonization of hexanoic acid to diesel-blendable 6-undecanone on the stable zirconia aerogel catalyst. Appl. Catal. A Gen. 506, 288–293 (2015) 75. Saravanan, K., Tyagi, B., Bajaj, H.C.: Nano-crystalline, mesoporous aerogel sulfated zirconia as an efficient catalyst for esterification of stearic acid with methanol. Appl. Catal. B. Env. 192, 161–170 (2016) 76. Song, J.H., Han, S.J., Yoo, J., Park, S., Kim, D.H., Song, I.K.: Hydrogen production by steam reforming of ethanol over Ni-SrAl2O3-ZrO2 aerogel catalyst. J. Mol. Catal. A Chem. 424, 342–350 (2016) 77. Ben Nsir, S., Younes, M.K., Rives, A., Ghorbel, A.: Characterization and reactivity of zirconia-doped phosphate ion catalyst prepared by sol–gel route and mechanistic study of acetic acid esterification by ethanol. J. Sol-Gel Sci. Techn. 84, 349–360 (2017) 78. Posada, O.M., Tate, R.J., Grant, M.H.: Toxicity of cobalt–chromium nanoparticles released from a resurfacing hip implant and cobalt ions on primary human lymphocytes in vitro. J. Appl. Toxicol. 35, 614–622 (2015) 79. Evans, A.G., Heuer, A.H.: Transformation toughening in ceramics: martensitic transformations in crack-tip stress fields. J. Am. Ceram. Soc. 63, 241–248 (1980) 80. Tiainen, H., Eder, G., Nilsen, O., Haugen, H.J.: Effect of ZrO2 addition on the mechanical properties of porous TiO2 bone scaffolds. Mater. Sci. Eng. C. 32, 1386–1393 (2012) 81. Liu, H., Li, S., Li, Q., Li, Y., Zhou, W.: Microstructure, phase stability and thermal conductivity of plasma sprayed Yb2O3, Y2O3 co-stabilized ZrO2 coatings. Solid State Sci. 13, 513–519 (2011) 82. Ortiz, A.L., Sánchez-González, J., González-Méndez, L.M., Cumbrera, F.L.: Determination of the thermal stability and isothermal bulk modulus of the ZrO2 polymorphs at room temperature by molecular dynamics with a semi-empirical quantum-chemical model. Ceram. Int. 33, 705–709 (2007)

475 83. Zhou, Y., Guan, X., Zhou, H., Ramadoss, K., Adam, S., Liu, H., Lee, S., Shi, J., Tsuchiya, M., Fong, D.D., Ramanathan, S.: Strongly correlated perovskite fuel cells. Nature. 534, 231–234 (2016) 84. Guo, D., Lu, Y., Zhao, Y., Zhang, X.: Synthesis and physicochemical properties of graphene/ZrO2composite aerogels. RSC Adv. 5, 11738–11744 (2015) 85. Zhao, S., Ma, F., Song, Z., Xu, K.: Thickness-dependent structural and optical properties of sputter deposited ZrO2 films. Opt. Mater. 30, 910–915 (2008) 86. Lai, L.J., Lu, H.C., Chen, H.K., Cheng, B.M., Lin, M.I., Chu, T.C.: Photoluminescence of zirconia films with VUV excitation. J. Electron. Spectros. Relat Phenomena. 144, 865–868 (2005) 87. Zinatloo-Ajabshir, S., Salavati-Niasari, M., Hamadanian, M.: Praseodymium oxide nanostructures: novel solvent-less preparation, characterization and investigation of their optical and photocatalytic properties. RSC Adv. 5, 33792–33800 (2015)

19

Lassaad Ben Hammouda is an assistant professor in chemistry at the Faculty of Sciences of Tunis where he studied chemistry and also obtained his PhD in 2003. His research interests are focused on the synthesis and characterization of modified zirconia catalysts and their potential applications as acid catalysts or supports in different reactions such as isomerization of n-alkane, esterification, and transesterification in addition to the biomass valorization.

Imen Mejri studied chemistry at the Faculty of Sciences of Tunis where she obtained her PhD in chemistry in 2009. Currently she is an assistant professor in chemistry at the High Institute of Medical Technology in Tunis. Her research focuses on the development of oxides synthesized by sol–gel route. Currently she is working in the field of selective catalytic reduction of NOx.

476

Mohamed Kadri Younes is a professor of chemistry and member of the Laboratory of Chemistry of Materials and Catalysis at the Faculty of Sciences of Tunis – Tunisia. He obtained his PhD in 1994 and the university habilitation in 2000 in chemistry of materials and catalysis. Now he is interested with his team in the field of “acid catalysis and its applications.”

L. Ben Hammouda et al.

Abdelhamid Ghorbel carried out research work at IRCELYON to obtain his Thèse d’Etat at the Université Claude Bernard Lyon 1 (France) in 1974. He was then professor at the Faculté des Sciences de Tunis to teach physical chemistry and catalysis courses. He initiated in 1974 the creation of the Laboratoire de Chimie des Matériaux et Catalyse (LCMC) that he managed until the end of 2011 and directed the works of many researchers on themes related to energy and environmental chemistry. He contributed also to several reflections on higher education and scientific research. Currently he is professor emeritus at the Université Tunis El Manar.

Part IV Synthetic Polymer Aerogels

Phenolic Aerogels and Their Carbonization

20

Chariklia Sotiriou-Leventis, Nicholas Leventis, and Sudhir Mulik

Contents 20.1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 480

20.7

20.2 20.2.1 20.2.2

The Resorcinol-Formaldehyde (RF) Paradigm . . . . . . . 481 The Chemistry of Base-Catalyzed Gelation [37] . . . . . . . . 481 The Chemistry of Acid-Catalyzed Gelation [25, 40] . . . . 481

20.7.1

20.3 20.3.1 20.3.2

RF Aerogels Via Base Catalysis . . . . . . . . . . . . . . . . . . . . . . . . 483 Typical Synthesis of RF Aerogels Via Base Catalysis . . . 483 Factors Affecting the Structure and Properties of RF Aerogels Via Base-Catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . 484

20.4

RF Aerogels Via Acid Catalysis . . . . . . . . . . . . . . . . . . . . . . . . 486

20.5

Comparison of RF Aerogels Via the Base- Versus the Acid-Catalyzed Routes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 488 Chemical Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 488 Morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489

20.5.1 20.5.2 20.6 20.6.1 20.6.2 20.6.3 20.6.4 20.6.5

The Emerging Polybenzoxazine (PBO) Paradigm of Phenolic Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PBO Aerogels Via Thermal Gelation of Ishida’s BO Monomer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PBO Aerogels Via Acid-Catalyzed Gelation of Ishida’s BO Monomer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Comparison of PBOs Obtained Via Thermal Gelation Versus the Acid-Catalyzed Route . . . . . . . . . . . . . . . . . . . . . . . . Ring-Fusion Aromatization of PBO Aerogels during Curing in Air . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Carbonization and Properties of Carbon Aerogels Derived from PBO Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

C. Sotiriou-Leventis (*) Department of Chemistry, Missouri University of Science and Technology, Rolla, MO, USA e-mail: [email protected] N. Leventis (*) Department of Chemistry, Missouri University of Science and Technology, Rolla, MO, USA Department of Research and Development, Aspen Aerogels, Inc., Northborough, MA, USA e-mail: [email protected] S. Mulik The Dow Chemical Company, Collegeville, PA, USA

490 490 490 490 491 494

20.7.2

The Effect of Air Oxidation on Other Phenolic Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 495 Ring-Fusion Aromatization in Other Mainstream Phenolic Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 495 Carbonization and Properties of Carbons Derived from RF, PF, TPOL, and FPOL . . . . . . . . . . . . . . . . . . . . . . . . . . 497

20.8

Further Studies and Applications of Phenolic Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 499

20.9

Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 502

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 503

Abstract

Phenolic aerogels comprise an important class of organic aerogels with potential for use in thermal superinsulation applications and which serve as precursors to electrically conducting carbon aerogels of importance to applications including filtration, energy generation (e.g., electrodes in fuel cells), energy storage (e.g., electrodes for batteries and supercapacitors), and other green energy technologies. Historically, the most important variety of phenolicresin-based aerogels has been based on resorcinolformaldehyde (RF) chemistry, but in recent years aerogels based on polybenzoxazine (PBO) chemistry have become increasingly important as well. In this chapter, we present a broad overview of these materials, focusing on how the chemical, microscopic, and macroscopic characteristics of RF aerogels, and thereby carbon aerogels, can be tailored to achieve desired application-specific structure-property relationships via variation of processing conditions such as monomer chemical identity, monomer concentration, pH, and catalyst-to-monomer ratio. Emphasis is placed on chemical transformations that occur during processing as well as on how chemical composition and structure drive materials properties. Discussion of PBOs focuses on a recently developed room-temperature acid-catalyzed synthetic route that enables deconvolution of polymerization

© Springer Nature Switzerland AG 2023 M. A. Aegerter et al. (eds.), Springer Handbook of Aerogels, Springer Handbooks, https://doi.org/10.1007/978-3-030-27322-4_20

479

480

C. Sotiriou-Leventis et al.

of the monomer from subsequent curing steps, which led to the discovery of ring-fusion aromatization as a reaction pathway available to benzoxazine monomers that proves to be a main property-determining factor during subsequent carbonization. Ring-fusion aromatization in PBOs was extended to and its effect studied in other mainstream phenolic aerogels derived not only from resorcinol but also from phenol and from phloroglucinol. Keywords

Phenolic aerogels · Carbonization · Carbon aerogels · Phenol · Resorcinol · Formaldehyde · Phloroglucinol · Terephthalaldehyde · Polybenzoxazine · Acid catalysis · Base catalysis · Aromatization

20.1

Introduction

Aerogels comprise a special class of low-density, open-cell foams with large internal void space (up to 99.9% v/v) that gives rise to useful material properties such as high internal surface area, low thermal conductivity (2–3 orders of magnitude less than glass), and high acoustic impedance [1, 2]. These open-cell foams are derived from wet gels typically prepared by sol–gel processes and subsequently dried by a technique that preserves porosity such as drying from a supercritical fluid (SCF) such as CO2 (▶ Chap. 4) [3, 4]. The most successful commercial application for aerogels to date has been as thermal insulation, however other potential applications of aerogels have been explored as well, including as catalysts and catalyst supports, gas filters and gas storage materials, electrically conducting and dielectric materials, lightweight structural materials, and acoustic insulation, among others [5]. Historically, most aerogels have been based on inorganic metal/metalloid oxide frameworks. Purely polymer-based aerogels were first prepared by Kistler [6, 7], however organic aerogels did not receive major attention until the development of synthetic polymer aerogels based on a framework consisting of a resorcinol-formaldehyde resin first described by Pekala and co-workers [8]. For quite some time, Pekala’s RF aerogels became synonymous with organic aerogels, but subsequently other types of materials in that class also became available, such as phenol-formaldehyde [9, 10], melamine-formaldehyde [11–13], cresol-formaldehyde [14], and phenol-furfural [15]. More recently, the class of organic aerogels has been even further expanded and today includes polyimides, polyamides, polyacrylamides, polyacrylonitriles, polyacrylates, polystyrenes, polyurethanes, polyureas, a wide variety of biopolymers, and others [16–22]. Resorcinol-formaldehyde (RF) aerogels were first synthesized by base-catalyzed poly-condensation of resorcinol

with formaldehyde in aqueous solutions using Na2CO3 (soda ash) as catalyst followed by SCF drying of the resulting wet gels from CO2. Resorcinol was considered as an ideal starting material because, even though it has three reactive sites (the 2, 4, and 6 positions of the aromatic ring) like phenol, it is about 10–15 times more reactive than phenol and therefore can react with formaldehyde at lower temperatures. Although the mechanism of the resorcinol-formaldehyde polycondensation reaction is very different from the reactions that lead to inorganic gels, the physicochemical processes underlying formation of RF wet gels are analogous to those that take place with typical inorganic gels such as silica and titania [4]. Following Pekala’s initial report on Na2CO3-catalyzed RF aerogel synthesis, several more recent literature reports discuss the synthesis of RF aerogels with both alkaline and acid catalysts [8, 23, 24]. RF wet gels have been produced by several different variations of essentially the same procedure. Initially, resorcinol (R) and formaldehyde (F) are mixed at the appropriate molar ratio in the presence of a catalyst (usually a base, or in very few cases an acid). The solution is then heated in a closed container to a predetermined temperature for a sufficient period of time to form a stable (crosslinked) wet gel. Wet gels are subsequently solvent-exchanged (washed) with a suitable organic solvent to remove the gelation solvent from the pores (in most cases water) and then finally dried with a technique that preserves pore structure upon drying such as SCF CO2 (▶ Chap. 4), freeze drying (▶ Chap. 5), or in some cases controlled ambientpressure evaporative drying to yield RF aerogels, or alternatively, dried evaporatively at ambient pressure with unhindered shrinkage to yield RF xerogels. Studies show that varying the processing conditions such as the amount of catalyst (through the resorcinol-to-catalyst ratio R/C) results in RF aerogels with different nanomorphologies. RF aerogels prepared with higher catalyst concentrations (e.g., R/C ¼ 50) lead to formation of a mesoporous skeletal framework which is very similar to that of typical basecatalyzed silica aerogels consisting of a pearl-necklace-like structure of secondary particles (40–70 nm), which in turn consist of primary particles (10–12 nm in diameter) (see also ▶ Chap. 13) [25–27]. The mechanical properties of RF aerogels, studied by several groups [28–33], have been found to depend on their microstructure, which is tunable via synthetic conditions. Overall, RF aerogels demonstrate typical aerogel-like properties such as high mesoporosity, low thermal conductivity, etc. In addition, RF aerogels serve as precursors to electrically conducting carbon aerogels (produced through pyrolytic carbonization, ▶ Chaps. 6 and ▶ 35), which perhaps is the main aspect of RF aerogels that has fueled the intense research interest these materials have seen (▶ Chap. 33).

20

Phenolic Aerogels and Their Carbonization

20.2

The Resorcinol-Formaldehyde (RF) Paradigm

Resorcinol-formaldehyde polymer is classified as a phenolic resin. In fact, phenol-formaldehyde condensates were the first synthetic polymers introduced commercially (Bakelite) at the beginning of the Twentieth Century. Thus, phenolic resin chemistry, about which there is a vast amount of information, can be used to understand the chemistry of RF aerogel synthesis. In the first step, resorcinol reacts with formaldehyde to form hydroxymethylated resorcinol (formaldehyde addition); in the second step the hydroxymethyl groups condense with each other to form nanometer-size clusters, which then crosslink by the same chemistry to produce a gel. The formation of clusters is influenced by typical sol–gel parameters such as the temperature, pH, and concentration of reactants. Although the first phenolic-resin aerogels were synthesized via aqueous polycondensation of resorcinol with formaldehyde with sodium carbonate as catalyst [8, 34, 35], now several literature reports exist for aerogels made from phenol and formaldehyde as well [9, 10, 36]. Phenolic resins made from phenol and formaldehyde are classified as either a “resole” or “novolac” based on the presence of dimethylene ether (-CH2OCH2-) or methylene (-CH2-) linkages between the aromatic moieties, respectively. The presence either of ether or methylene linkages in phenolformaldehyde resins is based on the ratio of the reactants, the pH, and the catalyst. In this context, resorcinol undergoes most of the typical elementary reactions of phenol, reacting with formaldehyde under alkaline and/or acid conditions to produce a mixture of addition and condensation products. However, as it turns out, the reaction of resorcinol and formaldehyde leads predominantly to formation of methylene bridges between resorcinol moieties. The reason lies with the 1,3 relative placement of the OH groups in resorcinol and is discussed in detail below. In terms of electrophilic aromatic substitution, resorcinol is a trifunctional monomer, which means it can react with up to three equivalents of formaldehyde. These substituted resorcinol derivatives condense into nanoclusters, which then crosslink through reaction among their surface functional groups. Generally, in the case of base catalysis, this reaction is carried out in aqueous media in the presence of sodium carbonate, or a similar base catalyst, with gelation requiring heating at elevated temperatures for prolonged periods. With acid catalysts, the monomers are mixed either in non-aqueous or in aqueous media with the catalyst (e.g., hydrochloric acid, acetic acid, perchloric acid), and gelation can take place even at room temperature and is generally faster.

481

20.2.1 The Chemistry of Base-Catalyzed Gelation [37] Under basic conditions, resorcinol is deprotonated to the resorcinol anion. As shown in Fig. 20.1, due to resonance, the electron density at the 4 (or 6) positions of the resorcinol anion is increased (Equation 1). Electron donation from these positions to the partially positively charged carbonyl carbon of formaldehyde leads to hydroxymethylation (i.e., addition of –CH2OH groups). In turn, hydroxymethylation activates other positions, and an excess of formaldehyde leads to di-hydroxymethylation (Equation 2) [38]. The base catalyst causes deprotonation of hydroxymethylated resorcinol, leading to a very reactive and unstable o-quinone methide intermediate (Equation 3). The o-quinone methide reacts further with another resorcinol molecule to form a stable methylene linkage (Equation 4). The formation of o-quinone methides and the presence of high electron densities at the 2-, 4-, and 6-positions of the resorcinol ring are the reasons for the enhanced reactivity of resorcinol compared to phenol. Alternatively to Equation 4, o-quinone methide intermediates may be attacked by deprotonated hydroxy methyl groups (– CH2O, Equation 4a), leading to ether linkages (–CH2–O– CH2–) between phenyl rings; however, this reaction entails a reaction between a hard nucleophile (–O) and a soft electrophile (the carbon atom of the ¼ CH2 group of the o-quinone methide) and is slower than the nucleophilc attack of the soft aromatic carbon on the soft electrophilic center of the o-quinone methide. Indeed, only a very small amount of ether linkages can be observed in RF aerogels, and –CH2– bridges are favored (see the solid-state 13C NMR spectra below). After the initial condensation of two R moieties (via Equation 4), there is a greater tendency for continuous condensation as long as there are active sites on resorcinol molecules or RF clusters. For both reasons just outlined, RF resins produced via base catalysis predominantly contain methylene-bridged novolac structures [39]. The overall chemical process leading to gelation is summarized by Equation 6.

20.2.2 The Chemistry of Acid-Catalyzed Gelation [25, 40] In contrast to the base-catalyzed route, which is based on activation of the aromatic ring toward electrophilic aromatic substitution by increasing the electron donating ability of its substituent groups (from –OH to O), the acid-catalyzed route is based on acceleration of the reaction of resorcinol with formaldehyde by increasing the electrophilicity of formaldehyde. As shown in Fig. 20.2, protonation of formaldehyde (Equation 7) is followed by nucleophilic attack by the

20

482

C. Sotiriou-Leventis et al.

Fig. 20.1 Mechanism of the base-catalyzed gelation of resorcinol and formaldehyde for RF aerogel synthesis

O–

OH

O

Base

–

HO

HO O

HO

HO O–

O

H –HC O

(1)

H

HO

O–

OH

(2)

HO

O–

O OH

CH2

– OH –

HO

(3)

HO o-Quinone methide

O

O–

O

CH2 – +

OH (4)

HO

HO

OH

O–

OH

O

OH OH

CH2 – + O

O OH

HO O–

OH

HO

HO

O

OH H

OH

–

H

OH HO

OH

Base, CH2O HO

C O as in (2)

(5)

OH

OH

Overall:

(4a)

n

HO

(6)

RF

π-system of resorcinol leading to hydroxymethylation of resorcinol (Equation 8). In an acidic environment, a subsequent protonation of hydroxymethyl groups leads to formation of OHþ 2 (Equation 9), which is a good leaving group. These groups cleave unimolecularly to form o-quinonemethide-type intermediates (Equation 10a). o-Quinone methides can lead to condensation via methylene bridging (Equation 11 – the same as Equation 4), or alternatively protonated hydroxymethylated groups (before they cleave to yield o-quinone-methides) can be attacked by the π-system of another resorcinol, again forming methylene linkages (Equation 10b). These addition and condensation

reactions in tandem result in the crosslinked structure of the resulting resorcinol-formaldehyde aerogels (Equation 12). Overall, it is noted that due to the presence of a second phenolic –OH group in resorcinol, both the base- and the acidcatalyzed condensation pathways go through a common oquinone methide intermediate (Equation 3 or Equation 10a, respectively), which leads to connection of the phenolic moieties mainly through methylene bridges (–CH2–) in both routes. In addition, since there is no good O-based nucleophile present in the acid-catalyzed route, the amount of ether bridging (–CH2–O–CH2–) occurs to an even lesser extent than in the base-catalyzed route, as discussed in Sect. 20.5 below.

20

Phenolic Aerogels and Their Carbonization

Fig. 20.2 Mechanism of the acid-catalyzed gelation of resorcinol and formaldehyde for RF aerogel synthesis

483

CH2O

H 2O –H2O

CH2(OH)2

H+, –H2O H2O, –

H+

OH

+ CH2OH

(7)

OH + + CH2OH

–H+

HO

OH

(8)

+ OH2

(9)

HO

R OH

OH OH

H+

HO

HO OH

+ OH2

O

HO

(10a)

HO o-Quinone methide OH OH

OH + OH2 HO O

CH2

–H3O+

R –H3O+

(10b) OH

HO OH

CH2

OH

R

(11) HO

HO OH

OH

OH

OH CH2O,

HO

OH

H +,

–H3O+

R

n

(12)

HO RF

20.3

RF Aerogels Via Base Catalysis

Aerogels prepared by Pekala’s base-catalyzed route have been studied extensively for effects of process parameters on materials properties. Typically, based on the desirable final physical properties of the RF aerogels, two ratios need to be selected carefully: the concentration of the catalyst, expressed as the resorcinol-to-catalyst ratio (R/C), and the concentration of the monomer, expressed by the resorcinolto-water ratio (R/W). The pH of the sol also plays an important role. Both the microstructure as well as properties such as density, surface area, particle size and pore size distribution are influenced by these parameters. Low-density RF aerogels (~0.03 g cm3) prepared by this method exhibit high porosities (>80%), high surface areas (400–900 m2g1), and ultrafine pore size (30%) than the carbon yield of PBO-H-130 that had been cured at 200  C in air prior to pyrolysis (referred to as PBO-H-200). This

Ar

200 nm

200 nm

25 μm

behavior was reminiscent of an analogous prior literature observation wherein the carbonization yield of PBO aerogels was higher by a similar percentage than the carbonization yield of bulk PBO polymer [76]. Furthermore, as shown in Fig. 20.16, carbon aerogels obtained from direct pyrolysis of PBO-A-RT were deformed and lost the microstructure of

20

Phenolic Aerogels and Their Carbonization

493

their parent PBO-A-RT aerogels. However, by first curing PBO-A-RT at 200  C in air, although initially thought unnecessary, carbon aerogels that preserved both the shape of the monoliths and their nanostructure were obtained. Therefore, it was concluded that curing in air of as-prepared PBO-H-130 aerogels did something more than just complete polymerization, as previously assumed in the literature. Furthermore, as far as carbonization yield is concerned, when PBO-A-RT aerogel samples were cured at 200  C in air (such samples are referred to as PBO-A-200), the carbonization yield became equal to that of PBO-H-200 (about 61% w/w). Based on spectroscopic evidence and CHN analysis, curing at 200  C in air of either PBO-H-130 or PBO-A-RT brings about fusion of the aniline and phenolic moieties of every polymer repeat unit along the polymeric backbone [85]. Indeed, solid-state 15N NMR with 15N-enriched PBO aerogels (Fig. 20.17) demonstrates convincingly that the Hand A- processes yield chemically different products (the spectra of PBO-H-130 and PBO-A-RT were completely different), but the 200  C air-curing step is the equalizer: the spectra of PBO-H-200 and PBO-A-200 were identical, corresponding to a fully aromatized polymer (Aox-II-T, see Fig. 20.18) originating from either process. In fact, 15N NMR showed that as-prepared PBO-A-RT was a simple, welldefined product whose treatment at 200  C in air converted it to a second, single-type, distinguishable product (PBO-A200). On the other hand, as-prepared PBO-H-130 showed

Fig. 20.17 Solid-state 15N NMR of 15N-enriched BO monomer and PBO aerogels prepared with 5% (w/w) BO monomer at room temperature (RT), 130  C, and 200  C via the heat- (H-) and acidcatalyzed (A-) processes

PBO-H-130 293

135

several resonances, suggesting a mixture of products; however, treatment at 200  C in air yielded the same product as the one that was obtained from PBO-A-RT. To confirm that as-prepared PBO-H-130 was indeed a mixture of PBO and partially oxidized products, as-prepared PBO-A-RT was subjected to incomplete oxidation, and indeed a spectrum similar to that of as-prepared PBO-H-130 was obtained (compare with the spectra of PBO-A-200 partially cured) and of PBO-H-130 in Fig. 20.17). It is noted that similar observations have been made and similar conclusions have been arrived at via solid-state 13C NMR, however, the 13C NMR spectra were much more complicated [85] and are not shown here. Mechanistically (Fig. 20.18), 200  C in air treatment of either PBO-A-RT or PBO-H-130 aerogels causes oxidation of the –CH2– bridges along the polymeric backbone and induces electrocyclic ring closure (Equation 13, Fig. 20.18). The product of Equation 13 is further oxidized by O2-producing superoxide, O 2 , as a side product (Equation 14), which, following typical superoxide chemistry, reacts as a base, abstracts a proton, and generates hydroperoxyl radicals HO_2 , which form H2O2 via another H atom abstraction (Equation 15). In turn, the product of Equation 15 yielded the fully aromatized polymer, Aox-II-T, (Equation 16), whereas the phenolic and aniline aromatic rings of the as-prepared PBO backbone fused together [85].

129

PBO-H-200 128

75.4

127

64.2

PBO-A-200 PBO-A-RT 62.8

PBO-A-200 (partially cured) 291

15N

enriched BO monomer 400

300

126

200

100

Chemical shift of 15N (ppm)

0

400

300

200

100

Chemical shift of 15N (ppm)

0

20

494

C. Sotiriou-Leventis et al.

OH

OH O N

OH O– O + H N

O N

200 °C Air n

n

–

–

As-prepared PBO network

O n

H

O

_

+ H N

O

O

–

O

+ H N

O

O O2

. n (– O2 –)

H

–

O

H

–

+ H N

OH O

O

O2.–

O n

H

–

O

O

n – H2O2

_

(14)

O N

H N n

H

(13)

Oxidized PBO backbone

OH O– O + H N –

n

n

_

(15)

+ HO2.

Superoxide as a base

Hydroperoxyl radical

OH O

OH O– O + N

O N n

–

–O

–

Aox-II Aromatization

O

O + c Nd

n n

–

H

j

n

(16)

Aox-II-T

H-transfer tautomerization

Fig. 20.18 Proposed mechanism for the oxidative ring-fusion aromatization of the PBO network (for clarity, only half of the bisphenol A moiety is shown) [85]

20.6.5 Carbonization and Properties of Carbon Aerogels Derived from PBO Aerogels In general, carbon aerogels are made by pyrolysis (also referred to as carbonization) of certain organic aerogels under inert atmosphere. PBO aerogels have been carbonized at 800  C under argon. Importantly, the carbonization efficiency (up to 61% w/w), nanomorphology, and pore structure of the resulting carbon aerogels critically depend on the

curing step of PBO aerogels as-prepared at 200  C in air (Fig. 20.16), which, as discussed in the previous section, does more than just complete polymerization. Carbon aerogels from air-cured (i.e., aromatized) PBO aerogels are microscopically similar to their respective parent PBO aerogels; however, exhibit greatly enhanced surface areas which, for carbon aerogels derived from HCl-catalyzed PBOs, can be as high as 520 m2 g1 with up to 83% of that surface area attributed to newly created micropores [85].

20

Phenolic Aerogels and Their Carbonization

20.7

495

The Effect of Air Oxidation on Other Phenolic Aerogels

Using Kanatzidis’s work on phloroglucinol-terephthalaldehyde porous polymers as a point of departure [86], Majedi-Far et al. studied the effect of air curing (oxidation) on other phenolic aerogels derived from phenol-formaldehyde (PF), phloroglucinol-formaldehyde (FPOL), and phloroglucinolterephthalaldehyde (TPOL), using resorcinol-formaldehyde (RF) aerogels as a reference system [87]. The structures of these monomers are shown in Fig. 20.19. Collectively, this group of phenolic aerogels is referred to as RES, and were all prepared via the fast HCl-catalyzed process described above [87]. As with PBOs, studying the effect of air oxidation on phenolic resins via RES aerogels was advantageous because their open porosity allowed unobstructed air circulation through their bulk. The chemical composition of the four model systems over the course of pyrolysis was followed with solid-state 13C NMR, FTIR, and elemental analysis. The evolution of the skeletal framework and the pore structure was followed with SEM and N2-sorption porosimetry. It was found that at 240  C in air, all four RES were oxidized (Fig. 20.20). In that regard, X-ray photoelectron spectroscopy (XPS) proved to be a powerful tool in the elucidation of the fate of phenolic oxygens into fused heteroaromatic pyrylium rings (the cases of TPOL and FPOL).

The postulated mechanism during oxidation and ringfusion aromatization is illustrated with TPOL in Fig. 20.21 [87]. For clarity, the full mechanism is separated into three stages. In Stage 1 (Fig. 20.21a), bridging –CH– groups along the phenolic backbone are converted into carbonyls via H-atom abstraction by O2 followed by addition of hydroperoxyl (–OOH) groups at the benzylic positions and subsequent homolytic cleavage of the O–OH bond and of the adjacent C–C bond. In addition to creating carbonyls, H2O2, and phenyl radicals, this process breaks down the polymer chain at several places and disrupts crosslinking. At this point, polymer chains are free to relax into new positions, where phenyl radicals can couple and establish a new crosslinked configuration that remains similar in terms of connectivity to, but is more compact than, the original TPOL. In Stage 2 (Fig. 20.21b), newly created carbonyl groups undergo 1,5-proton transfer tautomerization with the ortho OH groups of the phologlucinol moieties, followed by electrocyclic ring closure that restores aromaticity. At that point, air (O2) oxidizes the newly created cyclic-ether bridges (–O– + O2 ! –O+– + O: 2 ), and the byproduct of that oxidation (superoxide: O: 2 ) acts as a Brønsted base and abstracts phenolic protons yielding hydroperoxyl radicals HO 2 . The hydroperoxyl radicals

RES (as-prepared)

20.7.1 Ring-Fusion Aromatization in Other Mainstream Phenolic Aerogels Besides PBOs, ring-fusion aromatization was observed only in FPOL and TPOL aerogels. In these two systems, initial oxidation was followed by a sequence of steps leading to ring fusion and formation of pyrylium O+-heteroaromatic rings in every repeat unit of their polymeric backbone. Under the same conditions (240  C in air), RF and PF aerogels did not undergo aromatization; instead, they just went through an autooxidation-like process that converted the –CH2– bridges between phenolic moieties into carbonyls (C¼O) (Fig. 20.20).

OH

OH

OH

CHO O OH H

OH HO P

R

POL

Fig. 20.19 Structures of monomers used for phenolic aerogels, and their abbreviations

OH O

n

O+ m OH

HO

_

O

n OH

HO O

OH n OH

FPOL HO

_

+O OH

TPOL

_

OH

O+

n O

PF (X=H)

OH OH O

OH

RF (X=OH)

n x

CHO T

HO e´

OH

OH

H F

e

HO

RES-240 (air)

n x O

Fig. 20.20 Structures of as-prepared RES and their air-oxidized products at 240  C [87]

20

496

C. Sotiriou-Leventis et al.

a HO

OH

HO OH

HO OH

HO OH

OH

OH

HO OH

HO

HO OH

O

OH O

O

OH

OH

O

OH O

OH OH O

HO OH

HO

O OH

OH OH

HO OH

OH

HO OH

O

Chain Relaxation

O

OH

HO O OH

OH OH

OH

O OH

HO OH

O HO

OH

HO OH

HO

OH

– H2O2

O

OH

OH

HO

OH

HO OH

HO

OH

O OH OH

HO O OH

HO O

OH

OH

O

O OH

OH

240 °C

TPOL

HO OH

HO OH

O2 (air)

OH

OH

OH HO

HO OH

HO OH

OH

HO HO OH O

HO OH

O OH OH

HO

Reorganization

OH

O

Radical HO OH

O

OH

O OH OH

HO

Coupling

OH

O

O

OH OH

HO

OH

HO OH

O OH OH

HO

b HO

HO

OH O

OH OH

π

HO

OH

1,5 H-transfer O Tautomerization

OH π

O2.–

HO

Fig. 20.21 (continued)

Electrocyclic O Ring closure

O

_

OH

HO2.– – H2 O 2

O +

π

HO –O

π

O

HO

OH

OH OH

OH

HO

HO

HO

OH

O2

– O –O2

OH O

HO

O +

π

OH

_

HO

HO

O

OH

+O

π

HO –O

OH – OH

OH

Radical + Coupling O

+ O

HO

HO HO

OH

H – HO2.– O+

HO

π

OH

HO

OH

HO

OH

O

O

+O H

OH

O

OH

HO

HO

OH

O +

+ OH OH

HO –O

O+ OH

20

Phenolic Aerogels and Their Carbonization

497

c O+

O+ –O

e

OH e OH O–

⊕ HO HO H

– H2 H OH OH –

HO HO

e

O–

e

20

OH O–

–O

O+

HO HO

O+

OH OH O–

O+

–O

HO

+O

O+ –O

n

HO

OH

O+

TPOL-240(air) Fig. 20.21 Oxidation mechanism of RES demonstrated using TPOL. The polymer repeat units are shown in bold; adjacent units are included in order to show connectivity. (a) Stage 1: initial oxidation leads to a change in connectivity along the polymer chain. (b) Stage 2: further

oxidation leads to ring-fusion aromatization (the Dewar-benzene-like canonical form in the first step was used for brevity). (c) Stage 3: interchain coupling at the e’-positions [87]

abstract H atoms from the positions adjacent to –O+– yielding pyrylium rings. FPOL goes through Stages 1 and 2, yielding ring-fusion aromatization to pyrylium analogous to TPOL. Finally, since TPOL, unlike FPOL, possesses hydrogen atoms at the “e” positions (refer to Fig. 20.20), the reaction proceeds to Stage 3, in which the tip of one fused aromatic system undergoes an electrophilic aromatic substitution with the tip of another one (Fig. 20.21c). In contrast, RF and PF go through Stage 1, but oxidation stops there (Fig. 20.20). It is remarkable that, despite the immense complexity of the process, elemental analysis data of TPOL oxidized at 120  C in air (C: 64.7%, H: 1.7%, O: 33.6%) matched extremely well with the theoretical values based on the structure of Fig. 20.20 (C: 64.2%, H: 1.6%, O: 34.2%), which was considered as strong supporting evidence for the proposed structure. Direct evidence for pyrylium ring formation was based on XPS. Figure 20.22 shows the O 1 s energy region of the XPS spectrum of TPOL in comparison with RF, before and after air oxidation [87]. As expected, all as-prepared RES showed only a single type of oxygen (with maxima in the 531.52–532.66 eV range), assigned to phenolic –OHs [88–91]. The deconvolved XPS spectrum of TPOL after oxidation at 240  C in air [TPOL-240(air)] revealed three peaks: one at higher binding energies (533.62 eV, assigned to pyrylium O+ [92]); one somewhat more intense peak at intermediate binding energies (532.64 eV, assigned to phenolic –OH [88–91]); and, one peak at lower binding energies (531.39 eV, assigned to –O [93, 94]). Albeit the pyrylium and phenoxide peaks were of about equal intensity, as expected from charge balance, the

phenoxide peak was slightly more intense, perhaps because of overlapping with oxygens from C¼O moieties. Importantly, the O 1 s energy region of RF-240(air) consisted of only two chemistries, one at 531.53 and one at 529.75 eV, assigned to phenolic –OH and to C¼O, respectively.

20.7.2 Carbonization and Properties of Carbons Derived from RF, PF, TPOL, and FPOL Although only FPOL and TPOL aerogels yielded ring-fusion aromatization at 240  C in air, upon further pyrolysis under Ar, the chemical compositions of all four RES aerogels (RF, PF, FPOL, TPOL), either as-prepared, or after air oxidation at 240  C, converged to a common chemical composition at temperatures of 600  C (see Fig. 20.23). By XPS (Fig. 20.24), the carbon aerogels from all four systems bore only two types of O: pyrylium O+ and charge compensating phenoxide O in 1:1 atomic ratio [87]. Figure 20.25 shows a plausible structure for these carbons. It is noteworthy that the presence of pyrylium in extensively-studied carbons derived from RF aerogels has been overlooked in the literature. The only significant difference achieved by the air oxidation step at 240  C was in the surface areas of carbons derived from the aromatizable systems (TPOL and FPOL), which were found to be higher than those derived from direct pyrolysis of as-prepared aerogels. It was concluded that early ring-fusion aromatization rigidizes the polymeric structure and creates microporosity that contributes toward production of porous carbons with higher surface areas than those obtained by direct pyrolysis of as-prepared materials.

498

C. Sotiriou-Leventis et al.

Fig. 20.22 The O 1 s energy region of XPS spectra and peak assignments for as-prepared TPOL and RF aerogels and their 240  C air-oxidized versions. (Reprinted with permission from [87], copyright 2017, Royal Society of Chemistry)

TPOL-240(air)

TPOL Intensity (a.u)

O

OH

–

OH

⊕ O

540

537

534

531

528

RF

540

537

534

531

528

RF-240(air)

Intensity (a.u)

OH

Ph

540

a

TPOL

O

OH

534 531 537 Binding energy (eV)

-D-600

-O-600

-D-500

b

528

RF

540

Ph

537 534 531 Binding energy (eV)

528

-D-600

-O-600

-O-500

-D-500

-O-500

-D-400

-O-400

-D-400

-O-400

-D-300

-O-300

-D-300

-O-300

–240(air) As-prepared 200 150 100 50 δ (ppm)

0 200 150 100 50 δ (ppm)

–240(air) As-prepared

0

Fig. 20.23 Solid-state CP-MAS 13C NMR spectra for TPOL aerogels (a) and RF aerogels (b) along pyrolysis at different temperatures as shown. The -D extension designates samples from direct pyrolysis of as-prepared TPOL and RF aerogels, and the -O extension designates samples from pyrolysis of samples oxidized at 240  C in air. Similar

200 150 100 50 δ (ppm)

0 200 150 100 50 δ (ppm)

0

spectra have been published for FPOL and PF. It is noted that irrespective of the starting material (TPOL versus RF) or synthesis process (D or O), beyond 600  C all samples converge on a common chemical composition. (Adapted with permission from [87], copyright 2017, Royal Society of Chemistry)

20

Phenolic Aerogels and Their Carbonization

Fig. 20.24 Representative O 1 s XPS spectra for carbon aerogels derived from direct (D-) pyrolysis of TPOL and RF aerogels at 800  C under flowing Ar. Note the equal amounts of pyrylium and phenoxide. (Adapted with permission from [87], copyright 2017, Royal Society of Chemistry)

499

C-TPOL-D-800 Intensity (a.u.)

O

–

Aromatic repeat layer

Fused aromatic core

Aromatic repeat layer O+ O–

O–

Fig. 20.25 Idealized structure of carbon aerogels derived from pyrolysis of RES samples at 800  C, showing two repeat layers of the graphitic core, one in red, one in blue [87]

Upon further reactive etching with CO2 at 1,000  C (i.e., activation) [87], these surface areas reached values as high as 2,778  209 m2 g1. Figure 20.26 shows typical N2 sorption isotherms of TPOL and the variation in surface areas at different pyrolysis temperatures as a function of direct pyrolysis versus pyrolysis after oxidation at 240  C in air. These findings are directly relevant to high-surface-area carbons for gas sorption (e.g., capture and sequestration of CO2 [95]) and for ion-exchange applications that leverage the fixed O+ and O sites on their carbon backbones, as has been proposed for biochar materials bearing pyrylium moieties [92].

20.8

O

537 531 534 Binding energy (eV)

O+

Further Studies and Applications of Phenolic Aerogels

Present and emerging applications of phenolic aerogels generally focus on their thermal, mechanical, and carbonization properties. RF aerogels are better insulators than commercial

–

⊕

O

O– O+

O

⊕

540

20

C-RF-D-800

528

540

537 534 531 Binding energy (eV)

528

fiberglass, with approximately six times higher thermal resistance. When compared with silica aerogels, RF aerogels have been shown to exhibit thermal conductivities as low as 0.012 W m1 K1 at ambient conditions, whereas comparable minimum values reported for silica aerogels are typically about 0.014–0.016 W m1 K1 [96–98]. RF aerogels are also stiffer and stronger materials than silica aerogels – properties that play an important role in the design of better heat-transfer inhibitors [99, 100]. Efficient separation of gases and liquids with nanoporous membranes is another emerging technology where aerogels can serve as a convenient design platform [101, 102]. Organic aerogels like RF not only provide the freedom to tune pore properties to enable such applications, but also to tune the physical shape and form factor of the material (monolith, particle, film) to fit the engineering requirements of a separation process. As mentioned, to date RF aerogels have mainly been pursued as precursors for production of nanoparticulate (amorphous) carbon aerogels and related materials. Since carbon aerogels combine electrical conductivity with typical aerogel properties such as low density, open porosity, pore sizes less than 50 nm, and high surface area, they are highly desirable for a wide variety of applications including separation media (e.g., high pressure liquid chromatography, HPLC), non-reflective panels, materials for hydrogen [103] and methane storage [104], humidity sensors [105], fuel cell applications, anodes for lithium-ion batteries, and electrodes for supercapacitors [106–111] (▶ Chap. 50). This wide variety of potential commercial applications is largely due to the tunable properties of RF-derived carbon aerogels. These properties, which include surface area, high pore volume, and tailorable pore-size distribution, are directly related to the synthesis and processing conditions of their parent RF aerogels, which enables facile production of a wide spectrum of nanostructured materials with unique properties.

500

C. Sotiriou-Leventis et al.

Etched C-TPOL-O-800

1000 800 600 400

C-TPOL-O-800

200 0 0.0

0.2 0.4 0.6 0.8 Relative pressure (P/Po)

1.0

Fig. 20.26 (a) N2 sorption isotherms of TPOL aerogels after oxidation at 240  C in air followed by carbonization at 800  C (C-TPOL-O-800, red curve) as well as after etching at 1,000  C under flowing CO2 (blue curve). (b) BET surface areas and micropore surface areas of TPOL aerogels pyrolyzed at the temperatures indicated. Dotted lines connect pyrolyzed and their corresponding etched samples. Note that the surface

Efforts to achieve and tune desirable physical properties have led to the development of new organic aerogels prepared with similar raw materials and analogous characteristics to resorcinol and formaldehyde. The major requirements for such organic aerogels include the need for multifunctional aromatic monomers and the ability to achieve a high crosslinking density. In that context, several literature reports have explored phenol-furfural organic aerogels as a cheaper alternative to RF aerogels [10, 15]. The dark red color of RF aerogels limits their use in applications where materials need to be transparent and/or colorless. This specific limitation led to the development of melamine-formaldehyde aerogels (Fig. 20.27). Similar to RF, melamine-formaldehyde (MF) aerogels are lightweight and highly porous, however with the added advantage of optical transparency. In this system, hexafunctional melamine reacts with formaldehyde under alkaline conditions to form hydroxymethyl (-CH2OH) groups (Equation 17). The role of the base in this reaction is to remove and shuttle the proton from the positive N to the negative O in –NH2+–CH2O. The kinetics of the reaction of formaldehyde with melamine indicates that the presence of one hydroxymethyl group on N deactivates it toward a second reaction by a factor of 0.6 [112]. The reaction of 6 mol equivalents of formaldehyde with 1 mol of melamine thus yields a mixture of products including all levels of hydroxymethylation. Following hydroxymethylation, the base catalyst is neutralized, and condensation through etherification of the hydroxymethyl groups is facilitated via acid catalysis (Equation 18). In some cases,

3000

O samples D samples As-made

2000 1000

1000

500 Etched

500

Micropore area (m2 g–1)

b BET surf. area (σ, m2 g –1)

Volume adsorbed (cc/g STP)

a

0

100

0

200 400 600 800 1000 Pyrolysis temperature (°C)

areas of samples that were oxidized at 240  C in air prior to pyrolysis (blue symbols) were higher than those obtained by direct pyrolysis of as-prepared samples (red symbols). The same trends have been reported for FPOL, but not for RF and PF aerogels. (Adapted with permission from [87], copyright 2017, Royal Society of Chemistry)

NH2 N H2N

3 CH2O, OH–

N N

NH2

in H2O NHCH2OH N

Melamine HOH2CHN

N

(17)

NHCH2OH

N

Hydroxymethylated melamine(HMM ) HMM H +, –H 2 O N N

N HN

NHCH2OCH2NH

N N HN CH2 O H2C NH

N

N N N N NHCH2OCH2NH

N NCH2OH CH2 O (18) H2C NH N N N NH

Crosslinked polymer Fig. 20.27 Gelation mechanism of the melamine-formaldehyde (MF) system

20

Phenolic Aerogels and Their Carbonization

depending upon the required properties, methyl or butyl alcohol is used along with the acid catalyst. Note that there is substantial disagreement in the literature about the mechanism of this etherification. To this end, both specific and general acid catalysis have been suggested as possible mechanisms [112, 113]. Overall, the principal crosslinking reactions include the formation of diaminomethylene (-NHCH2NH-) and diaminomethylene ether (-NHCH2OCH2NH-) bridges [11, 12]. All melamine-formaldehyde solutions develop a bluish haze (similar to silica aerogels) as they are cured (2 days at 50  C followed by 5 days at 95  C) [13]. This phenomenon is associated with Rayleigh scattering [114, 115] from MF clusters formed in solution, noting that scattering intensity scales inversely with wavelength to the fourth power thus highly favors scattering of shorter wavelengths such as blue and violet. These clusters possess surface functional groups (e.g., -CH2OH) that eventually crosslink with one another to form a gel. The aggregation and crosslinking processes underlying gelation of MF gels show strong pH dependence. Finally, melamine-formaldehyde wet gels are converted into aerogels by drying with SCF CO2 or another suitable drying process. In addition to phenol, resorcinol, and phloroglucinol, other phenol derivatives that have been explored include cresol [14] and 5-methylresorcinol [116, 117]. In an effort to further rigidize the structure of RF, Mulik et al. reported crosslinking of RF aerogels with isocyanates (▶ Chap. 28). In this work, the hydroxyl groups of resorcinol as well as surface hydroxymethyl groups of the RF polymer react with isocyanate to form urethane and interparticle polyurea linkages. Polyurea linkages formed between RF nanoparticles hold particles from collapsing during supercritical drying, resulting in lower shrinkage and thus lower density. Where native (i.e., non-crosslinked RF) aerogels shrank significantly (up to 39% in linear dimensions), isocyanate-crosslinked RF aerogels shrank by only 13–22% relative to their mold (linear) dimensions [71]. Along with varying starting materials as a means of controlling the properties and structure of phenolic aerogels, some researchers have reported methods of introducing porosity by incorporating templating agents based on silica or polystyrene beads during gelation; such templating agents are then later removed from the resulting wet gels by dissolving them with HF or toluene, respectively [109, 110, 118, 119]. In order to expand the potential applications of phenolic aerogels and their derivative carbon aerogels, efforts have focused on microstructural modifications via doping or nanocompositing. Areas attracting significant interest include the incorporation of graphene oxide [120], carbon nanotubes [121–123], metals, metal oxides, and metal carbides into the RF framework, with the aim of modifying the structure,

501

electrical conductivity, and/or catalytic activity of the resulting carbon materials [124–128]. Job et al. incorporated metal salts of Pt, Fe, or Ni into RF precursor solutions and observed that the pore texture of the resulting aerogels could be controlled with different salts. This finding can be attributed to changes in the pH of the precursor solution due to the presence of the Lewis-acidic metals salts prior to gelation [129, 130]. It has been noted that there are often interactions present between dopants and the resulting carbon. For example, metals can catalyze partial graphitization of the surrounding carbon matrix, and the carbon matrix in turn can act as a reducing agent for the reduction of metal oxide precursors to metals and carbides during heat treatment. Leventis et al. demonstrated smelting of iron oxide to form metallic iron nanoparticles (▶ Chap. 44) [131]. The inertness of carbon nanoparticles or carbon aerogels derived from RF aerogels also make them potential candidates for adsorbent materials, electrodes for capacitive deionization of aqueous solutions, electrochemical double-layer capacitors and supercapacitors, gas diffusion electrodes in proton exchange membrane (PEM) fuel cells, anodes in rechargeable lithium ion batteries, and ozonolysis [23, 108, 131, 132]. Incorporation of metal precursors into carbon aerogels can be achieved via incorporation of metal salts in phenolic precursor solutions before the sol–gel process. Simple physical adsorption of metal salts on wet–gel matrices caused leaching of the metal ions during the wash steps required by the sol–gel process. This inspired researchers to use derivatives of resorcinol (e.g., potassium 2,4-dihydroxybenzoate), which can complex with the metal ions (e.g., Fe3+, Co3+, Ni2+) and retain them throughout processing [127, 133–135]. The resulting metal-doped phenolic aerogels can in turn be pyrolyzed to produce metalnanoparticle-doped carbon aerogels. Both metal-doped phenolic and carbon aerogels show promise for a number of interesting potential applications. For example, Steiner III et al. showed that metal-doped and metal-oxide-doped carbon aerogels can catalyze CVD growth of carbon nanotubes (CNTs) and can serve as a versatile screening platform for the discovery of new CNT growth catalysts [135– 138]. Attia et al. have prepared a series of composite RF aerogels doped with CoFe2O4 and studied their DC/AC electrical conductivity and dielectric properties [139], while Sanchez-Polo et al. have used metal-doped carbon aerogels for water treatment [140]. Thus, there is great interest in novel functional dopants and understanding the interaction between dopants and carbon in such aerogel systems. Along these lines, recently Leventis et al. prepared monolithic nanoporous iron aerogels via carbothermal reduction of interpenetrating networks of polybenzoxazine/iron oxide nanoparticles (PBO-FeOx) (Fig. 20.28) [141]. The synthesis of interpenetrating PBO-FeOx was based on the fact that

20

502

C. Sotiriou-Leventis et al.

a Formation of the FeOx network O Cl [Fe(H2O)6]3+

(Epichlorohydrin)

[Fe(H2O)5 (OH)]2+

–H+

b Formation of the PBO network

O

O

N

N

[Fe(H2O)5 (OH)]2+

FeOx gel

–H2O

OH O– + N

OH N ]3+

n

[Fe(H2O)6

n

O2 200 °C

80 °C N

Benzoxazine (BO) monomer

OH PBO

N

m

O–

OH

m

Curred (aromatized) PBO

Fig. 20.28 Chemical processes involved in the co-gelation of FeOx (a) and PBO (b) interpenetrating networks

hydrated metal salts, like [Fe(H2O)6]Cl3, are fairly strong Brønsted acids, thereby able to catalyze co-gelation of the BO monomer during their gelation. The final metallic iron aerogel products were filled with LiClO4 and behaved as explosives or thermites depending on the carbothermal reduction conditions (▶ Chap. 44). An emerging application of benzoxazine aerogels is in space exploration. A lightweight superinsulating benzoxazine organic-inorganic hybrid aerogel blanket was developed by Berthon-Fabry et al. in a one-pot sol–gel synthesis by impregnating a polyethylene terephthalate (PET) mat with a mixture of resorcinol, formaldehyde, and 3-aminopropyltriethoxysilane (APTES) in ethanol [142]. The hybrid nature of the organicinorganic benzoxazine compound, as opposed to a nanocomposite, was confirmed via FTIR (912 cm1 and 1374 cm1, oxazine peaks). These aerogel blankets showed thermal conductivities as low as 0.019 W m1 K1. Replacing part of the APTES with methyltriethoxysilane (MTES) or methyltrimethoxysilane (MTMS) led to materials with lower densities (as low as 0.04 g cm3) while maintaining the same low level of thermal conductivity [143].

20.9

Summary

Phenolic aerogels can be nanoengineered to exhibit a range of desired characteristics via tuning of synthetic and processing conditions. Tunable properties such as surface area, pore

volume, and pore size distribution can produce a wide spectrum of nanostructured materials with unique properties, e.g., catalyst supports, gas filters, gas-storage materials, humidity sensors, acoustic insulation, and thermal insulation. Though phenolic aerogels are often seen as attractive for being precursors to carbon aerogels, they are equally promising in their own right. Most of the research on phenolic aerogels to date has focused on RF aerogels. The most critical factors that determine the final characteristics of RF aerogels are the type and concentration of the catalyst (with respect to the reactants), the pH of the sol, and the concentration of the monomers in the sol. Increasing the R/C ratio increases the particle size with a significant effect on the surface area. By increasing the pH of the sol, the surface area, pore volume, and electrochemical double-layer capacitance of the final carbon aerogels all increase. Adjusting the pH has also led to flexible RF aerogels. Proper control of gelation and curing conditions is essential for completing the polymerization reactions and associated crosslinking of the resulting particles. Studies of other phenolic-type aerogels such as PBO, TPOL, and FPOL have led to the discovery of oxidative ring-fusion aromatization, taking place at about 200–240  C in air, as a key step toward rigidization of the polymeric backbone, which may critically affect several properties of carbon aerogels derived from those materials. Along the way it was also discovered that phenolicresin-derived carbon aerogels contain oxygen only in two forms, in equal amounts: as positively charged heteroaromatic pyrylium cation (O+) and as negatively charged phenoxide (O).

20

Phenolic Aerogels and Their Carbonization

References 1. Fricke, J.: Aerogels – highly tenuous solids with fascinating properties. J. Non-Cryst. Solids. 100, 169–173 (1988) 2. Fricke, J.: Aerogels and their applications. J. Non-Cryst. Solids. 147–148, 356–362 (1992) 3. Hench, L., West, J.: The sol-gel process. Chem. Rev. 90, 33–72 (1990) 4. Brinker, C., Scherer, G.: Sol-Gel Science: The Physics and Chemistry of Sol-Gel Processing. Academic Press Inc. (1990) 5. Carlson, G., Lewis, D., McKinley, K., Richardson, J., Tillotson, T.: Aerogel commercialization: technology, markets and costs. J. Non-Cryst. Solids. 186, 372–379 (1995) 6. Kistler, S.S.: Coherent expanded aerogels and jellies. Nature. 127, 741 (1931) 7. Kistler, S.S.: Coherent expanded-aerogels. J. Phys. Chem. 36, 52–64 (1932) 8. Pekala, R.W.: Organic aerogels from the polycondensation of resorcinol with formaldehyde. J. Mater. Sci. 24, 3221–3227 (1989) 9. Wu, D., Fu, R., Sun, Z., Yu, Z.: Low-density organic and carbon aerogels from the sol-gel polymerization of phenol with formaldehyde. J. Non-Cryst. Solids. 351, 915–921 (2005) 10. Mendenhall, R., Andrews, G., Bruno, J., Albert, D.: Phenolic aerogels by high-temperature direct solvent extraction. U S Pat Ser No 221520. (2000) 11. Pekala, R.W.: Melamine-formaldehyde copolymer aerogels. US5081163 A19920114. (1992) 12. Nguyen, M., Dao, L.: Effects of processing variable on melamine formaldehyde aerogel formation. J. Non-Cryst. Solids. 225, 51–57 (1998) 13. Alviso, C., Pekala, R.W.: Melamine formaldehyde aerogels. Polym Preprints. 32, 242–243 (1991) 14. Li, W., Reichenauer, G., Fricke, J.: Carbon aerogels derived from cresol-resorcinol-formaldehyde for supercapacitors. Carbon. 40, 2955–2959 (2002) 15. Pekala, R.W.: Organic aerogels from the sol-gel polymerization of phenolic-furfural mixtures. US5476878 A19951219. (1995) 16. Pierre, A., Pajonk, G.: Chemistry of aerogels and their applications. Chem. Rev. 102, 4243–4265 (2002) 17. Donthula, S., Mandal, C., Leventis, T., Schisler, J., Saeed, A.M., Sotiriou-Leventis, C., Leventis, N.: Shape-memory poly(isocyanurate-urethane) (PIR-PUR) aerogels for deployable panels and biomimetic applications. Chem. Mater. 29, 4461–4477 (2017) 18. Bang, A., Buyback, C., Sotiriou-Leventis, C., Leventis, N.: Flexible aerogels from hyperbranched polyurethanes: probing the role of molecular rigidity with poly(urethane acrylates) versus poly (urethane norbornenes). Chem. Mater. 26, 6979–6993 (2014) 19. Leventis, N., Chidambareswarapattar, C., Bang, A., SotiriouLeventis, C.: Cocoon-in-web-like superhydrophobic aerogels from hydrophilic polyurea and use in environmental remediation. ACS Appl. Mater. Interfaces. 6, 6872–6882 (2014) 20. Chidambareswarapattar, C., Xu, L., Sotiriou-Leventis, C., Leventis, N.: Robust monolithic multiscale nanoporous polyimides and conversion to isomorphic carbons. RSC Adv. 3, 26459–26469 (2013) 21. Sadekar, A.G., Mahadik, S.S., Bang, A.N., Larimore, Z.J., Wisner, C.A., Bertino, M.F., Kalkan, A.K., Mang, J.T., Sotiriou-Leventis, C., Leventis, N.: “Green” aerogels and porous carbons by emulsion gelation of acrylonitrile. Chem. Mater. 24, 26–47 (2012) 22. Leventis, N., Sotiriou-Leventis, C., Chandrasekaran, N., Mulik, S., Larimore, Z.J., Lu, H., Churu, G., Mang, J.T.: Multifunctional polyurea aerogels from isocyanates and water. A structure-property case study. Chem Mater. 22, 6692–6710 (2010) 23. Al-Muhtaseb, S., Ritter, J.: Preparation and properties of resorcinol-formaldehyde organic and carbon gels. Adv. Mater. 15, 101–114 (2003)

503 24. Laskowski, J., Milow, B., Ratke, L.: Subcritically dried resorcinolformaldehyde aerogels from a base-acid catalyzed synthesis route. Microporous Mesoporous Mater. 197, 308–315 (2014) 25. Mulik, S., Sotiriou-Leventis, C., Leventis, N.: Time-efficient acidcatalyzed synthesis of resorcinol-formaldehyde aerogels. Chem. Mater. 19, 6138–6144 (2007) 26. Barbieri, O., Ehrburger-Dolle, F., Rieker, T., Pajonk, G., Pinto, N., Venkateswara Rao, A.: Small-angle X-ray scattering of a new series of organic aerogels. J. Non-Cryst. Solids. 285, 109–115 (2001) 27. Brandt, R., Fricke, J.: Acetic-acid-catalyzed and subcritically dried carbon aerogels with a nanometer-sized structure and a wide density range. J. Non-Cryst. Solids. 350, 131–135 (2004) 28. Gross, J., Scherer, G.W., Alviso, C.T., Pekala, R.W.: Elastic properties of crosslinked resorcinol-formaldehyde gels and aerogels. J. Non-Cryst. Solids. 211, 132–142 (1997) 29. Pekala, R.W., Alviso, C.T., LeMay, J.D.: Organic aerogels: microstructural dependence of mechanical properties in compression. J. Non-Cryst. Solids. 125, 67–75 (1990) 30. Yang, J., Li, S., Yan, L., Liu, J., Wang, F.: Compressive behaviors and morphological changes of resorcinol-formaldehyde aerogel at high strain rates. Microporous Mesoporous Mater. 133, 134–140 (2010) 31. Schwan, M., Naikade, M., Raabe, D., Ratke, L.: From hard to rubber-like: mechanical properties of resorcinol-formaldehyde aerogels. J. Mater. Sci. 50, 5482–5493 (2015) 32. Alshrah, M., Tran, M.-P., Gong, P., Naguib, H.E., Park, C.B.: Development of high-porosity resorcinol formaldehyde aerogels with enhanced mechanical properties through improved particle necking under CO2 supercritical conditions. J. Colloid Interface Sci. 485, 65–74 (2017) 33. Aghabararpour, M., Mohsenpour, M., Motahari, S., Abolghasemi, A.: Mechanical properties of isocyanate crosslinked resorcinol formaldehyde aerogels. J. Non-Cryst. Solids. 481, 548–555 (2018) 34. Pekala, R.W., Schaefer, D.: Structure of organic aerogels 1. Morphology and scaling. Macromoleculres. 26, 5487–5493 (1993) 35. Gebert, M., Pekala, R.: Fluorescence and light-scattering studies of sol-gel reactions. Chem. Mater. 6, 220–226 (1994) 36. Jirglova, H., Perez-Cadenas, A., Maldonado-Hodar, F.: Synthesis and properties of phloroglucinol-phenol-formaldehyde carbon aerogels and xerogels. Langmuir. 25, 2461–2466 (2009) 37. Durairaj, R.: Resorcinol: Chemistry, Technology and Applications, pp. 186–187. Springer, Berlin (2005) 38. Sprung, M.: Reactivity of phenols toward paraformaldehyde. J. Am. Chem. Soc. 63, 334–343 (1941) 39. Pizzi, A., Mittal, K.: Resorcinol Adhesive, Handbook of Adhesive Technology, 2nd edn. Marcel Dekker, Inc., New York (2003) 40. Mulik, S., Sotiriou-Levetis, C., Leventis, N.: Acid-catalyzed timeefficient synthesis of resorcinol-formaldehyde aerogels and crosslinking with isocyanates. Polym Preprints. 47, 364–365 (2006) 41. Yamamoto, T., Nishimura, T., Suzuki, T., Tamon, H.: Control of mesoporosity of carbon gels prepared by sol-gel polycondensation and freeze drying. J. Non-Cryst. Solids. 288, 46–55 (2001) 42. Tamon, H., Ishizaka, H., Mikami, M., Okazaki, M.: Porous structure of organic and carbon aerogels synthesized by sol-gel polycondensation of resorcinol with formaldehyde. Carbon. 35, 791–796 (1997) 43. Fung, A.W.P., Reynolds, G.A.M., Wang, Z., Dresselhaus, M., Dresselhaus, G., Pekala, R.W.: Relationship between particle size and magnetoresistance in carbon aerogels prepared under different catalyst conditions. J. Non-Cryst. Solids. 186, 200–208 (1995) 44. Horikawa, T., Hayashi, J., Muroyama, K.: Controllability of pore characteristics of resorcinol-formaldehyde carbon aerogel. Carbon. 42, 1625–1633 (2004) 45. Horikawa, T., Hayashi, J., Muroyama, K.: Size control and characterization of spherical carbon aerogel particles from resorcinolformaldehyde resin. Carbon. 42, 169–175 (2004)

20

504 46. Pahl, R., Bonse, U., Pekala, R.W., Kinney, J.: SAXS investigations on organic aerogels. J. Appl. Crystallogr. 24, 771–776 (1991) 47. Saliger, R., Bock, V., Petricevic, R., Tillotson, T., Geis, S., Fricke, J.: Carbon aerogels from dilute catalysis of resorcinol with formaldehyde. J. Non-Cryst. Solids. 221, 144–150 (1997) 48. Fairen-Jimenez, D., Carrasco-Marin, F., Moreno-Castilla, C.: Interand intra-primary-particle structure of monolithic carbon aerogels obtained with varying solvents. Langmuir. 24, 2820–2825 (2008) 49. Mirzaeian, M., Hall, P.: The control of porosity at nano scale in resorcinol formaldehyde carbon aerogels. J. Mater. Sci. 44, 2705–2713 (2009) 50. Tamon, H., Ishizaka, H.: Porous characterization of carbon aerogels. Carbon. 36, 1397–1399 (1998) 51. Lin, C., Ritter, J.: Effect of synthesis pH on the structure of carbon xerogels. Carbon. 35, 1271–1278 (1997) 52. Job, N., Pirard, R., Marien, J., Pirard, J.: Porous carbon xerogels with texture tailored by pH control during sol-gel process. Carbon. 42, 619–628 (2004) 53. Feng, Y., Miao, L., Tanemura, M., Tanemura, S., Suzuki, K.: Effects of further adding of catalysts on nanostructures of carbon aerogels. Mater. Sci. Eng. B. 148, 273–276 (2008) 54. Schwan, M., Ratke, L.: Flexibilization of resorcinol-formaldehyde aerogels. J. Mater. Chem. A. 1, 13462–135468 (2013) 55. Tannert, R., Schwan, M., Ratke, L.: Reduction of shrinkage and brittleness for resorcinol-formaldehyde aerogels my means of a pH-controlled sol-gel process. J. Supercrit. Fluids. 106, 57–61 (2015) 56. Schwan, M., Tannert, T., Ratke, L.: New soft and spongy resorcinol-formaldehyde aerogels. J. Supercrit. Fluids. 107, 201–208 (2016) 57. Tannert, R., Schwan, M., Rege, A., Eggeler, M., Cesar da Silva, C., Bartsch, M., Milow, B., Itskov, M., Ratke, L.: The threedimensional structure of flexible resorcinol-formaldehyde aerogels investigated by means of holotomography. J. Sol-Gel Sci. Technol. 84, 391–399 (2017) 58. Conceicao, F., Carrott, P.J.M., Ribeiro Carrott, M.M.L.: New carbon materials with high porosity in the 1–7 nm range obtained by chemical activation with phosphoric acid of resorcinolformaldehyde aerogels. Carbon. 47, 1874–1877 (2009) 59. Merzbacher, C., Meier, S., Pierce, J., Korwin, M.: Carbon aerogels as broadband non-reflective materials. J. Non-Cryst. Solids. 285, 210–215 (2001) 60. Fairen-Jimenez, D., Carrasco-Marin, F., Moreno-Castilla, C.: Porosity and surface area of monolithic carbon aerogels prepared using alkaline carbonates and organic acids as polymerization catalysts. Carbon. 44, 2301–2307 (2006) 61. Berthon, S., Barbieri, O., Ehrburger-Dolle, F., Geissler, E., Achard, P., Bley, F., Hecht, A.-M., Livet, F., Pajonk, G., Pinto, N., Rigacci, A., Rochas, C.: DLS and SAXS investigations of organic gels and aerogels. J. Non-Cryst. Solids. 285, 154–161 (2001) 62. Brandt, R., Petricevic, R., Proebstle, H., Fricke, J.: Acetic acid catalyzed carbon aerogels. J. Porous. Mater. 10, 171–178 (2003) 63. Reuss, M., Ratke, L.: Subcritically dried RF-aerogels catalyzed by hydrochloric acid. J. Sol-Gel Sci. Technol. 47, 74–80 (2008) 64. Baumann, T., Satcher, J., Gash, A.: Preparation of hydrophobic organic aerogels. US Pat Appl US2002173554 A1 20021121. (2002) 65. March, J.: Advanced Organic Chemistry, Reactions Mechanisms and Structure, 4th edn, pp. 548–550. Wiley, New York (1992) 66. Lermontov, S.A., Malkova, A.N., Sipyagina, N.A., Straumal, E.A., Baranchikov, A.E., Yorov, K.E., Ivanov, V.K.: Facile synthesis of fluorinated resorcinol-formaldehyde aerogels. J. Fluor. Chem. 193, 1–7 (2017) 67. Moudrakovski, I., Ratcliffe, C., Ripmeester, J., Wang, L., Exarhos, G., Baumann, T., Satcher, J.: Nuclear magnetic resonance studies of resorcinol-formaldehyde aerogels. J. Phys. Chem. B. 109, 11215–11222 (2005)

C. Sotiriou-Leventis et al. 68. Werstler, D.: Quantitative carbon-13 NMR characterization of aqueous formaldehyde resins: 2 resorcinol-formaldehyde resins. Polymer. 27, 757–764 (1986) 69. Pekala, R.W., Alviso, C.T., Kong, F.M., Hulsey, S.S.: Aerogels derived from multifunctional organic monomers. J. Non-Cryst. Solids. 145, 90–98 (1992) 70. Al-Muhtaseb, S.A., Ritter, J.A.: Preparation and properties of resorcinol-formaldehyde organic and carbon aerogels. Adv. Mater. 15, 101–114 (2003) 71. Mulik, S., Sotiriou-Leventis, C., Leventis, N.: Macroporous electrically conducting carbon networks by pyrolysis of isocyanatecross-linked resorcinol-formaldehyde aerogels. Chem. Mater. 20, 6985–6997 (2008) 72. Holly, F.W., Cope, A.C.: Condensation products of aldehydes and ketones with o-aminobenzyl alcohol and o-hydroxybenzylamine. J. Am. Chem. Soc. 66, 1875–1879 (1944) 73. Ning, X., Ishida, H.J.: Phenolic materials via ring-opening polymerization: synthesis and characterization of bisphenol-A based benzoxazines and their polymers. Polym Sci. Part A Polym. Chem. 32, 1121–1129 (1994) 74. Ning, X., Ishida, H.: Phenolic materials via ring-opening polymerization of benzoxazines: effect of molecular structure on mechanical and dynamic mechanical properties. J. Polym. Sci. B Polym. Phys. 32, 921–927 (1994) 75. Kumar, K.S.S.: Polybenzoxazines and state-of-the-art hightemperature polymers in Polybenzoxazines: chemistry and properties, Chapter 1. In: Kumar, K.S.S., Raghunadhan Nair, C.P. (eds.) iSmithers Rapra Technology: Shrewsbury, Shropshire (2010) 76. Lorjai, P., Chaisuwan, T., Wongkasemjit, S.: Porous structure of polybenzoxazine-based organic aerogel prepared by sol-gel process and their carbon aerogels. J. Sol-Gel Sci. Technol. 52, 56–64 (2009) 77. Takeichi, T., Kano, T., Agag, T.: Synthesis and thermal cure of high molecular weight polybenzoxazine precursors and the properties of the thermosets. Polymer. 46, 12172–12180 (2005) 78. Andronescu, C., Garea, S.A., Deleanu, C., Lovu, H.: Characterization and curing kinetics of new benzoxazine monomer based on aromatic diamines. Thermochim. Acta. 530, 42–51 (2012) 79. Dunkers, J., Ishida, H.: Reaction of benzoxazine-based phenolic resins with strong and weak carboxylic acids and phenols as catalysts. J. Polym. Sci. Part A Polym. Chem. 37, 1913–1921 (1999) 80. Wang, Y.-X., Ishida, H.: Cationic ring-opening polymerization of benzoxazines. Polymer. 40, 4563–4570 (1999) 81. Wang, Y.-X., Ishida, H.: Synthesis and properties of new thermoplastic polymers from substituted 3,4-dihydro-2H-1,3benzoxazines. Macromolecules. 33, 2839–2847 (2000) 82. Liu, C., Shen, D., Sebastián, R.M., Marquet, J., Schönfeld, R.: Mechanistic studies on ring-opening polymerization of benzoxazines. Macromolecules. 44, 4616–4622 (2011) 83. Ran, Q.-C., Zhang, D.-X., Zhu, R.-Q., Gu, Y.: The structural transformation during polymerization of benzoxazine/FeCl3 and the effect on the thermal stability. Polymer. 53, 4119–4127 (2012) 84. Liu, C., Shen, D., Sebastián, R.M., Marquet, J., Schönfeld, R.: Catalyst effects on the ring-opening polymerization of 1,3-benzoxazine and on the polymer structure. Polymer. 54, 2873–2878 (2013) 85. Mahadik-Khanolkar, S., Donthula, S., Sotiriou-Leventis, C., Leventis, N.: Polybenzoxazine aerogels. 1. High-yield room-temperature acid-catalyzed synthesis of robust monoliths, oxidative aromatization, and conversion to microporous carbons. Chem. Mater. 26, 1303–1317 (2014) 86. Katsoulidis, A.P., Kanatzidis, M.G.: Phloroglucinol based microporous polymeric organic frameworks with –OH functional groups and high CO2 capture capacity. Chem. Mater. 23, 1818–1824 (2011)

20

Phenolic Aerogels and Their Carbonization

87. Majedi Far, H., Donthula, S., Taghvaee, T., Saeed, A.M., Garr, Z., Sotiriou-Leventis, C., Leventis, N.: Air-oxidation of phenolic resin aerogels: backbone reorganization, formation of ring-fused pyrylium cations, and the effect on microporous carbons with enhanced surface areas. RSC Adv. 7, 51104–51120 (2017) 88. López, G.P., Castner, D.G., Ratner, B.D.: XPS oxygen 1s binding energies for polymers containing hydroxyl, ether, ketone and ester groups. Surf. Interface Anal. 17, 267–272 (1991) 89. Briggs, D., Beamson, G.: XPS studies of the oxygen 1s and 2s levels in a wide range of functional polymers. Anal. Chem. 65, 1517–1523 (1993) 90. Ganguly, A., Sharma, S., Papakonstantinou, O., Hamilton, J.: Probing the thermal deoxygenation of graphene oxide using highresolution in situ x-ray-based spectroscopies. J. Phys. Chem. C. 115, 17009–17019 (2011) 91. Ciriello, R., Guerrieri, A., Pavese, F., Salvi, A.M.: Electrosynthesized, non-conducting films of poly(2-naphthol): electrochemical and XPS investigations. Anal. Bioanal. Chem. 392, 913–926 (2008) 92. Lawrinenko, M., Laird, D.A.: Anion exchange capacity of biochar. Green Chem. 17, 4628–4636 (2015) 93. Kim, C.M., Jeong, H.S., Kim, E.H.: NEXAFS and XPS characterization of molecular oxygen adsorbed on Ni(100) at 80 K. Surf. Sci. 459, L457–L461 (2000) 94. Wu, L.Q., Li, S.Q., Li, Y.C., Li, Z.Z., Tang, G.D., Qi, W.H., Xue, L.C., Ding, L.L., Ge, X.S.: Presence of monovalent oxygen anions in oxides demonstrated using x-ray photoelectron spectra. Appl. Phys. Lett. 108, 021905 (2016) 95. Saeed, A.M., Rewatkar, P.M., Majedi Far, H., Taghvaee, T., Donthula, S., Mandal, C., Sotiriou-Leventis, C., Leventis, N.: Selective CO2 sequestration with monolithic bimodal micro/ macroporous carbon aerogels derived from stepwise pyrolytic decomposition of polyamide-polyimide-polyurea random copolymers. ACS Appl. Mater. Interfaces. 9, 13520–13536 (2017) 96. Lu, X., Arduini-Schuster, M., Kuhn, J., Nilsson, O., Fricke, J., Pekala, R.: Thermal conductivity of monolithic organic aerogels. Science. 255, 971–972 (1992) 97. Yoldas, B., Annen, M., Bostaph, J.: Chemical engineering of aerogel morphology formed under nonsupercritical conditions for thermal insulation. Chem. Mater. 12, 2475–2484 (2000) 98. Alviso, C., Pekala, R.W., Gross, J., Lu, X., Caps, R., Fricke, J.: Resorcinol-formaldehyde and carbon aerogel microspheres. Mater. Res. Soc. Sym. Proc., 521–525 (1996) 99. Hrubesh, L., Pekala, R.: Thermal properties of organic and inorganic aerogels. J. Mater. Res. 9, 731–738 (1994) 100. Rettelbach, T., Ebert, H., Caps, R., Fricke, J., Alviso, C., Pekala, R.: Thermal conductivity of resorcinol-formaldehyde aerogels. Therm. Cond. 23, 407–418 (1996) 101. Homonoff, E.: New filtration materials for the new millennium. Book of Papers – International Nonwovens Technical Conference, Dallas (2000) Sept 26–28, 2000, 8.1–8.6 102. Yuan, Z., Govind, R.A., Misra, R.P., Drahushuk, L.W., Strano, M.S., Blanksctein, D., Agrawal, K.V.: Mechanism and prediction of gas permeation through sub-nanometer graphene pores: comparison of theory and simulation. ACS Nano. 11, 7974–7987 (2017) 103. Kabbour, H., Baumann, T.F., Satcher Jr., J.H., Saulnier, A., Ahn, C. C.: Toward new candidates for hydrogen storage: high-surface-area carbon aerogels. Chem. Mater. 18, 6085–6087 (2006) 104. Anas, M., Unsal, S., Erkey, C.: Investigation of various aerogels as adsorbents for methane storage. (2017). https://doi.org/10.1016/j. supflu.2017.11.032 105. Chung, V.P., Yip, M.-C., Fang, W.: Resorcinol-formaldehyde aerogels for CMOS-MEMS capacitive humidity sensor. Sensors Actuators B Chem. 214, 181–188 (2015)

505 106. Berthon-Fabry, S., Langohr, D., Achard, P., Charrier, D., Djurado, D., Ehrburger-Dolle, F.: Anisotropic high-surface-area carbon aerogels. J. Non-Cryst. Solids. 350, 136–144 (2004) 107. Farmer, J., Fix, D., Mack, G., Pekala, R.W., Poco, J.: Capacitive deionization of NH4ClO4 solutions with carbon aerogel electrodes. J. Appl. Electrochem. 26, 1007–1018 (1996) 108. Petricevic, R., Glora, M., Fricke, J.: Planar fiber reinforced carbon aerogels for application in PEM fuel cells. Carbon. 39, 857–867 (2001) 109. Gierszal, K., Jaroniec, M.: Carbons with extremely large volume of uniform Mesopores synthesized by carbonization of phenolic resin film formed on colloidal silica template. J. Am. Chem. Soc. 128, 10026–10027 (2006) 110. Tao, Y., Endo, M., Kaneko, K.: Hydrophilicity-controlled carbon aerogels with high mesoporosity. J. Am. Chem. Soc. 131, 904–905 (2009) 111. Marie, J., Berthon-Fabry, S., Achard, P., Chatenet, M., Pradourat, A., Chainet, E.: Highly dispersed platinum on carbon aerogels as supported catalysts for PEM fuel cell-electrodes: comparison of two different synthesis paths. J. Non-Cryst. Solids. 350, 88–96 (2004) 112. Wicks, Z., Jones, F., Pappas, S.: Organic Coatings: Science and Technology, Vol. 1: Applications, Properties, and Performance, pp. 84–87. Wiley, New York (1994) 113. Raetzsch, M., Bucka, H., Ivanchev, S., Pavlyuchenko, V., Leitner, P., Primachenko, O.: The reaction mechanism of the transetherification and crosslinking of melamine resins. Macromol. Symp. 217, 431–443 (2004) 114. Mandal, C., Donthula, S., Soni, R., Bertino, M., Sotiriou-Leventis, C., Leventis, N.: Light scattering and haze in TMOS-co-APTES silica aerogels. J. Sol-Gel Sci. Technol. 90, 127–139 (2018) 115. Mandal, C., Donthula, S., Majedi Far, H., Saeed, A.M., SotiriouLeventis, C., Leventis, N.: Transparent, mechanically strong, thermally insulating cross-linked silica aerogels for energy-efficient windows. J. Sol-Gel Sci. Technol. 92, 84–100 (2019) 116. Perez-Caballero, F., Peikolainen, A.-L., Uibu, M., Kuusik, R., Volobujeva, O., Koel, M.: Preparation of carbon aerogels from 5-methylresorcinol-formaldehyde gels. Microporous Mesoporous Mater. 108, 230–236 (2008) 117. Peikolainen, A.-L., Perez-Caballero, F., Koel, M.: Low-density organic aerogels from oil shale by-product 5-methylresorcinol. Oil Shale. 25, 348–358 (2008) 118. Tanaka, S., Katayama, Y., Tate, M., Hillhouse, H., Miyake, Y.: Fabrication of continuous mesoporous carbon films with facecentered orthorhombic symmetry through a soft templating pathway. J. Mater. Chem. 17, 3639–3645 (2007) 119. Baumann, T., Satcher, J.: Template-directed synthesis of periodic macroporous organic and carbon aerogels. J. Non-Cryst. Solids. 350, 120–125 (2004) 120. Lei, Q., Song, H., Chen, X., Li, M., Li, A., Tang, B., Zhou, D.: Effects of graphene oxide addition on the synthesis and supercapacitor performance of carbon aerogel particles. RSC Adv. 6, 40683–40690 (2016) 121. Worsley, M.A., Satcher, J.H., Baumann, T.F.: Synthesis and characterization of monolithic carbon aerogel composites containing doubled-walled carbon nanotubes. Langmuir. 24, 9763–9766 (2008) 122. Worsley, M.A., Satcher, J.H., Baumann, T.F.: Enhanced thermal transport in carbon aerogel nanocomposites containing doublewalled carbon nanotubes. J. Appl. Phys. 105, 1–5 (2009) 123. Haghgoo, M., Yousefi, A.A., Zohouriaan, M., Mohammad, J., Celzard, A., Fierro, V., Leonard, A.F., Leonard, A., Job, N.: Characterization of multi-walled carbon nanotube dispersion in resorcinol-formaldehyde aerogels. Microporous Mesoporous Mater. 184, 97–104 (2014) 124. Bekyarova, E., Kaneko, K.: Structure and physical properties of tailor-made Ce, Zr-doped carbon aerogels. Adv. Mater. 12, 1625–1628 (2000)

20

506 125. Baumann, T., Fox, G., Satcher, J., Yoshizawa, N., Fu, R., Dresselhaus, M.: Synthesis and characterization of copper-doped carbon aerogels. Langmuir. 18, 7073–7076 (2002) 126. Baumann, T., Worsley, M., Han, T., Satcher, J.: High surface area carbon aerogel monoliths with hierarchical porosity. J. Non-Cryst. Solids. 354, 3513–3515 (2008) 127. Baumann, T., Satcher, J.: Homogeneous incorporation of metal nanoparticles into ordered macroporous carbons. Chem. Mater. 15, 3745–3747 (2003) 128. Maldonado-Hodar, F., Perez-Cadenas, A., Moreno-Castilla, C.: Morphology of heat – treated tungsten doped monolithic carbon aerogels. Carbon. 41, 1291–1299 (2003) 129. Job, N., Pirard, R., Marien, J., Pirard, J.: Synthesis of transition metal-doped carbon xerogels by solubilization of metal salts in resorcinol-formaldehyde aqueous solution. Carbon. 42, 3217–3227 (2004) 130. Maldonado-Hodar, F., Ferro-Garcia, M., Rivera-Utrilla, J., Moreno-Castilla, C.: Synthesis and textural characteristics of organic aerogels, transition metal-containing organic aerogels, and their carbonized derivatives. Carbon. 37, 1199–1205 (1999) 131. Leventis, N., Chandrasekaran, N., Sotiriou-Leventis, C., Mumtaz, A.: Smelting in the age of nano: iron aerogels. J. Mater. Chem. 19, 63–65 (2009) 132. Moreno-Castilla, C., Maldonado-Hodar, F.: Carbon aerogels for catalysis applications: an overview. Carbon. 43, 455–465 (2005) 133. Job, N., Pirard, R., Vertruyen, B., Colomer, J., Marien, J., Pirard, J.: Synthesis of transition metal-doped carbon xerogels by cogelation. J. Non-Cryst. Solids. 353, 2333–2345 (2007) 134. Fu, R., Baumann, T., Cronin, S., Dresselhaus, G., Dresselhaus, M., Satcher, J.: Formation of graphitic structures in cobalt- and nickeldoped carbon aerogels. Langmuir. 21, 2647–2651 (2005) 135. Steiner III, S.A., Baumann, T.F., Kong, J., Satcher Jr., J.H., Dresselhaus, M.S.: Iron-doped carbon aerogels: novel porous substrates for direct growth of carbon nanotubes. Langmuir. 23(9), 5161–5166 (2007) 136. Steiner III, S.A., Baumann, T.F., Bayer, B.C., Blume, R., Worsley, M.A., Moberly Chan, W.J., Shaw, E.L., Schloegl, R., Hart, A.J., Hofmann, S., Wardle, B.L.: Nanoscale zirconia as a non-metallic catalyst for graphitization of carbon and growth of single- and multiwall carbon nanotubes. J. Am. Chem. Soc. 131, 12144–12154 (2009) 137. Kudo, A., Steiner III, S.A., Bayer, B.C., Kidambi, P.R., Hofmann, S., Strano, M.S., Wardle, B.L.: CVD growth of carbon nanostructures from zirconia: mechanisms and a method for enhancing yield. J. Am. Chem. Soc. 136, 17808–17817 (2014) 138. Steiner III, S.A.: Engineering carbon nanostructures: development of novel aerogel-nanotube composites and optimization techniques for nanotube growth. SM Thesis, Massachusetts Institute of Technology (2006). http://hdl.handle.net/1721.1/36216. Last accessed 21 June 2020 139. Attia, S.M., Sharshar, T., Abd-Elwahed, A.R., Tawfik, A.: Study of transport properties and conduction mechanism of pure and composite resorcinol formaldehyde aerogel doped with co-ferrite. Mater. Sci. Eng. B. 178, 897–910 (2013) 140. Sanchez-Polo, M., Rivera-Utrilla, J., Mendez-Diaz, J., LopezPenalver, J.: Metal-doped carbon aerogels new materials for water treatments. Ind. Eng. Chem. Res. 47, 6001–6005 (2008) 141. Leventis, N., Donthula, S., Mandal, C., Ding, M.S., SotiriouLeventis, C.: Explosive versus thermite behavior in iron (0) aerogels infiltrated with perchlorates. Chem. Mater. 27, 8126–8137 (2015)

C. Sotiriou-Leventis et al. 142. Berthon-Fabry, S., Hildenbrand, C., Ilbizian, P.: Lightweight superinsulating resorcinol-formaldehyde-APTES benzoxazine aerogel blankets for space applications. Eur. Polym. J. 78, 25–37 (2016) 143. Berthon-Fabry, S., Hildenbrand, C., Ilbizian, P., Jones, E., Tavera, S.: Evaluation of lightweight and flexible insulating aerogel blankets based on resorcinol-formaldehyde-silica for space applications. Eur. Polym. J. 93, 403–416 (2017)

Chariklia Sotiriou-Leventis received her Ph.D. from Michigan State University in organic chemistry in 1987. She was a postdoctoral fellow at Northeastern (1987–1989) and Harvard (1989–1992) University, and worked at Ciba Corning Diagnostics (1992–1994). Currently, she is a Professor of Chemistry at Missouri University of Science & Technology. Her research interests are in synthesis of organic materials, aerogels and supramolecular chemistry.

Nicholas Leventis received his Ph.D. from Michigan State University in organic chemistry in 1985. Currently, he is Director, Research of Aspen Aerogels. In 2019 he retired from the Missouri University of Science and Technology as a Curators’ Distinguished Professor. His aerogel work includes polymer-crosslinked aerogels, aerogels from most classes of organic polymers, interpenetrating organic-inorganic aerogels, and metallic, ceramic and carbon aerogels.

Sudhir Mulik received his Ph.D. from Missouri University of Science & Technology in 2008. He is currently Research Scientist at The Dow Chemical Company at Collegeville, PA. His work on aerogels introduced large-scale acid-catalyzed synthesis of phenolic aerogels, polymer-crosslinked organic and templated silica aerogels, and crosslinking in one pot via surface initiated free radical polymerization.

Isocyanate-Derived Aerogels and Nanostructure– Materials Properties Relationships

21

Nicholas Leventis

Contents 21.1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 508

21.2

The Gelation Process and Nanomorphology . . . . . . . . . . 509

21.3 21.3.1

The Chemistry of the Isocyanate Group . . . . . . . . . . . . . . Reaction of Isocyanates with Nitrogen-Based Nucleophiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Reaction of Isocyanates with Oxygen-Based Nucleophiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Condensation of Isocyanates: Polyisocyanurate and PIR-PUR Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

21.3.2 21.3.3 21.4 21.4.1 21.4.2 21.4.3

Polyurea Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Polyurea Aerogels from Isocyanates and Amines . . . . . . . Polyurea Aerogels from Isocyanates and Water . . . . . . . . . Polyurea Aerogels from Isocyanates and Mineral Acids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

512 513 515 520 521 521 526 538

21.5.2

Polyurethane Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 540 Polyurethane Aerogels from a Flexible Triisocyanate (Desmodur N3300A) and Short Aliphatic Diols . . . . . . . . . 543 Deconvolving the Properties of the Polymeric Nodes from the Functionality of the Monomer: Poly(Urethane-Acrylate) and Poly(Urethane-Norbornene) Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 548

21.6 21.6.1 21.6.2

Polyimide Aerogels and Polyamide Aerogels . . . . . . . . . 553 Polyimide Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 553 Polyamide Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 555

21.7

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 560

21.5 21.5.1

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561

Abstract

Isocyanates are a class of inexpensive yet versatile organic compounds that serve as the basis for a wide variety of familiar everyday products such as polyurethane foams,

N. Leventis (*) Department of Chemistry, Missouri University of Science and Technology, Rolla, MO, USA Department of Research and Development, Aspen Aerogels, Inc., Northborough, MA, USA

varnishes, sealants, etc. that can be found in virtually every home. Isocyanate-derived aerogels were first reported in the mid-1990s, followed by only a handful of further studies over the next 15 years. Isocyanate-derived aerogels have received renewed interest over the past decade, however, accompanied by a rapid growth in the number of scholarly works and patents. This expanding interest in isocyanate-derived aerogels can be attributed in part to the extremely broad array of nanostructures, morphologies, and functional properties that can be embodied by these materials and can be achieved even by a single chemical composition simply by tuning the synthetic parameters. The subject of isocyanate-derived aerogels now encompasses all categories of isocyanate-derivable polymers, including polyureas, polyisocyanurates, polyurethanes, polyimides, and polyamides. For these reasons, isocyanate-derived aerogels provide a fertile ground for fundamental studies of nanostructure-material property relationships. Additionally, numerous isocyanate-derived aerogels are carbonizable with high yields, making them compelling alternatives to resorcinol-formaldehyde aerogels. Several isocyanate-derived aerogels are in fact already in different stages of commercialization. This chapter places isocyanate-derived aerogels in the general historical context of aerogels and reviews the various types of these materials that have been produced to date through a nanostructure-material property perspective. Applications of polymeric isocyanate-derived aerogels and their carbonized derivatives that are reviewed include thermal and acoustic insulation, oil adsorption, shapememory biomedical devices, catalysts and catalyst supports, water vapor absorption, and CO2 sequestration. Keywords

Isocyanate · Aerogel · Polyurea · Polyurethane · Polyamide · Polyimide · Mineral acid · Structure-property relationships · K-index

© Springer Nature Switzerland AG 2023 M. A. Aegerter et al. (eds.), Springer Handbook of Aerogels, Springer Handbooks, https://doi.org/10.1007/978-3-030-27322-4_21

507

508

21.1

N. Leventis

Introduction

Aerogels are low-density solid materials that exhibit a high degree of open porosity and are typically prepared by drying a gel in a way that preserves most of the gel’s initial volume in the final dry form [1, 2]. Aerogels were invented in the late 1920s as a means of studying the solid framework of wet gels [3]; however owing to their unique material properties, soon after they took on a life of their own, independent of fundamental colloidal studies. By their nature, aerogels are a subclass of the much broader domain of porous materials. Traditional macroporous foams used in insulation applications, for example, can be prepared by using blowing agents [4]. A variety of pore-solid architected aerogel-like materials featuring micron-scale and submicron-scale porosity can also now be prepared through a variety of techniques (see, e.g., ▶ Chaps. 37, ▶ 43, and ▶ 45). This chapter focuses specifically on aerogels obtained by drying of sol–gel-derived wet gels. Depending on the degree of shrinkage that results upon drying, which in turn depends on the drying method and the physicochemical properties of the gel, the resulting porous materials are often distinguished as xerogels, ambigels, or aerogels [5–7] (see ▶ Chap. 1). Briefly, xerogels and ambigels are typically obtained by evaporating the porefilling solvent of a wet gel under ambient pressure. For xerogels, the evaporating solvent is the gelation solvent itself (e.g., methanol, acetone, water, etc.) that, upon evaporating, usually results in a large degree of drying shrinkage and thus higher-density materials. For ambigels, prior to drying, the gelation solvent is exchanged with a low-surface-tension solvent (e.g., pentane, hexane, cyclohexane, etc.) resulting in much lower-density materials [6]. Aerogels are typically obtained by using a pressure vessel to convert the pore-filling solvent of a gel into a supercritical fluid (SCF) and then subsequently venting off the supercritical fluid like a gas [5] (see ▶ Chaps. 4 and ▶ 65). This process, known as supercritical drying, eliminates the surface tension (capillary) stresses that are exerted on the network when the solvent is merely evaporated; thus any network shrinkage that does result can primarily be attributed to molecular-level interactions of the network with itself that occur when solvent is removed from its pores [8]. It is noted that the above-listed drying methods do not define the corresponding materials (xerogels, ambigels, and aerogels) but rather represent how such materials are most commonly made; indeed, other drying methods, such as freeze drying, can yield materials with similar properties to those obtained via supercritical drying (see ▶ Chaps. 1, ▶ 4 and ▶ 5). Xerogels, ambigels, and aerogels comprise a continuum in porosity and share compositional and morphological features with their common parent wet gels, with aerogels exhibiting the lowest density of the three.

The skeletal framework of aerogels can be inorganic or organic in nature or a combination of both. Inorganic aerogels include materials based on oxides, chalcogenides, metals, ceramics (e.g., carbides, nitrides), and carbon. Organic aerogels include materials based on synthetic and natural polymers, the first examples of which (cellulose, nitrocellulose, gelatin, agar, egg albumin, and rubber aerogels) were described by Kistler himself in his pioneering publications in the early 1930s [3, 9]. However, subsequent focus on oxide aerogels, and especially silica, diverted attention away from polymeric aerogels for over 50 years. This changed in the late 1980s with the introduction of aerogels based on phenolic resins, particularly those based on the condensation of resorcinol with formaldehyde (see ▶ Chap. 2). Such was the momentum of phenolic-type aerogels that, despite reports describing new types of polymeric aerogels, resorcinolformaldehyde aerogels were practically synonymous with organic aerogels for almost 20 years. However, this was to change again as a handful of isolated publications and patents in the late 1990s and through the 2000s would plant the seeds for the rapid development of polymeric aerogels that has occurred over the past decade. Included in these works were the first examples of aerogels based on polyurethanes [10], polyimides [11], polydicyclopentadiene [12], polyureas [13], polybenzoxazine [14], conducting polymers [15], and importantly a new class of organic-inorganic nanocomposite aerogels referred to as polymer-crosslinked aerogels (see ▶ Chap. 29) [16]. The latter materials in particular seem to have catalyzed a field-wide shift of attention toward polymeric aerogels, including those based on isocyanate chemistry, which is the subject of this chapter. The first organic aerogels employing isocyanate chemistry were disclosed in a 1994 patent assigned to Imperial Chemical Industries (ICI) [17]. Our interest in using isocyanate chemistry for aerogels however, stemmed from the dramatic increase in mechanical strength observed in polyureacrosslinked silica aerogels [16]. In these materials, a preformed inorganic 3D network of sol–gel-derived silica nanoparticles (i.e., a silica wet gel) served as a template for the deposition of a conformal, nanothin polyurea coating over the entire gel skeletal network. While the morphology of the network did not change, the resulting nanocomposite was extremely strong mechanically: where silica aerogels had historically been notoriously fragile, polyurea-crosslinked silica aerogels were strong enough to be used in armor (see ▶ Chap. 29). At this point, it was reasoned that since the dramatic improvement in strength was brought about by the crosslinking polymer, if the polymer alone could be rendered into an analogous form comprising similarly sized nanoparticles and interparticle connectivity as in the polymercrosslinked oxide aerogels, the resulting all polymer aerogel should exhibit similar mechanical properties. Early polymercrosslinked silica aerogels in the literature were obtained via

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

reaction of various triisocyanates with the hydroxyl groups on the surface of silica; however, the bulk of the polymer in the resulting nanocomposites was identified as polyurea arising from the reaction of isocyanate with strongly adsorbed gelation water remaining on the nanoparticles of the silica wet gel framework [18]. Thus, based on this chemical-topological rationale, a new class of polyurea aerogels was synthesized via reaction of triisocyanates with water, and the property of interest was mechanical strength [19, 20]. Next, owing to the wide variety of inexpensive commercially available diols and triols, the focus shifted toward polyurethane aerogels, which became a new point of departure for exploring the effect of the monomer size and molecular structure on the mechanical properties of the resulting aerogels. Indeed, the mechanical properties of these new polyurethane aerogels spanned from extremely strong materials (in terms of modulus, ultimate strength, and energy absorption capability) to flexible and superelastic materials displaying the shape-memory effect. Concurrently, the field of isocyanate-derived aerogels was also expanded to include polyimides and polyamides as well. Overall, isocyanate-based chemistry has yielded an extremely broad variety of nanostructured materials, and having run full circle, polymeric aerogels again serve as a powerful tool for studying the structure-property relationships of gels in the spirit of Kistler’s pioneering work almost 90 years ago. This review starts with a high-level overview of the nanomorphology of isocyanate-derived aerogels as a function of gelation parameters (Sect. 21.2). Section 21.3 proceeds with an overview of isocyanate chemistry, which, under the right conditions, can be used to achieve the various morphologies presented in Sect. 21.2. Finally, the various categories of isocyanate-derived aerogels are reviewed in detail, including polyurea aerogels (Sect. 21.4), polyurethane and poly

Acetone

1 μm

ρ b = 0.13 g cm–3 Π = 90% v/v σ = 169 m2 g–1

ρ b = 0.11 g cm–3 Π = 91% v/v σ = 307 m2 g–1

Fig. 21.1 SEM of polyurea aerogels at about the same bulk densities (ρb), prepared with an aliphatic triisocyanate (Desmodur N3300A) in three different solvents as shown. Notice the change in morphology from fibrous to particulate. All insets at 5
magnification relative to the

509

(isocyanurate-urethane) (PIR-PUR) aerogels (Sect. 21.5), and polyimide and polyamide aerogels (Sect. 21.6).

21 21.2

The Gelation Process and Nanomorphology

The sol–gel process as it applies to preparation of isocyanatederived aerogels, and most aerogels for that matter, entails reaction in a liquid medium of soluble monomers into insoluble polymers that form a solid network that will eventually become the skeletal framework of the aerogel. The morphology of the skeletal framework of isocyanate-derived aerogels can be extremely diverse, even when the chemical composition of its constituent polymer and density are held constant. This point is illustrated in Fig. 21.1, which shows various possible nanostructures of a certain type of polyurea aerogel, all possessing the same chemical composition and bulk density (further details on this type of aerogels are provided in Sect. 21.4.2) [21]. Through correlations of structure-property relationships in isocyanate-derived aerogels, this morphological diversity has been identified to arise from parameters including the molecular structure of the monomers from which the polymer is derived (e.g., aromatic versus aliphatic, rigid versus flexible, difunctional versus polyfunctional, the functional group density at the monomer molecular level), the solubility properties of the medium (polarity, ability to develop dispersion forces, and hydrogen bonding), and the gelation temperature. Representative examples and associated data are discussed in the reviews of polyurea, polyurethane, polyimide, and polyamide aerogels presented in Sects. 21.4, 21.5, and 21.6.

DMF

DMSO

1 μm

1 μm

ρ b = 0.13 g cm–3 Π = 89% v/v σ = 280 m2 g–1 main images. Π: porosity. σ: surface area via the BET method (N2 sorption). (Reproduced from [21], Copyright 2014 The American Chemical Society)

510

N. Leventis

As illustrated by Fig. 21.2, polymerization of monomers in solution can lead to phase separation of either solid colloidal primary particles (Fig. 21.2 – left branch) or an oily phase of insoluble oligomers (Fig. 21.2 – right branch) [22–24]. In the former case, primary particles tend to be smaller as the functionality and functional group density of the monomers increase, quickly leading to a highly crosslinked, insoluble polymer. This has been demonstrated most vividly with poly

(urethane-acrylate) aerogels derived from trifunctional versus 9-functional (nonafunctional) star and dendritic monomers, respectively: both types of monomers, made using the same rigid aromatic triisocyanate core, yielded primary particles with 17 nm versus 7 nm in radius, respectively [25, 26]. Phase-separated primary colloidal particles move randomly in the sol and upon encountering one another may connect together through chemical bonding of yet unreacted,

Fig. 21.2 Phase separation and growth processes involved in the formation of the skeletal network of gels

Monomers + Catalyst (sol) Phase separation of: Solid polymer

Liquid polymer

Aggregation

Mass-fractal Mass-fr f actal secondary r particles Agglomeration

Skeletal framework f amework fr Accumulation of unreacted monomers and oligomers

Skeletal framework f amework of fr fused mass -fractal -fr f actal nanoparticles

Morphological M Morpho logicall evolution evolution by b coalescence of o liquid polymer poly l mer phases phas a es Growth off bi-continuous bi-continuous network Growth of micro -droplets

Small primary particles

f aments fil Liquid filaments

Bi-continuous network

Small micro-droplets

Larger micro-droplets

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

monomer-related functional groups on their surfaces. If the reaction is fast (and thereby the sticking probability for each encounter is high), the kinetics of network growth is limited only by diffusion, and the process is referred to as diffusionlimited aggregation. Aggregates of this sort are mass-fractal objects [24], the size of which is limited by the fact that their density falls rapidly with distance from the nucleation point at their center; thus, beyond a certain size, there is not enough mass in their outer “layers” to support further growth. These mass-fractal aggregates of primary particles are referred to as secondary particles. On geometric grounds, the pore sizes inside secondary particles are on the same order as the primary particles, and typically they fall in the mesopore range (2–50 nm). Secondary particles likewise random walk, connecting to one another via chemical bonding to form higher aggregates. The sol–gels once secondary particles, or higher aggregates, have formed a continuous interconnected three-dimensional path spanning the liquid medium within the mold. At that point, assuming all monomers have been consumed, the skeletal network consists of interconnected secondary particles or higher aggregates; experimentally the primary particle size identified via small angle x-ray or neutron scattering (SAXS or SANS, respectively) and the particle size calculated from skeletal density and N2-sorption data (via r ¼ 3/(ρs
σ)) should be about equal to one another (r, particle radius; ρs, skeletal density from He-pycnometry;

a

511

σ: BET surface area from N2-sorption porosimetry). This situation is commonplace with oxide aerogels and is likewise encountered with all types of isocyanate-derived aerogels within all the systems reviewed in this chapter (the reader is directed to the original reports in the literature for additional details). However, if unreacted monomers or soluble oligomers remain in the pores after gelation, these species can eventually find their way onto the protoskeletal framework where they can accumulate via reaction with still live functional groups on the surface of the skeletal particles of said framework. Eventually, accumulation of monomer and/or oligomers fills the porous space within secondary particles, and experimentally, the particle size calculated from skeletal density and N2-sorption data reflects a size closer to that of the secondary particles as determined by SAXS or SANS. Under SEM these structures appear as if they consist of random assemblies of particles with a layer of polymer casted over the entire network. This is exemplified here by aerogels made of poly(urethane-norbornene), which are based on a three-branched star monomer featuring an aliphatic trifunctional core that is terminated with norbornene moieties at the ends of all three of its branches (aL-3-NBE; see Fig. 21.3 in Sect. 21.5.1 and ▶ Chap. 23) [25]. The topography of such an arrangement actually resembles that of polymer-crosslinked oxide aerogels, in which silica primary particles are embedded within crosslinker-derived polymer

b

1 mm

1 mm

 b = 0.128  0.002 g cm–3

 b = 0.792  0.010 g cm–3

r = 118 nm

r = 103 nm

R(1) = 16.42  0.92 nm

R(1) = 18.8  2.25 nm

R(2) = 75.1  8.42 nm

R(2) = 53.7  4.57 nm

 = 21

 = 24 m2 g–1

m2 g–1

Fig. 21.3 SEM images at 50 k magnification of poly(urethanenorbornene) aerogels (aL-3-NBE, see Sect. 21.5.2) for two density extremes [25]. Other material data are provided below the images (ρs, bulk density; r, radii from skeletal density/N2-sorption data; R(1), R(2),

radii of primary and secondary particles, respectively, from SAXS data; σ: BET surface area). (Reproduced from [25], Copyright 2014 The American Chemical Society)

21

512

N. Leventis

that fills the space within the aerogel’s secondary particles [27, 28]. Comparisons of particle sizes determined from skeletal density/N2-sorption data and SAXS data have proven useful in understanding the growth mechanisms underlying the formation of polymeric aerogels in general. A third configuration is encountered when, within a series of aerogel samples, the particle size determined using skeletal density/N2-sorption data matches the primary particle size from SAXS data for high-density aerogels but diverges, sometimes by orders of magnitude, at lower densities. This type of density-dependent trend is typically accompanied by obvious morphological changes observable under SEM and a large decrease in BET surface area as density decreases. A representative example of this situation occurs in polyurethane aerogels derived from tris(4-isocyanatophenyl) methane (Desmodur RE – see Fig. 21.38 in Sect. 21.4.2 below), which is a rigid aromatic triisocyanate monomer, and sulfonyl diphenol (SDP) [29], abbreviated as aR-SDPxx (where “xx” stands for the monomer concentration in the sol, see Sect. 21.5 below). Figure 21.4 shows SEMs of aRSDP-xx samples at two extreme densities [29]. To distinguish the low-density morphology from liquid-phase separation (Fig. 21.2, right branch), relevant morphological analysis was performed over a wide range of targeted bulk densities for this composition, described in more detail in the literature (see the Supporting Information for [29] for SEM images of intermediate densities of aR-SDP-xx aerogels). Phase separation of oligomers as an oily liquid (Fig. 21.2, right branch) occurs for large flexible monomers with low functional group density. The “oil” continues to react internally until it solidifies (arrested). Depending on the speed of solidification and the properties of the medium, the structural diversity in such cases can be extremely broad. For example, Fig. 21.4 Selected scanning electron microscopy (SEM) images for aR-SDP-xx aerogels (see text and Sect. 21.5) at two density extremes, at both low and high magnification (a and b). Length scales have been selected to capture the relative sizes of the building blocks and morphology of the networks. Other pertinent data are given below the images (ρs, bulk density; r, radii from skeletal density/N2-sorption data; R(1), radii of primary particles from SAXS; σ: BET surface area). (Adapted from [29], Copyright 2013 The American Chemical Society)

the oily phase may start out as thin hair-like filaments that evolve into small particles embedded within fibers; among other possibilities, the latter morphology may evolve into large particles with fibers emanating from their surface (refer to the discussion surrounding polyureas in Sect. 21.4.2 below). Alternatively, by choosing the right monomers and adjusting the reaction rate accordingly (e.g., via adjustment of monomer concentration or by using an appropriate catalyst), a skeletal framework with a bicontinuous morphology can be “frozen” (kinetically trapped) in place before the hair-like filaments have time to evolve into small droplets. At even slower reaction rates, small liquid droplets may eventually coalesce into larger ones before the polymer solidifies (see Fig. 21.55 in Sect. 21.5.1 below). These various skeletal morphologies translate directly into a wide range of surface areas, texture-induced hydrophobicity, and diverse mechanical properties that include superelasticity and the shape-memory effect. Although the examples raised in this section are all based on isocyanate-derived polymers, the principles can actually be generalized to other types of aerogel as well. For isocyanates, the phase separation process underlying gel formation is ultimately driven by reaction of the isocyanate group with other reactive groups, which is the subject of the next section.

21.3

The Chemistry of the Isocyanate Group

The chemistry of the isocyanate group, –N¼C¼O, is rich and has been studied extensively. This section provides a brief overview of the chemistry of isocyanate as it pertains to the synthesis of aerogels.

a

b

aR-SDP-10

aR-SDP-25 20

1 m

10 m

1 m

10 m

 b = 0.190  0.005 g cm–3

 b = 0.541  0.004 g cm–3

r = 812 nm

r = 74 nm

R(1) = 98.1  2.7 nm

R(1) = 60.2  1.7 nm

 = 2.8 m2 g–1

 = 28 m2 g–1

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

Fig. 21.5 Resonance within the isocyanate (–N¼C¼O) group

–

N

C

N

O

513

+

C

–

+

O

N

C

O

21

N

C

O

+

:Nu

N

C

O

–

Nu Fig. 21.6 Addition of a nucleophile (:Nu) to the isocyanate group

The carbon atom of the isocyanate group is in the 4+ oxidation state as it is in CO2 and thus is oxidatively stable. The reactivity of isocyanate is governed by electron withdrawing attributable to inductive and resonance effects caused by both oxygen and nitrogen (Fig. 21.5), leaving the carbonyl carbon with a partial positive charge and thus susceptible to attack by nucleophiles (:Nu; see Fig. 21.6). Typical nucleophiles are based on N or O atoms. Table 21.1 summarizes the relative reactivities of common N- and O-based active hydrogen nucleophiles by comparing reaction rates with –N¼C¼O without the use of catalysts according to Fig. 21.6 [30]. (“Active hydrogen” refers to nucleophiles that carry all atoms necessary to be stable neutral species. For instance, an alcohol, ROH, is an active hydrogen nucleophile, but the corresponding alkoxide, RO, albeit more reactive, is not.) The reactivity of the isocyanate group (–N¼C¼O) is modulated further by the electron-withdrawing or the electron-donating ability of the groups attached to its N. Aromatic isocyanates are generally more reactive than aliphatic ones [31]. Electron-withdrawing substituents on the aromatic ring increase the positive charge on the carbon atom of –N¼C¼O even further and, all other factors being equal (e.g., steric effects), increase its reactivity toward nucleophiles [32]. Conversely, electron donation due to either resonance (e.g., MeO–) or induction (e.g., CH3–) reduces the reactivity of the –N¼C¼O group. Figure 21.7 cites various aromatic and aliphatic isocyanates in order of their relative reactivity [31]. Of particular interest are also the often overlooked reactions of isocyanate with isocyanate-derived reaction products, e.g., the reaction of isocyanate with urea or urethane to form biuret or allophanate, respectively (see Table 21.1 for relative reaction rates). Finally, in addition to extraneous nucleophiles such as those listed in Table 21.1, the isocyanate group can also react with itself in a self-addition/self-condensation fashion [33].

Table 21.1 Relative reactivities of active hydrogen nucleophiles based on N and O toward –N¼C¼O [30]. Formulas have been written to emphasize the nucleophilic centers Active hydrogen nucleophile Primary aliphatic amines (R–NH2) Secondary aliphatic amines (R–NH–R′) Primary aromatic amine (Ar–NH2) Primary hydroxyl (e.g., R–OH) Water (H–OH) Carboxylic acids (R (C¼O)–OH) Secondary hydroxyl (e.g., RCH(OH)R) Ureas (–NH(C¼O) NH–) Tertiary hydroxyl (R3C–OH) Urethane (–NH(C¼O) O–)

Relative reaction rates (uncatalyzed at 25  C) 100,000

Product classification Urea

20,000–50,000

Urea

200–300

Urea

100

Urethane

100 40

Urea Amide

30

Urethane

15

Biuret

0.5

Urethane

0.3

Allophanate

21.3.1 Reaction of Isocyanates with NitrogenBased Nucleophiles Nitrogen-based nucleophiles include amines, ureas, and urethanes, which upon reaction with isocyanates yield ureas, biurets, and allophanates, respectively.

Reaction of –N5C5O with Amines: Formation of Ureas Nucleophilic addition of an amine group to an isocyanate yields a urea (Fig. 21.8). The reaction is self-catalytic and is typically very fast and exothermic. The 1,3-proton transfer tautomerization between the two nitrogens is catalyzed by another yet unreacted molecule of the amine, which acts as a base. Accordingly, aromatic amines, being weaker bases than aliphatic amines, react more slowly than their aliphatic counterparts [34]. Reaction of –N5C5O with Ureas: Formation of Biurets A urea, such as the reaction product of the reaction of Fig. 21.8, is an N-based nucleophile in and of itself and can

514

N. Leventis

NCO

O2N

Me

NCO

NCO

CnH2n+1NCO

NCO

MeO

Fig. 21.7 The reactivity of the isocyanate group as a function of substitution on the nitrogen of NCO

H R

N

C

O

+

N

NH2

R'

H

–

N + R'

R

H N

H N

R'NH2

R'

R O Urea

O

Fig. 21.8 Urea formation via reaction of an isocyanate with an amine

a R

H N N

C

O

+

H N

O

O R"

R'

100–150 °C

R"

R N H

N

N H

O

R' Biuret

b R

H N N

C

O

+

O

O

O

R'

R" O

120–150 °C

R"

R N H

N

O

R' Allophanate

Fig. 21.9 Reaction of isocyanates with ureas to biurets (a) and with urethanes to allophanates (b). Arrows show the sites of the initial nucleophilic attacks. The overall reactions involve several intermediates. Atoms are color-coded for tracking purposes

attack excess isocyanate to yield a biuret (Fig. 21.9a). As noted in Table 21.1, however, the reactivity of ureas as nucleophiles toward –N¼C¼O is low, and as a result the reaction requires more aggressive conditions (100–150  C) [31, 35]. Formation of biurets serves as a crosslinking mechanism for polyureas. Desmodur N3200 from Covestro, LLC (formerly Bayer MaterialScience), a triisocyanate that has been used extensively in the synthesis of polyurea aerogel and polyurea-crosslinked oxide aerogels (see ▶ Chap. 29), is a

biuret derived from hexamethylene diisocyanate (HDI) (Fig. 21.10) [36].

Reaction of –N5C5O with Urethanes: Formation of Allophanates Similar to ureas, isocyanates react with urethanes to yield allophanates (Fig. 21.9b) [31, 35]. Again, the reaction is slow and requires elevated temperatures (120–150  C). Analogous to biurets for polyureas, formation of allophanates serves as a crosslinking mechanism for polyurethanes.

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

21.3.2 Reaction of Isocyanates with OxygenBased Nucleophiles Of particular relevance to aerogel synthesis, oxygen-based nucleophiles include water, alcohols, epoxides, carboxylic acids, mineral acids, and carboxylic anhydrides, which yield polyureas, polyurethanes, polyoxazolidones (fivemember cyclic urethanes), polyamides, oxide- or metaldoped polyureas, and polyimides, respectively.

Reaction of –N5C5O with Water: Synthesis of Symmetric Polyurea Aerogels Water acting as a nucleophile attacks the isocyanate carbonyl to yield a carbamic acid, which is unstable and decomposes into an amine and carbon dioxide (Fig. 21.11). The hydrolytic reaction leading to the carbamic acid is slow and in practice is usually catalyzed by non-nucleophilic tertiary amines, e.g., triethylamine (Et3N). The primary amine formed by decomposition of the carbamic acid reacts rapidly with a second unreacted isocyanate according to Fig. 21.8 to form a symmetric urea [37, 38]. Since the sequence of reactions in Fig. 21.11 and Fig. 21.8 bypasses the use of extraneous

O OCN

6

NCO

OCN

6N

N

H

HDI

O N H

6

NCO

6

OCN

Desmodur N3200 Fig. 21.10 Hexamethylene diisocyanate (HDI) and the primary component of Desmodur N3200 (a biuret derivative of HDI) manufactured by Covestro, LLC

515

amines for the synthesis of ureas, it represents a cost- and time-efficient alternative to the process described in Fig. 21.8 for synthesizing symmetric polyurea aerogels. By the same token, however, it is noted that the reaction sequence of Figs. 21.11 and 21.8 is not an atom-efficient process, as a carbon atom is lost as CO2; as such, this reaction pathway is not typically used for the synthesis of bulk polyureas. Nevertheless, a small amount of water is often deliberately added as a foaming agent in formulations used to prepare bulk polyurethane foams [39]. When it comes to the synthesis of polymeric aerogels, however, loss of CO2 is not a deterrent, since aerogels being low-density materials only require small amounts of monomers and the cost of the mass loss due to CO2 evolution is a small fraction of the overall cost. Although CO2 evolution during polyurea aerogel synthesis can potentially result in unwanted voids in the resulting monolithic parts, in practice this has not been reported as a problem.

Reaction of –N5C5O with Alcohols: Synthesis of Polyurethane Aerogels Addition of alcohols to isocyanates yields urethanes (Fig. 21.12) [38–40]. Polyurethanes are an extremely versatile class of polymers and a household term. The properties of polyurethanes can be tailored by varying the chemical identity of the polymeric backbone and in addition by using chain extenders and/or crosslinkers. Polyurethanes are employed extensively for rigid plastics, foams, elastomers, fibers, sealants, adhesives, and coatings [31, 33, 38, 39]. For steric reasons, the reactivity of alcohols toward – N¼C¼O decreases from primary to secondary (Table 21.1). Phenols are even less reactive due to resonance delocalization of the electron pair on oxygen to the aromatic ring. Owing to the relatively low reactivity of

H R

N

C

O

+

N

H2 O

–

+

R

H N

Et3N

O

OH

R

H

O

O Carbamic acid

H N

O H

R O Carbamic acid

Et3N –Et3

N+H

H N

O

R Et3N+H

–

–CO2 –Et3N

O

Fig. 21.11 In situ formation of amines from the reaction of isocyanates with water. The 1,3-proton transfer tautomerization of the first step is catalyzed by non-nucleophilic tertiary amines such as Et3N. In the

R

NH2

In-situ primary amine

second step, Et3N first undergoes an acid-base reaction with carbamic acid. The resulting carbamate expels CO2, while it takes back a proton from [Et3NH]+ to form an amine

21

516

N. Leventis

alcohols with isocyanates, formation of the urethane group is typically catalyzed with Lewis acids or bases. The reaction in Fig. 21.12 can also be carried out at higher temperatures without a catalyst. Polyurethanes Via Base Catalysis Activation of isocyanate via direct attack at the carbonyl carbon of isocyanate by tertiary amines used as catalysts was previously postulated but has led to several contradictions and is generally considered as not valid [41–45]. Rather, it is generally accepted that the reaction starts with nucleophilic attack of alcohol O to the carbonyl carbon (i.e., in the same fashion as water does in the first step of Fig. 21.11), wherein the role of the base catalyst (e.g., a tertiary amine) is to transfer the proton from the O to the N (1,3-proton transfer tautomerization just like in Fig. 21.11). However, if the alcohol is relatively acidic, e.g., a phenol derivative, a tertiary amine removes the proton from the phenolic –OH quantitatively, and the isocyanate is attacked by the resulting phenoxide (Fig. 21.13) [46].

Catalyst

R N C O + HO R'

R

H N

O

R'

O Urethane Fig. 21.12 Formation of urethanes from isocyanates and alcohols

O Ar

BH

H

Ar

Polyurethanes Via Acid Catalysis Many metallic compounds are effective Lewis acid catalysts for isocyanate-hydroxyl coupling. A list of various metalbased catalysts in a rough order of descending catalytic activity, including bases, is as follows: Bi, Pb, Sn, DABCO (1,4-diazabicyclo [2.2.2]octane), strong bases, Ti, Fe, Sb, U, Cd, Co, Th, Al, Hg Zn, Ni, trialkylamines, Ce, Mo, Va, Cu, Mn, Zr, and trialkyl phosphines [47]. Among these, lead 2-ethylhexoate, lead benzoate, lead oleate, sodium trichlorophenate, sodium propionate, lithium acetate, and potassium oleate can also be used as trimerization catalysts (see Sect. 21.3.3) [47]. Tin catalysts in particular, such as dibutyltin dioctanoate, dibutyltin dilaurate (DBTDL), stannous oleate, and stannous octoate, are many times more powerful for the isocyanate-hydroxyl reaction than tertiary amines but are not strong catalysts for the isocyanate-water foaming reaction that results in urea and CO2 (see Fig. 21.11) [47]. Mechanistically, the catalytic activity of metal salts depends on their ability to form complexes with both reacting groups, i.e., the alcohol and the isocyanate [48]. Synthesis of commercial polyurethane foams makes wide use of DBTDL, with which the rate of urethane formation in polar solvents such as DMF is proportional to the square root of the dibutyltin dilaurate concentration [49]. Two mechanisms have been proposed to fit the experimental data: the so-called insertion mechanism and the Lewis acid/base mechanism. A good summary of these two pathways can be found in reference [39]. Essentially, the difference between these two mechanisms is what

O

–

H

B+

O– R

N

C

R

O

N

R

C O

O

–

O R

N

O

–

C

H –B

O Ar

B+

Ar

R

–

O C O

Ar Ar

N

N O H Urethane

Fig. 21.13 Formation of urethanes from isocyanates and aromatic alcohols (ArOH) via base catalysis

Ar

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

517

a The insertion mechanism X

Bu 2 SnX2

ROH

Bu 2 Sn

X

R

R'NCO

Bu 2 Sn RO

OR

N

+XH

OR Tin alkoxide (the actual catalytic species)

X

R'

R'

Bu 2 Sn

Insertion

N

ROH

C

C

RO

O

O

O

R'

b The lewis acid/base mechanism

Bu

X R'NCO

Sn Bu

Bu

X

Bu

Bu O

C

R N O H Urethane

R

X

– Sn

21

Bu 2 Sn

XH

OR

X

Bu 2 Sn

X –XH

Bu 2 Sn

O

X

DBTDL Bu: butyl X: laurate

X

H

N

R'

ROH

+

X

X

– Sn

H

O

O

C +

Bu

X

N

R'

Urethane + Bu 2 SnX2

Fig. 21.14 DBTDL-catalyzed formation of urethane via the insertion mechanism (a) and via the Lewis acid/base mechanism (b) [39]

DBTDL interacts with first: the alcohol or the isocyanate. The distinguishing feature of the first mechanism is that it is initiated by alcoholysis of DBTDL to a tin alkoxide, which is the actual catalytic species (Fig. 21.14a). Subsequently, the isocyanate coordinates to Sn via N to form a fourmember intermediate reminiscent of those in Wittig reactions, which then gets inserted into the Sn-O bond of the tin alkoxide [50]. Next, urethane is replaced in the complex by a new alcohol molecule, and the sequence propagates. The Lewis acid/base mechanism is initiated by coordination of the isocyanate group to Sn via O (Fig. 21.14b). This increases the partial positive charge on the isocyanate carbon, thereby enhancing its electrophilicity toward the incoming alcohol.

Reaction of Isocyanates with Epoxides: Synthesis of Cyclic Urethanes The reaction of the isocyanate group with epoxides yields oxazolidones (Fig. 21.15), which can be classified as fivemember ring cyclic urethanes [51]. The reaction requires heating and is complicated by parallel trimerization of the isocyanate to isocyanurate (see Sect. 21.3.3) and by

homopolymerization of the epoxide. The trimerization reaction is favored with catalysts such as imidazole or tertiary amines. The resulting isocyanurate ring is fairly stable in the presence of epoxides. Aerogels via reaction of isocyanates with epoxides according to Fig. 21.15 have been described in recent patent applications filed by Henkel AG [52].

Reaction of Isocyanates with Acids Isocyanates react differently with carboxylic acids than with mineral acids. Reaction with the former has been used for the synthesis of polyamide aerogels, while reaction with the latter yields polyurea aerogels doped with metals or metal oxides; an exception is the reaction of isocyanates with boric acid where the product is practically the same polyurea as that obtained from the reaction of isocyanates with water. Reaction of –N5C5O with Carboxylic Acids: Synthesis of Polyamide Aerogels Typically, polyamides are prepared from activated carboxylic acids, e.g., via reaction of acid chlorides with amines [53]. They can also be prepared in high yield from the reaction of

518

N. Leventis

O R' O

C

N

R

N

C

O

R

+

N O

Diisocyanate

N

R'

O

O

Diepoxide

O Oxazolidone

Fig. 21.15 The reaction of isocyanates with epoxides to form oxazolidones

–O R

O

+ NR′

– CO2 R

O O

R′

O

O R

N H Amide

O C N R′ OH

O

O

NHR′

R O O Carbamic-anhydride adduct

R

NR′

[1,3] OH Shift

O

N

R

O

– CO2 R

O

2x

R′ OH

N-formyl amide

CO2 R′

H N

H N O Urea

O R′ +

R

O O

Anhydride

– CO2 R

R′ N H Amide

O 2 R

R′ N H Amide

Fig. 21.16 The three mechanistic pathways proposed for the reaction of isocyanates with carboxylic acids

carboxylic acids with isocyanates [54–57]. Kinetic studies have shown that high-polarity solvents increase the reaction rate and that aromatic isocyanates are more reactive than aliphatic isocyanates [58]. The reaction can take place without a catalyst even at room temperature (23  C); however temperatures in the range of 90  C to 135  C are also often used, especially for high-molecular-weight polymers. Catalysts include alkoxy metal salts, alkali metal lactamates, and mono alkali metal salts of dicarboxylic acids [59, 60]. Polymerization of dicarboxylic acids with aromatic diisocyanates using such Lewis acid catalysts at relatively low temperatures (135  C) [65]. All three pathways expel one molecule of CO2 per pair of monomers. Formation of ureas (Fig. 21.16, pathway c) has been observed in all cases of polyamide aerogels synthesized via the isocyanate route. This pathway can be minimized in certain situations but not eliminated altogether. Reflective of the fact that both the urea and the anhydride group become parts of the polymeric network, and therefore are not free to diffuse and react with one another (especially at the later

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

stages of the gelation process), it was observed that a significant amount of urea survives in the final aerogel when gelation is carried out at room temperature along with polyfunctional carboxylic acids having –COOH groups on adjacent carbon atoms, whereas stable intramolecular cyclic anhydrides can be formed. In these cases, formation of an anhydride from these carboxylic acids (Fig. 21.16, pathway c) can also lead to parallel formation of imides (see Sect. 21.3.2.5 below) that comprise a major component of the final aerogel. Reaction of –N5C5O with Mineral Acids: Alternative Synthesis of Polyurea Aerogels The reaction of isocyanates with mineral acids, summarized in Fig. 21.17, was first reported in 2016 [66] and stemmed from an effort to prepare boramic aerogels by reproducing a procedure reported in a 1962 patent [67], disclosing the presumed synthesis of boramides (i.e., materials with – B–NH– linkages) via reaction of isocyanates with boric acid in analogy to Fig. 21.16. The product was quite different though: what was obtained was pure polyurea aerogel. The proposed mechanism starts with formation of the carbamic boric-anhydride adduct, which can then react with another molecule of the same kind (Fig. 21.18, route 1) or with an isocyanate monomer (Fig. 21.18, route 2). The two routes converge to a common intermediate (Fig. 21.18, Int-1) that then reacts with another molecule of boric acid. The next intermediate (Int-2) rearranges itself into urea and boric oxide (B2O3, the anhydride of boric acid). Overall, the isocyanate/ mineral acid system behaved as if it proceeded along pathway c of Fig. 21.16 but stopped at the urea/anhydride stage. B2O3 can be removed quantitatively during post-gelation washes, leaving a pure polyurea aerogel [66]. This new pathway to polyurea aerogels is quite general: besides H3BO3, reaction of the same triisocyanate with H3PO4, H3PO3, H2SeO3, H6TeO6, H5IO6, and H3AuO3 always yielded the same polyurea aerogel; however, while the anhydride side product from the reaction with H3BO3 (B2O3) could be washed off easily from the porous structure of the wet gels, the other oxides were insoluble and remained as nano-dispersed dopants in the final polyurea aerogels and in the corresponding carbon aerogels derived from them by

NCO

OCN

Oxide Mineral or metal acid Polyurea + nano–1.5 CO2 particles NCO

Fig. 21.17 General gelation pathway from reaction of a triisocyanate (Desmodur RE – see Fig. 21.38) with mineral acids

519

pyrolysis. An exception was gelation with auric acid, where, due to the instability of the oxide, both the polyurea aerogel and the derived carbon aerogel were doped with nanodispersed metallic Au.

Reaction of Isocyanates with Anhydrides: Synthesis of Polyimide Aerogels Polyimide aerogels were first prepared [11, 68, 69] via the two-step DuPont route based on the reaction of dianhydrides

B

O C N R OH + 23 °C

O B

NHR

O

O

R N C O NHR O Carbamic-boric Route 2 anhydride adduct B

Route 1

B B

NHR O–

O B

O + NHR O

N

R

H N+ O R

O

O NHR B

OH

RN B

Int-1

O O

O

HO

B

R –O

HN H B O O N O + B R

Int-2

O

O R

N N H H Urea

R

+

O B

O B

B

O

B

+ CO2

Anhydride Fig. 21.18 Proposed mechanism for the formation of urea from isocyanate and H3BO3 (Note, boric acid is used here as a proxy for any of several possible mineral acids)

21

520 Fig. 21.19 The pathway for the reaction of isocyanates with anhydrides to imides, emphasizing the initial nucleophilic attack and the formation of the seven-member ring intermediate

N. Leventis

O

O

O + R' N C O

O C N R' IH

_

+

O

O

O

O

O Imide

with diamines [70, 71]. Subsequently, polyimide aerogels were also prepared using the method referred to as polymerization of monomeric reactants (PMRs), in which a norbornene end-capped imide monomer was polymerized and gelled via ring-opening metathesis polymerization [72] (ROMP, see ▶ Chap. 23). Finally, polyimide aerogels have been also prepared via a less known reaction of isocyanates with anhydrides (Fig. 21.19), both intentionally [73–75] and unintentionally as copolymers together with polyurea and polyamides (see section “Reaction of –N¼C¼O with Carboxylic Acids: Synthesis of Polyamide Aerogels”) [76]. Arguably, the first report of the reaction of an isocyanate (ethyl isocyanate) with an anhydride (acetic anhydride) toward an imide (N-ethyldiacetimide) dates back in 1854 [77]. In the late 1950s, systematic studies showed that imides can be prepared from isocyanates and acyclic anhydrides in fair yields (71%) [78, 79]. The polymerization of aromatic isocyanates such as 4,4′-methylenediphenyl diisocyanate (MDI) with cyclic aromatic anhydrides such as pyromellitic dianhydride (PMDA) was first reported by Meyers in the late 1960s [80]. The reaction was carried out in dimethylformamide (DMF) by stepwise heating of the reaction mixture from 40  C to 130  C. The final product was obtained in good yields (78% w/w). The rate of formation of the imide is increased significantly by the presence of water, which has been attributed to the hydrolysis of the isocyanate either to urea or amine [81]. Meyers was also the first to propose that the reaction proceeds through a sevenmember ring intermediate (Fig. 21.19). In 2010, Leventis et al. reported similar solid-state 13C NMR spectra of polyimide aerogels synthesized via the isocyanate route from PMDA and 4,4′-methylenediphenyl diisocyanate and via the DuPont route from PMDA and 4,4′-methylenedianiline, confirming that the two routes yielded the same product [73]. Although it can be argued that the isocyanate route to polyimides is not as atom-efficient as the DuPont route, since the C atom of the isocyanate is lost as CO2 upon decomposition of the seven-member ring intermediate, it

–CO2

O + O C _ N R' O O

O N R'

O

O N

R' O 7-Member ring intermediate

+

O C _ N O R'

O

does provide several other advantages over the DuPont route for the synthesis of aerogels: (a) it is an one-step, low-temperature process; (b) it does not require sacrificial dehydrating agents (such as acetic anhydride) in order to achieve gelation; (c) CO2 is the only by-product; and (d) it bypasses the formation of a polyamic acid intermediate, the formation of which leads to highly viscous solutions (especially at higher concentrations) that limit the maximum achievable aerogel density and complicates homogeneity.

21.3.3 Condensation of Isocyanates: Polyisocyanurate and PIR-PUR Aerogels Isocyanates can react with other isocyanates and, depending on the reaction conditions (temperature, acid vs. base catalysis), can yield an open dimer (a carbodiimide), a four-member ring dimer (a uretdione), a cyclic trimer (an isocyanurate), or a polymer (a polyisocyanate) [51]. Figure 21.20 summarizes these possibilities. Formation of uretdiones requires heating and is catalyzed by bases (phosphines, substituted pyridines, and trialkylamines) [82, 83]. At higher temperatures, uretdiones react with amines to form biurets and with alcohols to form allophanates [51]. Formation of carbodiimides is induced by heating isocyanates to 180–200  C [84] and is accompanied by loss of CO2. Catalysts include transition metals, phosphine oxides, amides, urea, and other isocyanate derivatives. Polyisocyanates are prepared under anionic polymerization conditions, and the reaction can be carried out between 40  C and 100  C. The reaction is reversible, yielding back the monomer and cyclic trimers [51, 85]. Formation of isocyanurates is catalyzed by alkali metal alkoxides as well as by tertiary amines and certain tin compounds. However, the reaction is slower than formation of, for example, urethanes. As a result, if formation of a urethane

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

521

Polymerization Polyisocyanates

21 O

Trimerization R

N

C

O

R O

R

N

N N

O

Isocyanurates O

R Cyclization

R

RNH 2 N

N

R

ROH

Dimerization – CO 2

O Uretdiones R

N

C

N

Biurets Allophanates

R

Carbodiimides Fig. 21.20 Possible pathways for the self-condensation of isocyanates

employs an excess of isocyanate, the end product consists of rigid isocyanurate rings linked by more flexible polyurethane tethers. These polymeric materials are referred to as poly(isocyanurate-urethanes) (PIR-PUR). Formation of PIR-PURs is self-catalyzed, as the urethane also serves as a catalyst for the trimerization of dangling isocyanate groups into isocyanurate [86]. This strategy for isocyanurate formation in situ has been also extended to polyurea aerogels via reaction of polyisocyanates with sub-stoichiometric amounts of aromatic diamines and has been combined with the reaction of excess isocyanate groups with water (see section “Reaction of –N¼C¼O with Water: Synthesis of Symmetric Polyurea Aerogels” above) for the formation of copolymers with symmetric polyureas [87]. The isocyanurate ring is thermally very stable, with decomposition temperatures that can reach over 400  C [88, 89]. Therefore, isocyanurates play an important role in fire-resistant foams, coatings, and adhesives [89–92]. A reason that in use are PIR-PURs rather than PIR-only foams is because the latter are friable (easily crumbled), whereas PIR-PURs, owing to their flexible polyurethane bridging between isocyanurate nodes, are mechanically resilient and in many cases elastomeric [93, 94]. This design principle has recently been employed in the production of superelastic shape-memory aerogels [95–97]. The isocyanurate ring enters synthesis of aerogels very frequently, yet indirectly: Several small-molecule diisocyanates are commercially available as highermolecular-weight nonvolatile isocyanurate trimers, which serve the dual purpose of decreasing toxicity due to the low vapor pressure of such trimmers, as well as providing a ready-made isocyanurate ring for formulating foams, coatings, and adhesives [98]. Figure 21.21 shows three such isocyanurate-based trimers and their common commercial

names. Note that most isocyanate-derived aerogels reported in the literature to date have been based on these monomers.

21.4

Polyurea Aerogels

As reviewed in the previous section, polyurea aerogels can be prepared via the three routes summarized in Fig. 21.22. Polyurea aerogels can be produced in various form factors ranging from monoliths to particles, in most cases dried from SCF CO2. Polyurea aerogels have also been produced via freeze drying by replacing the pore-filling solvent with tert-butanol followed by freezing and subsequent sublimation [99–101]. The freeze drying process provides easier access to large monolithic panels than supercritical drying; however, the process parameters seem to have significant influence on the properties of the resulting polyurea aerogels.

21.4.1 Polyurea Aerogels from Isocyanates and Amines Polyurea aerogels have been prepared via reaction of isocyanates with amine-bearing monomers. Properties of isocyanate-derived aerogels were first reported by Biesmans et al. in the late 1990s, which mentioned polyureas but did not include chemical evidence of such [10]. With an eye toward thermal insulation, the first comprehensive study in this subclass of polyurea aerogels was published in 2009 by Lee et al. at Aspen Aerogels [102], using two types of isocyanates (4,4′-methylenediphenyl diisocyanate (MDI) and polymeric methane diphenyl diisocyanate (pMDI)) and two types of commercial polyoxypropylene triamines, Jeffamine T3000 and T5000, where the numerical extension refers to

522

N. Leventis

NCO

Me

NCO

OCN 6

O

N

O

O

O

O

N

O OCN

N

NCO OCN

N 6

N

NCO 6

N

OCN

N

Me

O

Desmodur N3300A

O

O

Me

NCO

Desmodur RC

Fig. 21.21 Commercially available triisocyanates (from Covestro LLC) based on the isocyanurate ring. Desmodur N3300A: the trimer of hexamethylene diisocyanate (HDI, see Fig. 21.10), supplied as a pure

Fig. 21.22 The three routes to polyurea aerogels via reaction of diisocyanates with (a) diamines, (b) water, and (c) mineral acids, exemplified here with boric acid

N

N

Desmodur Z4470BA

compound. Desmodur RC: the trimer of toluene diisocyanate (TDI), supplied as a solution in ethyl acetate. Desmodur Z4470BA: the trimer of isophorone diisocyanate, supplied as a solution in butyl acetate

O

a

OCN

R

NCO + H2N

Diisocyanate

OCN

NCO

–CO2

Diisocyanate

O R

NH2

N H

Diamine O H 2O

R

b

R′

c

OCN

NCO

R′

N H

n

Polyurea O N H

R

N H

N H

R

N H

n

Symmetric polyurea O

R

N H

N H

+ B(OH)3

Diisocyanate

the approximate molecular weight of the monomer [103]. Gelation was carried out in acetone using triethylamine as the catalyst. The resulting polyurea aerogels exhibited low shrinkage factors (1.14  f  2.95, where f ¼ final density/ target density), bulk densities in the range of 0.098–0.116 g cm3, high porosities (90–91% v/v), and good hydrophobicity. It was reported that polyurea monoliths based on Jeffamine-T3000 showed smaller shrinkage and lower thermal conductivity values, which was attributed to higher crosslinked structures. Polyureas prepared from pMDI were more flexible and less dusty than those prepared from MDI. Under electron microscopy, these polyurea aerogels looked similar to silica aerogels, consisting of ~50 nm clusters of nanosized particles; however, high-density polyurea aerogels exhibited larger pores than those of similar-density silica aerogels (Fig. 21.23).

O N H

R

N H

N H

R

N H

n

+ B2O3

Symmetric polyurea

Thermal conductivity as a function of bulk (final) density (Fig. 21.24a) showed a classic U-shaped curve typical of aerogels, reflecting a balance of solid conductivity, aerogel pore size, and morphology. The thermal conductivity minimum (13 mW m1 K1) was located at around 0.20 g cm3, which is in the range of other polymeric aerogels and was comparable to the thermal conductivity of resorcinolformaldehyde aerogel monoliths (12 mW m1 K1) and opacified silica aerogels (13 mW m1 K1) that have been reported previously. The thermal conductivity of various polyurea-based aerogels in this study was compared over a wide pressure range at two different target densities (0.07 and 0.1 g cm3) along with analogous polyurethane-based aerogels prepared from MDI and a 3400-molecular-weight polyether polyol modified with ethylene oxide (Multranol 9185, Supplied by Covestro LLC) using acetone as the

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

a

523

c

b

21

Fig. 21.23 SEM images of (a) silica aerogel at 0.0902 g cm3 and (b), (c) polyurea aerogels derived from the reaction of isocyanate with amine, synthesized with constant EW ratio (¼ equivalent weight of amine/equivalent weight of NCO). The polyurea aerogel in (B) was

b

23 21 19 17 15 13 11 0.06 0.09 0.12 0.15 0.18 0.21 0.24 0.27 0.30 Final density (g/cm3)

Thermal conductivity (mW/m K)

Thermal conductivity (mW/m K)

a

prepared from pMDI and Jeffamine-T3000 (aerogel bulk density ¼ 0.06 g cm3), whereas the aerogel in (c) was synthesized from pMDI and Jeffamine-T5000 (bulk density ¼ 0.1 g cm3). (Scale bars are 200 nm.). (Adapted from [102], Copyright 2009 Springer Nature)

35 Polyurea 1-LD Polyurea 1-HD Polyurea 2-LD Polyurea 2-HD Polyurethan 1-LD Polyurethan 1-HD

30 25 20 15 10 5 0 10–2

10–1

100 101 102 103 Pressure (torr (=mmHg))

104

Fig. 21.24 (a) Thermal conductivity versus bulk (final) density of various polyurea aerogels prepared with pMDI at constant EW ratio (¼ equivalent weight of amine/equivalent weight of NCO) and catalyst (triethylamine) content at 5% w/w. (b) Thermal conductivity of polyurea

and analogous polyurethane aerogels prepared with target densities of 0.07 g cm3 (LD) and 0.1 g cm3 (HD) as a function of pressure. (Adapted from [102], Copyright 2009 Springer Nature)

solvent [104]. Figure 21.24b shows that, irrespective of density and pressure, the polyurea aerogels had lower thermal conductivities than the polyurethane aerogels. More recently, this type of polyurea aerogel has been expanded to include polyurea/polyurethane copolymers and isocyanurate-crosslinked polyisocyanate aerogels (the latter being synthesized using trimerization catalysts) [105]. All three types of amine-derived polyurea aerogels were prepared in molds containing nonwoven fibrous battings to produce fiber-reinforced polyurea aerogel composites. The resulting materials were evaluated in terms of their thermal conductivity, flexibility, durability, and dustiness for applications in diver gloves and extravehicular activity spacesuit assemblies.

Among the three types of aerogels, polyurea aerogels showed the lowest thermal conductivities but were also the stiffest of the three: both thermal conductivity and flexural modulus decreased with increasing target density over the range of 0.04 to 0.08 g cm3. Increasing the target density beyond this range resulted in decreased flexibility to undesirable levels. A lower isocyanate-to-amine ratio led to improved thermal performance and reduced deformation under impact. Optimization of other formulation variables, including isocyanate functionality, resulted in relatively flexible polyurea aerogels with thermal conductivities of 18 mW m1 K1, which remained unchanged after one laundering/drying cycle. The analogous polyurea/polyurethane aerogels (crosslinked with

524

N. Leventis

polymeric backbones more hydrophobic than those based on ODA. In both cases, the NCO-capped oligomers were crosslinked with 1,3,5-tris(4-aminophenoxy)benzene (TAB) according to Fig. 21.25. The materials in this study were formulated via the molar ratio of the diamine to MDI, which controls the crosslinking density of the polymer for a constant oligomer-to-TAB ratio at 3:2. The average bulk density of the resulting aerogels was in the range of 0.19–0.26 g cm3 and exhibited porosities of 79–86% v/v and surface areas between 106 and 309 m2 g1. Microscopically, the ODA-based polyurea aerogels produced in this study consisted of string-ofbeads-like, low-aspect-ratio interconnected fibrils with diameters in the range of 75–125 nm. In contrast, the morphology of DMBZ-based polyurea aerogels varied from fibrilar, with fibril diameters in the range of 10–75 nm, to clusters of particles. Spectroscopic evidence suggested that

both polyamine and polyol) were ten times less dusty than their all polyurea counterparts and remained flexible, exhibiting flexural moduli of less than 0.5 MPa at 130  C at a bulk density of 0.08 g cm3. On the downside, the ambient pressure thermal conductivity of these materials was never lower than 26 mW m1 K1. In another study designed to explore the effect of segmental hydrogen (H–) bonding on the properties of polyurea aerogels, gels were prepared in NMP from MDI and one of two different diamines (4,4′-oxydianiline (ODA) and 2,2′-dimethylbenzidine (DMBZ)) [106]. The choice of these two amines was based on the fact that ODA is a less bulky monomer that renders polymeric backbones more flexible thanks to its central oxygen atom and that flexibility allows for greater segmental motion within the polymer and thereby a higher possibility for organization via H bonding. On the other hand, an advantage of using DMBZ is that it renders

a R: + OCN

H2N

NCO MDI

R

DMBZ

NH2

O

Diamine

ODA

Dissolve in anhydrous NMP, then combine O OCN

N H

O N RN H H

NCO

N H

n Polyurea oligomer b

Stir for 5 min, then add TAB crosslinker

H2N

NH2

O

NH2

O

O O

O

O

TAB O

O N H

O N H

N H

O N H

R

N H

O

O N H

n Network polyurea

N H

O

N H

Fig. 21.25 Synthesis of crosslinked polyurea aerogels, showing the polyurea oligomer (a), the crosslinker (TAB), and the network structure (b) [106]

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

shrinkage was linked to hydrogen bonding between urea groups and was observed to decrease as hydrogen bonding increased in linear oligomers of increasing molecular weight. Interestingly, the extent of H-bonding in ODA-based aerogels was higher than in DMBZ-based aerogels, which had the effect of stabilizing the aerogel network and reducing shrinkage compared to DMBZbased aerogels (14–21% for ODA-based aerogels versus 21–23% for DMBZ-based aerogels). Accordingly, the density of the DMBZ-based aerogels did not vary with increasing crosslink density. The DMBZ-based aerogels were reported as being somewhat translucent, exhibiting a mean pore size around 15 nm, an onset of thermal decomposition at 250  C, and a compressive modulus in the range of 38.5–69.4 MPa, compared to 11.7–48.2 MPa for the corresponding ODA-based aerogels. In an extension of this study, polyurea-co-polyurethane block copolymers were developed with an eye toward improving flexibility [107]. Materials were prepared through a modified version of the procedure shown in Fig. 21.25 as follows: MDI was first reacted with two aliphatic diols (polycaprolactone diol and polytetramethylene glycol, with molecular weights of 530 and 650 g mol1, respectively) to form isocyanate end-capped oligomers, which in turn were reacted with ODA, following, after that point, the exact procedure as described in Fig. 21.25, starting from the top. The resulting aerogel densities varied from 0.20 to 0.35 g cm3 showing porosities ranging from 71 to 85% v/v and surface area values between 47 and 163 m2 g1. Shrinkage was higher for these materials (19–33%) than for the corresponding all polyurea aerogels described above (14–21%) [107]. It was reported that shrinkage data correlated with the extent of H-bonding between urea and urethane groups, leading to the conclusion that the weight fraction of the aliphatic diol can be used as a parameter to control shrinkage. The extent of H-bonding could be also influenced by the degree of crosslinking, which in turn was affected by the amount of TAB used: reducing the amount of TAB brought about a reduction of urea H-bonding and an increase of urethane H-bonding. The materials were thermally stable up to 300  C and exhibited a compressive modulus between 12 and 52 MPa. Microscopically, all formulations

525

comprised networks of branching and interconnected small-aspect-ratio fibrils, which in turn appeared to consist of assemblies of particles. Samples with a bulk density of 0.22 g cm3 exhibited a porosity of 82% v/v, a surface area of 71 m2 g1, and a compressive modulus of 10.6  2.0 MPa and were flexible as small monoliths. A more recent study by Wu et al. targeted improving the thermal conductivity of polyurea aerogels by manipulating the solid framework through controlling gelation time, which in turn was controlled via the gelation temperature [108]. Polyurea aerogels in this study were prepared from polymethylene polyphenyl isocyanate (PPI) and polymer MDA (poly-MDA) (Fig. 21.26). The total monomer concentration varied from 5% w/w to 20% w/w in DMF (sols with a monomer concentration of 3% w/w did not gel). Gelation was carried out at ambient temperature, 0  C, and  20  C. All aerogels exhibited a fibrous morphology consisting of entangled worm-like nanostructures. Gelation times increased by 60–80% by decreasing temperature from ambient to 0  C and by another 50% by decreasing temperature to 20  C. Interestingly, shrinkage decreased with decreasing gelation temperature, and for every sol concentration, the bulk density decreased correspondingly. (All bulk densities were in the range of 0.2 and 0.08 g cm3.) The thermal conductivity as a function of density showed a typical U-shaped curve for samples synthesized at all three temperatures, the minimum of which was 0.0192  0.0002 W m1 K1, displayed by samples with a bulk density of 0.096  0.002 g cm3 prepared at 20  C using 10% w/w concentration of monomers. The particle size of those 10% w/w samples decreased from about 8 nm to 4 nm to 2 nm when gelled at ambient temperature, 0  C, and  20  C, respectively. The particle size/thermal conductivity data of these materials supports that the latter is controlled by the interparticle contacts along the skeletal framework (more on this in Sect. 21.5. below). Finally, in an interesting variation of polyurea aerogel synthesis from isocyanates and amines, a triisocyanate (Desmodur N3300A, see Fig. 21.21) and a diamine (ethylene diamine) were first physically separated in two immiscible liquids; the former dissolved in propylene carbonate, and the latter was suspended in mineral oil. The isocyanate solution was then added dropwise (via

Fig. 21.26 The isocyanate and amine used in the preparation of polyurea aerogels in Wu et al. [108]

H 2N

NH2

n

Polymer-MDA

NH2

OCN

NCO n

Polyisocyanate

NCO

21

526

N. Leventis

dripping) into the diamine suspension, resulting in the formation of spherical beads on contact with the mineral oil suspension. As ethylene diamine diffused inside the incoming droplets, the entire volume of the droplets was converted into a polyurea wet gel. Drying the resulting wet gel beads with SCF CO2 yielded polyurea aerogel beads with a mean diameter of 2.7 mm and a narrow particle size distribution (full width at half maximum ¼ 0.4 mm) (Fig. 21.27a) [109]. The bulk density of the beads was 0.166  0.001 g cm3 with a porosity of 87% v/v and a BET surface area of 197 m2 g1 all within the range of the monolithic isocyanate/amine-derived polyurea aerogels discussed so far. Microscopically, the internal texture of the beads was fibrilar in nature accompanied by a so-called skin effect [110, 111] in which the external surface of the beads was found to be denser than their interior and the fibrous morphology was almost completely lost (Fig. 21.27b).

21.4.2 Polyurea Aerogels from Isocyanates and Water Polyurea aerogels have been prepared via reaction of isocyanates with water, thus eliminating the need to add a separate amine-bearing monomer. In this process, water reacts with isocyanate to form primary amine groups in situ that in turn react with unreacted isocyanates to form polyurea. A schematic of this approach is shown in Fig. 21.22 reaction b (see also Fig. 21.11). Polyurea aerogels produced this way comprise a large segment of polymer-based aerogels and by conceptual extrapolation of composite organic-inorganic aerogels as well, the latter via the relevance of reaction b of Fig. 21.22 to polymer crosslinking of inorganic, polymer, and biopolymer skeletal networks (see ▶ Chap. 29). Mechanically strong polyurea aerogel panel based on this approach is commercially available under the trade name Airloy ® X103 [112].

20 a

Normal Mean = 2.68 Std. Dev. = 0.164 N = 100

6

Frequency

15 5

10

2 5

0

2.10 2.30 2.50 2.70 2.90 3.10 Diameter (mm)

b

Skin

20 mm

2 mm

Fig. 21.27 (a) Photographs and bead size distribution curve for spherical polyurea aerogel beads prepared via the dripping method. (b) SEM images of spherical polyurea aerogel beads showing the denser outer skin of the beads (left) and their porous fibrilar interior (right) [109]

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

The first encounter with this class of aerogels was somewhat accidental and occurred during efforts to crosslink resorcinol-formaldehyde (RF) aerogels in a fashion analogous to polymer crosslinked oxide aerogels (e.g., silica). In these experiments, RF wet gels were solvent-exchanged using baths containing the triisocyanate Desmodur N3300A (see Fig. 21.21) and triethylamine (Et3N) as catalyst in an effort to form a conformal polyurea coating over the RF network via reaction of dangling surface –OH groups with the isocyanate. Surprisingly, the crosslinking baths themselves gelled, the cause of which was traced to a small amount of water present in the solvent [19]. Systematic studies of materials based on this chemistry followed shortly thereafter [20, 113], resulting in the development of rationally designed polyurea aerogels derived from the reaction of isocyanates with water. Such Desmodur N3300A/water-derived aerogels are closely related chemically to the polyurea aerogel beads

527

discussed in the previous section (Fig. 21.27), with the difference being that they lack three –CH2– groups per polymer repeat unit and include only 1.5 C¼O groups per polymer repeat unit instead of three. These small structural differences are enough to leave a clear signature in the solid-state 13C NMR spectra (Fig. 21.28) [109], exemplifying the fact that solid-state NMR techniques are instrumental in the chemical characterization of polymeric aerogels (see also ▶ Chaps. 20, ▶ 23, and ▶ 29). In the synthesis of Desmodur N3300A/water-derived polyurea aerogels, an approximately stoichiometric amount of water was used (relative to isocyanate) [20]. Large excess of water led to precipitation rather than gelation. Varying the concentration of the isocyanate yielded aerogels with densities ranging from 0.016 g cm3 to 0.55 g cm3 (see the photograph in Fig. 21.28a) with porosities ranging from 54 to 99% v/v and specific surface areas between 54 and 244 m2 g1 [20, 113]. The variation of surface area with

b′-e′

a i′ e′ 16

34 72 126 190 550

Desmodur N3300A

h′

f′

c′

Et3N

N

O

a′ b′

i′ O N

O

H2O

d′

N h′ H

a′,f ′

N O

1.00 : 1.87 e

b

f

h i

Desmodur N3300A

en

d

c b a

N i O

O N

O

g

2.27 : 5.33 b-e a,f,g

N h N H H

N O

2.97 : 3.87

1.00 : 0.92 150

100

50

0

ẟ (ppm) Fig. 21.28 Comparison of the solid-state 13C NMR spectra of polyurea aerogels derived from the reaction of the triisocyanate Desmodur N3300A (see Fig. 21.21) with water (a spectrum) and with ethylene diamine (en; b spectrum). Relative integrated peak intensities are given

underneath each resonance. Inset: photograph of polyurea aerogel monoliths synthesized from Desmodur N3300A and water spanning the density range of 0.016 g cm3 to 0.550 g cm3 [109]

21

528

N. Leventis

Stress (MPa)

45 30 15

Stress (MPa)

60 2 1 0 0.4 0.0 0.2 0.6 Engineering strain (mm/mm)

0 0.0

0.2 0.4 0.6 0.8 Engineering strain (mm/mm)

1.0

@ RT @ –173 °C Load-unload Fig. 21.29 Quasistatic compression testing (strain rate at 0.05 s1) of polyurea aerogel monoliths with bulk densities at 0.13 g cm3 derived from the reaction of the triisocyanate Desmodur N3300A with water in acetone, under various conditions as color coded. Inset: Magnification of the elastic region and of the beginning of the plastic deformation region. Other pertinent data for the sample: (Density: 0.13 g cm3; Porosity: 90 % v/v; Surface area: 169 m2 g1; Morphology: fibrous; Elastic modulus: 20  3 MPa; Poisson’s ratio: 0.23; Energy absorption: 21  9 J g1). (Adapted from [20], Copyright 2010 The American Chemical Society)

density strongly suggests morphological changes over this density range, as discussed below. Polyurea aerogels produced through the water route were found to be mechanically strong materials. Figure 21.29 shows a typical compressive stress-strain curve at room and cryogenic temperatures [20]. In similar fashion to polyureacrosslinked silica and vanadia aerogels (▶ Chap. 29), at room temperature, Desmodur N3300A/water-derived polyurea aerogels were linearly elastic at small compressive strains (0.2 g cm3) burned only in direct contact with the flame, selfextinguishing once the flame was removed (Fig. 21.31b); and (c) density gradient samples ignited from their lowerdensity end propagated the flame and burned only until the flame reached the higher-density side, where it selfextinguished (Fig. 21.31c) [116]. It was speculated that the origin of this density-dependent flame retardancy was related to the nanomorphology dependence of the air circulation through the porous structure.

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

a

Wire O-ring

Pump 2

Pump 1 Low concentration sol High concentration sol

Density-gradient sol Mold

Magnetic stirrer

b

0.05 0.03

0.10 0.15

0.02

0.20 0.01 0.25 0.00

Fig. 21.31 Density-dependent flame propagation of polyurea aerogel monoliths derived from Desmodur N3300A and water. Upon ignition (middle frames), (a) samples with bulk densities 0.2 g cm3 selfextinguished shortly after removal of the flame. (c) Density gradient aerogels (Fig. 21.30) burned only until the flame reached the highdensity region. (Adapted from [116], Copyright 2011 Materials Research Society)

529

0

0.13 g cm–3

0.30 10 20 30 40 50 60 Distance (mm)

Ignition

Density (g cm3)

Fig. 21.30 (a) Preparation of density gradient polyurea wet gel monoliths [116]. (b) Left: magnetic resonance imaging (MRI) of a density gradient polyurea wet gel after it was solvent-exchanged with water. (The high-density end of the monolith was at the bottom.) Middle: density variation by analysis of the MRI images and by direct measurement (red line, by cutting out and weighing discshaped coupons along the length of the monolith). Right: SEM images showing the fibrilar morphology at the low-density (ρb) end and the particulate morphology at the high-density end of the monolith. (Reprinted from [116], Copyright 2011 Materials Research Society)

Mean intensity

21

High r b end

Low r b end

Time lapsed: 2´14’’

a

0.19 g cm–3

Time lapsed: 1´6’’

Density gradient

Time lapsed: 1´30’’

b

c

Subsequently, the simple synthetic protocol of polyurea aerogels via the water route, together with their good mechanical properties, flame retardancy, and intriguing nanomorphology, became the point of departure for numerous other studies on their processing, structure-property relationships, and applications.

Most commonly, polyurea aerogels come as monoliths. Higher-density monolithic samples (>0.1 g cm3) can be machined to desirable shapes with regular tools (see ▶ Chap. 6). Noncontact slicing of lower-density polyurea aerogel samples has been performed at ambient conditions using an 800-nm Ti:sapphire femtosecond laser [117]. The

21

530

N. Leventis

ablation rate was investigated at different energy levels and found to be on the order of tens of microns per 40-fs pulse. Periodic grooves noted on the newly exposed cut surfaces were attributed to a material removal mechanism that involves melting and vaporization. The fact that the morphology of this class of polyurea aerogels changes with density became only more intriguing when it was observed that the nanostructure can also be varied simply by changing the gelation solvent holding all other parameters constant (see Fig. 21.1 for examples). Specifically, it was noted that Desmodur N3300A/waterderived polyurea aerogels synthesized in acetonitrile could take on a cocoon-like nanostructure (Fig. 21.32a), wherein a solid core (Fig. 21.32b) is trapped inside a fibrous web [21]. These acetonitrile-synthesized polyurea aerogels were flexible (Fig. 21.32c) and superhydrophobic (water contact angle of 150 ) and demonstrated the rose petal effect (Fig. 21.32d). The diverse morphologies accessible by Desmodur N3300A/water-based polyurea aerogels enable study of structure-property relationships in nanostructured matter while holding chemical composition constant. In this context, the Reichenauer group at ZAE Bayern investigated the

a

b

5 mm

10 mm c

PUA-ACN-109

q =150°

d H2O

Fig. 21.32 (a) SEM of a Desmodur N3300A/water-derived polyurea aerogel (bulk density ¼ 0.172 g cm3) prepared in acetonitrile. (b) SEM of a similar sample as in (a) after cross-sectioning with an Ar-ion beam, revealing that the aerogel’s constituent particles were dense with no internal structure. (c) Low-density (0.073 g cm3) polyurea aerogel monoliths synthesized in acetonitrile were flexible. (d) Contact angle of a water droplet on the acetonitrile-synthesized polyurea aerogel monolith from frames (a) and (b). The water droplet remained attached to the surface even when the sample was flipped upside down (the rose petal effect). (Adapted from [21], Copyright 2014 American Chemical Society)

thermal conductivity of Desmodur N3300A/water-derived polyurea aerogels as a function of nanomorphology [118]. Thermal conductivity values between 0.027 and 0.066 W m1 K1 were obtained using the hot-wire method for samples of densities between 0.04 and 0.53 g cm3. The total thermal conductivity was deconvolved into gaseous, radiative, and solid-conductive transport contributions as a function of pressure and temperature. The solid thermal conductivity scaled with density with an exceptionally low exponential factor α ¼ 1 (compared to the typical exponent for aerogels of α ¼ 1.5 [119, 120]). It was concluded that unlike macroporous foams, which exhibit linearly increasing solidphase thermal conductivity with increasing density because of their regular pore structure [121], the transition of microstructure in N3300A/water-derived polyurea aerogels from fibrilar to particulate with increasing density counteracts this trend. Specifically, the particulate nanomorphology in higherdensity samples results in numerous interparticle contact points, which serve as effective heat resistors in a network model of thermal conduction. These added contact points work to partially counteract the increase in solid-phase thermal conductivity that would otherwise result from increased density. Thus, the reason why the exponent α ¼ 1 for N3300A/water-derived polyurea aerogels is not related to the reason why α ¼ 1 for foams. Subsequently, the same group investigated the relationship between thermal conductivity and compressive stiffness as a function of density, following two different approaches. In the first approach [122], polyurea aerogels with an initial density of 0.027 g cm3 were compressed uniaxially to various strains up to a maximum density of 0.400 g cm3. The properties of interest in this study included microstructure, thermal conductivity, and elastic modulus. The important finding in this study was that thermal pathways along the skeletal backbone of the aerogels can be decoupled from the mechanism that supports external compressive loads. (Note that similar conclusions had been reported before with silica aerogels [123].) Thus, at low densities, the stiffness of the compressed polyurea aerogels and their solid thermal conductivity followed a quadratic relationship with one another, which was attributed to the fact that microscopic deformations were dominated by bending, which does not alter the length of thermal conduits in the material. At higher densities (achieved by higher compressive strains in the context of these experiments), the relationship between stiffness and solid thermal conductivity transitioned from quadratic to linear, which was attributed to network compaction. The key parameters that allowed for a microstructural decoupling of mechanical and thermal properties were identified as the microscopic homogeneity of the solid network and the curvature of its constituent network elements. In the second approach [124], similar studies were conducted with two series of samples prepared in the density range of 0.03 to 0.3 g cm3. Again, the elastic modulus was found to

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

correlate with solid thermal conductivity, irrespective of microstructural details. It was speculated that possibilities for structural decoupling of the elastic modulus and heat transport through the solid phase, as suggested by these studies, could open new approaches to applications, presumably in the spirit of reference [123]. The mechanical properties of N3300A/water-derived polyurea aerogels have also been investigated from the perspective of acoustic attenuation and shockwave absorption. Lu et al. have reported sound transmission loss values of over 30 dB cm1 over the range of 1 to 4 kHz for such polyurea aerogels with a bulk density of 0.25 g cm3 at thickness of 5 mm, which is considered extremely high for a material (see Fig. 21.33 for a comparison with other relevant materials) [125, 126]. Inspired by the nanostructure of Fig. 21.32a, and noting striking similarities with acoustic metamaterials, polyurea aerogels were modeled first with a one-dimensional multidegree-of-freedom mass-spring system. Wave transmission loss results were produced for different configurations modeled to reflect various polyurea aerogel nano- and microstructures. Significant wave attenuation was observed with a random spring distribution. Based on these results, the authors introduced polyurea aerogels of different bulk densities and porosities into laminated composites in the form of thin sheets, for example, placed between two gypsum wallboards of the type typically used in soundproofing applications [127]. The sound transmission loss of the sandwich structure reached 40 dB cm1 of aerogel at 2 kHz after the implementation of only two 5-mm thick aerogel layers of bulk densities of 0.15 and 0.25 g cm3, respectively. In parallel, an exact analytical time-harmonic plane-strain

Sound transmission loss (dB/cm)

40 35 30 25 20 15 10

PUA (0.45 g/cm3) PUA (0.25 g/cm3) PUA (0.11 g/cm3) Spaceloft® blanket Acoustic foam

5 0 500 1000 1500 2000 2500 3000 3500 4000 Frequency (Hz)

Fig. 21.33 Experimental sound transmission loss data of Desmodur N3300A/water-derived polyurea (PUA) aerogels at the densities indicated in the legend compared with two other relevant commercial porous materials (Spaceloft ® Blanket and Acoustic Foam). (Reprinted from [125], Copyright 2017 Elsevier B. V)

531

solution for the diffused wave propagation through the multilayered structure was developed using the theories of linear elasticity and Biot’s dynamic poroelasticity. The theoretical results were well-supported by experiments, and the authors suggested that such materials could be utilized for the design of future lightweight multifunctional composite structures [127]. In a similar study incorporating both experiment and theory, commercially available Desmodur N3300A/waterbased polyurea aerogels (Aerogel Technologies Airloy X103 [112]) have been considered for use in explosive shockwave mitigation and have been tested using singleand two-stage gas-gun driven plane impact experiments [128–131]. In response to the intense interest in the mechanical properties of the Desmodur N3300A/water-based polyurea aerogels, more recently the nonlinear mechanical properties, deformation mechanisms, and failure modes of such polyurea aerogels were investigated in detail using a multiscale approach that combined experimental nanoindentation, analytical modeling, and computational modeling [132]. First, primary particles were built from the monomer and up, and their mechanical interactions were investigated with molecular dynamics simulations. From nanoindentation, four deformation and failure modes were identified: free ligament buckling, cell ligament bending, stable cell collapse, and ligament crush induced strain hardening. The corresponding structural evolution during indentation and strain hardening was analyzed and modeled. The material scaling properties were found to be dependent on both the bulk density and the secondary particle size. Best fit of the data was achieved with secondary particles comprising ten primary particles. Using a porosity-dependent material constitutive model, a linear relationship was found between the strain hardening index and the secondary particle size instead of a conventional power-law relationship. Finally, the structural efficiency of Desmodur N3300A/ water-based polyurea aerogels with respect to their energy absorption capability was evaluated as a function of structural parameters and base polymeric material properties. Because of their high mechanical strength, Desmodur N3300A/water-based polyurea xerogels and aerogels have been considered for use in concrete confinement [133]. Later, inspired by the structure of nacres, this literally bricksand-mortar concept of composite design employing Desmodur-N3300A/water-based polyurea aerogels was translated into nanocellular composites by infiltrating preformed polyurea networks in acetonitrile with a hard inorganic phase, namely, magnesium phosphate cement particles. In essence, the material design here was a conceptual inverse to polymer-crosslinked oxide aerogels (see ▶ Chap. 29). Owing to nanoconfinement effects, the effective compressive modulus and compressive strength of the polyurea ligaments (measured using nanoindentation) were found to be eight to ten times higher than those in the non-infiltrated aerogels,

21

532

while both the low density (0.285 g cm3) and the high porosity (86.4% v/v) of the parent aerogel were preserved. Overall, this bioinspired nanocomposite design exhibited synergistic properties, offering both high strength and deformability. This nanoconfinement effect was subsequently modeled analytically [134]. With an eye toward biomedical applications, Yin and Rubenstein evaluated the biocompatibility of Desmodur N3300A/water-derived polyurea aerogels as a function of their density and morphology in the range of 0.035–0.21 g cm3 in the form of their effects on the vascular system by looking at their hemolysis, platelet activity, endothelial cell activity, and inflammatory responses [135]. Desmodur N3300A/water-derived polyurea aerogels did not absorb proteins, alter blood cells, or cause inflammatory responses. All polyurea aerogels were compatible with endothelial cells, although changes were observed in the gel morphology after contact with human platelet poor plasma for 48 hours followed by washing with deionized water and air-drying. It was not clear whether those morphological changes, akin to structural collapse during xerogelling, were brought about by the contact with the plasma or by the postprocessing method and drying. Overall it was concluded that Desmodur N3300A/water-derived polyurea aerogels are suitable for cardiovascular applications. Based on the superhydrophobicity of certain polyurea aerogels of this class (see Fig. 21.32d above), samples with the cocoon-in-web morphology (made in acetonitrile, [21]) were tested for their oil absorption capacity, and results were compared with those from samples made using acetone that exhibit the typical morphological variability from fibrous to particulate as discussed throughout this section (see Fig. 21.30 [20, 116]). Low-density cocoon-in-web samples (0.073 g cm3) were found to uptake 11 times their weight in pump oil (Fig. 21.34), competing favorably with other aerogel materials evaluated for such applications, especially in terms of volumetric capacity [21]. As shown in Fig. 21.34, both density and morphology again play a role in oil uptake, with the ultimate oil capacity found to be related to overall porosity. More recently, the Jana group at the University of Akron revisited the subject of oil uptake from Desmodur N3300A/ water-based polyurea aerogels with an innovative twist. They developed a new class of hierarchically porous composite materials referred to as open cell aerogel foams (OCAFs) that combine the attributes of open cell polymer foams (pore size >1 μm) with those of mesoporous Desmodur N3300A/water-based polyurea aerogels (pore size ~50 nm) [136]. OCAFs were prepared using templating with co-continuous immiscible polymer blends. The open cell macropores were created by selective dissolution of polystyrene from a co-continuous blend of polyethylene oxide with tetrahydrofuran. The polyurea network was synthesized

N. Leventis

inside the macropores from an acetone sol. The polyethylene oxide phase was then dissolved away with water, and the resultant material was dried with SCF CO2 to obtain polyurea OCAFs. The microstructure of a typical polyurea OCAF is shown in Fig. 21.35a. Internally, the walls that define the macroporous network consisted of typical mesoporous Desmodur N3300A/water-derived polyurea fibers exhibiting material properties expected for the given densities. The surface of these walls clearly showed a skin effect comprising denser polymer at the outer geometric boundary (compare with Fig. 21.27). As shown in Fig. 21.35b, OCAF samples with a bulk density of 0.073 g cm3 showed a 14
w/w of oil uptake versus the 11
w/w uptake displayed by cocoon samples of the same bulk density (see Fig. 21.34). This difference in uptake performance can be attributed in part to the extra void space available in the OCAFs. Notably, the corresponding non-templated polyurea aerogels showed a much lower oil uptake, in the same range as the study by Leventis et al. (Fig. 21.34, lower left frame [21]), highlighting the reproducibility of these materials. Finally, in response to the broad interest in the structure and properties of Desmodur N3300A/water-based polyurea aerogels, the Leventis group sought to find ways to express structure-property relationships quantitatively. That quest led to the issue of how to prepare nanostructures at will. In this context, it was realized that in order to establish procedures that deterministically relate nanomorphology to synthetic conditions, it would be necessary to express nanostructure numerically. Guided by a statistical design-of-experiments (DoE) model, a large array of Desmodur N3300A/waterbased polyurea aerogels were prepared (188 samples initially) in which the solvent, the concentrations of the monomers in solution, and the catalyst were systematically varied [137]. At first, the structural variability seemed overwhelming. However, upon reflection on the SEM images of those samples, it was realized that one’s first impression about a nanostructure is related to its openness and the texture; the former is quantifiable by the porosity (Π), and the latter is often times related to hydrophobicity, which in turn can be quantified by the contact angle (θ) of water droplets resting on the material (see Fig. 21.32d). The ratio of θ to Π, referred to as the K-index (“K” being phonetic for correlator), could then be used to place all 188 polyurea aerogels of the DoE model into eight K-index groups associated with nanomorphologies ranging from caterpillar-like assemblies of nanoparticles to random assemblies of fused nanoparticles, entangled thin nanofibers, microspheres with hair, cocoonlike structures (like the one shown in Fig. 21.32a), and large bald microspheres (see Fig. 21.36). At first, the K-index was validated as a morphology descriptor by compressing samples to different strains and observing that as the porosity decreased, the water-contact angle decreased proportionally and thus the K-index

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

533

b

a Oil (1 g) on water (5 g)

30 sec later....

21

Aerogel

+ PUA-ACN-109

....Oil removed

Oil : aerogel (w/w)

12 -PUA-acetone-xxx

10

-PUA-ACN-xxx 8 6 4 2 0.1

0.2

0.3

0.4

r b (g cm–3)

w/w oil uptake vs. theoretical

F

P

1.0 0.8 0.6 0.4

-PUA-acetone-xxx -PUA-ACN-xxx

0.2 0.1

0.2

0.3

0.4

r b(g cm–3)

Fig. 21.34 (a) Removal of pump oil from water with a chunk of polyurea (PUA) aerogel made in acetonitrile (ACN; alphanumerical extensions in sample names are related to the sol concentration). The aerogel sample in this figure had a weight of 0.087 g, a volume of 1.19 cm3, a density of 0.073 g cm3, and a porosity of 94% v/v and exhibited an uptake oil-to-aerogel weight ratio of 11.5 w/w. (See Fig. 21.32a for an SEM image representative of the nanomorphology of this material.) (b) Left, gravimetric oil absorption as a function of

bulk density. Right: ratio of experimental versus theoretical oil uptake (the latter was calculated from the sample porosities and the density of the oil, which was 0.924 g cm3). The arrow above the right frame shows the direction of the morphological transition from fibrilar (F) to particulate (P). The labels PUA-acetone and PUA-ACN refer to polyurea aerogels made in acetone and acetonitrile, respectively. (Reprinted from [21], Copyright 2014 The American Chemical Society)

remained constant as expected by the fact that morphology should not change by compression – at least in the early stages [137]. The predictive power of the K-index was demonstrated by preparing 20 randomly formulated Desmodur N3300A/water-based polyurea aerogels in eight randomly selected binary solvent systems; the K-indices of the resulting 20 randomized samples were then determined experimentally (calculating θ/Π). The expected structures based on the K-index values (as shown in Fig. 21.36) were then compared with SEM images of the samples and were found to match. At this point, these additional 20 samples prepared in binary solvents were added to the pool of the 188 samples studied initially, increasing the domain of samples, which to extract material properties from, to 208 samples in all. Next, several

material properties of interest were correlated to nanomorphology using the K-index (see Fig. 21.37). Eventually, the structural diversity of this type of polyurea aerogels was related to the phase separation mechanism involved in gel formation (refer to Fig. 21.2 in Sect. 21.2 above). Furthermore, it is noted that numerous attempts in the literature to infer nanomorphology from quantifiable material properties have focused on mechanical properties, which, therefore, have been assumed as the link between nanomorphology and synthetic conditions [138, 139]. However, according to Fig. 21.37, most material properties of interest of polyurea aerogels are not single-valued functions of the K-index (i.e., the nanomorphology). Therefore, because at least in one case (polyurea) those properties cannot be used as unique

534

N. Leventis

1

b 1400

2 1

1 μm

2

Pump oil uptake/%

a

OCAF ρb = 0.073 g cm –3 Π = 94.1 % v/v

1200 1000 800 600 400

Aerogel

ρ b = 0.118 g cm–3 Π = 90.9 % v/v

200 0

20 μm

0

1 μm

3000

6000

9000

12000 15000

Time (s) Fig. 21.35 (a) SEM image of a polyurea OCAF (ρb ¼ 0.0073  0.003 g cm3, porosity Π ¼ 94.1% v/v). Inset 1: fractured surface of an aerogel wall domain. Inset 2: denser skin layer of the aerogel domain. (b) Weight percent of pump oil absorbed as a function Fig. 21.36 The eight nanomorphology groups identified from 188 formulations of Desmodur N3300A/waterderived polyurea aerogels prepared using eight different solvents (acetone, acetonitrile, nitromethane, propylene carbonate, THF, DMF, 2-butanone, ethyl acetate). K-index is defined as the ratio of water contact angle to percent porosity for a given material (θ/ Π). Each morphological group shown in this figure was associated with a unique K-index value. (Adapted from [137], Copyright 2019 The American Chemical Society)

of time by OCAFs and corresponding non-templated Desmodur N3300A/water-derived aerogels at the densities and porosities shown within the frame. (Adapted from [136], Copyright 2017 Elsevier Ltd.)

K = 1.6

K = 1.5 Assemblies of fused nanoparticles

Entangled nanofibers 200 nm

K = 1.4

200 nm

5 μm

K = 1.7 Cocoons of nanofibers

200 nm

Short worm-like assemblies

5 μm

Microspheres with hair 200 nm

K = 1.3

5 μm

K = 1.8

Caterpillar-like assemblies of nanoparticles

Bald microspheres 200 nm

K = 1.2

descriptors of nanomorphology, they cannot be assumed a priori as such descriptors in any other case. Next, all material properties of practical interest in Fig. 21.37 were fitted to the six independent variables of the system using quadratic models: the concentrations of Desmodur N3300A, water, and catalyst used in each sol

200 nm

5 μm

K = 1.9

and the respective three Hansen solubility parameters (HSP) of the sol (i.e., the three parameters calculated for the specific mixture comprising solvent, Desmodur N3300A, water, and triethylamine) [137]. Correlating aerogel properties (bulk density and thermal conductivity) to the properties of the solvent via the HSPs was first reported with PIR-PUR

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

a

b

lTotal (mW m–1 K–1)

r b (g cm–3)

0.6

0.4

0.2

535

70 60

21

50 40 30 20

0.0

1.2

1.4

1.6 K-index

1.8

c

d

1.2

1.4

1.6 K-index

1.8

1.2

1.4

1.6 K-index

1.8

1.2

1.4

1.6 K-index

1.8

160 140

2.0

q (deg.)

Log (E (MPa))

2.4

1.6 1.2

120 100

0.8 80 0.4 1.4

1.6 K-index

1.8

e 250

4

3

2.4

Log (r (nm))

200

s (m2 g–1)

f

Log (tgel (min))

1.2

150 100

2.0 1.6

50

2

1 1.2

0 1.2

1.4

1.6 K-index

1.8

Fig. 21.37 Selected material properties for all 208 Desmodur N3300A/ water-derived polyurea aerogel samples prepared using single and binary solvent systems as a function of their K-indices. (a) Bulk density, ρb. (b) Total thermal conductivity, λTotal. (c) Compressive Young’s modulus, E. (d) Contact angle, θ, of a water droplet on an exposed internal surface of a monolith of the given material. Inset: total reflection

from the layer of air trapped on the surface of a hydrophobic aerogel block submerged in water. Yellow arrow shows which group of K-index values the sample corresponds to. (e) BET surface area, σ, and particle radius, r. (f) Phenomenological gelation time, tgel. (K-index values were pinned to single-digit decimals in order to facilitate discussion.) (Adapted from [137], Copyright 2019 The American Chemical Society)

536

N. Leventis

aerogels prepared with sub-stoichiometric amount of polyol; with no quantitative tool available though, only qualitative observations were made in passing between the properties of interest and the aerogel nanomorphology [140]. On the contrary, in reference [137], the resulting six equations (one for each property) enabled the synthesis of polyurea aerogels with up to six prescribed material properties at a time, including nanomorphology, bulk density, BET surface area, compressive modulus, ultimate compressive strength, and thermal conductivity. (If a solution to the system of six equations with the six unknowns included a complex number or a negative value, materials with that particular combination of properties could not be made.) This study helped identify the lowest thermal conductivity value of Desmodur

NCO

NCO OCN Desmodur RE (TIPM) Fig. 21.38 Desmodur RE (tris(4-isocyanatophenyl)methane, abbreviated as TIPM in the literature), a rigid aromatic triisocyanate used in the preparation of isocyanate-derived aerogels

10¥

rb = 0.023 g cm–3

10¥

5 mm

rb = 0.037 g cm–3

10¥

5 mm

10¥

10¥

rb = 0.15 g cm–3

N3300A/water-based polyurea aerogels within the domain of this study as 18.5 W m1 K1 for a material exhibiting a caterpillar-like fibrous morphology (K-index ¼ 1.22) and a low bulk density (0.082 g cm3) synthesized in THF [137]. Formation of polyurea aerogels via reaction of triisocyanates with water has been applied successfully with other triisocyanates as well, including Desmodur N3200 (see Fig. 21.10), Desmodur RE (Fig. 21.38), and toluene diisocyanate [20]. Using the fully aromatic triisocyanate Desmodur RE (tris(4-isocyanatophenyl)methane), it was possible to prepare aromatic polyurea aerogels over the entire density range obtained with Desmodur N3300A (from about 0.02 g cm3 to over 0.25 g cm3). Using Desmodur N3200 and toluene diisocyanate, only aerogels with bulk densities of about 0.2 g cm3 and higher could be made. Polyurea aerogels from Desmodur RE are particularly interesting, because they can be pyrolyzed to carbon aerogels, and therefore are reviewed in more detail below (see also ▶ Chaps. 6 and ▶ 35). For cross-referencing purposes, polyurea aerogels derived from the reaction of Desmodur RE with water are referred to as PUA-yy, and their properties are used again in Sect. 21.4.3 as a benchmark for the properties of polyurea aerogels derived from the reaction of Desmodur RE with mineral acids (referred to as BPUA-xx). At low densities, PUA-yy exhibited fibrilar morphologies, progressively turning particulate with increasing density (Fig. 21.39). All other parameters held constant, the transition from fibrilar to particulate

5 mm

rb = 0.18 g cm–3

rb = 0.062 g cm–3

5

10¥

5 mm

rb = 0.25 g cm–3

5 mm

Fig. 21.39 SEM images of polyurea aerogels prepared from Desmodur RE using 3.0 mol equivalents of H2O and 0.6% (w/w) Et3N in acetone [20] ordered by increasing bulk density. (All scale bars are 5 microns.). (Reprinted from [20], Copyright 2010 The American Chemical Society)

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

morphology occurred at lower densities than for Desmodur N3300A/water-derived polyurea aerogels: by 0.062 g cm3, Desmodur RE/water-derived materials were particulate. The Desmodur RE-derived fibrils consisted of well-defined strings of nanobeads connected by narrow necks (see, e.g., the 0.023 g cm3 sample in Fig. 21.39). As the density increased to 0.037 g cm3, the constituent bead size remained about the same, but the interparticle neck zones became wider. Upon further increase in density, Desmodur RE/ water-derived polyurea aerogels appeared nanoparticulate, only exhibiting a very faint trace of the strings-of-beads nanostructuring, if any. Generally, the common feature of carbonizable polymers is that they can either cyclize or undergo ring fusion and subsequent chain coalescence by heating [141]. For these processes to happen, the polymer chain should either contain aromatic moieties linked by just one carbon atom or be aromatizable (usually oxidatively, e.g., as in the case of polyacrylonitrile, polybenzoxazines, and certain phenolic resins – see ▶ Chap. 20). Desmodur RE-derived polyurea aerogels fall into the former category. Their pyrolytic yield (at 800  C under Ar) was found to be around 56% w/w [20]. The resulting carbon aerogels consisted of a mixture of C (78–82% w/w), N (5–9% w/w), and O (~5–9% w/w) and contained no detectable H. The atomic ratio of O to N was

10¥

from rb = 0.023 g cm–3 10¥

0.62 ± 0.08 g cm–3

537

about one-to-one as determined by XPS. XRD showed only very broad diffractions. Raman spectroscopy showed both the G-band (graphitic) and D-band (disordered) peaks at 1597 cm1 and 1352 cm1, respectively, with an integrated peak intensity ratio ID/IG of 1.12. Together, these data suggest nanocrystalline/amorphous carbon, supported by skeletal density measurements (1.78–1.89 g cm3) that were found to be in line with amorphous carbon (1.8–2.0 g cm3) [142]. The bulk densities of these carbon aerogels were higher (in the range of 0.3–0.8 g cm3) than the densities of their parent polyurea aerogel precursors (in the range of 0.02–0.25 g cm3). By SEM (Fig. 21.40), all Desmodur RE/water-derived carbon aerogels appeared to be macroporous materials and were all very similar in appearance irrespective of their parent polyurea aerogel. In particular, the two lowest-density samples completely lost the stringof-beads structure exhibited by their parent polyurea aerogels, although they did retain a faint trace of the wider necks observed in, for example, 0.037 g cm3 samples (compare Figs. 21.39 and 21.40). The morphology of the nanoparticulate structures in samples with densities above 0.062 g cm3 (refer to Fig. 21.39) was retained more closely. It was proposed that the morphology equalizer across densities was sintering processes along pyrolysis.

10¥

10¥

0.29 ± 0.06 g cm–3 10¥

10¥

0.72 ± 0.03 g cm–3

Fig. 21.40 SEM images of carbon aerogels derived from polyurea aerogels via reaction of Desmodur RE and H2O in acetone. Common scale bar for all images in the lower right-hand corner is 5 μm. These SEMs correspond frame-by-frame to those of their parent polyurea aerogels shown in Fig. 21.39 [20]. (Note the carbon aerogel derived

0.40 ± 0.02 g cm–3

0.78 ± 0.01 g cm–3

from the polyurea aerogel with a density of 0.023 g cm3 broke to pieces during processing, and so the density of its carbonized derivative was not measured.). (Adapted from [20], Copyright 2010 The American Chemical Society)

21

538

N. Leventis

21.4.3 Polyurea Aerogels from Isocyanates and Mineral Acids As discussed in section “Reaction of –N¼C¼O with Mineral Acids: Alternative Synthesis of Polyurea Aerogels,” multifunctional isocyanates react with anhydrous mineral acids (e.g., H3BO3, H3PO4, H3PO3, H2SeO3, H6TeO6, H5IO6, and H3AuO3) to yield urea wet gels, typically doped with the corresponding oxide (i.e., the anhydride) of the mineral acid; exceptions were boric acid, wherein the oxide was washed out during solvent exchange, and auric acid, wherein the resulting dopant was metallic gold (Fig. 21.41) [66, 143].

Fig. 21.41 (A) Polyurea aerogel monoliths prepared in DMF from tris(4-isocyanatophenyl)methane (TIPM, see Fig. 21.38) and the acids listed underneath. (B) A polyurea aerogel monolith prepared in DMF from TIPM and H3AuO3 (a); its microstructure (b); and the residue obtained after pyrolysis at 600  C in air, consisting of pure gold (c). Although the material underwent partial sintering and significant shrinkage, panel B(c) shows that Au was evenly distributed throughout the monolith. (C) TEM of Au clusters in the Au-doped polyurea aerogel shown in panel B(a). The d-spacing of the clusters was found to be close to the literature value (0.2355 nm) for the (111) face of fcc Au(0) [66]. (Reprinted from [66], Copyright 2016 The American Chemical Society)

The model system for these studies was based on materials prepared using TIPM (see Fig. 21.38) and boric acid in DMF at room temperature (see Fig. 21.22, reaction c), which yielded nanoporous polyurea networks that were dried with SCF CO2 to produce robust polyurea aerogels (referred to as BPUA-xx). BPUA-xx were chemically (CHN, solid-state 13C NMR) and nanoscopically (SEM, SAXS, N2-sorption) very similar to the reaction product of the same triisocyanate (TIPM) and water (referred to as PUA-yy). For example, the bulk densities of BPUA-xx and PUA-yy at the same total weight percent of monomer concentration in the sol (4–16% w/w) were 0.28–0.58 g cm3 and 0.39–0.60 g cm3, respectively.

A

H3PO4 B

a

H3PO3

H2ScO3

H5IO6

Te(OH)6

b

c

Au(OH)3

200 nm

C

1 mm

0.233 nm (111)

50 nm

10 nm

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

106 BPUA-xx

O N H

H O N

PUA-yy NH2

H2N

160 140

53

NH2

120 100 80 60 40 20 15 Chemical shift of N (ppm)

0

Fig. 21.42 Solid-state CPMAS 15N NMR spectra of boric acid-derived BPUA-xx (top spectrum) and water-derived PUA-yy (middle spectrum) along with the liquid-phase 15N NMR spectrum of methylene dianiline in CD3NO2 (bottom spectrum) used as a control for identifying dangling aromatic amines. Chemical shifts are reported versus liquid ammonia. (See text for sample number taxonomy.). (Adapted from [66], Copyright 2016 The American Chemical Society)

539

Microscopically, both materials consisted of assemblies of nanoparticles as seen in PUA-yy aerogels of the same density range (see Fig. 21.39). Minute differences were detected in the primary particle radii (6.2–7.5 nm for BPUA-xx vs. 7.0–9.0 nm for PUA-yy) and the micropore size within these primary particles (6.0–8.5 Å for BPUA-xx vs. 8.0–10 Å for PUA-yy). A significant difference was noted in the solidstate 15N NMR spectra, in which only PUA-yy showed some dangling -NH2 groups; no such -NH2 groups were detected in the spectra of BPUA-xx (Fig. 21.42). Together, all data were consistent with exhaustive reaction of the isocyanate groups in BPUA-xx materials in accordance with Fig. 21.18. This exhaustive reaction removed dangling functional groups by stitching the ends of polymeric strands together, which was considered as being the reason BPUA-xx aerogels are the stiffest polymeric aerogels we are aware of at all densities (Fig. 21.43). Residual boron in BPUA-xx aerogels was quantified with prompt gamma neutron activation analysis (PGNAA) and was found to be very low (about 0.05% w/w) and primarily attributable to B2O3 as determined by 11B NMR [66]. Thus, any mechanism for systematic incorporation of boric acid into polymeric chains analogous to carboxylic acids was ruled out. (In fact, it was derived mathematically that boron-terminated star polyurea from TIPM should contain about 3.3% w/w of boron, irrespective of the size of the star

0.576 g/cc 0.545 g/cc

Young’s modulus, E (MPa)

600 500 400 0.407 g/cc 300 200 0.283 g/cc 100

BPUA -4 aR -BPA -15 aR -POL -10 aR -DHB-15 aR -DHB-20 a R-SDP -15 aR -HPE -15 20 -aR -PAC -EG 20 -aR -PAC -HD 20 -aL -PAC -EG 20 -aL -PAC -HD 20 -aR -NOR 15 -aL -NOR 20 -aL -NOR PA -10 PA -15 aR -BTDA-6 PI-AMN-90 -20 BPUA -8 aR -BPA -20 aR -DHB-25 aR -SDP -20 aR -HPE -20 aR -RES -15 30 -aR -PAC -HD 30 -aR -NOR PA -25 aR -BTDA-20 aR -PMDA-6 BIS-NAD-10 PNB-30(10:50) BPUA -12 aR -POL -15 aR -SDP -25 30 -aR -PAC 30 -aR -PAC -EG 30 -aL-PAC -EG 30 -aL-PAC 30 -aL-PAC -HD 30 -aL-NOR aL-PUA BIS-NAD-15 PNB -30 (30:70) PNB -30 (0:100) BPUA -16 aR -BPA -25 aR -HPE -25 aL-HPE -25 40 -aL-PAC -EG

0

Fig. 21.43 Compressive Young’s moduli for four density groups (xx) of polyurea aerogels obtained from the reaction of Desmodur RE (TIPM) with boric acid (BPUA-xx, indicated with yellow arrows),

along with compressive Young’s moduli for other polymeric aerogels. (Refer to the text of this chapter and the Supporting Information for reference [66] for sample taxonomy)

21

540

N. Leventis

monomer (refer to Appendix II in the Supporting Information of reference [66])). Upon reflection, it was concluded that it was fortuitous that the gelation study of isocyanates with mineral acids was conducted with H3BO3, whereas the resulting by-product, B2O3, was easily removed from the resulting gels during post-gelation solvent exchanges, thereby leaving behind relatively easy-to-identify pure polyurea. With other mineral acids and with dense polymers that cannot be washed via solvent exchange, experimental results could have been misleading, as the corresponding oxides are insoluble and remain within the polymer (confirmed with both skeletal density measurements and EDS). This said, forming of polyureas using mineral acids other than boric acid provided a convenient method for in situ doping of mechanically robust porous polymeric networks and carbon aerogels derived from them with a diverse array of oxides and pure metal nanoparticles (Au in the case of H3AuO3) with possible applications in catalysis. This method of doping polyurea and carbon aerogels stands open for further investigation.

21.5

Polyurethane Aerogels

Traditional polyurethane foams provide a good combination of thermal insulating performance and cost [144]. As such, polyurethane aerogels that could potentially further extend the insulating performance of porous polyurethanes by leveraging the Knudsen effect have been a logical area of interest for aerogels. Polyurethane (PU) aerogels were first reported in 1998 by Biesmans et al., using Suprasec DNR (an aromatic polymeric isocyanate produced by ICI Polyurethanes) in CH2Cl2 employing 1,4-diazabicyclo[2.2.2]octane (DABCO) as a catalyst [10, 145]. Curiously, the report makes no specific mention of an alcohol but instead emphasizes the role of DABCO as a trimerization catalyst that converts isocyanates to isocyanurates. Subsequently, in 2001, Tan et al. reported cellulose aerogels crosslinked with toluene diisocyanate (TDI) exhibiting an impact strength ten times higher than that of resorcinol-formaldehyde (RF) aerogels [146]. In 2002, Yim et al. reported silica-polyurethane hybrid aerogels produced by cogelation of tetramethoxysilane (TMOS) and polymeric methylene diisocyanate (MDI) [147]. Those aerogels had a bulk density of 0.07 g cm3 and a thermal conductivity of 0.0184 W m1 K1 at 1 torr; however mechanical characterization was not included in that report. In 2004, Rigacci revisited PU aerogels with an emphasis on thermal superinsulation synthesizing materials using Lupranat M20S (a polymeric MDI) with two aliphatic polyols, saccharose and pentaerythritol, and DABCO as a catalyst in DMSO/ ethyl acetate mixtures [148]. Materials produced using both supercritical and subcritical drying were compared in terms

of bulk density, pore volume, and thermal conductivity. The latter was less than that of standard polyurethane foams (0.022 versus 0.030 W m1 K1 at room temperature and atmospheric pressure). It was also shown that the aerogel morphology depended on the solubility of the precursors as well as the Hildebrand solubility parameter of the reaction medium (δm). When δm was lower than the solubility parameter of the polyurethane (δPU ¼ 10 (cal cm3)1/2), the aerogel consisted of aggregates of micron size particles (Fig. 21.44); when δm was greater than δPU, smaller size particles and mesoporous structures were reported (Fig. 21.45) [148]. In 2009, Lee et al. reported polyurethane aerogels made from 4,4′-methylenediphenyl diisocyanate (MDI) and a polyether polyol (Multranol 9185) catalyzed by triethylamine [102]. The properties of the resulting PU aerogels were then compared to those of silica aerogels and polyurea aerogels prepared from the reaction of MDI or polymeric MDI with Jeffamines T3000 and T5000 as discussed in Sect. 21.4.1 above. As shown in Fig. 21.24b, the polyurethane aerogels in this work exhibited consistently higher thermal conductivities than MDI-derived polyurea aerogels of similar densities over a wide pressure range. Representative polyurethane aerogels in that report had a bulk density of 0.128 g cm3, a surface area of 47 m2 g1, an average pore diameter (in the mesoporous range) of 13 nm, and a thermal conductivity of 0.027 W m1 K1 at atmospheric pressure [102].

Fig. 21.44 SEM of polyurethane (PU) aerogels synthesized in low solubility reaction media, i.e., δm < δPU, from saccharose and polymeric MDI (a) and from pentaerythritol and polymeric MDI (b). (Reprinted from [148], Copyright 2004 Elsevier B. V)

Fig. 21.45 SEM of polyurethane (PU) aerogels synthesized in high solubility reaction media, i.e., δm > δPU, from saccharose and polymeric MDI (a) and from pentaerythritol and polyMDI (b). (Reprinted from [148], Copyright 2004 Elsevier B. V)

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

Up until this point, the synthesis of polyurethane aerogels mostly involved the use of oligomeric isocyanates used in commercial bulk polyurethane synthesis and/or highmolecular-weight polyols. In 2013, Leventis et al. reported a broad series of PU aerogels derived from multifunctional small-molecule monomers [29, 149]. The objective of that work was to force early insolubility of developing oligomers during gelation (refer to the left branch of Fig. 21.2), thus enabling control over particle size, morphology, pore structure, and the mechanical properties of the resulting aerogels. The first molecular parameter of interest was the molecular rigidity/flexibility of the isocyanate, which was varied by using either the rigid aromatic monomer TIPM (Desmodur RE – refer to Fig. 21.38) or the flexible aliphatic monomer Desmodur N3300A (refer to Fig. 21.21). The alcohols selected for reaction with these two isocyanates were all small-molecule aromatics (Fig. 21.46), selected to allow for three additional degrees of design freedom: (a) the number of -OH groups, n, per monomer; (b) the ratio of -OH groups to aromatic rings, r; and (c) the degree of steric crowding at the bridge between aromatic rings. The rationale was that parameters n and r should manifest as a real OH-group density on the surface of the nanoparticles that make up the gel, which in turn affects interparticle connectivity and mechanical strength. Polyurethane formation was catalyzed by DBTDL (see Fig. 21.14) and was carried out in mixed solvent solutions of acetone and ethyl acetate. Samples based on aromatic TIPM (Fig. 21.38) or aliphatic Desmodur N3300A (Fig. 21.21) were referred to with the prefixes aR- or aL-,

OH OH Triols

HO

OH OH

HO POL (n=3, r=3)

HPE (n=3, r=1)

OH

O

O S

Diols

OH HO RES (n=2, r=2)

HO

OH SDP (n=2, r=1) O

OH

OH HO BPA (n=2, r=1)

DHB (n=2, r=1)

Fig. 21.46 Polyfunctional small-molecule alcohols and their associated n and r parameters (see text) [29]

541

respectively, and abbreviated as aR-ALC-xx or aL-ALC-xx accordingly, where ALC is an abbreviation referring to the specific alcohol used (Fig. 21.46) and the suffix -xx indicates the weight percent of the monomers used in the sol. All material properties were found to depend strongly on the choice of aR- or aL- and subsequently on n and r. Microscopically, aR- aerogels consisted of discrete particles, whereas the skeletal particles of aL- aerogels were coated and fused with additional polymer. Within the aR series of aerogels, material properties depended strongly first on n and second on n þ r. For materials with the same value for n þ r, properties were further dependent on the molecular flexibility of ALC. For example, BET surface areas for aRPOL-xx (n ¼ 3, n þ r ¼ 6) were in the range of 200–240 m2 g1; for aR-HPE-xx (n ¼ 3, n þ r ¼ 4) were in the range of 130–235 m2 g1; for aR-RES-xx (n ¼ 2, n þ r ¼ 4) were in the range of 33–120 m2 g1; for aR-SDPxx (n ¼ 2, n þ r ¼ 3) were in the range of 3–28 m2 g1; for aR-BPA-xx (n ¼ 2, n þ r ¼ 3) were in the range of 1–49 m2 g1; and for aR-DHB-xx (n ¼ 2, n þ r ¼ 1) were less than 1.0 m2 g1. An opposite trend was observed for primary particle sizes as determined from skeletal density and N2-sorption porosimetry data. Primary particle sizes from this method matched well with primary particle sizes calculated from small-angle X-ray scattering (SAXS) for aR- aerogels with n þ r ¼ 4 (cases of aR-POL-xx, aR-HPE-xx, and aR-RES-xx), which was attributed to a combination of (a) fast reactions leading to consumption of all monomers before the sol underwent gelation and (b) drastic decrease of solubility of the developing polymer with increasing n þ r. According to SAXS, primary particles in those samples were packed into surface-fractal secondary particles, while rheology suggested that their skeletal networks were formed by higher aggregates of secondary particles, as illustrated in Fig. 21.47. The situation was markedly different for materials made with lower values of n and r (specifically aR-SDP-xx, aR-BPA-xx, and aR-DHB-xx, n þ r ¼ 3). At lower densities (e.g., aR-SPD-10), the skeletal building blocks were large and featureless (~1.6 μm in diameter as determined by both SEM and BET data); however SAXS showed that these building blocks still consisted of smaller particles (~200 nm in diameter). At higher densities (e.g., aR-SDP-25), particle sizes determined with SEM, N2-sorption, and SAXS converged (all showing particle radii in the 60–70 nm range). These data were considered as being consistent with low-density aerogels in this group consisting of small primary particles embedded in a medium (polymer) of lower density (e.g., aR-SPD-10). This was attributed to the fact that when n þ r is low, oligomers are more soluble, and phase separation is delayed resulting in generally larger particles that begin to aggregate while a significant number of oligomeric species are still in solution. These oligomers then

21

542

N. Leventis

Primary particle

Secondary particle Dense packing

Aerogel network Mass-fractal aggregate of secondary particles Fig. 21.47 Proposed microstructure of polyurethane aerogels derived from small-molecule monomers [29]

b 2400 0.12 2000

aR-ALC-15 aR-ALC-20 aR-ALC-25

POL RES

1600 E (MPa)

Thermal conductivity (λ,W m–1 K–1)

a

0.08

1200 800

0.04

aR-POL-xx aR-HPE-xx aR-SDP-xx 0.6 0.2 0.4 Bulk density (ρ b, g cm–3)

0.8

HPE SDP

400 0 0.00

0.04

0.08

0.12

C (W m–1 K–1)

Fig. 21.48 (a) Thermal conductivity as a function of density of polyurethane aerogels derived from small-molecule monomers using the aromatic triisocyanate Desmodur RE (denoted with the prefix aR, see Fig. 21.38; alcohols are abbreviated as indicated in Fig. 21.46). (b) Compressive Young’s modulus, E, under dynamic compression versus

interconnectivity parameter C (from thermal conductivity data) for aRALC-xx aerogels. Inset: a flexible aR-HPE-5 aerogel (with ρb, 0.094 g cm3, Π ¼ 92% v/v; surface area, 132 m2 g1; primary particle size ¼ 16–18 nm; λ ¼ 0.041 W m1 K1). (Adapted from [29], Copyright 2013 The American Chemical Society)

accumulated over the resulting protonetwork and filled the interparticle pores. Thereby, mesoporosity was lost, and skeletal particles appeared larger and smoother in SEM. The mechanical properties of this last subclass of polyurethane aerogels ranged from exhibiting soft, foam-like flexibility to exhibiting energy absorption values in compression three times higher than armor-grade ceramics (up to 100 J g1). The lowest thermal conductivity measured for

these aerogels was 0.031 W m1 K1 for aR-POL-10 at a bulk density of 0.298 g cm3 (Fig. 21.48a). Subsequently, the trends in thermal conductivity noted in Fig. 21.48a were used in conjunction with compressive modulus values to assess interparticle connectivity as a function of n and r. For this, the solid thermal conductivity, λs, of each aR-ALC-xx aerogel was calculated from the total thermal conductivity measured experimentally using the laser flash method. The variation of

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

λs with the bulk density, ρb, was modeled with an exponential expression [119, 150]: λ s ¼ C ð ρb Þ

a

wherein the exponent α depends on the way in which matter in the material occupies and fills space and the coefficient C represents the amount of interparticle contact area per unit volume, which depends on the chemical composition of the particles and the nature of the interparticle coupling (degree of interconnectivity and interparticle bonding). As shown in Fig. 21.48b, the elastic modulus, E, varied linearly with C for fixed “xx” values (i.e., sols comprising different concentrations of monomers). Interestingly, despite the fact that HPE has a value n ¼ 3, like POL, interconnectivity (C) and stiffness (E) of aR-HPE-xx were lower than for both aRRES-xx and aR-POL-xx at all densities. For flexible materials (lower-density aR-HPE-xx and aR-SDP-xx), both interconnectivity and stiffness were low and numerically close to one another. Considering thermal conductivity and elastic moduli together, it was concluded that the most important molecular parameter for improving interparticle connectivity (and thus increasing stiffness) was the monomer functional group density r, which translates into functional group density on the surface of the primary particles that comprise the aerogel, which leads to more interparticle bonds. It is noted that the previous study, albeit comprehensive, mostly focused on polyurethane aerogels based on the aR-(omatic) triisocyanate Desmodur RE. This was because the properties of polyurethane aerogels based on the aL-(iphatic) triisocyanate Desmodur N3300A and the aromatic diols listed in Fig. 21.46 were more difficult to control in the solvent system used for gelation (acetone). With this point of departure, subsequently, efforts surrounding polyurethane aerogels derived from small-molecule monomers branched out into two parallel paths. The first path (Sect. 21.5.1) focused on the effect of diol size on the gelation of Desmodur N3300A. Fig. 21.49 Monomers used for the synthesis of superelastic shape-memory aerogels: the isocyanurate-based triisocyanate Desmodur N3300A. (Reproduced from Fig. 21.21) and short diol derivatives of ethylene glycol [151])

543

The size of the diol was varied systematically using short oligomers of ethylene glycol (EG). The second path (Sect. 21.5.2) focused on deconvolving three aspects pertaining to the molecular structure of the polyurethane network: (a) deconvolving the effect of the functionality of the polyurethane core of the polymer repeat unit from the degree of functionality of the monomer by evaluating trifunctional versus 9-functional (nonafunctional) monomers synthesized from the same triisocyanates; (b) deconvolving the rigidity/ flexibility of the core from the flexibility of the polymer chain making up the aerogel framework; and (c) deconvolving the rigidity of the core from its aromatic character.

21.5.1 Polyurethane Aerogels from a Flexible Triisocyanate (Desmodur N3300A) and Short Aliphatic Diols This line of research effectively served as a bridge between polyurethane aerogels derived from small-molecule monomers and polyurethane aerogels derived from polymeric starting materials. Initially, materials were prepared by reacting Desmodur N3300A with high-molecular-weight diols (in the range of 3000 amu) using acetone as the solvent. The resulting gels shrunk excessively upon drying and yielded materials that resembled dense plastics. To understand the effects of diol molecular weight (size), investigations resorted to the simplest possible aliphatic diol, ethylene glycol (EG), and its dimer, trimer, and tetramer (Fig. 21.49). In the process, a fortuitous event occurred: solvent that was delivered for use in these experiments that was supposed to be acetone was in fact an acetonitrile/acetone mixture. Poly (isocyanurate-urethane) aerogels made from Desmodur N3300A and these various short EG derivatives in that solvent were superelastic and showed a shape-memory effect. These highly desirable material properties became a strong motivating factor to resolve severe reproducibility issues that

Isocyanurate-based triisocyanate

O

Ethylene glycol-based diols : HO–R–OH (ALC)

HO

(CH2)6NCO N O

OCN(H2C)6 N

EG HO

N (CH ) NCO 2 6 O

HO

O DEG

OH

O O TEG

N3300A HO

OH

O

O TTEG

OH O

OH

21

544

N. Leventis

102

G‫׳‬

101

G‫״‬

100

b

2.0 1.5 1.0

10–1

0.5

10–2 10–3

2.5 39.9 °C

0.0

tan d

G‫ ׳‬, G ‫״‬

a

after the first stretching cycle, which was traced to a stretching-induced increase in H-bonding interactions (NHO5C and NHO(CH2)2). Above the Tg transition zone, the modulus of all formulations decreased by about a thousandfold. This gave rise to a robust shape-memory effect (SME), which was quantified via several figures of merit calculated from tensile stretching measurements over five thermomechanical cycles between Tg þ 10  C and Tg – 40  C as shown in Fig. 21.52a. The strain fixity was always >98%, pointing to very low creep. After the first cycle, strain recovery (a measure of fatigue) improved from 80–90% to about 100%. The fill factor, a cumulative index of performance calculated from the ratio of the shaded area to the entire area of the projection of the 3D plot of Fig. 21.52a onto the temperature-strain plane (Fig. 21.52b), reached 0.7, which is in the range of fast elastomers. In studying these properties, an additional figure of merit, the shape recovery rate Rt(N), wherein N is the thermomechanical cycle number) was introduced, which is calculated from the maximum slope at the inflection point of segment 4 in Fig. 21.52b. The robust shape-memory effect of these materials was demonstrated with deployable panels and a bionic hand capable of mimicking coordinated muscle function (Fig. 21.53). It is not difficult to see at this point how deployable panels like the one shown in Fig. 21.53 can be configured, for example, as orthopedic casts, especially if the material is based on high Tg aerogels. With an eye toward mass production of this class of superelastic shape-memory aerogels, large panels (70 cm
70 cm
5 cm) of Desmodur N3300A-TEGbased aerogels were prepared using ambient pressure drying

Tg (max in tan d, °C)

arose when attempting to repeat the same experiments using actual acetone (i.e., what the original solvent was supposed to be) from a different batch of solvent wherein all gels collapsed upon drying. Moving synthesis to acetonitrile explicitly enabled production of materials that exhibited most of the superelasticity and shape-memory properties of the initial accident, but not fully. Thus, a systematic study that separately investigated each of the four diols was undertaken in order to find the optimum volume ratio (VCH3CN/Vsol) of various acetonitrile/acetone mixtures as the gelation solvent while simultaneously studying the properties of the resulting aerogels as a function of the total monomer concentration in the sol [151]. These two independent variables were varied systematically over the ranges of 0.5  VCH3CN/Vsol  1.0 and 15% w/w  total monomer 25% w/w, respectively, using a statistical design-of-experiments (DoE) approach and a central composite rotatable design (CCRD) model. Poly(isocyanurate urethane) (PIR-PUR) aerogels from all four diols consisted of micrometer size particles; however, it is noted also that the skeletal frameworks of most gels synthesized with TTEG as the diol in high-acetone-containing solvent mixtures collapsed into dense plastic-like objects. Bulk densities for the PIR-PUR aerogels that did not collapse were in the range of 0.2–0.4 g cm3, with typical porosities between 70% and 80% v/v. Glass transition temperatures (Tg) from dynamic mechanical analysis (DMA) data varied from about 30  C (using TTEG) to 70  C (using EG) (Fig. 21.50). Tg values did not depend on the density or the morphology of the samples, only on their chemical makeup. At and above Tg, all aerogels in this study showed rubber-like elasticity (i.e., superelasticity, see Fig. 21.51). They also became stiffer

80 70

60 DEG 50 40

Fig. 21.50 (a) Storage (G′) and loss (G00 ) moduli and tangent δ (¼G00 /G′) curves obtained from DMA of a representative TEG-based PIR-PUR aerogel as a function of temperature. The glass transition temperatures, Tg, were reported as the maxima of tan δ. The arrow points to the maximum in the G00 curve (generally located 10 to 13  C below the maximum of tan δ), which marks the onset of segmental motion of

TEG TTEG

30 20

–150 –100 –50 0 50 100 150 Temperature (°C)

EG

1

2 3

4

5

6

7

8

9 10

Sample no. on the rotatable domain the polymer and may be considered either as the upper temperature limit for fixing the shape of the polymer or as the onset of shape recovery. (b) Glass transition temperatures (as maxima in tan δ) for all noncollapsed PIR-PUR aerogels based on the diols shown in Fig. 21.49. (Adapted from [151], Copyright 2017 The American Chemical Society)

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

t = 0.0

t = 0.5 s

545

t = 1.0 s

t = 3.0 s

21

Fig. 21.51 Room temperature superelasticity with a TEG-formulated sample (sample dimensions 200
800
0.400 (5 cm
20 cm
1 cm); other properties: ρb ¼ 0.41  0.02 g cm3, Π ¼ 67  1% v/v). (Adapted from [151], Copyright 2017 The American Chemical Society)

a

b

2

0.020

70

3

2

60 0.015 0.010

1 0.005 0.000

Tf

Strain (%)

Stress (MP

a)

3

4 0

Te 10 20 mp era 30 40 tur 50 e( °C )

0

10

20

40 30

70 60 50

)

n

rai

St

(%

Td = T r

50 40

1

4

30 20 10 0

0

10

20

30

40

50

Temperature (°C)

Fig. 21.52 (a) Three-dimensional representation of a five-cycle thermomechanical experiment for the quantification of the figures of merit for the shape-memory effect. As-prepared samples were heated to the deformation (¼ recovery) temperature Td (¼ Tr), and then they were stretched (Stage 1). In Stage 2, samples were cooled to the fixing temperature (Tf) under constant stress. In Stage 3, stress was released while continuously monitoring strain. In Stage 4, samples were again

heated to Tr, and the procedure was repeated. Note that samples were stiffer initially but settled after one cycle. (b) Projection of the 3D plot shown in (a) onto the temperature-strain plain. The fill factor (FF) was calculated as the ratio of the shaded area over the enclosing box. FF is sensitive to the slope of segment 4, which was quantified by introducing an additional figure of merit referred to as shape recovery rate (Rt(N)). (Adapted from [151], Copyright 2017 The American Chemical Society)

of pentane-exchanged wet gels and were tested under both quasistatic and dynamic loading conditions [97]. The samples did not shrink more than their SCF CO2-dried counterparts (22–25% in both cases) despite ambient pressure drying. Bulk densities of the panels ranged from 0.28 to 0.37 g cm3, and all porosities were above 70%

v/v. Microscopically the materials comprised fused spherical particles around 10 μm in size and were hydrophobic, exhibiting water contact angles around 130 . Compressive Young’s modulus increased with density, displaying an unusually high slope (>6.0) on a log-log plot of modulus versus density. Samples compressed to >80% strain

546

N. Leventis

recovered to >80% of their shape within 30 s. Aerogels at the lower end of the density range evaluated in this study had a Poisson’s ratio of 0.22. Under dynamic conditions, the complex moduli and stress-strain curves were highly frequencyand rate-dependent. For example, in a Hopkinson pressure bar experiment, materials were three orders of magnitude

a Deployable panels t = 0.0 77 K

t = 6.0’

t = 10’

t = 16’ 23 °C

b A bionic hand 77 K

23 °C

Fig. 21.53 (a) Shape-memory aerogel panel similar to that in Fig. 21.51. The permanent shape of the specimen was flat. The sample was heated above Tg, folded as shown in the first frame, and then dipped in liquid N2. Unfolding of the panel back to its original shape took place as the sample returned to room temperature. (b) Bionic hand based on a shape-memory aerogel similar to the one used for the deployable panel. The permanent shape (right frame) of the bionic hand was programmed to hold a pen. (Adapted from [151], Copyright 2017 The American Chemical Society)

stiffer than under quasistatic compression. These results suggested potential applications for such aerogels in lightweight flexible energy absorbers. With the latter application in mind, the next objective became to reduce the bulk density of those superelastic PIR-PUR aerogels with a minimal compromise in stiffness. That was achieved by providing the aerogel bulk with a foam-like structure and by replacing TEG with ethylene glycol (EG). The foam-like structure was prepared without chemical foaming agents or templates, by injecting pressurized air (7 bar) into the sol, which was allowed to gel under pressure, followed by slow depressurization. Voids (see Fig. 21.54) were created from the air bubbles formed during depressurization. Such aerogel foams exhibited lower bulk densities by about 25% and higher porosities by about 10% in comparison with their PIR-PUR aerogel counterparts prepared under ambient pressure drying. The thermal conductivities of aerogel foams were also found reduced significantly (by 25%) from 0.104 to 0.077 W m1 K1 at a bulk density of 0.25 g cm3 and porosity at 80% v/v [152]. In parallel, the properties of the shape-memory aerogels based on Desmodur N3300A and the short diols listed in Fig. 21.49 were explored further with aerogels prepared from multiple-diol mixtures of DEG, TEG, and TTEG [96]. Again, samples were prepared based on DoE modeling with three independent variables: (a) the total monomer concentration, (b) the mol fraction of DEG (the most rigid diol), and (c) the mol fraction of TTEG (the most flexible diol), wherein TEG acted as a filler. Screening tests indicated that using pure acetonitrile to synthesize PIR-PUR aerogels comprising multiple diols as described in the previous paragraph produced materials with

25 μm

50 μm Aerogel foam

Fig. 21.54 SEM images of a PIR-PUR aerogel foam prepared from Desmodur N3300A and ethylene glycol using the pressurized gelation method [152], in comparison with a regular aerogel with the same

5 μm

50 μm Regular aerogel

formulation. The void size is similar to that noted in polyurea OCAFtype foams in Fig. 21.35. (Reprinted from [152], Copyright 2020 Elsevier Ltd)

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

slower elastic recovery and pointed to an optimal VCH3CN/ Vsol ratio of 0.875, which was then used for all sols. Microscopically, all aerogels in this study could be placed into one of two groups: one consisting of variable diameter micron size particles connected by large necks and another one, which was classified as bicontinuous (Fig. 21.55). These two sets of morphologies were attributed to late versus early solidification of phase-separated oligomers (refer to Fig. 21.2 right branch). Irrespective of microstructure, all samples showed a robust shape-memory effect with shape fixity and shape recovery ratios close to 100%. Larger variations (0.35–0.71) in the fill factor were traced to variability in shape recovery rates, Rt(N), which at first glance seemed to

a

547

be related to microstructure; however, it was noted that materials with bicontinuous microstructures (Fig. 21.55a) were stiffer and showed slower recovery rates. Thereby, using the tensile (elastic) modulus, E, as a local proxy for microstructure, and based on the fact that the energy profiles of elastic deformation are parabolic (Fig. 21.56a), a general relationship was derived between the shape recovery rate, Rt(N), and the elastic modulus, E. That correlation of Rt(N) with E was traced to the relationship between the activation barrier for shape recovery, ΔA#, and the specific energy of deformation (the reorganization energy, λ see Fig. 21.56a). Based on the geometry of these crossing parabolas, it was calculated that the reorganization energy, λ, is proportional to the elastic

b

c

20 mm

Fig. 21.55 Representative microstructures of mixed diol PIR-PUR aerogels ranging from bicontinuous (a, from high-concentration sols that included all three DEG, TEG, TTEG) to small spheroidal (b, from high-concentration sols based on a combination of DEG and TEG in a 1:

Deformation

b

Fixing

A E2 > E1 E1

2

1 D A# DA

l

DA D A

em

1 mol:mol ratio) to large spheroidal (c, from low-concentration sols based on the same combination of DEG and TEG as in the middle frame). (Adapted from [96], Copyright 2018, The American Chemical Society)

e

Recovery

Fig. 21.56 (a) Free energy (A) surfaces upon elastic deformation (ε) at the deformation temperature (Td > Tg, red parabolas) and at the fixing temperature (Tf < Tg, blue parabola); E1, E2 represent two possible elastic moduli at Td. (b) Correlation of log10 [Rt(N ¼ 5)/ρb] and E/ρb.

Log ((R ( t /rb ) (cm3min–1 g –1))

a

50 mm

20 mm

2.0 DEG - sole alcohol TEG - sole alcohol TTEG - sole alcohol MIX-xx

1.5

1.0

0.5

0.0 –0.5

Slope: –0.180 ± 0.008 Intercept: 1.445 ± 0.023 R2 = 0.92

0.0

0.5

1.0

E(r b) (MPa

1.5

cm3

2.0

2.5

g–1)

The graph includes data from single- and mixed diol PIR-PUR aerogels [96, 151] as indicated. Statistical information related to the fitting (slope, intercept, correlation) is provided within the plot. (Adapted from [96], Copyright 2018, The American Chemical Society)

21

548

N. Leventis

modulus at the deformation temperature (Td, see Fig. 21.52a). Shape recovery rate data from aerogels made both with mixed diols [96] and with single diols [151] fitted well the derived equations (Fig. 21.56b, R2 ¼ 0.92). Along these studies, it was recognized that the inverse correlation between Rt(N) and the elastic modulus, E, is a thermodynamickinetic relationship analogous both in form and in essence to the Marcus equation for outer sphere single electron transfer. Accordingly, this equation was proposed as a means for qualitative prediction of shape recovery rates, fill factors, and overall performance of the shape-memory effect.

21.5.2 Deconvolving the Properties of the Polymeric Nodes from the Functionality of the Monomer: Poly (Urethane-Acrylate) and Poly(UrethaneNorbornene) Aerogels Polyurethane aerogels have been prepared from star-type and dendritic-type urethane monomers bearing three or nine end groups, respectively, comprising either norbornene (NBE) [25, 153–155] or acrylate (ACR) [25, 26, 156] (Fig. 21.57).

The core of these polyfunctional monomers, and therefore the polymeric nodes across the aerogel network, was based on a flexible aliphatic triisocyanate (Desmodur N3300A, denoted with aL-), a rigid aliphatic triisocyante (Desmodur Z4470BA, denoted with IP-), or a rigid aromatic triisocyanate (Desmodur RE, denoted with aR-) (see Fig. 21.57). Figure 21.58 shows typical 9-functional versus trifunctional NBE- and ACR-terminated monomers, exemplified using Desmodur N3300A as the core. Poly(urethane norbornene) aerogels were synthesized via ring-opening metathesis polymerization (ROMP) of NBE-terminated monomers (e.g., Fig. 21.58) employing first-generation Grubbs’ catalyst (GC-I) for the 9-NBE-terminated monomers in toluene [155] and secondgeneration Grubbs’ catalyst (GC-II) in acetone for 3-NBE-terminated monomers [25]. (Polymerization of aL-3-NBE and aR-3-NBE monomers using GC-I in toluene was not possible as these monomers are insoluble in toluene. See ▶ Chap. 23) All aerogels in this study consisted of aggregates of nanoparticles, whose size depended on the aliphatic/aromatic content of the monomer, rigidity/flexibility of the polymeric backbone, but most importantly, the number of terminal

Triisocyanate cores (H2C)6N=C=O O N O

Monomer design

O=C=N(H2C)6 N

N

Desmodur N3300 Aliphatic/flexible (aL)

(CH2)6N=C=O

O

9- vs 3-

N=C=O

NBE or acrylate

Desmodur RE Aromatic/rigid (aR)

Urethane bridge

Triisocyanate core

O=C=N

N=C=O

O=C=N

N=C=O O N O

Desmodur Z4470BA N

N

Aliphatic/rigid (IP) O

N=C=O Fig. 21.57 Design of monomers for polymeric node studies featuring either nine or three dangling norbornene (NBE) or acrylate (ACR) groups along with the three triisocyanate cores considered for those studies and the rationale for their selection

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

Norbornene-terminated

549

Acrylate-terminated

21

aL-9-NBE

aL-9-ACR

aL-3-NBE

aL-3-ACR

Fig. 21.58 Monomers terminated with 9-(dendritic-type) or 3-(star-type) polymerizable groups exemplified using a flexible aliphatic triisocyanate core (Desmodur N3300A)

functional groups of the monomers. Morphology generally varied with density. Typical SEM images from this class of materials were used in Fig. 21.3 as examples of morphology as a function of the gelation process parameters. Focusing on the 9-NBE-terminated monomers, at low densities ( 90% v/v) and were macroporous; aerogels based on the flexible aliphatic core were fragile, while aerogels based on the rigid aromatic core were described as plastic-like and at even lower densities (0.03 g cm3) were described as foamy. At higher densities (0.2–0.7 g cm3), all materials were stiff and strong. At low monomer concentrations, both categories of samples (both aL-9-NBE and aR-9NBE) consisted of discrete primary particles that formed spherical secondary aggregates. At higher monomer concentrations, the structure of these two categories of samples consisted of similarly sized fused particles of the same overall

size as secondary aggregates in the samples prepared from low monomer concentrations. It was concluded that it was neither the aliphatic nor aromatic core that determined the point of phase separation but rather the solubility of the polymeric backbone (polynorbornene) that, in both cases, was the same. Accordingly, it was proposed that the solubility of the growing polymer was low, resulting in early phase separation of a primary particle network that later became coated with unreacted monomers and oligomers, thereby fusing the particles together. Corroborating evidence for the growing polymer solubility hypothesis was provided by a comparison of the material properties of aL-9-NBE and aR-9-NBE aerogels with those of aL-3-NBE and aR-3-NBE aerogels (see Fig. 21.58) [25]. For example, for aerogels produced from sols containing the same or similar monomer concentrations (10–20% w/w), materials from 9-NBE-terminated monomers had higher

550

porosities than those derived from 3-NBE-terminated monomers (68–98% v/v versus 44–89% v/v, respectively) and exhibited significantly higher BET surface areas (54–300 m2 g1) than analogs derived from aL-3-NBE or aR-3-NBE (21–60 m2 g1). This difference in BET surface area is reflective of the smaller particle radii observed for materials derived from 9-NBE-terminated monomers (28 nm) as compared to materials derived from 3-NBE-terminated monomers (118 nm). (For more specific information on the aL-3-NBE aerogels, see Sect. 21.2 and the discussion around Fig. 21.3.) Poly(urethane-acrylate) aerogels were synthesized via free radical polymerization in acetone using two different trifunctional star monomers (aL-3-ACR and aR-3-ACR) [25] and three 9-functional (nonafunctional) dendritic monomers (aL-9-ACR [26], aR-9-ACR [26], and IP-9-ACR [156]). Poly (urethane-acrylate) aerogels from low monomer concentration sols were softer than aerogels made from higher monomer concentration sols, which were more rigid. Aerogels made at various densities with aR-9-ACR as the monomer were mostly macroporous in nature exhibiting porosities >76% v/v that included a small amount of microporosity (4–8% of the total BET surface area was allocated to micropores). Such aerogels were thermally stable to 300  C (by TGA under inert atmosphere) and could be pyrolyzed into primarily microporous carbon aerogels (BET surface areas, 640–740 m2 g1; micropore surface areas, 360–430 m2 g1) with low but satisfactory carbon yields (20–30% w/w). Comparing the effects of acrylate versus norbornene termination for nonafunctional aromatic core monomers, aR-9-ACR aerogels exhibited higher bulk densities (0.041–0.219 vs. 0.032–0.170 g cm3), lower skeletal densities (1.40–1.45 vs. 1.42–1.80 g cm3), higher BET surface areas (311–488 vs. 188–294 m2 g1), lower average pore diameters (8.2–13.4 vs. 11–123 nm), and smaller particle radii (4.5–6.8 vs. 6.0–9.9 nm) than analogous aR-9-NBE aerogels prepared from sols in the same concentration range (1.5–12% w/w). Comparing the effects of acrylate versus norbornene termination for nonafunctional aliphatic core monomers (see Fig. 21.58), aL-9-ACR aerogels shrank less upon drying (16–27 vs. 27–34%), had remarkably lower average pore diameters (10–13 nm vs. 13–88 nm), and smaller particle radii (9.0–12.6 vs. 6.9–28.4 nm) than the corresponding aL-9-NBE aerogels. Finally, comparison of the effects of 9- versus 3-functionality for acrylate-terminated monomers proved particularly interesting. At similar sol concentrations, aR-9ACR aerogels exhibited higher BET surface areas (311 vs. 139 m2 g1) and consisted of particles with smaller radii (6.8 vs. 17 nm) than aR-3-ACR aerogels. However, for the corresponding aerogels derived from monomers with a flexible aliphatic core, microstructural differences were much more dramatic: at the same sol concentration (12.5% w/w), aL-9-ACR aerogels, despite exhibiting a higher bulk density

N. Leventis

a

b

2 mm

2 mm

Fig. 21.59 SEM imaging (25 k magnification) of aL-3-ACR (b) and aL-9-ACR (a) aerogels prepared with equal monomer concentrations in their respective sols (12% w/w). Notice the difference in mean particle size arising from increased monomer functionality while all other factors remain constant: 2.4 μm for aL-3-ACR (b) versus 18 nm for aL-9-ACR (a) as determined from skeletal density and N2-sorption data

(0.310 vs. 0.171 g cm3) and lower porosity (76 vs. 86% v/v) than aR-3-ACR aerogels, had a remarkably higher BET surface area (260 vs. 2 m2 g1) and much smaller particle radii (9 nm vs. 1.2 microns) than analogous aL-3-ACR aerogels (Fig. 21.59). In fact, this was the first example of a polyurethane aerogel based on Desmodur N3300A that comprised nanosized primary particles. In all other cases of polyurethane aerogels where Desmodur N3300A was employed, including superelastic shape-memory aerogels [25, 96, 97, 151, 152], the aerogel’s constituent particles were micron sized, reflective of a different phase separation mechanism (Fig. 21.2, right vs. left branch). These findings were considered as a strong evidence that increasing the number of functional groups on the monomer (i.e., the functional group density of the monomer) increases crosslinking, thereby decreasing solubility of the developing polymer and forming networks comprising smaller particles that exhibit much higher BET surface areas, practically independent of the chemical identity of the monomer. To probe this conjecture further, poly(urethane-acrylate) aerogels were also prepared using a dendritic monomer based on the aliphatic, yet rigid core of Desmodur Z4470BA (IP-9-ACR) (refer to Figs. 21.21 and 21.57 for the molecular structure of this monomer) [156]. Among aerogels derived from nonafunctional acrylate-terminated dendritic monomers [26, 156], aerogels from monomers comprising a rigid triisocyanate core, irrespective of whether the core is aromatic (from Desmodur RE) or aliphatic (from Desmodur Z4470BA), all exhibited similar values for particle radii (4.5–6.8 nm) and BET surface areas (310–490 m2 g1). (For reference, similar aL-9-ACR aerogels based on flexible aliphatic cores exhibited a 2
larger mean particle size of 9.0–12.6 nm and lower surface

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

area values, in the range of 190–260 m2 g1, compared to aR-9-ACR and IP-9-ACR aerogels based on rigid cores [156].) Together, these studies support the conclusion that smaller primary particles and higher surface areas are decisively determined by the degree of crosslinking at the polymer chain level, which in turn arises directly from the functional group density of the monomer, whereas the molecular rigidity versus flexibility of the monomer core and the chemical identity of the polymer chains still play a significant, yet only secondary role. From a practical perspective, not only was mean particle size smaller with higher monomer functionality, but particles also exhibited a higher number of dangling surface acrylates resulting in more efficient crosslinking of adjacent particles. This permits gelation at low monomer concentrations (in the range of 1.5–12% w/w), whereas gelation only occurred at higher monomer concentrations for aerogels based on triacrylate monomers (in the range of 9–40% w/w). Thermal conductivities and mechanical properties have thus far only been measured for aerogels produced using 3-NBE and 3-ACR monomers. The pattern displayed in Fig. 21.60a indicates that aerogels with smaller mean particle sizes exhibit lower overall thermal conductivities. Notably, select low-density aerogels derived from these monomers are highly flexible and foldable (Fig. 21.60b). Elastic moduli were correlated with the C-coefficients for solid thermal conduction (similar to the discussion regarding Fig. 21.48 above), supporting the somewhat intuitive conclusion that flexibility in an aerogel requires lower total contact area between particles (which is reflected by lower C-values).

a

λ (W m–1 K–1)

Mechanically strong, machinable, aromatic polyurethane aerogel panels based on the small-molecule approach (Fig. 21.46) are commercially available under the trade name Airloy X134 [112]. Durable superinsulating aerogel panels, also referred to as polyurethanes, have been developed by BASF for building and construction applications under the name Slentite ® (see ▶ Chap. 64). An emerging application of polyurethane aerogels based on rigid aromatic Desmodur RE and cage-shaped α- or β-cyclodextrin is as desiccants (Fig. 21.61a) [157]. Gelation was carried out in DMF with DBTDL as catalyst (Fig. 21.61b). Wet gels were dried to aerogels (abbreviated as α- or β-CDPU-xx) with SCF CO2. Extension “xx” stands for the percent weight of the total monomers in the sol and was varied at two levels for each cyclodextrin: 2.5% and 15%. All materials had relatively higher capacities for water adsorption from high-humidity environments (99%) than typical commercial desiccants like silica (35–40% w/w) or Drierite (10–14% w/w) [157]. However, α-CDPU-2.5 aerogels did stand out with a water uptake capacity reaching 1 g of H2O per gram of aerogel (Fig. 21.61c). Adsorbed water could be released quantitatively without heating, by just reducing the relative humidity of the environment to 10%. All α- and β-CDPU-xx aerogel samples were cycled between humid and dry environments ten times. Samples settle over the first two cycles, and their moisture uptake was stable afterward (Fig. 21.61c). The moisture uptake was traced to filling smaller mesopores with water and was attributed to a delicate balance of enthalpic (H-bonding) and entropic factors, whereas the latter were a function of pore size (for further discussion of this mechanism, refer to Sect. 21.6.2 about CO2 adsorption from

b

aR-3-ACR aR-3-NBE aL-3-ACR aL-3-NBE

0.12

551

0.09

aR-3-ACR-EG

0.06 0.03 0.2

0.4 Bulk density (g

0.6

0.8

cm–3)

Fig. 21.60 (a) Thermal conductivity versus bulk density of poly(urethane-acrylate) and poly(urethane-norbornene) aerogels prepared from star-shaped monomers with aromatic and aliphatic cores terminated with three functional groups (either acrylate, ACR, or norbornene, NBE, as

indicated). (b) Photograph demonstrating the flexible and foldability of low-density aR-3-ACR-EG aerogels (ρb ¼ 0.14 g cm3) (ethylene glycol, EG, was used as a chain extender in this formulation) [25]. (Adapted from [25], Copyright 2014, The American Chemical Society)

21

552

N. Leventis

a

HO O

OH

OH O

O

OH OH OH O O OH HO

HO

O O O OH OHOHO OH O OH HO O O OH HO HO O OH OH O OH HO OH OH OH O OH O O O HO O OH

OH O

O

HO OH O HO OH O OH OHHO O O OH O HO

HO

 -CD

-CD

b NCO

OH

X

+ HO

O

DBTDL

O

DMF 25 °C

OH O

X

n OCN

X O n

NCO TIPM

 or

-CDPU

n = 6:  -CD X=

n = 7: -CD

O H O C N

Y CH Y

Y

TIPM-derived urethane c

 -CDPU-xx

-CDPU-xx

xx = 2.5 Water uptake (%w/w)

100 80 60

xx = 2.5 xx = 15

40

xx = 15

20 0 Number of cycles (1-10)

Fig. 21.61 (a) α- and β-cyclodextrins. (b) Synthesis of cyclodextrinbased polyurethane aerogels (α- and β-CDPU-xx), from α- and β-cyclodextrins, and TIPM (Desmodur RE). (c) Ten consecutive cycles of dynamic water uptake between a high (99%) and a low (10%) relative

humidity environment by the four aerogels as indicated. The weight of the four aerogels was monitored every 24 h. (Extensions “-xx” refer to the weight percent of total monomers in the sol.). (Adapted from [157], Copyright 2019 The American Chemical Society)

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

carbon aerogels derived from polyamide-polyimide-polyurea copolymers).

21.6

Polyimide Aerogels and Polyamide Aerogels

Polyimide and polyamide aerogels obtained from isocyanates are reviewed together because, from a mechanistic point of view, the processes used to make such materials cross paths frequently, and therefore oftentimes polyamide aerogels also contain polyurea and polyimide.

21.6.1 Polyimide Aerogels The first reported synthesis of polyimide aerogels via the isocyanate route included a comparison of such materials with polyimide aerogels obtained via the classic DuPont route (i.e., chemical dehydration of a polyamic acid intermediate with acetic anhydride/pyridine). Both materials were prepared using pyromellitic dianhydride (PMDA), which was reacted with either methylene diisocyanate (MDI) or methylene dianiline (MDA) [73]. Figure 21.62 compares the two pathways and includes the respective key

intermediates: a seven-member ring in the isocyanate route and polyamic acid in the DuPont route. The isocyanate route (ISO) for synthesizing polyimides (PI) at room temperature was evaluated using a mixture of Nmethyl-2-pyrrolidone (NMP) and acetonitrile (3:1 w/w); the amine route (AMN) and the ISO route at elevated temperatures were evaluated using straight NMP. The resulting materials were abbreviated as PI-ISO and PI-AMN, respectively. To complete imidization of the PI-AMN samples, wet gels were heated at 190  C, and samples were referred to as PI-AMN-190. Aerogels from the isocyanate route processed at room temperature are referred to as PI-ISO-RT; those processed at 90  C are referred to as PI-ISO-90. Figure 21.63 compares the solid-state 13C NMR spectra of PI aerogels produced through the two routes. It was reported that simulations showed that C-16 moves upfield as the polymeric chain length increases; therefore it was concluded that PI-ISO-RT consisted of longer polymeric chains. The spectra of PI-ISO-90 and PI-AMN-190 (both prepared in NMP) were practically identical, signifying the importance of solvent polarity in polymerization. In terms of material properties, the PI-AMN sols reached a solubility limit around 20% w/w of solids resulting in aerogel bulk densities in the range of 0.090–0.376 g cm3. PI-ISO sols could be prepared with concentrations containing up to 50% w/w solids and produced gels exhibiting less shrinkage,

PI-ISO

PI-AMN

O

O

PMDA

O +

O OCN

O

O

O

O +

O

553

H2N

NCO

O

MDI

O

PMDA

O

MDA

Room temperature O

R

O

N

Room temperature O

R N

O O O

O

N H HOOC

O O

NH2

N H COOH

O

H2

C

Polyamic acid –2n CO2

O N O

1. AcOAc, pyridine 2. 190 °C O 15 16 19 19 H 17 N 18 20 C2 9 O

n

Fig. 21.62 Comparison of two different synthetic routes for synthesizing the same polyimide: the PI-ISO route via a diisocyanate and the PI-AMN route via a diamine [73] (Atom labeling refers to the solid-state 13C NMR spectra shown in Fig. 21.63)

21

554

N. Leventis

PI-ISO-90

PI-ISO-RT

19

18 17 16

200

150

PI-AMN-190 20

15

100

9

50

0

d (ppm) Fig. 21.63 Solid-state 13C-NMR of polyimide aerogels prepared through the isocyanate route and the amine route. Samples were prepared from sols containing a solid concentration of 15% w/w. PI-AMN-190 and PI-ISO-90 were prepared in NMP; PI-ISO-RT was prepared in NMP/CH3CN (3:1 w/w). (For peak assignments, see Fig. 21.62.). (Reprinted from [73], Copyright 2010 The Royal Society of Chemistry)

resulting in aerogels with bulk densities ranging from 0.047 to 0.679 g cm3 and porosities ranging from 53% v/v to 97 v/v %. Irrespective of the synthetic route (PI-ISO versus PI-AMN), BET surface areas of aerogels in the same density range (0.1–0.4 g cm3) were between 200 and 400 m2 g1 for both routes, with the PI-AMN-190 aerogels having slightly higher surface areas. Under low-magnification SEM (Fig. 21.64), PI-AMN-190 exhibited a particulate morphology, whereas PI-ISO-RT exhibited a fibrilar morphology. At a higher magnification, the latter aerogels appeared to consist of entangled worm-like building blocks similar to those having been observed in both vanadia aerogels [158] and polyurea aerogels with a K-index of 1.2 (see Fig. 21.36). At high resolution, PI-AMN-190 consisted of particle aggregates that formed lowaspect-ratio strings of beads. Interestingly, SANS showed that both PI-AMN-190 and PI-ISO-RT, despite their apparent differences when viewed under SEM, consisted of similarly sized primary and secondary particles (7.53 and 45.4 nm for PI-AMN-190 and 6.10 and 54.0 nm for PI-ISO-RT,

respectively). The fact that similarly sized particle assemblies take two different routes to higher-level aggregation (as noted in Fig. 21.64) was attributed to the difference in rigidity between the two intermediates, although the solvent polarity (NMP in the case of PI-AMN-190 and a mixture of NMP/CH3CN in the case of PI-ISO-RT) likely also played a role in the same vein as the discussion regarding the K-index for polyurea aerogels presented earlier in this chapter (Fig. 21.36). Both types of polyimide aerogels could be pyrolyzed into carbon aerogels and provided similar carbon yields (54–58% w/w). The fibrous morphology of the PI-ISO samples was more or less preserved through carbonization, while PI-AMN showed pronounced evidence of melting effects during pyrolysis (Fig. 21.64). These morphological differences were reflected in the BET surface areas of the resulting carbon aerogels. For example, carbon aerogels produced from PI-AMN-190, which in turn were derived from sols containing 15% w/w solids, had a surface area of 113 m2 g1, whereas carbon aerogels produced from PI-ISO-RT coming from sols containing 10% w/w solids had surface areas at 361 m2 g1. Introduction of micropores via postpyrolysis activation with CO2 increased the surface area of the former carbon aerogels to 417 m2 g1 and of the latter carbon aerogels to 1010 m2 g1 [73]. Work surrounding isocyanate-derived polyimides was then extended to include triisocyanates (Desmodur RE, Fig. 21.38, and Desmodur N3300A, Fig. 21.21) as well as a less rigid dianhydride, benzophenone-tetracarboxylic dianhydride (BTDA) [75, 159]. The nanomorphology of the resulting PI-ISO aerogels changed completely from fibrilar to particulate throughout, for both anhydrides and for both triisocyanates. For the same range of sol concentrations (6–20% w/w) aromatic triisocyanate-based polyimide aerogels made with PMDA shrunk more (35–51% in linear dimensions) than those made with BTDA (25–47%) and were denser materials (0.44–0.72 g cm3) than those prepared with BTDA (0.26–0.43 g cm3). However, BET surface areas followed an opposite trend, with PMDA samples exhibiting a surface area of 328–427 m2 g1 versus 146–209 m2 g1 with BTDA. In both materials, about 5–10% of the BET surface area could be attributed to micropores. Microporosity was probed using N2-sorption porosimetry under low-pressure dosing. Pores were “visualized” with fully atomistic molecular dynamic simulations (Fig. 21.65) using (a) the simulated versus the experimental XRD patterns for guiding the simulation toward the correct stacking and packing of oligomers in the computational polyimide particles and (b) the calculated versus the experimental skeletal density of the particles for deciding the arresting point for the simulation. A good agreement was found between micropore sizes determined from N2-sorption analysis and simulations (see legend of Fig. 21.65).

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

Polyimide aerogels

555

Carbon aerogels

PI-AMN-190

21

200 nm

1 mm

200 nm

1 mm

200 nm

PI-ISO-RT

1 mm

1 mm

200 nm

Fig. 21.64 SEM of polyimide aerogels produced through the isocyanate route (ISO) and the amine route (AMN) along with the corresponding carbon aerogels, as indicated. Both types of aerogels were prepared from sols containing solids concentrations of 15% w/w. Each sample is shown at two different magnifications: low (left)

and high (right). Bulk densities of the parent polyimide aerogels were 0.23 g cm3 for PI-AMN-190 and 0.12 g cm3 for PI-ISO-RT. Bulk densities of the corresponding carbon aerogels were 1.018 g cm3 and 0.967 g cm3, respectively. (Adapted from [73], Copyright 2010 The Royal Society of Chemistry)

Consistent with all data reviewed so far, polyimide aerogels based on the flexible aliphatic monomer Desmodur N3300A shrunk more and were thus denser than those prepared with the rigid aromatic monomer Desmodur RE. The BET surface area of aerogels based on Desmodur N3300A and 20% w/w sol concentration using PMDA as the dianhydride was only 16 m2 g1; using BTDA as the dianhydride, the BET surface area was still low (53 m2 g1). All aromatic Desmodur RE-based polyimide aerogels regardless of being made with PMDA or BTDA were carbonizable, providing carbon yields in the range of 52–59% w/w. Porosities of carbon aerogels from PMDAbased polyimide aerogels were in the range of 32–63% v/v; porosities of carbon aerogels from BTDA-based polyimide aerogels were in the range of 70–79% v/v. The BET surface areas of polyimide-derived carbon aerogels followed an opposite trend from the trend seen for the BET surface areas of the parent polyimide aerogels: surface area for carbon aerogels from PMDA-derived polyimide aerogels was in the range of 113–533 m2 g1, while those from BTDA were

in the range of 147–253 m2 g1. In most carbon aerogel samples, the BET surface area included a 15–20% of area attributable to micropores. Along with polyurea and polyurethane aerogels, a product based on a triisocyanate-derived polyimide aerogel is commercially available under the trade name Airloy X114 for low-volume high-specialization applications [112]. The product is stated to be mechanically durable and nonflammable with a thermal conductivity of 0.020 W m1 K1 at room temperature.

21.6.2 Polyamide Aerogels Polyamide aerogels derived from isocyanates were reported almost concurrently with isocyanate-derived polyimide aerogels. In accordance with the reaction scheme described in section “Reaction of –N¼C¼O with Carboxylic Acids: Synthesis of Polyamide Aerogels,” the reaction of Desmodur RE and trimesic acid (TMA) yielded a polyamide gel

556

N. Leventis

a

b

c

XRD of aR-PMDA-6 Intensity (a.u.)

2nd generation: pack-4 (via simulation) 10.2 Å 10.9 Å

20

40 60 Deg. (2)

80

XRD of aR-BTDA-6 Intensity (a.u.)

2nd generation: pack-8 (via simulation)

4.1 Å

5.0 Å 20

40 60 Deg. (2)

80

Fig. 21.65 (a) Best match of simulated XRD patterns with experimental data as indicated. (“Generation” refers to the growth pattern/size of the hyperbranched oligomers out of the triisocyanate core. “Pack-4” and “pack-8” refer to the number of hyperbranched oligomers introduced in the molecular dynamics simulations.) (b) Structures corresponding to the simulated XRD patterns on the left. (c) Magnification of the voids

enclosed by dashed ovals in the middle frame. For comparison, assuming cylindrical pores, the experimental micropore width of PMDAbased polyimides was calculated to be around 10 Å; assuming slitshaped pores, the experimental micropore width of BTDA-based polyimides was calculated to be around 6 Å. (Reprinted from [75], Copyright 2013 The Royal Society of Chemistry)

network (Fig. 21.66a) [160]. The gelation process was performed in DMF at different temperatures. As shown in Fig. 21.66b, at lower gelation temperatures, the network included a significant amount of polyurea, the content of which decreased as the gelation temperature increased. This led to the conclusion that the polymerization process into a polyamide network included pathway c of Fig. 21.16. The degree of linear shrinkage measured for isocyanate-derived polyamide aerogels trended inversely to the concentration of monomers used in the sol, decreasing from 41 to 11% over the range of 5 to 25% w/w of monomers in the sol. The bulk density of isocyanate-derived polyamides ranged from 0.2 to 0.4 g cm3 corresponding to porosities of 84% to 69% v/v, respectively. The aerogel nanomorphology was particulate. BET surface areas ranged from 380 m2 g1 for 0.2 g cm3 samples to 155 m2 g1 for 0.4 g cm3 samples. Thermal

conductivity at a bulk density of 0.280  0.009 g cm3 was measured as 0.028  0.002 W m1 K1. This line of research was extended to include polyamide aerogels derived from the reaction of the rigid aromatic monomer Desmodur RE with two other carboxylic acids: pyromellitic acid (PMA), a tetrafunctional carboxylic acid, and ferrocene dicarboxylic acid (Fig. 21.67). Both carboxylic acids were introduced with an eye toward improving the properties of carbon aerogels produced from isocyanatederived polyamide aerogels. The rationale for carbon aerogels prepared from polyamides derived from PMA was that additional rigidization brought about by the higher functionality of the carboxylic acid and the introduction of O and N heteroatoms could increase the carbonization yield, modify the pore structure, and line the internal surfaces with functional groups that would facilitate

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

557

TMA a

NCO

NHCO COOH 90 °C, in anh. DMF

+ HOOC COOH OCN NCO Desmodur RE (TIPM) Trimesic acid (TMA)

OCHN

(–CO2 ) HNOC

CONH

TIPM

TIPM

NHCO TMA

b Urea C=O Amide

C=O 25 °C

90 °C

135 °C

180

160

140 120 d (ppm)

100

80

Fig. 21.66 (a) The reaction of Desmodur RE with trimesic acid to form a polyamide network. (b) Solid-state 13C NMR spectra of polyamide aerogels prepared via the reaction of Part (a) with 15% w/w of total

monomers in DMF at the three temperatures indicated. (The amount of polyurea side product decreased with increasing gelation temperature.). (Adapted from [160], Copyright 2011 The Royal Society of Chemistry)

adsorption. Carbonization of polyamide aerogels prepared with ferrocene dicarboxylic acid would result in carbon aerogels doped with metallic iron suitable for applications in catalysis. In the case of the reaction of PMA with Desmodur RE [161], solid-state 13C NMR suggested (Fig. 21.68a) and

solid-state 15N NMR confirmed (Fig. 21.68b) that the skeletal framework of as-prepared polyamide aerogels (referred to as PA-15, whereas the numerical extension designated the weight percent of monomers in the sol) was a statistical copolymer of polyamide, polyurea, polyimide, and the primary condensation product of the two reactants, a

21

558

N. Leventis

carbamicanhydride adduct. This latter adduct decomposed upon heating of the dried aerogels at 150  C, giving off CO2 (which was captured with an aqueous solution of Ca(OH)2 and quantified) to yield aerogels consisting of a combination of polyamide, polyurea, and polyimide [161, 162]. This compositional blend was not anticipated based on the previous aerogels made with trimesic acid (see above) [160, 163] nor on patent literature describing the synthesis of polyimide aerogels via reaction of polyisocyanates with polycarboxylic acids [164]. The formation of this polymeric blend was attributed in part to the fact that the –COOH groups of PMA are in a 1,2-position relative to one another, and thus anhydride intermediates formed via pathway c in Fig. 21.16 could imidize easily [161].

HOOC

COOH

HOOC

COOH

COOH Fe

Pyromellitic acid (PMA)

COOH Ferrocene dicarboxylic acid

Fig. 21.67 The structures of pyromellitic acid (PMA) and ferrocene dicarboxylic acid

a

b

154.5 156.3

104.5

165.0

PA-15 Urea C=O

PA-15 Heated at 150 °C

102.3

–

Residual DMF

Polyurea

Imide C=O

170.8

54.4

132.8

PA-15 As-prepared

105.3

–

Polyurea

Polyimide

c

b

DMSO-d6

a 3 5 2

200

Stepwise pyrolytic decomposition of the polyimide, polyamide, and polyurea components yielded carbon aerogels with both open and closed microporosity [161]. The open micropore surface area increased from 70%. Conceptual comparisons were made regarding the design of these catalysts and the structure of catalytic converters, although it was recognized that for high-temperature applications, the carbon-aerogel support should be replaced with a more oxidation-stable ceramic aerogel such as silicon carbide [1]. It was suggested that this approach for the design of catalytic systems (support plus catalyst) could be expanded to other metallocenes

Catalytic monolith

Active catalyst

5× Redeploy

Reactants

≥ 70% yield per cycle

Products

Fig. 21.72 The use of metal-doped graphitic aerogel monoliths as redeployable self-supporting catalysts. The catalytic monolith can be easily removed from solution using a pair of tweezers and redeployed in a fresh reaction mixture where it can continue its catalytic role [166]

noting, however, that such an approach might be less attractive for expensive metals. This said, transmetalation via galvanic replacement is atom-efficient; can be performed at room temperature, thus eliminating the risk of thermally induced particle coarsening; and results in an effective reduction in metal particle size – a desirable process feature that enhances the catalytic activity of such particles.

21.7

Conclusions

Over the last 10 years, isocyanate-derived aerogels have proven to be a versatile and important class of aerogel materials of significance to both fundamental studies of nanostructured matter and practical applications. In terms of fundamentals, the extremely broad range of nanomorphologies attainable by isocyanate-derived aerogels combined with the compositional and functional versatility exhibited by this class of materials has provided a powerful

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

new platform for studying the synthesis and behavior of nanostructured matter in the spirit of Kistler’s invention and early work. Questions surrounding how monomer size, flexibility, and functional group density translate to nanoscale structures, and thus macroscale properties, have been addressed conclusively. From a material design perspective, the functional group density of monomers was found to be a controlling factor in determining primary particle size and thus all material properties that depend on primary particle size. Smaller particles are favored by small-molecule monomers featuring more functional groups per molecule, which translates into higher surface area aerogels, with a higher degree of interparticle connectivity that in turn results in higher mechanical strength and lower thermal conductivity. The extremely broad range of nanomorphologies that can be taken on by a single chemical composition of polyurea inspired the possibility that a complex nanomorphology can be expressed quantitatively, raising the question of how such geometric complexity could be designed a priori. From this came the development of the K-index, which successfully addressed both of these questions. Isocyanate-derived aerogels have also found new applications for frequently overlooked reactions of isocyanates, including the formation of polyimides from the reaction of isocyanates with carboxylic acid anhydrides and the formation of polyamides from the reaction of isocyanates with carboxylic acids. These materials served as the impetus for new chemistry, as seen in the studies of isocyanates with mineral acids and with polyfunctional carboxylic acids as described in this chapter. The importance of solid-state 15N NMR in such studies cannot be over emphasized, and the reader is encouraged to consider its use in future studies. In addition to the time-tested application of aerogels as thermal superinsulation, the impressive mechanical properties of isocyanate-derived aerogels suggest that such materials have a promising future in applications that range from ballistic protection, blast wave mitigation, and acoustic insulation to shape-memory deployable panels for space exploration, biomedicine (e.g., orthopedic casts), and so on. Functional carbon aerogels produced from isocyanate-derived aerogels facilitate entry of this type of aerogels into the energy cycle in the form of electrodes for electrochemical energy storage devices (batteries, capacitors, fuel cells), as media for pre- and postcombustion sequestration of CO2 in power plants and as catalyst supports. When suitably activated and doped, carbon aerogel monoliths inherently resolve the problems of how to package and engage microporous materials for 3D service in an easy-to-use and reusable form factor in the spirit, for example, of catalytic converters. Isocyanate-derived aerogels possess all the material properties – chemical, physical, and morphological – to enable these roles. Owing to their versatility and desirable combination of material properties, isocyanate-derived aerogels have already

561

entered into various stages of commercial production. The future of these materials in academia and industry is undoubtedly bright, as advanced isocyanate-derived aerogels are well poised to address and resolve many of society’s current and emerging challenges related to energy, healthcare, and the environment. Acknowledgments This work would not have been possible without the contribution and hard work of many talented and dedicated postdoctoral, graduate, and undergraduate students at MS&T over the last 10 years. The author thanks the US Army Research Office (W911NF-14-10369; W911NF12-2-0029; W911NF-10-1-0476), the National Science Foundation (CHE-0809562 and CMMI-0653919), and BASF Polyurethanes GmbH for financial support. Special thanks go also to Covestro, LLC (formerly Bayer MaterialScience, USA), for their generous supply of isocyanates to all those who have worked with isocyanate-derived aerogels.

References 1. See Supporting Information for Leventis N, Sadekar, A., Chandrasekaran, N., Sotiriou-Leventis, C.: Click synthesis of monolithic silicon carbide aerogels from polyacrylonitrilecrosslinked 3D silica networks. Chem. Mater. 22, 2790–2803 (2010) 2. Vareda, J.P., Lamy-Mendes, A., Durães, L.: A reconsideration on the definition of the term aerogel based on current drying trends. Microporous Mesoporous Mater. 258, 211–216 (2018) 3. Kistler, S.: Coherent expanded aerogels and jellies. Nature. 127, 741 (1931) 4. Painter, P.C., Coleman, M.: Fundamentals to Polymer Science, An Introductory Text, 2nd edn, p. 39. CRC Press LLC, Boca Raton, FL (1997) 5. Pierre, A.C., Pajonk, G.M.: Chemistry of aerogels and their applications. Chem. Rev. 102, 4243–4266 (2002) 6. Long, J.W., Swider-Lyons, K.E., Stroud, R.M., Rolison, D.R.: Design of pore and matter architectures in manganese oxide charge-storage materials. Electrochem. Solid-State Lett. 3, 453–456 (2000) 7. Mansour, A.N., Smith, P.H., Baker, W.M., Balasubramanian, M., McBreen, J.: A comparative in situ X-ray absorption spectroscopy study of nanophase V2O5 aerogel and ambigel cathodes. J. Electrochem. Soc. 150, A403–A413 (2003) 8. Mandal, C., Donthula, S., Rewatkar, P.M., Sotiriou-Leventis, C., Leventis, N.: Experimental deconvolution of depressurization from capillary shrinkage during drying of silica wet gels with SCF CO2why aerogels shrink? J. Sol-Gel Sci. Technol. 92, 662–680 (2019) 9. Kistler, S.: Coherent expanded aerogels. J. Phys. Chem. 36, 52–64 (1932) 10. Biesmans, G., Randall, D., Francais, E., Perrut, M.: Polyurethanebased organic aerogels’ thermal performance. J. Non-Cryst. Solids. 225, 36–40 (1998) 11. Rhine, W., Wang, J., Begag, R.: Polyimide aerogels, carbon aerogels, and metal carbide aerogels and methods of making same. U.S. Patent No 7,074,880. (2006) 12. Lee, J.K., Gould, G.L.: Polydicyclopentadiene based aerogel: a new insulation material. J. Sol-Gel Sci. Technol. 44, 29–40 (2007) 13. Lee, J.K., Gould, G.L., Rhine, W.: Polyurea based aerogel for a high performance thermal insulation material. J. Sol-Gel Sci. Technol. 49, 209–220 (2009)

21

562 14. Lorjai, P., Chaisuwan, T., Wongkasemjit, S.: Porous structure of polybenzoxazine-based organic aerogel prepared by sol-gel process and their carbon aerogels. J. Sol-Gel Sci. Technol. 52, 56–64 (2009) 15. An, H., Wang, Y., Wang, X., Zheng, L., Wang, X., Yi, L., Bai, L., Zhang, X.: Polypyrrole/carbon aerogel composite materials for supercapacitor. J. Power Sources. 195, 6964–6969 (2010) 16. Leventis, N.: Three dimensional Core-Shell superstructures: mechanically strong aerogels. Acc. Chem. Res. 40, 874–884 (2007) 17. De Vos, R., Biesmans, G.L.J.G.: Organic aerogels. U.S. Patent No 5,484,818. (1994) 18. Zhang, G., Dass, A., Rawashdeh A-M, M., Thomas, J., Counsil, J. A., Sotiriou-Leventis, C., Fabrizio, E.F., Ilhan, F., Vassilaras, P., Scheiman, D.A., McCorkle, L., Palczer, A., Johnston, J.C., Meador, M.A.B., Leventis, N.: Isocyanate cross-linked silica aerogel monoliths: preparation and characterization. J. Non-Cryst. Solids. 350, 152–164 (2004) 19. Leventis, N., Mulik, N.S., Sotiriou-Leventis, C.: Macroporous electrically conducting carbon networks by pyrolysis of isocyanate cross-linked resorcinol-formaldehyde aerogels. Chem. Mater. 20, 6985–6997 (2008) 20. Leventis, N., Sotiriou-Leventis, C., Chandrasekaran, N., Mulik, S., Larimore, Z.L., Lu, H., Churu, G., Mang, J.T.: Multifunctional polyurea aerogels from isocyanates and water. A structure-property case study. Chem. Mater. 22, 6692–6710 (2010) 21. Leventis, N., Chidambareswarapattar, C., Bang, A., SotiriouLeventis, C.: Cocoon-in-web-like superhydrophobic aerogels from hydrophilic polyurea and use in environmental remediation. ACS Appl. Mater. Interfaces. 6, 6872–6882 (2014) 22. Trappe, V., Prasad, V., Cipelletti, L., Segre, P.N., Weitz, D.A.: Jamming phase diagrams for attractive particles. Nature. 411, 772–775 (2001) 23. Lu, P.J., Zaccarelli, E., Ciulla, F., Schofield, A.B., Sciortino, F., Weitz, D.A.: Gelation of particles with short-range attraction. Nature. 499, 453–503 (2008) 24. Rouwhorst, J., Ness, C., Stoyanov, S., Zaccone, A., Schall, P.: Nonequilibrium continuous phase transition in colloidal gelation with short-range attraction. Nat. Commun. 11, 3558 (2020) 25. Bang, A., Buback, C., Sotiriou-Leventis, C., Leventis, N.: Flexible aerogels from hyperbranched polyurethanes: probing the role of molecular rigidity with poly(urethane acrylates) versus poly(urethane norbornenes). Chem. Mater. 26, 6979–6993 (2014) 26. Papastergiou, M., Kanellou, A., Chriti, D., Raptopoulos, G., Paraskevopoulou, P.: Poly(urethane-acrylate) aerogels via radical polymerization of dendritic urethane-acrylate monomers. Materials. 11, 2249 (2018) 27. Mohite, D.P., Larimore, Z.J., Lu, H., Mang, J.T., Sotiriou-Leventis, C., Leventis, N.: Monolithic hierarchical fractal assemblies of silica nanoparticles cross-linked with polynorbornene via ROMP: a structure-property correlation from molecular to bulk through nano. Chem. Mater. 24, 3434–3448 (2012) 28. Mohite, D.P., Mahadik-Khanolkar, S., Luo, H., Lu, H., SotiriouLeventis, C., Leventis, N.: Polydicyclopentadiene aerogels grafted with PMMA: II. Nanoscopic characterization and origin of macroscopic deformation. Soft Matter. 9, 1531–1539 (2013) 29. Chidambareswarapattar, C., McCarver, P.M., Luo, H., Lu, H., Sotiriou-Leventis, C., Leventis, N.: Fractal multiscale Nanoporous polyurethanes: flexible to extremely rigid aerogels from multifunctional small molecules. Chem. Mater. 25, 3205–3224 (2013) 30. Lee, S.T., Ramesh, N.S.: Polymeric Foams: Mechanism and Materials, pp. 150–267. CRC Press, Florida (2004) 31. Saunders, J.H., Frisch, K.C.: Polyurethane Chemistry and Technology: Chemistry, pp. 63–118. Interscience Publishers (1964) 32. Stevens, M.P.: Polymer Chemistry. An Introduction. Oxford University Press, New York (1990) 33. Hepburn, C.: Polyurethane Elastomers, 2nd edn. Elsevier Science Publishing Co, Inc., New York (1992) Chapter 1

N. Leventis 34. Davis, T.L., Ebersole, F.: Relative velocities of reaction of amines with phenyl isocyanate. J. Am. Chem. Soc. 56, 885–886 (1934) 35. Neumann, W., Fischer, P.: Carbodiimide aus isocyanaten. Angew. Chem. 74, 801–806 (1962) 36. https://coatings.specialchem.com/product/r-covestro-desmodurn-3200 37. Wicks Jr., Z.W., Jones, F.N., Pappas, S.P., Wicks, D.A.: Organic Coatings, Science & Technology, 3rd edn. Wiley-Interscience, Hoboken (2007) 38. Odian, G.: Principle of Polymerization. Wiley-Interscience, New York (2004) 39. Dodge, J.: Polyurethanes and Polyureas. In: Rogers, M.E., Long, T.E. (eds.) Synthetic Methods in Step-Growth Polymers, pp. 197–263. Wiley, New York (2003) 40. Bayer, O.: Das Di-Isocyanat-polyadditionsverfahren (polyurethane). Angew. Chem. A59, 257–272 (1947) 41. Baker, J.W., Bailey, D.N.: 935. The mechanism of the reaction of aryl isocyanates with alcohols and amines. Part VII. The “spontaneous” reaction with amines. J. Chem. Soc., 4652–4662 (1957) 42. Baker, J.W., Davies, M.M., Gaunt, J.: The mechanism of the reaction of aryl isocyanates with alcohols and amines. Part IV. The evidence of infra-red absorption spectra regarding alcohol–amine association in the base-catalysed reaction of phenyl isocyanate with alcohols. J. Chem. Soc., 24–27 (1949) 43. Baker, J.W., Holdsworth, B.: Mechanism of aromatic side-chain reactions with special reference to the polar effects of substituents. Part XVI. Hyperconjugation of groups of the type CH2X in the benzaldehyde–cyanohydrin equilibria. J. Chem. Soc., 713–726 (1947) 44. Farkas, A., Flynn, K.G.: The catalytic effects of 1,4-diaza[2.2.2] bicycloöctane for isocyanate reactions. J. Am. Chem. Soc. 82, 642–645 (1960) 45. Farkas, A., Mills, G.A., Erner, W.E., Maerker, J.B.: Triethylenediamine – a new bicyclic intermediate and catalyst for making polyurethane foam. Ind. Eng. Chem. 51, 1299–1300 (1959) 46. Schwetlick, K., Noack, R., Stebner, F.: Three fundamental mechanisms of base-catalysed reactions of isocyanates with hydrogenacidic compounds. J. Chem. Soc. Perkin Trans. 2, 599–608 (1994) 47. Britain, J.W., Gemeinhardt, P.G.: Catalysis of the isocyanatehydroxyl reaction. J. Appl. Polym. Sci. 4, 207–211 (1960) 48. Smith, H.A.: Catalysis of the formation of urethanes. J. Appl. Polym. Sci. 7, 85–95 (1963) 49. Borkent, G., Van Aartsen, J.J.: Kinetics and mechanism of urethane formation in DMF: the reaction of 4,4′-diphenylmethane diisocyanate and alcohols catalyzed by dibutyltin dilaurate. In: Platzer, N.A.J. (ed.) Polymerization Kinetics and Technology. Advances in Chemistry Series, vol. 128, pp. 274–280. American Chemical Society, Washington, DC (1973) 50. Houghton, R.P., Mulvaney, A.W.: Mechanism of tin(IV)-catalysed urethane formation. J. Organomet. Chem. 518, 21–27 (1996) 51. Delebecq, E., Pascault, J.-P., Boutevin, B., Ganachaud, F.: On the versatility of urethane/urea bonds: reversibility, blocked isocyanate, and non-isocyanate polyurethane. Chem. Rev. 113, 80–118 (2013) 52. Huguet, M.L., Perez, J.G., Torres Cano, E., Salhi, F., Turon, S.B., Sakalyte, A.: Organic aerogels based on isocyanate and cyclic ether polymer networks. International Patent Publication WO 2017/ 016755 Al. (2017) 53. Joullie, M.M., Lassen, K.M.: Evolution of amide bond formation. ARKIVOC. VIII, 189–250 (2010) 54. Furukawa, I., Hashiyama, M., Hashimoto, S.: Synthesis of polyamides with [(1,2,3-Benzotriazine-4-one)-3-Yl] diphenyl phosphate as a new activating agent. J. Macromol. Sci–Chem. A26, 761–771 (1985) 55. Onder, K., Chen, A.T.: A new semicrystalline reinforced polyamide engineering thermoplastic. Polym. Eng. Sci. 25, 942–946 (1985)

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

56. Simionescu, C., Comaniţǎ, E., Vâcţǎ, M.: Amides and polyamides of m-phenylenedioxydiacetic acid. Angew. Makromol. Chem. 46, 135–142 (1975) 57. Agre, C.L., Dinga, G., Pflaum, R.: Organic anti-foaming agents. J. Org. Chem. 20, 695–699 (1955) 58. Xiao, H., Xiao, H.X., Frisch, K.C., Malwitz, N.: Kinetic studies of the reactions between isocyanates and carboxylic acids. High Perform. Polym. 6, 235–239 (1994) 59. Onder, K., Smith, C.P.: Catalytic condensation of isocyanates and carboxylic acids or anhydrides. U. S. Patent No 4,156,065. (1979) 60. Toyoda, T., Kawamata, M., Osawa, M., Itoh, M., Mizutani, K.: Process for the preparation of polyamide compounds by the reaction of polycarboxylic acids with diisocyanates in the presence of at least one mono-alkali metal salt of dicarboxylic acid. U. S. Patent No 4,395,531. (1983) 61. Wei, Y., Jia, X., Jin, D., Davis, F.A.: A facile synthesis of polyamides from aromatic diisocyanates and dicarboxylic acid catalyzed by Lewis acids. Macromol. Rapid Commun. 17, 897–903 (1996) 62. Jones, G.O., Li, X., Hayden, A.E., Houk, K.N., Danishefsky, S.J.: The coupling of isonitriles and carboxylic acids occurring by sequential concerted rearrangement mechanisms. Org. Lett. 10, 4093–4096 (2008) 63. Naegeli, C., Tyabji, A.: Über den Umsatz aromatischer Isocyansäure-ester mit organischen Säuren. I. Theorie und Anwendung der Reaktion fÜr die präparative Darstellung von Säure-anhydriden. Helv. Chim. Acta. 17, 931–957 (1934) 64. Naegeli, C., Tyabji, A.: Über den Umsatz aromatischer Isocyansäure-ester mit organischen Säuren II. Isolierung einiger Carbaminsäure-carbonsäure-anhydride. Helv. Chim. Acta. 18, 142–160 (1935) 65. Sorenson, W.R.: Reaction of an isocyanate and a carboxylic acid in dimethyl sulfoxide. J. Org. Chem. 24, 978–980 (1959) 66. Leventis, N., Sotiriou-Leventis, C., Saeed, A.M., Donthula, S., Majedi Far, H., Rewatkar, P.M., Kaiser, H., Robertson, J.D., Lu, H., Churu, G.: Nanoporous polyurea from a triisocyanate and boric acid: a paradigm of a general reaction pathway for isocyanates and mineral acids. Chem. Mater. 28, 67–78 (2016) 67. Aries, R.S.: Polymers from boric acid and organic diisocyanates. U.S. Patent No 2,945,841. (1960) 68. Kawagishi, K., Saito, H., Furukawa, H., Horie, K.: Superior nanoporous polyimides via supercritical CO2 drying of junglegym-type polyimide gels. Macromol. Rapid Commun. 28, 96–100 (2007) 69. Meador, M.A.B., Malow, E.J., He, Z.J., McCorkle, L., Guo, H., Nguyen, B.N.: Synthesis and properties of nanoporous polyimide aerogels having a covalently bonded network structure. Polym. Prepr. (Am. Chem. Soc. Div. Polym. Chem.). 51, 265–266 (2010) 70. Edwards, W.M., Robinson, I.M.: Polyimides of pyromellitic acid. U.S. Patent No 2,710,853. (1955) 71. Sroog, C.E.: Polyimides. Prog. Polym. Sci. 16, 561–694 (1991) 72. Leventis, N., Sotiriou-Leventis, C., Mohite, D.P., Larimore, Z.J., Mang, J.T., Churu, G., Lu, H.: Polyimide aerogels by ring opening metathesis polymerization (ROMP). Chem. Mater. 23, 2250–2261 (2011) 73. Chidambareswarapattar, C., Larimore, Z., Sotiriou-Leventis, C., Mang, J.T., Leventis, N.: One-step room-temperature synthesis of fibrous polyimide aerogels from anhydrides and isocyanates and conversion to isomorphic carbons. J. Mater. Chem. 20, 9666–9678 (2010) 74. Chidambareswarapattar, C., Sotiriou-Leventis, C., Leventis, N.: One-step polyimide aerogels from anhydrides and isocyanates. Polym. Prepr. 51, 638–639 (2010) 75. Chidambareswarapattar, C., Xu, L., Sotiriou-Leventis, C., Leventis, N.: Robust monolithic multiscale nanoporous polyimides and conversion to isomorphic carbons. RSC Adv. 3, 26459–26469 (2013)

563

76. Saeed, A.M., Rewatkar, P.M., Majedi Far, H., Taghvaee, T., Donthula, S., Mandal, C., Sotiriou-Leventis, C., Leventis, N.: Selective CO2 sequestration with monolithic bimodal micro/ macroporous carbon aerogels derived from stepwise pyrolytic decomposition of polyamide-polyimide-polyurea random co-polymers. ACS Appl. Mater. Interfaces. 9, 13520–13536 (2017) 77. Wurtz, M.A.: Memoire sur les ethers cyaniques et cyanuriques, et sur la constitution des amides. Ann. Chim. Phys. 42, 43–70 (1854) 78. Hurd, C.D., Prapas, A.G.: Preparation of acyclic imides. J. Org. Chem. 24, 388–392 (1959) 79. Ötvös, L., Marton, J., Meisel-Agoston, J.: Investigation of the reaction between isocyanates and acid anhydrides with 14 C. Tetrahedron Lett. 1, 15–18 (1960) 80. Meyers, R.A.: The polymerization of pyromellitic dianhydride with diphenylmethane diisocyanate. J. Polym. Sci. Part A-1 Polym. Chem. 7, 2757–2762 (1969) 81. Carleton, P.S., Farrissey Jr., W.J., Rose, J.S.: The formation of polyimides from anhydrides and isocyanates. J. Appl. Polym. Sci. 16, 2983–2989 (1972) 82. Arnold, R.G., Nelson, J.A., Verbanc, J.J.: Recent advances in isocyanate chemistry. Chem. Rev. 57, 47–76 (1957) 83. Querat, E., Tighzert, L., Pascault, J.P., Dusek, K.: Blocked isocyanate. Reaction and thermal behaviour of the toluene 2,4–diisocyanate dimer. Angew. Makromol. Chem. 242, 1–36 (1996) 84. Farrissey Jr., W.J., Ricciardi, R.J., Sayigh, A.A.R.: Reactions of 1, 3-diphenyl-4-(phenylimino)-2-uretidinone. J. Org. Chem. 33, 1913–1917 (1968) 85. Bur, A.J., Fetters, L.J.: The chain structure, polymerization, and conformation of polyisocyanates. Chem. Rev. 76, 727–746 (1976) 86. Kresta, J.E., Hsieh, K.H.: The kinetics of co–catalysis of cyclotrimerization of isocyanates by carbamates. Mactomol. Chem. 180, 793–797 (1979) 87. Fricke, M., Weinrich, D., Loelsberg, W.: Process for producing porous materials. U.S. Patent No 10,717,841 B2. (2020) 88. Kordomenos, P.I., Kresta, J.E.: Thermal stability of isocyanatebased polymers. 1. Kinetics of the thermal dissociation of urethane, oxazolidone, and isocyanurate groups. Macromolecules. 14, 1434–1437 (1981) 89. Kordomenos, P.I., Kresta, J.E., Frisch, K.C.: Thermal stability of isocyanate-based polymers. 2. Kinetics of thermal dissociation of model urethane, oxazolidone, and isocyanurate block copolymers. Macromolecules. 20, 2077–2083 (1987) 90. Skowronski, M.J., DeLeon, A.: Isocyanurate foam-the role of the isocyanate. Techn. Pap. Soc. Plast. Eng. 23, 167–170 (1977) 91. Modesti, M., Simioni, F., Checchin, M., Pielichowski, J., Prociak, A.: Thermal stability and fire performance of modified polyisocyanurate foams. Cell. Polym. 18, 329–342 (1999) 92. Petunova, M.D., Luchkina, L.V., Askadskii, A.A., Kovriga, O.V.: Synthesis and properties of polyisocyanurate networks based on 2,4-toluene diisocyanate and poly(oxytetramethylene) glycol. Polym. Sci. Ser A. 51, 542–549 (2009) 93. Liszkowska, J., Czuprynski, B., Paciorek-Sadowska, J.: Thermal properties of polyurethane-polyisocyanurate (PUR-PIR) foams modified with tris(5-hydroxypentyl) citrate. J. Adv. Chem. Eng. 6, 1000148 (2016) 94. Polyurethane and Polyiso: the Difference? PIMA Technical Bulletin #202; http://c.ymcdn.com/sites/www.polyiso.org/resource/resmgr/ technical_bulletins/tb202_jun30.pdf. Accessed 10 June 2020 95. Donthula, S., Mandal, C., Leventis, T., Schisler, J., Saeed, A.M., Sotiriou-Leventis, C., Leventis, N.: Shape-memory poly(isocyanurate-urethane) (PIR-PUR) aerogels for deployable panels and biomimetic applications. Chem. Mater. 29, 4461–4477 (2017) 96. Donthula, S., Mandal, C., Schisler, J., Leventis, T., Meador, M.A.B., Sotiriou-Leventis, C., Leventis, N.: Nanostructuredependent marcus-type correlation of the shape recovery rate and the Young’s modulus in shape memory polymer aerogels. ACS Appl. Mater. Interfaces. 10, 23321–23334 (2018)

21

564 97. Malakooti, S., Rostami, S., Churu, H.G., Luo, H., Clark, J., Casarez, F., Rettenmaier, O., Daryadel, S., Minary-Jolandan, M., Sotiriou-Leventis, C., Leventis, N., Lu, H.: Scalable, hydrophobic and highly-stretchable poly(isocyanurate–urethane) aerogels. RSC Adv. 8, 21214–21223 (2018) 98. Dearth, R.S., Mertes, H., Jacobs, P.J.: An overview of the structure/ property relationship of coatings based on 4,4′-dicyclohexylmethane diisocyanate (H12MDI). Prog. Org. Coat. 29, 73–79 (1996) 99. Steiner III, S.A., Griffin, J.S., Wunsch, B.H., Schneider, J.N.: Systems and methods for producing aerogel materials. PCT International Publication No WO 2016/127084 A1 (2016) 100. Członka, S., Bertino, M.F., Kośny, J., Shukla, N.: Freeze-drying method as a new approach to the synthesis of polyurea aerogels from isocyanate and water. J. Sol-Gel Sci. Technol. 87, 685–695 (2018) 101. Steiner III, S.A., Griffin, J.S., Wunsch, B.H., Schneider, J.N.: Systems and methods for producing aerogel materials. U.S. Patent No 10,563,035 B2. (2020) 102. Lee, J.K., Gould, G.L., Rhine, W.: Polyurea based aerogels for a high performance thermal insulation material. J. Sol-Gel Sci. Technol. 49, 209–220 (2009) 103. http://www.huntsman.com/portal/page/portal/performance_prod ucts/Media%20Library/global/files/jeffamine_polyetheramines.pdf 104. https://solutions.covestro.com/en/products/multranol/multranol9185_05213088-12942099?SelectedCountry¼US 105. Trifu, R., Gould, G., White, S.: Flexible polyisocyanate based aerogels. MRS Adv. 2, 3479–3484 (2017) 106. Shinko, A., Jana, S.C., Meador, M.A.: Crosslinked polyurea aerogels with controlled property. RSC Adv. (2015). https://doi. org/10.1039/C5RA20788F 107. Shinko, A., Jana, S.C., Meador, M.A.: Crosslinked polyurea-copolyurethane aerogels with hierarchical structures and low stiffness. J. Non-Cryst. Solids. 487, 19–27 (2018) 108. Wu, X., Wu, Y., Zou, W., Wang, X., Du, A., Zhang, Z., Shen, J.: Synthesis of highly cross-linked uniform polyurea aerogels. J. Supercrit. Fluids. 151, 8–14 (2019) 109. Chriti, D., Raptopoulos, G., Papastergiou, M., Paraskevopoulou, P.: Millimeter-size spherical polyurea aerogel beads with narrow distribution. Gels. 4, 66 (2018) 110. Petricevic, R., Glora, M., Fricke, J.: Planar fibril reinforced carbon aerogels for application in PEM fuel cells. Carbon. 39, 857–867 (2001) 111. Petricevic, R., Glora, M., Möginger, A., Fricke, J.: Skin formation on RF aerogel sheets. J. Non-Cryst. Solids. 285, 272–276 (2001) 112. http://www.aerogeltechnologies.com/airloy/products 113. Leventis, N., Sotiriou-Leventis, C., Mulik, S.: Three-dimensional porous polyurea networks and methods of manufacture. U.S. Patent No 10,301,445 B2. (2019) 114. American Society for Metals: ASM Engineering Materials Handbook, Composites, vol. 1, p. 178. ASM International, Materials Park (1998). Table 2 115. Jones, S.M.: A method for producing gradient density aerogel. J. Sol-Gel Sci. Technol. 44, 255–258 (2007) 116. Leventis, N., Sotiriou-Leventis, C., Chandrasekaran, N., Mulik, S., Chidambareswarapattar, C., Sadekar, A., Mohite, D., Mahadik, S. S., Larimore, Z.J., Lu, H., Churu, G., Mang, J.T.: Isocyanatederived organic aerogels: polyureas, polyimides, polyamides. Mater. Res. Soc. Symp. Proc. Mater. Res Soc. 1306 (2011). https://doi.org/10.1557/opl.2011.90 117. Bian, Q., Chen, S., Kim, B.-T., Leventis, N., Lu, H., Chang, Z., Lei, S.: Micromachining of polyurea aerogel using femtosecond laser pulses. J. Non-Cryst. Solids. 357, 186–193 (2011) 118. Weigold, L., Mohite, D.P., Mahadik-Khanolkar, S., Leventis, N., Reichenauer, G.: Correlation of microstructure and thermal conductivity in nanoporous solids: the case of polyurea aerogels synthesized from an aliphatic tri-isocyanate and water. J. Non-Cryst. Solids. 368, 105–111 (2013)

N. Leventis 119. Lu, X., Nilsson, O., Fricke, J., Pekala, R.W.: Thermal and electrical conductivity of monolithic carbon aerogels. J. Appl. Phys. 73, 581–584 (1993) 120. Fricke, J., Lu, X., Wang, P., Buettner, D., Heinemann, U.: Optimization of monolithic silica aerogel insulants. Int. J. Heat Mass Transf. 35, 2305–2309 (1992) 121. Gibson, L.J., Ashby, M.F.: Cellular Solids – Structure and Properties, 2nd edn. Cambridge University Press, Cambridge (1999) 122. Weigold, L., Reichenauer, G.: Correlation between mechanical stiffness and thermal transport along the solid framework of a uniaxially compressed polyurea aerogel. J. Non-Cryst. Solids. 406, 73–78 (2014) 123. Gould, G.L., Lee, J.K., Stepanian, C.J., Lee, K.P.: High strength nanoporous bodies reinforced with fibrous materials. U.S. Patent No 7,560,062. (2009) 124. Weigold, L., Reichenauer, G.: Correlation between the elastic modulus and heat transport along the solid phase in highly porous materials: theoretical approaches and experimental validation using polyurea aerogels. J. Supercrit. Fluids. 106, 69–75 (2015) 125. Malakooti, S., Churu, H.G., Lee, A., Xu, T., Luo, H., Xiang, N., Sotiriou-Leventis, C., Leventis, N., Lu, H.: Sound insulation properties in low-density, mechanically strong and ductile nanoporous polyurea aerogels. J. Non-Cryst. Solids. 476, 36–45 (2017) 126. Lu, H., Xiang, H., Leventis, N., Sotiriou-Leventis, C.: Acoustic attenuators based on porous nanostructured materials. U.S. Patent No 9,068,346 B1. (2015) 127. Malakooti, S., Gitogo, C.H., Lee, A., Rostami, S., May, S.J., Ghidei, S., Wang, F., Lu, Q., Luo, H., Xiang, N., Sotiriou-Leventis, C., Leventis, N., Lu, H.: Sound transmission loss enhancement in an inorganic-organic laminated wall panel using multifunctional low-density nanoporous polyurea aerogels: experiment and modeling. Adv. Eng. Mater. 20, 1700937 (2018) 128. Price, M.A., Aslam, T.D., Quirk, J.J.: Analysis of steady compaction waves in polyurea aerogel. AIP Conf. Proc. 1979, 110016/ 1–110016/7 (2018) 129. Whitworth, N., Lambourn, B.: A single-phase analytic equation of state for solid polyurea and polyurea aerogels. AIP Conf. Proc. 1979, 030007/1–030007/7 (2018) 130. Pacheco, A.H., Gustavsen, R.L., Aslam, T.D., Bartram, B.D.: Hugoniot based equation of state for solid polyurea and polyurea aerogels. AIP Conf. Proc. 1793, 120029/1–120029/5 (2017) 131. Aslam, T.D., Gustavsen, R.L., Bartram, B.D.: An equation of state for polyurea aerogel based on multi-shock response. J Phys. Conf. Ser. 500, 32001/1–32001/6 (2014) 132. Wu, C., Taghvaee, T., Wei, C., Ghasemi, A., Chen, G., Leventis, N., Gao, W.: Multi-scale progressive failure mechanism and mechanical properties of nanofibrous polyurea aerogels. Soft Matter. 14, 7801–7808 (2018) 133. Fakharifar, M., Lin, Z., Wu, C., Mahadik-Khanolkar, S., Leventis, N., Chen, G.: Microstructural characteristics of polyurea and polyurethane xerogels for concrete confinement with FRP system. Adv. Mater. Res. 742, 237–242 (2013) 134. Li, Y., Liao, W., Taghvaee, T., Wu, C., Ma, H., Leventis, N.: Bioinspired strong nanocellular composite prepared with magnesium phosphate cement and polyurea aerogel. Mater. Lett. 237, 274–277 (2019) 135. Yin, W., Lu, H., Leventis, N., Rubenstein, D.A.: Characterization of the biocompatibility and mechanical properties of Polyurea organic aerogels with the vascular system: potential as a blood implantable material. Int. J. Polym. Mater. Polym. Biomater. 62, 109–118 (2013) 136. Gu, S., Jana, S.C.: Open cell aerogel foams with hierarchical pore structures. Polymer. 125, 1–9 (2017) 137. Taghvaee, T., Donthula, S., Rewatkar, P.M., Majedi Far, H., Sotiriou-Leventis, C., Leventis, N.: K-index: descriptor, predictor,

21

Isocyanate-Derived Aerogels and Nanostructure–Materials Properties Relationships

and correlator of complex nanomorphology to other material properties. ACS Nano. 13, 3677–3690 (2019) 138. Rege, A., Preibisch, I., Schestakow, M., Ganesan, K., Gurikov, P., Milow, B., Smirnova, I., Itskov, M.: Correlating synthesis parameters to morphological entities: predictive Modeling of biopolymer aerogels. Materials. 11, 1670 (2018) 139. Tripathi, A., Parsons, G.N., Khan, S.A., Rojas, O.J.: Synthesis of organic aerogels with tailorable morphology and strength by controlled solvents swelling following Hansen solubility. Sci. Rep. 8, 2106 (2018) 140. Zhu, Z., Snellings, G.M.B.F., Koebel, M.M., Malfait, W.J.: Superinsulating polyisocyanate based aerogels: a targeted search for the optimum solvent system. ACS Appl. Mater. Interfaces. 9, 18222–18230 (2017) 141. Jenkins, G.M., Kawamura, K.: Polymeric Carbons, Carbon Fibre, Glass and Char. Cambridge University Press, New York (1976) 142. McKenzie, D.R., Muller, D., Pailthorpe, B.A.: Compressive-stressinduced formation of thin-film tetrahedral amorphous carbon. Phys. Rev. Lett. 67, 773–776 (1991) 143. Loelsberg, W., Fricke, M., Weinrich, D., Leventis, N., SotiriouLeventis, C., Saeed, A.M.: Process for producing isocyanate-based xerogels and aerogels with mineral acids. US Patent Application 20180305491 A1. (2018) 144. Jarfelt, U., Ramnäs, O.: Thermal conductivity of polyurethane foam – best performance. In: 10th International Symposium on District Heating and Cooling, pp. 1–11. Chalmers University of Technology, Göteborg (2006) 145. Biesmans, G., Mertens, A., Duffours, L., Woignier, T., Phalippou, J.: Polyurethane based organic aerogels and their transformation into carbon aerogels. J. Non-Cryst. Solids. 225, 64–68 (1998) 146. Tan, C., Fung, B.M., Newman, J.K., Vu, C.: Organic aerogels with very high impact strength. Adv. Mater. 13, 644–646 (2001) 147. Yim, T.J., Kim, S.Y., Yoo, K.P.: Fabrication and thermophysical characterization of nano-porous silica-polyurethane hybrid aerogel by sol-gel processing and supercritical solvent drying technique. Korean J. Chem. Eng. 19, 159–166 (2002) 148. Rigacci, A., Marechal, J.C., Repoux, M., Moreno, M., Achard, P.: Preparation of polyurethane-based aerogels and xerogels for thermal superinsulation. J. Non-Cryst. Solids. 350, 372–378 (2004) 149. Leventis, N., Sotiriou-Leventis, C., Chidambareswarapattar, C.: Porous polyurethane networks and methods of preparation. U.S. Patent No 8,927,079. (2015) 150. Lu, X., Caps, R., Fricke, J., Alviso, C.T., Pekala, R.W.: Correlation between structure and thermal conductivity of organic aerogels. J. Non-Cryst. Solids. 188, 226–234 (1995) 151. Donthula, S., Mandal, C., Leventis, T., Schisler, J., Saeed, A.M., Sotiriou-Leventis, C., Leventis, N.: Shape-memory superelastic poly(isocyanurate-urethane) aerogels (PIR-PUR) for deployable panels and biomimetic applications. Chem. Mater. 29, 4461–4477 (2017) 152. Malakooti, S., Zhao, E., Tsao, N., Bian, N., Soni, R.U., ud Doulah ABM, S., Sotiriou-Leventis, C., Leventis, N., Lu, H.: Synthesis of aerogel foams through a pressurized sol-gel method. Polymer. 208, 122925 (2020) 153. Leventis, N., Sotiriou-Leventis, C., Bang, A.: Flexible to rigid nanoporous polyurethane-acrylate (PUAC) type materials for structural and thermal insulation applications. U.S. Patent No 9,994,516. (2018) 154. Leventis, N., Sotiriou-Leventis, C., Bang, A.: Flexible to rigid nanoporous polyurethane-acrylate (PUAC) type materials for structural and thermal insulation applications. U.S. Patent No 9,550,846. (2017)

565

155. Kanellou, A., Anyfantis, G.C., Chriti, D., Raptopoulos, G., Pitsikalis, M., Paraskevopoulou, P.: Poly(urethane-norbornene) aerogels via ring opening metathesis polymerization of dendritic urethane-norbornene monomers: structure-property relationships as a function of an aliphatic versus an aromatic core and the number of peripheral norbornene moieties. Molecules. 23, 7 (2018) 156. Papastergiou, M., Chriti, D., Damalas, D.E., Raptopoulos, G., Paraskevopoulou, P.: Poly(urethane-acrylate) aerogels from the isocyanurate trimer of isophorone diisocyanate. J. Supercrit. Fluids. 148, 42–54 (2019) 157. Rewatkar, P., Saeed, A.M., Majedi Far, H., Donthula, S., SotiriouLeventis, C., Leventis, N.: Polyurethane aerogels based on cyclodextrins: high capacity desiccants regenerated at room temperature by reducing the relative humidity of the environment. ACS Appl. Mater. Interfaces. 11, 34292–34304 (2019) 158. Leventis, N., Sotiriou-Leventis, C., Mulik, S., Dass, A., Schnobrich, J., Hobbs, A., Fabrizio, E.F., Luo, H., Churu, G., Zhang, Y., Lu, H.: Polymer nanoencapsulated mesoporous vanadia with unusual ductility at cryogenic temperatures. J. Mater. Chem. 18, 2475–2482 (2008) 159. Leventis, N., Sotiriou-Leventis, C., Chidambareswarapattar, C.: Porous nanostructured polyimide networks and methods of manufacture. U.S. Patent No 9,745,198. (2017) 160. Leventis, N., Chidambareswarapattar, C., Mohite, D.P., Larimore, Z.J., Lu, H., Chariklia Sotiriou-Leventis, C.: Multifunctional porous aramids (aerogels) by efficient reaction of carboxylic acids and isocyanates. J. Mater. Chem. 21, 11981–11986 (2011) 161. Saeed, A.M., Rewatkar, P.M., Majedi Far, H., Taghvaee, T., Donthula, S., Mandal, C., Sotiriou-Leventis, C., Leventis, N.: Selective CO2 sequestration with monolithic bimodal micro/ macroporous carbon aerogels derived from stepwise pyrolytic decomposition of polyamide-polyimide-polyurea random copolymers. ACS Appl. Mater. Interfaces. 9, 13520–13536 (2017) 162. Leventis, N., Sotiriou-Leventis, C., Saeed, M.A.: Novel porous polymer compositions for the synthesis of monolithic bimodal microporous/macroporous carbon compositions useful for selective CO2 sequestration. US Patent Application Publication No 2018/0272312 A1 (2018) 163. Leventis, N., Sotiriou-Leventis, C., Chidambareswarapattar, C.: Multifunctional porous aramids (aerogels) and fabrication thereof. U.S. Patent No 8,877,824. (2014) 164. Cristadoro, A., Pietruschka, R., Fricke, M., Elbing, M.: Porous branched/highly branched polyimides. U.S. Patent Application No. US 2014/0171526 A1. (2014) 165. Majedi Far, H., Rewatkar, P.M., Donthula, S., Taghvaee, T., Saeed, A.M., Sotiriou-Leventis, C., Leventis, N.: Exceptionally high CO2 adsorption at 273 K by microporous carbons from phenolic aerogels: the role of heteroatoms in comparison with carbons from polybenzoxazine and other organic aerogels. Macromol. Chem. Phys. 220, 1800333 (2019) 166. Saeed, A.M., Wisner, C.A., Donthula, S., Majedi Far, H., SotiriouLeventis, C., Leventis, N.: Reuseable monolithic nanoporous graphite-supported nanocatalysts (Fe, Au, Pt, Pd, Ni, and Rh) from pyrolysis and galvanic transmetalation of ferrocene-based polyamide aerogels. Chem. Mater. 28, 4867–4877 (2016) 167. Leventis, N., Sotiriou-Leventis, C., Saeed, M.A.: Multifunctional porous aramids (aerogels), fabrication thereof, and catalytic compositions derived therefrom. U.S Patent No 9,593,225. (2017) 168. Leventis, N., Sotiriou-Leventis, C., Saeed, M.A.: Multifunctional porous aramids (aerogels) and fabrication thereof. U.S Patent No 9,260,581. (2016)

21

566

Nicholas Leventis received his Ph.D. from Michigan State University in organic chemistry in 1985. He is Director of Research at Aspen Aerogels. In 2019 he retired from the Missouri University of Science and Technology as a Curators’ Distinguished Professor of Chemistry. His aerogel work includes polymer-crosslinked aerogels, aerogels from most classes of organic polymers, interpenetrating organic-inorganic aerogels, and metallic, ceramic, and carbon aerogels. Professor Leventis has over 200 scientific publications and over 30 patents. His published work has received over 10,000 citations. His current H-index is 54.

N. Leventis

Aerogels from Engineering Polymers: Polyimide and Polyamide Aerogels

22

Mary Ann B. Meador, Stephanie L. Vivod, Baochau Nguyen, Haiquan Guo, and Rocco P. Viggiano

Contents 22.1

Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 567

22.2 22.2.1 22.2.2

Covalently Networked Polymers . . . . . . . . . . . . . . . . . . . . . . . 568 Highly Crosslinked Network Structures . . . . . . . . . . . . . . . . . 568 Crosslinked Step Growth Polymers . . . . . . . . . . . . . . . . . . . . . . 569

22.3

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 589

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 592

Abstract

As a class of materials, polymer aerogels have many of the same properties as silica aerogels, including high surface area, low density, low dielectric constants, and low thermal conductivity. However, many polymer aerogels also possess mechanical properties far exceeding those of silica aerogels, making them much more applicable as low dielectric substrates, durable insulation, or lightweight multifunctional structures. The use of engineering polymers, such as polyimide and polyamide, as the aerogel backbone, adds the potential for higher use temperatures, combined with good mechanical properties. In addition, by tuning the backbone chemistry, polyimide and

M. A. B. Meador (*) Materials and Structures, NASA Glenn Research Center, Cleveland, OH, USA e-mail: [email protected] S. L. Vivod · R. P. Viggiano NASA Glenn Research Center, Cleveland, OH, USA e-mail: [email protected]; [email protected] B. Nguyen Ohio Aerospace Institute, NASA Glenn Research Center, Brookpark, OH, USA e-mail: [email protected] H. Guo Ohio Aerospace Institute, NASA Glenn Research Center, Cleveland, OH, USA e-mail: [email protected]

polyamide aerogels provide an endless ability to dial in properties for specific aerospace and other applications. Keywords

Polyimide aerogels · Polyamide aerogels · Dielectric · Thermal conductivity · Antennas · Insulation

22.1

Background

Aerogels are low density solids with nanoscale porosity which imparts them with some extraordinary properties, including very low thermal conductivity, very high surface areas, and dielectric constants approaching that of air [1, 2]. Because of these properties, the potential applications of aerogels are seemingly endless and include superior insulation materials for everything from building and construction to high performance clothing, catalyst supports, low dielectric substrates, sensor platforms, materials for environmental remediation, drug delivery materials, tissue scaffolds, and many others [3]. The well-studied silica aerogels, however, are very fragile materials, making many of these applications out of reach. In the last decade, aerogels made from many other materials have emerged with nanoscale porosity, high surface area, and low density similar to silica, especially when made using a similar process – gelation followed by supercritical fluid extraction. Of all the materials studied in the form of aerogels, those derived from polymers have been reported with the best mechanical properties, making all of the potential applications of aerogels closer to reality. Materials in the form of aerogels have been known since the work of Kistler in the 1930s [4, 5]. Kistler was exploring the use of supercritical fluid extraction to remove the liquid from gels without collapsing their pore structure. Once he was able to demonstrate this technique to dry silica gels and coined the term aerogel, he began exploring the technique with many other gels, including gelatin, agar, cellulose,

© Springer Nature Switzerland AG 2023 M. A. Aegerter et al. (eds.), Springer Handbook of Aerogels, Springer Handbooks, https://doi.org/10.1007/978-3-030-27322-4_22

567

568

M. A. B. Meador et al.

nitrocellulose, and egg albumin. Thus, the first polymer aerogels were created. Kistler described the nitrocellulose aerogel as very light and translucent even in pieces several millimeters thick and “the strongest and toughest aerogel that has so far been produced.” Nevertheless, the focus of aerogel research through the early part of this century largely centered on silica aerogels. Two major improvements to the fabrication of aerogels – sol-gel processes to make the gels [6] and the use of liquid CO2, with a supercritical point as low as 32
C and 73 bar, as the extraction solvent [7] – simplified the development of new aerogels. Since then, work on aerogels has grown at a rapid pace, and interest is such that many porous materials that are not made in the way in which Kistler originally coined the term are now identified as aerogels. For the sake of this review, we are defining aerogels in the way that Kistler defined them – air gels – where the liquid that was once contained in the gel has been replaced with air by some means. We will focus mainly on those made using supercritical fluid extraction, although we will compare these to some ambient dried and freeze-dried materials. As a class of materials, polymer aerogels have many of the same properties as silica aerogels, including high surface area, low density, low dielectric constants, and low thermal conductivity. Other properties of polymer aerogels can be tailored for many different applications which are not suitable for silica aerogels. For example, polymer aerogels have been studied as precursors to carbon aerogels and as materials for environment cleanup (oil-water separation, etc.). Many polymer aerogels have mechanical properties far exceeding those of silica aerogels, making them much more applicable as low dielectric substrates or durable insulation. Research into polymer aerogels has exploded to include crosslinked network forming polymers, as well as those that form a gel without a covalently bonded network structure. The latter are virtually crosslinked through crystallization or other supramolecular assembly. Other forms of polymer aerogels derive their network structure from fillers such as clay, nanocellulose, graphene, or carbon nanotubes. These composite aerogels utilize polymers as a glue to hold the nanoparticle scaffolding together. The different classes of polymer aerogels will be discussed in detail herein.

22.2

Covalently Networked Polymers

22.2.1 Highly Crosslinked Network Structures The study of polymer aerogels began in earnest with the work of Pekala and others in the late 1980s, after the introduction of supercritical CO2 drying. The first polymer aerogels extensively studied were derived from highly crosslinked networks such as resorcinol-formaldehyde [8], melamine-

OH

OH

O

OH H HO

OH H HO

OH

Acid or base

OH

OH OH

OH

Resorcinol/formaldehyde HO

HN N

H2N N

NH2 N

NH2

H

H

Acid

H N

N N

O

OH

N

O

NH

N

NH

N Melamine/formaldehyde HN N

H N

HN N N NH

N

O HN NH N O N N H

N

NH

Fig. 22.1 Reaction schemes for aerogels made from resorcinol with formaldehyde and melamine with formaldehyde

formaldehyde [9], and phenol-furfural [10]. Some examples are shown in Fig. 22.1. These all organic aerogels are lower than silica aerogels in thermal conductivity [11], but the most prevalent use of these polymer aerogels has been as a precursor to fabricate carbon aerogels through pyrolysis at temperatures up to 2100
C [12]. Properties of the aerogels are highly dependent on the type of catalyst, ratio of monomers to catalyst, total monomer concentration, and reaction conditions. Many studies have focused on these variables over the years [13]. While the highly crosslinked polymer aerogels from resorcinolformaldehyde and melamine-formaldehyde are similar to silica aerogels in mechanical properties at the same density, the carbon aerogels produced from them are about an order of magnitude higher in modulus. Carbon aerogels have utility as absorbers of pollutants for environmental remediation [14], as hydrogen storage media, [15] and most notably as electrodes for batteries [16] and supercapacitors [17]. For example, in the latter case, carbon aerogel made from a resorcinolformaldehyde aerogel precursor is infused into a carbon fiber fabric for application as an electrode material. Figure 22.2 shows the electrochemical performance of the electrode material with and without the aerogel.

Aerogels from Engineering Polymers: Polyimide and Polyamide Aerogels

As-received CAG-modified (press) CAG-modified (infusion)

0.12

l (A/g)

0.08

24 16

0.04

8

0.00

0

–0.04

–8

–0.08

–16

–0.12

–0.2

–0.1 0.0 0.1 E vs Ag/AgCl (V)

0.2

569

market by BASF with the trade name Slentite ®. Company literature reports the thermal conductivity of these materials to be 17 mW/m-K.

C (F/g)

22

–24

Fig. 22.2 Cyclic voltammograms of as received and carbon aerogel modified carbon fabrics at 5 mV/S. (Reprinted with permission from [17]. Copyright 2013 American Chemical Society)

Other highly crosslinked polymer aerogels have been fabricated from polyurea and polyurethanes [18]. Typically, polyurethane gels are produced from the condensation reaction of polyols with polyfunctional isocyanate in the presence of Lewis acid or Lewis base catalysts in solution. Polyurea gels are produced from the reaction of di-isocyanates and polyamines in solution. Drying the gels is carried out either under ambient condition or with supercritical fluid extraction. The first such porous polyureas produced in this way were based on aromatic di-isocyanates and aromatic polyamines using ambient drying for use as core materials for vacuum insulation panels [19]. Thermal conductivities of these xerogels under vacuum were as low as 7 mW/m-K for samples with densities around 0.15 g/cm3 but increased to 33 mW/m-K under ambient pressure. Later, polyurethane aerogels produced using supercritical fluid extraction from liquid CO2 possessed much lower thermal conductivities at ambient pressure (16 mW/m-K) and high surface areas similar to silica aerogels [20]. These aerogels were made in dichloromethane using polymeric aromatic di-isocyanates and catalysts to produce cyanurate rings as crosslinks. Polyurea and polyurethane aerogels have been made using similar approaches and have obtained similar thermal conductivity and density. For example, Pirard et al. [21] used polyols combined with 4,4′-methylene di (phenylisocyanate) to produce aerogels with density of 0.12 g/cm3 and thermal conductivity as low as 22 mW/mK, while Lee et al. [22] obtained thermal conductivities as low as 14 mW/m-K from a combination of isocyanates in acetone with triethylamine as the catalyst. Leventis et al. [23] used tri-isocyanates reacted with water which transforms some of the isocyanate groups into amines to produce aerogels with densities as low as 0.028 g/cm3, but thermal conductivity was not reported. Polyurethane aerogels have recently been introduced into the commercial insulation

22.2.2 Crosslinked Step Growth Polymers Polyimides Polyimides are a family of high-temperature stable, high performance engineering polymers [24, 25]. They have been used in many rigorous applications, including as the matrix resins for carbon fiber composites for aircraft engines and other structural parts of aircraft and as ablators for thermal protective systems. They also have been used as adhesives and electrical insulation in the electronics industry. Polyimides are typically formed from the step growth polymerization of diamines with dianhydrides to form polyamic acids, which undergo condensation at elevated temperatures or with catalyst to form the imide rings. Aromatic polyimides are among the most thermally stable polymers known and are also inherently nonflammable. Hundreds of different diamines and dianhydrides have been used to tailor the properties of polyimides for myriad applications. Polyimides can be thermosets or thermoplastics, depending on backbone structure and whether or not they are crosslinked. The use of polyimide as the backbone structure of aerogels can lead to a desirable combination of properties, including high heat resistance, low flammability, and mechanical stability combined with the typical properties of aerogels, including low thermal conductivity and low dielectric constants. Highly crosslinked polyimide aerogels have been made from the condensation of aromatic dianhydrides with tri-isocyanates [26] or ring-opening metathesis polymerization of bis-nadimides [27]. These approaches resulted in a high degree of shrinkage during the processing. Hence, the densities of the polyimide aerogels ranged from 0.1 to 0.7 g/ cm3. The latter approach also required heating to affect cure, and onsets of decomposition temperatures of the resulting aerogels were quite low due to the high aliphatic content in their final structure (see “Isocyanate-Derived Aerogels and Applications”). In contrast to aerogels made using monomers which lead to highly crosslinked structures, aerogels have recently been produced from telechelic oligomers of polyimide with 10–60 repeat units and crosslinked using multifunctional amines. By controlling the crosslink density of the polyimide network, greater control over physical properties can be achieved including the ability to fabricate thin films of the polyimide aerogel. Figure 22.3 shows a typical scheme for the polyimide aerogels using 1,3,5-triaminophenoxybenzene (TAB) [28] as the crosslinker and anhydride-capped polyimide oligomers. The oligomers were made using 4,4′-oxydianiline (ODA), para-phenylenediamine (PPDA), or 3,3′-dimethylbenzidine (DMBZ) as the diamine and either

22

570

M. A. B. Meador et al.

O

O

X

R O + H2N NH2

O O

n

O

n+1

O

O

O

O

OH OH H H N N R

O X O

O X O

O

O

n

O

O

or mixture

or

R=

H 2N

O

O HN OH O

HN HO

O X

HO OH HN R

NH2

O

NH2

ODA DMBZ X = C=O (BTDA) or nil (BPDA) O

O

O

X H N

O O O

HN

HN

O n

O O N O

Acetic anhydride, pyridine

O X

N R

O N

OO X

O

n

O

O

N O

N

O N

Fig. 22.3 General scheme for producing polyimide aerogels crosslinked with TAB (Reprinted with permission from [28]. Copyright 2012 American Chemical Society)

benzophenone-3,3′,4′4′-tetracarboxylic dianhydride (BTDA) or biphenyl-3,3′,4,4′-tetracarboxylic dianhydride (BPDA) as the anhydride. Polyimide aerogels have also been made in this way using 1,3,5-tris(aminophenyl)benzene (TAPB) [29], octa-aminophenylsilsesquioxane (OAPS) [30, 31], or triphenylpyridine (TPP) [32] as the crosslinker. Densities of different formulations of polyimide aerogel range from 0.08 to 0.4 g/cm3 depending on backbone chemistry which affects shrinkage during processing and polymer

concentration in the solution. Brunauer-Emmett-Teller (BET) surface areas have been reported ranging from about 150 to as much as 800 m2/g. The crosslinked polyimide aerogels have been reported with thermal conductivity as low as 14 mW/m-K under ambient pressure and 4 mW/m-K in vacuum [30] for a formulation made using OAPS as crosslinker and an oligomer backbone comprised of BPDA and bis (aniline)-p-xylidene (BAX). That formulation had a density of 0.1 g/cm3 and a surface area of 240–260 m2/g.

22

Aerogels from Engineering Polymers: Polyimide and Polyamide Aerogels

Fig. 22.4 Polyimide aerogels (a) can be made in thin film form or as thick molded parts and are strong enough (b) to support the weight of a car (Reprinted with permission from [28]. Copyright 2012 American Chemical Society)

Uses of the polyimide aerogels depend on their backbone structure and form factor. For example, thin films (~0.5 mm), which are very flexible, can be cast using roll-to-roll processing and can even be rolled for supercritical fluid extraction or used to wrap around a structure for insulation. As thicker substrates, the polyimide aerogels are stiff and strong, making them suitable as rigid insulation panels. Figure 22.4 shows a picture of a thin film and a rigid block of polyimide aerogel. The rigid block is also shown supporting the weight of a car.

571

Scanning electron micrographs of polyimide aerogels show a different morphology from silica aerogels. As seen in Fig. 22.5, rather than the “beads on a string” morphology as seen with silica aerogels, the polyimide aerogels appear more fibrous. Depending on backbone structure and the density of the aerogel, the fibers can appear more or less closely packed and/or change in diameter. For example, Fig. 22.5a is an SEM of an aerogel made using DMBZ as the diamine, 5b of an aerogel made using PPDA, and 5c of an aerogel made using ODA. Since PPDA in the backbone causes much more shrinkage during gelation, the strands appear more closely packed than the other two. Figure 22.5d and e are micrographs of aerogels made using combinations of 50/50 DMBZ and ODA and PPDA and ODA, respectively. The micrographs appear similar to those made using PPDA or DMBZ alone. Figure 22.5f is of a cross section of the aerogel film made using 50/50 DMBZ and ODA, showing how the surface is denser than the inside. This is typical of the polyimide aerogels, which seem to develop a smoother skin on the surface. Polyimide aerogels made using at least 50 mol% DMBZ are quite moisture resistant as well. As shown in Fig. 22.6, thin films of aerogels made using 100 mol % ODA quickly absorb water and sink to the bottom of the jar, whereas replacing half the ODA with DMBZ renders the film moisture resistant, and the film remains floating. Using only 50 mol % DMBZ also keeps the films more flexible, while utilizing 100 mol % DMBZ makes the aerogels more stiff and brittle as a thin film. Moisture-resistant polyimide aerogels can also be made by replacing half of the ODA with poly-(propylene glycol)bis(2-aminopropyl ether) (PPG) [33]. Notably, water uptake in those aerogels with 50 mol % PPG is less than 10 w/w%, which accounts for less than 1.5% of the porosity. In addition, PPG in the backbone reduces shrinkage during fabrication to 7–8%. Even more hydrophobic polyimide aerogels have been fabricated by simply reducing the density of the aerogels
[34]. Water contact angles greater than 140 were measured for aerogels made using combinations of ODA and BPDA in the backbone when the densities were reduced to 0.01– 0.02 g/cm3. These aerogels also exhibited petal-like behavior, while aerogels made using a combination of PPDA and 4,4′-(4,4′-isopropylidenediphenoxy)bis(phthalic anhydride)
(BPADA) had contact angles greater than 150 and exhibited a lotus effect. Other approaches to fabricate polyimide aerogels utilize amine-capped oligomers crosslinked with 1,3,5benzenetricarbonyl trichloride (BTC) [35], tri-isocyanate based on hexamethylene di-isocyanate, (Desmodur N3300A) [36], or polymaleic anhydride (PMA) [37]. All three of these will react with the amine end groups to form a network structure. Unlike previously discussed amine crosslinkers, these crosslinkers are both inexpensive and

22

572

M. A. B. Meador et al.

Fig. 22.5 SEMs of different formulations of polyimide aerogels produced BPDA and different diamines: (a) DMBZ, (b) PPDA, (c) ODA, (d) DMBZ and ODA, (e) PPDA and ODA, and (f) DMBZ and ODA shown in cross section at lower magnification. (Reprinted with permission from [31]. Copyright 2012 American Chemical Society)

0.09g/cm3, 94% porous, 498 m2/g

0.28g/cm3, 84% porous, 396 m2/g

0.13g/cm3, 91% porous, 295 m2/g

0.10g/cm3, 93% porous, 392 m2/g

0.25g/cm3, 83% porous, 355 m2/g

0.17g/cm3, 88% porous, 320 m2/g

commercially available, making these aerogels more attractive for scale up. In addition, properties of these aerogels are the same or better than those made using other crosslinkers. Figure 22.7 illustrates a typical scheme for the synthesis of polyimide aerogels using either ODA or DMBZ as the diamine with BTC as the crosslinker. Properties of the aerogels using different crosslinkers but the same backbone chemistry are typically very similar. The backbone chemistry has a much greater influence on the properties. For example, the compressive modulus of both BTC and Desmodur N3300A crosslinked aerogels is

compared in Fig. 22.8. Aerogels made using more rigid DMBZ in the backbone are much higher modulus than those made with ODA or a mixture of DMBZ and ODA, when compared at the same densities, independent of crosslinker. Similar results are seen with other properties. For example, DMBZ-derived aerogels are higher in surface area than ODA-derived aerogels, regardless of the crosslinker used. Flexible thin films of the polyimide aerogels have been considered by NASA as part of the insulation system for inflatable aerodynamic decelerators, one concept for entry,

22

Aerogels from Engineering Polymers: Polyimide and Polyamide Aerogels

ODA

50% ODA+50% DMBZ

Fig. 22.6 Comparison of moisture resistance between polyimide aerogels made using ODA or a 50/50 mixture of ODA and DMBZ. (Reprinted with permission from [31]. Copyright 2012 American Chemical Society)

and descent and landing operations to land large payloads on Mars [38]. Thermally, the polyimide aerogels compared favorably with silica aerogel composite blankets made by Aspen Aerogels. Using thermal gravimetric analysis, the composite blankets begin to outgas at 350
C, while the polyimide aerogels have onsets of decomposition of 500 to 600
C depending on backbone chemistry. In addition, the polyimide aerogel films do not break down or shed dust particles when handled like the composite blankets [39]. A high heat flux test of a proposed layered insulation stack for an inflatable aerodynamic decelerator is shown in Fig. 22.9. The polyimide aerogel insulation is used in combination with 3 M™ Nextel™, a ceramic oxide based fabric, and Saffil ®, a higher temperature alumina-based insulation. In the graph are shown the temperatures measured by thermocouples at various levels in the insulation stack. The top surface of insulation quickly reached about 1300
C during the test, while it took about 240 seconds for the bottom of the insulation to reach 300
C. After the test, examination of the insulation layers showed only minimal charring of the top surface of polyimide aerogel, which saw a peak temperature of about 600
C. Though the polyimide aerogels perform well under high heat flux conditions up to 600
C for short time intervals, isothermal aging at temperatures up to 200
C causes the

573

polyimide aerogels to shrink, depending on the polyimide backbone [40]. For example, aging an aerogel made with BPDA as the dianhydride and different combinations of ODA and DMBZ cause shrinkages as high as 40% at 200
C resulting in densities as high as 0.7 g/cm3 as seen in the graphs in Fig. 22.10. This shrinkage is surprising in that it occurs well below the glass transition of the polyimides, which is typically about 300
C. It is theorized that drying the liquid from the gels must leave some residual stresses that are relieved on heating, causing the shrinkage. Shrinkage does occur in the initial hour of aging and is stable after that, indicating that using the aerogel at elevated temperatures requires preconditioning at those temperatures. Interestingly, using a sterically hindered monomer, such as 9,9′-bis(4-aminophenyl) fluorene (BAPF), in the backbone of the polyimide results in less shrinkage on both fabrication and aging as seen in Fig. 22.10, resulting in much less change in density [40]. BAPF is much bulkier and clearly disrupts the linearity of polymer chains, reducing the potential for close packing of neighboring polymer chains as seen from energy-minimized structures created using Chem3D 16.0 shown in Fig. 22.11a. Furthermore, computational analysis of bond rotation of the cardo-carbon of the fluorene group and the carbon atoms on the two adjacent phenyl rings in BAPF using the dihedral driver calculation from Chem3D 16.0 shows an increase in conformational energy compared to that around the ether linkage in ODA (10 kcal/mol compared to 4 kcal/mol as the angles are rotated 180
). This large barrier to rotation restricts possible chain conformations and ultimately restricts motion and packing within the porous aerogel architecture. In addition to the reduction in shrinkage, the fine pore structure of the aerogel is maintained after aging, as evidenced by bar graphs of BET surface area shown in Fig. 22.11b both before and after aging. Figure 22.11b shows the surface areas of four formulations containing 50 mol % BAPF. For all four formulations, the surface areas are maintained above 300 m2/g even after aging at 200
C, whereas those made with ODA, DMBZ, or combinations of those diamines are reduced to below 100 m2/g after aging at the same temperature. Another application for polyimide aerogel films is as a component of the layered insulation for extravehicular activity (EVA) suits, especially for future manned missions to Mars [41]. Multilayer insulation used in EVA suits for the Moon or in space performs well in a vacuum. Mars poses a bigger challenge since the atmosphere averages about 8 Torr, while temperatures average about 40
C. The polyimide aerogel film can replace the scrim layers in multilayer insulation (MLI) as shown in Fig. 22.12. As seen in the graph, replacing the scrim with polyimide aerogel reduces the thermal conductivity of the MLI over a wide temperature range at simulated Mars atmosphere.

22

574

M. A. B. Meador et al.

O

O R NH2 O + H2N

O

O

n

O

n +1

H2N

O

O

R NH HO

HN R OH NH2 O n

O

O

R=

Acetic anhydride, TEA

or O ODA

DMBZ

R N H2N

O NH2 N R n

O

O O Cl O Cl

Cl O N R

O

HN

NH

O

O O N

O

HN NR

NH O R N O

O

O O

HN NR

N RN H O

O

O

n

O

O

O O

O

O

N

O

Fig. 22.7 Typical scheme for the synthesis of polyimide aerogels with BTC crosslinker (Reprinted from [35]. Author Choice Open Access)

Dhara and Banerjee [42] showed that incorporation of trifluoromethyl groups into the backbone of aromatic polyimides can lead to more optically transparent films. The fluorine also creates low surface energy and cohesive energy resulting in low water uptake, water and oil repellence, and abrasion resistance. Polymer aerogels with fluorinated groups have been formulated with a variety of fluorinated diamines and dianhydrides including 4,4′-hexafluoroisopropylidene diphthalic anhydride (6FDA), 2,2′-bis(trifluoromethyl)benzidine (6FBZ), 2,2-Bis (4-aminophenyl)hexafluoropropane (Bis-AAF), and 2,2-bis [4-(4-aminophenoxyphenyl) hexafluoropropane (6FBAPP)] in the backbone to give more optically transparent aerogels [43]. Figure 22.13 shows the reaction scheme to fabricate one version made with up to 50 mol % 6FBZ replacing DMBZ as the diamine, combined with pyromellitic dianhydride (PMDA) as the dianhydride. The number of repeat units, n, could be as

high as 60, and the oligomers are formulated with n + 1 equivalents of diamine and n equivalents of dianhydride to give amine end caps in order to bond with BTC. To force an alternating structure, the 6FBZ fraction is first reacted with all of the PMDA to create an n ¼ 1 oligomer, followed by addition of the DMBZ. The resulting polyamic acid was then fully imidized chemically by addition of acetic anhydride (AA) and triethylamine (TEA) to form a diamine end-capped polyimide chain. Reaction of the BTC with the polyimide creates the three-dimensional architecture with amide crosslinking. Figure 22.14a shows photos of polyimide aerogels (about 2 mm thick) made with increasing amounts of 6FBZ, demonstrating an increase in optical clarity, high light transmission, and low haze. Clarity is the see-through quality of a material measured at small angles, while haze measures the wide angle scattering of the light through a material. Surface

22

Aerogels from Engineering Polymers: Polyimide and Polyamide Aerogels

areas of the aerogels also increase from 535 m2/g to 706 m2/g with increasing 6FBZ concentration, indicating a finer pore structure which would be expected as clarity increases and haze decreases. Figure 22.14b shows a graph of the empirical model for optical clarity of the aerogels from an experimental design study varying polymer concentration and number of repeat units, n, between crosslinks in addition to 6FBZ concentration. The graph shows a small though significant synergistic effect of polymer concentration with 6FBZ

concentration. For formulations with 0–25% 6FBZ, optical clarity goes down with increasing polymer concentration, while there is a slight increase in optical clarity with increasing polymer concentration for aerogels made using 50% 6FBZ. While optical clarity is improved by tailoring the backbone chemistry to favor smaller, more uniform porosity, other applications require different pore structures. Tuning of the pore structure of polyimide aerogels made using DMBZ and PMDA can be accomplished by use of different solvents such as DMF or DMAC and by the addition of surfactants [44, 45]. In this case, the use of these other solvents and surfactants increases the amount of macro-pores which improves the performance of the aerogel for air filtration applications. As with silica or resorcinol formaldehyde aerogels [46], polyimide aerogels also possess low dielectric constants. This property scales linearly with density regardless of backbone structure [47]. This is shown in Fig. 22.15 for a series of polyimide aerogels made using different amounts of 6FDA in the backbone. Ronova et al. [48], Simpson and St. Clair [49], Hougham et al. [50], and others have shown that the presence of fluorinated groups slightly decreased the relative dielectric constant, attributing this to an increase in the free volume and lower polarizability of the fluorine. Another recent study discussed the fabrication of polyimides containing polyhedral oligomeric silsesquioxanes with fluorinated side chains [51], which decreased the relative dielectric constants from 3.2 to 2.3 with the addition of 15% of these side groups. The decrease in the dielectric constant was suggested to be due

Modulus (MPa)

100

10

1

N3300A - DMBZ

N3300A - DMBZ/ODA

N3300A - ODA

BTC - DMBZ, ref 26

BTC - DMBZ/ODA, ref 26

0.1 0.06 0.07

BTC - ODA, ref 26

0.085 0.1 Density (g/cm3)

575

0.15

Fig. 22.8 Comparison of compressive modulus of polyimide aerogels with different backbone chemistry and different crosslinkers. (Reprinted with permission from [36]. Copyright 2017 American Chemical Society)

1500

Nextel BF-20

TC3R TC4K TC5K TC6K

Nextel BF-20 Saffil Pl aerogel Pl aerogel Kapton

Temperature (°C)

TC2R

TC2R TC3R TC4K TC5K TC6K

1200 900 600 300 0 0

100

200

300

400

500

Time (s) Fig. 22.9 Polyimide aerogel performance in insulation stack under high heat flux testing as shown by thermocouple (TC) measurements at each layer

22

576

M. A. B. Meador et al.

a 50

As fabricated 150 °C 200 °C

Shrinkage (%)

40 30 20 10 0 DMBZ

ODA/ DMBZ Composition

b 1.0

As fabricated 150 °C 200 °C

0.8 Density (g/cm3)

ODA/ BAPF

ODA

0.6 0.4 0.2 0.0 DMBZ

ODA

ODA/ DMBZ Composition

ODA/ BAPF

Fig. 22.10 Cumulative (a) shrinkage and (b) resulting density of different formulations of polyimide aerogels after fabrication and aging at 150 and 200
C for 500 hours. (Reprinted with permission from [40]. Copyright 2017 American Chemical Society)

to a combination of higher free volume, as well as the lower polarizability and greater hydrophobicity of the fluorine groups, but it is not clear from these studies which of these factors plays a larger role. For the polyimide aerogels, the fluorine content seems to play no role, and only density is the controlling factor in the dielectric constant as seen in Fig. 22.15. The low dielectric constant combined with superior mechanical strength makes polyimide aerogels suitable for use as substrates for lightweight antennas [52, 53]. These aerogels have dielectric constants as low as 1.08, allowing for antenna designs with higher bandwidth and gain, and one tenth the weight of those made using conventional substrates. An example of an aerogel antenna 32-element array being tested in the far field antenna range at NASA Glenn Research

Center is shown in Fig. 22.16, along with some test data. The graphs in Fig. 22.16 compare the antenna gain and aperture efficiency of an 8-element Duroid™ antenna with a similar array using an aerogel substrate. The aerogel arrays outperform the Duroid™ array at a wider range of frequencies. The data from the 32-element array made by putting together four 8-element aerogel arrays demonstrates the scalability of the antenna. To save space in aircraft, especially smaller unmanned air vehicles or future urban air mobility vehicles, it may be desirable to design conformal phased array antennas that can be directed with beamforming and conform to a wing or fuselage. This will allow smaller aircraft to communicate with a satellite and travel beyond line of sight of the pilot (further than 300 miles). To enable conformal antenna designs, aerogels that have greater flexibility in a thicker form are needed. To this end, it has been found that incorporating flexible aliphatic chains in the backbone of a polyimide aerogel can increase the flexibility of the aerogel in thicknesses up to 3 mm [54–56]. An example is shown in Fig. 22.17 where 1,3-bis(aminophenoxy)neopentane (BAPN) is used in place of 50 mol % of DMBZ in the backbone in TAB crosslinked aerogels. The polyimide aerogel is much more pliable even with only 25% BAPN in the backbone as shown in the picture of Fig. 22.18 and is moisture resistant when used in combination with DMBZ and BPDA. Increasing the amount of aliphatic diamine up to as much as 75% does typically lead to more shrinkage in the aerogel and a decrease in surface area. Another application of polyimide aerogels is as the tribonegative material in a triboelectric nanogenerator (TENG) [57]. TENGs are one of a handful of devices capable of harvesting mechanical waste energy from the environment. TENGs work by the surface charge transfer that occurs when two dissimilar materials in tribopolarity come in contact with each other. As shown in Fig. 22.19, a small TENG device using a polyimide aerogel as the positive material and a chitosan or cellulose nanofiber aerogel as the negative material can harvest as much as 60 V from a simple motion. Furthermore, as shown in Fig. 22.19c, the more porous the polyimide aerogel, the higher the voltage output. From left to right, the output for the TENG device shown is for polyimide aerogels that are 78%, 84%, and 92% porous, and the output voltage nearly doubles over this small range of porosities. The energy harvesting and output performance of the aerogel TENG devices have been optimized by adding multiple layers of polyamide nanofiber mats in combination with the polyimide aerogel [58]. Figure 22.20 shows the capabilities of the TENG device to power up to 60 LED lights. In

Aerogels from Engineering Polymers: Polyimide and Polyamide Aerogels

a

b

ODA

BAPF BAPF

BAPF

Surface area (m2/g)

22

577

600

As fabricated 150 °C 200 °C

500 400

22

300 200 100 10%/n=40

ODA

10%/n=20

7%/n=30

10%/n=30

0

Composition

Fig. 22.11 (a) Energy-minimized structures of BAPF and ODA in the polyimide backbone created using Chem3D 16.0 and (b) bar graph showing that surface area before and after aging is retained. (Reprinted with permission from [40]. Copyright 2017 American Chemical Society)

b

a

Baseline MLI (Mylar + scrim)

Pi aerogel + Mylar

0.030

c Thermal conductivity (W/m-K)

Fig. 22.12 MLI shown (a) without and (b) with polyimide aerogel sheets incorporated and (c) a comparison of the thermal conductivity of the two forms of MLI under simulated Mars pressure

0.028

Baseline MLI (Mylar with scrim) PI aerogel + Mylar

0.026 0.024 0.022 0.020 0.018 0.016 0.014 –140 –120 –100 –80

–60

–40

–20

Temperature (°C)

0

20

40

H N

O

H N

O

O

O

O

O

O AA

PMDA

H N

N O

O

O

N

O

TEA

N

O + n+1 (x) H2N

O

O

N

O

F3C

CF3

H2N

F3C 6FBZ

CF3

O

N

O

O

N

O

(1-x)

F3C

CF3

O

N

O

O n

N

O

O

F3C

CF3

OH

HN

NH HO O

O

DMBZ

O

NH2 + (1-x) H2N

O

N

O

n

NH2

F3C (x)

CF3

NH2

F3C

CF3 N H

O

O

n

N H

O Cl Cl O BTC

Cl 2O

O OH

HN

NH HO O

O

O

NMP, RT

O N H

F3C

CF3

CF3

NH2

Fig. 22.13 Fabrication of polyimide aerogels with fluorinated groups in the backbone (Reprinted (adapted) with permission from [43]. Copyright 2020 American Chemical Society)

F3C

H2N

O

n O

O

578 M. A. B. Meador et al.

22

Aerogels from Engineering Polymers: Polyimide and Polyamide Aerogels

a

0% 6FBZ

579

25% 6FBZ

50% 6FBZ

22

Porosity=91 %, Shrinkage=25 % ρ = 0.15 g/cm3, Pore size=20 nm Surface area= 535m2/g

Porosity=93 %, Shrinkage=16 % ρ = 0.12 g/cm3, Pore size= 13nm Surface area= 663 m2/g

Porosity= 93 %, Shrinkage=12% ρ = 0.12 g/cm3, Pore size= 9 nm Surface area= 706 m2/g

b

110

Clarity

100

90

80 10.0 9.5 9.0 8.5

Repea t unit

%

.(

nc 7.5

45

(n)

40 7.0

Po

50

lym

8.0

55

co

60

)

0% 6FBZ 25% 6FBZ 50% 6FBZ

er

70

Fig. 22.14 (a) Pictures of aerogel blocks about 2 mm thick showing transparency and shrinkage at increasing fluorine content and (b) empirical models of clarity

addition, Fig. 22.20 also shows that the output voltage measured over 8 weeks for the device is quite stable. Yet another potential application for polyimide aerogels is as a separator for lithium batteries. Polyimide aerogels have a porous structure similar in appearance to

commercial polyolefin separators as shown in Fig. 22.21, although polyimide aerogels can be much higher in porosity. Ideally, in addition to porosity, battery separators for lithium-based batteries should provide mechanical and dimensional stability for manufacturability and to limit

Relative dielectric constant (x-band)

580

M. A. B. Meador et al.

1.35 1.30 1.25 1.20 no 6FDA 0.25 eq 6FDA 0.375 eq 6FDA 0.5 eq 6FDA

1.15 1.10 1.05 0.05

0.10

0.15 0.20 Density (g/cm3)

0.25

Fig. 22.15 Dielectric constants of a series of polyimide aerogels scale with density regardless of backbone structure. (Reprinted with permission from [47]. Copyright 2014 American Chemical Society)

formation of lithium dendrites within the separator over time. They should also be resistant to degradation by electrolytes and ideally have a broad thermal stability window but also have a “shut-off” temperature where the porosity collapses [59, 60]. While polyolefin separators possess many of these properties, a problem plaguing their use is the fact that they are highly flammable, especially when combined with highly flammable aliphatic carbonates as the battery electrolytes. Polyimide aerogels on the other hand are intrinsically nonflammable. In addition, they are able to absorb room temperature ionic liquids (RTIL), which are a safer alternative to carbonates as the electrolyte [61]. Figure 22.22 shows the ionic conductivities of several RTIL containing polyimide aerogel films. As shown, ionic conductivity is dependent on the viscosity of the RTIL as well as the porosity of the aerogel. The highest ionic conductivities are comparable with those achieved with polyolefin separators and polycarbonates.

Polyamides Aramids are a class of rigid rod polyamides developed in the 1960s as strong, heat-resistant polymers [62]. Poly (m-phenylene isophthalate) made from m-phenylenediamine (mPDA) and isophthaloyl chloride (IPC) was commercialized by DuPont with the trademark, Nomex™, followed by the commercialization of poly(p-phenylene terephthalamide) (PPTA) made from PPDA and terephthaloyl chloride (TPC) with the trademark Kevlar™. There are many applications of Nomex and Kevlar, including astronaut spacesuits, bulletproof vests, bicycle tires, firefighting equipment, etc. In a similar approach to the step growth polyimide aerogels,

polyamide aerogels have been fabricated using amine-capped polyamide oligomers from the reaction of mPDA and either IPC or TPC, followed by crosslinking using BTC [63] as shown in Fig. 22.23. Highly crosslinked polyamide aerogels have also been made from the reaction between an aromatic tri-isocyanate and aromatic tricarboxylic acid [64], but high temperatures are required, and attempts to make oligomeric versions result in precipitation (see “Isocyanate-Derived Aerogels and Applications”). While more susceptible to distortion occurring during the solvent exchange and supercritical drying steps, the polyamide aerogels are the strongest polymer aerogels fabricated thus far when the same densities are compared. To illustrate, Fig. 22.24 shows modulus vs density plotted for polyimide aerogels made using BTC or OAPS as crosslinkers and the BTC crosslinked polyamide aerogels derived from TPC and mPDA compared to silica aerogels. The polyamide aerogels shown in the graph are almost three orders of magnitude higher in modulus than the silica aerogels [65]. Polyamide aerogels derived from pPDA and TPC should have the backbone chemistry of Kevlar and might therefore be expected to be even higher strength than those derived from mPDA. Initially, attempts to fabricate these aerogels in the same way failed because the oligomers would rapidly fall out of solution. As shown in Fig. 22.25, the oligomers could be made soluble so they can react to completion by including up to 40% CaCl2 by polymer weight in the reaction mixture [66, 67]. At least 30% CaCl2 was needed to allow polymerization to go to completion as evidenced by the lower modulus of the polyamide aerogels derived from only 20% CaCl2 [68]. The all para polyamide aerogels are not crosslinked as gelation occurs too quickly if crosslinker is added. However, post-gelation crosslinking is possible by soaking the gels in solutions of BTC or other crosslinkers before supercritical fluid extraction. Morphologies of the all para polyamide aerogels are different from the polyimide aerogels, especially when large amounts of CaCl2 are used to fabricate the gels. While polyimide aerogels typically appear very uniform and fibrous, those made with polyamide at lower polymer and CaCl2 concentration have smoother, more ribbon-like fibers as seen in Fig. 22.26a, b. The polyamides made using the highest polymer and CaCl2 concentration (Fig. 22.26g, h) look more sheet-like or papery. At higher magnification, it appears that the aerogels made with low CaCl2 concentration (Fig. 22.26b, d) have smaller diameter fibers, while the fibers appear to coalesce into larger bundles and sheets in the aerogels made with higher CaCl2 concentration (Fig. 22.26f, h). Though the pore structures appear to vary widely in the polyamide aerogels, the BET surface areas only varied from 250 to 300 m2/g, with the highest

22

Aerogels from Engineering Polymers: Polyimide and Polyamide Aerogels

581

22

Antenna aperture efficiency (%)

100

Antenna gain (dBi)

20 16 12 8 4400

4600

4800

5000

5200

Frequency (MHz)

80 60 40 20 0 4400

4600

4800

5000

5200

Frequency (MHz)

Antenna array Aerogel 8´4 element Aerogel 4´2 element #1 Aerogel 4´2 element #2 Aerogel 4´2 element #3 Aerogel 4´2 element #4 Duroid 4´2 element

Aerogel 8´4 element Aerogel 4´2 element #1 Aerogel 4´2 element #2 Aerogel 4´2 element #3 Aerogel 4´2 element #4 Duroid 4´2 element

Fig. 22.16 An antenna array made using a polyimide aerogel as the low dielectric substrate is shown being tested in the far-field range at NASA Glenn and test comparing the aerogel antennas with those made

using a Duroid™ substrate with a relative dielectric constant of 2.1. (Reprinted from Ref. [53])

surface areas measured for those aerogels fabricated from the highest polymer and CaCl2 concentration. It may be possible to further improve the surface area or morphology by the addition of dimethylacetamide (DMAC) in the solution before gelation. This has been shown to improve surface area of polyamides made using melamine and IPC by binding the hydrochloric acid by-product of the

condensation, although NMP may provide the same function [69]. While polyimide aerogels and other polymer aerogels typically have low thermal conductivity similar to silica aerogels at the same density, the all para polyamide aerogels were found to have unusually high thermal conductivity. As shown in Fig. 22.27, thermal conductivity of polyamide

O

O

O

O

O

O

N

OH H N

OH H N

N

OH H N

OH H N

0.5 n

O

O

O

O

O

O

NH2

OH O

O

O

O

O

O

O

O

N

O

O

O

O

NH2

OH H N

O

Acetic anhydride, TEA

O

NH2

H2N

OH N

O

OH

OH N

O

O

O

N O

O

O

O

O

N H HO

N H HO

O

H2N

OH H N O

O

N O n

O

O

O

OH n

n

O

O

O

O

H N

O

0.5 n

O

O

O

O

O

O

O

O

O

NH2

Fig. 22.17 Reaction scheme for flexible polyimide aerogels incorporating BAPN in the backbone (Reprinted (adapted) with permission from [54]. Copyright 2020 American Chemical Society)

O

N

O

O

O

O +H N 2 O

O

n+1

O

O

O

O

O

O

582 M. A. B. Meador et al.

22

Aerogels from Engineering Polymers: Polyimide and Polyamide Aerogels

583

22

Fig. 22.18 A polyimide aerogel made using 25% BAPN and 75% DMBZ with BPDA can flex and recover even in thicknesses up to 3 mm (Reprinted (adapted) with permission from [56]. Copyright 2019 American Chemical Society)

a

P-PDMS or P-PI P-CNF or P-CTS

Spacer

ITO PET b

c

90

P-CTS/P-PI78 34.7 V

P-CTS/P-PI84 44.7 V

P-CTS/P-PI92 60.6 V

Voltage (V)

60 30 0 –30 0.0

0.4

0.80.0 0.4 0.8 0.0 Time (s)

0.4

0.8

Fig. 22.19 (a–c) TENG device made using porous polyimide aerogels (P-PI) with either porous chitosan (P-CTS) or porous cellulose nanofiber (P-CNF) demonstrates higher voltage as porosity of the polyimide aerogel increases. (Reprinted with permission from [57]. Copyright 2018 Wiley)

TENG

Voltage (V)

0

1

2

3

4

5

6

0

LED

50 Charging time (s)

100 150 200 250 300 350

6.8 μF 10 μF 22 μF

e

Fig. 22.20 The energy harvesting and output performance of an optimum TENG consisting of six layers of PA nanofiber mats and one layer of polyimide aerogel with a contact area of 2 cm2. (a) Twenty-six LEDs that spelled “UW” were instantly turned on through a bridge rectifier by a TENG under a compressive stress of 30 kPa. (b) The TENG could light a maximum of 60 green LEDs. (c) Output voltage stability test for the TENG over 8 weeks. (d) Charge capacitors with

d

TENG

b

7 6 5 4 3 2 1 0

–40

–20

0

20

40

60

80

100

120

0

2 Weeks

8 Weeks

Light LEDs

4 Weeks

100 200 300 400 500 600 Time (s)

1 Week

different capacitances (22 μF, 10 μF, and 6.8 μF) using the TENG through a bridge rectifier. (e) A timer and two LEDs powered by the energy stored in a capacitor (22 μF), which was charged by the TENG, through a bridge rectifier. (Reprinted with permission from [57]. Copyright 2018 American Chemical Society)

Load

TENG

c

Voltage (V) Voltage (V)

a

584 M. A. B. Meador et al.

22

Aerogels from Engineering Polymers: Polyimide and Polyamide Aerogels

585

a

22

2.00 µm

b

2.00 µm

Fig. 22.21 Morphology of (a) polyimide aerogel compared to (b) commercial polyolefin battery separators Fig. 22.22 Room temperature ionic conductivities of various RTIL in polyimide aerogels

0.010 58.7 cP

Ionic conductivity (S/cm)

0.008

N

⊕ (CF3SO2)2N– N

72.1 cP 39.4 cP

0.006 0.004 N (CF3SO2)2N– ⊕

0.002 0.000

N (CF3SO2)2N– ⊕

–0.002 84

86

88

90 Porosity (%)

92

94

96

586

M. A. B. Meador et al.

O

NH2

Cl

H2N

+

H2N

O

mPDA n +1

O

O

Cl IPC or TPC n

NH2

H N

H N

n

Amine capped oligomer O

Cl

Cl O H N

HN O

O

O

O

Cl

O

H N

H N

H N

NH

n

H N O

BTC crosslinker O

H N O

H N n

Crosslinked PA aerogel

Fig. 22.23 Reaction scheme for polyamide aerogels made from IPC or TPC. (Reprinted with permission from [63]. Copyright 2014 American Chemical Society)

Modulus (MPa)

100 10 1 PA aerogels PI aerogels/OAPS PI aerogels/BTC Silica

0.1

4 0.

3 0.

2 0.

0.

05 0. 06 0. 0 0. 7 0 0. 8 09 0. 1

0.01 Density (g/cm3) Fig. 22.24 Comparison of moduli of polyimide and polyamide aerogels with silica aerogels. (Data for silica aerogels is taken from [63])

22

Aerogels from Engineering Polymers: Polyimide and Polyamide Aerogels

587

NH2 +

H2N

Modulus (MPa)

Cl

n+1

100

O 10

H2N

0.2 Density (g/cm3)

n CaCl2 NMP

0.3

O

22

H N NH2 O Ethanol

20 wt% CaCl2 30 wt% CaCl2 40 wt% CaCl2

1

0.1 0.09 0.1

N H

Cl

O

n

Supercritical CO2 extraction

Fig. 22.25 Compressive modulus of polyamide aerogels made using pPDA and TPC and varying amounts of CaCl2 and synthetic scheme. (Reprinted with permission from [68]. Copyright 2017 American Chemical Society)

588 Fig. 22.26 SEM at low (a, c, e, g) and high (b, d, f, h) magnification of aerogels made using different concentrations of CaCl2 and polymer in solution. (Reprinted with permission from [68]. Copyright 2017 American Chemical Society)

M. A. B. Meador et al.

a

b

5 w% polymer, 20 w% CaCl2, density = 0.143 g/cm3, surface area = 260 m2/g c

d

10 w% polymer, 20 w% CaCl2, density = 0.190 g/cm3, surface area = 250 m2/g e

f

5 w% solids, 40 w% CaCl2, density = 0.097 g/cm3, surface area = 258 m2/g g

h

10 w% solids, 40 w% CaCl 2, density = 0.265 g/cm 3, surface area = 290 m 2/g

Aerogels from Engineering Polymers: Polyimide and Polyamide Aerogels

Thermal conductivity (mW/m-K)

22

589

and galley furnishings in aviation interiors might all benefit from higher thermal conductivity to help dissipate excess heat.

75 70 65

22.3

60

Conclusions

22

55 50 45 4

5

6 7 8 9 10 Polymer concentration (wt %)

11

Fig. 22.27 Plots of thermal conductivity vs. polymer concentration for aerogels prepared with 40% CaCl2 concentration and n ¼ 40 (standard deviation ¼ 2.62 mW/m-K, R2 ¼ 0.82). (Reprinted with permission from [68]. Copyright 2017 American Chemical Society)

aerogels made using 40% CaCl2 increased linearly with increasing polymer concentration (and density) from 55 to 67 mW/m-K. These values of thermal conductivity are much higher than any aerogel reported in the literature at the same densities. In fact, the specific conductivity is three to four times higher than that of polyimide or silica aerogels [30]. The high effective thermal conductivity can be attributed to efficient phonon transport in the polymer backbone. The backbone thermal conductivity, as defined by Debye, increases with the longitudinal wave speed in the solid material [70]. It is expected that Kevlar with its rigid rod polymer structure and ability for the polymer chains to associate would also have high thermal conductivity. Comparison of bulk polymer properties supports this explanation, as Kevlar fiber has a thermal conductivity of about 3.8 W/m-K at 20
C [71] compared to Kapton ® polyimide film, for example, which has a thermal conductivity of 0.12 W/m-k [72]. The high thermal conductivity of the all para polyamide aerogels, compared to other aerogels of similar density, allows them to be used in applications where the insulating properties of such aerogels are disadvantageous. For example, lightweight plastic replacements in consumer electronics, lightweight structures in automobile engine compartments,

As a class of materials, polymer aerogels have many of the same properties as silica aerogels, including high surface area, low density, low dielectric constants, and, in some cases, low thermal conductivity. However, many polymer aerogels also possess mechanical properties far exceeding those of silica aerogels, making them much more applicable as low dielectric substrates, durable insulation, energyharvesting devices, battery separators, or lightweight multifunctional structures. The use of engineering polymers, such as polyimide and polyamide, as the aerogel backbone adds the potential for higher use temperatures, combined with good mechanical properties. Table 22.1 summarizes the various step growth polyimide and polyamide aerogels synthesized thus far along with some of their advantages and disadvantages. As seen in the table, many different crosslinkers, diamines, and dianhydrides or diacid chlorides have been used to fabricate the aerogels, and many other combinations are possible. Properties typically depend more on the backbone of the oligomers than on the crosslinker. By tuning the backbone chemistry, polyimide aerogels provide an endless ability to tailor properties for specific aerospace and other applications, including moisture resistance, dielectric constant, thermal conductivity, optical clarity, and flexibility. Both polyimide and polyamide aerogels are easily fabricated at room temperature from a single reaction, leading to ease of fabrication and scale-up in various form factors, such as thin films, thick plates, or complex molded parts, making them uniquely promising for commercial applications. The polyamide aerogels additionally can be produced at lower cost with even higher mass normalized strength and stiffness properties than the polyimide aerogels. They also can be produced with higher thermal conductivity, which may lead to some interesting applications in consumer electronics where lightweight and heat dissipation might be needed.

590

M. A. B. Meador et al.

Table 22.1 Summary of step growth polyimide and polyamide aerogels made with different crosslinkers

Crosslinker

Diamine (Ref.)

H2N

Dianhydride/diacid chloride O

NH2

H 2N

NH2

O O O

TAPB H 2N

NH2

O O O

DMBZ (34)

O

O ODPA (29) O

O O O O

O

BPADA (34)

Reacts with terminal anhydrides, room O temperature imidization O O with catalyst; O mechanical O properties depend on BPDA (28, 33, 44, 47, 52) oligomer structure; can be made moisture O O O resistant; good thermal stability; O O crosslinker not O O commercially available BTDA (28, 47, 52) (custom synthesis) O

O NH2

H2N

O

H 2N

O

NH2

n

PPG (33) TAB

O O

NH2 ODA (29, 34)

NH2

O

O O PMDA (29)

O H 2N

Reacts with terminal anhydrides, n = 0-3, thermally imidized at 180 °C, only porosity reported in Ref 29; very low density, superhydrophobic aerogels produced in Ref (34)

O

O

pPDA (29, 34)

Synthesis/unique properties

NH2

F 3C NH2

H2N CF3 6FBZ* O

O NH2

H2N BAPN*

O

F3C CF3

O

O

DMBZ, ODA (28, 43, 44, 47, 52) pPDA (28, 47)

O

O

6FDA (47)

O

PMDA (43, 44) H2N H2N H2N

Si O O Si O

H2N

O

Si O Si O O i O O S O Si O Si O

H2N

NH2

O BAX (30)

ODA, DMBZ, PPDA (31) NH2 NH2

OAPS

NH2

O

NH2

BPDA (30, 31)

Reacts with terminal anhydrides, room temperature imidization with catalyst; mechanical properties depend on oligomer structure; can be made moisture resistant; good thermal stability; commercially available but expensive crosslinker (continued)

22

Aerogels from Engineering Polymers: Polyimide and Polyamide Aerogels

591

Table 22.1 (continued)

NH2

H2N

N N H

Reacts with terminal anhydrides, good thermal stability; crosslinker not commercially available (custom synthesis)

BPDA (32)

NH2

APBI (32)

N NH2

H2 N TPP

O

O

Cl

N H2

H2N Cl

BPDA (35, 40)

O

O

Cl

Cl Cl

O

O

BAPF (40)

BTC

ODA, DMBZ (35)

TPC (63)

Cl

Cl

O

Reacts with terminal amines, room temperature imidization with catalyst; mechanical properties depend on oligomer structure; not as moisture resistant; less thermal stability; commercially available and inexpensive crosslinker

IPC (63)

H2N

NH2

mPDA (63) O C N

O

C

O N O N N

N

N

O

C

ODA, DMBZ (36)

BPDA (36)

ODA, DMBZ (37)

BPDA (37)

O

Desmodur N3300

O

O

R R O

O O R R

O PMA

O

R R O n

O

Reacts with terminal amines, room temperature imidization with catalyst; mechanical properties depend on oligomer structure; not as moisture resistant; less thermal stability; commercially available and inexpensive crosslinker Reacts with terminal amines, room temperature imidization with catalyst; mechanical properties depend on oligomer structure; can be moisture resistant; less thermal stability; commercially available and inexpensive crosslinker

22

592

References 1. Husing, N., Schubert, U.: Aerogels – airy materials: chemistry, structure and properties. Angew. Chem. Int. Ed. 37, 22–45 (1998) 2. Du, A., Zhou, B., Zhang, Z., Shen, J.: A special material or a new state of matter: a review and reconsideration of the aerogel. Materials. 6, 941–968 (2013) 3. Pierre, A.C., Pajonk, G.M.: Chemistry of aerogels and their applications. Chem. Rev. 102, 4243–4265 (2002) 4. Kistler, S.S.: Coherent expanded aerogels and jellies. Nature. 127, 741 (1931) 5. Kistler, S.S.: Coherent expanded-aerogels. J. Phys. Chem. 36, 52– 64 (1932) 6. Nicolaon, G.A., Teichner, S.J.: New preparation process for silica xerogels and aerogels, and their textural properties. Bull. Soc. Chim. Fr. 5, 1900–19006 (1968) 7. Tewari, P.H., Hunt, A.J., Lofftus, K.D.: Ambient-temperature supercritical drying of transparent silica aerogels. Mater. Lett. 3, 363– 367 (1985) 8. Pekala, R.W.: Organic aerogels from the polycondensation of resorcinol with formaldehyde. J. Mater. Sci. 24, 3221–3227 (1989) 9. Alviso, C.T., Pekala, R.W.: Melamine-formaldehyde aerogels. Polym. Prepr. 32, 242–243 (1991) 10. Pekala, R.W., Alviso, C.T., Lu, X., Gross, J., Fricke, J.: New organic aerogels based upon a phenolic-furfural reaction. J. Non-Cryst. Solids. 188, 34–40 (1995) 11. Lu, X., Arduini-Schuster, M.C., Kuhn, J., Nilsson, O., Fricke, J., Pekala, R.W.: Thermal conductivity of monolithic organic aerogels. Science. 255, 971–972 (1992) 12. Pekala, R.W., Alviso, C.T., Kong, F.M., Hulsey, S.S.: Aerogels derived from multifunctional organic monomers. J. Non-Cryst. Solids. 145, 90–98 (1992) 13. Mulik, S., Sotiriou-Leventis, C.: Resorcinol-Formaldehyde Aerogels. Aerogel Handbook, pp. 215–234. Springer, New York (2011) 14. Wei, G., Miao, Y.E., Zhang, C., Yang, Z., Liu, Z., Tjiu, W.W., Liu, T.: Ni-doped graphene/carbon Cryogels and their applications as versatile sorbents for water purification. ACS Appl. Mater. Interfaces. 5, 7584–7591 (2013) 15. Kabbour, H., Baumann, T.F., Satcher, J.H., Saulnier, A., Ahn, C.C.: Toward new candidates for hydrogen storage: high-surface-area carbon aerogels. Chem. Mater. 18, 6085–6087 (2006) 16. Hao, F., Zhang, Z., Yin, L.: Co3O4/carbon aerogel hybrids as anode materials for lithium-ion batteries with enhanced electrochemical properties. ACS Appl. Mater. Interfaces. 5, 8337–8344 (2013) 17. Qian, H., Kucernak, A.R., Greenhalgh, E.S., Bismarck, A., Shaffer, M.S.P.: Multifunctional structural supercapacitor composites based on carbon aerogel modified high performance carbon fiber fabric. ACS Appl. Mater. Interfaces. 5, 6113–6122 (2013) 18. Rigacci, A., Achard, P.: Cellulosic and Polyurethane Aerogels. Aerogel Handbook, pp. 191–214. Springer, New York (2011) 19. Tabor, R.: Microporous isocyanate based polymer compositions and method of preparation. US Patent 5,478,867. (1995) 20. Biesmans, G., Randall, D., Francais, E., Perrut, M.: Polyurethanebased organic aerogels’ thermal performance. J. Non-Cryst. Solids. 225, 36–40 (1998) 21. Pirard, R., Rigacci, A., Marechal, J.C., Achard, P., Quenard, D., Pirard, J.P.: Characterization of porous texture of hyperporous polyurethane based xerogels and aerogels by mercury porosimetry using densification equation. Polymer. 44, 4881–4887 (2003) 22. Lee, J.K., Gould, G.L., Rhine, W.: Polyurea based aerogel for a high performance thermal insulation material. J. Sol Gel Sci. Technol. 49, 209–222 (2009)

M. A. B. Meador et al. 23. Leventis, N., Sotiriou-Leventis, C., Chandrasekaran, N., Mulik, S., Larimore, Z.J., Lu, H., Churu, G., Mang, J.T.: Multifunctional Polyurea aerogels from isocyanates and water. A structure-property case study. Chem. Mater. 22, 6692–6710 (2010) 24. Hergenrother, P.M.: The use, design, synthesis and properties of high performance/high temperature polymers: an overview. High Perform. Polym. 15, 3–45 (2003) 25. Meador, M.A.: Recent advances in the development of Processable high-temperature polymers. Annu. Rev. Mater. Sci. 28, 599– 630 (1998) 26. Chidambareswarapattar, C., Xu, L., Sotiriou-Leventis, C., Leventis, N.: Robust monolithic multiscale naoporous polyimides and conversion to isomorphic carbon. RSC Adv. 2, 26459–26469 (2013) 27. Leventis, N., Sotiriou-Leventis, C., Mohite, D.P., Larimore, Z.J., Mang, J.T., Churu, G., Lu, H.: Polyimide aerogels by Ring-Opening Metathesis Polymerization (ROMP). Chem. Mater. 23, 2250– 2261 (2011) 28. Meador, M.A.B., Malow, E.J., Silva, R., Wright, S., Quade, D.J., Vivod, S.L., Guo, H., Guo, J., Hamilton, B., Cakmak, M.: Mechanically strong, flexible polyimide aerogels cross-linked with Aromatic Triamine. ACS Appl. Mater. Interfaces. 4, 536–544 (2012) 29. Kawagishi, K., Saito, H., Furukawa, H., Horie, K.: Superior Nanoporous polyimides via supercritical CO2 drying of jungle gym type polyimide gels. Macromol. Rapid Commun. 28, 96–100 (2007) 30. Guo, H., Meador, M.A.B., McCorkle, L., Quade, D.J., Guo, J., Hamilton, B., Cakmak, M., Sprowl, G.: Polyimide aerogels crosslinked through amine functionalized Polyoligomeric Silsesquioxane. ACS Appl. Mater. Interfaces. 3, 546–552 (2011) 31. Guo, H., Meador, M.A.B., McCorkle, L., Quade, D.J., Guo, J., Hamilton, B., Cakmak, M.: Tailoring properties of cross-linked polyimide aerogels for better moisture resistance, flexibility and strength. ACS Appl. Mater. Interfaces. 4, 5422–5429 (2012) 32. Shen, D., Liu, J., Yang, H., Yang, S.: Highly thermally resistant flexible polyimide aerogels containing rigid rod biphenyl, Benzimidazole and Triphenylpyridine moieties: synthesis and characterization. Chem. Lett. 42, 1545–1547 (2013) 33. Meador, M.A.B., Agnello, M., McCorkle, L., Vivod, S.L., Wilmoth, N.: Moisture-resistant polyimide aerogels containing propylene oxide links in the backbone. ACS Appl. Mater. Interfaces. 8, 29073–29079 (2016) 34. Li, X., Wang, J., Zhao, Y., Zhang, X.: Template-free self-assembly of fluorine-free hydrophobic polyimide aerogels with lotus or petal effect. ACS Appl. Mater. Interfaces. 10, 16901–16910 (2018) 35. Meador, M.A.B., Aleman, C.R., Hanson, K., Ramirez, N., Vivod, S. L., Wilmoth, N., McCorkle, L.: Polyimide aerogels with polyamide cross-links: a low cost alternative for mechanically strong polyimide aerogels. ACS Appl. Mater. Interfaces. 7, 12240–11249 (2014) 36. Nguyen, B.N., Meador, M.A.B., Scheiman, D., McCorkle, L.: Polyimide aerogels using Triisocyanate as cross-linker. ACS Appl. Mater. Interfaces. 9, 27313–27321 (2017) 37. Guo, H., Meador, M.A.B., McCorkle, L., Scheiman, D., McCrone, J.D., Wilkewitz, B.: Synthesis and characterization of poly(maleic anhydride) cross-linked polyimide aerogels. RSC Adv. 6, 26055– 26065 (2016) 38. Del Corso, J.A., Bruce, W.E., Liles, K.A., Hughes, S.J.: Thermal Analysis and Testing of Candidate Materials for PAIDAE Inflatable Aeroshell, 21st AIAA Aerodynamic Decelerator Systems Conference, Dublin (2011). May 23–26, 2011 39. Del Corso, J.A., Bruce, W.E., Hughes, S.J., Dec, J.A., Rezin, M.D., Meador, M.A.B., Guo, H., Fletcher, D.G., Calomino, A.M., Cheatwood, F.M.: Flexible Thermal Protection System Development for Hypersonic Inflatable Aerodynamic Decelerators, 9th International Planetary Probe Workshop 16–22 June 2012, Toulous (2012)

22

Aerogels from Engineering Polymers: Polyimide and Polyamide Aerogels

40. Viggiano, R.P., Williams, J.C., Schiraldi, D.A., Meador, M.A.B.: Effect of bulky substituents in the polymer backbone on the properties of polyimide aerogels. ACS Appl. Mater. Interfaces. 9, 8287– 8296 (2017) 41. Tang, H.H., Orndoff, E.S., Trevino, L.A.: Thermal Performance of Space Suit Elements with Aerogel Insulation for Moon and Mars Exploration, 36th International Conference on Environmental Systems, Norfolk (2006). July 17–20, 2006, 2006-01-2235 42. Dhara, M.G., Banerjee, S.: Fluorinated high-performance polymers: poly(arylene ether)s and aromatic polyimides containing trifluoromethyl groups. Prog. Polym. Sci. 35, 1022–1077 (2010) 43. Vivod, S.L., Meador, M.A.B., Pugh, C.R., Wilkosz, M., Calomino, K.: Toward improved optical clarity of polyimide aerogels. ACS Appl. Mater. Interfaces. 12, 8622–8633 (2020) 44. Teo, N., Jana, S.C.: Solvent effects on tuning pore structures in polyimide aerogels. Langmuir. 34, 8581–8590 (2018) 45. Zhai, C., Jana, S.C.: Tuning porous networks in polyimide aerogels for airborne nanoparticle filtration. ACS Appl. Mater. Interfaces. 9, 30074–30082 (2017) 46. Hrubesh, L.W., Keene, L.E., Latorre, V.R.: Dielectric properties of aerogels. J. Mater. Res. 8, 1736–1741 (1993) 47. Meador, M.A.B., McMillon, E., Sandberg, A., Barrios, E., Wilmoth, N., Mueller, C.H., Miranda, F.A.: Dielectric and other properties of polyimide aerogels containing fluorinated blocks. ACS Appl. Mater. Interfaces. 6, 6062–6068 (2014) 48. Ronova, I.A., Bruma, M., Schmidt, H.W.: Conformational rigidity and dielectric properties of polyimides. Struct. Chem. 23, 219– 226 (2012) 49. Simpson, J.O., St. Clair, A.K.: Fundamental insight on developing low dielectric constant polyimides. Thin Solid Films. 308, 480– 485 (1997) 50. Hougham, G., Tesoro, G., Viehbeck, A., Chapple-Sokol, J.D.: Polarization effects of fluorine on the relative permittivity in polyimides. Macromolecules. 27, 5964–5971 (1994) 51. Ye, Y.S., Chen, W.Y., Wang, Y.Z.: Synthesis and properties of low Dilectric constant polyimides with introduced reactive fluorine polyhedral oligomeric Silsesquioxanes. J Polym. Sci. Part A Polym. Chem. 2006, 5391–5402 (2006) 52. Meador, M.A.B., Wright, S., Sandberg, A., Nguyen, B.N., Van Keuls, F.W., Mueller, C.H., Rodriguez-Solis, R., Miranda, F.A.: Low dielectric polyimide aerogels as substrates for lightweight patch antennas. ACS Appl. Mater. Interfaces. 4, 6346–6353 (2012) 53. Meador, M.A.B., Miranda, F.A.: Design and development of aerogel-based antennas for aerospace applications: final phase II technical report to NARI seedling fund. NASATM – 2014-218346 (2014) 54. Cashman, J.L., Nguyen, B.N., Dosa, B., Meador, M.A.B.: Flexible, hydrophobic polyimide aerogels derived from use of a neopentyl spacer in backbone. ACS Appl. Polym. Mater. 6, 2179–2189 (2020) 55. Guo, H., Meador, M.A.B., Cashman, J., Tresp, D., Dosa, B., McCorkle, L.S.: Flexible polyimide aerogels with dodecane links in the backbone structure. ACS Appl. Mater. Interfaces. 12(9), 33288–33296 (2020) 56. Pantoja, M., Boynton, N., Cavicchi, K.A., Meador, M.A.B.: Effect of methylene chains in the backbone of polyimide aerogels. ACS Appl. Mater. Interfaces. 11(9), 9425–9437 (2019) 57. Zheng, Q., Fang, L., Guo, H., Yang, K., Cai, Z., Meador, M.A.B., Gong, S.: Highly porous polymer aerogel film-based triboelectric Nanogenerators. Adv. Funct. Mater. 28, 1706365 (2018) 58. Mi, H.-Y., Jing, X., Meador, M.A.B., Guo, H., Turng, L.-S., Gong, S.: Triboelectric Nanogenerators made of porous polyamide nanofiber Mats and Polyimide aerogel film: output optimization and performance in circuits. ACS Appl. Mater. Interfaces. 10 (2018) ASAP 59. Arora, P., Zhang, Z.: Battery separators. Chem. Rev. 104, 4419– 4462 (2004)

593

60. Xiang, Y., Li, J., Lei, J., Liu, D., Xie, Z., Qu, D., Li, K., Deng, T., Tang, H.: Advanced separators for lithium-ion and lithium-sulfur batteries: a review of recent progress. Chem. Sus. Chem. 9, 3023– 3039 (2016) 61. Galinski, M., Lewandowski, A., Stepniak, I.: Ionic liquids as electrolytes. Electrochim. Acta. 51, 5567–5580 (2006) 62. Chae, G.C., Kumar, S.: Rigid-rod polymeric fibers. J. Appl. Polym. Sci. 100, 791–802 (2006) 63. Williams, J.C., Meador, M.A.B., McCorkle, L., Mueller, C., Wilmoth, N.: Syntheis and properties of step-growth polyamide aerogels cross-linked with Triacid chlorides. Chem. Mater. 26, 4163–4172 (2014) 64. Leventis, N., Chidambareswarapattar, C., Mohite, D., Larimore, Z. J., Lu, H., Sotiriou-Leventis, C.: Multifunctional porous aramids (aerogels) by efficient reaction of carboxylic acids and isocyanates. J. Mater. Chem. 21, 11981–11986 (2011) 65. Fricke, J.: Aerogels: highly tenuous solids with fascinating properties. J. Non-Cryst. Solids. 100, 169–173 (1988) 66. Bair, T.I., Morgan, P.W., Killian, F.L.: Poly (1,4-phenyleneterephthalamides): polymerization and novel liquidcrystalline solutions. Macromolecules. 10, 1396–1400 (1977) 67. Kwolek, S.L., Morgan, P.W., Sorenson, W.R.: Process of making wholly aromatic polyamides. U.S. Patent 3,063,966. (1962) 68. Williams, J.C., Nguyen, B.N., McCorkle, L., Scheiman, D., Griffin, J.S., Steiner III, S.A., Meador, M.A.B.: Highly porous, rigid rod polyamide aerogels with superior mechanical properties. ACS Appl. Mater. Interfaces. 9, 1801–1809 (2017) 69. Ren, H., Zhu, J., Bi, Y., Xu, Y., Zhang, L.: Facile fabrication of mechanical monolithic polyamide aerogels via a modified sol–gel method. J. Sol-Gel Sci. Technol. 82, 417–423 (2017) 70. Ebert, H.P.: Thermal properties of aerogels. In: Aegerter, M., Leventis, N., Koebel, M. (eds.) Aerogels Handbook, pp. 537–564. Springer, New York (2011) 71. Ventura, G., Martelli, V.: Thermal conductivity of Kevlar 49 between 7 and 290 K. Cryogenics. 49, 735–737 (2009) 72. DEC Kapton Polyimide Film H-38492 Summary of Properties; DuPont: Wilmington, (2006). Available from http://www.dupont. com/content/dam/dupont/products-and-services/membranesandfilms/polyimde-films/documents/DEC-Kapton-summaryofproperties.pdf. Accessed 10 May 2018

Mary Ann Meador has a BS from Duquesne University (1979) and a Ph.D. from Michigan State University (1983). She joined NASA in 1983. Her research interests have focused on development of polymers and polymer aerogels for future NASA missions. She is also an ACS Fellow, an Executive Editor for ACS Applied Materials and Interfaces, and Adjunct Professor in Polymer Engineering at the University of Akron.

22

594

Stephanie Vivod, a native of North East Ohio, is a Research Chemical Engineer for the National Aeronautics and Space Administration (NASA) in the Structures and Materials Division specializing in advanced polymeric materials for aerospace applications in extreme environments. She received her BS in Chemistry from Cleveland State University and is currently pursuing a PhD in Polymer Science at The University of Akron.

Baochau Nguyen received her Ph.D. from the University of Akron in 2000. She has worked on synthesizing and processing different types of polymers including unsaturated polyesters, sheet molding compounds, water-based polyurethane adhesives, high-temperature polyimides, polyimide composites, and aerogels. Her current focus is on developing organic aerogels for additive manufacturing at the NASA Glenn Research Center.

M. A. B. Meador et al.

Haiquan Guo received her PhD in chemistry from City University of New York. She is a senior scientist working at Ohio Aerospace Institute in Cleveland, Ohio, USA. She works on developing both inorganic and polymer aerogels.

Rocco Viggiano received his PhD from Case Western Reserve University in 2015. He is currently a research chemical engineer in the Materials Chemistry and Physics Branch of the Materials and Structures division at the NASA Glenn Research Center. He currently works on the development of novel aerogel chemistries and materials for advanced energy storage.

ROMP-Derived Aerogels

23

Nicholas Leventis and George L. Gould

Contents 23.1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 596

23.2 23.2.1

Polydicyclopentadiene (pDCPD) Aerogels . . . . . . . . . . . . The Chemical Composition and Structure of pDCPD Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gelation Time and Material Properties of Practical Interest as a Function of Synthetic Conditions . . . . . . . . . . . The Deformation Mechanism of Monolithic pDCPD Aerogels and Rectification of the Problem . . . . . . . . . . . . . . . The Mechanical Properties of pDCPD Aerogels Versus Polynorbornene (pNB) Aerogels and Relevance to Interparticle Connectivity . . . . . . . . . . . . . Addition to Backbone Olefins – Halogenation and Hydrogenation of pDCPD Aerogels – ROMP-Derived Polymers with Inherently High H:C Atomic Ratios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Rigidity of the Olefinic Backbone: Response of Mostly cis pDCPD Xerogels and Aerogels to Solvents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

23.2.2 23.2.3 23.2.4

23.2.5

23.2.6

598 598 600 600

603

607

611

23.3 23.3.1 23.3.2

ROMP-Derived Copolymers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 613 Aerogels from ROMP-Derived Random Copolymers . . . 613 Aerogels from Well-Defined ROMP-Copolymers . . . . . . . 614

23.4

Conclusions and Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 618

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 619

Abstract

Ring-opening metathesis polymerization (ROMP) furnishes polymers with unsaturated hydrocarbon backbones. With the advent of efficient air and solvent N. Leventis (*) Department of Chemistry, Missouri University of Science and Technology, Rolla, MO, USA

tolerant catalysts, ROMP can work under mild conditions and is being adopted for the synthesis of functional materials, including aerogels. Ring-opening metathesis reactions proceed via the driving force provided by strain relief, and are commonly utilized with smaller strained cyclic olefins. Dicyclopentadiene and norbornene are easily available and inexpensive monomers, and most studies of ROMP-derived aerogels described in the literature to date have focused on polymers and random copolymers of these two systems. Designer ROMP-copolymer aerogels include poly(imide-norbornene) materials based on a rod-like imide-norbornene monomer, and poly(urethanenorbornene) aerogels based on three- and nine-norbornene terminated start-like and dendritic monomers, respectively. Fundamental studies have investigated the relationship between nanostructure and bulk aerogel properties, as well as the relationship of both to the configuration and molecular make-up of the polymeric backbone (cis versus trans, interchain crosslinking). Addition of hydrogen to the backbone olefins via reduction in the wet-gel state has increased stability of ROMP-derived aerogels in air. Practical applications explored to date include thermal insulation, lining of ignition targets for inertial confinement fusion, liquid hydrogen storage, selective solvent absorption, and chemoresponsive actuators. Keywords

Ring-opening metathesis polymerization · ROMP · Aerogels · Swelling · Deformation · Catalysts · Thermal insulation · Solvent absorption · Actuators

Department of Research and Development, Aspen Aerogels, Inc., Northborough, MA, USA e-mail: [email protected] G. L. Gould Research and Development, Aspen Aerogels, Inc., Northborough, MA, USA e-mail: [email protected] © Springer Nature Switzerland AG 2023 M. A. Aegerter et al. (eds.), Springer Handbook of Aerogels, Springer Handbooks, https://doi.org/10.1007/978-3-030-27322-4_23

595

596

N. Leventis and G. L. Gould

Olefin metathesis

a

R1

Catalyst

R1

b Ring-opening metathesis polymerization (ROMP)

R2

+

R2

H

H

H

H

Catalyst n Examples:

Cross metathesis

n Cyclopentene Catalyst

R1

R2

+

R1

Ring closure metathesis

R2 n Norbornene See figures 2 and 5 Discyclopentadiene (DCPD)

Fig. 23.1 (a) Generalized scheme for olefin metathesis reactions. (b) Olefin metathesis as applied to ring-opening metathesis polymerization (ROMP). Dashed lines through the middle of double bonds are placed to direct the eye toward the common scission pattern involved in ROMP

23.1

Introduction

This chapter focuses on a class of polymeric aerogels whose synthesis is driven by the specific reaction pathway, olefin metathesis, rather than a type of aerogel material (as for example in ▶ Chap. 20), or a class of starting materials (as for example in ▶ Chap. 21). Olefin metathesis may involve two different olefins, which are “cut” in the middle of their double bonds, and the four fragments recombine via a catalytic pathway to form new olefins (Fig. 23.1a). Conversely, if one is dealing with a single olefin that is part of a (usually strained) ring, then recombination of the fragments may lead to polymerization via a combination of both ringopening and olefin metathesis; that process is referred to as ring-opening metathesis polymerization (ROMP: Fig. 23.1b) [1–3]. Typically, the driving force of ROMP is the strain energy relief that is realized by ring-opening. Olefin metathesis and ROMP comprise a pillar in the foundation of modern organometallic catalysis, and are applied from the synthesis of basic chemicals to functional materials including high-strength polymers, high-value added pharmaceuticals, and so on [4]. The importance of that synthetic pathway was recognized with the 2005 Nobel Prize in Chemistry to Y. Chauvin, R. H. Grubbs, and R. R. Schrock “for the development of the metathesis method in organic synthesis” [5]. The interest in adopting ROMP for the synthesis of aerogels stemmed in part from an intriguing photograph in Grubbs’ Nobel lecture showing a relatively thin polydicyclopentadiene (pDCPD) block with two 9 mm

bullets embedded in it (Fig. 23.2a) [6]. A similar densepolymer pDCPD block prepared by the authors is shown in Fig. 23.2b. As noted in Fig. 23.1, all metathetic reactions require catalysts. In the beginning (late 1950s and through the 1960s) the only common denominator between metathesis of small olefins and ROMP was the fact that in both cases catalysts were certain compounds of transition elements; this general type of catalyst is referred to as ill-defined [7], and include for example high-valence halides of Mo and W (MoCl6 and WCl6) activated with organometallic co-catalysts such as tetramethyl tin (SnMe4) and triethylaluminum (AlEt3) [8], or with later transition metal salts, as for example the industrially used RuCl3/alcohol catalytic system [9], or OsCl3 and IrCl3, which in fact were among the first catalysts developed for the ROMP reaction, almost 50 years ago [10]. The mechanism of the metathesis reaction was debated intensely [11], until it was realized that all reactions of Fig. 23.1, from cross-metathesis to ROMP, share metal alkylidenes (i.e., metallocarbenes containing a metal-to-carbon double bond [12]) as key common intermediates that react with olefins and form metallacyclobutanes (reminiscent of the four membered cyclic intermediates in the Wittig reaction), which in turn propagate the polymer growth as shown in Fig. 23.3. While with ill-defined metallocarbene catalysts are formed heterogeneously in situ, that mechanistic understanding led to the design of specific metallocarbenes as well-defined homogeneous metathesis catalysts. Well-defined metallocarbene olefin-metathesis catalysts fall in two categories: the so-called Schrock-type catalysts [13], which are based on

23

ROMP-Derived Aerogels

597

a

b

23

Fig. 23.2 (a) Ballistic protection: A dense-polymer pDCPD block with two 9 mm bullets embedded in it [6]. (b) A similar block prepared by the authors by adding GC-II (see Fig. 23.4 – dissolved in a few μL of toluene) into a glass mold containing neat monomer

M R Metal alkylidene M = Mo, W, Ru

M

M

M

+

R

R

R Metallacyclobutane

n M

M n

R

R

ROMP-derived polymer Fig. 23.3 The mechanism of the ring-opening metathesis polymerization (ROMP) of cyclic olefins. For metal alkylidene complexes that have been used as catalysts in the preparation of aerogels see Fig. 23.4a. With

ill-defined catalysts (Fig. 23.4b) metal alkylidene complexes are formed in situ. (Color coding is used as a guide to the eye)

Mo and W alkylidenes, and Grubbs-type catalysts [14], which are based on ruthenium alkylidenes. Although Schrock-type catalysts are quite active and allow good control of the metathesis reaction, they are also very sensitive to oxygen, moisture, and solvent heteroatoms, and have not been used in aerogel synthesis up to now. Grubbs’ catalysts on the other hand are often considered of three types, referred to as first, second, and third generation, and oftentimes are abbreviated as GC-I, GC-II (see Fig. 23.4a), and GC-III. Grubbs’ catalysts as a group are much less sensitive to oxygen and moisture, and much more tolerant to functional

groups than the Schrock-type catalysts [15, 16]. Among Grubbs’ catalysts, GC-I is the most sensitive, however, it is also the least expensive and has been the catalyst of choice in aerogel synthesis. GC-II is more solvent-tolerant than GC-I (e.g., it works in alcohols); however, it is also much more expensive (about 3.5
in 2020 prices) and it may give side reactions that chop the polymer chain up. GC-III is prepared from GC-II, and although it is an excellent ROMP catalyst [17], it is even more expensive than GC-II (about 2
in 2020 prices) and has not been used for the preparation of aerogels yet. Besides GC-I and GC-II, the list of catalysts that have

598

N. Leventis and G. L. Gould

Well-defined catalysts

a

P(Cy)3

Mes Cl

Cl Ru

Cl P(Cy)3

Ph

Cy:

Cl Ru

P(Cy)3

THF

Cl

Mes

Cl

GC-I

III-defined catalysts

b

Cl Ph

GC-II

Mes:

–

Cl

Cl

THF Cl W W Cl Cl Cl Cl Cl

WCl6

W2

W

Cl Cl

Initiators:

Na+ . 3THF

H Phenylacetylene (PA)

Norbornadiene (NBD)

Fig. 23.4 Well-defined (a) and ill-defined (b) catalysts that have been used in the synthesis of ROMP-derived aerogels and xerogels, together with relevant abbreviations, and the initiators that have been employed with the ill-defined catalysts

been used for the synthesis of aerogels and xerogels via ROMP includes two ill-defined catalysts, water-sensitive tungsten(VI) hexachloride (WCl6), and a surprisingly stable bimetallic tungsten compound, abbreviated as W2 (Fig. 23.4b); both WCl6 and W2 are activated with phenylacetylene (PA) or norbornadiene (NBD) that presumably form alkylidene complexes in situ [18]. The chains of ROMP-derived polymers obtained with WCl6 and W2 are terminated with a short-length of metathetically derived polyacetylene-like oligomer, or a ROMP-derived NBD oligomer, respectively. Owing to their function, PA and NBD are referred to as initiators and are included in Fig. 23.4b. The first ROMP-derived aerogels were reported by Lee and Gould of Aspen Aerogels in 2007 and were based on polydicyclopentadiene (pDCPD) [19–21]. The objective of that work was to explore the use of pDCPD aerogels in thermal and acoustic insulation. Shortly thereafter (2010), ROMP was employed by the Leventis group at the Missouri University of Science and Technology in the elucidation of the location of the polymer in polymer-crosslinked silica aerogels (see ▶ Chap. 29 [22, 23]; that work on silicaROMP-polymer composites was followed (2011) by ROMP-derived copolymer aerogels based on poly(imidenorbornene) [24]. Subsequent work on ROMP-derived aerogels branched out into fundamental studies of the nanostructure of pDCPD and its interparticle connectivity, the stabilization of backbone olefins against oxidation, and studies of extreme swelling of pDCPD xerogels and aerogels in selected solvents for environmental applications. Work on copolymers branched out into pure ROMPderived copolymer aerogels and to poly(urethanenorbornene) aerogels. Work on pDCPD aerogels is

reviewed in Sect. 23.2 and work on ROMP-derived copolymer aerogels in Sect. 23.3.

23.2

Polydicyclopentadiene (pDCPD) Aerogels

Thematic areas in this section include the chemical composition/structure of pDCPD aerogels (Sect. 23.2.1), studies of the gelation process and material properties of practical interest (porosity, thermal conductivity) as a function of synthetic conditions (Sect. 23.2.2), the deformation mechanism of monolithic pDCPD aerogels and the rectification strategy (Sect. 23.2.3), the mechanical properties of pDCPD versus polynorbornene (pNB) aerogels and their relationship to interparticle connectivity (Sect. 23.2.4), addition to backbone olefins: halogenation and hydrogenation of pDCPD aerogels – ROMP-derived polymers with inherently high H: C atomic ratios (Sect. 23.2.5), and the rigidity of the olefinic backbone – response of mostly-cis pDCPD xerogels and aerogels to solvents (Sect. 23.2.6).

23.2.1 The Chemical Composition and Structure of pDCPD Aerogels Figure 23.5 illustrates the ROMP of dicyclopentadiene (DCPD) into a linear polymer, and the various possibilities for crosslinking via reaction of the double bonds of the pendant cyclopentene rings. All catalytic systems of Fig. 23.4 have been applied to the polymerization of

23

ROMP-Derived Aerogels

599

Catalytic system

trans

GC-I

Dicyclopentadiene (DCPD) Catalyst

GC-II

ROMP

WCl6/PA

Linear pDCPD

cis or trans

n

n

n Crosslinked pDCPD via metathesis

Crosslinked pDCPD via olefin addition

Fig. 23.5 Ring-opening metathesis polymerization (ROMP) of dicyclopentadiene (DCPD) and possible pathways for crosslinking via the pendant cyclopentene ring

DCPD; however, it is pointed out that they are not chemically equivalent, and yield stereochemically different polymers with profound differences in the material properties of the corresponding aerogels. Solid-state 13C CPMAS NMR (Fig. 23.6) has shown that GC-I produces mostly trans olefins along the pDCPD chain; both GC-II and WCl6/PA yield about 1:1 mixtures of cis and trans double bonds (further elaborated via IR and Raman spectroscopies); and W2/PA yields high-cis polymer [25, 26]. With regard to crosslinking of the polymer chains, control experiments with 5,6-dihydrodicyclopentadiene (dhDCPD, Fig. 23.7 [27, 28]) gave no reaction with either GC-I or GC-II, either at room temperature or at 70  C, supporting resistance of the pendant cyclopentene rings to metathetic activity with either catalyst [26, 29]. The conclusion was extended to all four catalytic systems of Fig. 23.4, and was based on the fact that 1H NMR during gelation and aging [26, 29], as well as infrared and Raman spectra of the final aerogels with all four catalysts showed that the engagement of the pendant cyclopentene ring was similar in all four systems [25]. For example, the intensity ratios of the band at 708 cm1 (assigned to the cyclopentene ¼ C–H) and the band at 1450 cm1 (assigned to the –CH2– of the polymeric chain) were similar (in the range of 0.97–0.98) in all infrared spectra of pDCPD aerogels from all four catalytic systems [25]. Thermal analysis data also corroborated toward similar type of crosslinking in all four types of aerogels: (a) Glass transition temperatures of pDCPD obtained with the four catalytic systems were very similar (142–144  C for catalytic systems W2/ PA, GC-I, and GC-II, and 129  C for catalytic system WCl6/ PA) [25, 30]; and, (b) differential thermogravimetry showed similar thermal events in the 460–470  C range for all four

cis

23

W2/PA

17 160 150 140 130 120 110 100 0 90 80 70 60 50 40 30 20 10 0

cis or trans

d (ppm)

Fig. 23.6 Solid-state 13C CPMAS NMR spectra of ROMP-derived pDCPD sol–gel materials using the four catalytic systems of Fig. 23.4. (Adapted from [25], Copyright 2017 Elsevier B. V)

dhDCPD Fig. 23.7 5,6-Dihydrodicyclopentadiene (dhDCPD) used to study the mechanism of crosslinking at the molecular level in pDCPD aerogels [26, 29]

catalytic systems [25]. (Differential thermogravimetry also suggested that the thermal decomposition pathway of pDCPD aerogels involves prior isomerization of the cis to the trans configuration [25].) Overall, all experimental evidence considered together suggested that the pendant cyclopentene rings of pDCPD aerogels from all four catalytic systems did not participate quantitatively in crosslinking via olefin metathesis, leaving olefin addition as the only major pathway for interchain crosslinking (Fig. 23.5). In that context, it was noted that olefin addition converts sp2 carbons to sp3; thus, integration of the sp2 carbons (at 131 ppm) and sp3 carbons in the 30–60 ppm range of the 13C NMR spectra has allowed the estimation of the fraction of the cyclopentene double bonds involved in thermally induced crosslinking via olefin addition (Fig. 23.5). This fraction, x, was calculated via Eq. 23.1, and it was found in the same range for all four catalysts: 19–24% in pDCPD ð2  xÞ=ð3  xÞ ¼ ½alkene Cs =aliphatic Cs experimental ð23:1Þ prepared with GC-I; 19–26% in pDCPD derived with GC-II [26, 29]; and 21–33% in pDCPD derived with WCl6/PA and

600

W2/PA [25]. The overall conclusion was that the only significant difference at the polymer chain (molecular) level between the polymer obtained with the four catalytic systems of Fig. 23.4 was the relative amounts of cis versus trans backbone olefins. A second kind of crosslinking relevant to aerogels is crosslinking at the nanoscopic level, namely crosslinking related to how elementary building blocks are interconnected. This was addressed via top-down mechanical and thermal characterization of pDCPD aerogels as discussed in Sect. 23.2.4 below.

23.2.2 Gelation Time and Material Properties of Practical Interest as a Function of Synthetic Conditions Aerogel-related properties of the pDCPD system with practical interest included the gelation time of DCPD sols, and then the bulk density, shrinkage, and thermal conductivity of the resulting aerogels as a function of the synthetic conditions [19–21]. The independent variables of interest were the monomer concentration (referred to as the target density) and the catalyst-to-DCPD ratio. The catalyst was bis(tricyclohexylphosphine)-3-phenyl-1H-inden-1-ylidene ruthenium dichloride (i.e., GC-I – Fig. 23.4). Wet gels were prepared in toluene, which was exchanged with acetone before drying. Aerogels were obtained by drying acetoneexchanged wet gels with SCF CO2. Selected material properties like shrinkage, porosity, and thermal conductivity were also compared to those of xerogels obtained by slow evaporation of the pore-filing solvent under ambient conditions over a period of 2 months [19]. As expected, the gelation time became significantly shorter as the target density and the catalyst-to-DCPD ratio increased (Fig. 23.8a). Porosity varied almost linearly with the target density, and it was totally insensitive to the catalystto-DCPD ratio (Fig. 23.8b). The thermal conductivity depended strongly on the target density as well as the catalyst-to-DCPD ratio (Fig. 23.8c). Looking closer at the relationship between thermal conductivity and actual (experimental) density, Fig. 23.9 shows that the thermal conductivity values went through a minimum of 0.0165 W m1 K1 at about 0.25 g cm3. This behavior is common among different types of aerogels and it is associated with the chemical composition, the porosity, the pore structure, and the morphology of the network. For comparison, the minimum thermal conductivity that has been reported for resorcinolformaldehyde aerogels under similar conditions is 0.012 W m1 K1, and 0.013 W m1 K1 for opacified silica aerogels. The thermal conductivity of corresponding xerogels (i.e., at the same target density) was 2.0–2.5 times higher for a decrease in porosity by half, as in fact is expected from the

N. Leventis and G. L. Gould

linear relationship of porosity and bulk density, and in turn from the linear relationship between bulk density and thermal conductivity. An interesting observation in that first study of the properties of pDCPD aerogels, which actually became the point of departure for further research on these materials, was the fact that thermal conductivity values had a relatively large error associated with them (see error bars in Fig. 23.9). The presence of that large error could be paired with the fact that the thermal conductivity showed larger-than-normal variation at different locations over the pDCPD aerogel samples; that variation was put in contrast to the behavior of silica aerogels where the variation of the thermal conductivity as a function of location on the sample is typically less than 1 mW m1 K1. In turn, the large variation of the thermal conductivity as a function of location on the sample was attributed to nonuniform pore structure as a result of irregular shrinkage, which sometimes gave monoliths an uneven shape and appearance. These deformation phenomena became extremely severe when gelation of DCPD was induced with GC-II [26, 29].

23.2.3 The Deformation Mechanism of Monolithic pDCPD Aerogels and Rectification of the Problem It was noted that pDCPD aerogel monoliths prepared from GC-II-catalyzed sols came out extremely deformed. Deformation was already complete in the wet-gel state, and then simply followed the samples though drying with supercritical fluid CO2 into aerogels. A typical solvent for the ROMP of DCPD has been toluene. pDCPD wet gels prepared with GC-II in toluene swelled up to double their volume during post-gelation washings with fresh toluene (lasting up to 100 h); subsequently, in preparation for drying, swollen wet gels were solvent-exchanged with acetone, in which they de-swelled rapidly and deformed [26, 29]. Swelling of pDCPD wet gels was attributed to the affinity of the hydrocarbon backbone for toluene on one hand, and the flexibility of the polymeric network on the other. De-swelling was attributed to the prevalence of the hydrophobic-van der Waals interactions between polymeric strands over interaction with polar acetone [29]. Notably, at similar sol concentrations (e.g., >10% w/w) pDCPD wet gels prepared with GC-I did not swell or deform at all [26]. Deformation of GCII-derived wet gels and aerogels was rectified by grafting polymethylmethacrylate (PMMA) on the polymeric backbone by introducing methylmethacrylate (MMA) monomer in the toluene wash baths together with a free radical initiator [29, 31]. Growing PMMA engaged skeletal olefins, but differential scanning calorimetry (DSC) suggested that it was dangling of the pDCPD backbone rather than building

ROMP-Derived Aerogels

601

a 12

b

95

c Thermal conductivity (mW/m K)

23

93 Porosity (%)

6

3

91

89

87

85

0 0.04

0.07

Target density

(g/cm3)

0.07

Target density

Porosity (%)

Gel time (min)

0.04

91

89

87

0.0030

0.0065

0.0100

Catalyst/DCPD (wt/wt)

0.10

Target density (g/cm3)

23

20

17

14

85

6

0.07

26

93

9

23

17

(g/cm3)

15

12

20

0.10

95

18

23

14 0.04

0.10

Thermal conductivity (mW/m K)

Gel time (min)

9

26

0.0030

0.0065

0.0100

Catalyst/DCPD (wt/wt)

0.0030

0.0065

0.0100

Catalyst/DCPD (wt/wt)

Fig. 23.8 The effect of the target density (i.e., the sol concentration) and of the catalyst-to-monomer ratio on the gelation time (a), the porosity (b), and the thermal conductivity (c) of pDCPD aerogels

prepared with GC-I in toluene guided by a design-of-experiments model. (Adapted from [19], Copyright 2007 Springer Nature)

crosslinking bridges. Free PMMA was washed off during post-grafting washes. During and post-grafting with PMMA in toluene, pDCPD wet gels kept on swelling, but during solvent-exchange with acetone they de-swelled orderly, returned to their original mold size and did not deform. Swelling, deformation, and its lack thereof are shown in Fig. 23.10. The nanomorphology of all samples prepared with GC-I and GC-II, as well as those grafted with PMMA were similar:

fibrous at lower sol-concentrations, turning particulate as the concentration increased (Fig. 23.11) [29]. This kind of morphological switch is not unique; in fact it is well-established for a completely different class of aerogels – those based on certain polyureas obtained via reaction of isocyanates and water [32, 33]. Microscopy by itself did not offer a readily identifiable reason for deformation. The fact that PMMA did not produce visible differences in SEM, despite its significant amount (28% w/w) suggested, in analogy to polymer-

602

N. Leventis and G. L. Gould

Thermal conductivity (mW/m K)

crosslinked aerogels, that it occupied space between elementary building blocks (primary particles), and did not spill out of their immediate aggregates (secondary particles). Using small angle x-ray scattering (SAXS) it was found that pDCPD aerogels from both GC-I and GC-II at all densities consisted of primary particles of 60 nm [26, 34]. On the other hand, using rheology, it was found that gel networks were formed with mass-fractal assemblies [26, 34]. Since secondary particles

35 30

Final density vs. TC TC=36-162F.D.+337F.D2

25 20 15 10 0.04 0.08 0.12 0.16 0.20 0.24 0.28 0.32 Final density (g/cm3)

Fig. 23.9 Thermal conductivity of pDCPD aerogels as a function of bulk (final) density. Wet gels were prepared in toluene with a constant catalyst-to-DCPD monomer ratio of 0.0003 w/w. (Reproduced from [19], Copyright 2007 Springer Nature)

a

Wet gels (GC-II)

b

Fig. 23.10 Photographs of: (a) a pDCPD wet gel prepared with GC-II in a 30% w/w toluene sol immediately after it was removed from the mold (left) and of a similar gel swollen after four toluene washes (~32 h in toluene baths – right). (b) A deformed aerogel from a wet gel prepared as in (a) (left), and a non-deformed aerogel prepared from a wet gel

were not mass fractals, it was concluded that the network was not formed by secondary particles, but rather by higher massfractal aggregates of secondary particles. By post-gelation grafting the skeletal framework with PMMA, the primary particle size remained almost the same but the size of secondary particles was reduced substantially [26]. For example, using SAXS, the primary and secondary particle sizes in pDCPD aerogels made with GC-II at 30% monomer concentration were 9.5  0.1 nm and 61  5 nm, respectively; after grafting with PMMA the primary particles were found at 7.9  0.5 nm and the secondary particles at 20.6  0.1 nm [26]. The conclusion was that PMMA filled the space between closed-packed primary particles rendering secondary particles more rigid, thus preventing their merging during de-swelling. In pDCPD aerogels made with GC-I, there was no PMMA to have a similar effect; the only difference between pDCPD aerogels prepared with GC-I and GC-II was at the molecular level: as it was discussed in conjunction with Fig. 23.6, the configuration of the double bonds along the polymeric chain was mostly trans in pDCPD aerogels obtained using GC-I, and an about equal mixture of cis and trans in aerogels obtained with GC-II. It was concluded that a self-consistent model should explain lack of deformation in PMMA-grafted pDCPD aerogels obtained with GC-II, as well as in unmodified aerogels obtained with GC-I. It was proposed then that either filling the space around primary particles with PMMA or having a mostly trans backbone brought about the same degree of rigidization that prevented merging of mass-fractal aggregates of secondary particles. Those processes are summarized in Fig. 23.12. Specifically, merging of mass-fractal aggregates of secondary particles is

Aerogels (GC-II)

c

Aerogel (GC-I)

whose skeletal framework was grafted with 28% w/w of PMMA (right). (c) A pDCPD aerogel prepared with GC-I in a 30% w/w toluene sol. (Adapted from [26] Copyright 2013 The Royal Society of Chemistry, and from [29] Copyright 2015 Springer Nature)

23

ROMP-Derived Aerogels

Fig. 23.11 Scanning electron microscopy (SEM) of the interior of representative pDCPD aerogel monoliths prepared with GC-I (referred to as pDCPD-I-xx) versus those prepared with GC-II (pDCPD-II-xx). Extension “xx” refers to the weight percent concentration of the DCPD monomer in the toluene sol. The sample designated with the “XPMMA-28%” extension was grafted with 28% w/w of PMMA. Notice that the morphology of this sample is identical to that of the corresponding sample without PMMA – immediately above. Bulk densities are given within the frames. Bulk densities of pDCPDII aerogels could not be measured due to deformation. Using pDCPD-I-30 as a proxy for the density of deformed pDCPD-II30, it is noted that the density of PMMA-grafted pDCPD-II-30-XPMMA-28% was lower than that of the native aerogel, owing to reduced shrinkage. (Adapted from [26] Copyright 2013 The Royal Society of Chemistry, and from [29] Copyright 2015 Springer Nature)

603

pDCPD-I-xx

pDCPD-II-xx

xx: r b = 0.026 ± 0.002 g cm−3 2.5

500 nm

23

r b = 0.084 ± 0.002 g cm−3

10

500 nm

500 nm

r b = 0.551 ± 0.004 g cm−3

30

500 nm

500 nm r b = 0.395 ± 0.034 g cm−3

pDCPD-II-30-X-PMMA-28%:

500 nm

illustrated in Fig. 23.12a: mass-fractal aggregates include a large amount of empty space that is capable to accommodate smaller particles from neighboring mass-fractal aggregates, if said aggregates are deformable, and therefore able to squeeze past one another. Figure 23.12b illustrates how rigidization with PMMA can prevent deformation, and Fig. 23.12c illustrates the same effect when the polymer is mostly trans rendering the primary particles themselves more rigid and less deformable. It is noted that the concept of merging massfractal aggregates developed to explain deformation in pDCPD aerogels was engaged by the literature again later to explain the depressurization shrinkage occurring during drying of silica aerogels with supercritical fluid CO2 [35].

23.2.4 The Mechanical Properties of pDCPD Aerogels Versus Polynorbornene (pNB) Aerogels and Relevance to Interparticle Connectivity The mechanical properties of regularly-shaped pDCPD aerogels obtained with GC-II in toluene and grafted with PMMA were investigated in parallel to those of polynorbornene (pNB) aerogels. The latter were prepared with 30% w/w solutions of the monomer (NB) using GC-II as the catalyst [29, 34]. The sol concentration was selected in the middle of the range of the corresponding pDCPD aerogels (20–40% w/w monomer in the sol). pNB was soluble in

604

N. Leventis and G. L. Gould

a

Fractal aggregate of secondary particles

Collapsible cis and trans polymer (with GC-II): Secondary particle: Primary particles b

Deformable primary & secondary particles c

Dimensionally stable

Non-collapsible

No deformation

Mostly trans polymer (with GC-I): Secondary particle: Primary particles

Rigid secondary particles filled with PMMA Fig. 23.12 (a) The mechanism of shrinkage by mass-fractal aggregates merging into one another [34]. (b) Halting shrinkage by particle rigidization by filling empty space within secondary particles with

a

pNB-30(0:100)

b

r b = 0.449 ± 0.005 g cm–3

PMMA [34]. (c) Halting shrinkage by rendering the polymer itself more rigid (mostly trans) [26]

pNB-30(30:70)

5 mm

Non-deformable primary & secondary particles

5 mm r b = 0.449 ± 0.007 g cm–3

c

pNB-30(50:50)

10 mm r b = 0.684 ± 0.015 g cm–3

Fig. 23.13 SEM of pNB aerogels obtained with GC-II, in 30% w/w sols and various toluene:iPrOH v/v ratios (given in parentheses). Skeletal particle sizes were in the micron range and did not vary between

pure iPrOH (a) and a 30% mixture with toluene (b). The morphology became bicontinuous when the solvent ratio went to 1:1 v/v. (Adapted from [34] Copyright 2013 The Royal Society of Chemistry)

toluene, so wet gels were obtained in pure isopropanol (iPrOH) and in toluene:iPrOH mixtures up to 1:1 v/v. The pNB aerogels consisted of much larger (micron-size) particles than those comprising the corresponding pDCPD aerogels; that morphology changed to bicontinuous when

the toluene:iPrOH volume ratio was increased to 0.5 (Fig. 23.13). Freshly prepared, cylindrical pDCPD/PMMA and pNB aerogel samples were tested under both quasi-static and high strain-rate compression, and results were discussed

23

ROMP-Derived Aerogels

605

relative to thermal conductivity data [29]. All samples showed a short linear elastic region up to less than 3% strain, followed by plastic deformation, and inelastic hardening (Fig. 23.14). Samples absorbed energy up to 80% compressive strain, did not buckle and were absorbed within their own porosities. (See photographs in Fig. 23.14.) Under quasi-static compression the Young’s moduli, E, of the pDCPD/PMMA aerogels were in the 279–349 MPa range, while at comparable densities (around 0.5 g cm3) the Young’s moduli of spheroidal pNB aerogels were in the range of 92–152 MPa. (Bicontinuous pNB aerogels were

a 500 pDCPD-20-X-MMA-50 pDCPD-30-X-MMA-50 pDCPD-40-X-MMA-50

300 200

Stress (MPa)

Stress (MPa)

400

10

0

100

0.00

0.05 0.10 Compressive strain

0 b

pDCPD-20-X-MMA-50 pDCPD-30-X-MMA-50 pDCPD-40-X-MMA-50

Before

After

20

100

50

Stress (MPa)

Stress (MPa)

150

10

0 0.00 0.05 0.10 Compressive strain

0 c

pNB-30(50:50) pNB-30(30:70) pNB-30(10:90) pNB-30(0:100)

150 Stress (MPa)

Fig. 23.14 (a) Stress–strain curves of pDCPD-xx/PMMA aerogel monoliths under quasistatic compression (strain rate ¼ 0.01 s1; Extension “xx” refers to the weight percent concentration of the DCPD monomer in the toluene sol). Inset: magnification of the low-strain linear region. Photograph: a pDCPD-30/PMMA sample after the compression experiment. (b) Stress–stress curves of pDCPD-xx/PMMA aerogel monoliths under dynamic compression (strain rates in the 1200–1300 s1 range). Inset: as in part A. Photograph: a pDCPD-30/ PMMA sample, as shown. (c) Stress–strain curves of pNB-30 (toluene:iPrOH v/v) monoliths under dynamic compression (strain rates in the 1100 to 1200 s1 range). Photograph: (i) a representative sample before impact; (ii–v), (toluene:iPrOH v/v): (0:100), (10:90), (30:70) and (50:50), respectively. (Reproduced from [29] Copyright 2013 The Royal Society of Chemistry)

significantly denser (0.87 g cm3) and their Young’s moduli reached 1.5 GPa.) Young’s moduli are related to the cumulative neck area per unit volume in the material [36], and in that respect the ratio of the two groups of E values was not very different from the ratio of the C coefficients derived from the solid thermal conductivity (λs ¼ C(bulk density)α) of the same samples (CpDCPD/PMMA ~ 0.66CpNB). That was interpreted to mean that the contact area per unit volume and the chemical bridging between nanoparticles in the two kinds of materials, pDCPD/PMMA and pNB, were similar. That finding was considered unexpected, because, all other

i

100

ii

50

0 0.0

0.2 0.4 0.6 0.8 Compressive strain

1.0

iii

iv

v

23

606

N. Leventis and G. L. Gould

things being equal, pDCPD is capable of even some crosslinking (see Fig. 23.5), while pNB is not capable at all. In that respect, pNB aerogels should not have been able to carry any significant loads and they should have been much less stiff materials than pDCPD aerogels. Also, the presence of PMMA was not seen to cause any unusual increase in the stiffness of the pDCPD/PMMA aerogels, which was considered consistent with the conclusions reached from the DSC data, namely that PMMA was not involved in interparticle crosslinking to any appreciable extent. With PMMA-based crosslinking ruled out, it was proposed that there had to be a common mechanism for holding the two kinds of polymeric nanostructures, pDCPD and pNB, together. Considering that ROMP is a living process, phase-separated nanoparticles of either pDCPD in toluene, or pNB in toluene:iPrOH mixtures, were expected to be terminated with active catalyst, which could be engaged in cross-metathesis with polymer on the surface of another phase-separated nanoparticle, when the two nanoparticles came in contact. That process is summarized in Fig. 23.15, and its effect is to bring about crosslinking between particles by extending the polymeric network of the one particle into another. It was suggested that the development of this type of interparticle bridging would lead to a broad polydispersity for the core polymer. That was confirmed in the case of pNB aerogels, which could be

Cl

Cross metathesis

[Ru]

Ru

[Ru]

partially solubilized, and therefore polydispersity could be measured. It is known that pNB obtained with GC-I displays polydispersities in the range of 2.0–2.5, which have been attributed to cross-metathesis (backbiting and chain-transfer reactions) [37]. After using THF to dissolve pNB aerogels obtained from 30% w/w sols at various toluene:iPrOH ratios, the polydispersities were much higher – in the 8–13 range. Although porous materials in general appear stronger, stiffer, and tougher under dynamic loading conditions at higher strain rates [38], exactly the opposite was observed with pDCPD aerogels at various target densities and PMMA loadings up to 28% w/w (compare Fig. 23.14a, b). The case of pNB aerogels obtained in different toluene: iPrOH ratios was more complicated: with the exception of the pNB-30(50:50) samples, which had different micromorphology from the rest of the samples (Fig. 23.13), all others samples were stiffer under quasi-static loading (just like pDCPD/PMMA), but in general they appeared stronger and could absorb more energy under dynamic loading conditions (Fig. 23.14c). The mode of failure under different loading conditions was also quite interesting. Under quasi-static compression, pDCPD/PMMA samples failed by shattering in fragments; under dynamic loading, however, they held themselves together. On the other hand, pNB-30 samples shattered

IMes

Cl

[Ru]

[Ru]

Backbiting Metathesis

[Ru]

[Ru] Cl

Cl Ru

Self metathesis

IMes

IMes:

N

[Ru]

+

Cy3P

Cl Ru

Cl

+

IMes

N

[Ru]: Ru catalyst-terminated tip of a propagating pDCPD or pNB chain : Phase-separated pDCPD or pNB particle (terminated with 'live' Ru-catalyst)

Fig. 23.15 Interparticle crosslinking mechanism: Cross-metathesis effectively extends the polymer of one pDCPD or pNB skeletal particle into another [29]

23

ROMP-Derived Aerogels

607

under both quasi-static and dynamic loading conditions. The behavior of the pDCPD/PMMA versus the pNB samples was considered consistent with both the particle crosslinking mechanism of Fig. 23.15, whereas skeletal nanoparticles in pDCPD/PMMA and pNB aerogels are linked by sharing their core polymeric strands, and the fewer interparticle contacts in macroporous pNB samples. It was proposed that interparticle links can take various conformations. At slow strain rates the polymer is given time to get reorganized and thus takes more load. Under quasi-static compression pDCPD/MMA samples took on average 380 MPa at 86% strain, but only 86 MPa at 72% stain under dynamic loading. By the same token, it was proposed that at the highest strain (86%) most of the void space had been squeezed out of pDCPD/PMMA, thereby loads were applied directly to rigid but glassy PMMA-filled secondary particles, and the material displayed brittle-like behavior and shattered. Despite their strength, a major drawback in the practical implementation of pDCPD aerogels is air oxidation of the backbone olefins, which is actually facilitated by the open-pore aerogel structure. Over a period of a couple of months, pDCPD aerogels become friable. Resistance to oxidative degradation has been addressed primarily by increasing the H:C ratio by addition of hydrogen (hydrogenation).

23.2.5 Addition to Backbone Olefins – Halogenation and Hydrogenation of pDCPD Aerogels – ROMP-Derived Polymers with Inherently High H:C Atomic Ratios This part of the work on ROMP-derived aerogels was carried out at the Lawrence Livermore National Laboratory (LLNL). Addition of iodine and bromine to olefins is an elementary reaction that is typically carried out in solution. Electrophilic addition of halogens on unsaturated polymers in the solid state is uncommon, but the open structure of wet gels and aerogels in combination with nanometer-sized pore walls

leaves this possibility open. Thus, although this work was not conducted for stabilization against oxidation, addition of both Br2 and I2 has been attempted in both the liquid and the gas phase onto wet-gels and aerogels of pDCPD/norbornene random copolymers prepared at a target density of 50 mg cm3 and a 100:10 v/v DCPD-to-NB ratio. The overall scheme, illustrated by the addition of I2, is shown in Fig. 23.16 [39]. (ROMP-derived copolymers are discussed in more detail in Sect. 23.3 below.) Addition of iodine from toluene solutions of I2 was reversible and only a small amount of iodine ended up in the polymer (C10H12O1.5I0.005). Addition of bromine from its vapors was fast and the uptake was relatively high (C10H12Br2.8); however, monoliths collapsed during doping and therefore the value of this approach was deemed low. Most successful were attempts of doping with iodine from the gas phase and with bromine from Br2/toluene solutions, giving stoichiometries equal to C10H12I0.43 (1.9 atomic percent of iodine) and C10H12Br1.5 (7.6 atomic percent of bromine), respectively. (It is noted that assuming linear pDCPD aerogels, that is, with no crosslinking present, the maximum amount of halogen, X, uptake would give a formula of C10H12X4.) Hydrogenation was carried out on pDCPD wet gels prepared in 1,2-dicholorobenzene with GC-I [40]. After aging, wet gels were placed in a deaerated mixture of 1,2-dichlorobenzene and ethyl vinyl ether; ptoluenesulfonyl hydrazine (TSH) was added to that solution together with tripropylamine (TPA) and the solution was heated at 75  C under N2. The role of tripropylamine was to neutralize sulfinic acid side product, and prevent its reaction with the remaining olefins of pDCPD. The amounts of TSH and TPA were optimized at 8 and 32 equivalents, respectively. Figure 23.17a shows the evolution of the FTIR-ATP spectra of pDCPD aerogels prepared from 30% w/w sols and heated with the optimized amounts of TSH and TPA at 75  C for the time periods indicated. Absorptions associated with olefins (see reaction in Fig. 23.17b) went away after 3 days of heating (Fig. 23.17c), and that time period became part of the standard hydrogenation protocol. The H:C atomic ratio of pDCPD was equal to 1.2, while the

X2

X2

X2

X2

X2 a n

1-n

(X2: Br2 or I2)

a n

X2

1-n

p(DCPD-r-NB)

Fig. 23.16 Halogenation of random copolymers of pDCPD and polynorbornene (p(DCPD-r-NB): see Sect. 23.3.1; cross-reference with Fig. 23.25) [39]

23

972 cm–1

3050 cm–1

3050 cm–1

0.0

0.2

0.4

0.6

0.8

1.0

0

1

972 cm–1

H

7

C : H / mole ratio

40

80 %

–1

45 Concentration (mg mL )

1.60 1.55 97 % 95 % 1.50 1.45 pDCPD 1.40 H-pDCPD 1.35 1.30 1.25 1.20 25 30 35

50

74 %

hydrogenated pDCPD (H-pDCPD) aerogels of varying target densities. The theoretical H:C atomic ratio after complete hydrogenation is 1.6. The percent values associated with every point indicate the degree of hydrogenation as calculated from the C:H atomic ratios. (Adapted from [40] Copyright 2012 Elsevier Ltd)

6

3050 cm–1 972 cm–1 708 cm–1

H 708 cm–1

H

2 3 4 5 Reaction time (days)

Fig. 23.17 (a) The effect of hydrogenation as a function of the reaction time on the FTIR-ATR spectra of a pDCPD aerogels (prepared at target density ¼ 30 mg cm3 as described in the text). Specific absorbances marked in the spectra correspond to the structural features shown in the reaction (b). (c) A plot of the relative IR intensities versus reaction time shows consumption of the IR absorptions after approximately 3 days of reaction. (d) Elemental analysis of pDCPD and

Wavenumber (cm )

–1

4000 3500 3000 2500 2000 1500 1000 500

Initial

1 day

2 days

3 days

4 days

7 days

H

708 cm–1

H

Relative IR intensity

H

608 N. Leventis and G. L. Gould

23

ROMP-Derived Aerogels

609

theoretical ratio after complete hydrogenation of linear pDCPD is 1.6. As shown in Fig. 23.17d, the ratio decreased somewhat from 1.58 to 1.50 with increasing sol concentration (i.e., target density) from 25 mg cm3 to 30, 40, and 50 mg cm3, corresponding to 97%, 95%, 80%, and 74% of hydrogenation. At those low sol concentrations, pDCPD aerogels are fibrous and that morphology was preserved after hydrogenation (Fig. 23.18 – cross-reference with Fig. 23.11 above). However, hydrogenated pDCPD aerogels had higher shrinkage factors (¼ experimental density: target density) than their regular analogues: in the ranges of 1.9 to 3.3 versus 1.2 to 2.2, respectively; in both cases, shrinkage factors increased as the target density increased. Thus, the actual densities (surface areas) of pDCPD aerogels with target densities at 25, 30, 40, and 50 mg cm3 were 30 (303), 42(303), 78(282), and 109(284) mg cm3(m2 g1), respectively, while the corresponding values for the hydrogenated samples were 48(192), 71(226), 128(167), and 166(176) mg cm3(m2 g1). The resistance to oxidative degradation of the hydrogenated samples was confirmed by placing samples in an oven at 135  C in air for 2 h. The FTIR-ATR spectra of pDCPD at all target densities that were investigated developed a strong absorption band in the 1700 cm1 region assigned to carbonyl groups, and a broad absorbance in the 3400 cm1 region assigned to hydroxyls; similarly, the C-H stretching absorptions at the left of 3000 cm1 were intensified (Fig. 23.19a). The FTIR-ATP spectra of the corresponding hydrogenated samples showed no change (Fig. 23.19b). XPS showed that pDCPD samples (from the lowest concentration sols: target density at 25 mg cm3) were already partially

oxidized on the surface (about 76% C, 16% oxygen) but after treatment at 135  C for 2 h in air, the oxygen content increased to 21% w/w. The corresponding hydrogenated samples contained 4.6% oxygen initially, which increased to 4.7% after thermolysis. The glass transition of pDCPD aerogels decreased from 168  4  C to 111  3  C after hydrogenation. The decomposition temperature was not affected by hydrogenation; both types of samples showed thermal weight loss >400  C, and their TGA traces practically coincided. However, the most dramatic effect of any thermal treatment was the fact that hydrogenated pDCPD aerogels shrunk a lot. For example, upon heating at 135  C in air for 2 h the experimental density of hydrogenated pDCPD aerogels prepared with a target density equal to 25 mg cm3 increased to 710 mg cm3 (from 48 mg cm3, originally). The corresponding pDCPD aerogels did not suffer such shrinkage; their density went to 45 mg cm3 (from 30 mg cm3, originally), but it was also reported that they underwent severe deformation, while their hydrogenated analogues, despite their dramatic increase in density, did not. In an interesting extension of this work, the LLNL group reasoned that the H:C atomic ratio, and thereby the oxidative stability of ROMP-derived aerogels, would increase if one moved away from pDCPD and its copolymers (see Sect. 23.3.1 below), and instead considered new monomers (see Fig. 23.20) specifically designed with aliphatic spacers and multiple norbornene (NB) moieties for crosslinking [41]. In that regard, it is noted that the three difunctional monomers shown in Fig. 23.20a (NB(CH2)nNB for n ¼ 2, 6, 10) and the trifunctional monomer shown in Fig. 23.20b (NB3) were pure

pDCPD

LLNL

SE

SEM

LEI

2.0kV

Hydrogenated pDCPD

X15,000 WD 9.4mm

1mm

LLNL

SE

SEM

LEI

2.0kV

X15,000 WD 9.6mm

1mm

Fig. 23.18 SEM images as indicated on top, from sols made with the same target density (25 mg cm3). The fibrous aerogel structure (cross reference with Fig. 23.11) was unaffected by hydrogenation of the polymer backbone. (Reproduced from [40] Copyright 2012 Elsevier Ltd)

23

610

N. Leventis and G. L. Gould

pDCPD

a 25

25

30

30

40

40

50

50

4000

3500

3000 2500 2000 1500 Wavenumber (cm–1)

1000

500

Fig. 23.19 FTIR-ATR spectra of pDCPD (a), and of hydrogenated pDCPD (b). Spectra are overlaid from before (- - -) and after (____) thermolysis at 135 C/2 h in air. Thermal treatment of hydrogenated pDCPD aerogels shows the lack of formation of absorbances at

Fig. 23.20 Monomers designed in order to increase (a) the amount of crosslinking between ROMPderived polymer chains and (b) the H-to-C atomic ratio. (a) Three difunctional monomers with variable length of the aliphatic tether between two norbornene (NB) moieties. (b) A trifunctional monomer in the spirit of frame (a). (c) A mixture of three difunctional monomers obtained via a DielsAlder reaction between cyclopentadiene and 1,5-cyclooctadiene [41]

Hydrogenated pDCPD

b

4000 3500 3000 2500 2000 1500 1000 Wavenumber (cm–1)

500

3400 cm1 (OH) and 1700 cm1 (carbonyl) indicating that these aerogels are increasingly stable to oxidation versus pDCPD aerogels. Numbers at left of each spectrum indicate the target density of each sample in mg cm3. (Adapted from [40] Copyright 2012 Elsevier Ltd)

a

b

n

2

NB(CH2)nNB n H:C (atom ratio)

2 1.375

2 2

6 1.50

10 1.583 NB3 H:C (atom ratio)

1.529

c NON: H:C (atom ratio)

+ 1.333

compounds. Difunctional NON, prepared via the Diels-Alder reaction of 1,5-cyclooctadiene with cyclopentadiene, was a mixture of three products (Fig. 23.19c). The H:C atomic ratios of those monomers (included in Fig. 23.20), and thereby of their ROMP-derived polymers, were already up to those achieved by hydrogenation of pDCPD. Polymers were prepared in toluene with GC-II, and it is noted that while pDCPD is a mostly linear polymer with some crosslinks (referred to as “interspersed” [41]), these new

+ 1.333

1.20

ROMP-derived polymers in that work were expected to be highly crosslinked by design. Morphologically all samples consisted of fiber-like strings of nanobeads. Experimenting with target densities ranging from 5 to 50 mg cm3, the lowest density aerogels were obtained from the shortest-tether NB(CH2)2NB monomer (6 mg cm3), followed by NB(CH2)6NB (16 mg cm3) and NB(CH2)10NB (25 mg cm3). At the highest target-density end of that study (50 mg cm3) the experimental bulk

ROMP-Derived Aerogels

611

densities were 99, 150, and 160 mg cm3 for n ¼ 2, 6, and 10, respectively. The BET surface area of samples prepared at a target density equal to 30 mg cm3 were 330, 130, and 99 m2 g1, with actual densities equal to 40, 68, and 56 mg cm3 for n ¼ 2, 6, and 10, respectively. More recently, NB(CH2)2NB aerogels at a density of 20 mg cm3 were investigated for nanoconfinement of liquid hydrogen within their porosity [42]. The properties of the NB3 aerogels were intermediate of those of NB(CH2)2NB and NB(CH2)5NB. The lowest density NB3 aerogels (8.2 mg cm3) were obtained from 5 mg cm3 sols; with 30 mg cm3 sols, the BET surface areas were equal to 200 m2 g1. From the monomer system with the most rigid bridge (NON), stable gels were obtained down to a target density of 4 mg cm3, and no shrinkage was noted after drying with supercritical CO2. For comparison purposes, the BET surface area of NON-aerogels prepared from sols formulated with a target density at 30 mg cm3 was the highest in this NB-based series of samples, and equal to 370 m2 g1. In summary, the material properties of those highly crosslinked, highly aliphatic ROMP-derived aerogels have been considered as an interplay of crosslinker functionality (2 versus 3 norbornenes) and the rigidity or flexibility of the aliphatic crosslinks. This conclusion resonates with conclusions drawn from other ROMP-derived systems (see Sect. 23.3.2 below) as well as from a broad array of other polymeric aerogels (see for example ▶ Chap. 21), and therefore can be considered as conceptually portable. As far as the ROMP-derived polymeric backbone is concerned, in general that has been considered rigid, but that view has been challenged as reviewed in the next section.

As reviewed in the previous section, the question of how low one can go in density with ROMP-derived aerogels has been addressed via the number, length, and flexibility of interchain links. The olefinic backbone of ROMP-derived aerogels is generally considered rigid. However, as was described in Sect. 23.2.3, pDCPD aerogels prepared with GC-II swelled significantly during solvent exchanges with toluene (up to 200% in terms of volume – see Fig. 23.21a). This type of pDCPD aerogels consist of a mixture of cis and trans backbone olefins (Fig. 23.6). In contrast, mostly trans pDCPD gels prepared with GC-I did not swell at all. Since the two classes of gels did not differ in any other significant way, but only in the relative amount of the cis versus the trans configuration of the backbone olefins, the different swelling behavior (and subsequent disorderly shrinking and deformation) were attributed to the different backbone configuration. Following the trend established by pDCPD prepared with GC-II versus GC-I, the Paraskevopoulou group at the National and Kapodistrian University of Athens, Greece, investigated the swelling behavior of pDCPD sol–gel materials prepared with W2/PA (see Fig. 23.4), consisting of mostly cis backbone olefins (see Fig. 23.6). W2/PA-derived pDCPD wet gels swelled by 12,000–14,000% in toluene (Fig. 23.21b), meaning that their volume expanded by over 120 times (Fig. 23.22) [25]. This behavior was quite intriguing. A multi-solvent systematic study followed using W2/PA-

b 500

200 3’tol

2’tol

150 100 50

3tol 2tol 1tol

150

4tol

100

in MMA bath 1acet 2acet 3

50

acet

0

Diameter (% vs. mold)

200

0 0

15000

4’tol

1’tol

Volume (% vs. mold)

Diameter (% vs. mold)

a

23.2.6 The Rigidity of the Olefinic Backbone: Response of Mostly cis pDCPD Xerogels and Aerogels to Solvents

20 40 60 80 100 120 140 Time (h)

Fig. 23.21 (a) Swelling data in toluene for pDCPD wet gels prepared with GC-II, and their de-swelling in acetone. Washes and solvents are indicated with numerals and subscripts (tol for toluene and acet for acetone). Note that gels kept on swelling in a MMA bath for crosslinking with PMMA. The last point at the far right (filled-square)

400

12500

300

10000 7500

200 5000 100

2500

Volume (% vs. mold)

23

0

0 0

50 100 150 200 250 300 350 Time (h)

corresponds to the dry aerogel. (b) Analogous swelling data in toluene for pDCPD wet gels prepared with W2/PA in toluene and de-swelling in pentane. (Adapted from [25] Copyright 2017 Elsevier, and from [29] Copyright 2013 The Royal Society of Chemistry)

23

612

N. Leventis and G. L. Gould

Dry 0.8 cc

5 days 47 cc

8 days 102 cc

10 days 120 cc

Fig. 23.22 Swelling in toluene of a pDCPD xerogel prepared with W2/PA. (Adapted from [25] Copyright 2017 Elsevier B.V)

120 100 80

qmax

derived pDCPD xerogels, and focused mainly on aromatic, chlorinated, and brominated hydrocarbons [30]. As summarized in Fig. 23.23, in addition to toluene, extreme swelling was observed also with chloroform where swollen gels exceeded by more than 50 times the volume of the parent xerogel. It was noted that because of that extreme swelling, the volume of the pDCPD xerogel required for the uptake of a given volume of solvent is very small – in many cases 1/100 or less versus other absorbents reported in the literature including organic polymers or carbon-based materials. The swelling behavior of those mostly cis pDCPD materials was studied with the Hansen solubility parameter (HSP) theory and the Flory theory, each of which provided insight into the swelling mechanism and the parameters that affected it. Specifically, on the basis of 44 different aliphatic and aromatic organic solvents, including those included in Fig. 23.23, a correlation was derived between the swelling behavior of the mostly cis pDCPD xerogels in each solvent and the solvent HSP. Those correlations led to the estimation of the HSP of mostly cis pDCPD at δD ¼ 18.15, δP ¼ 3.69, and δH ¼ 3.55. On the basis of those results, W2/PA-derived pDCPD xerogels were used to separate organic solvents from water, but also from oil and hexane [30]. With an eye on the possible application of W2-derived sol–gel materials in chemical sensors and solvent responsive actuators, solvent swelling was studied in similar materials in aerogel form; the rationale was that for these applications the relative advantage of aerogels versus xerogels or dense

60 40

a. toluene b. chloroform c. bromobenzene d. carbon disulfide e. 1,3-dichlorobenzene f. carbon tetrachloride g. chlorobenzene h. 1,2-dibromoethane i. tetrahydrofuran j. benzene k. ethyl bromide l. 1-bromobutane

m. methylene dichloride n. 1,3,5-trimethylbenzene o. 1,4-dimethylbenzene p. 1,3-dimethylbenzene q. 1,2-dichlorobenzene r. benzyl chloride s. cyclohexane t. 1,4-dioxane u. cyclohexanone v. 1,2-dichloroethane w. pyridine

20 0

a b c d e f g h i j k l mn o p q r s t u vw

Fig. 23.23 Swelling behavior of W2/PA-derived pDCPD xerogels in various organic solvents. qmax is the ratio of maximum-to-initial volume of a xerogel monolith. Maximum volume is the volume reached either at the point of disintegration, or at the point of equilibrium. (Solvents in which gels kept on expanding until they disintegrated are marked with asterisks “*”.) Color-coding denotes the scores given to the various solvents according to the Hansen theory based on their ability to cause swelling (red, score “1”; blue, score “2”; green, score “3”; orange, score “4”; cyan, score “5” – see Reference [30]). (Adapted from [30] Copyright 2019 The American Chemical Society))

polymers is their open porosity, which provides rapid access of the solvent to their interior, which in turn would accelerate the swelling response to the solvents of interest [43].

23

ROMP-Derived Aerogels

e en lu to HF T e en yl l x p- PhC Br Ph e en yl -x e m ylen it e es id m hlor e ic id ed rom e n n e yl yl b tha eth eth me orm ro f lo ro e ch lo an di ch but 0 o om br 1

100

613

200 300 400 Volume increase (%)

500

600

Fig. 23.24 Volume increase of thin pDCPD aerogel disks prepared with the W2/NBD catalytic system in various organic solvents at 10 min (shaded length of each bar) and at 60 min (full length of each bar) [43]

Furthermore, in order to halt uncontrollable swelling to disintegration, ROMP was catalyzed with W2/NBD (see Fig. 23.4). The use of norbornadiene (NBD) as a co-initiator versus the more frequently used phenylacetylene (PA) enhances the degree of crosslinking between ROMPderived chains, yielding materials with increased resistant to expansion. Indeed, W2/NBD-derived pDCPD aerogels absorb selected organic solvents (e.g., toluene, chloroform, tetrahydrofuran) and swell rapidly, in some cases up to four times their original volume within just 10 min (Fig. 23.24).

23.3

ROMP-Derived Copolymers

For the purposes of this chapter, ROMP-derived copolymers are classified in two categories: random ROMP copolymers (Sect. 23.3.1) and well-defined ROMP copolymers (Sect. 23.3.2).

23.3.1 Aerogels from ROMP-Derived Random Copolymers Random copolymers of pDCPD with NB and various NB derivatives were investigated at the LLNL as part of a project related to aerogel coatings used for the fabrication of ignition targets for inertial confinement fusion experiments [44]. ROMP-derived polymers were selected for that application because of their mild polymerization conditions and their ability to form statistical copolymers with tunable properties (Fig. 23.25). Specifically, the primary objective of those efforts was to coat nonplanar surfaces with thin and homogeneous aerogel

films. That requires continuous rotation of the object to be coated, which translates into that without proper control of the viscosity and gel time, shear forces can damage the growing gel networks and prevent the formation of uniform coatings. As it turned out, both the moduli and viscosity at the gel point could be tuned through copolymerization of DCPD with several norbornene comonomers, a list of which is shown in Fig. 23.26 [44]. Along those efforts, a parallel objective was to develop a means of doping those coatings with high-Z elements for the nondestructive evaluation of their quality with x-ray radiography; iodine-containing monomer “I” and tin-containing monomer “J” (Fig. 23.26) were developed for this purpose [39]. All copolymer gels were obtained with GC-I in toluene. GC-I was chosen over GC-II because of the shorter gelation times. Nevertheless, gelation time remained more as an empirical parameter rather than a predictable quantity. For example, although NB (shown as comonomer A in Fig. 23.26) has a much higher ROMP-reactivity than DCPD, the gelation time of DCPD itself at a target density of 50 mg cm3 was 491 s (determined with rheometry), while p(DCPD-r-NB) copolymers prepared at the same target density with 100:10 to 100:40 w/w mixtures of DCPD:NB gelled in about 2100 s. (As mentioned in Sect. 23.2.4 above, pNB is soluble in toluene, and therefore NB itself does not gel in neat toluene.) On the other hand, addition of NB-based comonomer B (which, actually serves as a crosslinker – see Fig. 23.26) reduced the gelation time to about 1 min. Both the storage shear modulus (G’) and the viscosity at the gelation point increased over 100-fold even with the lowest concentration of NB (DCPD:NB ¼ 100:10 w/w). The higher viscosity of p(DCPD-r-NB) sols near the gelation point reduced the shear experienced by the growing polymer network and made it possible to fabricate uniformly coated wet-gel layers on the interior surfaces of cylindrical substrates (e.g., inside vials) [44, 45]. p(DCPD-r-NB) copolymer aerogels had a fibrous three-dimensional porous network similar to that of pure pDCPD aerogels. However, the length of the individual fiber strands decreased from a few μm in the case of pure pDCPD to hundreds of nm in p(DCPD-r-NB) aerogels, and kept on decreasing as the concentration of the NB monomer increased. In terms of other material properties, the change in morphology was accompanied by a drastic increase in density and an equally dramatic drop in surface areas. For example, at a constant target density of 50 mg cm3, and for DCPD:NB ratios equal to 100:0, 100:20, 100:30, and 100:40 w/w, the respective experimental densities (surface areas) of p(DCPD-r-NB) aerogels were: 37(103), 82(102), 102(57) and 244(22) mg cm3(m2 g1). The distinct features in p(DCPD-r-NB-R) copolymer networks were changes in the length of the fiber strands, and pore sizes and morphologies surrounded by the strands. Thus, aerogels prepared using comonomer “C” (R ¼ methyl

23

614

N. Leventis and G. L. Gould

+

GC-I

a

Toluene

R

R DCPD

n

NB-R

a

1-n

p(DCPD-r-NB-R)

Fig. 23.25 Formation of random copolymers, p(DCPD-r-NB-R), via ROMP of DCPD and various substituted norbornenes (NB-R: see Fig. 23.26) [44]

Fig. 23.26 Comonomers, NB-R, derivatives of norbornene, used for the synthesis of random copolymers with DCPD: p (DCPD-r-NB-R) – see Fig. 23.25 [39, 44]

O

R

R: CH3, C6H13, C12H25, benzyl, tosyl A (NB)

B

C

I

D

E

F

G

SnBu3

I I

H

– Fig. 23.26) showed the characteristic fibrous network morphology observed in pure pDCPD and p(DCPD-r-NB) at low NB concentrations, while comonomers “D” and “E,” which contain longer alkyl chain groups (hexyl and dodecyl, respectively), created aerogels with shorter fibrous segments between interconnects, and more entangled networks. In terms of pore structure, entanglement of longer linear ligaments in pDCPD led to the development of large, irregularshaped pores. However, the addition of comonomers such as NB and compounds “C”–“E” produced drastically different pore shapes and sizes as a result of the increased entanglement of shorter fibrous segments. Addition of comonomers “F” and “G,” with benzyl and tosyl groups, respectively, introduced strong hydrophobic and electronic interactions that resulted in spherical pores with sizes of about 50 nm. Addition of iodomethyl-monomer “H” produced extended long fibers reminiscent of those in pure pDCPD. Similarly, addition of diiodomethyl-monomer “I,” and tetrabutyltinmonomer “J” produced fibrous-shape morphologies, similar to other p(DCPD-r-NB-R) aerogels. Out of the entire series of comonomers of Fig. 23.26, only comonomer “B” comprised an aberration: albeit a similar reactivity of two NBs, the corresponding copolymers with DCPD did not show a

I

J

fibrous morphology, but rather a fine particulate morphology, often observed in, for example, silica aerogels.

23.3.2 Aerogels from Well-Defined ROMPCopolymers There are two classes of well-defined ROMP-copolymerbased aerogels: poly(imide-norbornene) aerogels based on a rod-like monomer (section “Poly(Imide-Norbornene) Aerogels”), and poly(urea-norbornene) aerogels based on star and dendritic monomers (section “Poly(UrethaneNorbornene) Aerogels”).

Poly(Imide-Norbornene) Aerogels Poly(imide-norbornene) were the first ROMP-derived aerogels in this class of materials, and were prepared from an imide-norbornene monomer in NMP at 60–90  C using GC-II as catalyst as shown in Fig. 23.27a [24]. The design of these polymers was inspired from PMR-15, a hightemperature thermoset polyimide resin rated at 290  C for 10,000 h and used by the aerospace industry in hightemperature applications [46]. PMR-15 is prepared by heat-

23

ROMP-Derived Aerogels

a

615

O

O

O N

GC-II NMP, 60-90 °C

N

O

Imide-norbornene monomer

O

O N

N

O

O

b

H2C

H2 C

2n

Poly(imide-norbornene)

23

O

O

N

N

H2 C

O N

O 2 O O O Norbornene-capped imidized-oligomer of molecular weight ~1,500 >300 °C Fully-crosslinked PMR-15 via olefin addition of the norbornene end caps Fig. 23.27 (a) Imide-norbornene monomer and the polymerization (gelation) conditions to a ROMP-derived network [24]. (b) Analogous thermoset resins used in aerospace industry that comprised the impetus for (a) [46, 47]

induced crosslinking (at >300  C) of norbornene capped imide oligomers with a molecular weight of ~1500 according to Fig. 23.27b [47]. It was reasoned that ROMP would induce crosslinking of the dangling NB moieties at milder conditions, and the imide-norbornene monomer of Fig. 23.27a was designed as a simplified version of the PMR-15 precursor. The concentration of the imide-norbornene monomer in the sol was varied from 2.5% to 20% w/w, and the density of the resulting aerogels varied in the 0.13–0.66 g cm3 range. Wet gels shrunk significantly relative to their molds (28–39% in linear dimensions and in reverse order to the sol concentration), but the final aerogels retained high porosities (50–90% v/v) and high surface areas (210–632 m2 g1, of which up to 25% was assigned to micropores). The pore size distributions in the mesoporous range had maxima from 20 nm to 33 nm. The skeletal framework was particulate (Fig. 23.28a). According to small angle neutron scattering (SANS – Fig. 23.28b) the finest structures in SEM consisted of primary particles 16–17 nm in diameter that assembled to form secondary aggregates 60–85 nm in diameter. As we saw in Sect. 23.2.3 above, densely packed surface-fractal secondary particles are not unusual in ROMP-derived aerogels. In that regard, poly(imide-norbornene) aerogels comprise an interesting case, which, within the available density range, captures a mass-to-surface fractal transition (determined from the slope of Region III of the spectra in Fig. 23.28b). That is, at lower densities (e.g., 0.26 g cm3), secondary particles were mass fractals, with mass-fractal dimension, Dm ¼ 2.34  0.03, turning to closed-packed surface fractal

objects with surface fractal dimension, DS ¼ 2.9  0.01 as the bulk density increased to 0.34 g cm3, suggesting a change in the network-forming mechanism from diffusionlimited aggregation of primary particles to a space-filling bond percolation model. At the highest density (0.660 g cm3), secondary particles were non-fractal objects (the slope of Region III was equal to 3.0). Poly(imidenorbornene) aerogels combined facile one-step synthesis with heat resistance up to 200  C, high mechanical compressive strength and specific energy absorption (168 MPa and 50 J g1, respectively, at 0.39 g cm3 and 88% ultimate strain), low speed of sound (351 m s1 at 0.39 g cm3) and styrofoam-like thermal conductivity (0.031 W m1 K1 at 0.34 g cm3 and 25  C).

Poly(Urethane-Norbornene) Aerogels Poly(imide-norbornene) aerogels were followed by poly(urethane-norbornene) aerogels. Those were prepared first from star, and subsequently from dendritic-type urethane monomers bearing respectively three and nine norbornene (NB) moieties as end groups (Fig. 23.29) [48–51]. Those monomers were prepared via Diels-Alder reactions of the corresponding urethane-acrylates [48, 52], and the poly(urethane-norbornene) aerogels were studied in parallel to their poly(urethane-acrylate) analogues (see ▶ Chap. 21). The core of the urethane-norbornene monomers, and therefore the nodes along the polymeric network, was based on a flexible/aliphatic (aL-), and a rigid/aromatic (aR-) triisocyanate (Desmodur N3300A and Desmodur RE,

616

N. Leventis and G. L. Gould

a

r b = 0.261 g cm–3

r b = 0.341 g cm–3

10

IV

III

II

I(Q) (cm–1)

100 Slope = –3.1

10

1 IV

III

I

r b = 0.660 g cm–3

1000

I(Q) (cm–1)

II

100

Slope = –3.0

10

1 IV

III

II

0.01

I 0.1

Q Fig. 23.28 (a) SEM of poly(imide-norbornene) aerogels at three different densities, as shown. (b) Corresponding small angle neutron scattering (SANS) profiles. Vertical lines serve as guides to the eye rather than delineating the exact cut offs between the four regions. The actual values of interest from each region were obtained by fitting the entire curves according to the Beaucage Unified Model. All slopes in Region I obeyed Porod’s law (they were equal to 4.0  0.1). The

I

r b = 0.341 g cm–3

1000

r b = 0.660 g cm–3

500 nm

Slope = –2.34

100

1

500 nm

2.5 ×

r b = 0.261 g cm–3

1000

500 nm

2.5 ×

SANS

b

I(Q) (cm–1)

2.5 ×

SEM

(Å–1)

curvature of Regions II and IV gives the size of the primary and secondary particles, respectively, and the slope of Region III gives the fractal dimension of the secondary particles (if |slope| < 3.0, then Dm ¼ | slope|; if |slope| ¼ 3.0, non-fractal object; if 6 < slope < 3, then DS ¼ 6 – |slope|). (Adapted from [24] Copyright 2011 The American Chemical Society)

23

ROMP-Derived Aerogels

617

Triisocyanate cores

Monomer design

(H2C)6N=C=O 9- vs 3-

NB moieties

O

O=C=N(H2C)6

O

N N

N

Flexible/aliphatic (aL)

(CH2)6N=C=O

O

Urethane bridge

Desmodur N3300A

23

N=C=O Desmodur RE

Triisocyanate core

Rigid/aromatic (aR) N=C=O

O=C=N

Fig. 23.29 The design of urethane-norbornene monomers with 9- versus 3-dangling norbornene (NB) moieties, based on two triisocyanate cores: a flexible/aliphatic (Desmodur N3300A), and a rigid/aromatic (Desmodur RE) [48–51]

O O O O

O

O

O

O

O HN(CH)6 O N O

O

HN(CH)6 O N O

O

O

O

O O

N (CH ) NH HN(H2C)6 N 2 6 O O O O O

O

O O

HN(H2C)6 O O

O

N

N O

(CH2)6NH

O

O

O

O

O

O

O

O O

O O

aL-3-NB

aL-9-NB

Fig. 23.30 Urethane-norbornene monomers with 3- and 9-polymerizable norbornene moieties illustrated with the flexible/aliphatic core of Fig. 23.29 [48–51]

respectively – Fig. 23.29). Figure 23.30 shows two of the four monomers involved in those studies. Polymerization of the 9-NB group terminated monomers was carried out in toluene and was catalyzed with GC-I [51]. 3-NB-terminated

monomers were insoluble in toluene and gelation was carried out in acetone with GC-II [48]. All aerogels of this study consisted of aggregates of nanoparticles, whose size depended on the aliphatic/aromatic

618 Fig. 23.31 SEM images at 50 k magnification of poly(urethanenorbornene) aerogels (aL-3-NB – refer to Fig. 23.30) at two density extremes. Other material data below the images. (ρs: bulk density; r: radii from skeletal density/N2-sorption data; R(1), R(2): radii of primary and secondary particles, respectively, from SAXS data; σ: BET surface areas from N2-sorption data.) (Adapted from [48] Copyright 2014 The American Chemical Society)

N. Leventis and G. L. Gould

1 mm r b = 0.128 ± 0.002 g cm−3 r = 118 nm R(1) = 16.42 ± 0.92 nm R(2) = 75.1 ± 8.42 nm s = 21 m2 g−1

content of the monomer, the rigidity/flexibility of the polymeric backbone, but most importantly on the number of functional groups at the tips of the branches of the monomers. Typical SEM images from this class of materials and other pertinent data of the corresponding samples are shown in Fig. 23.31. Focusing on the two 9-NB monomers, at low densities ( 90% v/v), and were macroporous materials; aerogels based on the flexible/ aliphatic core were fragile, whereas aerogels containing the rigid/aromatic core were reported as plastic, and at even lower densities (0.03 g cm3) as foamy [51]. At higher densities (0.2–0.7 g cm3) all materials were stiff and strong. In terms of micro/nanomorphology, at low monomer concentrations both aL-9-NB and aR-9-NB consisted of discrete primary particles that formed spherical secondary aggregates. At higher monomer concentrations the structure of both samples consisted of same-size fused particles, at the same overall size of the previous (i.e., at lower concentrations) secondary aggregates. It was thus concluded that it was not the aliphatic or aromatic core that determined the point of phase separation, but rather the solubility of the polymeric backbone that in both materials was the same (polynorbornene). It was proposed that irrespective of aL or aR, the solubility of the developing polymer was low, causing early phase-separation of a primary particle network that was later coated with yet unreacted monomer and oligomers fusing the particles together. Corroborating evidence for the solubility hypothesis came from comparison of the material properties of aerogels from aL-9-NB and aR-9-NB with those from aL-3-NB and aR-3-NB. For example, referring to aerogels from the same or similar monomer concentrations (10–20% w/w), materials from the

1 mm r b = 0.792 ± 0.010 g cm−3 r = 103 nm R(1) = 18.8 ± 2.25 nm R(2) = 53.7 ± 4.57 nm s = 24 m2 g−1

9-NB group of monomers had higher porosities (68–98% vs. 44–89% v/v in the 3-NB-terminated analogues), and significantly higher BET surface areas (54–300 m2 g1) than their analogues from aL-3-NB and aR-3-NB (21–60 m2 g1). The BET surface area comparison reflects smaller particle radii in the former group of materials (28 nm) than those in the 3-NB-based aerogels (118 nm – see Fig. 23.31). Overall poly(urethane-norbornene) aerogels strongly support conclusions reached with several other aerogels previously, namely that increasing the functional group density of the monomer (i.e., the number of functional groups of the monomer relative to its size) increases crosslinking per unit mass, and thereby decreases solubility of the developing polymer and causes formation of networks with smaller particles and higher BET surface areas, practically independent of the exact chemical identity of the monomer (see ▶ Chap. 21). Data with poly(urethane-norbornene) also support that aerogels with smaller particles have overall lower thermal conductivities [48].

23.4

Conclusions and Outlook

ROMP as a means to polymeric aerogels has been expanded from polymerization of “traditional” substrates like DCPD and NB to designer copolymers. Issues along the way, like extreme swelling of wet gels leading to deformed aerogels, have been studied in depth and addressed via fundamental studies of molecular-nanoscopic-bulk property relationships. The propensity of certain ROMP-derived sol–gel materials to swell in selected solvents was developed into selective solvent adsorption and investigated for solvent-responsive actuators. Other more conventional applications of aerogels, as

23

ROMP-Derived Aerogels

for example in thermal insulation, have been hampered by environmental instability through air oxidation of the backbone olefins. Saturation of the double bonds with hydrogen via reduction in the wet-gel state has addressed the oxidation issue, but as it currently stands the process requires solvent exchanges, the reaction time is long, and the resulting saturated random poly(DCPD-r-NB) copolymers it has been applied to are dimensionally unstable toward heating and shrinkage. However, hydrogen addition to ROMP-derived aerogels rigidized by other means (e.g., see Sect. 23.3.2 on well-defined copolymers) is probably a viable approach to bring ROMP-derived aerogels to commercialization.

References 1. Vougioukalakis, G.C., Grubbs, R.H.: Ruthenium-based heterocyclic carbene-coordinated olefin metathesis catalysis. Chem. Rev. 110, 1746–1787 (2010) 2. Furstner, A.: Olefin metathesis and beyond. Angew. Chem. Int. Ed. 39, 3012–3043 (2000) 3. Grubbs, R.H.: Olefin metathesis. Tetrahedron. 60, 7117–7140 (2004) 4. Sutthasupa, S., Shiotsuki, M., Sanda, F.: Recent advances in ringopening metathesis polymerization, and application to synthesis of functional materials. Polym. J. 42, 905–915 (2010) 5. https://www.nobelprize.org/prizes/chemistry/2005/summary/. Accessed 10 Apr 2020 6. https://www.nobelprize.org/uploads/2018/06/grubbs-slides.pdf. Accessed 10 Apr 2020 7. Slugovc, C.: The ring opening metathesis polymerization toolbox. Macromol. Rapid Commun. 25, 1283–1297 (2004) 8. Natta, G., Dall’Asta, G., Mazzanti, G.: Stereospecific homopolymerization of cyclopentene. Angew. Chem. Int. Ed. 3, 723–729 (1964) 9. Mutch, A., Leconte, M., Lefebvre, F., Basset, J.M.: Effect of alcohols and epoxides on the rate of ROMP of norbornene by a ruthenium trichloride catalyst. J. Mol. Catal. Chem. 133, 191–199 (1998) 10. Michelotti, F.W., Keaveney, W.P.: Coordinated polymerization of the bicyclo-[2.2.1]-heptene-2 ring system (norbornene) in polar media. J. Polym. Sci. A. 3, 895–905 (1965) 11. Chauvin, Y.: Olefin metathesis: the early days (Nobel lecture). Angew. Chem. Int. Ed. 45, 3741–3747 (2006) 12. Fischer, E.O., Maasböl, A.: On the existence of a tungsten carbonyl carbene complex. Angew. Chem. Int. Ed. 3, 580–581 (1964) 13. Schrock, R.R.: Multiple metal-carbon bonds for catalytic metathesis reactions (Nobel lecture). Angew. Chem. Int. Ed. 45, 3748–3759 (2006) 14. Grubbs, R.H.: Olefin-metathesis catalysts for the preparation of molecules and materials (Nobel lecture). Angew. Chem. Int. Ed. 45, 3760–3765 (2006) 15. Hine, P.J., Leejarkpai, T., Khosravi, E., Duckett, R.A., Feast, W.J.: Structure property relationships in linear and cross-linked poly (imidonorbornenes) prepared using ring opening metathesis polymerisation (ROMP). Polymer. 42, 9413–9422 (2001) 16. Kessler, M.R., White, S.R.: Cure kinetics of the ring-opening metathesis polymerization of dicyclopentadiene. J. Polym. Sci. Part A: Polym. Chem. 40, 2373–2383 (2002) 17. Walsh, D.J., Lau, S.H., Hyatt, M.G., Guironnet, D.: Kinetic study of living ring-opening metathesis polymerization with third-generation Grubbs catalysts. J. Am. Chem. Soc. 139, 13644–13647 (2017)

619 18. Raptopoulos, G., Grigoropoulos, A., Mertis, K., Paraskevopoulou, P., Pitsikalis, M.: Multinuclear transition metal catalysts for metathesis polymerization. Current developments and future perspectives. Recent Res. Dev. Polym. Sci. 12, 83–106 (2014) 19. Lee, J.K., Gould, G.L.: Polydicyclopentadiene based aerogel: a new insulation material. J. Sol-Gel Sci. Technol. 44, 29–40 (2007) 20. Lee, J.K., Gould, G.L.: Microporous polydicyclopentadiene-based aerogels. US Patent No 8,461,223 B2 (2013) 21. Lee, J.K., Gould, G.L.: Microporous polyolefin-based aerogels. US Patent No 9,469,739 B2 (2016) 22. Mohite, D., Larimore, Z., Leventis, N., Sotiriou-Leventis, C.: Strong silica aerogels crosslinked with polynorbornene via ring opening metathesis polymerization (ROMP). Polym. Preprints. 51, 469–470 (2010) 23. Mohite, D., Larimore, Z., Lu, H., Mang, J.T., Sotiriou-Leventis, C., Leventis, N.: Monolithic hierarchical fractal assemblies of silica nanoparticles cross-linked with polynorbornene via ROMP: a structureproperty correlation from molecular to bulk through nano. Chem. Mater. 24, 3434–3448 (2012) 24. Leventis, N., Sotiriou-Leventis, C., Mohite, D.P., Larimore, Z.J., Mang, J.T., Churu, G., Lu, H.: Polyimide aerogels by ring-opening metathesis polymerization (ROMP). Chem. Mater. 23, 2250–2261 (2011) 25. Raptopoulos, G., Anyfantis, G.C., Chriti, D., Paraskevopoulou, P.: Synthesis and structural characterization of poly(dicyclopentadiene) gels obtained with a novel ditungsten versus conventional W and Ru mononuclear catalysts. Inorg. Chim. Acta. 460, 69–76 (2017) 26. Bang, A., Mohite, D., Saeed, A.M., Leventis, N., Sotiriou-Leventis, C.: Polydicyclopentadiene aerogels from first- versus secondgeneration Grubbs’ catalysts: a molecular versus a nanoscopic perspective. J. Sol-Gel Sci. Technol. 75, 460–474 (2015) 27. Davidson, T.A., Wagener, K.B., Priddy, D.B.: Polymerization of dicyclopentadiene: a tale of two mechanisms. Macromolecules. 29, 786–788 (1996) 28. Davidson, T.A., Wagener, K.B.: The polymerization of dicyclopentadiene: an investigation of mechanism. J. Mol. Catal. A Chem. 133, 67–74 (1998) 29. Mohite, D.P., Mahadik-Khanolkar, S., Luo, H., Lu, H., SotiriouLeventis, C., Leventis, N.: Polydicyclopentadiene aerogels grafted with PMMA: I. Molecular and interparticle crosslinking. Soft Matter. 9, 1516–1530 (2013) 30. Chriti, D., Raptopoulos, G., Anyfantis, G.C., Paraskevopoulou, P.: An extreme case of swelling of mostly cis-polydicyclopentadiene by selective solvent absorption – application in decontamination and environmental remediation. ACS Appl. Polym. Mater. 1, 1648–1659 (2019) 31. Mohite, D., Larimore, Z., Sotiriou-Leventis, C., Leventis, N.: Polydicyclopentadiene aerogels grafted with PMMA. PMSE. 105, 1054–1056 (2011) 32. Taghvaee, T., Donthula, S., Rewatkar, P., Majedi Far, H., SotiriouLeventis, C., Leventis, N.: K-index: a descriptor, predictor, and correlator of complex nanomorphology to other material properties. ACS Nano. 13, 3677–3690 (2019) 33. Leventis, N., Sotiriou-Leventis, C., Chandrasekaran, N., Mulik, S., Larimore, Z.J., Lu, H., Churu, G., Mang, J.T.: Multifunctional polyurea aerogels from isocyanates and water. A structure-property case study. Chem. Mater. 22, 6692–6710 (2010) 34. Mohite, D.P., Mahadik-Khanolkar, S., Luo, H., Lu, H., SotiriouLeventis, C., Leventis, N.: Polydicyclopentadiene aerogels grafted with PMMA: II. Nanoscopic characterization and origin of macroscopic deformation. Soft Matter. 9, 1531–1539 (2013) 35. Mandal, C., Donthula, S., Rewatkar, P.M., Sotiriou-Leventis, C., Leventis, N.: Experimental deconvolution of depressurization from capillary shrinkage during drying of silica wet-gels with SCF CO2 – why aerogels shrink? J. Sol-Gel Sci. Technol. 92, 662–680 (2019)

23

620 36. Lu, X., Caps, R., Fricke, J., Alviso, C.T., Pekala, R.W.: Correlation between structure and thermal conductivity of organic aerogels. J. Non-Cryst. Solids. 188, 226–234 (1995) 37. Schwab, P., Grubbs, R.H., Ziller, J.W.: Synthesis and applications of RuCl2(CHR′)(PR3)2: the influence of the alkylidene moiety on metathesis activity. J. Am. Chem. Soc. 118, 100–110 (1996) 38. Luo, H., Zhang, Y., Wang, B., Lu, H.: Characterization of the compressive behavior of glass fiber reinforced polyurethane foam at different strain rates. J. Offshore Mech. Arctic Eng. 132, 021301 (2010) 39. Kim, S.H., Shin, S.J., Lenhardt, J.M., Braun, T., Sain, J.D., Valdez, C.A., Leif, R.N., Kucheyev, S.O., Wu, K.J.J., Biener, J., Satcher, J. H., Hamza, A.: Deterministic control over high-Z doping of polydicyclopentadiene-based aerogel coatings. ACS Appl. Mater. Interfaces. 5, 8111–8119 (2013) 40. Lenhardt, J.M., Kim, S.H., Nelson, A.J., Singhal, P., Baumann, T.F., Satcher Jr., J.H.: Increasing the oxidative stability of poly (dicyclopentadiene) aerogels by hydrogenation. Polymer. 54, 542–547 (2013) 41. Lenhardt, J.M., Kim, S.H., Worsley, M.A., Leif, R.N., Campbell, P. G., Baumann, T.F., Satcher Jr., J.H.: ROMP crosslinkeres for the preparation of aliphatic aerogels. J. Non-Cryst. Solids. 408, 98–101 (2015) 42. Kucheyav, S.O., Lenhardt, J.M.: Freezing hydrogen in nanoconfinement. Fusion Sci. Technol. 73, 293–297 (2018) 43. Chriti, D., Raptopoulos, G., Brandenburg, B., Paraskevopoulou, P.: Large, rapid swelling of high-cis polydicyclopentadiene aerogels suitable for solvent-responsive actuators. Polymers. 12, 1033 (2020) 44. Kim, S.H., Worsley, M.A., Valdez, C.A., Shin, S.J., Dawedeit, C., Braun, T., Baumann, T.F., Letts, S.A., Kucheyev, S.O., Wu, K.J.J., Biener, J., Satcher Jr., J.H., Hamza, A.V.: Exploration of the versatility of ring opening metathesis polymerization: an approach for gaining access to low density polymeric aerogels. RSC Adv. 2, 8672–8680 (2012) 45. Dawedeit, C., Kim, S.H., Braun, T., Worsley, M.A., Letts, S.A., Wu, K.J., Walton, C.C., Chernov, A.A., Satcher Jr., J.H., Hamza, A.V., Biener, J.: Tuning the rheological properties of sols for low-density aerogel coating applications. Soft Matter. 8, 3518–3521 (2012) 46. Meador, M.A.: Recent advances in the development of processable high-temperature polymers. Annu. Rev. Mater. Sci. 28, 599–630 (1998) 47. Meador, M.A.B., Johnston, J.C., Cavano, P.J.: Elucidation of the cross-linking structure of nadic-end-capped polyimides using NMR of 13C-labeled polymers. Macromolecules. 30, 515–519 (1997) 48. Bang, A., Buback, C., Sotiriou-Leventis, C., Leventis, N.: Flexible aerogels from hyperbranched polyurethanes: probing the role of molecular rigidity with poly(urethane acrylates) versus poly(urethane norbornenes). Chem. Mater. 26, 6979–6993 (2014) 49. Leventis, N., Sotiriou-Leventis, C., Bang, A.: Flexible to rigid nanoporous polyurethane-acrylate (PUAC) type materials for structural and thermal insulation applications. US Patent No 9,994,516 (2018) 50. Leventis, N., Sotiriou-Leventis, C., Bang, A.: Flexible to rigid nanoporous polyurethane-acrylate (PUAC) type materials for structural and thermal insulation applications. US Patent No 9,550,846 (2017)

N. Leventis and G. L. Gould 51. Kanellou, A., Anyfantis, G.C., Chriti, D., Raptopoulos, G., Pitsikalis, M., Paraskevopoulou, P.: Poly(urethane-norbornene) aerogels via ring opening metathesis polymerization of dendritic urethane-norbornene monomers: structure-property relationships as a function of an aliphatic versus an aromatic core and the number of peripheral norbornene moieties. Molecules. 23, 7 (2018) 52. Papastergiou, M., Kanellou, A., Chriti, D., Raptopoulos, G., Paraskevopoulou, P.: Poly(urethane-acrylate) aerogels via radical polymerization of dendritic urethane-acrylate monomers. Materials. 11, 2249 (2018)

Nicholas Leventis received his Ph.D. from Michigan State University in organic chemistry in 1985. He is Director, Research at Aspen Aerogels. In 2019 he retired from the Missouri University of Science and Technology as a Curators’ Distinguished Professor of Chemistry. His aerogel work includes polymer-crosslinked aerogels, aerogels from most classes of organic polymers, interpenetrating organic-inorganic aerogels, and metallic, ceramic, and carbon aerogels.

George L. Gould received his Ph.D. from Yale University in inorganic chemistry in 1989. He has led technology development at Aspen Aerogels since 2001, where he is currently the Chief Technology Officer. His primary aerogel-related work includes designing and commercializing fiber-reinforced aerogel composite products as well as their associated scaled manufacturing processes. Other aerogel interests include supercritical drying methodology development, aerogel property metrology, and new applications development for many classes of aerogel materials.

Part V Biopolymer Aerogels

Cellulose Aerogels: Monoliths, Beads, and Fibers

24

Lorenz Ratke, Kathirvel Ganesan, and Maria Schestakow

Contents 24.1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 623

24.2

Dissolution Agents for Cellulose . . . . . . . . . . . . . . . . . . . . . . . . 625

24.3

Gelation of Cellulose Solutions . . . . . . . . . . . . . . . . . . . . . . . . . 627

24.4

Drying Methods and Shrinkage . . . . . . . . . . . . . . . . . . . . . . . . 628

24.5 24.5.1 24.5.2 24.5.3 24.5.4 24.5.5

Monoliths: Preparation and Properties . . . . . . . . . . . . . . . Aerogels from Cellulose Derivatives . . . . . . . . . . . . . . . . . . . . . Aerogels from Alkali Hydroxide Routes . . . . . . . . . . . . . . . . . Aerogels from Salt Melt Hydrate Routes . . . . . . . . . . . . . . . . Aerogels from Ionic Liquid Routes . . . . . . . . . . . . . . . . . . . . . . Mechanical and Thermal Properties . . . . . . . . . . . . . . . . . . . . . .

629 629 630 631 635 638

24.6 24.6.1 24.6.2 24.6.3

Cellulose Aerogel Beads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods of Bead Fabrication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Microstructure and Properties of Beads . . . . . . . . . . . . . . . . . . Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

639 639 639 642

24.7

Cellulose Filaments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 642

24.8

Summary and Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 647

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 648

Abstract

Cellulose aerogels can be produced by using several methods, yielding materials with extremely low densities. Their structure can be described as a type of nanofelt, which means that nanosized fibrils of cellulose are arranged in a random three-dimensional (3D) network with a huge meso- to macro-porosity. The synthesis of cellulose aerogels typically has four steps: dissolution of cellulose precursors down to their polymeric level,

L. Ratke (*) Institute of Materials Research, DLR, German Aerospace Center, Cologne, Germany e-mail: [email protected] K. Ganesan · M. Schestakow German Aerospace Center, DLR, Cologne, Germany e-mail: [email protected]; [email protected]

gelation of the solution by different methods, regeneration in nonsolvents, and drying in a way to preserve the wet gel nanostructure. Cellulose aerogels are prepared as monoliths having sizes in the centimeter to decimeter range, beads with diameters ranging from a few tens of micrometers to a millimeter, and filaments for possible textile applications. This chapter describes the different methods developed in the last decades by research groups worldwide to produce low density cellulose monoliths, beads, and filaments. It presents different methods of cellulose dissolution, gelation, regeneration, and drying as well as the microstructures and properties of cellulose aerogels.

24.1

Introduction

Cellulose in its various modifications is a natural linear macromolecule produced mainly by plants in huge amounts per year, approximately a few teratons. Chemically it is a chain of 1–4linked β-D-glucopyranose, which means that the β-Dglucopyranose ring in its chair conformation is connected via oxygen bridges to a linear chain and hydrogen bonds stiffen the chain [1, 2]. The macromolecule of cellulose consists of repeating cellobiose units (two β-D-glucopyranose rings also called anhydrous glucose unit, AGU, connected by the oxygen bridge form a cellobiose unit). In the cellulose polymer, each AGU has three OH groups. Two of them are directly located at the carbon atoms of the AGU, and one is outward via a methylol group. A sketch of the cellulose base unit is shown in Fig. 24.1. Crystalline cellulose can exhibit essentially four crystallographically distinct polymorphic modifications, cellulose I, more exactly Iα and Iβ, II and III, and IV [4]. The walls of plant cells produce cellulose from glucose units which are a result of photosynthesis. Naturally plants produce so-called cellulose I, which means that the polymers are arranged parallel in sheets and have the same direction. Cellulose fibers in the cell walls together with hemicellulose and lignin

© Springer Nature Switzerland AG 2023 M. A. Aegerter et al. (eds.), Springer Handbook of Aerogels, Springer Handbooks, https://doi.org/10.1007/978-3-030-27322-4_24

623

624

L. Ratke et al.

OH OH O

HO

O

O

HO

O

OH OH

n

Fig. 24.1 Chemical structure of a cellobiose unit: two 1–4-linked β-Dglucopyranose rings. (Figure taken from [3])

are responsible for the extremely good mechanical properties of this three-phase composite allowing especially trees to grow to huge sizes. For commercial applications, cellulose is obtained from cotton, bast fibers, flax, hemp, sisal, and jute or wood. The production of cellulose fibers generally needs a separation of the cellulose from lignin and hemicellulose and other constituents of plants cells. This can be done chemically, and processes were developed to produce a viscous liquid that can be spun. Regeneration leads to fibers consisting of pure cellulose, but generally as a type II cellulose, meaning the polymers are arranged anti-parallel in sheets (rayon and viscose process [5, 6]). Many derivatives of cellulose in the form of esters and ethers were developed having a broad variety of applications (e.g., cellulose acetates and nitrocellulose). Cellulose is the basis for papermaking. Its fibers have high strength and durability. As Zugenmaier [4] writes, they are readily wetted by water, exhibit considerable swelling when saturated, and are hygroscopic, i.e., they absorb appreciable amounts of water, when exposed to the atmosphere. Even in the wet state, natural cellulose fibers show almost no loss in strength, which is important for textile applications. Cellulose of different natural sources exhibits different degrees of polymerization, DP. Cellulose from wood pulp has between 6,000 and 10,000 monomer units and cotton between 10,000 and 15,000. Chemical modifications during processing often lead to a reduction of the degree of polymerization giving products below 1,000 monomer units [4, 5]. All native cellulose is organized into fringes and fibrils [7], which means that there are areas of crystalline order being intermixed with amorphous ones such as shown schematically in Fig. 24.2. Typically the amount of amorphous regions varies in plant cellulose between 40% and 60%. Figure 24.3 shows a schematic breakdown of a cellulose fiber as it can be found in plant cells down to its polymeric unit. Although cellulose is hydrophilic and considerable swelling occurs in water, it is insoluble in water and most organic solvents but is biodegradable, and as such cellulose aerogels

Crystalline area

Amorphous area

Fig. 24.2 A schematic picture of the fringe-fibril model of Hearle [7] showing the local variation of crystalline order of the polymer chains; see [4–6]

Macro-fibril 60 - 400 nm

Micro-fibril Ø 20 - 30 nm length: ª10 µm

Elementary fibril ª 3 - 5 nm

Polymer chain

Fig. 24.3 The figure shows a schematic of a macro-fibril as existing in plant cells, which is a composite of micro-fibrils. These consist of elementary fibrils which are made of 30–40 polymeric linear cellulose chains. (Picture based on the botany visual resource library [3, 8, 9])

would also have the taste of a green material. The cellulose polymer can be broken down chemically into its glucose units with concentrated acids at elevated temperature. Since cellulose is chemically a very stable material, the production of aerogels from cellulose needs a technology or processing

24

Cellulose Aerogels: Monoliths, Beads, and Fibers

route to disintegrate the cellulose into the elementary fibrils and eventually down to the polymeric level without degradation and then to rebuild them into a suitable low density, open porous gel that can be dried to obtain a 3D mesoporous microstructure typical for aerogels.

24.2

Dissolution Agents for Cellulose

Dissolution of cellulose, as stated above, is not simple; however, this is also true for many synthetic polymers. What drives dissolution of a polymer into a solvent? Thermodynamically, always an increase in entropy. The dissolved state possesses a higher entropy compared with the undissolved one. The smaller the DP of a polymer, the higher the entropy gain. Therefore cellulose in its native form with often a very long polymer chain and a high DP has a low solubility since there is a lack of entropic gain during dissolution. But this is only one aspect. The other aspect is that dissolution of polymers might they be in a crystalline or amorphous state requires that bonds between the polymers can be broken (loss in inner energy). For cellulose this means that the complex patterns of hydrogen bonds made by the three hydroxyl groups per AGU must be broken. To break these, solvents with high hydrogen bonding capacity are necessary. A typical hydrogen bond has an energy of around 20 kJ/mol [10]. As pointed out by Olsson [11] for dissolution of cellulose, it is not sufficient to think about breakage of the hydrogen bond network but to take into account that cellulose has an amphiphilic nature [12, 13]. The CH-groups being essentially perpendicular to the plane of the polymer chain provide a hydrophobic interaction between the polymer chains, whereas the OH-groups provide the hydrogen bonds and are hydrophilic. Both must be altered by a suitable solvent for cellulose. In recent years, many papers appeared on solvents for cellulose discussing especially the effect of the amphiphilic nature [11, 14–17]. In principle there are two ways to dissolve cellulose: derivatization and dissolution without any chemical modification. Derivatization of cellulose is an old technique and still widely used in industry, as, for instance, cellulose acetates, cellulose esters, and ethers. The most common source of cellulose is cotton linters. The fibers are mixed with glacial acetic acid and acetic anhydride with sulfuric acid as a catalyst. This results in cellulose triacetate. In a subsequent step, water is added to stop the reaction and to partially hydrolyze the triacetate. Cellulose acetate is a crystal clear, tough, and flexible plastic and is the most stable cellulose derivative. Cellulose ethers are produced from wood pulp or cotton linters. The cellulose is treated with a solution of sodium hydroxide (mercerization, see below). In a subsequent step, the alkali cellulose is treated with an alkyl halide or an epoxide. The first method is frequently used to prepare

625

ethyl cellulose, whereas the second method is used to prepare hydroxyethyl and hydroxypropyl cellulose. Alternatively, the alkali cellulose can be treated with alkyl sulfate. For example, methylsulfate treatment is a common process for the manufacture of methyl cellulose, being the most important commercial cellulose ether. It is also the simplest derivative where methoxy groups have replaced the hydroxyl groups. A classical way discovered in the nineteenth century is the mercerization. Cellulose is soaked in strong alkali to the extent that the crystal structure changes from cellulose I to cellulose II. It is used to activate the hydroxyl groups of cellulose for further modification or dissolution. For cellulose to dissolve in alkaline aqueous media, it needs to be cooled well below room temperature. Hirosi Soube et al. [18] completed the phase diagram for the ternary system cellulose/ NaOH/H2O. They found that sodium hydroxide-water mixtures at low temperature and in a suitable concentration range can dissolve low molecular weight cellulose (
7  C to 1  C for cellulose contents up to 7 wt% and DP < 200). A utilization of the alkali cellulose is made in the well-known viscose process, in which a NaOH/water swollen cellulose liquid is the starting solution to spin fibers from the highly viscous solution. The alkali-cellulose solution is mixed with carbon disulfide to form cellulose xanthate. The resulting viscose is extruded into an acid bath through a spinneret to make rayon. The acid converts the viscose back into cellulose [5, 6]. Other solvents were developed in the last decades like metal-inorganic complexes, Cuoxam, Nioxam, salt melt hydrates, and ionic liquids to name a few (for more details, see [5, 11]). For aerogels, three dissolving agents are widely used: • NaOH/water with additions of, e.g., urea, thiourea, polyethylene glycol (PEG), ZnO or LiOH with similar additives as in the sodium hydroxide case • Ionic liquids, most frequently based on N-methylmorpholine-N-oxide (NMMO) with suitable stabilizers • Salt hydrate melts, like ZnCl23H2O or Ca(SCN)24H2O and others In a recent article, Budtova and Navard [12] reviewed the dissolution of cellulose in sodium hydroxide water solutions and develop an understanding of the chemical processes of swelling, dissolution, and the effects of various additives enhancing the dissolution power of NaOH-water solutions. The basic idea of NaOH-water solutions being able to dissolve cellulose was that the sodium cation in its solvation shell interacts with the hydroxyl groups of the polymer and thereby breaks the hydrogen bonds. Numerous studies have revealed a clear picture of this interaction (see [12]). These ideas were questioned recently by Medronho, Lindman, and co-workers [13, 14] bringing back into the discussion the amphiphilic

24

626

nature of cellulose. Owing to the strongly hydrophilic character of cellulose, it should be easily soluble in water-based solvents able to form hydrogen bonds with cellulose. The fact that in some cases the additives that are used (like urea) are known to disrupt hydrophobic interactions pushed them to dispute the mantra, that the dissolution always proceeds through the breaking up of hydrogen bonding. The amphiphilic character of cellulose is evident, aliphatic carbons being present on the edges of the pyranose rings together with highly polar groups on the side of the chain. Dissolution in NaOH-water at low NaOH concentrations means that a lot of water is present around the cellulose chains. The mutual polarization between water and hydroxyl groups of cellulose is important and may induce orientation correlations of species close to the cellulose chain, influencing how cellulose interacts with co-solutes such as NaOH. The polarizability effects can partly explain the influence of the temperature of dissolution reported in all studies because of the possible varying structure of the hydrated ions in contact with the cellulose chains. It is quite often observed that dissolved cellulose tends to gel after some time. Additives like urea, thiourea, PEG, and ZnO are used to impede the aggregation of polymers in solution. Indeed, it was found that small amounts of these additives can prohibit gelation for extended times. One idea behind the effect of additives is that urea interacts with the hydrogen bonds perpendicular to the AGU ring where it would diminish the hydrophobic association of cellulose polymer strands [13, 14]. Another idea discussing effects of ZnO in strong alkali solution, in which the amphoteric ZnO is transformed into sodium zincate ions [19], was recently speculated by Meera et al. [20], namely, that the diffusion of cellulose polymers with complexed zincate ions is reduced, thus leading to a reduced Brownian aggregation behavior. Ionic liquids are under investigation for cellulose processing since a few decades, and these are already used industrially to produce cellulose fibers and yarns (Lyocell or Tencel fibers). Ionic liquids consist of organic cations and organic or inorganic anions. They typically have a low melting point, a high viscosity, and a good thermal stability. The most prominent ionic liquid used in aerogel production is N-methyl-morpholine-N-oxide (NMMO), although many others have been studied in the last decade. As stated by Olsson [11], NMMO is completely soluble in water, and as a pure substance, it is extremely hygroscopic. The high polarity of the N-O bond results in a pronounced ability to form hydrogen bonds. The NMMO oxygen is able to form two hydrogen bonds with nearby hydroxyl groups such as in water or cellulose. The procedure for dissolving cellulose in NMMO usually includes a first step where a suspension of cellulose in NMMO and a large excess of water is well mixed. The excess water provides low viscosity and thereby superior mixing. Surplus water is then removed by heat

L. Ratke et al.

between 100 and 120  C and reduced pressure until the point of complete cellulose dissolution is reached. Up to approximately 14% cellulose can be dissolved in a mixture of 10% water and 76% NMMO [21, 22]. A problematic issue with NMMO is that it is an oxidant and is sensitive to all forms of catalytic impurities in the cellulose pulp. There are known side reactions using NMMO, and therefore stabilizing agents are used like propyl gallate. Salt hydrate melts are typically a mixture of an inorganic salt with an amount of water close to the coordination number of the salt cation. Examples are Ca(SCN)2 4H2O and ZnCl2 3H2O. The idea behind these salt hydrates being able to dissolve cellulose is mainly that they are able to break the intermolecular hydrogen bonds [23]. The dissolving capabilities of salt hydrate melts were first described and successfully realized by Phillip et al. [24] using the donator-acceptor concept, which describes the interactions between a polar solvent and the hydrogen bonds. Such a salt hydrate melt system is a solvent for the cellulose system, because of its polarity and acidity and its low melting point, being far below the crystalline amorphous transition point of cellulose/water mixtures being at 320  C [25]. They act in a certain sense like an ionic liquid, being however purely inorganic [16, 26]. The main idea put forward by Sen and co-workers for the case of zinc chloride-water is that the Zn cation is surrounded by a first hydration shell, and the hydroxyl groups of the cellulose molecule complete this hydration shell once they are broken. Attached to such an octahedron of hydrated water is a tetrahedron of Zn cation on whose vertices chloride ions are located. This special feature makes ZnCl23 H2O an excellent dissolving agent for cellulose. A sketch of this idea is shown in Fig. 24.4.

OH O HO

OH O

HO O

OH OH2

OH

n

Cl

Cl

H 2O Zn H 2O

O

Zn

OH2 OH2

Cl

Cl

Fig. 24.4 A sketch of the action of zinc chloride-water on the cellulose. The zinc cation has a hydration shell which is completed by the hydroxyl groups of the cellulose molecule. Attached to it is a tetrahedron of zinc chloride. (Redrawn after [26])

Cellulose Aerogels: Monoliths, Beads, and Fibers

It is reported in the literature that a deviation from the trihydrate to a four hydrate also allows to dissolve cellulose. This is in agreement with the phase diagram study of the ZnCl2 – H2O, showing that the trihydrate is a congruent melting phase and concentrations off this composition can lead to liquids with the trihydrate completely dissolved if the temperature is adjusted properly [27].

24.3

627

104

24

1000 tgel (min)

24

100 10

Gelation of Cellulose Solutions

Cellulose solutions are intrinsically unstable, meaning if they are left at rest after dissolution, they tend to either form precipitates, which then sediment, or simply gel, which also depends on the amount of cellulose brought into the solution. Gavillon and Budtova [28] studied the rheological behavior of cellulose dissolved in sodium hydroxide water solutions using an Avicel cellulose of a DP of 180 and a Borregaard cellulose of a DP of 500 and a 7.6 wt% NaOH in water. Their result with a 5 wt% cellulose indicates that gelation is temperature dependent in an exponential way. Gavillon and Budtova used the relation tgel ¼ 5.7  104 exp.(
0.345 T ), with the temperature measured in degrees Celsius to fit the data. In a more recent study Liu, Budtova and Navard [29] looked at the influence of ZnO on the gelation of cellulose dissolved in sodium hydroxide. They used Avicel microcrystalline cellulose with a DP of 170 and studied the gelation of 4 and 6 wt.% cellulose in an 8% NaOH-water solution. They also found an exponential dependence of the gelation time on temperature. Although the fits in both papers agree well with the data, one might question if it makes sense in terms of underlying physics. Plotting their data in a typical way, namely, plotting the gel time as a function of inverse temperature in Kelvin, yields even a much better fit. This is shown in Fig. 24.5. Extracting from these fits, the activation energy yields an astonishing large value, namely, a value around 250– 300 kJ mol
1, ten times larger than the activation energy of the viscosity of a 7.6 wt% NaOH solution as determined by Gavillon [30]. Nevertheless, assuming that diffusion of the polymer molecules in the solution is essential for their aggregation and gelation, one could argue that the mean square displacement is proportional to the diffusion coefficient and time. For gelation to occur, the molecules, which can be thought of as thin rods, would have to make a sufficient displacement to touch and interlock. Thus one could argue that the gelation time must be inversely proportional to the diffusion coefficient or proportional to the viscosity using the Stokes-Einstein relation between diffusion coefficient and viscosity. This would result in an Arrhenius fit as made in Fig. 24.5. Therefore the huge value of the activation energy or its strong difference to the activation energy of the viscosity is astonishing but might be a hint that gelation is more

1 0,1 0,003

0,0032

0,0034 0,0036 1/T (1/K)

0,0038

tgel (min) 5% cell 7.6% NaOH tgel (min) 4% cell 8%NaOH tgel (min) 6% cell 8%NaOH

Fig. 24.5 Gelation time as a function of temperature for 4, 5, 6% Avicel/NaOH/water solutions. Solid lines correspond to exponential fits. (Data taken from [28, 29])

complex than just diffusion and aggregation of rod-like molecules. In this context, it should be noted that the addition of urea, thiourea, and ZnO delays gelation considerably [12, 28, 29]. Taking the published data on the effect of ZnO to delay gelation, the activation energy would be reduced from the above value down to around 200 kJ mol
1. This still is a large value and also points to a missing model describing gelation of cellulose solutions. In a study of thermally induced gelation and coagulation-induced gelation of cellulose dissolved in NaOH/urea, Isobe and co-workers [31] used thin capillaries filled with the solution and either heated it to 105  C or overlap the solution with a Na2SO4 solution which induced gelation. The developing structure in the solution was followed by synchrotron X-ray radiation. They could show that both the temperature and the coagulation bath changed the arrangement of the polymers, leading in their interpretation to an alignment of the polymers parallel to their hydrophobic surface, and these aligned strands of polymers interact to fibrils leading to gelation. Generally, however, cellulose dissolved in alkali-based solution gelation is induced by a pH inversion, meaning by a neutralization reaction. The wet gel is submerged into an acidic regeneration bath containing different concentration of acids; mostly reported are H2SO4, HNO3, and HCl. For example, using NaOH-water-urea as solvent, coagulated cellulose beads were prepared in [32] by dripping a solution in an aqueous solution of 2 N nitric acid. By the same way, in the presence of ZnO nanoparticles, cellulose was dissolved in

628

L. Ratke et al.

an aqueous solution of NaOH-urea-water mixture under ice cold condition, and then particles were regenerated in 2 N hydrochloric acid solution [20]. Gelation of cellulose dissolved in NMMO or salt hydrate melts is different. As reported by Innerlohinger [33] and Liebner’s group [34–38], after dissolution of cellulose in NMMO at temperatures above 100  C, gelation occurs upon cooling. The same was observed for cellulose dissolved in calcium thiocyanate by Hoepfner et al. [39, 40] and for ZnCl2-tetrahydrate by Schestakow et al. [41]. In both cases, the reversibility of dissolution on increasing the solution temperature and gelation on cooling suggests that a miscibility gap exists and gelation could be a result of phase separation like nucleation, growth and aggregation, or even spinodal decomposition as argued by Pircher et al. [15]. Today this is, however, a speculation, since there are only a few studies on phase diagrams for NaOH-water, NMMO-water, or salt hydrate melts cellulose. Besides phase diagram studies, thermodynamic quantities would have to be collected (free enthalpy of mixing and specific heat, for instance). These would allow to calculate binodal and spinodal lines and compare them with polymer models like that of Flory and Huggins [42, 43] and discuss possible types of phase separation, especially with respect to gelation and duration as a function of temperature.

24.4

Drying Methods and Shrinkage

All wet gels prepared by any type of route have to be dried to obtain an aerogel. Drying is the last and the most critical step in the aerogel production. The gel is a highly porous structure in which the pores are filled with a liquid. The fraction of the volume liquid to the volume of the gel is generally more than 0.95. It is desirable to preserve the pore volume of the matrix, minimize shrinkage of the solid, and prevent collapse of the pores during drying to result in aerogels with desirable properties for a wide variety of applications. Essentially three drying processes are commonly used for biopolymer aerogel production: ambient drying, freeze drying, and supercritical drying. In all variants of ambient drying processes, the pore liquid simply evaporates. This always leads to a collapse of the microstructure inside the gel body due to the surface tension and surface tension gradients created by the solid-liquid-gas interface tension and capillary stress gradients stemming from the dispersion of pore sizes. It was recently reported for monolithic gel bodies that the drying processes effectively influenced the material properties and the porous structures [44, 45]. For polysaccharide wet gels, the ambient drying method provides more aggregated microstructures due to massive shrinkage of around 70–80% leading to densely packed solids with almost no porous structure [45]. Chemically modifying the

hydrophilic -OH functional groups to hydrophobic ones can assist the ambient drying. Recently low density, open porous, and hydrophobic cellulose materials were prepared via ambient drying by chemically modifying the –OH functional groups with triethyl chloride [46]. In this method, depending upon the degree of substitution, the chemical modification may lead to the development of an unusual microstructure due to the different manner of self-assembly of cellulose molecules and lack of hydrogen bonding. Ambient drying is generally not a suitable method for the preparation of biopolymer aerogels due to the almost total collapse of the pore structure, and therefore it rarely is discussed in the literature. In the case of freeze drying, the liquid in the gel body is frozen and sublimed under vacuum. The volume shrinkage can be limited to 40–50%. The critical step here is the freezing of the gel liquid. Since it is a solution, on freezing typically dendritic structures appear, having even at high cooling rates sizes in the micrometer to 10 micrometer range and thus destroying the nanostructure of the wet gel. The critical step in freeze drying is the cooling of the solution, and this is a solidification problem. In many papers on aerogel production by freeze drying or ice-templating, how it is sometimes called, this aspect is mostly overlooked. Solidification of a solution means that upon cooling ice crystals are nucleated below the liquidus temperature of the solution having a composition of the solidus line in the phase diagram. Upon further cooling, these ice crystals grow, and the composition of the solid phase changes, whereas the liquid solution follows in its composition the liquidus line. The morphology of the ice crystals is dictated by the cooling rate; the phase diagram, meaning essentially the partition coefficient; the interface tension between ice crystal and liquid solution; diffusion coefficient; the aspect ratio of the sample (surface area divided by the sample volume, Chvorinov’s rule); and some more parameters. Details can be read in any textbook on solidification [47–49]. Important in this context here are two issues: first the cooling rate does not only depend on the temperature difference between the as-prepared cellulose solution and the freezer in which it is stored but also on the volume of the liquid to be cooled. Secondly, the polymeric strands inside the solution are not built into the growing ice crystals but are shifted by the advancing solidification front [50]. Therefore after complete solidification, they mimic the shape of the ice crystal network, and any memory of the wet gel structure is lost. Typically freeze drying may therefore lead to an open porous material with a pore size in the range of micrometers, which might be termed open porous foams, since the amount of mesoporosity is most often rather small. In supercritical drying, the pore liquid from a polysaccharide gel body can be extracted under supercritical conditions

24

Cellulose Aerogels: Monoliths, Beads, and Fibers

using a supercritical fluid. A fluid reaches its supercritical state when it is compressed and heated above its critical temperature and pressure. Supercritical fluids (SCFs) have liquid-like densities and gas-like viscosities. Supercritical carbon dioxide (SC-CO2) is very attractive among other supercritical fluids and most employed even in many industries as it has relatively easy accessible critical conditions and is nontoxic, environmentally friendly, widely available, and cheap. In this drying process, the gel network can be preserved without cracks as there is no existence of capillary stresses. With the great advantage of this drying process, many industries produce commercial aerogel products in different forms mostly sheets or panels. In any case, all types of drying are associated with different types of shrinkage. Firstly there always is a shrinkage of the wet gel, in as much as the gel network forms and stiffens, secondly there is a shrinkage related to the exchange of the original pore liquid with a liquid that either can easily be evaporated or is miscible with supercritical carbon dioxide, and finally drying itself leads to a shrinkage. The amount of shrinkage differs from case to case and has to be taken into account in order to obtain a material with the desired properties. A recent review on methods to design porosity and pore size distribution taking into account the effects of shrinkage and drying was published by Smirnova and Gurikov [51].

24.5

Monoliths: Preparation and Properties

Several methods are described in the literature to prepare cellulose aerogel monoliths, and still many new methods are developed, partly depending on the raw material used, since hemicellulose needs a different process than raw cellulose or lignocellulosic polymer mixtures. First attempts to preserve the swollen structure of cellulose pulp go back to the work of Weatherwax and co-workers [52, 53] and Alince [54]. They used different pulps from paper, cotton, and rayon. The materials were swollen by distilled water, ethylenediamine, and aqueous sodium hydroxide (a standard solvent agent to produce a pre-material for the spinning of filaments in the viscose process [5, 6]). The swollen and wet materials were dried mainly by solvent exchange (water against alcohols or acetone) and then slowly evaporated. In all cases reported by them, they were able to produce porous materials but could not avoid considerable shrinkage. The materials were only characterized by adsorption isotherms, and no other structural or physical property measurements were made. Weatherwax and Caulfield [52, 53] used, after solvent exchange with alcohol, carbon dioxide supercritical drying to obtain cellulose aerogels with a specific surface area of around 200 m2 g
1.

629

24.5.1 Aerogels from Cellulose Derivatives The first cellulose aerogels which became well known and popular were prepared by Tan and co-workers [55]. They used cellulose acetate as a starting material and de-esterified it. The cellulose ester was crosslinked in an acetone solution using toluene-2,4-di-isocyanate. Tan and co-workers observed that they could form gels if the cellulose concentration was larger than five and less than 30 wt%. The supercritically dried aerogels had specific surface areas of less than 400 m2/g and densities in the range of 100–350 kg m
3. The amount of cellulose to volume of acetone had an effect on the density and the shrinkage as well as the ratio of cellulose to crosslinker. The smaller the cellulose ester concentration and the larger the crosslinker amount, the larger the shrinkage. This means that the cellulose backbone is rather rigid and does compensate shrinkage stress gradients. Their work became popular and was mentioned even in newspapers, since they measured the impact strength of the cellulose aerogel sheets (5 mm thick) and could show that although their material had a high porosity, its strength exceeds that of resorcinol-formaldehyde (RF) aerogels. The technique to measure the strength might be questionable from an engineering point of view, since an impact test is not really appropriate for aerogels. Bending or tensile tests would be a better suited benchmark [56]. The comparison with RF-aerogels is not indicative for a high strength material. It just shows that the RF aerogels produced by Tan et al. are as brittle as many other aerogels. Pinnow et al. [57] prepared cellulose aerogels from cellulose carbamate. They used two methods. One is thermal decomposition of cellulose carbamate at above 100  C in sodium hydroxide solution, and the other method is chemical precipitation of cellulose carbamate like pH inversion in acidic medium in the presence of sodium sulfate. Cellulose carbamate was then converted to cellulose at room temperature in the presence of sodium hydroxide. The aerogels were characterized with scanning electron microscopy, small-angle X-ray scattering (SAXS), mercury intrusion, and nitrogen adsorption. They were able to produce aerogels with a density as low as 60kgm
3 from cellulose carbamate having a broad pore size distribution ranging from around 1 nm to 1 mm. The effects of different preparation routines and processing parameters including drying and pyrolysis were studied by the methods mentioned. They showed that the preparation of the gel state from cellulose carbamate is an interesting alternative to the thiocyanate route (see Sect. 24.5.3) to produce highly porous cellulose materials with a low density and a specific surface area of about 430 m2 g
1. A similar study was performed recently by Gan et al. [58]. They prepared the cellulose carbamate from Kenaf (Hibiscus cannabinus) by themselves and thus were able to vary the amount of the carbamate at the cellulose polymer by

24

630

treating the cellulose pulp with different amounts of urea. To prepare hydrogels from the cellulose carbamate, the solutions were mixed with epichlorohydrin as a crosslinker. The so-called aerogels were prepared by freeze drying of the hydrogels. The microstructure of the aerogels as revealed by SEM shows that they were able to produce a foam-like material with all cellulose carbamate in the cell walls and large pores in the range of 100 nm. In the understanding used here, their material is a macroporous foam and not an aerogel, which should mainly consist of mesopores.

L. Ratke et al.

a

b

c

24.5.2 Aerogels from Alkali Hydroxide Routes Gavillon and Budtova reported on the preparation and the morphology of an aerogel-like pure cellulose material [28]. The material they called aerocellulose was obtained from cellulose dissolved in nonpolluting solvents, such as aqueous NaOH or N-methyl-morpholine-N-oxide (NMMO) monohydrate, and dried supercritically with CO2. The studies were focused on the preparation of cellulose aerogels from cellulose/NaOH/water solutions and on the influence of the preparation conditions on their morphology. As a starting material, they used several microcrystalline celluloses with degree of polymerization ranging from 180 to 950 (Solucell cellulose from Lenzing R&D labs, Austria). The cellulose was dissolved and swollen in mixtures of water and sodium hydroxide at low temperatures. The cellulose concentrations were varied between 3 and 8 wt%. The solutions were cast into cylindrical molds of 8–10 mm diameter and 30–40 mm length. After gelation, the samples were regenerated in water. During regeneration, the samples shrunk by about 10% in volume. The gelation time varied exponentially with temperature. Although at 10  C the gelation takes around 3 days, at 40  C it takes only 6 min (see Sect. 24.3). The regenerated and swollen cellulose gels were subjected to a solvent exchange using acetone, which was then replaced with CO2 and supercritically dried. The resultant aerogels show porosities above 90%, specific surface areas above 200m2 g
1, and densities in the range of 120–140kgm
3. A special surfactant (Simulsol, alkyl polyglycoside, APG) was also used in various concentrations, and it could be shown that it changes the morphology from a more open porous network of thin fibrils to a more clamped compact one, often observed in freeze-dried samples [39, 59]. The surfactant reduced the density slightly but increased the average pore size by an order of magnitude. It is generally observed that aerogels from cellulose gels regenerated in water at ambient temperature show a fibrillar structure, while those regenerated at 70  C show a cloudy structure irrespectively of the cellulose solvent. This can be attributed to a higher diffusivity of all components in the gelling system being accelerated at higher temperatures.

Fig. 24.6 Photographs of cellulose gels according to reference [60]: (a), hydrogel in water; (b), alcogel in EtOH; and (c), aerogel obtained by supercritical carbon dioxide drying. (With permission from Wiley-VCH Verlag GmbH & Co)

Probably because of this, cellulose chains have less time to pack, and rapid precipitation causes a more disorganized structure and a cloudy appearance. Gavillon and Budtova also tested if the acidity of the regenerating bath (adjusted as in the industrial viscose process with sulfuric acid) changes the aerogels. The higher the regenerating bath acidity, the lower are both the average pore diameter and the total porosity and thus the density. Cai et al. [60] were able to prepare transparent cellulose aerogels using an aqueous alkali hydroxide/urea solution as the dissolving and gelling agent. Figure 24.6 shows one of their impressive results, a transparent cellulose aerogel. They used cellulose of different sources and NaOH-ureawater or LiOH-urea-water solutions to dissolve the cellulose at low temperature by vigorously stirring and ultracentrifugation to remove air bubbles. The amount of cellulose was varied between 0.5 and 7 wt%. The clear solution was cast onto a glass plate yielding sheets of 0.5 mm thickness. The material was regenerated in various solutions such as alcohols and dilute sulfuric acid. Regenerated cellulose was washed with water, a solvent exchange performed with ethanol, and the samples were finally dried supercritically with carbon dioxide. Table 24.1 shows an overview of their materials and properties and process parameters. This table already shows that the optical transmission can be better than 80% which comes into the range of inorganic aerogels. Nitrogen adsorption has shown a distinct hysteresis in the capillary condensation regime. As pointed out by Reichenauer and Scherer [61] for silica aerogels, this is a result of the deformation on loading of the aerogel leading to an elastic/plastic behavior whose origins are capillary stresses in the mesopores stemming during the desorption cycle from the evaporating liquid nitrogen. The results of Table 24.1 also

24

Cellulose Aerogels: Monoliths, Beads, and Fibers

631

Table 24.1 Extract of the results achieved by Cai and co-workers [60] Solvent bath Aqueous NaOH Urea Aqueous LiOH Urea

Cellulose concentration [wt%] 6 4 6 6 6 4 4

Regeneration 5 wt% H2SO4 EtOH H2O EtOH H2SO4 EtOH BuOH

260

420 3

Density (kg/m )

400

220

380

200

2

SBET (m /g)

180

360 340

160

320

140

300

120

0

20

40 60 80 Amount EtOH (%)

SBET (m2/g)

Density (kg/m3)

240

280 100

Fig. 24.7 Density and specific surface area of cellulose aerogels prepared with aq.LiOH/urea and regenerated at 20  C. The cellulose concentration was 6 wt% after [60]

show that in contrast to NaOH, the use of aqueous LiOH as a cellulose solvent allowed the formation of more transparent cellulose hydrogels and aerogels on regeneration by organic solvents (alcohols and acetone). Gels prepared with LiOH and alcohols as regenerators also showed a more homogeneous fine fibrillar and highly porous network structure. The variability of the properties with the composition of the regeneration bath is demonstrated by a plot extracted from their tabulated data and shown in Fig. 24.7. The strong variation of the density and the specific surface area with ethanol concentration in the bath clearly demonstrate how important it is to control the regeneration conditions precisely.

24.5.3 Aerogels from Salt Melt Hydrate Routes Jin and co-workers [59] developed another technique to produce high-quality cellulose aerogels. Their technique avoids the utilization of toxic isocyanates and allows in contrast to the method of Tan [55] to use lower amounts of cellulose. Jin et al. [59] used the above-described salt hydrate melt technique for dissolution. The hot solution of calcium thiocyanate and water Ca(SCN)2 and H2O with 1:4 mol/mol dissolves cellulose probably by complexing the hydroxyl

Optical transmission [%] 49.1 19.6 43.9 84.1 60 78.6 53

Density kg/m3 260 140 160 190 260 120 400

SBET [m2/g] 364 260 406 410 381 406 304

groups of the cellulose molecule as described above for the zinc chloride water system. This particular salt hydrate cellulose solution had a temperature around 110–120  C and undergoes a reversible sol–gel transition at approximately 80  C. A schematic of the process developed by Jin and co-workers [59] is shown in Fig. 24.8. The team prepared the gels by quickly spreading the hot solution on a glass plate to form a layer of 1 mm thickness. After the solution solidified, the gel plate was immersed in a methanol bath, which extracted the salt and regenerated cellulose as a gel. The gel was washed with deionized water until the electrical conductivity of the effluent became negligible. The films prepared by Jin and co-workers were dried by freeze drying using slightly different procedures. The samples either were freeze-dried regularly after immersion into liquid nitrogen or after a series of solvent exchanges with various alcohols prior to liquid nitrogen cooling or were contacted with a copper plate kept at liquid nitrogen temperature and then transferred to the freeze drying unit. Jin and co-workers did not use supercritical drying, which is reflected somewhat in their results. Essential to their process was the use of low cellulose concentration (0.53 wt%), and therefore they obtained extremely low density aerogels. The density increased with cellulose content. Their data suggest a parabolic increase with concentration, which in view of newer results suggest that the low cellulose content materials exhibited larger shrinkage (see below). The specific surface area increases too with increasing cellulose content from around 105–200 m2 g
1 from 0.5 wt% cellulose to 3 wt%. Freeze drying always leads to materials with higher density and lower surface area. The variation of tensile strength measured on thin but large samples with cellulose content is shown in Fig. 24.9. In contrast to the statement of Tan [55], the strength of these aerogels is rather low. A value of 120 kPa is easily achieved with RF aerogels if dried subcritically. The low strength might also be a result of the drying process and the modification of the nanostructure inside the wet gel. Although a high cooling rate was used, which typically leads to very small dendrites [50], the microstructure will have changed and at least modified the fibrillar network as explained above. As an example of the extensive microstructural analysis, Fig. 24.10 shows a comparison of

24

632

L. Ratke et al.

Fig. 24.8 Schematic of the process developed by Jin and co-workers [59] to produce cellulose aerogels. Instead of rhodanide, one can use also other salts, like zinc chloride tetrahydrate

Mixing

to 110°C

H2O

Ca(SCN)2 Cellulose or ZnCl2

Tensile strength (kPa)

120 100 80 60 40 20 0

0

Heating up

1 2 3 Cellulose concentration (wt%)

Suspension

Clear solution with dissolved colloidal cellulose

varying between 0.5 and 3 wt%. The linear relation shows that only the cellulose concentration in the salt hydrate melt determines the density. The temperature of the melt bath, the annealing time before cooling to the gel point and not the aging time in ethanol played a detectable role. Note that the density is only eight times higher than the density of air and is in the range of styrofoams [62]. A linear variation of the aerogel density of cellulose concentration should be observed on simple theoretical grounds. If ρA denotes the density of the aerogel and ωc is the salt hydrate melt weight fraction having a density ρL, then for small concentration, ωc  1, the aerogel density should obey the relation: ρ A ffi ρ L ωc

Fig. 24.9 Strength of cellulose aerogels as prepared by Jin and co-workers using a salt hydrate melt to dissolve cellulose (calcium thiocyanate/water) and regeneration of the cellulose in alcohols. (Redrawn from [59] different units used here)

the microstructures as revealed by SEM for a material with 2 wt% cellulose. One is freeze-dried (left panel), and the other is solvent exchanged and freeze-dried (right panel). Hoepfner et al. [39] used the process of Jin [59] to produce monolithic cellulose aerogels. They also used Ca(SCN)2hydrate as described above with a water concentration near the coordination number of the cation and dissolved microcrystalline cellulose of a DP of 150 in that solvent by increasing the temperature of the salt hydrate melt to 110–120  C until the solution became clear. The hot cellulose solution was poured into polyacrylic multiwall sheets with a cross section of 15  15 mm yielding monoliths with a length of around 100 mm. The cellulose is then regenerated by immersion in ethanol. Freeze drying was used after rapid solidification of the wet gel in liquid nitrogen and supercritical drying after solvent exchange by CO2. The monolithic aerogels were white, soft, and easily deformable by indentation. Figure 24.11 shows the average density of aerogels aged in ethanol. The density varies between 10 and 60 kgm
3 with the cellulose concentrations

ð24:1Þ

which has also been observed by Innerlohinger [33], whose data points, however, did not cross the coordinate origin at zero concentration. Hoepfner et al. also observe a rather large shrinkage during supercritical drying, being not as large as those of Innerlohinger but around 15% and varying linearly with cellulose concentration. A large shrinkage at the lower concentration shows that the gel network is not very stable and tends to contract significantly during the subsequent gel processing. With higher cellulose content, the shrinkage is smaller showing that the network is more stable. Using nitrogen adsorption, Hoepfner and co-workers measured specific surface areas of 210 m2 g
1, which were almost independent of the cellulose concentration. A simple calculation shows that the independence is theoretically expected, if the drying procedure does not harm the gels. To understand this, they assumed that all cellulose nanofibrils can be described as cylinders with average radius hRi and length hLi. Then the specific surface area per unit volume SV is given as [63]: SV ¼

2ϕA hRiρA

ð24:2Þ

Assuming that hRi/hLi  1. Assuming ρA ≈ ϕAρc, they obtain for the specific surface area per unit mass the relation:

24

Cellulose Aerogels: Monoliths, Beads, and Fibers

633

Fig. 24.10 Microstructure of a cellulose aerogel as prepared by Jin and co-workers [59] (with permission from Elsevier). The samples shown have a cellulose concentration of 2 wt%. The figure in the left panel shows a regularly freeze-dried sample and in the right a sample rapidly freeze-dried as described above with gel casting onto a copper plate kept at liquid nitrogen temperature

24

60

Density (kg/m3)

50 40 30 20 10 0

0

0.5 1 1.5 2 2.5 3 Cellulose concentration (wt%)

3.5

Fig. 24.11 Density of cellulose aerogels prepared by Hoepfner et al. [39] using the route of Jin [59]

Sm ¼

SV 2 ¼ ρA hRiρc

ð24:3Þ

with ρc the skeletal density of cellulose. If the fibril radius in the aerogel network is essentially independent of the cellulose concentration, the relation states that the specific surface area is a constant independent of the cellulose concentration, in agreement with their observations. One can put this in other words: if the individual fibrils are equally dispersed (independent of the concentration) and do not pack or aggregate together, then the surface area per mass of total fibrils is essentially independent of concentration. They also developed a simple model to describe the effect of networking on specific area. All points of contact between fibrils reduce the specific area in a first approach linearly dependent on the number density of contacts points. This simple model also shows routines, how to increase the specific surface area: one should avoid aggregation or packaging of the cellulose fibrils

and reduce the number density of contacts of the fibrils and, of course, the fibril diameter hRi. The challenge here is how to achieve this in the gel preparation and the subsequent processing steps of washing and drying. Figure 24.12 shows the SEM picture of SCD aerogels prepared with 2 wt % cellulose. The picture clearly shows that the nanofelt indeed consists of cellulose fibrils having thicknesses in the range of 20–50 nm. In a recent study, Schestakow and co-workers used zinc chloride tetrahydrate (ZnCl24H2O) as a solvent system and studied the influence of several regeneration bath fluids on the microstructure and properties. Cellulose aerogels with concentrations of 1–5 wt.% cellulose were produced. After gelation the salt hydrate was washed out using water, ethanol, isopropanol, or acetone. Cellulose aerogels regenerated in acetone show a specific surface area of around 340 m2g
1 which is 60% higher than those regenerated in water. The onset of irreversible plastic deformation under compressive load is around 0.8 MPa for acetone-regenerated aerogels and thus a factor of two larger compared to ethanol-regenerated ones. This shows that the zinc chloride dissolution leads to rather strong aerogels compared to other solvent systems used. Young’s modulus depends almost linearly on the cellulose concentration which is observed for all regenerative fluids with the exception of water. A comparison of the microstructures observed by SEM after carefully breaking the samples in liquid nitrogen to avoid deformation of the fibrillar network during cracking is shown in Fig. 24.13. In all cases, the cellulose has arranged itself into a fibrillar network during gelation. The pores are, however, typically in a submicron size range. The mesoporosity is not directly observable in the SEM pictures. These aerogels and also some prepared with Ca (SCN)24H2O were used to develop a new micro-mechanical model describing the mechanical response of fibrillar structured aerogels by Rege and co-workers in two papers concerned with compression and tension testing

634

L. Ratke et al.

[64, 65]. Their generalized micro-mechanical model is constructed as follows: the fibrillar, cellular microstructures of cellulose aerogels are modeled by a network of squareshaped cells. These cells can vary their size determined by the

200 nm

EHT = 2.00 kV WD = 3.7 mm

Signal A = InLens Photo No. = 120 I Probe = 15 pA

Date :11 Jan 2012 Time :9:34:09 Mag = 25.00 K X

Fig. 24.12 SEM picture of a supercritically dried aerogel with 2 wt% cellulose as prepared according to the route of Hoepfner et al. [39]

Water

Isopropanol

measured pore size distribution, and the cell walls are beams, whose thickness varies as the fibril diameter, which is determined from the specific surface area per unit volume. The cells can vary their orientation with respect to the load direction statistically, and a special method was developed to describe this properly. Each beam in such a square cell undergoes nonlinear beam deflection according to the wellestablished Euler-Bernoulli theory. In contrast to scaling theories developed by Gibson and Ashby [66], this model is able to predict the whole deformation curve of a mesoporous material, and thus a more detailed comparison with experiments is possible. Figure 24.14 shows an example of such a deformation curve in comparison with an experimental one. This physically motivated constitutive model, nonlinear beam bending combined with statistical distribution of cells with respect to orientation and size, describes the compressive response of different cellulose aerogels in good agreement with the experimental data for cellulose aerogels with varying cellulose content. A similar good agreement was found for tension loading of cellulose aerogels, which is more difficult to model, because the cells and their walls undergo both a bending and stretching deformation. A recent review about modeling and simulation of polysaccharide

1 Pm

Ethanol

1 Pm

1 Pm

Acetone

1 Pm

Fig. 24.13 The figure shows the fibrillar network of samples with 5% cellulose regenerated in four different fluids

Cellulose Aerogels: Monoliths, Beads, and Fibers

Nominal stress P (MPa)

2.5 2 1.5 1

250

Model prediction Experimental data 5% cellulose 3% cellulose

0.5 0

635

Density (kg/m3)

24

200 150 100 50 0

1 0.95 0.9 0.85 0.8 0.75 0.7 0.65 0.6 0.55 Stretch O (–)

Fig. 24.14 Model predictions vs. experimental data for ZnCl2-based cellulose aerogels

aerogels summarizes the current opinion on mechanical behavior of cellulose and other polysaccharides [?].

24.5.4 Aerogels from Ionic Liquid Routes Innerlohinger et al. [33, 67] were the first to produce aerogels from cellulose using an ionic liquid, N-methyl-morpholineN-oxide (NMMO), as a solvent. Cellulose was obtained from 13 different pulps by Lenzing AG having different degrees of polymerization ranging from 180 to 4,600. The typical process of Innerlohinger and co-workers consisted of a solution cellulose and NMMO, shaping, regeneration and washing in alcohol and water, solvent exchange, and supercritical drying with carbon dioxide. They used cellulose solutions with 0.5– 13 wt% and produced three different shapes of aerogels: large cylindrical monoliths with 26 mm in diameter, spherical beads, 2–4 mm in diameter, and thin films by casting the hot aerogel solution onto glass plates. From their extensive studies, a few results are reproduced here. Figure 24.15 shows the density of aerogels as a function of initial cellulose content. The density increases in an almost linear fashion with the cellulose content. The lowest density is achieved with 50kgm
3. Innerlohinger and co-workers reported a rather large shrinkage of their wet gels of around 40%. Such values are extremely large compared to supercritical drying of other aerogels and show that their gels are rather soft and plastically deformable. The reported internal surface areas range between 100 and 400 m2 g
1. The SEM figures shown in their work reveal a more flake-like appearance of the cellulose fibrils being strongly interconnected. Therefore, it seems that their technique to dissolve cellulose can be improved especially with respect to the solvent exchange and

24

0

1 2 3 4 5 6 7 8 9 Initial cellulose concentration (wt%)

Fig. 24.15 Density of cellulose aerogels prepared by Innerlohinger et al. [33, 67] using NMMO as a solvent and supercritical drying with carbon dioxide. (Redrawn on the basis of their data using different units)

regeneration procedures, which both have a marked influence on the aerogel structure and properties. Liebner and co-workers [36, 37] report on their approach to also use NMMO as a solvent and N-benzyl-morpholine-Noxide (NBnMO) as a stabilizer against oxidation. From NMMO solutions with cellulose contents between 1 and 12 wt%, dimensionally stable cellulose gels were produced. The cellulose solutions were produced at temperatures varying from 100 to 120  C increasing with the cellulose content. The cellulose contents were 3, 5, 10, and 12 wt% and thus rather large compared to the salt hydrate melt route. They performed the regeneration with ethanol or a mixture of ethanol and dimethylsulfoxide (DMSO) in different concentrations. The dissolution, regeneration, and solvent exchange routes could take up to several days. Their aerogels exhibit densities in a range from 50 to 260 kg m
3 and specific surface areas from 172 to 284 m2 g
1. All their regenerated cellulose gels were dried supercritically. The SEM microstructures presented look, compared to other cellulose aerogel structures, rather dense and not as open porous as aerogels prepared by other authors. Liebner and co-workers [34] improved their methods to prepare ultra-lightweight aerogels using solutions with 3% and 6% cellulose dissolved in NMMO-H2O mixtures and stabilized with 1 wt.% N-benzyl-morpholine-N-oxide (NBnMO). The clear viscous melts were poured into cylindrical molds and cooled to room temperature. Regeneration was performed using ethanol and after several steps of washing with ethanol supercritically dried with carbon dioxide. They observed that aerogels regenerated with water and ethanol mixtures had a higher density of 0.09 gm
3 after drying compared with samples washed only with pure ethanol 0.06gm
3. The average shrinkage was around 10–20%. They claim that shrinkage can be diminished by

636

(1) preventing any mechanical stress during regeneration, (2) avoiding water in the regeneration baths, and (3) using regeneration times long enough to establish an equilibrium between NMMO and the regenerating solvent, even in the smallest pores. Average shrinkage related to supercritical drying of the aerogels was between 5% and 32% depending on the cellulose concentration used. They found that interactions between cellulose and the regeneration bath are relevant for the amount of shrinkage. Replacing ethanol by supercritical CO2 results in the destruction of dipole-dipole interactions between ethanol and the cellulose matrix, whose forces are necessary to maintain the swollen cellulose network structure obtained in the regeneration step. The final removal of CO2 might also influence shrinkage to some extent, but the interaction between CO2 and the cellulose matrix is much weaker because the dipole moment of CO2 is zero. Looking at the microstructures they have obtained, the cellulose network is rather irregular, barely nanostructured. The pores appear to be beyond the macroporous range. It should be noted that the DP of the cellulose pulps used reduced during treatment by around 100, which shows that the cellulose polymer is unstable at the high temperatures and extended times used. The mechanical strength, measured in compression, varies between 20 kPa and 940 kPa depending on the pulp used and the amount of cellulose. The specific surface area varies between 55 and 310 m2 g
1, which are rather large values, but looking at the adsorption isotherms, there seems to be some microporosity and little or no mesoporosity. Aaltonen and Jauhiainen [68] prepared cellulose aerogels from microcrystalline cellulose, spruce wood, and mixtures of cellulose, lignin, and xylan using an ionic liquid as a solvent. The work is interesting because of the use of a mixture of cellulose, lignin, and xylan, as well as direct use of wood. They mixed cellulose from 1 to 4 wt% with respect to the amount of 1-butyl-3-methylimidazolium chloride (C4mimCl which is identical to BmimCl used by Deng and co-workers [69]), lignin in the range of 1.4–3 wt%, and xylan of 2 wt%. The wood spruce was used in two concentrations of 3 and 4 wt%. The dissolution was performed at 130  C and took between 2 and 27 h. The gelation of the lignocellulosic polymers was performed in water/ethanol baths of different compositions. Gels could not be obtained in all cases, but some mixtures of cellulose for, lignin, and xylan gels were obtained. The suspensions were cast in Petri dishes giving sheets of 10 mm thickness. After gelation, pieces of 5  5 mm2 were cut and supercritically dried after solvent exchange with ethanol and carbon dioxide. The bulk densities of the biopolymer aerogels ranged from 25 to 114 kg m
3 and the internal surface areas (BET) from 108 to 539 m2 g
1. The aerogels exhibit the typical 3D fibrillar network structure, which varied with the processing conditions. The materials were generally mechanically soft and did not recover

L. Ratke et al.

their shape after slight compression with fingers. In contrast to the material described by Hoepfner [39], Aaltonen and Jauhiainen noted that their aerogels could easily be disintegrated to a fibrous or powder-like material and rubbed between fingers. This is similar to some inorganic aerogels. They also observed that the aerogels obtained from wood were comparably hard and exhibited much higher mechanical strength, without, however, quantifying this. Liebner and co-workers [34] extended their research on cellulose aerogels described above. They used again a preparation method with hot solutions of cellulose/NMMO/water at 110–120  C and cast the clear solutions into molds. After gelation, the samples were regenerated in an ethanol bath, removing the NMMO/water solution. Supercritical drying was performed with carbon dioxide. The densities were in the range of 46–69 kgm
3 for 3 wt% of cellulose, the pore size around 10 nm, and the specific surface area around 190– 310 m2 g
1. The low density shows that their process preserves the wet gel structure excellently. They applied a compression test and found that their aerogels can withstand a yield strain of around 5% before fracture, which is rather large compared with inorganic aerogels; on the other hand, a Young’s modulus of 5–10 MPa is small. Doubling the amount of cellulose changes the elastic modulus by a factor of 10. Deng et al. [69] used an ionic liquid to dissolve cellulose and prepared aerogels by freeze drying. Their procedure starts from microcrystalline cellulose of a low degree of polymerization (DP 186) dissolved at 130  C in 1-butyl-3methyl-imidazolium chloride (BmimCl) until a clear solution was obtained. The solution was cast onto a glass plate to obtain films of 1 mm thickness and then regenerated in water until no chloride was detected further. The hydrogels were dried according to different routes, like rapid freezing with liquid nitrogen, guessed to be best to preserve the hydrogel network structure, conventional freeze drying, typically leading to large ice crystals destroying the soft network, and ambient drying in a furnace at 50  C. The results mirror the expectations and knowledge about the preparation of cellulose aerogels described above. SEM and nitrogen adsorption reveal that rapid freezing and conventional freeze drying produce a foam having a 3D open fibrillar network structure with a specific surface area of 186 m2 g
1 and a porosity of 99%. A completely different approach was later used by Liebner and co-workers, using bacterial cellulose [35]. Bacterial cellulose is fascinating since it provides cellulose I with a huge degree of polymerization, namely, around one million. The essential difference to any other way to produce aerogels is that the Gluconacetobacter xylinum used produces in a suitable culture medium by itself an open porous foam. The wet gel simply has to be dried appropriately. Liebner and co-workers used with carbondioxide supercritical drying

24

Cellulose Aerogels: Monoliths, Beads, and Fibers

after solvent exchange with ethanol. Bacterial cellulose wet gels were found to feature a far-reaching dimensional stability upon repeated solvent replacement with ethanol and subsequent CO2 drying. Average shrinking over these two process steps was only about 6.5% which was surprisingly low considering the extremely low cellulose density of the samples being around 0.008 g cm
3 at 1% cellulose content. The microstructure is clearly not classical aerogels, since the pores are all in the micrometer range. An interesting study was performed by Pircher and co-workers [15, 70]. They compared different solvents for cellulose and looked at their impact on the porous microstructure. As a cellulose starting material, they used cotton linters and four different solvent systems, namely, mixtures of (1) tetrabutylammonium fluoride and DMSO (TBAF/ DMSO); (2) 1-ethyl-3-methyl-1H-imidazolium acetate and DMSO (EMIm/DMSO); (3) calcium thiocyanate octahydrate, which in contrast to Jin [59] and Hoepfner [39] was modified with lithium chloride; and (4) molten N-methylmorpholine-N-oxide monohydrate (NMMO) . All gels prepared after dissolution and gelation were regenerated in ethanol and then supercritically dried with carbon dioxide. From their very extensive and interesting study, a few results are important. The SEM analysis of aerogels from solutions

637

of cotton linters in NMMO or Ca(SCN)28 H2O/LiCl revealed random networks of cellulose nanofibrils. The comparison is shown in Fig. 24.16. They studied the bulk shrinkage in all stages of the aerogel production, performed small- and wide-angle x-ray scattering (WAXS) studies, and analyzed the porosity using thermoporosimetry and mechanical properties with compression test. All aerogels exhibited a total shrinkage from gel preparation to the dry state between 23% and 52%. The solvent system used had a dramatic effect. Whereas the ionic liquid EMIm showed with 1.5% cotton linter a shrinkage of 52%, the use of rhodanide/LiCl as a solvent exhibited a shrinkage of 23%. The final envelope densities are 0.042 g cm
3 and 0.030gcm
3. The crystallinity, determined by WAXS, varied interestingly also dramatically from 15 to 45% for both solvents. The compression yield strength is different in the same way, namely, 22 kPa for 1.5% cotton linter dissolved in EMIm and 67 kPa if dissolved in rhodanide/LiCl. The general interpretation of their results is based on the idea that for all solutions, a phase separation occurs. They assume that there is microphase separation of free and bound solvent, accompanied by cellulose alignment to larger nanofibrils. While the free solvent is easily removed upon addition of a suitable cellulose nonsolvent, contact of the regeneration

Fig. 24.16 Scanning electron micrographs of aerogels derived from solutions of 3.0 w% cotton linters in TBAF/DMSO, EMIm/ DMSO, and NMMO and 1.5 w% cotton linters in calcium thiocyanate octahydrate/LiCl after Pircher [15]

CL-TBAF

10 Pm

CL-EMIm

10 Pm

CL-NMMO

10 Pm

CL-CTO

10 Pm

24

638

network with pores below 500 nm and relatively thick, interconnected fibrils.

24.5.5 Mechanical and Thermal Properties The mechanical properties of cellulose II aerogels depend on their envelope density. Although in many papers data are presented, they are often given as a function of cellulose content without specifying the envelope density. We have collected the currently available data on the elastic modulus and plotted them into a single curve to especially look for a scaling behavior. The result is shown in Fig. 24.17. These data follow a scaling law as derived by Gibson and Ashby for open porous materials, namely, an increase of Young’s modulus with a power of 2. The value of 1.8 is close enough to this predicted value, especially with respect to the different data sources. A similar plot can be made for the thermal conductivity. Putting all available data together leads to the picture shown in Fig. 24.18. First it is interesting to note that the only systematic study, namely, that of Karadagli and co-workers [40], exhibits an almost linear increase in thermal conductivity with envelope density. They, however, used the Hot Disk method, whereas in all other cases, a stationary method was used. These data show that cellulose II aerogels have a thermal conductivity being always higher than that of still air at room temperature, shown in the figure as a horizontal line. Thus, in contrast to many other aerogels, cellulose aerogels seem always not to be superinsulating. Since in aerogels the thermal conductivity can be modeled as a sum of solid state conductivity, gas phase contribution, and radiation, one might ask here for the origin of this behavior. Probably the gas phase contribution in these fibrillar network is not that much reduced by the Knudsen effect as in other aerogels.

14 Young’s modulus (MPa)

fluid with bound solvent causes additional cellulose coagulation in agreement with observations of Gavillon and Budtova [28] and Hoepfner [39]. They suggest that coagulation of cellulose from solutions in TBAF or EMIm occurs by spinodal decomposition without, however, specifying the underlying thermodynamics or the kinetics in further detail. From the analysis of the SAXS scattering curves, they suggest that all aerogels are mass fractals, whereas the salt hydrate melt dissolution process leads to a clustered network or a rough surface fractal. They state that from all systems used, calcium thiocyanate-octahydrate with lithium chloride is the most promising solvent due to its low costs and environmental impact. In two recent papers, Qin-yong Mi and co-workers [71] and Ayadi and co-workers [72] produced transparent cellulose aerogels using very different routes. Whereas Qin-yong Mi used 1-allyl-3-methylimidazolium chloride (AMIMCl) to dissolve the cellulose, Ayadi used trifluoroacetic acid as a solvent. The team of Qin-yong Mi dissolved 2 wt% cellulose in AMIMCl and gelled the solution by using a high concentration of AMIMCl as a regeneration bath with varying concentration of 20, 40, and 60% AMIMCl in water. They made a comparison with a material regenerated in pure water. All aerogels had after supercritical drying in carbon dioxide a porosity of around 98%, a Young’s modulus of 15.2 to 27.3 MPa, and a compressive stress at 50% strain of larger than 1.28 MPa up to 1.99 MPa for the gel regenerated in a solution with 60 wt% AMIMCl. The transmittance increased with increasing AMIMCl content of the regeneration bath from around 48.4% for the water regenerated aerogels and 80% if regenerated in a solution with 60 wt% AMIMCl (the transparency was measured at 800 nm and thus is measured in the near-infrared (NIR) region). The microstructure of the gels regenerated in a 40% or 60% AMIMCl solution in the SEM figures appears homogeneous and mesoporous, which is confirmed by the specific surface area being around 175.5– 227.2 m2 g
1. The aerogels produced by Ayadi and team are interesting, since a nonconventional dissolving agent is used. 2,4 and 5.6 wt% microcrystalline cellulose were dispersed in trifluoroacetic acid and then stored at 0  C for 24 h and then kept at room temperature for 10 days until a clear solution was obtained. To this solution, ethanol was added, stirred, and cast into molds. Within 24 h a free standing and transparent organogel was produced. After solvent exchange with ethanol, the gels were dried supercritically with carbon dioxide. The aerogels had densities between 70 mg cm
3 and 220 mg cm
3 and a specific surface area decreasing from around 500 m2 g
1 to 200 m2 g
1 in as much as the density increases. Interestingly the optical transmittance is almost independent of the cellulose concentration and is in the NIR around 80% but decreases toward zero for wavelength of 300 nm. The microstructure (SEM) looks like a fibrillar

L. Ratke et al.

12 10

E = E0 m = 0.00194 1.8

8 6 4 2 0 0

20

40

60 80 100 Density (kg/m3)

120

140

Fig. 24.17 Young’s modulus as a function of envelope density. The data points are collected from [15, 40, 45, 70]

24

Cellulose Aerogels: Monoliths, Beads, and Fibers

639

Thermal conductivity (10–3 W/mK)

80 70 Karadagli Ayadi Nguyen Innerlohinger Qin-Yong Demilecamps

60 50 40 30 20 10

0

50

100 150 Density (kg/m3)

200

250

Fig. 24.18 Thermal conductivity as a function of envelope density as measured by various authors. The data points are collected from [40, 67, 71–74]

24.6

Cellulose Aerogel Beads

Porous cellulose beads and their derivatives have been developed since 1988 [75]. Cellulose aerogel beads in size range micro- to millimeter have attracted great interests due to the fact that they are biodegradable and renewable material and can be an alternative to oil-based plastic beads. There are several advantages beads have in comparison with monoliths. First a large-scale production is easy to achieve. Secondly diffusion and loading of desired molecules into beads are faster than in monoliths, since the characteristic time for diffusion and diffusive transport scales with the square of the bead size, meaning the smaller, the faster. Third, the reactive -OH chemical functional groups of cellulose can offer chelating functions, and they can be chemically modified as hydrophobic or hydrophilic.

the drops produced on the sharp tip fall into the regeneration bath. The droplet size depends on the nozzle diameter, the surface energy, and the viscosity of the cellulose solution. Adopting the same principle of droplet production, i.e., the formation of liquid droplets in a gaseous phase, a scale-up can be achieved by applying additional forces. In commercial droplet production devices, several techniques are used or even combined: the extrusion of a liquid through a nozzle is forced by a defined pressure given to the liquid chamber. This can be combined with an electrostatic cutting tool [77–79] or a vibrational one [80–82] or a mechanical one [83–86]. Also a co-axial gas flow around the nozzle can enhance the production rate, or a gas pressure through an atomizer nozzle [87, 88] can be used to improve the speed of particles disintegration from the orifice. In all these droplet production methods, the particle shape and size vary from disc-like plates to spheres depending upon various parameters such as cellulose concentration, viscosity, conditions of the regeneration bath, and the distance between the orifice and the surface of the regeneration bath. These parameters were discussed in depth in the literature [89]. In an alternative way, the cellulose liquid is dispersed in a nonsolvent liquid. It is most employed for the derivatized cellulose solution which is dispersed in protic nonsolvent medium having surfactants [75] in order to induce coagulation leading to gel network formation. Then the cellulose derivative is converted to cellulose under an acidic condition. The regeneration bath for the gelation process is most often an acidic aqueous-based medium or protic nonsolvent liquid (alcohol) medium having nonionic surfactants. In the latest report, liquid droplets were sprayed into cryogenic liquid nitrogen medium. Depending upon the conditions used for the gelation processes in regeneration bath, the gelation methods were named as reversing pH method, coagulation and/or regeneration method [90], and cryogelation method [87, 88]. The wet gel beads of cellulose are washed to remove the reagents used in their preparation. Generally the pure cellulose beads undergo a solvent exchange step before they finally are dried either under supercritical or ambient conditions.

24.6.1 Methods of Bead Fabrication 24.6.2 Microstructure and Properties of Beads Most often cellulose aerogel beads were prepared by dissolving cellulose in a solvent medium, producing droplets of cellulose by some mechanism, coagulation or regeneration to produce a wet gel network, washing, and solvent exchange followed by supercritical drying. The production of wet gel beads of cellulose was discussed in detail in a recent review article [76]. We therefore summarize some techniques briefly. One of the most employed lab techniques to prepare cellulose droplets is a conventional dripping method using tools like pipettes and syringes. Under the action of gravity,

In the case studies of cellulose aerogel beads, the mesoporous structure of the interconnected nanofibrillar network is essentially maintained after supercritical drying or cryogelationfreeze drying method [87, 88]. The specific surface area is greater than 250 m2 g
1 even though the preparation and drying methods differ in its way. To the best of our knowledge, the production of aerogels of cellulose beads has been described for the first time in 1988 [75], but at that time the materials were termed porous

24

640

L. Ratke et al.

cellulose beads. The filed patent describes that the particles were prepared by dispersing an alkaline aqueous solution of cellulose or cellulose derivative (cellulose xanthate) into a coagulation bath (containing alcohol and nonionic surfactants) having temperature range 30–70  C. Then the regeneration of wet cellulose beads was achieved using a neutralization (acidic) medium. That means finally cellulose xanthate was chemically converted to cellulose using hydrolysis under acidic conditions. The size of the particles was less than 500 μm. The percentages of crystallinity of porous particles were in the range 23–46%. The crystalline domain phase was composed of type cellulose II phase and an amorphous phase. The pore diameter was in the range between 6 nm and 1 μm. Cellulose aerogel beads were prepared by Gavillon [30] using the JetCutter technology from geniaLab, Germany. In this technique, a liquid jet of the solutions is cut by thin wires fixed in a rotating ring. Drop size and production rate depend on the rotation speed, the jet speed and diameter, the viscosity of the solution, and its surface tension. Gavillon prepared a solution of cellulose with 8% NaOH-water and jet cut it into H2SO4 solution. Wet gel beads were washed with water and acetone and dried under supercritical CO2 conditions. The SEM images showed spherical particles and a homogeneous distribution of a fibrillar network (Fig. 24.19). The same jet cutting technology was used to prepare cellulose aerogel beads starting from a solution of cellulose derivative, i.e., cellulose carbamate dissolved in NaOH-water [57]. Finally before drying, in order to convert the cellulose derivative to cellulose, the cellulose carbamate beads were treated either thermally or chemically in an acid bath. Beads of about 0.5 mm were formed. After washing, the beads were dried

a

Acc.V Spot Magn 12.0 kV 3.0 61x

using supercritical CO2. Depending on the polymer concentration, the specific surface area varied from 350 to 540 m2 g
1. Sescousse and co-workers reported on the preparation of cellulose aerogel beads from the solution of cellulose-NaOHwater mixture by coagulating the cellulose solution in water. It was shown that the shape of the beads varied from very flat plates to spheres (Fig. 24.20) [89] depending upon several parameters such as cellulose concentration, delay time, bath temperature, the distance between the pipette and the bath surface, and the mechanical shock of the droplets on the bath surface. The influence of the preparation conditions on the shape of the beads was discussed. A high viscosity, a short distance from the nozzle to the bath, and a high coagulation bath temperature in comparison with that of the cellulose solution are favorable parameters to prepare beads with a spherical shape. Various inorganic powders were encapsulated into cellulose beads. Herein the fibrillar network was used as supporting material for the growth of an inorganic sol. As a result organic-inorganic aerogel particles were prepared (see Fig. 24.21). Recently it was reported that the cellulose aerogels can be prepared from NaOH-urea-water mixture containing ZnO nanoparticles [20]. The beads were regenerated in an acidic aqueous medium. This study showed that the addition of 0.5 wt % of ZnO to the NaOH-urea-water mixture can effectively improve the properties of cellulose aerogel beads/particles (see Fig. 24.22). Very recently, using the combination of atomization, cryogelation, and freeze drying method, the production of finely distributed cellulose nanofibril aerogels was reported [87, 88, 91]. The cellulose nanofibrils were crosslinked with Kymene, and then the cellulose nanofibril solution was

b

Dot WD SE 16.0

1 mm

Acc.V Spot Magn 12.0 kV 3.0 5000x

Dot WD SE 15.6

10 mm

Fig. 24.19 SEM micrographs of aerocellulose prepared from 5%Avicel/7.6NaOH/water solutions regenerated in 10% H2SO4 bath: (a) aerocellulose bead and (b) the cross-sectioned image [30]

24

Cellulose Aerogels: Monoliths, Beads, and Fibers

641

24

Fig. 24.20 Illustrating the photographic images of wet cellulose particles (coagulated cellulose in water) [89]

sprayed at 40 MPa pressure through a 1-mm inner diameter steel nozzle into a liquid nitrogen bath. After freeze drying, the crosslinked cellulose nanofibril aerogels showed high specific surface area of 390 m2 g
1 with a cellulose nanofibril concentration of 0.6% [87]. In an another study, TEMPO-oxidized cellulose nanofibril aerogel microspheres were prepared by using cryogelation, ultrasonic atomization, and freeze drying method, shown in Fig. 24.23 [88]. The size of aerogel particles was in the range of 2– 7 μm. This ultralight cellulose aerogel microsphere with its high water storage capacity was used as novel micro-reactor to extract phenol and copper (II) ions from an external water phase. Beaumont et al. [92] produced nanostructured cellulose II gel from NMMO solution of Lyocell fibers by high-pressure homogenization. This method provides particles with irregular shape having particle sizes in the range between 0.5 and 10 μm. The stability of cellulose particles was dealt with alcohols in order to avoid agglomeration of particles. After freeze drying from tert-butanol medium, the particles showed

a continuous open porous structure (Fig. 24.24) with high specific surface area of about 298 m2 g
1. The particles were termed as cryogels by authors. Omura et al. [93] have reported on spongy cellulose particles which can be used for the encapsulation of hydrophilic and hydrophobic fluorescent molecules. The particles were prepared by solvent releasing method from cellulose droplets which contain ionic liquid, 1-butyl-3-methylimidazolium chloride, and N,N-dimethylformamide. This non-derivatized cellulose solution was dispersed in n-hexane containing a dissolved surfactant in order to make spherical droplets. The cellulose droplets were employed for coagulation by mixing with excess amount of a protic nonsolvent medium, 1-butanol. The specific surface area of supercritically dried samples showed a higher value of 371 m2 g
1 in comparison with vacuum dried sample (a dense structure having 1m2 g
1). 1-allyl-3-methylimidazolium chloride, an ionic liquid, was used as solvent to prepare cellulose beads from cellulose paper wastes [94]. Using a syringe needle, the cellulose solution was dropped into a water coagulation medium. The

642 Fig. 24.21 Photographic image of TiO2-encapsulated (a) and its SEM image (b) and SEM images of iron-encapsulated (c, d) aerocellulose beads [89]

L. Ratke et al.

c

a

500 mm

1 cm b

d

10 mm

specific surface area was varied from 101 to 478 m2 g
1. The authors employed different drying techniques and compared the microstructures. The aerogel of cellulose beads showed open porous nanofibrillar structure with high surface area in comparison with oven-dried and air-dried samples. In addition, aerogels showed high loading capacity of a dye molecule, curcumin, which is about 0.55 mg/g.

20 mm

demanding for sorption, separation, and filter techniques. As it has been demonstrated, the -OH functional groups can be effectively modified to hydrophobic or hydrophilic with chelating functions [76]. The efficiency of sorbent materials is controlled by the porosity that entails the aerogels possessing high porosity and high surface area. Cellulose beads can also be used as supporting materials to encapsulate the inorganic particles [89] or as biological cell culture scaffolds [87].

24.6.3 Applications Cellulose aerogel beads with their physical, chemical, and functional properties can be used in applications such as separation techniques or as carriers for drug delivery. The porous cellulose beads are commercialized in the market with desired chemical functionalities for efficient purification techniques such as sorbents, filters, and chromatographic techniques and personal care products (Iontosorb; Rengo Co. Ltd.; JNC Corporation to name a few). With their regular or irregular spherical shape, they can be used in chromatographic applications as fillers with effective flow resistance. The commercial porous cellulose beads showed that the high flow resistance depends on the particle size and mechanical strength, i.e., the larger particle sizes provide a larger flow rate in comparison with smaller particles at constant pressure. The chemical affinity of the functional groups is

24.7

Cellulose Filaments

Cellulose is one of the oldest materials used to produce fibers, filaments, and yarns, from which fabrics of all kinds are manufactured. Today filament and yarn production is a field for both natural and synthetic polymers [95]. All fibers used today have a compact, dense microstructure and a very large aspect ratio. Therefore it would be a rather promising endeavor to try and manufacture open porous fibers from cellulose solutions with an aerogel structure. The first attempts have been made by Hoepfner and Ratke [96] and taken up by Schmenk [97] and Hacker [98]. Schulz, Schestakow, and Karadagli published in 2015 a comprehensive study on cellulose fiber production using different salt melt hydrate routes [40, 99].

24

Cellulose Aerogels: Monoliths, Beads, and Fibers

a

b

c

200 nm

Fig. 24.22 Cellulose aerogel beads prepared in the presence of 0.5% of ZnO nanoparticles. The wet gels of cellulose beads/particles (a), SEM image of cellulose aerogel particle (b), SEM micrograph of the inner fractured surface (c) [20]

Fig. 24.23 SEM images of TEMPO-oxidized cellulose aerogel microsphere prepared by freeze drying method [88]

643

Polymeric fibers are typically prepared from a melt and spun in a spinneret to yield thin fibers. For cellulose aerogel fibers, a classical wet spin method must be applied, and the wet gel fibers must be regenerated, the solvent must be washed out, and the fibers have to be dried supercritically. Several techniques were developed both with respect to the cellulose solution used and the spinning technique. In the first technique, the cellulose aerogels were synthesized from microcrystalline cellulose in a hydrated calcium thiocyanate salt melt, which upon cooling forms a gel at around 80  C. Schulz, Karadagli, and co-workers used a twin screw extrusion setup, shown in Fig. 24.25. The extrusion experiments were performed with hot cellulose solutions at different temperature from 90  C to 110  C yielding thin and wet cellulose filaments of 0.5 and 1 mm diameter. Washing and coagulation of the wet gels in ethanol were followed by supercritical drying with CO2 yielding cellulose aerogel filaments. Although the microstructure could be described as an open porous network of nanofibrils with pore sizes ranging from 10 to 100 nm and fibril diameters of around 10 to 25 nm, there is a speciality in these fibers compared to, for instance, cellulose aerogel monoliths. In the extrusion process, the hot melt is directly pressed into the regeneration bath, which is at room temperature. Therefore, the regeneration fluid used, e.g., the ethanol, evaporates immediately close to the wet gel fiber surface, and a part of this ethanol steam enters the fibers and leaves behind macrochannels directly beneath the fiber surface. Figure 24.26 shows a few examples of the fibers. The fiber core shows the typical nanostructured fibrillar structure (see Fig. 24.27). The densities of supercritically dried cellulose aerogels were in the range of 9–137 kg m
3, and the BET-specific surface areas were between 120 and 230 m2 g
1. The cellulose aerogel possesses a fiber strength in tension between 4.5 and 6.4 MPa. The tensile strength of

b

a

1 Pm

1 Pm

24

644

L. Ratke et al.

Freeze-dried from water

a

Freeze-dried from tert-butanol

b

60 nm

40 nm

40 nm

40 nm

50 nm

50 nm 40 nm

5 Pm

50 nm

50 nm

5 Pm

60 nm

40 nm

1 Pm

Fig. 24.24 SEM pictures of freeze-dried particles obtained from hydrogel (a) and alcogel (b). The higher magnification of freeze-dried alcogel (middle) showed features of fibrillary network. (Reprinted with permission from [92]. Copyright 2016 American Chemical Society)

Fig. 24.25 Scheme of the extrusion facility: (a, b) two conical, fully intermeshing corotating twin screws, (c) filling of a viscous cellulose salt melt hydrate solution into an extruder. (With permission from Elsevier)

a

aerogel filaments increases with the increasing cellulose amount in the spin dope. However, the cellulose amount is limited to the solubility of cellulose in the chosen solvent as well as to a viscosity range that can be processed. Their strength is not high enough to allow any further textile processing. Much better results can be obtained using a piston pump device (see Fig. 24.28) or a gear pump technique. Both techniques were developed and applied by Schulz [99] and Schestakow [100] to cellulose dopes prepared from calcium

b

c

thiocyanate or zinc chloride hydrates as solvents. The essential difference between both devices is that a piston pump runs discontinuously, meaning only a given amount of solution in the heated cylinder (see Fig. 24.28) can be pressed through the nozzle, whereas in the gear pump device, the solution is continuously fed into the pump consisting of special gears, and the solution is pumped through the nozzle under a suitable pressure. Different types of cellulose were used and first pre-swelled in water followed by dissolution with adding the salt. After

24

Cellulose Aerogels: Monoliths, Beads, and Fibers

645

24 a

100 Pm

c

100 Pm

b

20 Pm

d

10 Pm

Fig. 24.26 SEM images of 3 wt.% cellulose aerogel fibers produced by twin screw extrusion exhibiting a mantle-core structure with different extrusion conditions: (a, b) 105  C, 100 rpm, fiber diameter 1.0 mm and

(c, d) 100  C, 200 rpm, fiber diameter 0.5 mm. (With permission from Elsevier)

homogenization of the mixture at 70  C in the case of zinc chloride or 110  C in the case of calcium thiocyanate, the cellulose salt melt could be directly spun into the regeneration bath, followed by winding up the fibers to give spindles (see Fig. 24.28). Filaments with diameters of 18 to 300 μm in diameter can be achieved. An example of wound up filaments before and after supercritical drying is shown in Fig. 24.29. Several regeneration media, like ethanol, water, and isopropanol, were tested. Ethanol and water lead to disintegration of the cellulose fibrils into a packed bed of precipitates, whereas isopropanol left the fibers intact. Figure 24.30 shows a SEM figure of the thin fibers prepared using the zinc chloride route. Essentially the fiber surfaces appear smooth, but a few fibers look undulated pointing to inhomogeneous flux of the viscous melt through the nozzle. The fibers had a fineness of 2.7 to 3.7 dtex (dtex is the mass in grams per 10,000 m of filament). Due to the fact that salt hydrate melts, especially zinc chloride, create a harsh environment for cellulose degeneration, processing effects on the cellulose macromolecules are

inevitable. During processing times of 30 minutes at 70  C, the initial DP of 201 is reduced to 180. Nitrogen adsorption results show that cellulose aerogel fibers from the zinc chloride route exhibit internal surface areas of 70 to 268 m2 g
1. The measured tensile strengths of cellulose aerogel fibers are in the range of wool fibers with values of about 12 cN/tex. Figure 24.31 exemplarily shows force strain diagrams of cellulose aerogel fibers. The fibers are very flexible with elongations up to 50%. For any textile processing, it is important that fibers are so flexible that knotting is possible. The thin fibers made with this process allow indeed to make knots without fiber breaking. The fracture surface of these fine fibers reveals a fibrillar, open porous structure with cellulose nanofibrils having diameters around 10–20 nm. A few figures showing the nanostructure inside the filaments are shown in Fig. 24.32. In contrast to the extrusion experiments, Schulz and co-workers were also able to use multi-threading nozzles, i.e., producing fiber bundles or multifilaments consisting of 50 filaments with diameters in the range of only a few ten

646

L. Ratke et al.

a

10 Pm

b

1 Pm

d

c

1 Pm

2 Pm

Fig. 24.27 SEM images of the core region of the cellulose aerogel fibers shown in Fig. 24.26 in higher magnifications: (a, b) the core of the fibers shown in Fig. 24.26a, b, (c, d) the core of the fibers shown in Fig. 24.26c, d. (With permission from Elsevier)

Metering pump

a

Gear head

b wet gel filaments

supercritically dried aerogel filaments

Heating

Nozzle head Regeneration Guide bath

Fiber

Bobbin

Fig. 24.28 Scheme of a piston pump setup. The cellulose solution is in a heated tube and pressed through the heated nozzle tube directly into the regeneration bath. A bobbin winds up the filaments

Fig. 24.29 Pictures of the wound up cellulose wet gel (a) and the supercritically dried cellulose aerogel filament (b)

24

Cellulose Aerogels: Monoliths, Beads, and Fibers

a

b

d

e

647

100 Pm

10 Pm

Fig. 24.30 Comparison of cellulose aerogel fibers prepared from a piston pump extrusion process. Images (a–c) show aerogels prepared from Ca(SCN)2 illustrating monofilaments of 300 μm in diameter, while

c

f

1 Pm

2 Pm

images (d–f) show ZnCl2-derived multifilaments of cellulose aerogels of 95 μm diameter

3

Force (cN)

2,5 2 1,5 1 0,5 0

200 Pm

0

10

20

30 40 Strain (%)

50

60

Fig. 24.31 Force strain diagrams of fibers tested in tension

micrometers pointing the way from mono- to multifilament production.

24.8

Summary and Outlook

Cellulose aerogels in the shape of monoliths, beads, or filaments are at the beginning of intensive research, and owing to their fascinating microstructure and properties, they promise to have many applications. Various routes have been tested to produce aerogels. They all start with dissolving cellulose in a

Fig. 24.32 SEM picture of fibers or filaments prepared in the piston pump device after supercritical drying

solvent, like sodium hydroxide, salt hydrate melts, or ionic liquids, followed by a washing step and most often supercritically drying. Freeze drying rarely leads to cellulose materials with a large amount of mesoporosity. Summarizing the literature on cellulose gel formation from any of the solvent media used, it seems that the mechanism of dissolution is essentially understood, namely, the breaking of the intermolecular hydrogen bonds and the complexation of the hydroxyl groups by, for instance, sodium and its hydration shell or rhodanide or zinc cations and their hydration shell. The gelation process is, however, not understood and this in several aspects. First the aggregation of the dissolved

24

648

polymers to fibrils is not clear, and the understanding of this process would be important to control the fibril diameter. As mentioned above, statements are made in the literature that the polymers arrange parallel each other along their hydrophobic planes, but the point is how do they do this and why? Since the dissolved polymers are not naked polymers but are surrounded by ions and a hydration shell, one could imagine them as polyelectrolytes. If this would be true, the charge distribution on the cellulose polymer would lead to an aggregation. This might lead to a parallel alignment or an endto-end alignment, depending on the charge, as discussed theoretically in the literature on polyelectrolytes [101, 102]. Second, once fibrils are formed by, for instance, aggregation, how do these interact to form a network and thus a gel? If a phase separation process leads to the gel network, it would be essential to first understand the thermodynamics of cellulose solvent systems, and for our understanding, such investigations have rarely been done, or they are extremely old. If the thermodynamic functions of cellulose-solvent systems are known, one could easily set up theoretical models of the kinetics of phase separation as has been done in the context of other polymer systems [43]. One might even think that especially using the pH inversion method, the formation of gels from cellulose might be a reaction-diffusion problem [103]. There is another open set of questions concerning the microstructure and some properties of cellulose aerogels. In many cases, the microstructure is described as a felt or fleece with nanosized fibrils. Till now no theoretical description of such a fleece with strong bonds at the node points is available. Such descriptions are used in the context of papers and card boards but have never been applied to aerogels. This is, however, essential especially with respect to properties like thermal conductivity and gas permeability. For nanoporous materials, typically the gas phase contribution to thermal conductivity is described by the so-called Knudsen effect, meaning the ratio of mean free path of a gas molecules and the pore diameter determines the reduction of thermal diffusion and thus thermal conductivity [104]. The question in the context of cellulose aerogels is, however, what are pores? The cellulose aerogels are open porous, without doubt, but the pores have nothing to do with any type of cylindrical pores; at best they look like cylinders with many holes in their walls. Therefore the open question to be solved is: how to describe the Knudsen effect in such a network of fibrils? This also is essential to describe the permeability, which often is done by a type of Carman-Kozeny relation and thus related to the pore diameter [104]. In many cases studied till now, microcrystalline cellulose is used to prepare aerogels. In order to exploit cellulose as a renewable and biodegradable material, the variety of possible cellulose pulp sources has to be studied in much more detail and much more extensively. Concerning beads and filaments,

L. Ratke et al.

open issues are crosslinking to stabilize and strengthen the dry aerogel structure and thus to enhance processes of filament drawing and yarn production. In addition, the stability of dry cellulose aerogels against humidity has to be improved, if cellulose aerogels would have to be used, for instance, as a natural isolation material for buildings or even a pizza pack. Since almost all cellulose aerogels need supercritical drying, the processing route has to be optimized and made cost-effective.

References 1. Glucose. URL https://en.wikipedia.org/wiki/Glucose 2. Kräassig, H.A.: Cellulose. Gordon and Breach Science Publishers, Amsterdam (1993) 3. Cellulose. URL https://en.wikipedia.org/wiki/Cellulose 4. Zugenmaier, P.: Crystalline Cellulose and Derivatives: Characterization and Structures. Springer Series in Wood Science. Springer, Berlin/Heidelberg (2008) 5. Klemm, D., Phlipp, B., Heinze, T., Heinze, U., Wagenknecht, W.: Comprehensive Cellulose Chemistry: Fundamentals and Analytical Methods, vol. 1 and 2. Wiley-VCH Verlag GmbH (1998) 6. Marsh, J., Wood, F.C.: An Introduction to the Chemistry of Cellulose, 2nd edn. Chapman & Hall Ltd., London (1942) 7. Hearle, J.: A fringed fibril theory of structure in crystalline polymers. J. Polym. Sci. 28(117), 432–435 (1958) 8. Moore: Botany Visual Resource Library. McGraw-Hill Companies (1997) 9. Uno, G., Storey, R., Moore, R.: Principles of Botany. McGraw-Hill (2001) 10. Israelachvili, J.N.: Intermolecular and Surface Forces. Academic, Toronto (1985) 11. Olsson, C., Westman, G.: Direct dissolution of cellulose: background, means and applications. In: de Wen, T.G.V. (ed.) Cellulose – Fundamental Aspects, Chap. 6. INTECH (2013). https://doi.org/ 1 0 . 5 7 7 2 / 5 2 1 4 4 . h t t p s : / / w w w. i n t e c h o p e n . c o m / b o o k s / howtoreference/cellulose-fundamental-aspects/direct-dissolutionof-cellulose-background-means-and-applications 12. Budtova, T., Navard, P.: Cellulose in NaOH-water based solvents: a review. Cellulose. 23, 5–55 (2016) 13. Medronho, B., Lindmann, B.: Brief overview on cellulose dissolution/regeneration interactions. Adv. Colloid Interf. Sci. 222, 502–508 (2015) 14. Kihlman, M., Medronho, B., Romano, A., Germgard, U., Lindman, B.: Cellulose dissolution in an alkali based solvent: influence of additives and pretreatments. J. Braz. Chem. Soc. 24(2), 295–303 (2013) 15. Pircher, N., Carbajal, L., Schimper, C., Bacher, M., Rennhofer, H., Nedec, J.M., Lichtenegger, H.C., Rosenau, T., Liebner, F.: Impact of selected solvent systems on the pore and solid structure of cellulose aerogels. Cellulose. 23, 1949–1966 (2016) 16. Sen, S., Martin, J.D., Argyropoulos, D.S.: Review of cellulose non-derivatizing solvent interactions with emphasis on activity in inorganic molten salt hydrates. ACS Sustain. Chem. Eng. 1(8), 858–870 (2013). https://doi.org/10.1021/sc400085a 17. Wang, S., Lu, A., Zhang, L.: Recent advances in regenerated cellulose materials. Prog. Polym. Sci. 53, 169–206 (2016) 18. Hirosi, S., Heinz, K., Kurt, H.: Das system cellulose– natriumhydroxyd–wasser in abhängigkeit von der temperatur. Z. Phys. Chem. 43B(1), 309–328 (1939). https://doi.org/10.1515/ zpch-1939-4324

24

Cellulose Aerogels: Monoliths, Beads, and Fibers

19. Degen, A., Kosec, M.: Effect of ph and impurities on the surface charge of zinc oxide in aqueous solution. J. Eur. Ceram. Soc. 20, 667–673 (2000) 20. Mohamed, S.M.K., Ganesan, K., Milow, B., Ratke, L.: The effect of zinc oxide (zno) addition on the physical and morphological properties of cellulose aerogel beads. RSC Adv. 5(90), 193–90,201 (2015) 21. Cibik, T.: Untersuchungen am system nmmo/h2o/cellulose. Ph.D. thesis, TU Berlin (2003) 22. Walker, M., Zimmerman, R., Whitcombe, G., Humbert, H.: N-methylmorpholinoxid (nmmo) – die entwicklung eines losemittels zur industriellen produktion von zellulosefasern. Lenzinger Berichte. 97, 76–80 (1997) 23. Fischer, S.: Anorganische salzhydratschmelzen – ein unkonventionelles lose- und reaktionsmedium fur cellulose. Ph. D. thesis, TU Bergakademie Freiberg, Freiberg (2004) 24. Philip, B., Schleicher, H., Wagenknecht, W.: ChemTech. 7, 702–709 (1977) 25. Deguchi, S., Tsujii, K., Horikoshi, K.: Crystalline-to-amorphous transformation of cellulose in hot and compressed water and its implications for hydrothermal conversion. Green. Chem. 10, 191–196 (2008). https://doi.org/10.1039/B713655B 26. Sen, S., Losey, B.P., Gordon, E.E., Argyropoulos, D.S., Martin, J.D.: Ionic liquid character of zinc chloride hydrates define solvent characteristics that afford the solubility of cellulose. J. Phys. Chem. B. 120, 1134–1141 (2016) 27. Wilcox, R.J., Losey, B.P., Folmer, J.C., Martin, J.D., Zeller, M., Sommer, R.: Crystalline and liquid structure of zinc chloride trihydrate: a unique ionic liquid. Inorg. Chem. 54, 1109–1119 (2015) 28. Gavillon, R., Budtova, T.: Aerocellulose: new highly porous cellulose prepared from cellulose-naoh aqueous solutions. Biomacromolecules. 9, 269–277 (2008) 29. Liu, W., Budtova, T., Navard, P.: Influence of ZnO on the properties of dilute and semi-dilute cellulose-NaOH-water solutions. Cellulose. 18, 911–920 (2011) 30. Gavillon, R.: Preparation et caracterisation der materiaux cellulosiques ultra poreux. Ph.D. thesis, Ecole des Mines de Paris, Paris (2007) 31. Isobe, N., Kimura, S., Wada, M., Kuga, S.: Mechanism of cellulose gelation from aqueous alkali-urea solution. Carbohydr. Polym. 89, 1298–1300 (2012) 32. Trygg, J., Fardim, P., Gericke, M., Makila, E., Salonen, J.: Physicochemical design of the morphology and ultrastructure of cellulose beads. Carbohydr. Polym. 93(1), 291–299 (2013). https:// doi.org/10.1016/j.carbpol.2012.03.085. 2nd European Polysaccharide Network of Excellence International Conference, August 2011 33. Innerlohinger, J., Weber, H.K., Kraft, G.: Aerocellulose: aerogels and aerogel-like materials made from cellulose. Macromol. Symp. 244, 126–135 (2006) 34. Liebner, F., Haimer, E., Potthast, A., Lodl, D., Tschegg, S., Neouze, M.A., Wendland, M., Rosenau, T.: Cellulosic aerogels as ultralightweight materials. Part 2: synthesis and properties. Holzforschung. 63, 3–11 (2009) 35. Liebner, F., Haimer, E., Wendland, M., Neouze, M.A., Schlufter, K., Miethe, P., Heinze, T., Potthast, A., Rosenau, T.: Aerogels from unaltered bacterial cellulose: application of scco2 drying for the preparation of shaped, ultra-lightweight cellulosic aerogels. Macromol. Biosci. 10, 349–352 (2010) 36. Liebner, F., Potthast, A., Rosenau, T., Haimer, E., Wendland, M.: Ultralightweight cellulose aerogels from NBnMO-stabilized Lyocell dopes. Res Lett. Mater. Sci. 2007(ID 7324) (2007). https://doi.org/10.1155/2007/73724 37. Liebner, F., Rosenau, T., Wendland, M., Potthast, A., Haimer, E.: Cellulose aerogels: highly porous, ultra-lightweight materials. Holzforschung. 62(2), 129–135 (2008)

649 38. Perrut, M., Smirnova, I. (eds.): Aerogels from bacterial cellulose: a new dimension in preparing shaped cellulose aerogels. ISASF, International Society for Advancement of Supercritical Fluids (2010) 39. Hoepfner, S., Ratke, L., Milow, B.: Synthesis and characterization of nanofibrillar cellulose aerogels. Cellulose. 15, 121–129 (2008) 40. Karadagli, I., Schulz, B., Schestakow, M., Milow, B., Gries, T., Ratke, L.: Production of porous cellulose aerogel fibers by an extrusion process. J. Supercrit. Fluids. 106, 105–114 (2015) 41. Schestakow, M., Karadagli, I., Ratke, L.: Cellulose aerogels prepared from an aqueous zinc chloride salt hydrate melt. Carbohydr. Polym. 137, 642–649 (2016) 42. Principles of Polymer Chemistry. Cornell University Press, Ithaca, New York (1953) 43. Teraoka, I.: Polymer Solutions. Wiley, New York (2002) 44. Buchtová, N., Budtova, T.: Cellulose aero-, cryo- and xerogels: towards understanding of morphology control. Cellulose. 23(4), 2585–2595 (2016). https://doi.org/10.1007/s10570-016-0960-8 45. Ganesan, K., Dennstedt, A., Barowski, A., Ratke, L.: Design of aerogels, cryogels and xerogels of cellulose with hierarchical porous structure. Mater. Des. 92, 345–355 (2016) 46. Pour, G., Beauger, C., Rigacci, A., Budtova, T.: Xerocellulose: lightweight, porous and hydrophobic cellulose prepared via ambient drying. J. Mater. Sci. 50(13), 4526–4535 (2015). https://doi. org/10.1007/s10853-015-9002-4 47. Dantzig, J., Rappaz, M.: Solidification, 2nd edn. EPFL Press (2018) 48. Kurz, W., Fisher, D.: Fundamentals of Solidification, 4th edn. Trans. Tech. Publisher, Aedermannsdorf (1998) 49. Glicksman, M.E.: Principles of Solidification. Springer Science and Business Media LLC, New York (2011) 50. Stefanescu, D.M.: Science and Engineering of Casting Solidification. Springer Science and Business Media LLC, New York (2009) 51. Smirnova, I., Gurikov, P.: Aerogels in chemical engineering: strategies toward tailor-made aerogels. Ann. Rev Chem. Biomol. Eng. 8(1), 307–334 (2017). https://doi.org/10.1146/annurevchembioeng-060816-101458. PMID: 28375771 52. Weatherwax, R., Caulfield, D.: Cellulose aerogels: an improved method for preparing a highly expanded form of dry cellulose. TAPPI. 54(6), 985–986 (1971) 53. Weatherwax, R.C.: Collapse of cell-wall pores during drying of cellulose. J. Colloid Interface Sci. 62(3), 432–446 (1977). https:// doi.org/10.1016/0021-9797(77)90094-7 54. Alince, B.: Porosity of swollen solvent-exchanged cellulose and its collapse during final liquid removal. Colloid Polymer Sci. 253, 720–729 (1975) 55. Tan, C., Fung, B.M., Newman, J.K., Yu, C.: Organic aerogels with very high impact strength. Adv. Mater. 13(9), 644–646 (2001) 56. Dieter, G.E.: Mechanical Metallurgy, 3rd edn. McGraw Hill, New York (1986) 57. Pinnow, M., Fink, H.P., Fanter, C., Kunze, J.: Characterization of highly porous materials from cellulose carbamate. Macromol. Symp. 262, 129–139 (2008) 58. Gan, S., Zakaria, S., Chia, C.H., Chen, R.S., Ellis, A.V., Kaco, H.: Highly porous regenerated cellulose hydrogel and aerogel prepared from hydrothermal synthesized cellulose carbamate. PLOS One. (2017). https://doi.org/10.1371/journal.pone.0173743 59. Jin, H., Nishiyama, Y., Wada, M., Kuga, S.: Nanofibrillar cellulose aerogels. Colloids Surf. A Physicochem. Eng. Asp. 240, 63–67 (2004) 60. Cai, J., Kimura, S., Wada, M., Kuga, S., Zhang, L.: Cellulose aerogels from aqueous alkali hydroxide-urea solution. ChemSusChem. 1, 149–154 (2008) 61. Reichenauer, G., Scherer, G.W.: Nitrogen sorption in aerogels. J. Non Cryst. Solids. 285(1–3), 167–174 (2001). https://doi.org/ 10.1016/S0022-3093(01)00449-5 62. Polystyrene. URL https://en.wikipedia.org/wiki/Polystyrene

24

650 63. Underwood, E.E.: Quantitative Stereology. Addison-Wesley, Reading (1970) 64. Rege, A., Itskov, M.: A microcell-based constitutive modeling of cellulose aerogels under tension. Acta Mech. (2017). https://doi. org/10.1007/s00707-017-1987-0 65. Rege, A., Schestakow, M., Karadagli, I., Ratke, L., Itskov, M.: Micro-mechanical modelling of cellulose aerogels from molten salt hydrates. Soft Matter. 12, 7079–7088 (2016) 66. Gibson, L.J., Ashby, M.F.: Cellular Solids, 2nd edn. Cambridge University Press, Cambridge, UK (1997) 67. Innerlohinger, J., Weber, H.K., Kraft, G.: Aerocell: aerogels from cellulosic materials. Lenzinger Berichte. 86, 137–143 (2006) 68. Aaltonen, O., Jauhiainen, O.: The preparation of lignocellulosic aerogels from ionic liquid solutions. Carbohydr. Polym. 75, 125–129 (2009) 69. Deng, M., Zhou, Q., Du, A., van Kasteren, J., Wang, Y.: Preparation of nanoporous cellulose foams from cellulose-ionic liquid solutions. Mater. Lett. 63(21), 1851–1854 (2009). https://doi.org/ 10.1016/j.matlet.2009.05.064 70. Pircher, N., Fischhuber, D., Carbajal, L., Strauß, C., Nedec, J.M., Kasper, C., Rosenau, T., Liebner, F.: Preparation and reinforcement of dual-porous biocompatible cellulose scaffolds for tissue engineering. Macromol. Mater. Eng. 300(9), 911–924 (2015) 71. Yong Mi, Q., Rong Ma, S., Yu, J., Song He, J., Zhang, J.: Flexible and transparent cellulose aerogels with uniform nanoporous structure by a controlled regeneration process. ACS Sustain. Chem. Eng. 4, 656–660 (2016) 72. Ayadi, F., Ceseracciu, L., Scarpinelli, A.: Mechanically flexible and optically transparent three-dimensional nanofibrous amorphous aerocellulose. Carbohydr. Polym. 149, 217–223 (2016) 73. Demilecamps, A., Beauger, C., Hildenbrand, C., Rigacci, A.: Cellulose-silica aerogels. Carbohydr. Polym. 122, 293–300 (2015) 74. Nguyen, S.T., Feng, J., Ng, S.K., Wong, J.P., Tan, V.B., Duong, H. M.: Advanced thermal insulation and absorption properties of recycled cellulose aerogels. Colloids Surf. A Physicochem. Eng. Aspects. 445, 128–134 (2014) 75. Ookuma, S., Igarashi, K., Hara, M., Aso, K., Yoshidome, H., Nakayama, H., Suzuki, K., Nakajima, K.: Porous ion-exchanged fine cellulose particles, method for production thereof, and affinity carrier. US Patent US5196527A. (1988) 76. Gericke, M., Trygg, J., Fardim, P.: Functional cellulose beads: preparation, characterization, and applications. Chem. Rev. 113, 4812–4836 (2013) 77. Bugarski, B., Li, Q., Goosen, M.F.A., Poncelet, D., Neufeld, R.J., Vunjak, G.: Electrostatic droplet generation: mechanism of polymer droplet formation. AICHE J. 40, 1026–1031 (1994) 78. Halle, J.P., Leblond, F.A., Pariseau, J.F., Jutras, P., Bradant, M.J., Lepage, Y.: Studies on small (¡300 micrometers) microcapsules. parameters governing the production of alginate beads by highvoltage electrostatic pulses. Cell Transplant. 3, 365–372 (1994) 79. Poncelet, D., Neufeld, R., Bugarski, B., Amsden, B.G., Zhu, J., Goosen, M.F.A.: A parallel plate electrostatic droplet generator: parameters affecting microbead size. Appl. Microbiol. Biotechnol. 42, 251–255 (1994) 80. Cicco, F.D., Russoa, P., Reverchon, E., Garcia-Gonzalez, C.A., Aquino, R.P., Gaudio, P.D.: Prilling and supercritical drying: a successful duo to produce core-shell polysaccharide aerogel beads for wound healing. Carbohydr. Polym. 147, 482–489 (2016) 81. Gaudio, P.D., Colombo, P., Colombo, G., Russo, P., Sonvico, F.: Mechanisms of formation and disintegration of alginate beads obtained by prilling. Int. J. Pharm. 302, 1–9 (2005) 82. Ghosal, S.K., Talukdar, P., Pal, T.K.: Standardization of a newly designed vibrating capillary apparatus for the preparation of microcapsules. Chem. Eng. Technol. 16, 395–398 (1993) 83. Bilancetti, L., Bučko, M., Bugarski, B., Bukowski, J., Gemeiner, P., Lewińska, D., Manojlovic, V., Massart, B., Nastruzzi, C.,

L. Ratke et al. Nedovic, V., Poncelet, D., Siebenhaar, S., Tobler, L., Tosi, A., Vikartovská, A., Vorlop, K.D.: Comparison of different technologies for alginate beads production. Chem. Pap. 62, 364 (2008) 84. Prüsse, U., Bruske, F., Breford, J., Vorlop, K.D.: Improvement of the jet cutting method for the preparation of spherical particles from viscous polymer solutions. Chem. Eng. Technol. 21, 153–157 (1998) 85. Prüsse, U., Fox, B., Kirchhoff, M., Bruske, F., Breford, J., Vorlop, K.D.: New process (jet cutting method) for the production of spherical beads from highly viscous polymer solutions. Chem. Eng. Technol. 21, 29–33 (1998) 86. Prüsse, U., Jahnz, U., Wittlich, P., Breford, J., Vorlop, K.D.: Bead production with jetcutting and rotating disc/nozzle technologies. Landbauforschung Vüolkenrode, 1–10 (2002) 87. Cai, H., Sharma, S., Liu, W., Mu, W., Liu, W., Zhang, X., Deng, Y.: Aerogel microspheres from natural cellulose nanofibrils and their application as cell culture scaffold. Biomacromolecules. 15(7), 2540–2547 (2014). https://doi.org/10.1021/bm5003976 88. Zhang, F., Ren, H., Dou, J., Tong, G., Deng, Y.: Cellulose nanofibril based-aerogel microreactors: a high efficiency and easy recoverable w/o/w membrane separation system. Sci. Rep. 7, 40,096 EP (2017). https://doi.org/10.1038/srep40096 89. Sescousse, R., Gavillon, R., Budtova, T.: Wet and dry highly porous cellulose beads from cellulose-NaOH-water solutions: influence of the preparation conditions on beads shape and encapsulation of inorganic particles. J. Mater. Sci. 46(3), 759–765 (2011) 90. Bidoret, M.E., Martens, E., Smet, B.D., Poncelet, D.: Cell Microencapsulation. Methods Mol. Biol. 1479. Humana Press (2017) 91. Jiménez-Saelices, C., Seantier, B., Cathala, B., Grohens, Y.: Spray freeze-dried nanofibrillated cellulose aerogels with thermal superinsulating properties. Carbohydr. Polym. 157, 105–113 (2017). https://doi.org/10.1016/j.carbpol.2016.09.068 92. Beaumont, M., Rennhofer, H., Opietnik, M., Lichtenegger, H.C., Potthast, A., Rosenau, T.: Nanostructured cellulose ii gel consisting of spherical particles. ACS Sustain. Chem. Eng. 4(8), 4424–4432 (2016). https://doi.org/10.1021/acssuschemeng.6b01036 93. Omura, T., Imagawa, K., Kono, K., Suzuki, T., Minami, H.: Encapsulation of either hydrophilic or hydrophobic substances in spongy cellulose particles. ACS Appl. Mater. Interfaces. 9(1), 944–949 (2017). https://doi.org/10.1021/acsami.6b13261 94. Voon, L.K., Pang, S.C., Chin, S.F.: Porous cellulose beads fabricated from regenerated cellulose as potential drug delivery carriers. J. Chem. 2017, 11 (2017) 95. Cook, J.G.: Handbook of Textile Fibres. Woodhead Publishing Series in Textiles. Woodhead Publishing Limited, Cambridge (2001) 96. Hoepfner, S., Ratke, L.: Open porous cellulose aerogel fibers. Tech. rep., German Aerospace Center DLR, Cologne, Germany (2008). https://doi.org/10.13140/RG.2.2.12012.26246 97. Schmenk, B., Ratke, L., Gries, T.: Solution spinning process for porous cellulose aerogel filaments. In: Dörfel, A. (ed.) Proceedings of the 2nd Aachen-Dresden International Textile Conference. Institute of Textile and Clothing Technology, TU Dresden (2008) 98. Hacker, C., Gries, T., Popescu, C., Ratke, L.: Solution spinning process for highly porous, nanostructured cellulose fibers. Chem. Fibers Int. 59, 85–87 (2009) 99. Schulz, B.: Cellulose-aerogelfasern. Ph.D. thesis, RWTH Aachen University, Aachen (2015) 100. Schestakow, M., Schulz, B., Ratke, L.: Cellulose aerogel fibers using aqueous zinc chloride salt hydrate melts. In: 7th AachenDresden International Textile Conference, Aachen-Dresden International Textile Conference ADITC, Aachen (2013) 101. Fazli, H., Golestanian, R.: Aggregation kinetics of stiff polyelectrolytes in the presence of multivalent salt. Phys. Rev. E. 76, 041,801 (2007)

24

Cellulose Aerogels: Monoliths, Beads, and Fibers

651

102. Fazli, H., Mohammadinejad, S., Golestanian, R.: Salt-induced aggregation of stiff polyelectrolytes. J.Phys. Condens. Matt. 21, 424,111 (2009). https://doi.org/10.1088/0953-8984/21/42/424111 103. Grzybowski, B.A.: Chemistry in Motion: Reaction-Diffusion System for Micro- and Nanotechnology. Wiley (2009) 104. Kaviany, M.: Principles of Heat Transfer in Porous Media, 2nd edn. Springer (1995)

Lorenz Ratke is a professor in Materials Physics at the Technical University of Aachen and retired head of the department “Aerogels” in the Institute of Materials Research at the German Aerospace Center DLR in Cologne. He received his diploma in Physics at the University of Muenster, PhD in Metal Physics from the University of Aachen. Professor Ratke held positions at the Vereinigte Aluminium-Werke Bonn, the Technical University in Clausthal, and the Max Planck Institute for Metals Research in Stuttgart and visiting positions at the Technical University of Miskolc, Hungary, and the University of Iowa. He has received the Georg-Sachs-Award of the German Society of Materials, a honoris causa doctorate degree of the University of Miskolc, the Innovation award of the City of Cologne, the Bavarian State Innovation award. He has published over 300 papers in the areas of corrosion, powder metallurgy, in situ composites, polyphase solidification, coarsening, and aerogels.

24

Kathirvel Ganesan is a scientific co-worker in the Department of Aerogels and Aerogel Composites, Institute of Materials Research at the German Aerospace Center (DLR), Cologne, Germany. He completed his M.Sc. in Chemistry at the Bharathidasan University, Trichy, Tamil Nadu, India. He received his Ph.D. in Inorganic Chemistry from the University of Duisburg-Essen, Essen, Germany. He has published 14 papers in decent international journals in the area of biominerals and biopolymers. His current research interests are biopolymer-based porous materials.

Maria Schestakow is a PhD student at the German aerospace center (DLR). Currently, she is working on various types of composite materials involving cellulose aerogels. She is most interested in gelation mechanisms of cellulose and the preparation of bio-based composites and especially their mechanical properties.

Biopolymer-Silica Aerogel Nanocomposites

25

Shanyu Zhao, Wim J. Malfait, Chunhua Jennifer Yao, Xipeng Liu, Matthias M. Koebel, and William M. Risen

Contents 25.1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 654

25.2

Cellulose-Silica Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 654

25.3

Pectin-Silica Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 656

25.10

Pharmaceutical and Biomedical Applications of Biopolymer-Silica Aerogel Composites . . . . . . . . . . . . 663

25.11

25.11.5

Multi-Functionalization of Biopolymer-Silica Aerogel Composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Transition Metal-Containing Aerogels . . . . . . . . . . . . . . . . . . Gold-Containing Chitosan-Silica Aerogels . . . . . . . . . . . . . Diffusion Control of Chemical Reactions in NanoDomains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Attachment of Polymers to Chitosan-Silica Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Graphene–Cellulose-Silica Aerogels . . . . . . . . . . . . . . . . . . . .

25.12

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 671

25.4

Chitosan-Silica Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 656

25.5

Alginate-Silica Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 657

25.11.1 25.11.2 25.11.3

25.6

Pullulan-Silica Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 657

25.11.4

25.7

Protein-Silica Aerogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 658

25.8

Biopolymer-Clay Hybrid Aerogels . . . . . . . . . . . . . . . . . . . . 658

25.9

Properties of the Biopolymer-Silica Aerogel Composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 659

663 664 665 666 670 671

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 672

Abstract S. Zhao (*) Building Energy Materials and Components Laboratory, EMPA, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland Department of Chemistry, Brown University, Providence, RI, USA e-mail: [email protected] W. J. Malfait Building Energy Materials and Components Laboratory, EMPA, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland e-mail: [email protected] C. J. Yao Firestone Building Products Co. LLC, Indianapolis, USA e-mail: [email protected] X. Liu Northboro Research and Development Center, Northboro, MA, USA e-mail: [email protected] M. M. Koebel Building Energy Materials and Components, Laboratory for Building Technologies, Empa, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland e-mail: [email protected]; [email protected] W. M. Risen (*) Department of Chemistry, Brown University, Providence, RI, USA e-mail: [email protected]

SiO2 aerogels are attractive materials for a wide range of applications due to their exceptional physical properties. However, despite many decades of research, the use of silica aerogels is still restricted by their poor mechanical properties and limited range in surf"ace chemistry. Because of their flexibility, biocompatibility, hydroxyl reactivity, low density, toughness, and easy formability, biopolymers aerogels have great potential, but they also have inherent limitations related to their high flammability, compressibility, and inherent hygroscopicity. Hybrid biopolymer-silica aerogels can address the limitations of pure silica and pure biopolymer aerogels. The complementarity of the respective material properties has led to a growing interest in the development of biopolymer-silica hybrids with superior properties in three main ways. First, biopolymers can alleviate the inherently fragile structure of silica aerogels and can introduce chemical multi-functionality to the silica surface. Second, the compressibility, flammability and inherent hygroscopicity of pure biopolymer aerogels can be reduced by the incorporation of inorganic components. Third, there are a large number of possible combinations of biopolymers and silica, and there is a large range of functional groups available in

© Springer Nature Switzerland AG 2023 M. A. Aegerter et al. (eds.), Springer Handbook of Aerogels, Springer Handbooks, https://doi.org/10.1007/978-3-030-27322-4_25

653

S. Zhao et al.

Aerogels · Biopolymers · Gels · Multifunctional · Composites

25.1

Introduction

Combining materials is an established method for producing new materials with properties that are superior to their pure counterparts. Pure silica aerogels are attractive candidates for thermal, catalytic and environmental purification applications [1–3], because of their exceptional physical properties, such as their low density (0.1 g/cm3), thermal conductivity (12– 15 mW
m1
K1), high porosity (>95%), and specific surface area (800–1000 m2/g) [4–6]. However, the widespread adoption of silica aerogels is hindered by their poor mechanical properties [7], concerns about dust-release [8, 9], and very limited surface chemistry. They also are brittle, this has been related directly to the “weak pearl-necklace structure” [10], but can be modified by combining the silica phase with various polymers to engineer strong and thermally superinsulating materials. The use of biopolymers for the fabrication of aerogels has been researched intensively in recent years [3, 11–13]. Pure biopolymer aerogels have the advantage of low density, high flexibility, and toughness, and they also are easy to shape by subtractive manufacturing. Certain biopolymers are also ideal materials for the mechanical reinforcement of fragile silica aerogel [14–16]. They can display a high biocompatibility and hydroxyl reactivity, and a consequent high potential to extend the application of silica aerogels to the fields of biomedicine, catalysis, and chemistry [17, 18]. Obviously, the disadvantages of neat biopolymer aerogels, such as the high deformability, flammability, and inherent hygroscopicity [19], can be addressed by hybridizing with silica. Recently, there has been a strong renewed interest in biopolymer-silica hybrid aerogels, with a significant number of studies published since 2015. Cellulose and chitosan are by far the most popular biopolymer used for the production of hybrid aerogels (Fig. 25.1). Cellulose is of particular interest because of its ubiquity in plant matter (including waste), and cellulose is also by far the most popular biopolymer for the production of neat aerogels [20]. The use of chitosan is motivated by the availability of abundant amino-groups,

Pr ot ei n Pu llu la n

Keywords

40 35 30 25 20 15 10 5 0

Pe ct in

biopolymers. Consequently, large variations in structure, surface chemistry, and chemical functionality of aerogels are possible in silica-biopolymer aerogels and this is described in detail for functionalized chitosan-silica aerogels. Thanks to their versatile performance, hybrid biopolymersilica aerogels are promising candidate materials for thermal insulation, pharmaceutical, biomedical, catalytic, and separation applications.

% of silica-biopolymer studies A lg in at e Ce llu lo se Ch ito sa n

654

Fig. 25.1 Fraction of biopolymer-silica aerogel studies in our database, grouped according to the nature of the biopolymer

which enables specific applications (e.g., CO2 sorption) or an easy entry point for post-functionalization (see Sect. 9). With respect to the silica source, both alkoxide precursors (e.g., tetraethoxysilane, TEOS) and waterglass have been applied, where the latter benefits from the higher solubility of most biopolymers in water compared to ethanol, which is the most common solvent for TEOS-based aerogels. In this chapter, we first will review the state of the art of biopolymer-silica aerogel composites, grouped according to the type of biopolymer (Table 25.1), and then compare their physical, thermal, and mechanical properties. We then highlight the further functionalization of biopolymer-silica aerogels through the addition of various metallic or polymeric species, with a particular focus on chitosan-silica aerogels.

25.2

Cellulose-Silica Aerogels

The first cellulose-silica aerogels have been prepared from regenerated cellulose either by mixing cellulose solutions in alkali/urea rich media with sodium silcate aqueous solution [24] (Fig. 25.2), or through the impregnation of a preformed coagulated cellulose gel with tetraethyl orthosilicate (TEOS) [23]. The cellulose-silica aerogel composites display increased specific surface area and Young’s modulus [24], but they have thermal conductivities above 25 mW
m1
K1. More recently, nanofibrillated cellulose was dispersed in or impregnated with a silica-based sol and hydrophobized. The resulting materials display excellent thermal properties, but only a moderate mechanical reinforcement [19, 26]. In addition, a surface modification of cellulose nanofibers was necessary to improve the compatibility at the cellulose-silica interface. Very recently, three-dimensional (3D) scaffolds based on regenerated cellulose [23, 24], nanofibrous cellulose [19, 25,

HO

O

O HO

HO

O

OH

NH2

HO

HO

OH

HO

OH

O O

O O

NH

CH3

n

HO O O

O O COOCH3 n

OH O HO

O

n

O HO O O

O O

OH

OH

OH O O

n

Maleki et al. (2018) [40] Maleki et al. (2018) [41] Maleki et al. (2019) [42] Zhao (2018) [43]

Silica

Silica

Veres et al. (2017) [39]

Cai et al. (2012) [23] Demilecamps (2015) [24] Fu et al. (2016) [25] Hayase et al. (2014) [26] Markevicius et al.(2017) [27] Nguyen et al. (2014) [28] Zhao et al. (2015) [19] Cai et al. (2018) [29] Finlay et al. (2008) [16] Mi et al. (2018) [30] Ayers et al. (2001) [31] Hu et al.(2001) [32] Risen et al. (2001) [33] Wang et al. (2002) [34] Wang et al. (2015) [35] Yao et al. (2011) [17] Zhao et al.(2013) [36] Zhao et al. (2016) [15] Ebisike et al. (2018) [37] Zhao et al. (2015) [14] Chen et al. (2013) [38]

Silica

Silica Clay

Clay + protein Silica + GO Silica

Silica

Silica

Reference Ulker et al. (2014) [21] Onbas et al. (2019) [22]

ScCO2 ScCO2 ScCO2 ScCO2

ScCO2

ScCO2 ScCO2 APD ScCO2 ScCO2 and APD FD ScCO2 – FD FD ScCO2 ScCO2 ScCO2 ScCO2 ScCO2 ScCO2 ScCO2 ScCO2 APD ScCO2 FD

Drying ScCO2

0.12–0.19 0.08–0.30 1.25–1.54 0.10–0.18

–

0.14–0.58 0.16–0.23 0.06–0.30 0.02–0.19 0.11–0.13 0.04 0.12–0.15 0.06–0.12 0.07–0.12 0.01–0.03 – 0.25 0.25–0.27 0.27 0.10–0.17 – 0.09–0.20 0.08–0.11 0.44 0.13–0.19 0.03–0.19

Density g/cm3 – –

311–798 335–920 433–531 560–617

644

356–664 750–810 11–700 525–732 571–680 – 454–726 – – 30–120 472–750 600 605–750 – 247–567 – 429–812 603–674 237 752–837 –

SBET m2/g 345–407 600–682

33–39 32–42 – 16.3–17.7

–

25–45 26–28 – 15.3–24.3 15.3–22.3 29–32 17.5–20.1 29.2–35.1 – – – – – – – – – 16.9–18.3 – 14.2–16.9 –

λ mW/(mK) – –

4.7–38.5 0.2–84.2 4.03–7.3 0.23

–

7.9–12 11.5 1.74–5.93 – – 0.01 0.48–1.95 – 0.01–8.86 0.02–30 – – – – – – – 0.44–0.90 – 0.65–8.23 0.04–114

E-mod. MPa – –

1.19–7.45 0.01–14 0.36–1.6 0.82

–

0.7–1.8 6.3 0.19–1.44 0.014–1.481 – 1.47 0.117–0.642 0.485 0.029–0.108 0.012–0.045 – – – – – – – 0.07–7.27 – 0.06–1.88 0.19–8.17

σmax MPa – –

90–95 84–97 – 80

–

85 60 – 50 – – 17–30 60 – 50 – – – – – – – 26–80 – 8–80 0.04–114

Strain % – –

Biopolymer-Silica Aerogel Nanocomposites

ScCO2 stands for Supercritical CO2 drying, APD stands for Ambient pressure drying, FD stands for Freeze drying

HO

HO

Pullulan+PVA

Gelatin Protein Silk Protein

HO HO

HO HOOC O O

Pectin

HO

HO O

NH2

HO

HO O O

Chitosan

HO

HO

O

OH m HO

Cellulose

HO

HO O O

Precursors Alginate

Table 25.1 Biopolymers used for the production of biopolymer-silica hybrid aerogels

25 655

25

656

S. Zhao et al.

a b

NA2SiO3 solution

Washing in water, exchange with ethanol

Acid solution Formation of SiO2 particles,

c Spontaneous cellulose Cellulose- gelation

Washing in acid

CO2 drying

d

SC

Wet composite

8% NaOH1% ZnO

Organicinorganic composite aerogel

4cm

Fig. 25.2 Cellulose–silica hybrid aerogels by in-situ formation of silica in cellulose gel [23]. (Reproduced with permission. Copyright 2017, Wiley)

26, 30], cellulose macro fiber [27, 28], or bacterial cellulose [29] have been proposed for the reinforcement of silica aerogels. Those biocompatible and biodegradable cellulose enhanced the mechanical properties of the pristine materials under compressive [17, 21] or tensile [20] loading, but thermal conductivity values (λ) tended to be above 25 mW/(m
K) as a result of the significantly enhanced skeletal conductivity [17, 19, 21, 22]. Chemical functionalization of the cellulose (Fig. 25.3a) simultaneously improved the low thermal conductivities and it significantly improved mechanical properties of the products [19] by (i) improving the interfacial adhesion between the cellulosic surface and the silica matrix; and (ii) reducing the inherent hygroscopicity of the cellulosic scaffold, which generally has a negative effect on the thermal conductivity of porous solids [23].

materials are widely used for immobilization of cells [44], drug delivery [45], and biodegradation of toxic compounds [46]. With respect to aerogels, however, there are only limited studies on their pectin-silica hybrids. Risen et al. [33] first reported pectic acid-silica hybrid aerogels and their loading with metal ions of Er3+, Yb3+, Sm3+, and Dy3+, as well as the applications for gas detectors, liquid and gas absorbing objects, and optical devices, which will be discussed in detail in Sect. 8. Recently, a one-step process at pH 1.5 was developed for synthesizing pectin-silica hybrid aerogels with high compressive strength, high stiffness, and minimal dust release without compromising the thermal superinsulating properties [14] (Fig. 25.3b). The resulting materials surpass current state-of-the-art composites with their unique hydrophobic, thermal, and mechanical properties.

25.3

25.4

Pectin-Silica Aerogels

Pectins consist of α(1–4)-linked galacturonic acids and are water-soluble down to pH ≈ 4, but they are insoluble in most organic solvents. Pectin by itself has been reported as a high potential candidate for aerogels with thermal conductivities close to the performance of silica aerogels [11]. Furthermore, pectin and silica are both negatively charged at neutral pH and therefore no attractive electrostatic interactions are expected to arise between the polymer and the silicon. This should favor the formation of homogeneous hybrid materials rather than precipitation. Indeed, pectin-silica hybrid

Chitosan-Silica Aerogels

The first reported synthesis of chitosan-silica composite aerogels combined an aqueous acidic chitosan solution (75– 100% de-acylated chitin, HCl/HF/acetic acid) with an alkoxide-based silica sol [31, 47]. Typically, the alkoxidebased sols are prepared in alcoholic solvents with a trace of water [48], but because chitosan coagulates very quickly in the alcohol [49], water was used as the solvent (Fig. 25.4a). Alternatively, and similar to the pectin-silica aerogel case, chitosan also has been combined with an aqueous sodium silicate-based silica sol. Recently, chitosan was directly

25

Biopolymer-Silica Aerogel Nanocomposites

657

1. Silylated cellulose bioscaffold

a

2. Reinforced silica aerogel composite

Polysiloxane layer

25 Impregnation and ScCO2

NFC substrate 1 cm

2 mm

50 mm

b

1. Silica-pectin sol

2. Reinforced silica aerogel composite

One-pot

Interpenetrating

500 nm

Fig. 25.3 (a) Multiscale assembly of superinsulating silica aerogels within silylated nanocellulosic scaffolds [19]; (b) one-pot synthesis of pectinsilica hybrid aerogels [19]. (Reproduced with permission. Copyright 2018, Wiley)

dissolved in an ion-exchanged silicic acid solution to produce a non-brittle, low thermal conductivity chitosan-silica aerogel composite (Fig. 25.4b) [15].

25.5

be used as a hybrid with alginate, where the high biocompatibility, moldability, and permeability of the silica could retain or even improve the biological activities of the immobilized substances on the alginate.

Alginate-Silica Aerogels 25.6

Alginate is a linear anionic polysaccharide with α-Lguluronate and β-D-mannuronic acid residues (shown in Table 25.1). Gel formation of alginate is most often based on ionic crosslinking with calcium cations [52]. Alginate presents good mucoadhesive properties, and alginate aerogels are often considered for pharmaceutics applications, such as drug delivery [53] or enzyme immobilization [22]. The environmental changes of temperature, pH, or buffer condition can affect calcium alginate gel and lead to a loss of drug release control and enzyme encapsulation [54]. Therefore, inorganic silica aerogel has the potential to

Pullulan-Silica Aerogels

Cellulose, pectin and chitin/chitosan are the most commonly studied polysaccharides for hybridization of silica aerogels, and alternative polysaccharides have received far less attention in aerogel research. Pullulan is a starch-derived polysaccharide based on maltotriose units connected by α-1,6 glycosidic bonds. Recently, silylated, nanofibrous pullulan scaffold was used as a template and impregnated with a silica sol to produce a pullulan-silica aerogel hybrid that combines the excellent thermal conductivity of silica aerogel with the flexibility of nanofibrous pullulan (Fig. 25.5) [43].

658

S. Zhao et al.

a H3CH2CO

Si

H3CH2CO

OCH2CH3

TEOS

OCH2CH3 Gel

X=Chitosan

b

1. NH3/H2O 2. EtOH ex/

1% (v/v) in HAc/H2O

HO Si HO

OH

X-SiO2

SCF X-SiO2 Aerogel CO2 Monolithic

X-SiO2

SCF X-SiO2 Aerogel CO2 Monolithic

Silicic acid

OH Gel

1. NH3/H2O 2. EtOH ex/

X=Chitosan

c H3CH2CO

Si

H3CH2CO

OCH2CH3

TEOS

OCH2CH3 Gel

X=Chitosan

1% (v/v) in HAc/H2O

1. NH3/H2O X-SiO – SCF X-SiO2-Au 2 Aerogel 2. EtOH ex/ Au(III) CO2 Monolithic gel

Au(III)/EtOH/H2O

d X-SiO2-Au-e Aerogel Monolithic

SCF

Au(III)/EtOH

X-SiO2 gel

CO2

e H3CH2CO H3CH2CO

Si

OCH2CH3

TEOS

OCH2CH3

2– Aniline S2O8 /HCL Chitaline +Chitosan

Gel

1. NH3/H2O X-SiO2 Au(III)/EtOH X-SiO2– SCF X-SiO2-Au Au(III) CO Aerogel 2. EtOH ex/ gel 2 Monolithic gel

Fig. 25.4 Different synthesis routes for chitosan-silica composite aerogels developed by Risen et al. [15, 17, 18, 36, 50, 51]

25.7

Protein-Silica Aerogels

Proteins are an alternative option to polysaccharides for the reinforcement of silica aerogel for insulation applications. In fact, silk fibroin (SF) extracted from silkworm cocoons has been used to reinforce silica aerogel (Fig. 25.6) [40, 41]. The composites display excellent mechanical properties, comparable to those of the polysaccharide-silica composite aerogels described above. For biomedical applications, protein-silica composite aerogels present a number of exciting opportunities. For example, the addition of gelatin improves the rate of release of ibuprofen (IBU) and ketoprofen (KET) by a factor of 10 compared to neat silica aerogels [39]. Heme protein

cytochrome C encapsulated into silica aerogels exhibits rapid gas-phase recognition of NO [55], and the fluorescent protein DsRed [56] has also been incorporated into silica aerogels with 85% retained fluorescent activity.

25.8

Biopolymer-Clay Hybrid Aerogels

Biopolymer-clay hybrid aerogels comprise a closely related class of high strength/density ratio thermal insulation materials, although they are not exclusively made of silica. The addition of biopolymers, whether protein-based, i.e., soy silk, silk latte, or cellulose-based, i.e., hemp, ramie, bamboo top,

25

Biopolymer-Silica Aerogel Nanocomposites

659

HO Si O

O Si O Si O OH O OH Si HO Si O HO O O Si Si Si O Si O O Si O O OH HO O O H H O HO Si Pu llu O O

b

a

Pu ll na ulan no /P fib VA er

10 mm

O Si O

lan /PV A

OH

100 nm

Fig. 25.5 (a) Pullulan/ Polyvinyl alcohol (PVA)-silica hybrid aerogels that are easy to shape; (b) Scanning electron microscope (SEM) image and a schematic illustration of how the silylation of the pullulan

nanofibers promotes pullulan-silica compatibility [43]. (Reproduced with permission. Copyright 2018, Elsevier)

reinforces clay aerogels [16, 38, 57], and the compressive strength and moduli of the hybrids are improved fivefold through the addition of 5 wt.% of bio-fibers, while the bulk density is increased by a factor of 50%: chitin m > 50%: chitosan Fig. 26.2 Structural formula of chitin and chitosan. (Adapted from [10], Progress in Polymer Science, 34, C.K.S. Pillai, Willi Paul, Chandra P. Sharma, Chitin and chitosan polymers: Chemistry, solubility and fiber formation, 641–678, Copyright (2009), with permission from Elsevier)

Chitin is difficult to extract without much degradation, and, as cellulose, it is insoluble in many common solvents due to intra- and intermolecular hydrogen bonding. Chitin is soluble in strong polar protic solvents, such as trichloroacetic acid and dichloroacetic acid, and highly polar solvents such as hexafluoroisopropyl alcohol, hexafluoroacetone sesquihydrate, and methane sulfonic acid [10]. As cellulose, chitin is soluble in LiCl/N,N-dimethylacetamide. Chitosan is the N-deacetylated derivative of chitin with degree of acetylation usually less than 0.35. The dissolution of chitosan is much simpler as compared to chitin. Chitosan is soluble in hydrochloric, dilute nitric, acetic, formic (found to be the best), and lactic acids; the most commonly used solvent is 1% acetic acid at pH around 4 [10]. Chitosan is used as dietary fibers, as biopesticide, adsorbent, antibacterial, and hemostatic agent. Fibers can be spun using traditional and electrospinning methods. Both chitin and chitosan can be either chemically modified or blended with other polymers to vary material final properties. For example, trimethyl chitosan is used in nonviral gene delivery. β-Glucan is homoglycan with (1 ! 4)- and (1 ! 3)-β-linked anhydroglucose units. The type of linkage and the physicochemical properties (branching, molecular weight, solubility in water, viscosity, ability to gel) depend on the source from which β-glucan is extracted. It exists in cell walls of cereals (mainly oat and barley) and in some fungi and bacteria. β-Glucan is used as soluble dietary fibers in food with the main advantage in decreasing the level of cholesterol and also for its ability to regulate glucose and insulin levels. Aqueous solutions of β-glucan can gel upon cooling. Dextran. It is a branched homoglycan consisting of mainly (1 ! 6)-α-linked glucose units. It is produced by bacteria mainly from sucrose. Dextran has high molecular weight around 106–108 g/mol with high polydispersity depending on branch density. Branches are located at O-3 position of the glucosyl units. The degree of branching of commercial dextran is about 5%. It is water soluble and often used in hydrogel form. Dextran main applications are in pharma and medicine as antigen delivery due to self-adjuvant properties, biocompatibility, and biodegradability. Dextran is a neutral polymer but is often derivatized (acetylated, cationized) to be used as drug delivery matrix. No neat dextran aerogels are reported in literature till now. Pullulan. It is another polysaccharide composed of repeating (1 ! 4)- and (1 ! 6)-α-linked AGU: three glucose units are connected by (1 ! 6) bond and three next units by (1 ! 4) bond. Pullulan is produced by yeast-like fungi from starch, and it is a water-soluble polymer. It forms flexible films with good mechanical properties and is thus used in edible films and coatings and also as food additive. No neat pullulan aerogels are known till now.

26

Polysaccharide (Non-cellulosic) Aerogels

681

b

a

CO2H O OH HO

HO2C O

b-D-Mannuronic acid

O

O OH OH

O O2C O −

OH O OH

a-L-Guluronic acid

O

O

Ca2+

26

HO HO O

O

O CO2−

Fig. 26.3 (a) Alternating mannuronic and guluronic blocks in alginate and (b) egg-box model together with example of a dimer and aggregated dimers. (Adapted from [12])

Heteroglycans (or Heteropolysaccharides) Alginate. It is a linear polyuronic acid consisting of β-Dmannuronic acid and α-L-guluronic acid in various proportions (Fig. 26.3a); in some alginates, these acids are alternating, and in some of them, they make block copolymers. Their ratio practically does not vary within species but may vary between 0.45 and 1.85 among different algae sources. α-Lguluronate regions in alginate are almost mirroring α-Dgalacturonate sequences in pectin, and thus some properties are similar, such as so-called “egg-box” gel model (Fig. 26.3b) [11]. Alginate is extracted from brown algae and can also be produced by some bacteria. It is used in films in food and pharma applications and also as adhesives and in paper coating and dental applications. Its main function is gelation and modification of viscosity. Alginate gel is used for wound healing, in tissue engineering, as coating agent to prevent dehydration, and as ion exchanger. One of the classical ways to form alginate gels is via ionic gelation with polyvalent metal ions; the most commonly used are calcium salts. Calcium ions are binding to guluronate blocks which in turn form junction zones with guluronate blocks of neighbor chains resulting in “egg-box” structure [12]. This gelation mechanism is a two-step process: formation of strongly linked dimers associating two crosslinked chains followed by weaker inter-dimer aggregates of several chains (Fig. 26.3b). If calcium salts are used alone, gelation may be too quick, and heterogeneous gels are formed. To slow down gelation, buffer salt solutions are used in which ions are screening carboxylate groups, thus competing with calcium ions. Decreasing temperature also retards gelation. The slower the gelation, the higher is network structure order, and the stronger are the gels. Gel stiffness also depends on the number and type of sequences of guluronic and mannuronic acid blocks. In physiological conditions, alginate gels are aging because of the release of binding calcium ions.

OH OH

O O O

O OH

HO

O

Fig. 26.4 Repeating units of agarose

Alginate gels can also be formed via chemical crosslinking, for example, with poly(ethylene glycol)-diamines, poly(acrylamide-co-hydrazide), adipic acid dihydrazide, etc. [13]. The mechanical properties and swelling of gels depend on the degree of crosslinking and type of crosslinker. Another way is photo-crosslinking, for example, when alginate is modified with methacrylate or polyallylamine or partially modified with α-phenoxycinnamyldiene acetylchloride. Alginate is not intrinsically thermosensitive, but it may become if performing copolymerization with N-isopropylacrylamide or with poly(ethylene glycol)-co-poly(ε-caprolactone) by UV irradiation [13]. Agar. Agar consists of 50–90% agarose and 50–10% agaropectin. Agarose, which is the main component used in applications, is a linear polymer composed of alternating (1 ! 3)-β-linked galactose and (1 ! 4)-3,6-β-linked anhydro-L-galactopyranose (Fig. 26.4). Agar is extracted from seaweeds. Agar swells in cold water and dissolves above 95  C; it forms reversible gels upon cooling. The gelation mechanism is similar to that of carrageenans, i.e., coil-helix transition, except that it does not involve metal ions, and helixes are stabilized by hydrogen bonds. Agarose gels are turbid, with thick bundles of aggregated helix-helix agarose chains and large pores. The mechanical properties of agarose gels strongly depend on the thermal history of solution treatment [14] and on the type of low or high molecular weight

682

T. Budtova

OSO3– CH2OH H HO H H

OH H

H O

H

H O

O

H

OSO3– CH2OH H HO

O H OH

H H

OH H

H H

R

O OH CH2OH –O SOH HO 3 H H –O SO 3

H O

O

O H OSO3–

H H

O OSO3–

O H

H O

H

O H H

H H

Fig. 26.5 Repeating units in κ-, ι-, and λ-carrageenans

compounds added. For example, the addition of sucrose results in more homogeneous gels with a large increase in the strain at failure [15]. Carrageenans. They consist of alternating sulfated units of (1 ! 4)-α-linked and (1 ! 3)-β-linked D-galactopyranose units and varying amounts of 3,6-anhydro-Dgalactopyranose units. Carrageenans are polyelectrolytes and are divided into three main classes according to the degree of sulfation with one, two, or three sulfate ester groups corresponding to κ-, ι-, and λ-carrageenan, respectively (Fig. 26.5). Carrageenans are extracted from red seaweeds. They are widely used as thickener, emulsifier, and stabilizer in food, pharma, and cosmetics. Carrageenans are soluble in hot water and insoluble in organic solvents, oil, and fats. Gelation of κ- and ι-carrageenans occurs upon cooling via coil-helix transition followed by helix aggregation into double-helix junctions. Several alkali metal ions induce gelation, but the most often used are potassium and calcium. For steric reasons, λ-carrageenan is not gelling with viscosity depending on temperature, ionic strength, and polymer concentration. Contrary to κ- and ι-carrageenans, ions decrease solution viscosity of λ-carrageenan due to the screening of charges on polymer chain. To modify solution viscosity and gel properties, carrageenans are often chemically modified (with hydroxyalkyl groups, alkaline treatment to increase gelling properties, grafting with acrylic acid and/or acrylamide, etc.) [16]. Pectins. Pectins are a group of heteroglycans with major component being D-galacturonic acid (1 ! 4)-α linked in the main chain (Fig. 26.6). Depending on the plant source and extraction method, carboxyl groups can be methylated (characterized by the degree of esterification, DE) or amidated

R

H

H

OH

OH

H O H

C R

C O

H

O

O

O

H H

H

OH

H

H

OH

O

Fig. 26.6 Main repeating units of pectin, based on galacturonic acid

(degree of amidation). Pectins are thus divided into highmethoxyl (HM) pectins with DE > 50% and low-methoxyl (LM) pectins. Pectins may also contain branched rhamnogalacturonans. Pectins are present mainly in the middle lamella of cell walls in higher plants, fruits, and vegetables and less in grasses. For example, on dry matter basis, citrus peel contains 20–30% of pectins, apple pomace 10–15%, grass 2–10%, and woody tissue up to 5% [17]. Pectins have fundamental roles in both primary and secondary wall structure and function: in plant development and growth, defense against microorganisms, cell-cell adhesion, wall structure, signaling, cell expansion, wall porosity, binding of ions, pollen tube growth, seed hydration, leaf abscission, and fruit development [17]. Pectin is used in food and cosmetic as a gelling, thickening, and stabilizing agent; pectin dietary fibers have anti-cholesterol and anti-tumor properties and stimulate the immune response.

26

Polysaccharide (Non-cellulosic) Aerogels

683

Pectins are soluble in water, and solution viscosity and ability to gel depend on polymer molecular weight and concentration, DE, temperature, pH, and presence of ions and/or other low molecular weight compounds such as sugars. Gelation occurs via different mechanisms of chain association. At a given pectin concentration, at room temperature, and at low pH, gelation occurs due to intermolecular hydrogen bonding between protonated groups (carboxyl, hydroxyl) and is stabilized by hydrophobic interactions of methylated groups [18–20]; it is so-called acid gelation. If keeping pH low, the increase in temperature leads to gel “melting”; sol–gel transition is reversible with strong hysteresis at lower pH. The increase of pH leads to deprotonation of acid groups which prevents chains’ aggregation and gelation. The addition of calcium is known to induce ionic gelation via “egg-box” model as described for alginate. Gelation is more pronounced when (i) lowering pectin DE [21], (ii) increasing calcium concentration [22, 23], and (iii) solution pH is close and higher than pKa which is around 3–3.5 depending on the degree of esterification. Ionic gelation leads to more firm and brittle gels than their “acid-gelled” counterparts due to the strength of the bonds [22]. In particular, it is well known that low methylated pectins undergo a strong ionic gelation in the presence of calcium due to their high sensitivity to Ca2+ ions [19, 23–26]. Moreover, the strength of pectin gel crosslinked with calcium increases with calcium concentration up to a certain value, generating a denser pectin network [22, 23, 26]. Hyaluronan or hyaluronic acid. It is a linear polymer composed of disaccharide units of N-acetyl-D-glucosamine (1 ! 4) linked to D-glucuronic acid (Fig. 26.7). It is present in all vertebrates: in connective (cartilage), epithelial, and neural tissues. In vivo, it is in the form of a polyanion and not in the protonated acid form. Hyaluronic acid participates to proliferation and migration of cells, plays an important role in mechanical wear and adsorption of loads, and acts as lubricant in muscular tissues. Because of the “repairing” functions, hyaluronan is used in cosmetics and medicine. Hyaluronic acid dissolves in water, but the dissolution may be accompanied by degradation. In order to use hyaluronic acid hydrogels as tissue culture scaffolds, it is

OH O O HO

OH O

O

HO O

OH

NH O

Fig. 26.7 Repeating units of hyaluronan

usually stabilized by either chemical modifications or crosslinking. The latter is performed using carbodiimides, aldehydes, sulfides, and polyfunctional epoxides [27]. When chemically modified, hyaluronan esters can form membranes and fibers or processed by spray-drying to produce particles. To make dry porous materials, hyaluronan is freeze-dried to obtain sponges or foamed (supercritical CO2 can be used for this purpose), or porogens can be added [27]. To vary and reinforce hyaluronan scaffold, it is often mixed with other polymers. Till now, there is no literature reporting on hyaluronan aerogels. Xanthan is composed of (1 ! 4)-β-linked pentasaccharide units in backbone with trisaccharide side chains composed of mannose (β-1,4) glucuronic acid (β-1,2) mannose. It is obtained by aerobic fermentation of Xanthomonas campestris cultures. When D-glucuronic acid is deprotonated above pH 4.5, xanthan becomes negatively charged polyelectrolyte and can be ionically crosslinked with metal ions such as calcium. Xanthan can also form gels by reacting with citric acid, which is a nontoxic crosslinker for polysaccharides. Xanthan gum is used as emulsifier, lubricant, and thickener in food and as strong shear thinning fluids in oil drilling. Hemicelluloses. The main groups of hemicelluloses are xyloglycans (xylans), mammoglycans (mannans), xyloglucans, and mixed-linkage β-glucans [28]. They are non-crystalline polysaccharides and differ in structure and organization. Hemicelluloses are components in the cell walls of higher plants, often associated with proteins, cellulose, and lignin. Depending of the degree of purification, hemicelluloses can contain phenolic substances, bound proteins, and other polysaccharides. The solubility of hemicelluloses depends on the purity of the extracted polymer and its molecular weight, degree of branching, chain aggregation, etc. For example, several hemicelluloses are water dispersible, and some xylans are soluble in dimethyl sulfoxide-water mixtures [29]. Separation and characterization of hemicelluloses are complicated because of their huge diversity and complex structures. Overall, hemicelluloses are of low molecular weight resulting in materials with rather weak mechanical properties. However, as many polysaccharides, hemicelluloses can be derivatized which may diversify and modify their properties. Xylans are the largest family of hemicelluloses present in wood, terrestrial plants, and algae. Depending on the type of plant and mode of extraction, xylans can be weak polyelectrolytes and be acetylated or feruloylated, the latter being a promising component for wound treatment. Xylans are less thermally stable than cellulose with thermal decomposition starting around 150  C. Some xylan-type hemicelluloses possess tensioactive properties, for example, those that contain phenolic compounds that remain from lignin [29]. Xylans can stabilize protein foams (bovine serum albumin) against thermal disruption due to film-forming properties or form

26

684

T. Budtova

surface active complexes with surfactants [29]. Xylans have various biological activities: they can contribute to the action of dietary fibers such as detoxification ability, be antinutritive, inhibit the growth rate of tumors, and be used as components of drug delivery systems.

26.1.2 Aerogels Based on Polysaccharides or Bio-aerogels: Materials Developed in the Twenty-First Century The very first aerogels based on polysaccharides were synthesized by Kistler; he reported the possibility of making aerogels from several bio-based polymers such as cellulose, nitrocellulose, agar, gelatin, and egg albumin [30]. However, their properties were not studied. The increased interest to bio-based and renewable materials, in particular to polysaccharides, raised at the beginning of the twenty-first century. However, not always the term “aerogel” was used for mesoporous samples dried with supercritical CO2: for example, in the first publication on starch aerogels in 1995, they were called “microcellular foams”

a Number of publications per year

Fig. 26.8 (a) Number of publications on “aerogels +polysaccharide” from Web of Science, March 2019, with “polysaccharides” being chitosan, starch, cellulose I and cellulose II, alginate, and pectin, and (b) distribution percentage (%) by the type of the same polysaccharides in bio-aerogels. (Courtesy of Sophie Groult [33])

300 200 150 50 0

292

Chitosan Starch Cellulose II Cellulose I (Nanocellulose) Carrageenan Alginate Pectin

250

100

[31]. Cellulose aerogels obtained in 1993 from viscose were simply called “porous cellulose” [32]. In our days, the term “aerogel” is sometimes overused as far as porous, not necessarily mesoporous materials, are called “aerogels.” This often occurs when a polysaccharide gel is freeze-dried. The number of publications on polysaccharide-based aerogels per year and per type of polymer is presented in Fig. 26.8a, b. It shows a continuous strong increase starting from the year 2010. The majority of articles are devoted to cellulose aerogels because they include both cellulose I (nanoand microfibrillated cellulose and cellulose nanocrystals) which is a “hot” topic in cellulose area and cellulose II-based porous materials. Next in number are chitosan and alginate aerogels, followed by those based on starch and pectin. In the following sections, synthesis, structure, and properties of bio-aerogels, except cellulose-based, will be presented and discussed. Only native polysaccharides will be considered. The focus will be made on mainly mesoporous materials obtained using drying with supercritical CO2. Unless specifically mentioned, freeze-dried polysaccharides will not be analyzed.

68 26

16 100 m2/g). This chapter discusses the high compressive strength and superinsulation properties of CNM aerogels, which paves the way for future high-value applications. Possibilities in functionalization of CNMs for their controlled interaction with other nano-sized particles and/or biopolymers have significantly broadened the applications of CNM aerogels spanning from bio-based adsorbents, biomedical scaffolds, and insulation materials to carbon aerogels, energy-storage devices, or inorganic templates. This chapter illustrates a few of these promising application areas and highlights remaining challenges to address for advancing commercialization of CNM-based aerogels. Keywords

Nanocellulose · Cellulose nanomaterials · Aerogels · Mesoporous · Insulation · Absorbent · Tissue engineering · Supercritical drying · Sustainability

27.1

Introduction

Cellulose is the most abundant renewable polymer derived from biomass with a worldwide production estimated to be over 1011 tons each year [6]. Cellulose is a linear homopolysaccharide of β-1.4-linked anhydro-D-glucose units with a degree of polymerization of approximately 10,000 for cellulose chains in nature and 15,000 for native cellulose cotton [6, 72]. There are four polymorphs of cellulose, namely,

© Springer Nature Switzerland AG 2023 M. A. Aegerter et al. (eds.), Springer Handbook of Aerogels, Springer Handbooks, https://doi.org/10.1007/978-3-030-27322-4_27

707

708

cellulose I, II, III, and IV. Cellulose I is the native form of cellulose found in nature and occurs in two allomorphs, Iα and Iβ [93], while cellulose II, III, and IV can be obtained after chemical treatment of the polymorphs [53]. In this chapter, aerogels produced from cellulose I are discussed. Research and development on cellulose-based materials has reached an important milestone with the emergence of cellulose nanomaterials (CNMs). This new family of materials has properties and functionalities distinct from molecular cellulose and wood pulp, paving the way for the development of applications that were once thought impossible for cellulose materials [28, 33]. CNMs are commonly referred to as “nanocellulose.” This umbrella term describes different categories of nano- and micro-sized cellulose particles but preferably alludes to CNMs with at least one dimension in the size range 1–100 nm. While no standard has yet been agreed on by the community, the International Nanocellulose Standards Coordination Committee (INSCC) established in 2011 and housed by the Technical Association for the Pulp and Paper Industry (TAPPI) is actively working on a road map for the development of international standards for nanocellulose. In this chapter, we will refer to the terminology and classification of cellulose nanomaterials as defined by TAPPI (Standard Terms and Their Definition for Cellulose Nanomaterial WI 3012) and shown on Fig. 27.1. The terminology and classification proposed by TAPPI is valid regardless of the cellulose source, extraction/production method, and surface chemistry of the CNMs. These factors are nevertheless of importance for an accurate characterization and comparison of CNMs and related products [33]. CNMs can be produced from various sources: wood [55], plants (e.g., cotton, hemp, sisal) [63], industrial and crop wastes (e.g., rice straw, sunflower shells, sugarcane bagasse) [36], and living species (e.g., bacteria, algae, tunicate) [65]. While bacterial cellulose microfibrils (CMFs) can be directly synthesized as hydrogel by Gluconacetobacter bacterial strains (such as G. xylinus), other cellulose micro- and nanofibrils (CMFs/CNFs) are commonly produced in water by high-shear mechanical treatment of the cellulose source using, e.g., microfluidizer, grinder, or homogenizer as conventional technique [91]. Because the solely mechanical disintegration of cellulose fibers to CMFs/CNFs is highly demanding in energy and time [123], enzymatic [46] or chemical [56, 98, 121] pretreatments are usually performed prior to the mechanical defibrillation. Chemical posttreatments are often applied after mechanical disintegration for surface modification purposes. As such, CMFs/CNFs possess a wide variety of surface chemistries and charge densities. Unlike bacterial CMFs that are made of pure cellulose only, CMFs/CNFs may also contain hemicelluloses [57] and residual lignin [103]. From one unique cellulose source, over 50 grades of CMFs/CNFs can be produced,

N. Lavoine

when combining the different pre-, mechanical, and posttreatments available today [91]. In contrast, cellulose nanocrystals (CNCs) are rodlike nanoparticles of much higher crystallinity [28], obtained by acid hydrolysis of the cellulose source under strictly controlled conditions of temperature, agitation, and time [8, 11, 26]. The nature of the acid plays a major role in the CNC preparation and properties. Sulfuric acid is the most extensively used acid for producing CNC with surface sulfate charges [42], but hydrolyses with hydrochloric [140], phosphoric [15], hydrobromic [106], and phosphotungstic [79] acids have also been reported. The use of other acids, however, may influence the aqueous dispersion stability of the CNC suspension (e.g., flocculation occurred with hydrochloric and hydrobromic acids) and the charge density [42], resulting in different CNC aspect ratios and properties. The geometrical dimensions of CNCs (viz., the width/diameter, D, and the length, L, resulting in the aspect ratio L/D; Fig. 27.1) also vary with the cellulosic source. While CMFs/CNFs form physically entangled networks, even at very low concentration (e.g., critical gelation concentration of 0.8 wt% reported for mechanically treated CNFs [19]), CNCs self-assemble into chiral nematic liquid crystals above a critical concentration of ca. 4.5 wt% [23, 102] and form a gel on their own from 10 to 14 wt% [23, 50]. Figure 27.2 summarizes the main cellulose sources and production routes of CNMs, but for additional information, the reader is directed to the detailed reviews and book chapters published on the production and optimization strategies of CNMs, their characterization, main properties, and potential applications [23, 28, 29, 33, 36, 47, 64, 72, 75, 84, 88, 91, 104]. Compared with synthetic nanomaterials, CNMs address significant challenges in terms of sustainability, renewability, biocompatibility, and cost efficiency. They have been successfully used in a wide range of applications spanning from composites [114, 119, 120], packaging [75], coatings [12], and tissue engineering [29] to printed electronics [47], energy-storage devices [94], and [in]organic templating [124]. Characteristic properties of CNMs such as low density, high aspect ratio and specific surface area, gel-formation ability, and good strength promoted research and development on CNM foams and aerogels. These lightweight and porous materials are the current target of several projects from process optimization and property improvement to high-performance tailored materials for, e.g., energy-efficient building, water purification, scaffolds engineering, or food emulsion [23, 71, 81]. Originally, cellulose has been investigated for the production of cellulose aerogels, or the so-called aerocellulose, in response to the increasing demands in biocompatible and biodegradable materials [38]. In the preparation process of aerocellulose, however, the cellulose gel is commonly formed by dissolution and regeneration of

27

Nanocellulose Aerogels

709

Cellulose nanomaterials (CNMs)

Nano-objects

Nano-structured

27 Cellulose nanofiber

Cellulose nanocrystal (CNC) width: 3–10 nm L/D >5 a

Cellulose nanofibril (CNF) width: 5 –30 nm L/D >50

Cellulose microcrystal (CMC) width: 10 –15 nm L/D < 2

Cellulose microfibril (CMF) width: 10 –100 nm length: 0.5 –50 μm

c

d

b 50 μm

500 nm

5 μm

500 nm

Fig. 27.1 Cellulose nanomaterials: hierarchy, terminology, and definition. Adapted from the TAPPI Standard Terms and Their Definition for Cellulose Nanomaterial WI 3012 (2011). (a) Transmission electron micrograph (TEM) of ramie cellulose nanocrystals (CNCs). (Reprinted and adapted with permission from [43]); (b) TEM of TEMPO-oxidized cellulose nanofibrils (CNFs) from hardwood bleached Kraft pulp

[110]. (Copyright (2018) American Chemical Society); (c) scanning electron micrograph of cellulose microcrystals (CMCs) from Norway spruce. (Reproduced from [11] with permission from Springer Science +Business Media); (d) SEM of sisal cellulose microfibrils (CMFs). (Reprinted and adapted with permission from [118])

cellulose in an aqueous or organic solvent [117]. These aerogels may lack mechanical stability since they have lost their favorable cellulose I crystal structure or because of the reduced aspect ratio of the fibrils. Aerogels based on CNMs may thus offer advantages from an environmental standpoint, since CNMs are obtained from renewable resources and no harmful solvents are required during processing [116]. CNM aerogels may also cover broader property range due to the preservation of the full cellulose I structure and CNM high aspect ratios, in addition to more functionalization opportunities [96, 116], as it will be illustrated in this chapter. When it comes to CNM foams and aerogels, the terminology has been used interchangeably to describe any porous and lightweight materials. Nanocellulose aerogel was reported as “a highly porous solid of ultralow density and with nanometric pore sizes formed by replacement of liquid in a gel with gas,” during which “there should be no or

limited shrinkage [. . .] and the volume of the solid phase should be only of few percent of the total volume (0.2–20%)” [115, 116]. Nanocellulose foams were referred to as “solid porous materials with micrometric pore sizes which can be obtained by ice templating, i.e., after freezing of suspensions and sublimation of the formed ice crystals or by simple ambient drying from the suspensions” (in opposition to supercritical drying) [85]. Conventionally, as reported in ▶ Chap. 1, the term “aerogel” has been used to designate gels dried under supercritical conditions, in opposition to wet gels dried by evaporation, namely, xerogels [27, 66]. As a result, the dry samples keep the very unusual porous texture which they had in the wet stage [99]. Unlike foams, an aerogel is (almost) always derived from a wet gel (via a sol–gel process) and has pore size in the range 2–50 nm. More recently, however, materials dried by other techniques such as freeze-drying (see

710

N. Lavoine

Cellulose sources (a) + Purification and Homogenization (b)

Wood

Cotton

Wheat straw

Miscanthus

Jute

Sisal

Sugar beet

Bamboo

Acid hydrolysis (c) Crystalline domain

Banana rachis

Algae

Bacteria

Tunicate

Enzyme hydrolysis (d)

Disordered regions

Hydrolysis

glucose

Strong acid

crystalline

disordered regions

crystalline

cellulose reducing end groups cellulose non-reducing end groups

TEMPO-mediated oxidation (e) CH2OH

1/2 NaClO

cellobiohydrolase I

b-glucosidase

cellobiohydrolase II

OH

OH O

endoglucanase

Periodate-chlorite oxidation (f)

N

O HO

OH

O OH

O

NalO4 n

O

O

OH

OH O

O

2 NaClO2

O

O

n

O

O O− n + O−Na+ Na

+ N Nacl

NaBr

CHO

O

OH

O

O

NaBrO

NaClO NaBrO

N

COOH

OH O

O HO

COONa

NaOH at pH 10

OH

OR

OH

OH NaClO

OH

OH

Cl

O n

OH

alcoholic NaOH

HCl to pH 1

+ NaCl

RO O

R = H or •

OH

Homogenizer (h)

O

O

O OH

Microfluidizer (i)

OR n O− Na+

Grinder (j)

Inlet suspension

Mechanical treatment

Dialysis + Centrifugation

Pre-chemical modification

CNC

Impact ring Valve seat

Inlet suspension

Valve

Static stone Inlet suspension

Outlet suspension

Outlet suspension

Outlet suspension

Rotative stone

k1

k3

k2

k5

1 µm

0.5 µm

100 nm

k4

200 nm

k7

Post-treatments

300 nm

k6

200 nm

Cellulose nanomaterials (k) Fig. 27.2 (continued)

cellobiose

cello-oligosaccharides

k8

200 nm

200 nm

27

Nanocellulose Aerogels

▶ Chap. 5) have also been named aerogels, although “cryogels” are the usual term assigned to this type of materials [27]. Such an extension can nevertheless be argued, when considering, for example, the porosity (>90%), the specific surface area (SSA >100 m2/g), and the pore size (90%, but we will consider a wider pore size range, in the sub-hundred nanometer length scale (Fig. 27.3). We will also extend the focus to CNM solid foams of high SSA (>100 m2/g) and larger pore size (in the micrometer range), as novel processing routes have been investigated and represent, today, an interesting alternative for upscaling and/or further functionalization (Fig. 27.3, [71]).

27.2

Processing

Processing of CNM aerogels involves two main steps, (i) preparation of a hydro- or alcogel from a CNM dispersion/suspension and (ii) removal of the solvent by freezedrying or supercritical drying (Fig. 27.3). The preparation of a wet foam (e.g., Pickering foams) followed by oven drying is another possible pathway for the production of CNM porous solids, but commonly foams of pore size >50 μm and SSA 5000 m3

Blanket, thin films, powder, paste, and coating

–

1000 m3

Producer

Panel, blanket, paper, powder/ granulate, and products for the military

SCD

–

–

Producer

Blanket, panel, and granulate

–

–

2002

Research and production

Producer

Powder, composite board, and blanket

–

–

2013

Research and production

Producer

Granulate, powder, blanket, and coating

–

–

Producer or integrator Producer

Year founded 2015

Stage of scale and notes Research and production

Suzhou, Jiangsu (headquarters) Nanchang, Jiangxi (R&D and pilot plant)

2014

Producer

Xianyang, Shaanxic

2011

Research and production, affiliated with Suzhou Nano Research Institute, future unclear [17] Full production

Tianjind

2006

Jinzhong, Shanxic

Beijingd

Location Shenzhen, Guangdong (office) Xintao, Hubei (manufacturing)

Main products (composition is silica based unless specified) Powder, blanket, and paste

(continued)

64

The Aerogel Industry

1591

Table 64.2 (continued) Company name (alternate names)/ company name in Mandarin (transliteration) and website Zhejiang Surnano Aerogel Co., Ltd. [22] (Surnano)/浙江绍兴圣 诺节能技术有限公司 (Zhejiang Shaoxing Shengnuo Jieneng Jishu Youxian Gongsi) www.surnano.com Zhejiang UGOO Technology Co., Ltd. [23]/浙江岩谷科技有限 公司 (Zhejiang Yangu Keji Youxian Gongsi) www.ugootec.com CAS Advanced Materials (Nantong) Co., Ltd. [24] (Zhongke High-Tech Materials) – see also REM Tech Co., Ltd., South Korea/中科高新 材料(南通)有限责任公 司 (Zhong Ke Gaoxin Cailiao [Nantong] Youxian Zeren Gongsi) www.zhongke-tech.cn Ningbo Yushi New Material Co., Ltd.* (AeroClay China)/宁波 羽石新材料科技有限公 司 (Ningbo Yushi Xin Cailiao Keji Youxian Gongsi) www.aeroclay.cn Botianzirui New Technology Co., Ltd. [25]/北京博天子睿科技 有限公司 (Beijing Botianzirui Keji Youxian Gongsi) Xuntian Environmental Technology Co., Ltd. [26]/常州循天能源环境 科技有限公司 (Changzhou Xuntian Nengyuan Huanjing Keji Youxian Gongsi) Bellcoming New Material Co., Ltd. [27]/ 浙江贝来新材料有限公 司 (Zhejiang Beilai Xin Cailiao Youxian Gongsi) Canew Tech Co., Ltd. [28] – see also Bronxculture Pte., Ltd., Singapore, and affiliate

64

Location Shaoxing, Zhejiang

Year founded 2010 or earlier

Stage of scale and notes Production

Producer or integrator Producer

Yiwu, Zhejiang

2017

–

Producer

Nantong, Jiangsu (headquarters), and Daejeon, South Korea (manufacturing/ R&D)

2016

Research and production, joint venture with stateowned company

Producer

Ningbo, Zhejiang

2018

Research and production

Beijing

2009

Changzhou, Jiangsu

Main products (composition is silica based unless specified) Granulate, powder, blanket, panel, and also design and construction of aerogel production lines Blanket and panel

Drying method Continuous SCD

Reported capacity/ annual production volume –

–

4 million m2 (announced)

Powder, granulate, paste, blankets, and coatings

APD

~400 m3 as of 2016

Producer

Polymer/clay aerogel granules, films, and panels

“

1–5 metric tons

Production, current status unclear

Producer

Blanket and panel

SCD

–

2011

Research and production, current status unclear

Producer

Blanket

–

–

Zhuji, Zhejiang

2014

Production, current status unclear

Producer

Blanket and shaped parts

–

3000 m3

Shenzhen, Guangdong

2016

Research and possibly earlystage production

Primarily distributor and

–

–

–

(continued)

1592

R. A. Collins et al.

Table 64.2 (continued) Company name (alternate names)/ company name in Mandarin (transliteration) and website Bronxsinga Technology (Shenzhen) Co., Ltd./加 新科技(深圳)有限公司 (Jiaxin Keji [Shenzhen] Youxian Gongsi) Chengxiangfengkai New Materials Technology Co., Ltd. [29]/天津晨祥丰凯新材 料有限公司 (Tianjin Chenxiangfengkai Xincailiao Youxian Gongsi) Nano-Energy Solution Technology Co., Ltd. [30]/深圳市纳能科技有 限公司 (Shenzhen Shi Na Neng Keji Youxian Gongsi) www.naneng.net BCEG Advanced Construction Materials Co., Ltd. [31]/北京建工 新型建材科技股份有限 公司 (Beijing Jiangong Xinxing Jiancai Keji Gufen Youxian Gongsi) www.bceg.com.cn Guojia New Material Co., Ltd. [32]/ 珠海国佳新材股份有限 公司 (Zhuhai Guojia Xin Cai Gufen Youxian Gongsi) www.chinagel.com.cn Anhua New Materials Industrial Research Institute Center [29]/安 华新材料产业研究院 (Anhua Xincailiao Chanye Yanjiuyuan) www.anhuachun.com/

Location

Year founded

Stage of scale and notes

Producer or integrator

Main products (composition is silica based unless specified)

Drying method

Reported capacity/ annual production volume

possibly producer

Tianjin

2016

Production

Producer

Carbon aerogel

APD

100 metric tons

Shenzhen, Guangdong

–

–

Integrator

Blanket

–

–

Beijing

–

State-owned enterprise

Integrator

Blanket

–

–

Zhuhai, Guangdong

1999

Research

–

–

Zhenjiang, Jiangsu

2019

Research and production

Integrator

Aerogel powder-based fire extinguishers

–

APD

–

a

Information is directly provided by producer via questionnaire, all others via publicly available information SCD ¼ supercritical drying, APD ¼ ambient-pressure drying c Note that Shaanxi province should not be confused with neighboring Shanxi province d Note that Beijing, Shanghai, and Tianjin are municipalities and do not have an associated province b

been commercially available for many years but have only recently begun to see significant market adoption thanks to recent initiatives from leading manufacturers aiming to diversify into energy storage applications. Additionally, organic polymer aerogels are forecast to progress from pilot-plant explorations to full production between 2019 and 2029 [10]. Polymer aerogel materials being explored

in industry include aerogels based on polyureas, polyurethanes, polyisocyanurates, polyamides, and polyimides, typically taking the form of panels or thin films. The key differentiator between polymer aerogels and inorganic oxide aerogels such as silica is the mechanical robustness that organic polymer networks provide, enabling durable, dust-free monolithic forms to be produced. Material

64

The Aerogel Industry

1593

Fig. 64.4 Apparent capacity share of aerogel manufacturers in China based on company-reported values, normalized to percent of estimated total m3/year

64

2.5% 0.8% 0.3% 4.1% 4.1% 32.7%

8.2%

Nano Tech Co., Ltd. Yingde Alison Co., Ltd. Hebei Jinna Technology Co., Ltd. UniNano Advanced Materials Co., Ltd. Guizhou Aerospace Wujiang IBIH Advanced Materials Co., Ltd. Shenzen Aerogel Technology Co., Ltd. Bellcoming New Material Co., Ltd. Suzhou Tongxuan New Materials Co., Ltd. CAS Advanced Materials Co., Ltd.

8.2%

16.3% 22.9%

properties across these different compositions are diverse; however, such materials still provide many of the same advantages characteristic of inorganic mesoporous opencell materials such as low thermal conductivity and low density. Large-volume end uses for these materials have yet to be established, but ventures in defense, transportation, apparel, communications, and building and construction all show significant promise.

64.4.1 Impact of COVID-19 It should be noted that projections presented in this chapter reflect the state of the aerogel industry as it stood at the beginning of 2020 just prior to the COVID-19 pandemic. It remains to be seen what impact this event will have on the global aerogel industry in terms of both the supply side and the demand side. Many aerogel producers continued operating during the pandemic owing to being deemed essential businesses, as the aerogel industry services pipeline operators, chemical manufacturers, and defense industrial base. The pandemic did, however, expose underappreciated interdependencies in supply chains that underlie the aerogel industry. In the USA, for example, industrial carbon dioxide of the type used for supercritical drying of aerogels (among other things) was in short supply during the pandemic [37], as it is primarily produced as a by-product of fermenting corn sugar into ethanol for gasoline blending. Concurrent with rapidly changing oil prices and decreased demand for gasoline (down 30% at one point [38]), such CO2-producing ethanol plants were not profitable to operate. Additionally, a number of chemicals used in commercial manufacturing of aerogels were harder to get or more expensive than prior to the pandemic, as shipments of both reagents and their starting materials were slowed in part by reduced transportation route

frequency caused by decreased demand for airline travel, which squeezed supply chains and delayed purchase orders [39], and in part by labor shortages at docks and ports. In short, the economic effects of the pandemic highlighted how many aerogel products involve supply chains that are still in a nascent stage of development and that the price points the aerogel industry relies only exist because of demand for related products and services in even larger markets. The pandemic had a large impact on the Chinese aerogel market as well, where producers reported declines in sales compared to the same period in 2019. This was in large part because of decreased demand for industrial insulation due to a China-wide suspension of construction activity. Other markets such as barrier materials for Li-ion battery packs used in electric vehicles were also impacted, although to a lesser extent. Companies in the USA and Europe likewise anticipated decreased demand in the quarters to come for similar reasons. Whether or not such demand-side events in China and other parts of the world play a role in shaping the industry’s long-term future, however, remains to be seen.

64.5

Aerogel Producers

In this section, we present profiles of the world’s commercial producers of aerogels and aerogel-like materials as of 2020. Most profiles in this section are based on information received from a questionnaire created by the authors that was distributed to known aerogel manufacturers. To supplement some profiles or in instances where a questionnaire response was not received, a profile was created from secondary sources that the company has made publicly available. Where possible, statements in these profiles are kept in the voice of the manufacturer. Only the six largest Chinese producers of silica aerogel materials are profiled below, as

1594

little information is publicly available about China’s other silica aerogel producers (see Table 64.2 for a full list of known aerogel producers in China). It is important to note that, due to the proprietary nature of commercial activity, certain information about some companies could not be included. Additionally, please note that what follows should be considered accurate only as of the date of writing and that statements about companies and technical specifications regarding their products presented in this section are provided for reference only with no guarantee as to their accuracy or correctness.

64.5.1 Aspen Aerogels, Inc.

R. A. Collins et al.

production applications. The company offers multiple products under the brand names Pyrogel ® (used for its hightemperature products), Cryogel ® (used for its cryogenic insulation products), and Spaceloft ® (used for its near-ambient temperature and subsea products); each of their product variants and the typical target markets for them are outlined below. The company’s products contain various additives for enhanced fire reaction performance, infrared opacification, and corrosion protection, which render them optically opaque. The company’s blankets are also hydrophobic, typically exhibiting liquid or water vapor uptake values of 5 wt% or less following short-term exposure. A summary of the company’s aerogel blanket product offerings follows. Representative examples of Aspen Aerogels products are shown in Fig. 64.5:

Northborough, MA, USA | www.aerogel.com Aspen Aerogels reports to be the world’s largest manufacturer of aerogels as of 2019. The company was formed in 2001 and produces a range of flexible hydrophobic fiberreinforced silica aerogel composite blankets – a form factor first invented by the company’s progenitor Aspen Systems which, along with other numerous aerogel-related inventions, is protected by an extensive international IP portfolio. The company is at full-production scale. Aspen Aerogels blankets are used for thermal insulation applications ranging from cryogenic temperatures to 650  C. In addition, their composite configuration provides corrosion protection, passive pool fire and jet fire protection, acoustic insulation, and mechanical damping functions. The company states that its customers typically enjoy significantly faster installation rates relative to other forms of commercially available insulation products in the same application. Aspen Aerogels is primarily a producer of aerogel products, although the company also fabricates packaged insulation systems for direct application in subsea oil and gas

• Pyrogel XTE (20 mW/m-K at 25  C) is Aspen’s bestselling product, reinforced with a glass-fiber batting with an upper-use temperature of 650  C. Pyrogel XTE was initially designed for use in refineries and petrochemical facilities but has proven to have wide applicability throughout the energy infrastructure sector. Applications include corrosion-under-insulation (CUI) defense, medium- to high-temperature processes, pipes, vessels, district energy steam networks, aerospace, and defense systems. • Pyrogel HPS is optimized for applications in the power generation market that involve continuous operating temperatures greater than 400  C with an upper-use temperature of 650  C. Optimized for high-temperature hydrocarbon and chemical processing, high-pressure steam pipes, vessels, and gas and steam turbines. • Pyrogel XTF (20 mW/m-K at 25  C) provides thermal performance similar to Pyrogel XTE but is specially formulated to provide strong protection against fire. Applications include passive pool and jet fire protection and relief system sizing.

Fig. 64.5 Selection of fiber-reinforced silica aerogel composite blanket products from Aspen Aerogels [40]. (Image credit and copyright to Aspen Aerogels, Inc.)

64

The Aerogel Industry

1595

• Cryogel Z (15 mW/m-K at 25  C) is designed for sub-ambient and cryogenic applications in the energy infrastructure market including cryogenic pipelines, vessels, and equipment, as well as natural gas liquefaction and regasification facilities. Cryogel Z is reinforced with a batting comprising glass and polyester fibers and is produced with an integral outer vapor barrier. • Cryogel x201 (15 mW/m-K at 25  C) is similar in composition to Cryogel Z but produced without the integral vapor barrier. The product designed for use in cold systems where space is at a premium, including refrigerated applications, cold storage, and aerospace. • Spaceloft Subsea (14 mW/m-K at 25  C) is reinforced with a batting comprising glass and polyester fibers and is designed for use in pipe-in-pipe applications in offshore oil production to provide superior flow assurance. • Spaceloft Grey (15 mW/m-K at 10  C) is reinforced with a batting comprising glass and polyester fibers and is designed for use in the building material market. Applications include space-saving insulation for walls, floors, and roofs in residential and commercial buildings and miscellaneous OEM applications. Spaceloft Grey carries the European CE mark. • Spaceloft A2 is reinforced with a glass-fiber batting and is specifically designed to meet Euroclass A2 fire standards for the building material market. Spaceloft A2 is marketed and sold exclusively by Aspen Aerogels partner BASF under the trade name Slentex ® [41].

Aspen produces its products by integrating lofty fibrous battings into liquid-phase alkoxide-derived silica-based sols that contain suspended performance additives. Gelation of the sol is initiated inside the free volume of the fibrous batting. Finished products are obtained subsequent to further mechanical handling, chemical treatment, and removal of solvents by a proprietary supercritical CO2 extraction process. Table 64.3 summarizes product specifications and material properties for the company’s aerogel blanket products. As of late 2019, Aspen Aerogels states it was manufacturing approximately 5.1 million m2 (55 million ft2, ~50,000 m3) of aerogel blankets per year. The company has over 145 issued patents and over 96 patent applications pending with broad international coverage. Aspen Aerogels expects a 20% annual growth for its aerogel products over the next decade. Aspen Aerogels focuses its commercial efforts on the energy infrastructure market, where it says its products find utility in multiple high-value applications. Aspen also markets and sells aerogel products for the building and construction sector and numerous other end markets. Customers in these markets use the company’s products for applications as diverse as wall systems, military and commercial aircraft, trains, buses, appliances, apparel, footwear, and outdoor gear. Aspen sells its products globally, with about 60–70% of its business typically outside of the USA annually.

Table 64.3 Product specifications and materials properties for Aspen Aerogels products [40] Cryogel x201 200  C

Spaceloft 125  C

Spaceloft Subsea 200  C

Pyrogel HPS 650  C

Pyrogel XTE 650  C

Pyrogel XTF 650  C

Test ASTM C447 14 mW/m-K 14 mW/m-K – – – – – ASTM C177 17 mW/m-K 17 mW/m-K 16.5 mW/m-K 14.5 mW/m-K 21 mW/m-K 21 mW/m-K 21 mW/m-K ASTM C177 19 mW/m-K 19 mW/m-K – 18.5 mW/m-K 28 mW/m-K 28 mW/m-K 28 mW/m-K ASTM C177 – – – – 69 mW/m-K 89 mW/m-K 89 mW/m-K ASTM C177 90% 0.12–0.15 g/cc

~20 nm >90% 0.12–0.15 g/cc

~20 nm >90% 0.12–0.15 g/cc

Hydrophobic

Hydrophobic

Hydrophobic

600–800 m2/g 12 mW/m-K at 25  C

600–800 m2/g 12 mW/m-K at 25  C

600–800 m2/g 12 mW/m-K at 25  C

Table 64.7 Materials properties of Cabot Aerogel MT1100 and MT1200 silica aerogel particles

Description Target application Particle size range Pore diameter Porosity Bulk density Surface chemistry Oil absorption capacity

Cabot Aerogel MT1100 Fine particle Matting agent 2–40 μm ~20 nm >90% 0.025–0.050 g/cc Hydrophobic 540–650 g DBP/100 g particle

Cabot Aerogel MT1200 Fine particle Matting agent with enhanced solvent resistance 2–40 μm ~20 nm >90% 0.025–0.050 g/cc Hydrophobic with added solvent resistant 540–650 g DBP/100 g particle

fibers comprising an outer sheath that flows when heated, that then glues the aerogel particles together into a flexible composite blanket. Because of its polymer component, Thermal Wrap is typically limited to ambient-temperature-range applications of 200 to 125  C and has a slightly higher thermal conductivity than blankets prepared by sol infiltration. For comparison, 8.0-mm Thermal Wrap provides about the same thermal insulating performance as 5.0-mm Spaceloft at room temperature. This said, Thermal Wrap is a popular alternative to other aerogel blankets when particle shedding is not well tolerated and where thickness is not as constrained. Additionally, Thermal Wrap is easy to cut and handle, is hydrophobic, and provides lighttransmissive benefits. Thermal Wrap is used in a variety of applications ranging from industrial insulation to structural fabric roofing to water bottles. Thermal Wrap is produced in thicknesses of 3.5, 6.0, and 8.0 mm, with product numbers TW350, TW600, and

TW800, respectively. Table 64.9 summarizes properties of Cabot’s Thermal Wrap aerogel blankets [50].

In addition to the above products, Cabot’s Compression Pack (Fig. 64.7), produced in conjunction with Johns Manville, is a conformable insulation package loaded with aerogel particles designed for pipe-in-pipe infrastructure that expands to fill the spaces between pipes. Cabot’s Compression Pack provides superior thermal performance over even fiber-reinforced silica aerogel blankets for applications where space is at a high premium. Major applications for Cabot’s aerogel products include the following: • Interior and exterior insulating plasters for breathable building envelopes and façades • Insulation boards for internal insulating finishing systems • Thermal insulation coatings for safe-to-touch surfaces, energy efficiency, prevention of corrosion under insulation (CUI), thermal breaks, and condensation control • Insulation packs for oil and gas subsea pipelines • Architectural daylighting panels, glass units, and tensile roofing systems • Nonwovens for architectural membrane roofing • Industrial insulation • Ultralow-gloss matte paint coatings for industrial surfaces • Outdoor gear and apparel • Personal products including skin and beauty care Cabot has worked with hundreds of companies in support of integrating its products into a wide range of applications, with major users including Fixit, Johns Manville, Advanced Glazings, Tnemec, and Birdair, among many others. See Sect. 64.6 for additional companies that have integrated Cabot Aerogel’s particles into different applications.

64

The Aerogel Industry

1599

Table 64.8 Materials properties of Cabot Aerogel P100, P200, and P300 industrial-grade silica aerogel particles Description Target application Particle size range Pore diameter Porosity Particle density Bulk density Surface chemistry Surface area Thermal conductivity

Cabot Aerogel P100 Fine to coarse particle Industrial-grade insulative additive 0.1–4.0 mm ~20 nm >90% 0.12–0.15 g/cc 0.08–0.1 g/cc Hydrophobic 600–800 m2/g 19 mW/m-K at 12.5  C

Cabot Aerogel P200 Medium coarseness particle Industrial-grade insulative additive 0.1–1.2 mm ~20 nm >90% 0.12–0.18 g/cc 0.075–0.095 g/cc Hydrophobic 600–800 m2/g 19 mW/m-K at 12.5  C

Cabot Aerogel P300 Coarse particle Industrial-grade insulative additive 1.2–4.0 mm ~20 nm >90% 0.12–0.18 g/cc 0.065–0.085 g/cc Hydrophobic 600–800 m2/g 19 mW/m-K at 12.5  C

Table 64.9 Materials properties of Cabot Aerogel’s Thermal Wrap TW350, TW600, and TW800 silica aerogel/polymer fiber composite blankets Thickness Width Length Density Tensile strength Light transmission

TW350 3.5 mm 76 cm Up to 120 m 0.070 g/cc 517 kPa ~46%

TW600 6.0 mm 76 cm Up to 120 m 0.070 g/cc 517 kPa ~27%

TW800 8.0 mm 76 cm Up to 120 m 0.070 g/cc 517 kPa ~20%

Fig. 64.7 Real-world examples of applications benefitting from Cabot Aerogel products. (a) Insulative daylighting panels manufactured by Advanced Glazings incorporating Lumira LA1000 particles at Minneapolis-St. Paul International Airport. (Image credit and copyright to Metropolitan Airports Commission). (b) Pipe fitting coated with Tnemec’s Aerolon coating made with Cabot Aerogel particles, which

provide energy efficiency and safe-touch benefits. (Image credit and copyright to Tnemec Corporation). (c) Superinsulating Compression Pack from Johns Manville made with Cabot Aerogel particles installed in a pipe-in-pipe configuration. (Image credit and copyright to Cabot Corporation)

64.5.3 Nano Tech Co., Ltd.

shaped parts, granulates, and powders (Fig. 64.8). The company operates production lines using both supercritical and ambient-pressure drying technologies. In 2018, the company stated its annual production capacity to be 2,000,000 m2/year, which for comparison can be reasonably estimated as translating into about 40,000 m 3/year. The largest market for Nano’s products is thermal insulation for pipelines (Fig. 64.8) however, the company also sells into a number of other markets as well, with applications including building insulation, lithium-ion battery pack thermal management barriers for electric vehicles,

Shaoxing, Zhejiang Province, People’s Republic of China | www.nanuo.cn Founded in 2004, Nano Tech Co., Ltd. (also sometimes spelled as its Mandarin transliteration Nanuo, formerly Nano Hi-Tech), is China’s first and largest aerogel producer. Nano produces only silica-based aerogels on a commercial basis, with a variety of form factors including flexible fiber-reinforced composite blankets, panels,

64

1600

R. A. Collins et al.

Fig. 64.8 Silica aerogel products produced by Nano High-Tech and application examples. (Image credit and copyright to Nano Tech Co., Ltd.)

and thermal/acoustic insulation for high-speed trains. Since 2007, Nano has operated an aerogel R&D and technology transfer center in conjunction with the Bohr Institute of Solid State Physics at Tongji University. Today the company also maintains partnerships/collaborations with organizations including Chinese oil producer Sinopec; high-speed train manufacturer CRRC Co., Ltd.; and China General Nuclear Power Group (CGNPC). The company holds 28 Chinese patents providing coverage spanning methods for aerogel production technologies to applications.

aerogel granules, and silica aerogel powders. The company’s aerogel blankets are stated to combine silica aerogel with an inorganic fiber matrix. According to its website, as of 2017, the company’s annual production capacity was 28,000 m3/year, with primary end uses for its products being heated pipelines, furnaces, building façades, and clean-energy vehicles [11]. A summary of applications for Alison’s aerogel blanket and board products is provided in Table 64.10. A summary of materials properties for Alison’s aerogel blanket and board products is provided in Table 64.11 [11].

64.5.4 Yingde Alison Co., Ltd.

64.5.5 Shenzhen Aerogel Technology Co., Ltd.

Yingde, Guangdong Province, People’s Republic of China | www.ydalison.com

Shenzhen, Guangdong Province, People’s Republic of China | www.agel-tech.com

Alison Aerogel (also called YD Alison, Yingde Alison Hi-Tech, and Guangdong Alison) is one of the largest manufacturers of silica aerogel products in China. The company develops, manufactures, and distributes nanotechnologybased materials with an emphasis on novel material solutions for thermal management applications. The company’s target markets include pipelines for the oil and gas industry, building and construction, machinery and equipment, transportation, and home appliances. The company began production of aerogel materials in 2007 and today produces fiber-reinforced silica aerogel blankets and panels, silica

Shenzhen Aerogel Technology (also called Zhongning Aerogel, AGTech, and Aerogel Technology) is a Chinese aerogel producer headquartered in Shenzhen, Guangdong, manufacturing in Xintao, Hubei. The company has been engaged in aerogel production since 2015. The company’s main products include silica aerogel powders and blankets produced via ambient-pressure drying. According to its website, the company holds over 80 Chinese and international patents and has a current annual production capacity of 5000 m3/year for aerogel blankets and 100 metric tons/ year for aerogel powders [14]. The company plans to build

64

The Aerogel Industry

1601

Table 64.10 Aerogel products made by Alison Aerogel Product series DRT06 Series and GR10 Series

Form factor Aerogel blanket

GY10 Series

Aerogel panel

GY06 Series

Aerogel panel

Applications Flexible superinsulating blanket High-temperature insulating applications Prefabricated insulated pipes Tanks, pressure vessels, and plant equipment Steam pipelines Pipelines in thermal power, petrochemical, and chemical plants Medium- to high-temperature industrial furnaces Escape capsules for space launch systems Automotive applications, highspeed trains, and subway cars Building and construction Medium- to high-temperature industrial furnaces Escape capsules for space launch systems Specialized military equipment Shaped insulation covers Integrated insulation systems for buildings

Table 64.11 Materials properties and specifications for products made by Alison Aerogel [11] Form factor Thickness Width Thermal conductivity at 25  C Application temperature limit Density Hydrophobicity Chlorine content Flammability rating

Blanket 3, 5, 6, and 10 mm 150 cm 21 mW/m-K

Board 10, 20, and 30 mm 88  58 cm 90% 200  C to 1600  C

A summary of material properties for AeroVa aerogel powder is provided in Table 64.12. JIOS states that as of 2019, its commercial group had 50 full-time employees and was equipped with R&D capabilities to assist customers with product integration and commercial development. As of 2019, the company held 48 patents associated with the production of aerogel powder and applications of aerogels. In 2016, a joint venture between JIOS and Armacell was established for the production of composite aerogel blankets sold under the trade name ArmaGel, shown in Fig. 64.11 [51]. ArmaGel is a flexible insulation material comprising AeroVa silica aerogel powder incorporated into an e-glassfiber mat. This roll-type insulation product is stated to provide superior insulation performance over traditional insulation materials and covers a wide operating temperature range (200 to ~650  C). ArmaGel is stated to have various

Fig. 64.11 (a) Packaged AeroVa silica aerogel powder produced by JIOS Aerogel; (b) ArmaGel composite blanket produced in cooperation with Armacell. (Image credit and copyright to JIOS Aerogel Corporation)

advantages over conventional insulation materials including superior thermal and sound insulation properties, fire resistance, water resistance, and breathability. JIOS states that production capacity for ArmaGel was 750,000 m2/year as of July 2019 and that the company intends to scale production to 3 million m2/year by 2024 and to 10 million m2/year by 2029. JIOS is also developing its AeroVa powder product for incorporation into composite boards (gypsum, cement, perlite) and coatings. Table 64.13 lists properties for JIOS/ Armacell’s ArmaGel blanket [52]. JIOS identifies industrial sectors including oil and natural gas and chemical plants as their main sectors with 70% of use; others include building and construction (15%), ship building (10%), electronics (3%), textiles (1%), and

1604

R. A. Collins et al.

Table 64.13 Product specifications and materials properties for ArmaGel blanket produced by JIOS Aerogel and Armacell [52] Property Service temperature

Value 40  C to 650  C

Thermal conductivity

21 mW/m-K at 24  C 43 mW/m-K at 371  C