Comprehensive Organometallic Chemistry IV. Volume 14: Applications III. Materials Science, Nanoscience, Polymer Science and Surface Chemistry [14] 9780128202067

Citation preview

COMPREHENSIVE ORGANOMETALLIC CHEMISTRY IV

COMPREHENSIVE ORGANOMETALLIC CHEMISTRY IV EDITORS-IN-CHIEF

GERARD PARKIN Department of Chemistry, Columbia University, New York, NY, United States

KARSTEN MEYER Department of Chemistry and Pharmacy, Friedrich-Alexander-Universität, Erlangen, Germany

DERMOT O’HARE Department of Chemistry, University of Oxford, Oxford, United Kingdom

VOLUME 14

APPLICATIONS III. MATERIALS SCIENCE, NANOSCIENCE, POLYMER SCIENCE AND SURFACE CHEMISTRY VOLUME EDITOR

DEREK P. GATES Department of Chemistry, University of British Columbia, Vancouver, BC, Canada

Elsevier Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States Copyright © 2022 Elsevier Ltd. All rights reserved No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers may always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein.

Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISBN 978-0-12-820206-7 For information on all publications visit our website at http://store.elsevier.com

Publisher: Oliver Walter Acquisition Editor: Blerina Osmanaj Content Project Manager: Claire Byrne Associate Content Project Manager: Fahmida Sultana Designer: Christian Bilbow

CONTENTS OF VOLUME 14 Editor Biographies

vii

Contributors to Volume 14

xiii

Preface 14.01

xv

Applications III. Materials Science, Nanoscience, Polymer Science and Surface Chemistry

1

Derek P Gates

14.02

Ferrocene and Related Metallocene Polymers

3

André Schäfer

14.03

Conjugated Poly(metalla-ynes)

23

Ashanul Haque, Muhammad S Khan, Mei-Tung Lau, Zikang Li, Paul R Raithby, and Wai-Yeung Wong

14.04

Organoboron and Related Group 13 Polymers

71

Holger Helten

14.05

Organosilicon and Related Group 14 Polymers

135

Rebekka S Klausen and Ernesto Ballestero-Martí nez

14.06

Organophosphorus and Related Group 15 Polymers

166

Jordann AL Wells and Andreas Orthaber

14.07

Organometallic Dendrimers

233

Xu-Qing Wang, Xiao-Qin Xu, Wei Wang, and Hai-Bo Yang

14.08

Organometallic Functionalized MOFs - Reactivity and Catalysis

257

Thomas M Rayder and Casey R Wade

14.09

Organometallic Mesogens

285

Manuel Bardají and Silverio Coco

14.10

Organometallic Complexes for Optoelectronic Applications

339

Zhijun Ruan and Zhen Li

14.11

Organometallic Lanthanide Complexes as Single Molecule Magnets

383

Bijoy Dey and Vadapalli Chandrasekhar

14.12

Organometallic Receptors for Charged and Neutral Guest Species

418

Robert Hein and Paul D Beer

14.13

Surface Organometallic Chemistry and Catalysis

463

Walid Al Maksoud, Sandeep Mishra, Aya Saidi, Manoja K Samantaray, and Jean Marie Basset

14.14

Common Precursors and Surface Mechanisms for Atomic Layer Deposition

534

Seán Thomas Barry, Peter George Gordon, and Vincent Vandalon

v

EDITOR BIOGRAPHIES Editors in Chief Karsten Meyer studied chemistry at the Ruhr University Bochum and performed his Ph.D. thesis work on the molecular and electronic structure of first-row transition metal complexes under the direction of Professor Karl Wieghardt at the Max Planck Institute in Mülheim/Ruhr (Germany). He then proceeded to gain research experience in the laboratory of Professor Christopher Cummins at the Massachusetts Institute of Technology (USA), where he appreciated the art of synthesis and developed his passion for the coordination chemistry and reactivity of uranium complexes. In 2001, he was appointed to the University of California, San Diego, as an assistant professor and was named an Alfred P. Sloan Fellow in 2004. In 2006, he accepted an offer (C4/W3) to be the chair of the Institute of Inorganic & General Chemistry at the Friedrich-Alexander-University ErlangenNürnberg (FAU), Germany. Among his awards and honors, he was elected a lifetime honorary member of the Israel Chemical Society and a fellow of the Royal Society of Chemistry (UK). Karsten received the Elhuyar-Goldschmidt Award from the Royal Society of Chemistry of Spain, the Ludwig Mond Award from the RSC (UK), and the Chugaev Commemorative Medal from the Russian Academy of Sciences. He has also enjoyed visiting professorship positions at the universities of Manchester (UK) and Toulouse (F) as well as the Nagoya Institute of Technology (JP) and ETH Zürich (CH). The Meyer lab research focuses on the synthesis of custom-tailored ligand environments and their transition and actinide metal coordination complexes. These complexes often exhibit unprecedented coordination modes, unusual electronic structures, and, consequently, enhanced reactivities toward small molecules of biological and industrial importance. Interestingly, Karsten’s favorite molecule is one that exhibits little reactivity: the Th symmetric U(dbabh)6. Dermot O’Hare was born in Newry, Co Down. He studied at Balliol College, Oxford University, where he obtained his B.A., M.A., and D.Phil. degrees under the direction of Professor M.L.H. Green. In 1985, he was awarded a Royal Commission of 1851 Research Fellowship, during this Fellowship he was a visiting research fellow at the DuPont Central Research Department, Wilmington, Delaware in 1986–87 in the group led by Prof. J.S. Miller working on molecular-based magnetic materials. In 1987 he returned to Oxford to a short-term university lectureship and in 1990 he was appointed to a permanent university position and a Septcentenary Tutorial Fellowship at Balliol College. He has previously been honored by the Institüt de France, Académie des Sciences as a leading scientist in Europe under 40 years. He is currently professor of organometallic and materials chemistry in the Department of Chemistry at the University of Oxford. In addition, he is currently the director of the SCG-Oxford Centre of Excellence for chemistry and associate head for business & innovation in the Mathematics, Physical and Life Sciences Division. He leads a multidisciplinary research team that works across broad areas of catalysis and nanomaterials. His research is specifically targeted at finding solutions to global issues relating to energy, zero carbon, and the circular economy. He has been awarded numerous awards and prizes for his creative and

vii

viii

Editor Biographies

ground-breaking work in inorganic chemistry, including the Royal Society Chemistry’s Sir Edward Frankland Fellowship, Ludwig Mond Prize, Tilden Medal, and Academia–Industry Prize and the Exxon European Chemical and Engineering Prize. Gerard Parkin received his B.A., M.A., and D.Phil. degrees from the Queen’s College, Oxford University, where he carried out research under the guidance of Professor Malcolm L.H. Green. In 1985, he moved to the California Institute of Technology as a NATO postdoctoral fellow to work with Professor John E. Bercaw. He joined the Faculty of Columbia University as assistant professor in 1988 and was promoted to associate professor in 1991 and to professor in 1994. He served as chairman of the Department from 1999 to 2002. He has also served as chair of the New York Section of the American Chemical Society, chair of the Inorganic Chemistry and Catalytic Science Section of the New York Academy of Sciences, chair of the Organometallic Subdivision of the American Chemical Society Division of Inorganic Chemistry, and chair of the Gordon Research Conference in Organometallic Chemistry. He is an elected fellow of the American Chemical Society, the Royal Society of Chemistry, and the American Association for the Advancement of Science, and is the recipient of a variety of international awards, including the ACS Award in pure chemistry, the ACS Award in organometallic chemistry, the RSC Corday Morgan Medal, the RSC Award in organometallic chemistry, the RSC Ludwig Mond Award, and the RSC Chem Soc Rev Lecture Award. He is also the recipient of the United States Presidential Award for Excellence in Science, Mathematics and Engineering Mentoring, the United States Presidential Faculty Fellowship Award, the James Flack Norris Award for Outstanding Achievement in the Teaching of Chemistry, the Columbia University Presidential Award for Outstanding Teaching, and the Lenfest Distinguished Columbia Faculty Award. His principal research interests are in the areas of synthetic, structural, and mechanistic inorganic chemistry.

Volume Editors Simon Aldridge is professor of chemistry at the University of Oxford and director of the UKRI Centre for Doctoral Training in inorganic chemistry for Future Manufacturing. Originally from Shrewsbury, England, he received both his B.A. and D.Phil. degrees from the University of Oxford, the latter in 1996 for work on hydride chemistry under the supervision of Tony Downs. After post-doctoral work as a Fulbright Scholar at Notre Dame with Tom Fehlner, and at Imperial College London (with Mike Mingos), he took up his first academic position at Cardiff University in 1998. He returned to Oxford in 2007, being promoted to full professor in 2010. Prof. Aldridge has published more than 230 papers to date and is a past winner of the Dalton Transactions European Lectureship (2009), the Royal Society of Chemistry’s Main Group Chemistry (2010) and Frankland Awards (2018), and the Forschungspreis of the Alexander von Humboldt Foundation (2021). Prof. Aldridge’s research interests are primarily focused on main group organometallic chemistry, and in particular the development of compounds with unusual electronic structure, and their applications in small molecule activation and catalysis (website: http:// aldridge.web.ox.ac.uk). (Picture credit: John Cairns)

Editor Biographies

ix

Eszter Boros is associate professor of chemistry at Stony Brook University with courtesy appointments in radiology and pharmacology at Stony Brook Medicine. Eszter obtained her M.Sc. (2007) at the University of Zurich, Switzerland and her Ph.D. (2011) in chemistry from the University of British Columbia, Canada. She was a postdoc (2011–15) and later instructor (2015–17) in radiology at Massachusetts General Hospital and Harvard Medical School. In 2017, Eszter was appointed as assistant professor of chemistry at Stony Brook University, where her research group develops new approaches to metal-based diagnostics and therapeutics at the interfaces of radiochemistry, inorganic chemistry and medicine. Her lab’s work has been extensively recognized; Eszter holds various major federal grants (NSF CAREER Award, NIH NIBIB R21 Trailblazer, NIH NIGMS R35 MIRA) and has been named a 2020 Moore Inventor Fellow, the 2020 Jonathan L. Sessler Fellow (American Chemical Society, Inorganic Division), recipient of a 2021 ACS Infectious Diseases/ACS Division of Biological Chemistry Young Investigator Award (American Chemical Society), and was also named a 2022 Alfred P. Sloan Research Fellow in chemistry. Scott R. Daly is associate professor of chemistry at the University of Iowa in the United States. After spending 3 years in the U.S. Army, he obtained his B.S. degree in chemistry in 2006 from North Central College, a small liberal arts college in Naperville, Illinois. He then went on to receive his Ph.D. at the University of Illinois at Urbana-Champaign in 2010 under the guidance of Professor Gregory S. Girolami. His thesis research focused on the synthesis and characterization of chelating borohydride ligands and their use in the preparation of volatile metal complexes for chemical vapor deposition applications. In 2010, he began working as a Seaborg postdoctoral fellow with Drs. Stosh A. Kozimor and David L. Clark at Los Alamos National Laboratory in Los Alamos, New Mexico. His research there concentrated on the development of ligand K-edge X-ray absorption spectroscopy (XAS) to investigate covalent metal–ligand bonding and electronic structure variations in actinide, lanthanide, and transition metal complexes with metal extractants. He started his independent career in 2012 at George Washington University in Washington, DC, and moved to the University of Iowa shortly thereafter in 2014. His current research interests focus on synthetic coordination chemistry and ligand design with emphasis on the development of chemical and redox noninnocent ligands, mechanochemical synthesis and separation methods, and ligand K-edge XAS. His research and outreach efforts have been recognized with an Outstanding Faculty/Staff Advocate Award from the University of Iowa Veterans Association (2016), a National Science Foundation CAREER Award (2017), and a Hawkeye Distinguished Veterans Award (2018). He was promoted to associate professor with distinction as a College of Liberal Arts and Sciences Deans Scholar in 2020. Lena J. Daumann is currently professor of bioinorganic and coordination chemistry at the Ludwig Maximilian Universität in Munich. She studied chemistry at the University of Heidelberg working with Prof. Peter Comba and subsequently conducted her Ph.D. at the University of Queensland (Australia) from 2010 to 2013 holding IPRS and UQ Centennial fellowships. In 2013 she was part of the Australian Delegation for the 63rd Lindau Nobel Laureate meeting in chemistry. Following postdoctoral stays at UC Berkeley with Prof. Ken Raymond (2013–15) and in Heidelberg, funded by the Alexander von Humboldt Foundation, she started her independent career at the LMU Munich in 2016. Her bioinorganic research group works on elucidating the role of lanthanides for bacteria as well as on iron enzymes and small biomimetic complexes that play a role in epigenetics and DNA repair. Daumann’s teaching and research have been recognized with numerous awards and grants. Among them are the national Ars Legendi Prize for chemistry and the Therese von Bayern Prize in 2019 and the Dozentenpreis of the “Fonds der Chemischen Industrie“ in 2021. In 2018 she was selected as fellow for the Klaus Tschira Boost Fund by the German Scholars Organisation and in 2020 she received a Starting grant of the European Research Council to study the uptake of lanthanides by bacteria.

x

Editor Biographies

Derek P. Gates hails from Halifax, Nova Scotia (Canada) where he completed his B.Sc. (Honours Chemistry) degree at Dalhousie University in 1993. He completed his Ph.D. degree under the supervision of Professor Ian Manners at the University of Toronto in 1997. He then joined the group of Professor Maurice Brookhart as an NSERC postdoctoral fellow at the University of North Carolina at Chapel Hill (USA). He began his independent research career in 1999 as an assistant professor at the University of British Columbia in Vancouver (Canada). He has been promoted through the ranks and has held the position of professor of chemistry since 2011. At UBC, he has received the Science Undergraduate Society—Teaching Excellence Award, the Canadian National Committee for IUPAC Award, and the Chemical Society of Canada—Strem Chemicals Award for pure or applied inorganic chemistry. His research interests bridge the traditional fields of inorganic and polymer chemistry with particular focus on phosphorus chemistry. Key topics include the discovery of novel structures, unusual bonding, new reactivity, along with applications in catalysis and materials science. Patrick Holland performed his Ph.D. research in organometallic chemistry at UC Berkeley with Richard Andersen and Robert Bergman. He then learned about bioinorganic chemistry through postdoctoral research on copper-O2 and copper-thiolate chemistry with William Tolman at the University of Minnesota. His independent research at the University of Rochester initially focused on systematic development of the properties and reactions of three-coordinate complexes of iron and cobalt, which can engage in a range of bond activation reactions and organometallic transformations. Since then, his research group has broadened its studies to iron-N2 chemistry, reactive metal–ligand multiple bonds, iron–sulfur clusters, engineered metalloproteins, redox-active ligands, and solar fuel production. In 2013, Prof. Holland moved to Yale University, where he is now Conkey P. Whitehead Professor of Chemistry. His research has been recognized with an NSF CAREER Award, a Sloan Research Award, Fulbright and Humboldt Fellowships, a Blavatnik Award for Young Scientists, and was elected as fellow of the American Association for the Advancement of Science. In the area of N2 reduction, his group has established molecular principles to weaken and break the strong N–N bond, in order to use this abundant resource for energy and synthesis. His group has made a particular effort to gain an insight into iron chemistry relevant to nitrogenase, the enzyme that reduces N2 in nature. His group also maintains an active program in the use of inexpensive metals for transformations of alkenes. Mechanistic details are a central motivation to Prof. Holland and the wonderful group of over 80 students with whom he has worked. Steve Liddle was born in Sunderland in the North East of England and gained his B.Sc. (Hons) and Ph.D. from Newcastle University. After postdoctoral fellowships at Edinburgh, Newcastle, and Nottingham Universities he began his independent career at Nottingham University in 2007 with a Royal Society University Research Fellowship. This was held in conjunction with a proleptic Lectureship and he was promoted through the ranks to associate professor and reader in 2010 and professor of inorganic chemistry in 2013. He remained at Nottingham until 2015 when he was appointed professor and head of inorganic chemistry and co-director of the Centre for Radiochemistry Research at The University of Manchester. He has been a recipient of an EPSRC Established Career Fellowship and ERC Starter and Consolidator grants. He is an elected fellow of The Royal Society of Edinburgh and fellow of the Royal Society of Chemistry and he is vice president to the Executive Committee of the European Rare Earth and Actinide Society. His principal research interests are focused on f-element chemistry, involving exploratory synthetic chemistry coupled to detailed electronic structure and reactivity studies to elucidate structure-bonding-property relationships. He is the recipient of a variety of prizes, including the IChemE Petronas Team Award for Excellence in Education and Training, the RSC Sir Edward Frankland Fellowship, the RSC Radiochemistry

Editor Biographies

xi

Group Bill Newton Award, a 41st ICCC Rising Star Award, the RSC Corday-Morgan Prize, an Alexander von Humboldt Foundation Friedrich Wilhelm Bessel Research Award, the RSC Tilden Prize, and an RSC Dalton Division Horizon Team Prize. He has published over 220 research articles, reviews, and book chapters to date. David Liptrot received his MChem (Hons) in chemistry with Industrial Training from the University of Bath in 2011 and remained there to undertake a Ph.D. on group 2 catalysis in the laboratory of Professor Mike Hill. After completing this in 2015 he took up a Lindemann Postdoctoral Fellowship with Professor Philip Power FRS (University of California, Davis, USA). In 2017 he began his independent career returning to the University of Bath and in 2019 was awarded a Royal Society University Research Fellowship. His interests concern new synthetic methodologies to introduce main group elements into functional molecules and materials.

David P. Mills hails from Llanbradach and Caerphilly in the South Wales Valleys. He completed his MChem (2004) and Ph.D. (2008) degrees at Cardiff University, with his doctorate in low oxidation state gallium chemistry supervised by Professor Cameron Jones. He moved to the University of Nottingham in 2008 to work with Professor Stephen Liddle for postdoctoral studies in lanthanide and actinide methanediide chemistry. In 2012 he moved to the University of Manchester to start his independent career as a lecturer, where he has since been promoted to full professor of inorganic chemistry in 2021. Although he is interested in all aspects of nonaqueous synthetic chemistry his research interests are currently focused on the synthesis and characterization of f-block complexes with unusual geometries and bonding regimes, with the aim of enhancing physicochemical properties. He has been recognized for his contributions to both research and teaching with prizes and awards, including a Harrison-Meldola Memorial Prize (2018), the Radiochemistry Group Bill Newton Award (2019), and a Team Member of the Molecular Magnetism Group for the Dalton Division Horizon Prize (2021) from the Royal Society of Chemistry. He was a Blavatnik Awards for Young Scientists in the United Kingdom Finalist in Chemistry in 2021 and he currently holds a European Research Council Consolidator Grant. Ian Tonks is the Lloyd H. Reyerson professor at the University of MinnesotaTwin Cities, and associate editor for the ACS journal Organometallics. He received his B.A. in chemistry from Columbia University in 2006 and performed undergraduate research with Prof. Ged Parkin. He earned his Ph.D. in 2012 from the California Institute of Technology, where he worked with Prof. John Bercaw on olefin polymerization catalysis and early transition metal-ligand multiply bonded complexes. After postdoctoral research with Prof. Clark Landis at the University of Wisconsin, Madison, he began his independent career at the University of Minnesota in 2013 and earned tenure in 2019. His current research interests are focused on the development of earth abundant, sustainable catalytic methods using early transition metals, and also on catalytic strategies for incorporation of CO2 into polymers. Prof. Tonks’ work has recently been recognized with an Outstanding New Investigator Award from the National Institutes of Health, an Alfred P. Sloan Fellowship, a Department of Energy CAREER award, and the ACS Organometallics Distinguished Author Award, among others. Additionally, Prof. Tonks’ service toward improving academic safety culture was recently recognized with the 2021 ACS Division of Chemical Health and Safety Graduate Faculty Safety Award.

xii

Editor Biographies

Timothy H. Warren is the Rosenberg professor and chairperson in the Department of Chemistry at Michigan State University. He obtained his B.S. from the University of Illinois at Urbana-Champaign in 1992 and Ph.D. from the Massachusetts Institute of Technology in 1997. After 2 years of postdoctoral research at the Organic Chemistry Institute of the University of Münster, Germany with Prof. Dr. Gerhart Erker, Dr. Warren started his independent career at Georgetown University in 1999 where he was named the Richard D. Vorisek professor of chemistry in 2014. He moved to Michigan State University in 2021. Prof. Warren’s research interests span synthetic and mechanistic inorganic, organometallic, and bioinorganic chemistry with a focus on catalysis. His research group develops environmentally friendly methods for organic synthesis via C–H functionalization, explores the interconversion of nitrogen and ammonia as carbon-free fuels, and decodes ways that biology communicates using nitric oxide as a molecular messenger. Mechanistic studies on these chemical reactions catalyzed by metal ions such as iron, nickel, copper, and zinc enable new insights for the development of useful catalysts for synthesis and energy applications as well as lay the mechanistic groundwork to understand biochemical nitric oxide misregulation. Dr. Warren received the NSF CAREER Award, chaired the 2019 Inorganic Reaction Mechanisms Gordon Research Conference, and has served on the ACS Division of Inorganic Chemistry executive board and on the editorial boards of Inorganic Synthesis, Inorganic Chemistry, and Chemical Society Reviews.

CONTRIBUTORS TO VOLUME 14 Ernesto Ballestero-Martínez Escuela de Quí mica and Centro de Investigación en Ciencia e Ingenierí a de Materiales, Universidad de Costa Rica, San José, Costa Rica

Robert Hein Department of Chemistry, University of Oxford, Oxford, United Kingdom

Manuel Bardají IU CINQUIMA/Quí mica Inorgánica, Facultad de Ciencias, Universidad de Valladolid, Valladolid, Spain

Holger Helten Institute of Inorganic Chemistry & Institute for Sustainable Chemistry and Catalysis With Boron (ICB), Julius-Maximilians-Universität Würzburg, Würzburg, Germany

Seán Thomas Barry Department of Chemistry, Carleton University, Ottawa, ON, Canada

Muhammad S Khan Department of Chemistry, Sultan Qaboos University, Muscat, Oman

Jean Marie Basset King Abdullah University of Science and Technology (KAUST), Kaust Catalysis Center (KCC), Thuwal, Saudi Arabia

Rebekka S Klausen Department of Chemistry, Johns Hopkins University, Baltimore, MD, United States

Paul D Beer Department of Chemistry, University of Oxford, Oxford, United Kingdom Vadapalli Chandrasekhar Tata Institute of Fundamental Research Hyderabad, Hyderabad, India; Department of Chemistry, Indian Institute of Technology Kanpur, Kanpur, India Silverio Coco IU CINQUIMA/Quí mica Inorgánica, Facultad de Ciencias, Universidad de Valladolid, Valladolid, Spain Bijoy Dey Tata Institute of Fundamental Research Hyderabad, Hyderabad, India Derek P Gates Department of Chemistry, University of British Columbia, Vancouver, BC, Canada Peter George Gordon Department of Chemistry, Carleton University, Ottawa, ON, Canada Ashanul Haque Department of Chemistry, College of Science, University of Hail, Hail, Saudi Arabia

Mei-Tung Lau Department of Applied Biology and Chemical Technology and Research Institute for Smart Energy, The Hong Kong Polytechnic University, Hong Kong, People’s Republic of China Zhen Li Hubei Key Lab on Organic and Polymeric Opto-Electronic Materials, Department of Chemistry, Wuhan University, Wuhan, China; Institute of Molecular Aggregation Science, Tianjin University, Tianjin, China Zikang Li Department of Applied Biology and Chemical Technology and Research Institute for Smart Energy, The Hong Kong Polytechnic University, Hong Kong, People’s Republic of China Walid Al Maksoud King Abdullah University of Science and Technology (KAUST), Kaust Catalysis Center (KCC), Thuwal, Saudi Arabia Sandeep Mishra King Abdullah University of Science and Technology (KAUST), Kaust Catalysis Center (KCC), Thuwal, Saudi Arabia; Department of Physics & NMR Research Center, Indian Institute of Science Education and Research, Pune, India

xiii

xiv

Contributors to Volume 14

Andreas Orthaber Synthetic Molecular Chemistry—Department of Chemistry, Ångström Laboratories, Uppsala University, Uppsala, Sweden Paul R Raithby Department of Chemistry, University of Bath, Bath, United Kingdom Thomas M Rayder Department of Chemistry and Biochemistry, The Ohio State University, Columbus, OH, United States Zhijun Ruan College of Chemistry and Chemical Engineering, Huanggang Normal University, Huanggang, China Aya Saidi King Abdullah University of Science and Technology (KAUST), Kaust Catalysis Center (KCC), Thuwal, Saudi Arabia Manoja K Samantaray King Abdullah University of Science and Technology (KAUST), Kaust Catalysis Center (KCC), Thuwal, Saudi Arabia André Schäfer Department of Chemistry, Faculty of Natural Sciences and Technology, Saarland University, Saarbrücken, Saarland, Federal Republic of Germany Vincent Vandalon Department of Applied Physics, Eindhoven University of Technology, Eindhoven, The Netherlands Casey R Wade Department of Chemistry and Biochemistry, The Ohio State University, Columbus, OH, United States

Wei Wang Shanghai Key Laboratory of Green Chemistry and Chemical Processes, Chang-Kung Chuang Institute, School of Chemistry and Molecular Engineering, East China Normal University, Shanghai, China Xu-Qing Wang Shanghai Key Laboratory of Green Chemistry and Chemical Processes, Chang-Kung Chuang Institute, School of Chemistry and Molecular Engineering, East China Normal University, Shanghai, China Jordann AL Wells Synthetic Molecular Chemistry—Department of Chemistry, Ångström Laboratories, Uppsala University, Uppsala, Sweden Wai-Yeung Wong Department of Applied Biology and Chemical Technology and Research Institute for Smart Energy, The Hong Kong Polytechnic University, Hong Kong, People's Republic of China Xiao-Qin Xu Shanghai Key Laboratory of Green Chemistry and Chemical Processes, Chang-Kung Chuang Institute, School of Chemistry and Molecular Engineering, East China Normal University, Shanghai, China Hai-Bo Yang Shanghai Key Laboratory of Green Chemistry and Chemical Processes, Chang-Kung Chuang Institute, School of Chemistry and Molecular Engineering, East China Normal University, Shanghai, China

PREFACE Published 40 years ago in 1982, the first edition of Comprehensive Organometallic Chemistry (COMC) provided an invaluable resource that enabled chemists to become efficiently informed of the properties and reactions of organometallic compounds of both the main group and transition metals. This area of chemistry continued to develop at a rapid pace such that it necessitated the publication of subsequent editions, namely Comprehensive Organometallic Chemistry II (COMC2) in 1995 and Comprehensive Organometallic Chemistry III (COMC3) in 2007. Organometallic chemistry has continued to be vibrant in the 15 years following the publication of COMC3, not only by affording compounds with novel structures and reactivity but also by having important applications in organic syntheses and industrial processes, as illustrated by the awarding of the 2010 Nobel prize to Heck, Negishi, and Suzuki for the development of palladium-catalyzed cross couplings in organic syntheses. Comprehensive Organometallic Chemistry IV (COMC4) thus serves the same important role as its predecessors by providing an indispensable means for researchers and educators to obtain efficiently an up-to-date analysis of a particular aspect of organometallic chemistry. COMC4 comprises 15 volumes, of which the first provides a review of topics concerned with techniques and concepts that feature prominently in current organometallic chemistry, while 5 volumes are devoted to applications that include organic synthesis, materials science, bio-organometallics, metallo-therapy, metallodiagnostics, medicine, and environmental chemistry. In this regard, we are very grateful to the volume editors for their diligent efforts, and the authors for producing high-quality chapters, all of which were written during the COVID-19 pandemic. Finally, we wish to thank the many staff at Elsevier for their efforts to ensure that the project, initiated in the winter of 2018, remained on schedule. Karsten Meyer, Erlangen, March 2022 Dermot O’Hare, Oxford, March 2022 Gerard Parkin, New York, March 2022

xv

14.01 Applications III. Materials Science, Nanoscience, Polymer Science and Surface Chemistry Derek P Gates, Department of Chemistry, University of British Columbia, Vancouver, BC, Canada © 2022 Elsevier Ltd. All rights reserved.

The growing importance of materials science applications in organometallic chemistry is the focus of Section 8, entitled “Applications III. Materials Science, nanoscience, polymer science and surface chemistry.” With Comprehensive Organometallic Chemistry IV, these topics, many of which have appeared separately in previous editions of COMC, are now collected together within their own volume. Particularly important advances involve exploiting the unique magnetic, optoelectronic and catalytic properties that organometallic functional moieties can impart to nanostructures, polymers, surfaces and other materials. The chapters included highlight the great successes that have been achieved from both a fundamental and applied perspective with emphasis on developments in the new millennium. They are selected to emphasize the continued growth of the field of organometallic chemistry and its expanded impact on modern scientific research. The volume opens with a comprehensive overview by Schäfer on the exciting area of macromolecules containing ferrocene and related metallocenes. A fascinating advance involves the development of the crystallization driven self-assembly of poly (ferrocenylsilane) block copolymers and its use to selectively generate large nanostructures such as cylinders, platelets and vesicles with high precision. The theme of transition metal-based organometallic polymers continues with a detailed synopsis of the last decade of developments in the exciting area of conjugated poly(metalla-ynes). Haque, Khan, Lau, Li, Raithby and Wong cover the synthetic methods employed to access these materials and highlight their applications as sensors, memory devices, catalysts and biologically active materials. A growing area of organometallic polymers are those featuring p-block elements such as boron and group 13 elements. Helten describes how the facile interconversion between conjugated and non-conjugated states of boron has led to applications for boron polymers as fluorescent sensors, semiconductors, light-emitting diodes, field-effect transistors. Klausen and Ballestero-Martinex provide a comprehensive treatise on the important field of catenated group 14 elements. Highlighted are the fascinating optical and electronic properties of s-conjugated polysilanes, polygermanes and polystannanes which starkly contrast those of polyolefins. Poly(cyclohexasilanes) and corresponding cyclocyclic oligomers illustrate the dramatic increase in sophistication of synthetic methods and highlight the nature of s-conjugation in these fascinating organometallic materials. Orthaber and Wells illustrate the large variety of organophosphorus and related group 15 building blocks that have been developed and employed to generate novel macromolecular structures over the past decades. Diverse applications are highlighted ranging from catalysis and optoelectronic polymers to flame retardants and gene delivery agents. Wang, Xu, Wang and Yang highlight the fascinating structural complexity possible for organometallic macromolecules in a comprehensive overview of dendrimers featuring organometallic cores, branches and peripheries. The incorporation of poly (metalla-yne) moieties into either the branches of dendrimers or dendritic rotaxanes, both of which respond to external stimuli, represent particularly imaginative applications of organometallic functional groups. As the focus of the volume shifts slightly, Rayder and Wade provide extensive coverage of the rapidly expanding area of organometallic-functionalized metal-organic frameworks (MOFs). Applications of MOFs in catalysis highlighted therein include, for example: Suzuki-Miyaura coupling, olefin oligo- and poly-merization, oxidation of benzene, CO2 hydrogenation to alcohols, alkene and alkyne hydrogenation, hydroboration and hydrosilylation, olefin metathesis, Diels-Alder, among many others. In the next chapter, Bardaji and Coco detail the most recent advances in the development and application of organometallic mesogens. Organometallic liquid crystals featuring main group elements along with elements of groups 6, 7, 8, 9, 10 and 11 are highlighted in addition to their intriguing optical and electro-optical properties. Ruan and Li provide a detailed narrative of the wide variety of organometallic complexes that have been designed and employed in light-emitting diodes, photovoltaics, non-linear optics, photocatalysis, chemosensing, bioimaging among other fields of optoelectronics. This chapter summarizes the optical behaviors of organometallic complexes when stimulated by external mechanical forces (i.e. mechanoluminescence and mechanochromic). The synthesis and application of organometallic lanthanide complexes as single molecule magnets is treated by Dey and Chandrasekhar. A focus of this chapter involves the development of creative synthetic methods to combine organometallic ligands with lanthanide ions with high spin ground states along with magnetic anisotropy. In each case, a discussion of the magnetic relaxation parameters of the single molecule magnets and single ion magnets is provided. The volume continues with the development and utilization of highly functional structural scaffolds featuring organometallic receptors for use in host-guest chemistry. Hein and Beer demonstrate through recent examples how supramolecular receptors based on organometallic complexes has found widespread use in areas such as: extraction, membrane transport, catalysis, materials design, and sensing. In the penultimate chapter, a detailed account is provided on the state-of-the-art of surface organometallic chemistry for applications in catalytic transformations. Al Maksoud, Mishra, Saidi, Samantaray and Basset outline the latest developments in

Comprehensive Organometallic Chemistry IV

https://doi.org/10.1016/B978-0-12-820206-7.00170-0

1

2

Applications III. Materials Science, Nanoscience, Polymer Science and Surface Chemistry

this field which bridges the gap between homogeneous and heterogeneous catalysis. The authors highlight recent advances in: alkane metathesis, cross-metathesis of alkanes, hydroamination, reduction of dinitrogen, alkane aromatization, olefin epoxidation and catalytic CO2 chemistry. To close this volume, Barry, Gordon and Vandalon report on recent developments in the employment of organometallic compounds as precursors to thin metal-containing films via atomic layer deposition processes. Many designer ligands are highlighted that have been made specifically for ALD applications. In the past 15 years since COMC III, there has been increasing focus on the development and use of organometallic complexes to generate materials with unique function not possible for organic materials. Perhaps most striking are the applications of organometallic materials as sensors, catalysts, molecular magnets, semiconductors and optoelectronics. Clearly, in the past few decades contemporary organometallic chemistry has firmly established itself at the crossroads of molecular chemistry and materials science. Moreover, the wide range of countries represented by the authors of this volume serves as an indication of the global impact of organometallic materials chemistry. As the prominence of organometallic chemistry in the fields of polymers, surfaces, nanostructures and materials continues to flourish, it can be anticipated that developments featured in this volume will inspire continued imaginative contributions from a rapidly growing cohort of increasingly interdisciplinary researchers. Therefore, the next decades offer considerable opportunities for advancement of these and new themes in organometallic-based materials.

14.02

Ferrocene and Related Metallocene Polymers

André Schäfer, Department of Chemistry, Faculty of Natural Sciences and Technology, Saarland University, Saarbrücken, Saarland, Federal Republic of Germany © 2022 Elsevier Ltd. All rights reserved.

14.02.1 Introduction 14.02.2 Ferrocene-based polymers 14.02.3 Ruthenocene-based polymers 14.02.4 Cobaltocene-based polymers 14.02.5 Nickelocene-based polymers 14.02.6 Arenocene-based polymers 14.02.7 Summary Acknowledgment References

3 3 12 14 17 18 19 19 20

14.02.1 Introduction Historically, synthetic polymers were often perceived as organic compounds, although natural macromolecular structures with metal centers play a pronounced and essential role in basically all biological systems.1 However, for a long time, synthetic metallopolymers containing metal centers either in the main-chain or in the side-chains have suffered from preparative difficulties, often resulting in low-molecular weight, ill-defined and poorly-characterized materials, although they have been a desirable target with regards to their electronic and electrochemical properties, and possible optoelectronic and catalytic applications.2–8 One of the most important class of metallopolymers nowadays are metallocene-based systems, with the first reports of oligomeric/polymeric ferrocene compounds dating back to the 1960s.5,9 These early reports however offered little characterization and only in the last few decades, metallopolymers containing metallocene moieties have emerged as a highly interesting and versatile class of materials, due to new and improved synthetic routes.10 Ferrocene-based polymers represent the largest class of this family of compounds, but, more recently, related polymers containing ruthenocene, nickelocene and cobaltocene units, as well as other arenocene units have also been explored. This chapter aims to provide an overview of the historic development of these compounds and summarizes recent advancements in this area.

14.02.2 Ferrocene-based polymers In general, polymers containing ferrocene units can be divided into two classes: Polymers with ferrocene moieties in the main chain, and polymers with pendant ferrocene moieties in the side-chains (Fig. 1).10–15 Among the polymers with ferrocene units in the main-chain, polyferrocenylsilanes are the most prominent examples.10,11,13 The earliest attempts to synthesize these compounds date back to the 1960s, where in polycondensation-type salt-elimination reactions 1,10 -dilithioferrocene was reacted with dichlorosilanes at room temperature.5 The products of these reactions were however only of low molecular weight and the colors were described as brown or black, suggesting significant amounts of unidentified impurities, possibly iron metal. Similarly, the reaction of dilithio bis(cyclopentadienidyl)dimethylsilane with iron(II) chloride was also reported to give only low-molecular weight poly(ferrocenyldimethylsilane).9 The breakthrough came in the 1990s,9,13,16 when Manners and coworkers reported that the sila[1]ferrocenophanes, which can be obtained from the reaction of 1,10 -dilithioferrocene with dichlorosilanes at low temperates and were first reported in the 1970s,17–22 possess a large ring-strain (ca. 60 to 80 kJ mol−1) and began to investigate the application of these molecules in ring-opening polymerizations. The ring-strain in these systems is due to the loss of coplanarity of the Cp moieties and is often quantified by the dihedral angle a between the Cp planes, or by the Cpcentroid–metal–Cpcentroid angle d (Fig. 2).16,21,23–25 In addition, the deviation of the ansa-bridging atom (e.g. Si) from the Cp plane is also an indicator of ring-strain, whereby the angle between the bridging atom and the Cp plane is often referred to as the b angle in such systems (Fig. 2). Interestingly, the degree of tilt of the cyclopentadienyl moieties also influences the color of the complex. While unbridged ferrocene, which has coplanar cyclopentadienyl groups, is a yellow to orange solid, sila[1]ferrocenophanes are usually deep red in color.21,26 Due to this ring-strain, sila[1]ferrocenophanes can undergo ring-opening polymerizations upon heating, cationic or anionic initiation, irradiation or treatment with transition metal catalysts, to yield high-molecular weight, well-defined linear polymers (Scheme 1).

Comprehensive Organometallic Chemistry IV

https://doi.org/10.1016/B978-0-12-820206-7.00138-4

3

4

Ferrocene and Related Metallocene Polymers

(A)

(B)

Fig. 1 Ferrocene-based polymers with (A) main-chain ferrocene-units and (B) pendant side-chain ferrocene-units.

Fig. 2 Relevant angles in sila[1]ferrocenophanes.

Scheme 1 Synthesis of sila[1]ferrocenophanes and polyferrocenylsilanes.

With some exception, like dicarba[2]ferrocenophanes, ferrocenophanes with longer bridging motifs do not undergo ringopening polymerization since they do not possess enough/any ring-strain.27–29 Importantly, while thermal ring-opening polymerizations of sila[1]ferrocenophanes do give high-molecular weight polymers (Mn  104 to 105 g mol−1), the control over the molecular weight and molecular weight distribution is somewhat limited (PDI  1.5) and the exact mechanism of the polymerization and the propagating species remains unclear.30,31 While most sila[1] ferrocenophane monomers give predominantly linear materials, a rare exception is the spiro silacyclobutylidene-[1]ferrocenophane, which was found to act as a crosslinking agent and thus enables the preparation of crosslinked polyferrocenylsilane material.32 Shortly after the thermal ring-opening polymerization of [1]ferrocenophanes was reported, the possibility of transition metal catalyzed polymerization was discovered as well.33–37 This route represents an interesting alternative, since sila[1]ferrocenophanes will undergo ring-opening polymerization in solution at ambient conditions, thus not requiring high temperatures. In addition, this route is relatively tolerant toward impurities of the monomer or the used solvent, and offers access to regioregular material when a sila[1] ferrocenophane with one methylated and one non-methylated Cp group is applied.38 Several late transition metal compounds of rhodium, palladium and platinum, such as Karstedt’s catalyst, were found to catalyze this ring-opening polymerization reaction and produce high molecular weight material.39–41 Furthermore, anionic ring-opening polymerizations of sila[1]ferrocenophanes were developed roughly at the same time and are particularly important as they, among other advantages (vide infra), offer good control over the molecular weight of the polymer and give material with very narrow molecular weight distributions (PDI 1.05 to 1.2).42–44 It is noteworthy that this was the first example of a living anionic ring-opening polymerization of a metallopolymer with the metal centers in the main chain. In addition to dimethylsila[1]ferrocenophane,43 this method has also been applied to a variety of sila[1] ferrocenophanes with various substituents on the bridging silicon atom,45–48 although in case of sila[1]ferrocenophanes with super-bulky substituents, the ferrocenophanes were shown not to undergo anionic ring-opening reactions.22 Moreover, [1]ferrocenophanes with different elements in the bridging position have been reported and some were applied in ring-opening polymerizations.11,21,26 For instance a phenylphospha[1]ferrocenophane was described to undergo anionically-initiated ring-opening polymerization, when treated with phenyl lithium. Noteworthy however, the corresponding polyferrocenylphosphine was only of low-molecular weight, while a polycondensation-type reaction between dilithioferrocene and dichlorophenylphosphine gave the material in high molecular weight (Mw up to 161,000 g mol−1) (Scheme 2).49 Thus, this is a rare example of the polycondensationtype reaction giving higher molecular weights than the related ring-opening route.

Ferrocene and Related Metallocene Polymers

5

Scheme 2 Synthesis of polyferrocenylphosphines.

Yet, the one example that is unknown, presumably due to its instability owned to a very high ring-strain, is a carba[1]ferrocenophane.21 Thus, polyferrocenylmethylene cannot be prepared via ring-opening-polymerization and was unknown until recently. However, carba[1]magnesocenophanes offer a route to circumvent this difficulty. Treatment of a Me2C[1]magnesocenophane with iron(II) bromide gives the corresponding polyferrocenylmethylene in moderate molecular weights (Mn ¼ 8300 g mol−1) (Scheme 3).50 Similarly to polyferrocenylsilanes (vide infra), polyferrocenylmethylene is an air and water stable material that exhibits redox-active iron centers.

Scheme 3 Synthesis of polyferrocenylmethylene.

As mentioned before, a wide-array of sila[1]ferrocenophanes with different substituents on the silicon atom, such as hydrogen and halide atoms, as well as alkyl, aryl, silyl, alkoxy and amino groups are accessible and undergo ring-opening polymerization. Remarkably, many of these polyferrocenylsilanes exhibit a comparably low glass transition temperature (Tg). For instance the glass transition temperature of the prototypical poly(ferrocenyldimethylsilane) (PFDMS), one of the most prominent examples, is just 33  C (306 K).13,51 This is relatively surprisingly low for a material with a bulky main-chain group like ferrocenyl, but has been suggested to be the result of the free rotatability of the Fe-Cp bonds, which results in a high conformational flexibility. Similar glass transition temperatures are observed for many other polyferrocenylsilanes with a symmetrically substituted silicon atom bearing C3 groups. On the other hand, when longer alkyl chains are bonded to the bridging silicon atom, amorphe materials with very low glass transition temperatures are observed, with a general qualitative correlation between chain length of the alkyl substituents and glass transition temperatures (Fig. 3). For instance the dihexyl substituted derivative poly(ferrocenyldi-n-hexylsilane) possesses a glass transition temperature of just −26  C (247 K) and even lower values were reported for some alkoxy-substituted derivatives, such as poly(ferrocenyldi-n-hexyloxysilane), which has a glass transition temperature of as low as −51  C (222 K).51

Fig. 3 Polyferrocenylsilanes with different alkyl substituents on the silicon atom and corresponding Tg and Mn (g mol−1) values.

Alkoxy- or amino-substituted polyferrocenylsilanes are accessible via functionalization of the corresponding dichlorosilane.52 When dichlorosila[1]ferrocenophane is treated either with a secondary amine of an alcohol in the presence of triethylamine, selective substitution on the bridging silicon atom is possible without ring-opening reactions. The obtained dialkoxy- or diamino-substituted sila[1]ferrocenophanes can then be employed in thermal ring-opening polymerizations (Scheme 4).53–56

Scheme 4 Functionalization of Cl2Si[1]ferrocenophane with alcohols and amines and subsequent ring-opening polymerization (ROP).

6

Ferrocene and Related Metallocene Polymers

As mentioned before, the best researched, prototypical polyferrocenylsilane is poly(ferrocenyldimethylsilane), which, just like many other polyferrocenylsilanes, is a thermally relatively robust material and can be processed at temperatures above 150  C (423 K) into different shapes, making it interesting for various application. As an alternative to functionalization of the silicon atom in sila[1]ferrocenphane monomers, post-polymerization functionalization is also possible (Scheme 5).40,56–59 For example, treatment of poly(ferrocenylchlorosilanes) with organolithium compounds, such as phenyl lithium, or alcohols in the presence of triethylamine, gives the corresponding phenyl- and alkoxy-substituted polymers. Moreover, introduction of a cationic alkylammonium chain makes the polymer water-soluble, and treatment of poly(ferrocenyldimethylsilane) with a tert-butyl lithium potassium tert-butoxide mixture allows lithiation of the Cp-rings.59–61 Likewise, when the polyferrocenylsilane possesses an Si-H or an alkenyl functionality, hydrosilylation chemistry can be used for further polymer functionalization,62,63 as well as thiol-ene reactions.64

Scheme 5 Selected examples of postpolymerization functionalization of polyferrocenylsilanes.

The properties of polyferrocenylsilane materials with regard to their redox characteristics have been intensively probed by cyclic voltammetry. Poly(ferrocenyldimethylsilane) for example shows two (quasi-)reversible waves separated by 0.25 V, which suggests an interaction between the iron centers. This is different from polymers with pendant ferrocene groups bond to the side-chains, which generally only show one oxidation/reduction wave.65–68 Mechanistically, it has been suggested that the first oxidation occurs at alternating iron centers and is then followed by the oxidation of the other iron centers at higher potentials.69 It was further suggested that the iron-iron interactions might be mediated by the silicon atom linkers. This is however challenged by the fact that in a recent study of the carbon analog, polyferrocenylmethylene, by cyclic voltammetry, similar two oxidation wave voltamograms were observed, suggesting similar iron-iron interactions. Furthermore, partially oxidized polyferrocenylsilanes exhibit a significant increase in electrical conductivity of 10−3 to 10−4 S cm−1, which is in the range of semiconductor materials.70 Polyferrocenylsilanes can also be combined with a large variety of other polymers in block copolymers. This is often realized by utilizing the living anionic ring-opening polymerization route. Mechanistically, the anionic ring-opening polymerization of sila[1] ferrocenophanes initiated by an organolithium compound proceeds via Si-CCp bond cleavage. This bond is particularly reactive due to the geometric deformation owing to the silyl group’s deviation from the Cp plane, as indicated by angle b (Fig. 2). This results in the addition of the carbanion to the silicon atom and lithiation of the Cp ring. Because on this mechanism, anionic ring-opening polymerization of sila[1]ferrocenophanes have several advantages, so that they have become one of the most popular methods of polymerization of sila[1]ferrocenophanes. The main advantages are (1) molecular weight control via the used amount of initiator (RLi); (2) preparation of high molecular weight linear material with low polydispersities; (3) possibility of straight forward endgroup functionalization of the polymer; (4) preparation of block copolymers.71–73 By choosing different carbanions and quenching the lithiated polymer with a suitable element halide, different endgroups can be introduced into the polymer both on the terminal silicon atom and the Cp ring (Scheme 6). Furthermore, when the polymer with a lithiated Cp endgroup is treated with another anionically polymerizable monomer such as hexamethyltrisiloxane, the corresponding block copolymer, in this case polyferrocenylsilane-b-polysiloxane, can be obtained (Scheme 6). In addition to a sequential addition of other monomers to the polyferrocenylsilane with a living anionic endgroup, polyferrocenylsilane-containing block copolymers can also be obtained when an anionically polymerized polymer with a lithiated endgroup is treated with a suitable sila[1]ferrocenophane. For instance, when styrene is anionically polymerized by butyl lithium followed by the addition of dimethylsila[1]ferrocenophane, the corresponding block copolymer polystyrene-b-polyferrocenylsilane is obtained (Scheme 7).

Ferrocene and Related Metallocene Polymers

7

Scheme 6 Synthesis of poly(ferrocenyldimethylsilane) and poly(ferrocenyldimethylsilane)-b-poly(dimethylsiloxane) block copolymer via anionic ring-opening polymerization.

Scheme 7 Synthesis of poly(ferrocenyldimethylsilane)-b-polystyrene block copolymer via anionic ring-opening polymerization.

Such block copolymers containing a polyferrocenylsilane block have been prepared with different substituents on the silicon atom, and the same techniques can also be applied to phospha[1]ferrocenophanes. Furthermore, starting from dilithiated initiators like 1,10 -dilithioferrocene, even the preparation of triblock copolymers, such as poly(dimethyl)siloxane-b-poly(ferrocenyldimethylsilane)b-poly(dimethyl)siloxane (PDMS-b-PFDMS-b-PDMS), and pentablock copolymers, such as poly(dimethyl)siloxane-b-poly (ferrocenyldimethylsilane)-b-polystyrene-b-poly(ferrocenyldimethylsilane)-b-poly(dimethyl)siloxane (PDMS-b-PFDMS-b-PSb-PFDMS-b-PDMS), with a symmetrical block structure could be achied.43 This, or similar routes, can be applied to a large variety of different monomers and give access to many different block copolymers, including ones with polymethacrylate, polyisoprene, polypeptide and polyphosphazene blocks.72,74,75 In some cases however, like for example PMMA-base block copolymers, intermediate endgroup functionalization can be necessary. For instance, when an anionically polymerized poly(ferrocenyldimethylsilane) is endgroup-functionalized to carry a terminal OH-group, this OH-group can be deprotonated and the resulting terminal alkoxy anion can polymerize dimethylaminoethylmethacrylate to give the corresponding block copolymer poly(ferrocenyldimethylsilane)-b-polydimethylaminoethylmethacrylate (PFDMS-b-PDMAEMA) (Scheme 8).

Scheme 8 Synthesis of poly(ferrocenyldimethylsilane)-b-polydimethylaminoethylmethacrylate block copolymer.

In addition, post-polymerization functionalization of polyferrocenylsilane-based block copolymers is possible and was used to introduce addition metal centers in the side chains.76–78 For example, a poly(ferrocenyldi(tert-butylethynyl)silane)-b-poly(ferrocenyldimethylsilane) block copolymer can be functionalized with cyclopentadienylnickelcarbonyl to introduce cyclopentadienylnickel centers into the side chains via alkyne complexation. Furthermore, alkylthienyl groups can be introduced into a poly(ferrocenylmethylvinylsilane) polymer by a thiol-ene reaction under photolytic conditions (Scheme 9). This is however accompanied by cross-linking side reactions.

8

Ferrocene and Related Metallocene Polymers

Scheme 9 Synthesis of poly(ferrocenylmethylvinylsilane) and subsequent thiol functionalization.

In general, many of these polyferrocenylsilane-based block copolymers with different blocks exhibit amphiphilic properties and immiscibility in many common (organic) solvents. This often results in the formation of different nanostructures which possess iron-rich domains.79–81 Such self-assembly behavior which is induced by the crystallinity of the polyferrocenylsilane block— crystallization-driven self assembly (CDSA)—was studied for a vide array of different polyferrocenylsilane-based block copolymers and has been discussed for different applications including preparation of thin films for surface modifications, nanostructured ceramics, nanolithography and preparation of complex nano structures.43,46,48,72,82–96 For instance, in polyferrocenylsilane-b-polysiloxane block copolymer the polyferrocenylsilane block is crystalline, while the polysiloxane block is not. This amphiphilic character was found to give this material striking self-assembly properties, when using a solvent which is selective for only one of the blocks in the block copolymer. Thus, the material can form micellar structures if dissolved in hexane [a selective solvent for poly(dimethylsiloxane)] (Scheme 10). For example, when poly(ferrocenyldimethylsilane)-b-poly(dimethylsiloxane) block copolymer with a block ratio of 50:300 is heated in hexane, cylindrical micelles are obtained. However, depending on the exact composition of the block copolymer and the conditions, other core-shell micellar structures with spherical, cylindrical, or platelet forms can be grown as well (Fig. 4). In some cases, spherical micelles can be obtained from cylindrical micelles by heating. In general, the exact morphologies are dependent on multiple factors and are very difficult to predict in advance.

Scheme 10 Crystallization-driven self-assembly (CDSA) of poly(ferrocenyldimethylsilane)-b-poly(dimethylsiloxane) block copolymer.

In this case, the crystalline iron-rich core of the micelles consists of the polyferrocenylsilane block while the non-crystalline polysiloxane block forms the corona. The iron-rich core of these materials makes them easily visualizable by transmission electron microscopy (Fig. 5) and their crystalline nature allows for characterization by wide angle X-ray scattering (WAXS). Interestingly, the crystallinity of the polyferrocenylsilane block seems to be a decisive factor in the formation of onedimensional-like micellar structures, as block copolymers with asymmetrically substituted silicon atoms in the polyferrocenylsilane block, which exhibit poor crystallization properties, are described to form—if any—only spherical micellar structures.45 The length of the different block, or more specifically the block ratio, seems to also play an important role in determining which kind of nanostructures are formed.97 For example, a poly(ferrocenyldimethylsilane)-b-polyisoprene block copolymer with a block ratio of 53 to 320 (ca. 1:6) assembles into cylindrical micelles in hexanes, which is a polyisoprene-selective solvent, while the same block copolymer with a block ratio of 60:30 (2:1) assembles into platelet-like micelles. In this example, it is noteworthy that the polyferrocenylsilane blocks have about the same length, but the different block sizes of the non-crystalline corona-forming blocks seem to have a great impact in the shape of the micellar structures. This can be explained with increased repulsion between the polyisoprene blocks, which lead to curvature and thus formation of cylindrical- instead of platelet-like structures. In addition to the block ratios, the overall degree of polymerization, the nature of the solvent or solvent mixture composition was also

Fig. 4 Schematic representation of crystallization-driven self-assembly (CDSA) of polyferrocenylsilane-based block copolymers. Reproduced with permissions from MacFarlane, L.; Zhao, C.; Cai, J.; Qiu, H.; Manners, I. Chem. Sci. 2021, 12(13), 4661–4682.

Ferrocene and Related Metallocene Polymers

9

Fig. 5 TEM images of poly(ferrocenyldimethylsilane)-b-poly(dimethylsiloxane) block copolymer micelles (scale bars: 500 nm). Reproduced with permissions from Gilroy, J. B.; Gädt, T.; Whittell, G. R.; Chabanne, L.; Mitchels, J. M.; Richardson, R. M.; Winnik, M. A.; Manners, I. Nat. Chem. 2010, 2(7), 566–570.

investigated and shown to have distinct impact on the morphologies of the formed nanostructures. As previously mentioned, this self-assembly behavior has been investigated and shown for a large variety of polyferrocenylsilane-based block copolymers. Polyferrocenylsilane-based block copolymers with hydrophilic blocks were found to exhibit similar self-assembly properties in water.98,99 To preserve the micellar-type nanostructures in any solvent, crosslinking of the corona-forming blocks has been shown to be a working strategy. For example, when hydrosilylation is used to cross-link corona-forming blocks with vinyl groups, these micells are persistent even in solvents which are selective for the polyferrocenylsilane core-forming blocks or for both corona- and core-forming blocks.39,100–109 The initially formed micelles can be very different in length, thus can be polydisperse. It was found that extensive sonication of these micelles can break them up into much smaller entities and addition of more block copolymers, in this regard also referred to as unimers, results in growths of the nanostructures and often formation of very long micelles with relatively low dispersities with regards to their length.110–113 Essentially, the smaller nanostructures which are obtained through sonication act as seeds and enable the selective growth of cylindrical micelles, in some cases up to 2 mm long with dispersities of around 1.0 (Fig. 6). In this case, growth can occur in both directions, thus the “seed” will be in the center of the micelle. In order to grow micelles only in one direction, the previously discussed corona cross-linking strategy was used to cap one end of the micellar seeds.114 For example, in a triblock-micelle with an ABA-type composition, with A ¼ poly(ferrocenyldimethylsilane)-b-polyisoprene block copolymer and B ¼ poly(ferrocenyldimethylsilane)-b-poly(dimethylsiloxane) block copolymer, the polyisoprene corona of the

Fig. 6 Schematic representation of the growth of cylindrical monodisperse micelles.

10

Ferrocene and Related Metallocene Polymers

Fig. 7 (A) and (B) TEM images, (C) and (D) LSCM images and (E) schematic representation of a “nano-pixel” micelles (scale bars: (A), (B): 2 mm; (C): 10 mm; (D): 5 mm). Reproduced with permissions from Hudson, Z. M.; Lunn, D. J.; Winnik, M. A.; Manners, I. Nat. Commun. 2014, 5(1), 3372.

terminal A-blocks can be cross-linked. In case of sufficiently long polyisoprene corona-forming blocks, this caps the ends of the micelles. When the central B-block is then dissolved in a suitable solvent, this leaves seeds that can essentially only grow in one direction. With this techniques, non-centrosymmetric comicelles with several different blocks were grown. One of the most striking application of this techniques was the preparation of so-called “nano-pixels,” which are essentially very long comicelles with up to 11 blocks that carry red, blue or green fluorescent dye-substituents and exhibit tunable color emissions properties (Fig. 7).108 By sequential addition of different unimer, “nano-pixel” with different color sequences could be grown. Interestingly, the “nanopixel” micelles were found to be stable over prolong periodes of time, indication that no exchange of unimer between the different micelles occurs. In recent years, also more complex micron-scale three dimensional structures, including platelet-type micelles and long windmill-type “supermicelles,” have been grown, using crystallization driven self-assembly of polyferrocenylsilane-based block copolymers.115–123 In conclusion, since the first reports of oligomeric ferrocenylsilanes in the 1960s and the breakthrough of preparing these material with high-molecular masses via ring-opening polymerization in the 1990s, these ferrocene-based polymers with the iron center in the main-chain, have developed into one of the most important class of metal-containing polymeric materials. Although related ferrocenophanes and ferrocenyl-based polymers with different main-group element linkers have been prepared, the polyferrocenylsilanes remain the most important example of this class. Many different derivatives with different substituents on the silicon atom and different substitution patterns on the Cp-ligands have been prepared and have been combined with other organic and inorganic polymers in block copolymers, which give these materials a very broad application pattern.

Ferrocene and Related Metallocene Polymers

11

Fig. 8 Functionalized ferrocenes carrying polymerizable groups.

In addition to these types of ferrocene-based polymers in which the ferrocene moiety is part of the polymer main-chain, polymers with pendant ferrocenyl groups in the side-chains are also known (vide supra) and have also been studied extensively for their (opto)electronic and biological properties and applications in catalysis and smart materials, derived from the redox activity of the ferrocene/ferricenium couple.12,124,125 In general, a pendant ferrocenyl group attached to the side chain of the polymer can have just a large impact on the overall polymer properties, as ferrocenyl groups in the main-chain. In principle, there are two synthetic routes that provide access to polymers with pendant ferrocenyl groups in the side-chains: (1) Polymerization of functionalized ferrocenes with polymerizable groups, and (2) post-polymerization functionalization of an organic or inorganic polymer with a suitable ferrocenyl reagent. For the first route, a number of different monomers have been reported, among others vinylferrocene, ferrocenylmethyl methacrylate and ferrocenyl glycidyl ether (Fig. 8).126–133 In addition, difunctionalized ferrocenes carrying two different polymerizable groups, such as vinyl ferrocenyl glycidyl ether have been reported and are particularly useful as these are unprotected orthogonal monomers with two groups which allow polymerization under different conditions, such as radical polymerization of the vinyl-group and anionic polymerization of the epoxy function (Scheme 11).

Scheme 11 Radical and anionic polymerization of vinyl ferrocenyl glycidyl ether.

In addition to “classical” organic backbone polymers, inorganic P/B-polymers with pendant ferrocene groups have also been reported.134,135 These systems are prepared by metal-catalyzed dehydropolymerization, starting from the corresponding phosphine boranes. A similar route—the intermolecular dehydrocondensation polymerization—has also been used to prepare polycarbosilazane polymers with pendant ferrocenyl groups (Scheme 12).136 Compared to, for example, the polymerization of vinyl ferrocene, the dehydropolymerization of ferrocenylphosphine borane and ferrocenylphenylsilane with para-xylylendiamin does however only give moderate molecular masses (P/B: Mn  5000 to 8000 g mol−1; N/Si: Mn  10,000 to 20,000 g mol−1) and in case of the P/B polymers moderate polydispersities (PDI 1.6 to 2.0). (A)

(B)

Scheme 12 Dehydro polymerization of (A) ferrocenyl substituted phosphine borane and (B) ferrocenyl substituted silane with amine.

12

Ferrocene and Related Metallocene Polymers

In addition to the polymerization of functionalized ferrocenes with polymerizable groups, the ferrocenyl moiety may be introduced by postpolymerization functionalization. For instance, poly(dimethylphosphazenes) can easily be deprotonated with n-butyl lithium and then functionalized with ferrocenaldehyd.137 In a similar manner, hydrosilylation can be used to functionalize polysiloxanes with Si-H moieties with alkenylferrocenes (Scheme 13).138 (A)

(B)

Scheme 13 Postpolymerization ferrocenyl functionalization of (A) polyphosphazene polymers and (B) polysiloxane polymers.

Complementary, hydrosilylation can also be used in the opposite way to introduce ferrocenyl moieties into polymers, which possess side-chains with C]C double bonds, when ferrocenylhydridosilanes are used as starting materials.139,140 In addition, thiol-ene reactions of SH-functionalized polymers with vinylferrocene,141 coupling reactions of borylated polymers with stannylferrocenes142,143 and amide-coupling of amino substituted polymer side-chains144 may also be used to introduce ferrocenyl groups into preformed polymers. Attachment of the ferrocenyl groups with longer spacers between the polymer main-chain und the ferrocenyl moieties can be of interest and have great impact on the overall properties of the material. In this regard, the functionalization of polyethylenimine with (haloalkyl)ferrocenes allows for the easy introduction of C3 to C6 carbon linkers between the nitrogen atom in the main-chain and the ferrocenyl group.145 Overall, pendant ferrocenyl moieties can have great influence on the polymer properties. For instance, in addition to increasing thermal properties and stability in some cases, (partial) oxidation of the iron atoms can drastically change the polarity of the polymer, which will greatly influence the solubility and secondary interactions.141,146 With this, otherwise hydrophobic polymeric materials can be turned hydrophilic, simply by oxidation of the ferrocenyl groups. When a corresponding polymer is attached to a surface, this principle can be used to switch the surface from hydrophobic to hydrophilic. Another interesting application is the modifiable potential of some polymers with pendant ferrocenyl groups, with regard to their charge transport properties. This has, for example, been used successfully for surface coatings of electrode materials. In addition, due to the different color of ferrocene vs. ferrocenium, polymers with pendant ferrocenyl groups exhibit electrochromic properties, hence the color can be changed by applying electric potential.147,148 Because of the interaction of the ferrocenyl groups and secondary interactions of different polymers, even more than just the two colors of ferrocene and ferrocenium—orange and blue—are possible. Finally, polymers with pendant ferrocenyl groups can be used as precursors for iron-containing ceramic materials.138 With these points in mind, many different applications in material science are possible. Overall, polymers containing ferrocene-groups are the most researched class of metallocene-based polymers and hold immense potential for various application due to their electronic properties and stimuli-responsiveness, and have been researched intensively in the last three decades. Many examples of materials with both 1,10 -ferrocenylene-groups directly incorporated into the back-bone of the polymers and ferrocenyl-groups attached to the polymer side-groups have been reported.

14.02.3 Ruthenocene-based polymers With ruthenium being iron’s “heavy analog” and sharing many of its properties, it comes as no surprise that metallapolymers with ruthenocene groups are also known. However they remain much less explored than the previously discussed iron systems. In analogy to iron, ferrocene systems respectively, polymers with ruthenocenyl groups in the side chains, as well as with ruthenocenylidene-groups incorporated into the back-bone of the polymer are known. Unlike polyferrocenylsilanes, which can

Ferrocene and Related Metallocene Polymers

13

easily be prepared from sila[1]ferrocenophanes by ring-opening polymerization, only few examples of [1]ruthenocenophanes are known, some of which were described to be thermally non-polymerizable. However, a dicarba[2]ruthenocenophane, as well as a stanna[1]ruthenocenophane were reported to undergo thermal ring-opening polymerization at temperatures of  200  C ( 473 K), and an alumina[1]ruthenocenophane was reported to undergo transition metal catalyzed ring-opening polymerization, when treated with Karstedt’s catalyst (Scheme 14).149–151

(A)

(B)

(C)

Scheme 14 Ring-opening polymerization (ROP) of (A) a dicarba[2]ruthenocenophane, (B) a stanna[1]ruthenocenophane and (C) an alumina[1]ruthenocenophane.

The dicarba[2]ruthenocenophane was obtained with and without permethylated Cp rings. Both were reported to undergo thermal ring-opening polymerization when heated to temperatures of 220  C (493 K). The non-methylated poly(ruthenocenylethylene) was described to be insoluble in common organic solvents, whereas the permethylated derivative was found to be much more soluble and could be characterized by NMR and GPC. The material was found to be of moderate molecular weight (Mn  12,000 g mol−1) and exhibited relatively high polydispersities (PDI ¼ 3.59). Surprisingly, it was found to be less easy to oxidize than the polyferrocenylsilanes or polyferrocenylmethylene. Cyclic voltammetry measurements showed only one irreversible oxidation wave at 0.60 V. Furthermore, the stanna[1]ruthenocenophane was also found to undergo ring-opening polymerization at slightly lower temperatures of 200  C (473 K) and give high molecular weight polymer (Mn ¼ 2.7  105 g mol−1) with comparably high polydispersities (PDI ¼ 2.28). Surprisingly, compared to the afore discussed polyferrocenylsilanes, the material exhibits a remarkably high glass transition temperature of Tg ¼ 221  C (494 K). In addition, metallapolymers with pendant ruthenocenyl groups have been prepared on similar routes as their iron analogs (vide supra). For example, polymethacrylates with ruthenocenyl groups have been prepared by ATRP and RAFT polymerization, starting from 2-(methacryloyloxy)ethyl ruthenocenecarboxylate. The poly(2-(methacrylolyoxy)ethyl ruthenocenecarboxylate), which was obtained via RAFT polymerization, was described to have a high molecular weight (Mn ¼ 14,200 to 18,700 g mol−1) and low narrow polydispersity (PDI ¼ 1.15). Using chain extension techniques, it was also possible to prepare diblock copolymers with amphiphilic properties. Similar to afore discussed ferrocene systems, self-assembly behavior into micelle structures could be observed in some cases. Additionally, ruthenocenyl glycidyl ether, which is prepared thru functionalization of ruthenocene via the aldehyde and alcohol as intermediates, was copolymerized with ethylene oxide by treatment with cesium benzyloxyethoxide as anionic initiator. The resulting water-soluble material was of moderate molecular weight (Mn ¼ 1900 to 4300 g mol−1) but exhibited narrow polydispersity (PDI ¼ 1.10 to 1.15), with the amount of ruthenocenyl glycidyl ether varied between 2% and 8%. Furthermore, a terpolymer starting from ethylene oxide, ruthenocenyl glycidyl ether and previously discussed ferrocenyl glycidyl ether could be prepared via the same route. This material offered similar molecular weights (Mn ¼ 2100 to 4000 g mol−1) and similar polydispersities (PDI ¼ 1.09 to 1.15) (Scheme 15).

14

Ferrocene and Related Metallocene Polymers

Scheme 15 Copolymerization of ethyleneoxide with ruthenocenyl glycidyl ether and ferrocenyl glycidyl ether.

Electrochemical characterization of these ruthenocenyl-containing copolymers by cyclic voltammetry revealt a non-reversible redox behavior of the ruthenium centers, as is typical for ruthenocene-based materials, but a reversible oxidation peak for the ferrocenyl groups at 0.6 V.

14.02.4 Cobaltocene-based polymers With cobalt being a group 9 element, the corresponding metallocene—cobaltocene—is a paramagnetic 19 electron complex, whereas the corresponding cation—cobaltocenium—is an 18 electron system and thus is isoelectronic to ferrocene. Therefore, cobaltocene for example exhibits a relatively low ionization potential of just 5.56 eV,152,153 thus makes functionalizations of the neutral metallocenes much more challenging than in case of ferrocenes and polymers with cationic cobaltocenium groups are much more common than their neutral analogs. In many cases, the cationic nature of these metallocene moieties give the corresponding polymers interesting solubility properties.154 Both polymers with pendant cobaltocenium groups bond to the side-chains and cobaltocenium groups in the main-chain are known, similar to the ferrocene-based systems, although the latter are less common. The first reports of cobaltocenophanes and the corresponding cationic cobaltocenophanium compounds date back to the late 1990s, and include complexes with C[2], C[3] and Si[2] bridging motifs, but no ring-opening polymerization was initially described.155–159 Only in 2009, the first report of a ring-opening polymerization of a dicarba[2]cobaltocenophane compound was published (Scheme 16).154

Scheme 16 Ring-opening polymerization of a dicarba[2]cobaltocenophane (A− ¼ Cl− or NO3 − ).

Interestingly, the corresponding dicarba[2]cobaltocenophanium cation in form of its hexafluoro-phosphate did not undergo photoinduced ring-opening polymerization. This can be explained by the fact that the metal-Cp bond is stronger and thus shorter in the cationic 18 electron complex than in the neutral 19 electron complex, which has one unpaired electron of anti-bonding character and means that the cationic cobaltocenophanium possesses less ring strain than the neutral cobaltocenophane. This is also indicated by the fact that the neutral dicarba[2]cobaltocenophane exhibits a larger dihedral angle of the Cp planes of a ¼ 27.1(4) , while the dihedral angle of the Cp planes in the cationic dicarba[2]cobaltocenophanium is just a ¼ 21.4(2) (for the chloride salt). Notably, the molecular weight of the polycobaltocenylethylene (PCE) depends significantly on the reaction conditions of the ring-opening polymerization. For instance, treatment of dicarba[2]cobaltocenophane with a lithium cyclopentadienide at ambient conditions with or without UV photolysis gave only low molecular weight material (DPn ca. 5 to 9), while thermal ring-opening polymerization at 140  C (413 K) gave material of significantly higher molecular weight (Mw ¼ 55,000 g mol−1). In all cases, the oligomeric/polymeric crude product was oxidized with aqueous ammonium chloride and ammonium nitrate solution in air, to obtain the cobaltocenium material. This photocontrolled ring-opening polymerization can also be adopted for the synthesis of a hetero bimetallic poly (ferrocenyldimethylsilane)-b-poly(cobaltoceniumylethylene) block copolymer.160 Treatment of dimethylsila[1]ferrocenophane with sodium cyclopentadienide (NaCp) as an anionic initiator and subsequent photolysis results in anionic ring-opening polymerization and formation of a sodium cyclopentadienyl-terminated polyferrocenylsilane. Successive addition of the dicarba[2] cobaltocenophane monomer gives the corresponding block copolymer in which the cobalt centers were oxidized with methanol and ammonium triflate in air and the block copolymer was isolated as a poly cation in form of its triflate salt (Scheme 17).

Ferrocene and Related Metallocene Polymers

15

Scheme 17 Sequential ring-opening polymerization of a sila[1]ferrocenophane and dicarba[2]cobaltocenophane.

With this protocol, block copolymers with an iron cobalt ratio of 50:50 and 42:75, as determined by 1H NMR spectroscopy, could be obtained. As discussed before, polyferrocenylsilanes are oxidizable polymers that exhibit two distinct quasi-reversible oxidation/ reduction waves in the cyclic voltammogram. Cyclic voltammetry studies of the corresponding poly(ferrocenyldimethylsilane)b-poly(cobaltoceniumylethylene) block copolymer with a 50:50 block ratio revealed a third redox wave at a potential of −1.27 V corresponding to the cobaltocenyl moieties, in addition to the typical two oxidation waves of the ferrocenyl moieties. This is comparable to the reduction potential which was previously observed for the poly(cobaltoceniumylethylene) homo polymer (ca. −1.3 V). Thus, this block copolymer has remarkable electronic properties and can essentially exhibit three different electronic states: (1) all neutral iron(II) and cobalt(II) centers; (2) neutral iron(II) centers and positively charged cobalt centers; and (3) all positively charged iron(III) and cobalt(III) centers. Thermal analysis of the material revealed glass transition temperatures similar to those of poly(ferrocenyldimethylsilane) homopolymer (Tg ¼ 32 to 34  C). As discussed before, block copolymers with one crystalline polyferrocenylsilane block and one non-crystalline (solubilizing) block like polyisoprene or poly(dimethylsiloxane) possess remarkable self-assembly properties as they undergo crystallizationdriven self-assembly in solvents selective for only one of the blocks and form micellar structures. Due to the positively charged nature of poly(cobaltoceniumylethylene), it dissolves in polar solvents such as water or methanol, which polyferrocenylsilanes do not. Therefore, a water methanol mixture is a selective solvent for the poly(cobaltoceniumylethylene) block and self assembly into spherical micelles was observed in this solvent mixture. In addition to this, micelles with different morphologies from linear to branched and tapered micelles could be grown, depending on the solvent polarity and temperature.161 In addition, solution self-assembly was observed for poly(cobaltoceniumylethylene) homo polymer with anionic surfactants, such as N-palmitoyl162 L-alanine and N-palmitoyl-D-alanine. Similar to the before discussed ferrocene-based materials, there are not only polymers with cobaltocenium moieties in the main chain, but also with cobaltocenium groups bond to the side chains. In the following, some selected examples will be discussed. Polymers with pendant cobaltocenium groups are much less-researched compared to their iron analogs, among other things due to the greater difficulty in preparing substituted cobatocenium complexes, thus challenging monomer synthesis.162–167 Due to their cationic nature, side-chain-substitution of polymers with cobaltocenium groups gives the polymers interesting solubility properties. For example, hexafluorophosphates of polyacrylates carrying cationic cobaltocenium moieties are soluble in polar solvents such as water or acetone, but are insoluble in nonpolar organic solvents, such as chloroform.167 As with polymers carrying pendant ferrocene units, the synthesis of side-chain-cobaltocenium-functionalized polymers can either start from a polymerizable monomer or by post-polymerization functionalization. The latter was reported in 2010 for a poly(tert-butyl acrylate)-b-poly(2-hydroxyethyl acrylate) block copolymer, prepared by atom-transfer radical polymerization, where the terminal hydroxy groups were partially substituted with a cobaltocenium acyl chloride in the presence of triethylamine (Scheme 18).167 It is worth mentioning that the synthesis of this mono-substituted cobaltocenium acyl chloride was quite challenging and required extensive purification, yielding the compound in only 20%. Furthermore, the degree of cobaltocenium functionalization with regards to the hydroxy groups was about 70%, as determined by 1H NMR, most likely limited by steric hindrance.

Scheme 18 Side-chain functionalization of poly(tert-butyl acrylate)-b-poly(2-hydroxyethyl acrylate) with cobaltocenium.

As mentioned before, cationic cobaltocenium groups make the polymer soluble in polar solvents. When only one block of a block copolymer is functionalized accordingly, this can be used for solution self-assembly, similar to what discussed before for

16

Ferrocene and Related Metallocene Polymers

polymers with cobaltocenium groups in the main-chain. When the poly(tert-butyl acrylate)-b-poly(2-cobaltoceniumethyl acrylate) block copolymer was dissolved in an acetone/water mixture, with water being a selective solvent for the cationic cobaltocenium functionalized block, micellar aggregates with a vesicle-type morphology could be obtained. These were not very uniform however and exhibited diameters of 20 to 360 nm according to transmission electron microscopy images and dynamic light scattering experiments. On the other hand, when the poly(tert-butyl acrylate)-b-poly(2-cobaltoceniumethyl acrylate) block copolymer was dissolved in an acetone/chloroform mixture, with chloroform being a selective solvent for the neutral poly(tert-butyl acrylate) block, self-assembly into micelles with a nanotubular morphology was observed. These nanotubular micelles possessed a relatively uniform diameter of about 50 nm, and an electron-rich shell, most likely composed of the poly(2-cobaltoceniumethyl acrylate) block. In addition to cobaltocenium post-polymerization functionalization of polymers as described above, the preparation of such materials from polymerizable cobaltocenium-based monomers, for instance by reversible addition–fragmentation chain-transfer (RAFT) or ring-opening metathesis polymerization (ROMP), has also been reported.168–170 The latter was shown for a cobaltocenium-substituted norbornene monomer in combination with Grubbs’ third-generation olefin metathesis catalyst (Scheme 19A). (A)

(B)

Scheme 19 (A) Polymerization of cobaltocenium-containing norbornene monomer via ring-opening metathesis polymerization and (B) subsequent synthesis of block copolymer.

Using this route, high-molecular weight polymer with Mn up to 167,000 g mol−1 and low polydispersities of ca. 1.1 could be prepared. Furthermore, the polymerization reaction was found to be very rapid, with monomer conversion of almost 100% observed after just 10 min. Modification of this route allowed for the preparation of a block copolymer through chain extension via a one-pot two-step protocol with the subsequent addition of norbornene carboxylic acid (Scheme 19B). In a similar approach, a ferrocene cobaltocenium diblock copolymer could be obtained starting from ferrocenium and cobaltocenium-containing norbornene monomers.170 This material showed fragments up to about 10,000 Da in the MALDI-ToF mass spectrum and possessed very rich electrochemical properties, as indicated by four quasi reversible oxidation/reduction waves in the cyclic voltammogram. Another method that was reported, was the reversible-addition-fragmentation chain-transfer polymerization (RAFT) of a methacrylate-functionalized cobaltocenium hexafluorophosphate monomer (Scheme 20).169

Scheme 20 Reversible-addition-fragmentation chain-transfer polymerization (RAFT) of methacrylate-functionalized cobaltocenium.

Ferrocene and Related Metallocene Polymers

17

While the preparation of the monomer is an elaborate multi-step reaction sequence, the polymerization gave high-molecular weight material with Mn of 19,000 to 22,000 g mol−1, as determined by NMR and GPC. Similar to the cationic cobaltoceniumcontaining polymers discussed before, this material is a cationic hydrophilic polyelectrolyte, as it was reported to be insoluble in many nonpolar organic solvents such as dichloromethane and acetone, but soluble in polar solvents such as acetonitrile, methanol, dimethylformamide and dimethyl sulfoxide. Noteworthy, the compound, which was described to be a yellow to green solid—the typical color of cobaltocenium—was also characterized by UV-Vis and cyclic voltammetry. The latter shows a not fully reversible redox behavior of the material. As before with ring-opening metathesis polymerization, the reversible-addition-fragmentation polymerization technique also allows to prepare heterobimetallic ferrocene cobaltocenium containing block copolymers by sequential polymerization with high molecular weights Mn of up to 71,900 g mol−1, as determined by NMR (Scheme 21).171

Scheme 21 Reversible-addition-fragmentation chain-transfer polymerization (RAFT) of methacrylate-functionalized cobaltocenium and ferrocene monomers for the preparation of block copolymers.

This material possessed solution self-assembly properties, similar to what was described before. For example a block copolymer with a cobaltocenium: ferrocene block ratio of 62:122 self-assembled in a dimethylformamide/acetonitrile solvent mixture into spherical micelles with diameters of 45 to 65 nm. Furthermore, the counteranions to the cationic cobaltocenium moieties can play an important role with regards to the solubility and conformational properties of the materials, and ion exchange reactions have been studied.172

14.02.5 Nickelocene-based polymers In 2014, Manners et al. extended the concept of ring-opening polymerization of metallocenophanes to nickelocenophanes.172–177 With nickel being a late group 10 transition metal and nickelocenes therefore being 20 electron systems with two unpaired electrons in anti-bonding orbitals, these metallcocenes are paramagnetic and exhibit a triplet ground state. Similar to the afore discussed cobaltocenes, the partial occupation of these anti-bonding orbitals (with regard to the Ni-Cp bond) weaken the metal-Cp bond and result in an elongation of this bond. This is very vivid when comparing the metal-Cpcentroid bonds in unsubstituted ferrocene (18 e−), cobaltocene (19 e−) and nickelocene (20 e−): Fe-Cp: 166.06(11) pm178; Co-Cp: 166.71(1), 171.20(2), 172.23(2), 172.52(2) ppm179; Ni-Cp: 181.67(13), 181.77(4) pm180 (multiple values given due to non-equivalent bonds in the metallocene and/or disordering of the Cp rings in the crystal). The longer metal-Cp bonds in case of nickelocenes, and nickelocenophanes respectively, have an impact on possible ansa-bridging motifs. For instance, attempts to prepare a dicarba[2]nickelocenophane with an ethylene linker were reported to be unsuccessful,173 while the cobalt and iron derivatives are known. However, the preparation of a carba[3] nickelocenophane was successful. This molecule exhibits a dihedral angle a of the Cp planes (Fig. 2) of 16.6 , which is significantly larger than the a angle in the corresponding carba[3]cobaltocenophane (12.0 ) and carba[3]ferrocenophane (10.3 ), and was reported to be the largest of any carba[3]metallocenophane at the time. This is consistent with longer Ni-Cpcentroid bonds in the carba[3]nickelocenophane (180.38(4), 180.53(4) pm) than in the carba[3]cobaltocenophane (171.59(7), 171.89(6) pm) and carba[3]ferrocenophane (163.62(2), 163.78(2) pm) (multiple values given due to two non-equivalent M-Cp bonds in the metallocenophanes in the crystal), which suggests that the carba[3]nickelocenophane173 possesses a higher ring-strain than its cobalt(II)155 and iron(II)181 analogs. The elongation of the metal-Cp distance in combination with the induced ring-strain, results in a relatively weak bond, which was shown by stochiometric reactions of the carba[3]nickelocenophane with different donor molecules, such as phosphines and N-heterocyclic carbenes, that resulted in coordination of the donor molecule to the metal center. Attempts of thermally-induced ring-opening polymerization of carba[3]nickelocenophane gave only an insoluble material, which was speculated to be of an oligomeric/polymeric and possibly cross-linked nature, but further characterization was unsuccessfully. However, when carba[3]nickelocenophane was stirred in pyridine, a color change from blue to green was observed, which indicated coplanarization of the Cp rings, thus a ring-opening reaction, and soluble polymeric poly(nickelocenylpropylene) was obtained (Scheme 22).

18

Ferrocene and Related Metallocene Polymers

Scheme 22 Synthesis of polynickelocenylpropylene by pyridine-induced ring opening polymerization.

The material was characterized by paramagnetic 1H NMR, MALDI-ToF mass spectrometry and static light scattering, which suggested the material to be cyclic and contain up to 33 repeat units (MALDI-ToF), more specifically possess a molecular weight of ca. 40,000 g mol−1 (SLS). Mechanistically, the ring-opening polymerization reaction is believed to proceed via metal-Cp bond cleavage, which is promoted by the pyridine coordination to the metal center and Cp-backbiting produces the predominantly observed cyclic material. A detailed follow-up study of the ring-opening polymerization of the carba[3]nickelocenophane by paramagnetic 1H NMR in deuterated pyridine revealed a concentration dependent equilibrium between the carba[3]nickelocenophane monomer and poly(nickelocenylpropylene) polymer.174 A greater monomer conversion and therefore higher yields of the polymer were observed at higher concentrations. Furthermore, the equilibrium was found to be temperature dependent, with higher temperatures favoring the monomer and lower temperature favoring the polymer. This is an unusual example of a reversible ring-opening polymerization of a metallocenophane. Such reversible ring-opening polymerizations were previously only known for organic monomers, such as THF or cyclopentene.182,183 Entropic effects are often discussed to be a relevant factor in such systems, in-line with the observed temperature dependence. Furthermore, it was found that the ring-opening polymerization occurs significantly faster in coordinative solvents and was extremely slow in benzene or toluene. Subsequent studies into a vide array of [n]nickelocenophanes, exhibiting different tricarba[3] and disiloxa[3] bridging motifs, illustrated that the non-ring strained disiloxa[3]nickelocenophanes with coplanar Cp rings do not undergo ring-opening polymerization under the conditions described above.175 However, a later study showed that a tetracarba[4]nickelocenophane and a disila[2]nickelocenophane can undergo ring-opening polymerization in pyridine solutions (Fig. 9), although significantly longer reaction times are needed in case of the non-ring-strained tetracarba[4]-bridged system.176

Fig. 9 Polymerizable [n]nickelocenophanes.

Furthermore, the materials were characterized by SQUID magnetometry measurements that revealed paramagnetic character at higher temperatures and deviation from the Curie–Weiss behavior due to zero-field splitting at low temperatures. When combinations of different monomers were employed in the pyridine-induced ring-opening polymerization, such as a combination of the tricarba[3]nickelocenophane and the tetracarba[4]nickelocenophane, or a combination of the tetracarba[4]nickelocenophane and the disila[2]nickelocenophane, the corresponding copolymers could be obtained. Extending this concept to a mixture of tricarba[3] nickelocenophane and dicarba[2]cobaltocenophane, the heterobimetallic copolymer was obtained, as suggested by MALDI-ToF mass spectrometry. Due to the 20 electron nature of the nickel center in these nickelocene-based polymers, they possess interesting redox properties. The oxidation behavior of poly(nickelocenylpropylene) was studied by cyclic voltammetry, which revealed two single-electron redox processes. Thus, both the species with monocationic nickel(III) centers as well as the species with dicationic nickel(IV) centers could be observed. In summary, nickelocene-based polymers with the nickelocene moieties in the main chain are the latest addition to main-chain metallocene-based polymers, prepared by ring-opening polymerization. It is particularly interesting that [3]- and [4]nickelocenophanes with very small to virtually no ring-strain do also undergo ring-opening polymerizations. This is a distinct mechanistic difference to the previously reported and discussed ferrocene-systems.

14.02.6 Arenocene-based polymers In addition to varying the central metal atom and the bridging motif in metallocenophanes, the cyclopentadienide groups can also be substituted by other arenes, giving rise to [n]arenocenophanes. One of the earliest examples of such a compound that was later also used in ring-opening polymerizations was a sila[1]chromarenophane.184–186 This compound was synthesized via a similar route as its iron analog, sila[1]ferrocenophane, starting from dilithiated bis(benzene)chromium by treatment with the corresponding chlorosilane. Although initial investigations into the compounds reactivity did not show any thermally or anionically initiatable ring-opening reactions,184 it was later demonstrated that a corresponding polymer, with a molecular mass of Mn  4100 g mol−1, could be obtained by treatment with Karstedt’s catalyst at room temperature (Scheme 23).187

Ferrocene and Related Metallocene Polymers

19

Scheme 23 Synthesis of polyarenochromocenylsilane by ring-opening polymerization.

This represented the first example of a metallopolymer with bis(arene)chromium and organosilane groups in the main-chain. Shortly after, a poly(troticenylsilane) was reported, which could be obtained by ring-opening polymerization of a sila[1] cycloheptatrienyl-cyclopentadienyl-titanocenophane, by treatment with tris(triethylphosphine)platinum (Scheme 24).

Scheme 24 Synthesis of polycycloheptatrienylcyclopentadienyltitanocenylsilane by ring-opening polymerization.

This material could be obtained with molecular weights of about 6000 Da, repeat units of up to n ¼ 23 respectively, according to MALDI-ToF measurements.188 Mechanistically, the ring-opening reaction is believed to proceed via the Si-cycloheptatrienyl bond cleavage, as suggested by stochiometric reactions between the titanocenophane and tris(triethylphosphine)platinum, which yielded the corresponding insertion product. Furthermore, the concept of ring-opening polymerizations of [1]arenocenophanes was also adopted to sila[1]vanadoarenocenophanes.189 Similar to the above described chromium derivative, a sila[1]vanadoarenophane can undergo ring-opening polymerization when treated with Karstedt’s catalyst (Scheme 25).

Scheme 25 Synthesis of polyarenovanadocenylsilane by ring-opening polymerization.

Unlike in case of the afore discussed chromium analog, the ring-opening polymerization of the sila[1]vanadoarenophane required temperatures of 323 K (50  C). The obtained polymeric material was characterized by small-angle X-ray scattering (SAXS) which allowed to estimate a molecular wight of Mw ¼ 28,000 g mol−1. Overall, the ring-opening polymerization of arenocenophanes is still a new and underdeveloped research field, compared to the related ferrocenophane systems, but offers much potential for a variety of different metal-containing polymers in the future.

14.02.7 Summary In general, metallopolymers are considered to be a very important class of functional materials, which combine the functionality of a metal center with the properties of organic or inorganic polymers, and they have seen emerging applications in the past few decades. This chapter presents some key concepts and selected examples of different polymers containing metallocene groups, such as ferrocene, ruthenocene, cobaltocene or nickelocene. Such metallocene-based polymers have emerged tremendously in the last three decades, due to synthetic breakthrough. Without a doubt, one of the key innovations was the discovery that sila[1] ferrocenophanes can undergo ring-opening polymerizations and give the corresponding polyferrocenylsilanes in high molecular weights and with low polydisperties. This method has since been adopted for a variety of related metallocene- and arenocene-based systems. In addition, new synthetic routes to organic and inorganic polymers with pendant metallocene groups bonded to the side-chains have also been established in the recent past and have advanced the field of metallopolymers even further. Due to the afore mentioned synthetic breakthroughs and the ability to prepare these material in larger scale, in-depth studies of the properties became possible and have led to a rising interest in these functional metal-containing materials.

Acknowledgment Inga-Alexandra Bischoff is thanked for proof-reading of this chapter. Funding by Saarland University, Deutsche Forschungsgemeinschaft, DFG (Emmy Noether program SCHA1915/3-1), and Fonds der Chemischen Industrie, FCI, is gratefully acknowledged.

20

Ferrocene and Related Metallocene Polymers

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70.

Feldman, D. Des. Monomers Polym. 2008, 11 (1), 1–15. Whittell, G. R.; Hager, M. D.; Schubert, U. S.; Manners, I. Nat. Mater. 2011, 10 (3), 176–188. Eloi, J.-C.; Chabanne, L.; Whittell, G. R.; Manners, I. Mater. Today 2008, 11 (4), 28–36. Vorotyntsev, M. A.; Vasilyeva, S. V. Adv. Colloid Interf. Sci. 2008, 139 (1), 97–149. Neuse, E. W.; Rosenberg, H. J. Macromol. Sci. Part C 1970, 4 (1), 1–145. Gates, D. P. Annu. Rep. Sect. Inorg. Chem. 2005, 101, 452–471. Yan, Y.; Zhang, J.; Ren, L.; Tang, C. Chem. Soc. Rev. 2016, 45 (19), 5232–5263. Wang, Y.; Astruc, D.; Abd-El-Aziz, A. S. Chem. Soc. Rev. 2019, 48 (2), 558–636. Manners, I. Polyhedron 1996, 15 (24), 4311–4329. Hailes, R. L. N.; Oliver, A. M.; Gwyther, J.; Whittell, G. R.; Manners, I. Chem. Soc. Rev. 2016, 45 (19), 5358–5407. Bellas, V.; Rehahn, M. Angew. Chem. Int. Ed. 2007, 46 (27), 5082–5104. Pietschnig, R. Chem. Soc. Rev. 2016, 45 (19), 5216–5231. Manners, I. Chem. Commun. 1999, (10), 857–865. Hudson, R. D. A. J. Organomet. Chem. 2001, 637–639, 47–69. Rider, D. A.; Manners, I. Polym. Rev. 2007, 47 (2), 165–195. Foucher, D. A.; Tang, B. Z.; Manners, I. J. Am. Chem. Soc. 1992, 114 (15), 6246–6248. Osborne, A. G.; Whiteley, R. H. J. Organomet. Chem. 1975, 101 (2), C27–C28. Butler, I. R.; Cullen, W. R.; Ni, J.; Rettig, S. J. Organometallics 1985, 4 (12), 2196–2201. Bishop, J. J.; Davison, A.; Katcher, M. L.; Lichtenberg, D. W.; Merrill, R. E.; Smart, J. C. J. Organomet. Chem. 1971, 27 (2), 241–249. Fischer, A. B.; Kinney, J. B.; Staley, R. H.; Wrighton, M. S. J. Am. Chem. Soc. 1979, 101 (22), 6501–6506. Musgrave, R. A.; Russell, A. D.; Manners, I. Organometallics 2013, 32 (20), 5654–5667. Musgrave, R. A.; Russell, A. D.; Whittell, G. R.; Haddow, M. F.; Manners, I. Organometallics 2015, 34 (5), 897–907. Berenbaum, A.; Jäkle, F.; Lough, A. J.; Manners, I. Organometallics 2001, 20 (5), 834–843. Pudelski, J. K.; Foucher, D. A.; Honeyman, C. H.; Lough, A. J.; Manners, I.; Barlow, S.; O’Hare, D. Organometallics 1995, 14 (5), 2470–2479. Green, J. C. Chem. Soc. Rev. 1998, 27 (4), 263–272. Manners, I.; Vogel, U. Modern Cyclophane Chemistry; John Wiley & Sons, 2004; pp 415–433. Herbert, D. E.; Mayer, U. F. J.; Gilroy, J. B.; López-Gómez, M. J.; Lough, A. J.; Charmant, J. P. H.; Manners, I. Chem. Eur. J. 2009, 15 (45), 12234–12246. Gilroy, J. B.; Russell, A. D.; Stonor, A. J.; Chabanne, L.; Baljak, S.; Haddow, M. F.; Manners, I. Chem. Sci. 2012, 3 (3), 830–841. Nelson, J. M.; Rengel, H.; Manners, I. J. Am. Chem. Soc. 1993, 115 (15), 7035–7036. Jäkle, F.; Rulkens, R.; Zech, G.; Massey, J. A.; Manners, I. J. Am. Chem. Soc. 2000, 122 (17), 4231–4232. Baumgartner, T.; Jäkle, F.; Rulkens, R.; Zech, G.; Lough, A. J.; Manners, I. J. Am. Chem. Soc. 2002, 124 (34), 10062–10070. MacLachlan, M. J.; Lough, A. J.; Geiger, W. E.; Manners, I. Organometallics 1998, 17 (9), 1873–1883. Ni, Y.; Rulkens, R.; Pudelski, J. K.; Manners, I. Macromol. Rapid Commun. 1995, 16 (9), 637–641. Reddy, N. P.; Yamashita, H.; Tanaka, M. J. Chem. Soc. Chem. Commun. 1995, (22), 2263–2264. Temple, K.; Jäkle, F.; Sheridan, J. B.; Manners, I. J. Am. Chem. Soc. 2001, 123 (7), 1355–1364. Jaska, C. A.; Bartole-Scott, A.; Manners, I. Dalton Trans. 2003, (21), 4015–4021. Bartole-Scott, A.; Resendes, R.; Manners, I. Macromol. Chem. Phys. 2003, 204 (10), 1259–1268. Gómez-Elipe, P.; Resendes, R.; Macdonald, P. M.; Manners, I. J. Am. Chem. Soc. 1998, 120 (33), 8348–8356. Cyr, P. W.; Rider, D. A.; Kulbaba, K.; Manners, I. Macromolecules 2004, 37 (11), 3959–3961. Power-Billard, K. N.; Manners, I. Macromolecules 2000, 33 (1), 26–31. Temple, K.; Dziadek, S.; Manners, I. Organometallics 2002, 21 (21), 4377–4384. Rulkens, R.; Ni, Y.; Manners, I. J. Am. Chem. Soc. 1994, 116 (26), 12121–12122. Ni, Y.; Rulkens, R.; Manners, I. J. Am. Chem. Soc. 1996, 118 (17), 4102–4114. Lammertink, R. G. H.; Hempenius, M. A.; Manners, I.; Vancso, G. J. Macromolecules 1998, 31 (3), 795–800. Massey, J. A.; Temple, K.; Cao, L.; Rharbi, Y.; Raez, J.; Winnik, M. A.; Manners, I. J. Am. Chem. Soc. 2000, 122 (47), 11577–11584. Rider, D. A.; Cavicchi, K. A.; Power-Billard, K. N.; Russell, T. P.; Manners, I. Macromolecules 2005, 38 (16), 6931–6938. Acikgoz, C.; Hempenius, M. A.; Vancso, G. J.; Huskens, J. Nanotechnology 2009, 20 (13), 135304. Gwyther, J.; Manners, I. Polymer 2009, 50 (23), 5384–5389. Withers, H. P.; Seyferth, D.; Fellmann, J. D.; Garrou, P. E.; Martin, S. Organometallics 1982, 1 (10), 1283–1288. Winter, T.; Haider, W.; Schießer, A.; Presser, V.; Gallei, M.; Schäfer, A. Macromol. Rapid Commun. 2021, 42 (9), 2000738. Kulbaba, K.; Manners, I. Macromol. Rapid Commun. 2001, 22 (10), 711–724. Nguyen, P.; Stojcevic, G.; Kulbaba, K.; MacLachlan, M. J.; Liu, X.-H.; Lough, A. J.; Manners, I. Macromolecules 1998, 31 (18), 5977–5983. Wang, J.-J.; Wang, L.; Chen, T.; Wang, X.-J.; Wang, C.-L.; Dong, X.-C.; Yang, J.-H. Eur. Polym. J. 2006, 42 (4), 843–848. Wang, J.-J.; Wang, L.; Wang, X.-J.; Chen, T.; Yu, H.-J.; Wang, W.; Wang, C.-L. Mater. Lett. 2006, 60 (11), 1416–1419. Calleja, G.; Carré, F.; Cerveau, G.; Labbé, P.; Coche-Guérente, L. Organometallics 2001, 20 (20), 4211–4215. Zechel, D. L.; Hultzsch, K. C.; Rulkens, R.; Balaishis, D.; Ni, Y.; Pudelski, J. K.; Lough, A. J.; Manners, I.; Foucher, D. A. Organometallics 1996, 15 (8), 1972–1978. Cheng, Z.; Ren, B.; Zhao, D.; Liu, X.; Tong, Z. Macromolecules 2009, 42 (7), 2762–2766. Wang, Z.; Lough, A.; Manners, I. Macromolecules 2002, 35 (20), 7669–7677. Hempenius, M. A.; Robins, N. S.; Lammertink, R. G. H.; Vancso, G. J. Macromol. Rapid Commun. 2001, 22 (1), 30–33. Hempenius, M. A.; Brito, F. F.; Vancso, G. J. Macromolecules 2003, 36 (17), 6683–6688. Jäkle, F.; Wang, Z.; Manners, I. Macromol. Rapid Commun. 2000, 21 (18), 1291–1296. Hilf, S.; Cyr, P. W.; Rider, D. A.; Manners, I.; Ishida, T.; Chujo, Y. Macromol. Rapid Commun. 2005, 26 (12), 950–954. Liu, X.-H.; Bruce, D. W.; Manners, I. J. Organomet. Chem. 1997, 548 (1), 49–56. Chabanne, L.; Pfirrmann, S.; Lunn, D. J.; Manners, I. Polym. Chem. 2013, 4 (7), 2353–2360. Rulkens, R.; Lough, A. J.; Manners, I.; Lovelace, S. R.; Grant, C.; Geiger, W. E. J. Am. Chem. Soc. 1996, 118 (50), 12683–12695. Foucher, D. A.; Honeyman, C. H.; Nelson, J. M.; Tang, B. Z.; Manners, I. Angew. Chem. Int. Ed. Eng. 1993, 32 (12), 1709–1711. Nguyen, M. T.; Diaz, A. F.; Dement’ev, V. V.; Pannell, K. H. Chem. Mater. 1994, 6 (7), 952–954. Wang, X.-J.; Wang, L.; Wang, J.-J. J. Polym. Sci. Part B Polym. Phys. 2004, 42 (12), 2245–2253. Barrière, F.; Camire, N.; Geiger, W. E.; Mueller-Westerhoff, U. T.; Sanders, R. J. Am. Chem. Soc. 2002, 124 (25), 7262–7263. Rulkens, R.; Resendes, R.; Verma, A.; Manners, I.; Murti, K.; Fossum, E.; Miller, P.; Matyjaszewski, K. Macromolecules 1997, 30 (26), 8165–8171.

Ferrocene and Related Metallocene Polymers 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142.

21

Power-Billard, K. N.; Wieland, P.; Schäfer, M.; Nuyken, O.; Manners, I. Macromolecules 2004, 37 (6), 2090–2095. Kloninger, C.; Rehahn, M. Macromolecules 2004, 37 (5), 1720–1727. Massey, J. A.; Winnik, M. A.; Manners, I.; Chan, V. Z.-H.; Ostermann, J. M.; Enchelmaier, R.; Spatz, J. P.; Möller, M. J. Am. Chem. Soc. 2001, 123 (13), 3147–3148. Wang, X.-S.; Winnik, M. A.; Manners, I. Macromol. Rapid Commun. 2002, 23 (3), 210–213. Korczagin, I.; Hempenius, M. A.; Vancso, G. J. Macromolecules 2004, 37 (5), 1686–1690. Tanabe, M.; Vandermeulen, G. W. M.; Chan, W. Y.; Cyr, P. W.; Vanderark, L.; Rider, D. A.; Manners, I. Nat. Mater. 2006, 5 (6), 467–470. Chabanne, L.; Matas, I.; Patra, S. K.; Manners, I. Polym. Chem. 2011, 2 (11), 2651–2660. Lunn, D. J.; Boott, C. E.; Bass, K. E.; Shuttleworth, T. A.; McCreanor, N. G.; Papadouli, S.; Manners, I. Macromol. Chem. Phys. 2013, 214 (24), 2813–2820. Massey, J. A.; Power, K. N.; Winnik, M. A.; Manners, I. Adv. Mater. 1998, 10 (18), 1559–1562. MacFarlane, L.; Zhao, C.; Cai, J.; Qiu, H.; Manners, I. Chem. Sci. 2021, 12 (13), 4661–4682. Gilroy, J. B.; Gädt, T.; Whittell, G. R.; Chabanne, L.; Mitchels, J. M.; Richardson, R. M.; Winnik, M. A.; Manners, I. Nat. Chem. 2010, 2 (7), 566–570. Kloninger, C.; Rehahn, M. Macromolecules 2004, 37 (22), 8319–8324. Lammertink, R. G. H.; Hempenius, M. A.; van den Enk, J. E.; Chan, V. Z.-H.; Thomas, E. L.; Vancso, G. J. Adv. Mater. 2000, 12 (2), 98–103. Lammertink, R. G. H.; Hempenius, M. A.; Thomas, E. L.; Vancso, G. J. J. Polym. Sci. Part B Polym. Phys. 1999, 37 (10), 1009–1021. Temple, K.; Kulbaba, K.; Power-Billard, K. N.; Manners, I.; Leach, K. A.; Xu, T.; Russell, T. P.; Hawker, C. J. Adv. Mater. 2003, 15 (4), 297–300. Eloi, J.-C.; Rider, D. A.; Wang, J.-Y.; Russell, T. P.; Manners, I. Macromolecules 2008, 41 (23), 9474–9479. Gwyther, J.; Lotze, G.; Hamley, I.; Manners, I. Macromol. Chem. Phys. 2011, 212 (2), 198–201. Datta, U.; Rehahn, M. Macromol. Rapid Commun. 2004, 25 (18), 1615–1622. Massey, J.; Power, K. N.; Manners, I.; Winnik, M. A. J. Am. Chem. Soc. 1998, 120 (37), 9533–9540. Raez, J.; Manners, I.; Winnik, M. A. J. Am. Chem. Soc. 2002, 124 (35), 10381–10395. Frankowski, D. J.; Raez, J.; Manners, I.; Winnik, M. A.; Khan, S. A.; Spontak, R. J. Langmuir 2004, 20 (21), 9304–9314. Wang, X.; Wang, H.; Frankowski, D. J.; Lam, P. G.; Welch, P. M.; Winnik, M. A.; Hartmann, J.; Manners, I.; Spontak, R. J. Adv. Mater. 2007, 19 (17), 2279–2285. Raez, J.; Manners, I.; Winnik, M. A. Langmuir 2002, 18 (19), 7229–7239. Korczagin, I.; Hempenius, M. A.; Fokkink, R. G.; Cohen Stuart, M. A.; Al-Hussein, M.; Bomans, P. H. H.; Frederik, P. M.; Vancso, G. J. Macromolecules 2006, 39 (6), 2306–2315. Cai, P.; Wang, C.; Ye, J.; Xie, Z.; Wu, C. Macromolecules 2004, 37 (9), 3438–3443. Chen, T.; Wang, L.; Jiang, G.; Wang, J.; Wang, J.; Zhou, J. Polymer 2005, 46 (15), 5773–5777. Cao, L.; Manners, I.; Winnik, M. A. Macromolecules 2002, 35 (22), 8258–8260. Gohy, J.-F.; Lohmeijer, B. G. G.; Alexeev, A.; Wang, X.-S.; Manners, I.; Winnik, M. A.; Schubert, U. S. Chem. Eur. J. 2004, 10 (17), 4315–4323. Wang, X.; Winnik, M. A.; Manners, I. Macromolecules 2005, 38 (5), 1928–1935. Wang, X.-S.; Winnik, M. A.; Manners, I. Angew. Chem. Int. Ed. 2004, 43 (28), 3703–3707. Thurmond, K. B.; Kowalewski, T.; Wooley, K. L. J. Am. Chem. Soc. 1996, 118 (30), 7239–7240. Thurmond, K. B.; Kowalewski, T.; Wooley, K. L. J. Am. Chem. Soc. 1997, 119 (28), 6656–6665. Bütün, V.; Lowe, A. B.; Billingham, N. C.; Armes, S. P. J. Am. Chem. Soc. 1999, 121 (17), 4288–4289. Wang, X.-S.; Arsenault, A.; Ozin, G. A.; Winnik, M. A.; Manners, I. J. Am. Chem. Soc. 2003, 125 (42), 12686–12687. Wang, X.-S.; Wang, H.; Coombs, N.; Winnik, M. A.; Manners, I. J. Am. Chem. Soc. 2005, 127 (25), 8924–8925. Wang, X.; Liu, K.; Arsenault, A. C.; Rider, D. A.; Ozin, G. A.; Winnik, M. A.; Manners, I. J. Am. Chem. Soc. 2007, 129 (17), 5630–5639. Wang, H.; Wang, X.; Winnik, M. A.; Manners, I. J. Am. Chem. Soc. 2008, 130 (39), 12921–12930. Hudson, Z. M.; Lunn, D. J.; Winnik, M. A.; Manners, I. Nat. Commun. 2014, 5 (1), 3372. Rupar, P. A.; Cambridge, G.; Winnik, M. A.; Manners, I. J. Am. Chem. Soc. 2011, 133 (42), 16947–16957. Qian, J.; Guerin, G.; Cambridge, G.; Manners, I.; Winnik, M. A. Macromol. Rapid Commun. 2010, 31 (9–10), 928–933. Qian, J.; Guerin, G.; Lu, Y.; Cambridge, G.; Manners, I.; Winnik, M. A. Angew. Chem. Int. Ed. 2011, 50 (7), 1622–1625. Qian, J.; Lu, Y.; Cambridge, G.; Guerin, G.; Manners, I.; Winnik, M. A. Macromolecules 2012, 45 (20), 8363–8372. Qian, J.; Lu, Y.; Chia, A.; Zhang, M.; Rupar, P. A.; Gunari, N.; Walker, G. C.; Cambridge, G.; He, F.; Guerin, G.; Manners, I.; Winnik, M. A. ACS Nano 2013, 7 (5), 3754–3766. Rupar, P. A.; Chabanne, L.; Winnik, M. A.; Manners, I. Science 2012, 337 (6094), 559–562. Qiu, H.; Hudson, Z. M.; Winnik, M. A.; Manners, I. Science 2015, 347 (6228), 1329–1332. Gould, O. E. C.; Qiu, H.; Lunn, D. J.; Rowden, J.; Harniman, R. L.; Hudson, Z. M.; Winnik, M. A.; Miles, M. J.; Manners, I. Nat. Commun. 2015, 6 (1), 10009. Li, X.; Gao, Y.; Boott, C. E.; Winnik, M. A.; Manners, I. Nat. Commun. 2015, 6 (1), 8127. Hsiao, M.-S.; Yusoff, S. F. M.; Winnik, M. A.; Manners, I. Macromolecules 2014, 47 (7), 2361–2372. Mohd Yusoff, S. F.; Hsiao, M.-S.; Schacher, F. H.; Winnik, M. A.; Manners, I. Macromolecules 2012, 45 (9), 3883–3891. Presa Soto, A.; Gilroy, J. B.; Winnik, M. A.; Manners, I. Angew. Chem. Int. Ed. 2010, 49 (44), 8220–8223. Hudson, Z. M.; Boott, C. E.; Robinson, M. E.; Rupar, P. A.; Winnik, M. A.; Manners, I. Nat. Chem. 2014, 6 (10), 893–898. Cambridge, G.; Gonzalez-Alvarez, M. J.; Guerin, G.; Manners, I.; Winnik, M. A. Macromolecules 2015, 48 (3), 707–716. Qiu, H.; Gao, Y.; Boott, C. E.; Gould, O. E. C.; Harniman, R. L.; Miles, M. J.; Webb, S. E. D.; Winnik, M. A.; Manners, I. Science 2016, 352 (6286), 697–701. Gao, Y.; Shreeve, J. M. J. Inorg. Organomet. Polym. Mater. 2007, 17 (1), 19–36. Abd-El-Aziz, A. S.; Manners, I. J. Inorg. Organomet. Polym. Mater. 2005, 15 (1), 157–195. Satyanarayana, V. S. V.; Singh, V.; Kalyani, V.; Pradeep, C. P.; Sharma, S.; Ghosh, S.; Gonsalves, K. E. RSC Adv. 2014, 4 (104), 59817–59820. Gallei, M.; Tockner, S.; Klein, R.; Rehahn, M. Macromol. Rapid Commun. 2010, 31 (9–10), 889–896. Elbert, J.; Gallei, M.; Rüttiger, C.; Brunsen, A.; Didzoleit, H.; Stühn, B.; Rehahn, M. Organometallics 2013, 32 (20), 5873–5878. Kim, B. Y.; Ratcliff, E. L.; Armstrong, N. R.; Kowalewski, T.; Pyun, J. Langmuir 2010, 26 (3), 2083–2092. Ghimire, G.; Yi, Y.; Derylo, M. A.; Baker, L. A.; Ito, T. Langmuir 2015, 31 (44), 12307–12314. Bas, S. Z.; Maltas, E.; Sennik, B.; Yilmaz, F. Colloids Surf. A Physicochem. Eng. Asp. 2014, 444, 40–47. Alkan, A.; Natalello, A.; Wagner, M.; Frey, H.; Wurm, F. R. Macromolecules 2014, 47 (7), 2242–2249. Alkan, A.; Thomi, L.; Gleede, T.; Wurm, F. R. Polym. Chem. 2015, 6 (19), 3617–3624. Pandey, S.; Lönnecke, P.; Hey-Hawkins, E. Eur. J. Inorg. Chem. 2014, 2014 (14), 2456–2465. Pandey, S.; Lönnecke, P.; Hey-Hawkins, E. Inorg. Chem. 2014, 53 (16), 8242–8249. Morris, L. J.; Whittell, G. R.; Eloi, J.-C.; Mahon, M. F.; Marken, F.; Manners, I.; Hill, M. S. Organometallics 2019, 38 (19), 3629–3648. Kraiter, D. C.; Wisian-Neilson, P.; Zhang, C.; Crumbliss, A. L. Macromolecules 2012, 45 (9), 3658–3668. Zamora, M.; Bruña, S.; Alonso, B.; Cuadrado, I. Macromolecules 2011, 44 (20), 7994–8007. Jiménez, A.; Armada, M. P. G.; Losada, J.; Villena, C.; Alonso, B.; Casado, C. M. Sensors Actuators B Chem. 2014, 190, 111–119. de la Cruz, G.; Schüle, H.; Losada, J.; García-Armada, M. P.; Frey, H.; Alonso, B.; Casado, C. M. Eur. J. Inorg. Chem. 2013, 2013 (1), 44–53. Martínez-Montero, I.; Bruña, S.; González-Vadillo, A. M.; Cuadrado, I. Macromolecules 2014, 47 (4), 1301–1315. Parab, K.; Jäkle, F. Macromolecules 2009, 42 (12), 4002–4007.

22 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. 180. 181. 182. 183. 184. 185. 186. 187. 188. 189.

Ferrocene and Related Metallocene Polymers Li, H.; Sundararaman, A.; Pakkirisamy, T.; Venkatasubbaiah, K.; Schödel, F.; Jäkle, F. Macromolecules 2011, 44 (1), 95–103. Dubacheva, G. V.; Van Der Heyden, A.; Dumy, P.; Kaftan, O.; Auzély-Velty, R.; Coche-Guerente, L.; Labbé, P. Langmuir 2010, 26 (17), 13976–13986. Merchant, S. A.; Meredith, M. T.; Tran, T. O.; Brunski, D. B.; Johnson, M. B.; Glatzhofer, D. T.; Schmidtke, D. W. J. Phys. Chem. C 2010, 114 (26), 11627–11634. Abbott, N. L.; Whitesides, G. M. Langmuir 1994, 10 (5), 1493–1497. Akpınar, H.; Balan, A.; Baran, D.; Ünver, E. K.; Toppare, L. Polymer 2010, 51 (26), 6123–6131. Özdemir, S.; ¸ Balan, A.; Baran, D.; Dogan, Ö.; Toppare, L. React. Funct. Polym. 2011, 71 (2), 168–174. Bagh, B.; Schatte, G.; Green, J. C.; Müller, J. J. Am. Chem. Soc. 2012, 134 (18), 7924–7936. Nelson, J. M.; Lough, A. J.; Manners, I. Angew. Chem. Int. Ed. Eng. 1994, 33 (9), 989–991. Vogel, U.; Lough, A. J.; Manners, I. Angew. Chem. 2004, 116 (25), 3383–3387. Ketkov, S. Y.; Selzle, H. L. Angew. Chem. Int. Ed. 2012, 51 (46), 11527–11530. Herberich, G. E.; Bauer, E.; Schwarzer, J. J. Organomet. Chem. 1969, 17 (3), 445–452. Mayer, U. F. J.; Gilroy, J. B.; O’Hare, D.; Manners, I. J. Am. Chem. Soc. 2009, 131 (30), 10382–10383. Mayer, U. F. J.; Charmant, J. P. H.; Rae, J.; Manners, I. Organometallics 2008, 27 (7), 1524–1533. Drewitt, M. J.; Barlow, S.; O’Hare, D.; Nelson, J. M.; Nguyen, P.; Manners, I. Chem. Commun. 1996, (18), 2153–2154. Fox, S.; Dunne, J. P.; Tacke, M.; Schmitz, D.; Dronskowski, R. Eur. J. Inorg. Chem. 2002, 2002 (11), 3039–3046. Braunschweig, H.; Breher, F.; Kaupp, M.; Gross, M.; Kupfer, T.; Nied, D.; Radacki, K.; Schinzel, S. Organometallics 2008, 27 (24), 6427–6433. Ransom, P.; Ashley, A.; Thompson, A.; O’Hare, D. J. Organomet. Chem. 2009, 694 (7), 1059–1068. Gilroy, J. B.; Patra, S. K.; Mitchels, J. M.; Winnik, M. A.; Manners, I. Angew. Chem. Int. Ed. 2011, 50 (26), 5851–5855. Jarrett-Wilkins, C. N.; Musgrave, R. A.; Hailes, R. L. N.; Harniman, R. L.; Faul, C. F. J.; Manners, I. Macromolecules 2019, 52 (19), 7289–7300. Musgrave, R. A.; Choi, P.; Harniman, R. L.; Richardson, R. M.; Shen, C.; Whittell, G. R.; Crassous, J.; Qiu, H.; Manners, I. J. Am. Chem. Soc. 2018, 140 (23), 7222–7231. Sheats, J. E.; Rausch, M. D. J. Org. Chem. 1970, 35 (10), 3245–3249. Astruc, D.; Ornelas, C.; Ruiz, J. Acc. Chem. Res. 2008, 41 (7), 841–856. Ornelas, C.; Ruiz, J.; Astruc, D. Organometallics 2009, 28 (9), 2716–2723. Casado, C. M.; González, B.; Cuadrado, I.; Alonso, B.; Morán, M.; Losada, J. Angew. Chem. 2000, 112 (12), 2219–2222. Ren, L.; Hardy, C. G.; Tang, C. J. Am. Chem. Soc. 2010, 132 (26), 8874–8875. Ren, L.; Zhang, J.; Bai, X.; Hardy, C. G.; Shimizu, K. D.; Tang, C. Chem. Sci. 2012, 3 (2), 580–583. Yan, Y.; Zhang, J.; Qiao, Y.; Tang, C. Macromol. Rapid Commun. 2014, 35 (2), 254–259. Ciganda, R.; Gu, H.; Castel, P.; Zhao, P.; Ruiz, J.; Hernández, R.; Astruc, D. Macromol. Rapid Commun. 2016, 37 (1), 105–111. Zhang, J.; Ren, L.; Hardy, C. G.; Tang, C. Macromolecules 2012, 45 (17), 6857–6863. Zhang, J.; Pellechia, P. J.; Hayat, J.; Hardy, C. G.; Tang, C. Macromolecules 2013, 46 (4), 1618–1624. Baljak, S.; Russell, A. D.; Binding, S. C.; Haddow, M. F.; O’Hare, D.; Manners, I. J. Am. Chem. Soc. 2014, 136 (16), 5864–5867. Musgrave, R. A.; Russell, A. D.; Hayward, D. W.; Whittell, G. R.; Lawrence, P. G.; Gates, P. J.; Green, J. C.; Manners, I. Nat. Chem. 2017, 9 (8), 743–750. Musgrave, R. A.; Hailes, R. L. N.; Annibale, V. T.; Manners, I. Chem. Sci. 2019, 10 (42), 9841–9852. Hailes, R. L. N.; Musgrave, R. A.; Kilpatrick, A. F. R.; Russell, A. D.; Whittell, G. R.; O’Hare, D.; Manners, I. Chem. Eur. J. 2019, 25 (4), 1044–1054. Musgrave, R. A.; Russell, A. D.; Gamm, P. R.; Hailes, R. L. N.; Lam, K.; Sparkes, H. A.; Green, J. C.; Geiger, W. E.; Manners, I. Organometallics 2021, 40 (12), 1945–1955. Dunitz, J. D.; Orgel, L. E.; Rich, A. Acta Crystallogr. 1956, 9 (4), 373–375. Antipin, M. Y.; Boese, R.; Augart, N.; Schmid, G. Struct. Chem. 1993, 4 (2), 91–101. Seiler, P.; Dunitz, J. D. Acta Crystallogr. B 1980, 36 (10), 2255–2260. Kadkin, O.; Näther, C.; Friedrichsen, W. J. Organomet. Chem. 2002, 649 (2), 161–172. Tuba, R.; Grubbs, R. H. Polym. Chem. 2013, 4 (14), 3959–3962. Kubisa, P.; Penczek, S. Prog. Polym. Sci. 1999, 24 (10), 1409–1437. Hultzsch, K. C.; Nelson, J. M.; Lough, A. J.; Manners, I. Organometallics 1995, 14 (12), 5496–5502. Seyferth, D. Organometallics 2002, 21 (14), 2800–2820. Elschenbroich, C.; Hurley, J.; Metz, B.; Massa, W.; Baum, G. Organometallics 1990, 9 (4), 889–897. Berenbaum, A.; Manners, I. Dalton Trans. 2004, (14), 2057–2058. Tamm, M.; Kunst, A.; Herdtweck, E. Chem. Commun. 2005, (13), 1729–1731. Braunschweig, H.; Adams, C. J.; Kupfer, T.; Manners, I.; Richardson, R. M.; Whittell, G. R. Angew. Chem. Int. Ed. 2008, 47 (20), 3826–3829.

14.03

Conjugated Poly(metalla-ynes)

Ashanul Haquea, Muhammad S Khanb, Mei-Tung Lauc, Zikang Lic, Paul R Raithbyd, and Wai-Yeung Wongc, aDepartment of Chemistry, College of Science, University of Hail, Hail, Saudi Arabia; bDepartment of Chemistry, Sultan Qaboos University, Muscat, Oman; c Department of Applied Biology and Chemical Technology and Research Institute for Smart Energy, The Hong Kong Polytechnic University, Hong Kong, People’s Republic of China; dDepartment of Chemistry, University of Bath, Bath, United Kingdom © 2022 Elsevier Ltd. All rights reserved.

14.03.1 Preamble 14.03.2 Synthesis of poly(metalla-ynes) 14.03.3 Structure-property relationship 14.03.3.1 Pt(II)-based poly(metalla-ynes) 14.03.3.1.1 Pt(II)-based poly(metalla-ynes) incorporating phenylene, fluorene and carbazole spacers 14.03.3.1.2 Pt(II)-based poly(metalla-ynes) incorporating thiophene and related spacers 14.03.3.1.3 Pt(II)-based poly(metalla-ynes) incorporating pyridine and related spacers 14.03.3.1.4 Pt(II)-based poly(metalla-ynes) incorporating hybrid spacers 14.03.4 Non-platinum poly(metalla-ynes) 14.03.5 Applications 14.03.5.1 Optoelectronics 14.03.5.1.1 Bulk heterojunction polymer solar cells (PSCs) 14.03.5.1.2 Organic light-emitting diodes (OLEDs) 14.03.5.1.3 Organic field-effect transistors (OFETs) 14.03.5.1.4 Non-linear optics (NLO) 14.03.5.2 Sensors 14.03.5.3 Other devices 14.03.5.3.1 Memory devices 14.03.5.3.2 Catalysts 14.03.5.3.3 Biologically active compounds 14.03.6 Summary Acknowledgments References

23 24 27 27 27 32 33 33 37 41 41 41 50 52 53 57 61 61 63 65 66 67 67

14.03.1 Preamble Transition metal ions that are s-bound to acetylides have been known since the 1960s1 and the first metalla-yne polymers were reported by Hagihara et al. in the late 1970s.2–4 One of the reasons for the interest in these complexes and polymers is the range of chemical and physical properties that they exhibit.5–7 Since the 1980s they have found applications in a number of areas including liquid-crystalline materials,4 non-linear optical (NLO) materials,8 light-emitting diodes9,10 and one-dimensional conductors,11 and the range of their uses continues to expand until the present day.12,13 The advantages of metalla-ynes is that they are generally quite soluble, easily processable materials and it is relatively easy to tune their physical or chemical properties to optimize a desired functionality of the material consisting of either monomeric, oligomeric or polymeric molecules.14,15 Mononuclear and polynuclear complexes containing transition metal s-acetylide bonds have featured in all the previous editions of Comprehensive Organometallic Chemistry. In the first two editions of Comprehensive Organometallic Chemistry transition metal complexes with s-acetylide bonds were generally discussed in the chapters pertaining to the different transition metals involved, with the majority of the discussion concerning the heavier transition metals.16,17 In the 3rd Edition while examples of metalla-ynes can be found in the chapters on specific elements, there was a focus on rigid-rod metalla-yne polymers in Volume 12 where their non-linear optical properties were discussed.18 Because of the expansion in the field over the last 15 years, in the present Edition transition metal mono-, di-, oligo- and polymeric acetylides (conjugated poly (metalla-ynes)) have this chapter devoted to them. The majority of the discussion in the following pages describes developments in the areas of synthesis, characterization and the nature of the chemical and physical properties of poly(metalla-ynes) along with their applications in materials chemistry since 2007. However, to provide the reader with the context of the advances some earlier work is also described. As mentioned above, the ability to tune the chemical and physical properties of poly(metalla-ynes) through chemical manipulation is one of the key reasons why these materials, and their molecular precursors, have proved to be so versatile. The manipulation can be achieved by altering both the acetylide ligand and the nature and environment of the metal. The addition of functional groups to the acetylene spacer ligand (R) can alter the electronic and steric properties of the materials as well as their solubility in a range of solvents. The number and type of auxiliary ligands (Ln) (often called spectator ligands) also influence the steric and electronic properties of the systems and, of course, varying the metallic element(s) present together with their formal oxidation states and coordination geometries has major influence on the resultant properties.19 Ln M −C  C −R

Comprehensive Organometallic Chemistry IV

https://doi.org/10.1016/B978-0-12-820206-7.00148-7

23

24

Conjugated Poly(metalla-ynes)

In the development of the chemistry of transition metal s-acetylide complexes and their related polymers one of the fundamentals is the thermal stability of the metal-carbon bond which is a result of the dp-pp bonding interactions along the meta acetylide axis. The absence of b-hydrogen atoms on the acetylene ligand eliminates the possibility of the breaking of the M-C s-bond by b-hydrogen elimination. An early combined experimental and theoretical study indicated that there was a significant overlap between the metal d-orbitals and the filled p-orbitals of the acetylene ligand.20,21 The energy difference between the HOMO metal d orbitals and the LUMO p∗ acetylene orbitals appears to be too large, at ca. 15 eV, to act as an acceptor. Therefore, in these complexes, the acetylene behaves more like a chloride p-donor rather than a carbonyl p-acceptor ligand. The coordination geometry of the transition metal also has a major influence on both the thermodynamic and kinetic properties of the s-acetylide complexes and the metalla-yne polymers. The majority of the reported metalla-ynes contain heavier transition metals. There are two main reasons for this. Firstly, the heavy transition metals tend to form thermodynamically stable metal-carbon bonds. Secondly, the heavy transition metals exhibit high spin-orbit coupling that leads to relatively efficient intersystem crossing and, therefore, access to excited triplet states, that provides a route to efficient phosphorescence as well as fluorescence.22 This factor is important in the design of Organic Light Emitting Diodes (OLEDs) because for organic materials only singlet states, resulting in fluorescence, are available. Since the ratio of singlet to triplet states is 1:3, organic OLEDs have a maximum theoretical efficiency of 25%, whereas, if the triplet states are accessible the theoretical efficiency of the OLED is 100%. The presence of the heavy transition metals allows the quantum mechanically singlet-triplet transitions to be accessed, and in triplet emitters the additional 75% of excitons are harvested.23 For metalla-ynes that contain the transition metal ions Ru(II), Rh(III) and Ir(III) the coordination geometry around the metal is usually octahedral with acetylide groups occupying two trans positions to generate the linear, rigid-rod configuration. In these systems chelating phosphines or diimines usually act as the auxiliary supporting ligands.24 For the group 10 metals, in rigid-rod molecules and polymers, the most commonly observed metal oxidation states are Pd(II) and Pt(II) and both adopt the expected square planar geometry of a four coordinate d8 metal ion and the acetylide ligands occupy trans positions.24 For group 11 the commonly observed oxidation states are Ag(I) and Au(I) that give a linear two coordinate geometry, and there are an increasing number of related square planar Au(III) complexes that display luminescent properties.25 Linear Hg(II) complexes with two trans acetylide groups have also been developed and they often mimic the behavior of the analogous Au(III) complexes.26 One of the most significant effects of selecting a particular metal for inclusion in a poly(metalla-yne) is the effect that it has on the bandgap. Four coordinate, square planar d8 metal centers are favored over six coordinate, octahedral d6 metal centers because of the greater optical gap energy of the latter (ca. 0.4 eV).27 Depending on the particular application being targeted the tuning of the HOMO-LUMO band gap is one of the most important features of the poly(metalla-ynes) and the initial focus in the following sections will be on Pt(II) and Pd(II) systems.

14.03.2 Synthesis of poly(metalla-ynes) Following the first report on the synthesis of metal s-acetylide complexes,1 several strategies have been reported for the preparation of mono-, di-, oligo- and polymeric metal acetylides. Fig. 1 depicts some common protocols applied to access cis- and transpoly(metalla-ynes) incorporating one or more types of transition metals. In general, polymetalla-ynes are prepared mainly by three convergent strategies: dehydrohalogenation, oxidative coupling or by an alkynyl ligand exchange reaction. Since all these methods and their scope have already been reviewed in the past,24,28,29 here we describe representative examples of the methods. The very first reported method for soluble polymers containing a s-bonded transition metal-carbon bond in the main chain was Hagiharas’ dehydrohalogenation method.3,30 Undoubtedly, this is the most straight forward and frequently used method until now to afford linear (trans) and angular (cis) mono-, di-, oligo and poly(metalla-

Fig. 1 Methodologies for synthesizing trans poly(metalla-ynes) incorporating early and late transition metals (M). The geometry around the metal center can be linear, square planar, or octahedral; L ¼ mono- or bidentate p-back bonding or s-donating auxiliary ligands, X ¼ H or SnMe3.

Conjugated Poly(metalla-ynes)

25

Fig. 2 Some selected Pt(II) and Pd(II) complexes synthesized using a dehydro-halogenation condensation reaction.

ynes). The reaction is carried out in an amine solvent in the presence/absence of Cu(I) halide. Fig. 2 depicts some examples of p-conjugated Pt(II) and Pd(II) organometallic oligomeric and polymeric metalla-ynes obtained by the classical dehydrohalogenative coupling procedure. The chain length (oligomeric/polymeric) and nature (poly-yne/polyene) of the polymerized product depends on the configuration of the metal precursor (cis/trans), the presence/absence of catalyst (CuI), reaction time and also the type of metal.31–33 Advantages of this method include the formation of a polymeric product with high molecular weight, given that the two highly pure precursors are loaded in the exact ratio of 1:1. However, limited stability of the early transition metal complexes including Ni(II) in amine solvents often hampers the use of this method. In oxidative coupling, mono or (bis)acetylide monomers are converted to oligo- or polymeric products under Hay conditions. Oxidative coupling method produces polymers with high degree of polymerization, as the process depends on a single monomer only. Apart from the purity of the polymers obtained, another attractive feature of oxidative coupling is the lengthening of the conjugated organic spacer in that varying the inter-metal distance may result in slightly different polymer properties. This method has advantage that it takes place in the presence of oxygen (inert conditions are not required).2 Fig. 3 shows some oligomeric Pt(II) complexes realized by following oxidative homocoupling (O2/CuCl/TMEDA).34 Historically, a transmetalation/metathesis protocol was the first method used for synthesizing a molecular Pt(II) s-acetylide complex.1 In this method, an alkynyl transfer reagent [MdC^CdR] [M ¼ transition, alkali, alkaline earth etc.) delivers an alkynyl group to a transition-metal halide complex (Fig. 4).35 To obtain poly(metalla-ynes) incorporating Ni(II), Pd(II), Pt(II) and Rh(I) metals in the main chain, a modified Sonogashira method employing bis(trimethylstannyl)alkynyl reagents was proposed by Lewis and co-workers. 36 Later, several modifications of this method have been suggested. For instance, the CdH activation method for the synthesis of Rh(III),37 Ru(II),38 Fe(II),39 and Ir(III)40 and others.40–43 Particularly, the one reported by Marder37 was immensely useful for obtaining mono- and poly-nuclear Rh(III) complexes in high yields.

Fig. 3 A representative example of Pd(II) oligo-yne synthesis by oxidative coupling.

Fig. 4 A representative example of Group 9 and 10 metal-containing poly-yne synthesis using bis(trimethylstannyl) alkynyl precursor and metal halides.

26

Conjugated Poly(metalla-ynes)

The metathesis method using the stannylalkynyl derivative has advantages over the dehydrohalogenation method in that (i) higher molecular weights were achieved and (ii) a variety of transition metals (groups 8, 9 and 10) could be incorporated as a wide range of organic solvents could be used including CH2Cl2, THF, toluene, etc. and (iii) the good stability of the trimethylstannyl derivatives. The method is not restricted to amines whereas in dehydrohalogenation, amines are essential components for the synthesis. Therefore, the metathesis method may be applied to other metals earlier in the d-block. However, toxicity, sensitivity (air/moisture), purity of stannylalkynyl reagents, low yields, range of by-products, are some of the major drawbacks of this method. Apart from these, some metalla-ynes incorporating transition metals can also be prepared via an alternative route. For example, Szafert and co-workers reported that octahedral iridium(III) end-capped complexes can be obtained by oxidative addition of 1-iodopoly-ynes to Vaska’s complex (Fig. 5).44 Mild condition, high yield, up to 10 C atom chains and formation of only one isomer were the main advantages of this method. On-surface synthesis becomes a powerful approach to fabricate well-controlled molecular organic and organometallic nanostructures.45 It has been demonstrated that terminal alkynes undergo on-surface reactions at a reactive substrate such as a noble metal surface to yield low dimensional organometallic molecular wires nanomaterials. Metallapoly-ynes incorporating Ag(I), Au(I) have been reported in recent times.46 Liu and workers reported the formation Ag and Au-alkynyl networks on Ag(111) and Au(111) surfaces (Fig. 6). It was found that the alkyne precursor 1,3,5-tris(chloroethynyl)benzene on Ag(111) formed a honeycomb Ag-alkynyl networks formed at 393 K, and only short chain intermediates were observed. On the other hand, the same precursor formed honeycomb Au–alkynyl networks on Au(111) at 503 K.47

Fig. 5 Routes for obtaining metal end-capped poly-ynes.

Fig. 6 Stepwise growth mechanism of metal-alkynyl networks on Ag(111) and Au(111). Reproduced with permission from ref. Shu, C.-H.; He, Y.; Zhang, R.-X.; Chen, J.-L.; Wang, A.; Liu, P.-N. J. Am. Chem. Soc. 2020, 142(39), 16579–16586.

Conjugated Poly(metalla-ynes)

27

14.03.3 Structure-property relationship Metalla-ynes are well known for their attractive chemical and photophysical features leading to multi-disciplinary applications.12 This is due to the possibility of tuning properties via changing either the metal or the auxiliary ligands or the acetylide-based spacer groups. The fact that poly(metalla-ynes) having the structure [dM(L)ndC^CdRdC^Cd]1 possess a greater number of sub-units than the organic poly-ynes, they offer greater opportunities to tune the property of a material. In particular, the inclusion of a metallic core to the organic framework drastically improves the O-E properties and imparts the capability to resolve many hidden facets at the supramolecular level.48,49 For example, a recent study has demonstrated that adsorption and activity of a microbial lipase immobilized on noble-metal (Ag/Au) nanoparticles depends on the Pt(II) di-yne complexes used to stabilize the NPs. It has been shown that nature of the metal plays a role in enzyme adsorption, while enzyme activity is mostly influenced by the chemistry of the organic spacer.50 To date, a large number of mono-, di-, oligo- and polymeric metal acetylides with interesting properties and applications have been reported, which have been well reviewed from time to time.12,14,28,51–54 Due to the low stability, limited solubility, and lack of reliable synthetic pathway to high-molecular weight materials, functional poly(metallaynes) containing earlier transition metals are relatively scarce. However, their mono-, dinuclear complexes and oligomers have been studied extensively, which is beyond the scope of this chapter.12 This chapter highlights the versatility and structure dependent features/properties of poly(metalla-ynes) using some selected examples. In the following sub-sections, we highlight the structure-property relationships of poly(metalla-ynes). We divided this section into two parts: (i) Pt(II) containing poly-ynes and (ii) other metalla-ynes.

14.03.3.1 Pt(II)-based poly(metalla-ynes) A survey of the literature shows that poly(platina-ynes) with either cis or trans geometry, incorporating carbocyclic and heterocyclic spacers, and having phosphine/arsine/pyridine/bipyridine auxiliary ligands have been reported upon extensively together with their molecular precursors. In contrast to the majority of the other late transition metals, research on conjugated Pt(II) poly-ynes is widespread. Platinum(II) complexes represent an important type of transition metal complexes with d8 electron configuration and the square planar geometry. When organic ligands are bound to the metal center, a variety of organoplatinum(II) complexes with either mono- or multidentate character are obtained.12,55 Through structure/property relationship studies, it is now well established that the organic spacers, their functionalization and variation in chain length and molecular weight have significant effect on the property of the resulting materials. Since the intramolecular interaction between the two arylacetylide segments linked by a metal center affects the localization/delocalization of excitons, they are considered to be excellent systems for investigating the ground state absorption to the triplet state, ISC, triplet state absorption and phosphorescence 56–60. Indeed, compared to other transition metals, the Pt(II) fragment offer extra advantage: high stability, emission that extends to the near infrared (NIR) region, high emission quantum yields and long excited state lifetimes.61–63 In the sub-sections below, we discuss the effects, by virtue of which properties can be modulated.

14.03.3.1.1

Pt(II)-based poly(metalla-ynes) incorporating phenylene, fluorene and carbazole spacers

Extensive research has been carried out using platina-ynes with or without phenylene, fluorene and carbazole spacers and their derivatives in order to establish the underlying properties. In this context, platina-ynes such as p-phenyleneethynyl (PtdPEs)2,3,30 poly(p-phenyleneethynylenes) (PtdPPEs)64 and poly(p-phenylenebutadiynylene)s (Pt-PPBs)65 with varying metal to spacer ratios have been reported along with their PL properties. For instance, in poly-ynes Pt-1 to Pt-4 (Chart 1), it was found that p-conjugation is reduced by the insertion of phenylene rings into the backbone.66 This reflects the energy differences between the acetylenic and phenylene ring orbitals and also relates to the disposition of the phenylene ring p-orbitals with respect to the platinum-acetylenic conjugated chain orbitals. Decreases in conjugation can be seen as increases in the frequencies of acetylenic stretching vibrations between “pure” acetylenic polymers (Pt-1 and Pt-2) and the pseudo-acetylenic polymers (Pt-3 and Pt-4). Addition of acetylene moieties to pseudo-acetylenic materials emphasizes the break at the aromatic ring, where metal p-interactions are delocalized over a longer chain fragment.66 Later, it was demonstrated that poly(platina-ynes) incorporating phenylene spacer possess radiative triplet excited states, which are very weak in their organic counterparts.67,68 Photophysical investigation of Pt(II) poly-ynes incorporating fluorinated 1,4-diethynylbenzene derivatives revealed that the relative intensity of the triplet emission increases strongly with an increase in the electronegative fluorine content in such systems.69 Recently, solution processable polymers Pt-5 (R ¼ OC4H9, OC8H17, OC12H25, Chart 1) having molecular weights (Mn) ranging from 52,738 to 74,212, and platinum to phenylenebutadiynylene (PB) ratios of 1:4 to 1:6 have been reported.65 As expected, organic polymer showed only fluorescence in thin films, while strong phosphorescence was observed for polymers having a higher Pt(II) content at both room temperature and at low temperatures (Fig. 8). On the other hand, polymer Pt-5 (R ¼ OC4H9) with a decreased platinum content displayed strong fluorescence at both 290 and 5 K. In contrast to the organic PPB oligomer, it also showed weak phosphorescence (Fig. 7). Polymers Pt-7 to Pt-10 (Chart 1) with chalcogen atoms in the spacer group have been investigated.70 The introduction of chalcogen atoms effectively increases the accessibility of the 3p–p excited states leading to efficient phosphorescence at room temperature. A blue shift in the phosphorescence maxima was attributed to conjugation break unit (chalcogens). Pt(II) poly-ynes incorporating condensed aromatic spacers such as naphthalene (Pt-11, Chart 1) and anthracene (Pt-12, Chart 1) have lower Eg values compared to the biphenylene spacer (Pt-6, Chart 1). These electron-rich rings create strong intramolecular interactions

28

Conjugated Poly(metalla-ynes)

Chart 1 Examples of poly(platina-ynes) containing phenylene and related spacers.

Fig. 7 (A) Room temperature absorption and emission data, as well as emission at low temperature, in toluene solution and (B) in a polystyrene matrix as a thin film. Fl and Ph denote the fluorescence and phosphorescence, respectively. (C) chemical structure for ease of reference. Reproduced with permission from ref. Ahmed, M. K.; Rahman, M. M.; Naher, M.; Mehdi, S. S.; Khan, A. R.; Khan, M. M. R.; Islam, S. S.; Younus, M.; Wedler, S.; Bagnich, S., Macromol. Chem. Phys. 2019, 220(8), 1800494.

between the Pt units and conjugated organic groups along the poly-ynyl backbone. The Eg value decreases as the size of the aromatic spacer increases, owing to the greater p-delocalization within the anthryl group as compared to the phenylene spacer, but there is no evidence of phosphorescence in thin films of Pt-12. Thermal characterization of the materials indicates that the di-ynes possess a somewhat higher thermal stability than the corresponding poly-ynes. Both the Pt(II) di-ynes and the poly-ynes exhibit increasing thermal stability along the series of spacers from phenylene through naphthalene to anthracene.

Conjugated Poly(metalla-ynes)

29

It is now well established that the polymer chain length as well as the PL properties of the metal-containing poly-ynes can be significantly tuned by functionalization of the spacer. For example, anchoring a diphenylamine pendant as an electron-rich group and anthraquinone (AQ) as electron-deficient species has been shown to modulate the properties of the material. Thus functionalization of AQ (Pt-13, Chart 1) generated polymers Pt-14 and Pt-15 (Chart 1) with longer chain length (Mn ¼ 88,200, Mw ¼ 188,700 for Pt-14 and Mn ¼ 147,000, Mw ¼ 327,000 for Pt-15) and improved fluorescence intensity and lifetime.71 Polymer Pt-16 (Chart 1) is a recent example of a fluorescent Pt(II) poly-yne, which was never observed before for this type of chromophore.72 From the spectroscopic and photophysical viewpoints, the anchoring of the bis(4-methoxyphenyl)amine groups turned out to be beneficial since fluorescence arising from the lowest energy excited state, S1 was observed for the first time in this type of push −pull material. These polymers undergo an efficient oxidative photoinduced electron transfer in solution and in films. Fluorescence lifetime studies indicated that the value depends on the diphenylamine pendants that act as electron-rich groups (tF  8 ps for the polymer having one bis(4-methoxyphenyl)amine and tF ¼ 10 ps having two bis(4-methoxyphenyl)amine). Another rare example of a p-conjugated inorganic polymer Pt-13 (both transition metal and main group elements) has been reported recently (Fig. 8).73,74 The polymer, which was constructed from boron difluoride formazanates and Pt(II) di-ynes, showed excellent thermal stability, film-forming properties, controllable redox chemistry, a low optical band gap (1.4 eV) and an electron accepting nature. Poly(platina-ynes) incorporating carbazole and fluorene spacers have been reported by several research groups. Advantages of using these spacers include a rigid structure, multiple functionalization sites and excellent O-E properties (such as blue emission, tunable S1-T1 energy gaps).75–78 Studies using a functionalized fluorene spacer (Pt-18 – Pt-22, Chart 2) indicated that properties such as the energy of the S1 state of the material depends on the type of substituents at the C-9 position79–83 as well as the type of metal. For example, a decrease in ISC efficiency upon introduction of a Pd(II) ion in the back has been reported.84 Compared to their model compounds, polymers Pt-18 (R ¼ C4H9, C8H17) exhibited conjugation length dependent absorption, emission and phosphorescence properties. The temperature dependent emission profile in polymer Pt-18 (R ¼ C8H17) was attributed to the thermally activated diffusion of T-excitons along the polymer chain, which led to triplet-triplet annihilation or increased sampling of dissociation sites or non-radiative sites. The introduction of an electron withdrawing group (CN) to the C-9 position causes a significant shift in the absorption maxima and reduced the band gap. It was found that the inductive effect generated by strongly electron-withdrawing -CN moieties in the poly-yne lowered the S1 state significantly. Polymer Pt-18 (R ¼ C6H13) and Pt-23 (Chart 2) reportedly have lower band gap values than the group 11 and 12 metal-containing compounds.85,86 Zhou and co-workers

Fig. 8 (A) Chemical structure of platina-yne Pt-17, (B) X-ray crystal structure of a model compound, (C) UV–vis absorbance spectra recorded for a 80 mM solution of Pt-12 (blue line) and the fully reduced form of Pt-12 (1.5 equiv. cobaltocene, red line) in a 1:1 (v/v) mixture of ACN and DCM. Reproduced with permission from ref. Dhindsa, J. S.; Maar, R. R.; Barbon, S. M.; Avilés, M. O.; Powell, Z. K.; Lagugné-Labarthet, F.; Gilroy, J. B. Chem. Commun. 2018, 54(50), 6899–6902.

30

Conjugated Poly(metalla-ynes)

Chart 2 Poly(platina-ynes) containing some fluorene-derived spacer units.

λmax = 399 nm ФFl. = 0.6

λmax = 364 nm ФFl. = 0.1 λmax = 412 nm ФFl. = 9.9 Fig. 9 Effect of changing metals and auxiliaries in fluorene based metalla-ynes.

compared the photophysical properties and optical power limiting (OPL) ability of Pt(II) poly-ynes bearing fluorene-type ligands (Pt-24 – Pt-31, Chart 2).87 When ethynyl units were attached via the 3,6-position, its absorption spectra were blue-shifted compared to the 2,7-analogs and exhibited stronger triplet emissions. Compared to trans counterparts, fluorene containing polymers with a cis-configuration of the Pt(II) metal center have a zig-zag structure and less effective p-conjugation along the chain (Fig. 9) On the other hand, when Pd(II) was introduced, it showed a more red shifted absorption and a significant fluorescence quantum yield. The first example of poly(arylene ether) copolymers incorporating Pt(II) acetylide units in the polymer backbone have been reported recently (Pt-32 and Pt-33, Chart 3).88 The covalent attachment to the polymer effectively avoids aggregation and oxygen quenching. The polymers exhibited high molecular weights, thermal/mechanical stability, negligible absorption in the visible Ph region (lAbs max. ¼ 378 and 395 nm), strong phosphorescence (lmax. ¼ 541 and 561 nm) and long T1 state lifetimes (tPh  0.88 and 2.07 ms) at low temperatures and exceptional OPL properties. Following this work, several groups designed and synthesized a variety of poly(platina-ynes) with different spacers, connectivities (topology) and functionalized rings. A comparative study of the structural and photophysical properties of 2,7- and 3–6carbazole incorporated poly(platina-ynes) (Pt-34, Pt-35, R ¼ H, EH, Chart 4) highlights some interesting findings.89 For example, when the Pt(II) fragments was para-linked, the S1-T1 splitting was twice that of the meta-linked counterpart. Additionally, a larger decrease in exchange energy (i.e. the S1-T1 energy gap) after polymerization was seen in the meta-linked polymer compared to the para-linked counterpart. It was noted that the incorporation of carbazole-2,7-diyl units into rigid-rod organometallic polymers enhances electron transport along the chain more effectively than the carbazole-3,6-diyl counterparts as the former offer a better conjugation pathway than the latter (Fig. 10).

Chart 3 Poly(platina-ynes) containing some fluorene-derived spacers.

Chart 4 Poly(platina-ynes) containing carbazole spacer units.

Fig. 10 Schematic illustration of the electronic delocalization across the carbazole-2,7-diyl and carbazole-3,6-diyl spacers.

32

Conjugated Poly(metalla-ynes)

Irrespective of the metal used, harvesting of the organic triplet emission was reported in carbazole-containing systems Pt-34 and Pt-35 (R ¼ Bu, Chart 4). The organic localized phosphorescence emission in carbazole-based spacers can be readily triggered upon metallation with group 10–12 metals.90,91 Among the type of metals, emission efficiency was in the order of Pt > Au > Hg. However, these materials were soluble and thermally stable, the T-state of the materials was very high (2.68 eV) and was not much affected by the substituent at 9-aryl group. When N atom of the carbazole unit was replaced by S,S0 -dioxide, a rare phosphorescence was observed at room temperature.92 With such spacer, di-yne shows Eg of 3.27 eV, while it was 3.14 eV for poly-ynes. An electronic communication from the carbazole (Cz) chromophore to the fluorene (F) moiety in the conjugated backbone (Pt-36 – Pt-39, Chart 4, Table 1) has been reported by Aly et al.78,93 A fast S-electron and slow T-energy transfer takes place in the backbone having carbazole and fluorene moieties. The detailed PL spectral studies indicated that carbazole and fluorene chromophores acted as both S- and T-energy donors and acceptors, respectively. For the monomers, the values of p were 0.54, 0.25 and 0.17 while f were 0.0028, 0.0021 and 0.17 for Pt-36 to Pt-38, respectively. A recent detailed thermal study of carbazole based poly-ynes confirmed that 3,6-substituted polymers are more stable than 2,7-analogs.94

14.03.3.1.2

Pt(II)-based poly(metalla-ynes) incorporating thiophene and related spacers

Similarly, investigation of a series of Pt(II) poly-ynes incorporating oligothiophene spacers (Pt-40 – Pt-42, Chart 5) indicated that the Eg decreases as the number of thiophene rings increases in the spacer.95,96 This can be attributed to the increased delocalization of p-electrons along the polymer backbone. A similar trend in band gap was reported for Pt(II) poly-ynes incorporating ethylene dioxythiophene (EDOT) spacers (Pt-43 and Pt-44, Chart 5).97 The first absorption band of Pt-43 with its maximum at 2.90 eV has its origin in the p–p transitions of the conjugated polymer backbone and is associated with the first singlet excited state S1.98 This band is red-shifted to 2.65 eV in Pt-44, indicating greater conjugation in the backbone along the polymer chain. The energy of the optical gap of both the Pt(II) di-ynes and poly-ynes decreases in going from EDOT to the bi-EDOT linker unit. The reduction in optical gap was attributed to an increased delocalization of p electrons through the more conjugated bis-EDOT spacer along the polymer backbone. The introduction of the ethylene 1,2-dioxy group in EDOT over thiophene unit (EDOT) increases the electron density of the ring. The optical gaps of the poly-ynes are lowered compared to the corresponding di-ynes, indicating that the p conjugation of the ligand extends into and through the metal center. Table 1 Entry

UV–Vis absorption and emission data at 298 and 77 K in 2-MeTHF. Absorption labs (nm)

lem (nm) 298 K

Pt-18 Pt-34 Pt-35 Pt-36 Pt-37 Pt-38 Pt-39

310, 392 264, 390 304, 332 256, 282, 290, 324, 340 258, 320, 350 250, 294, 348 272, 288, 374

77 K a

409, 431, 550, 593 408, 428, 527,a 567 408, 430 382, 400, 420, 450, long tail at ca. 500a 380, 400, 420, 488,b 530 404, 420, 540,a 565 400, 422, 545,a 587

416, 443, 552,a 595, 606, 626, 654, 672 414, 436, 524,a 572 402, 426, 460,a 483, 497, 510 395, 422, 455,a 477, 490, 504, 525 395, 420, 458,b 483, 505, 540 401, 423, 454, 530,a 570, 615 407, 434, 543,a 591

a

Emission maximum of the 0–0 phosphorescence peak.

Reused with permission from ref. Aly, S. M.; Ho, C. L.; Wong, W. Y.; Fortin, D.; Harvey, P. D. Macromolecules 2009, 42(18), 6902–6916.

Chart 5 Poly(platina-ynes) incorporating fused and non-fused oligothienyl spacer units.

Conjugated Poly(metalla-ynes)

33

Devi et al.99 studied Pt(II) poly-ynes incorporating fused oligothiophene spacers. In both the fused and non-fused thiophene containing systems, which lacks p-p interactions in their molecular structure,10,100 the energy shifted from ligand through monomer to polymer suggesting the presence of D-A interaction and a moderate degree of conjugation. But the S1 and T1 and Tn excited states of the fused system (Pt-45 and Pt-46, Chart 5) were found at higher energy.99 This was attributed to the decreased number of double bonds in the fused system and to the presence of an additional sulfur atom in spacers with the same number of double bonds. Other studies on fused and non-fused oligothiophenes also proved that the conjugation length overrides the co-planarity effect of the fused ring spacers (I.e. a greater number of conjugated double bonds lowers the energy gap).99 There is also a marked effect on absorption and emission profiles when the S-atom of Pt-40 is substituted by other main group elements (viz. Se, Te) in Pt-47.101 For example, heavy-atom substitution causes a red-shift in optical absorbance, fluorescence, and phosphorescence spectra of the polymers. Furthermore, a decrease in the phosphorescence intensity was observed in moving down the chalcogenide group (S ! Se ! Te). In addition to these, long range coupling between the phosphorus and the chalcogenophene proton through the Pt center was also confirmed.

14.03.3.1.3

Pt(II)-based poly(metalla-ynes) incorporating pyridine and related spacers

The ligand/spacer topology has a substantial effect on the emission properties of the poly-yne materials. It has been demonstrated that in a series of Pt(II) poly-ynes incorporating oligo-pyridine spacers, the poly-yne with the alkynyl groups at the 5,50 -positions is fully conjugated as compared to 6,60 - or 6,600 -positions.9,102 This was attributed to the conjugation interruption between the pyridyl rings in the latter examples.102 Similarly, topology-dependent PL properties in 2,20 - and 3,30 -substituted carbazoles,103 meta−/paraphenylene104 linked polymers has already been reported. Studies have also been carried out to explore the effect of incorporating pyridine/oligo-pyridine spacers in the organometallic framework (Pt-48 – Pt-52, Chart 6).9,102 Compared to the monopyridine counterpart (Pt-48), there was a red shift in Pt-49 while other polymers underwent a blue shift. The poly-yne incorporating 2,5-diethynyl mono pyridine showed an Eg of 3.0 eV, while 5,50 -diethynyl-2,20 -bipyridine and 6,60 -diethynyl-2,20 -bipyridine containing poly-yne showed an Eg of 2.90 and 3.30 eV, respectively. Terpyridine also showed an Eg of 3.30 eV. Both fluorescence and phosphorescence were observed in linear compounds Pt-49 and corresponding poly-yne at low temperature, while phosphorescence was absent in kinked di-ynes and poly-ynes (Pt-50 – Pt-52, Chart 6). Due to the presence of ethynyl moieties at different positions (2,5; 5,50 ; 6,60 ; 2,20 :60 ,200 ) in the pyridine backbone, different levels and extents of conjugation were reported in these molecules (Pt-49 and corresponding poly-yne fully conjugated while the rest have hindered conjugation). Furthermore, the lack of fluorescence in the oligo-yne and weak fluorescence in the poly-yne suggested an efficient metal induced ISC in these systems.

14.03.3.1.4

Pt(II)-based poly(metalla-ynes) incorporating hybrid spacers

Various studies have demonstrated that insertion of a hybrid spacers is advantageous for obtaining tunable optical and electrochemical properties. The behavior of the Pt(II) center (donor/acceptor) in platina-ynes exclusively depends on the nature of the spacer.68,98,105 p-Conjugated polymer semiconductors with D-A moieties have wide range of OE applications. By a careful selection of donor and acceptor groups, the Eg and the strength of intramolecular charge transfer (ICT) interaction between the D and the A moiety in the poly-ynes can be effectively tuned. High Eg polymers with a high T-state energy normally favor the observation of T-emissions whereas the low-gap congener is not likely to be phosphorescent even at low temperatures.106 It was found that, in a given polymer, Pt(II) fragment acts as neutral fragment when connected to an arene ring, an acceptor with a thiophene ring and a donor with a pyridine spacer.9 For instance, photophysical data suggested a significant D-A interaction between the Pt(II) centers and the conjugated ligands in polymers Pt-53 to Pt-55 (Chart 7).107 Both the fluorescence and phosphorescence as well as the absorption bands decrease in energy with increasing electronegativity of the ligands along the series from Pt-53 to Pt-55 (the more electron-withdrawing the spacer, the smaller the Eg). A change in the T1 and S1 energy was seen as the structure of the spacer varied. For instance, as the electronegativity of spacer increased, both T1 and S1 values moved to lower energy. With decreasing Eg, the lifetimes and intensities of phosphorescence are dramatically reduced and the (knr)P value increases exponentially with decreasing T1-S0 gap. Use of fused-ring spacers is a powerful strategy to produce low band-gap (Eg) conjugated polymers, as the ring fusion creates strong electronic interactions with the metal fragment. Compared to their bithiophene and terthiophene counterparts,95 Pt-56 and Pt-57 exhibited blue shifted band gaps. Low S1-S0 gap (0.14 eV), S1-T1 gaps in the range of phenylene/thiophene-based systems and blue shift of the S-state were the main features of this system.108 Computational study suggested the important role of molecular

Chart 6 Poly(platina-ynes) with kinked and linear pyridine spacers.

34

Conjugated Poly(metalla-ynes)

Chart 7 Some selected examples of poly(platina-ynes) containing hybrid spacer units.

twisting and intramolecular interactions in the molecule. Thieno[3,4-b]pyrazine is an admix of diphenylpyrazine and thiophene units and acts as a strong acceptor. Platina-ynes containing thieno[3,4-b]pyrazine (Pt-59, Chart 7) have been reported with lower Egap, and CT type transition.68,108,109 This phenomenon was ascribed to the large thieno[3,4-b]pyrazine spacer being orthogonal to the chain and capable of pulling charge out of the main chain. Further enhancement in conjugation, reduction in Eg can be realized by combining highly electron-accepting and donating units at alternating position.28,110 Schanze et al.111 found absorption in visible region and an efficient ISC in model compound as well as in poly-yne incorporating 2,1,3-benzothiadiazole moiety flanked by 2,5-thienyl or 3,4-ethylenedioxy donor units (Pt-60 and Pt-61, Chart 7). A low, tunable Eg of 1.49 to 1.97 eV was achieved using alternating D-A-D and -Pt[P(Bu)3]2- system incorporating 2,5-thienylene as donor and pyrido[3,4-b]pyrazine, 2,1,3-benzothiadiazole and thieno[3,4-b]pyrazine as acceptor.112 Poly-ynes with at least one or two acceptor units in their backbone (Pt-59 and Pt-62 – Pt-65, Chart 7) showed strong ICT, broad absorption bands (extended into the NIR region), and ambipolar redox properties including small ionization potentials (4.82–5.23 eV). Optical absorption studies indicated that ICT strength increased with acceptor strength. The incorporation of different electron acceptors demonstrated the tunability of ICT absorption bands with lmax at a broad wavelength range of 543–680 nm in thin films, as well as a tunable optical Eg of 1.49–1.97 eV in the solid state. The absorption bands were further optimized by combining two D-A-D pairs so that the visible wavelength region could be covered. Similar modulation of the PL features and applications have been reported for the polymers (Pt-66 – Pt-69, Chart 7).113–117 Thus full color tunable emission can be easily realized in poly(platina-yes) (Fig. 11A) by spacer group modification. In some cases, the presence of highly extended heteroaryl rings effectively reduced the heavy metal effect.114,115 Polymers Pt-70 and Pt-71 (Chart 7) are examples of low band gap diketopyrrolopyrrole-containing polymers with absorption covering almost entire UV–Vis region and extended into near-IR region (Fig. 11B).118 These polymers exhibited fluorescence lifetimes on the picosecond timescale, and showed efficient photo-induced electron transfer to PCBM and thus high PV performance.

Conjugated Poly(metalla-ynes)

35

Fig. 11 (A) Fluorescence spectra of poly(platina-ynes) at room temperature (B) Normalized absorption spectra of Pt-70 and Pt-71 at 298 K in DCM. Photographs of the DCM solutions containing polymers. Reproduced with permission from the ref. Ho, C.-L.; Yu, Z.-Q.; Wong, W.-Y., Chem. Soc. Rev. 2016, 45 (19), 5264-5295 and Nos, M.; Marineau-Plante, G.; Gao, D.; Durandetti, M.; Hardouin, J.; Karsenti, P.-L.; Gupta, G.; Sharma, G. D.; Harvey, P. D.; Lemouchi, C. J. Mater. Chem. C 2020, 8(7), 2363–2380.

In the examples of poly(platina-ynes) described above, we have seen that changing the organic spacers, their orientation, and functionalization are some of the leading and powerful ways to modulate the luminescence features of a polymer. In addition to these, some studies attempted other strategies. One intriguing way is the inhibition of molecular interaction between the adjacent conjugated chains leading to aggregation among the phosphorescent units.119–123 This can be realized by insulating the polymer backbone using a cyclic scaffold (such as permethylated a-cyclodextrins PM a-CDs), or side chains/ pendant groups. Schanze et al. 124 insulated the Pt(II) acetylide polymers using iptycene units and studied their luminescent behavior. It was noticed that the insulated polymer demonstrated strong phosphorescence in the solid state compared to its non-insulated counterpart. PM a-CD covered polymer film showed phosphorescence as well as enhanced mechanical properties compared to those of the uninsulated counterparts such as Pt-72 and Pt-73 (Fig. 12).125 It is also notable that such insulated polymers undergo phosphorescenceto-fluorescence conversion between complementary colors by addition of an acid. Researchers also suggested the use of sterically bulky pentiptycene-bridged congeners to circumvent the issue of aggregation and interchain interactions on the energy of the triplet state.124 Several studies have also been carried out to fine tune the photophysics, excited-state dynamics and physico-chemical behavior by varying auxiliary ligands bonded to the metal.126,127 Using different auxiliaries (PBu3, PMe3, AsBu3, Cl, SCN, CO, H, etc.), electronic structures of mono, oligo- and poly(metalla-ynes) containing Mo(II), Fe(II), Ru(II), Co(III), Rh(III), Ni(II), Pd(II), Pt(II), etc. have been investigated.27 Later, Pt(II) acetylides bearing different substituted phosphines (butyl, cyclohexyl, phenyl, ethoxy and phenoxyphosphines) were also studied.128 It was concluded that the energies of the frontier orbitals as well as the conjugation length of the backbone can be controlled by the p-donating and p-accepting character of the auxiliaries. Wang and co-workers127

Fig. 12 Chemical structures and the photographic images (excitation at 365 nm) of emission under deoxygenated conditions for polymer Pt-72 and polymer Pt-73 films on SiO2 substrates. Reproduced with permission for ref. Kaneko, S.; Masai, H.; Yokoyama, T.; Liu, M.; Tachibana, Y.; Fujihara, T.; Tsuji, Y.; Terao, J. Polymers 2020, 12(1), 244.

36

Conjugated Poly(metalla-ynes)

Chart 8 An example of a Pt(II) poly-yne bearing a chiral side chain with chiro-optical properties.

studied the effect of the phosphine ligand on the secondary structures of platinum acetylides and found that the bulkiness of the trialkylphosphine ligands plays an important role in the supramolecular polymerization and macroscopic gelation behavior. In a recent study, Sanda and co-workers129 studied the effect of the phosphine ligand on the PL and chirotropical properties of poly(platina-ynes) Pt-74 (Chart 8). When the aryl group was bonded to phosphorus, more conjugation was noted in the metallayne. Nevertheless, photoluminescence intensity was also found to be dependent on the substituent attached. For instance, the photoluminescence quantum yield was found to be in the order 4-OMe-Ph < Ph < Cy < C8H17 < C4H9. The longer alkyl chains of the phosphine ligands enhanced the thermal stability of the polymers, presumably due to the larger van der Waals interactions and/or donating ability, leading to the suppression of thermal decomposition of the polymers. In contrast, as the alkyl chain length increased, the refractive index decreased. An additional and promising strategy to optimize the optoelectronic properties is to incorporate a second (hetero) metal ion into the main backbone or as a side chain. The introduction of a second metal into a metalla-yne induces donor-acceptor interactions and influences the energy levels of the frontier orbitals and hence the absorption/emission wavelength, conductivity and excitation in the visible region, and the redox behavior. In the literature, poly(platina-ynes) incorporating heterometals such as Ir(III)130,131 Re(I)132, Ln(III)133,134 coordinated through diimine ligands have been reported. It is well established that 5,50 -bis(ethnyl)2,20 -bipyridine is fully conjugated while its 6,60 -bis(ethynyl) counterpart has broken conjugation and the former offers more desirable optical properties.102 The structural and photophysical properties of hetero-trinuclear ‘Pt-Eu-Pt’ complexes Pt-75 and Pt76 (Chart 9) have been investigated recently.134 It was noted that the Pt2-bpy unit plays an important role in the luminescence sensitization of the Eu(III) ion by d ! f orbital energy transfer. Also, the effect of alkynyl connectivity clearly affects their PL properties. For example, the bimetallic dimeric complex Pt-75 displayed typical red emission of the Eu(III) ion, with excitation confined to the UV region. On the other hand, complex Pt-76 exhibited dual emission (red and green) and can be excited in both the UV and Vis region. These complexes exhibited lifetimes (tobs) longer than analogous complexes previously reported in the literature. When the study was carried out using poly-ynes instead of di-ynes (with a second transition metal and other N,N-donors), similar observations were made. Wong and co-workers135 found that the incorporation of Re(CO)3Cl unit as a pendant side-chain into a bithiazole-containing poly(platina-yne) (Pt-77, Chart 10) generates low-energy broad absorption bands in the visible region, reduces the band gap, and had marked effect on thermal stability. The number of thienyl rings present also had a major influence on the PL (Table 2). A similar observation was made when the bithiazole was replaced by 2,2-Bipy donors Pt-78 and Pt-79 (Chart 10). It was found that the incorporation of the Re(CO)3Cl pendant functionality in the 2,20 -bipyridine-containing main-chain platina-ynes has a synergistic effect on the optical properties (red shifting the absorption profile and introducing strong long-wavelength absorptions). It was noted that the extent of the synergy depends on the topology of the ligands.

Chart 9 “d-f-d” Lanthanide-coordinated Pt(II) diy-ynes.

Conjugated Poly(metalla-ynes)

Chart 10

37

Re(I)-coordinated Pt(II) poly-ynes.

Table 2

UV–Vis Absorption and Emission Data of Pt-77 at 293 K in DCM.

Number of thienyl rings (n) With Re(I) pendant 0 1 2 Without Re(I) pendant 0 1 2

Absorption labs (nm)

Bandgap Eg (ev)

lem (nm)

F (%)

tp (ns)

230, 306, 539 229, 362, 548 229, 400, 552

2.18 1.95 1.85

582 659 729

0.05 0.09 0.1

0.22 0.44 0.34

452, 474 346, 468 378, 476

2.40 2.18 2.10

497 557 592

6.40 5.7 5.10

1.93 1.16 1.34

Reproduced with permission from ref. Li, L.; Ho, C.-L.; Wong, W.-Y. J. Organomet. Chem. 2012, 703, 43–50.

14.03.4 Non-platinum poly(metalla-ynes) Although majority of the poly(metalla-yne) materials reported together with their photo-physical characterization for use in O-E applications are those containing group 10 metals (especially Pt(II)), non-platinum metalla-mono-, di- and oligo-ynes have attracted much attention. Some other organometallic polymers incorporating low valent main group metals such as Ti, Ge, and diy-yne precursors have been reported.28 Chart 11 exemplifies some well-known poly(metalla-ynes) containing early transition metals.136 Despite the good yield and high thermal stability of Zr-1 and Zr-2, low solubility of such polymers largely hindered their PL characterization and application. Because of their use in molecular electronics, p-conjugated rigid-rod molecules or “molecular wires,” with given length and nanometer charge transport properties has attracted a lot of research interest. Lissel et al.137 reported a linear, long and well-defined rigid-rod homometallic tetranuclear complex Fe-1 (Chart 12) with tunable redox features, Fe(II) ! Fe(III) intervalence charge transfer (IVCT) absorption and a high stabilization of its oxidized forms. Prior to this work, the same group reported a tetranuclear FedW complex Fe-2 (Chart 12) with redox-active termini and absorption in the NIR region (lmax  1185 nm of the mixed valent species).138 Owing to the intrinsically open terminal binding sites, these complexes offer further functionalization with more metal centers along a fully conjugated rigid-rod like metal −organic backbone that could pave the way to materials with unique electronic properties. Though not really s-alkynyl complexes, incorporating ferrocene as pendant ligand in poly(metalla-ynes) chain drastically modulates the property and imparts unique functionality.139 For example, polymers Fe-3,140 Fe-4,141 Fe-5142 (Chart 12) are some of the classical examples of heterobimetallic poly-ynes. It has been suggested that the good air stability, sensitivity, electrochemical response and solvent-independent redox activity of such materials make them suitable for cathode-active materials in organic lithium-ion batteries. Fe-6 and Fe-7 (Chart 12) are two examples of electrochemically active143 polymers. A novel air- and moisture-stable bimetallic polyferroplatin-yne precursor Fe-8 (Chart 12) can be utilized directly as a negative resist in both electron-beam lithography and UV photolithography.144 Group 8-containing poly(metalla-ynes) Ru-1 to Ru-3, and Os-1 to Os-3 (Chart 13) along with their molecular precursors were prepared by reacting bis-trimethylstannylalkyne organic linkers.35 To overcome the issue of low solubility and to endow novel materials properties, new oligomeric systems incorporating early and late transition metal have been reported.145 Besides, linear

Chart 11

An early transition metal based metalla-ynes.

38

Conjugated Poly(metalla-ynes)

Chart 12

Fe(II) containing rigid-rod or molecular wires and polymers.

Chart 13

Ru- and Os-containing poly(metalla-ynes).

monometallic poly(metalla-ynes) containing Co146, Rh36, Au147 as well as heterobimetallic poly(metalla-ynes) containing RudPd, M-ferrocene (M ¼ Ni, Pd, Pt)142, PtdAu148, Ag149 have been reported. These materials were mostly reported during late 20th century to early 21st century and their characterization and application were hindered by low solubility of the materials. Recently, using a new Cu-free dehydrohalogenation synthetic strategy, Lissel and co-workers150 reported poly(metalla-ynes) Ru-4 (Chart 13) containing Ru(II) centers bridged in the trans-position by the organic linker 3,30 -didodecyl-5,50 -diethynyl-2,20 -bithiophene (DDBT). Such polymers have been found to be highly soluble, allowing full structural and PL characterization. Ru-5,142 Ru-6,151 and Ru-7152 (Chart 13) are some of the earlier examples of Ru(II) poly-ynes that display potentially useful electronic properties. These include push–pull interactions, extended p-conjugation and absorption extending into the visible region. It was found that the introduction of Ru(II) bis(dppe) moieties in the polymer main backbone could effectively tune the opto-electronic properties of the resulting polymer. Polymer Ru-8 (Chart 13) incorporating a Ru(II) bis(acetylide) complex and boron dipyrromethene (BODIPY) moieties is an example of Ru poly-yne with bathochromatically shifted bands and a low band gap (Fig. 13).153 Iridium complexes with acetylide ligands show limited reactivity and thus there are only a few reports of their characterization and reactivity available in the literature.154–156 Metalla-ynes containing Au(I) species are unique as they adopt a linear geometry with a coordination number of two, which renders them suitable for making rigid-rod metal-acetylide polymers with aurophilic (Au–Au) interactions.157 Au(I) acetylide compounds usually exhibit excellent transparency in the visible region, owing to the weak

Conjugated Poly(metalla-ynes)

λ .. = . λ .= ФFl. = E = E =

409, 612 nm 645 nm 0.26 1.73 eV 1.51 eV

λ .. = . λ .= ФFl. = E = = E

39

375, 634 nm 653 nm 0.08 1.52 eV 0.87 eV

Fig. 13 The effect of incorporating a Ru(II) fragment in a BODIPY polymer.

interaction of the Au 5d orbitals with p orbitals of the organic ligands. Several neutral and charged small dimeric and oligomeric Au(I) metalla-ynes have been reported with interesting PL properties22,158 Despite this, polymeric materials are relatively scarce.159 Earlier studies on Au(I) poly(metalla-ynes) were limited to solid state characterization because of their limited solubility. In a recent study, three new Au(I) poly-ynes Au-1 – Au-3 (Chart 14) have been reported along with their NLO properties.160 The absorption spectra of the ligands, monomeric complexes and polymers are similar, indicating that polymerization does not induce any obvious bathochromic effect and therefore transparency can be successfully maintained in the Au(I) poly-ynes. All the Au(I) poly-ynes show stronger triplet emission in CH2Cl2 solution compared to their corresponding Au(I) acetylide complexes, indicating their higher triplet quantum yields. Fluorescence quantum yield of the polymer varied in the order Au-2 (0.69%) > Au-1 (0.51%) > Au-3 (0.12%) It was found that polymeric products not only maintain the high transparency of the corresponding Au(I) poly-ynes like those of their corresponding molecular Au(I) acetylides, but also effectively enhances their triplet (T1) emission ability. The Au(I) poly-ynes based on fluorene and triphenylamine ligands show remarkably high OPL performance than the state-of-the-art OPL material C60. When doped into a polystyrene (PS) solid matrix, the fluorene-based Au(I) poly-yne exhibited improved OPL activity compared to that in solution. It has been suggested that the positive contribution of transition metal ions to the OPL of metalla-yne compounds generally follows the order: Pt > Au > Hg > Pd.84 In the case of Au(I) phosphine, there is a weak interaction between 5d orbitals of the Au(I) centers and p orbitals of the ethynyl ligands. Also, tetrahedral diphosphine ligands have conjugation-breaking ability. Combining these two features in a polymer is likely to improve the transparency. Based on this idea, the same group prepared and assessed OPL performance of heterobimetallic Au(I)dPt(II) poly-ynes Au-4 and Au-5 bearing carbazole (Cbz), fluorene (Flu) and triphenylamine (Tpa) spacers (Chart 14),161 Absorption studies of the polymers indicated that the maximum absorption wavelengths (lmax) and the cut-off absorption wavelengths (lcut-off) of heterometallic systems show a blue shift compared with those of the corresponding homometallic Pt(II) poly-ynes, due to the conjugation-breaking ability of the propylene spacer. All the polymers absorb strongly in UV region (Table 3), but there is negligible absorption in visible region (Fig. 14A). Also, they exhibit a better

Chart 14

Some examples of gold and platinum-containing poly(metalla-ynes).

40

Conjugated Poly(metalla-ynes)

Table 3

Photophysical data of Au(I) and heterobimetallic Au(I) – Pt(II) poly-ynes.160, 161

Complexes

Absorptiona

Emissiona 298 K/77 K

FFb

Lifetimes S1 state (ns)/ T1 state (ms)

Au-1 Au-2 Au-3 Au-4 (Ar ¼ Flu) Au-4 (Ar ¼ Tpa) Au-4 (Ar ¼ Cbz) Au-5 (Ar ¼ Flu) Au-5 (Ar ¼ Tpa) Au-5 (Ar ¼ Cbz)

276, 306, 319, 344, 362 276, 309, 335sh, 366 276, 304sh, 315, 336, 357sh, 375 270, 276, 305, 319, 345, 362, 387 264, 276, 316sh, 377 253, 262, 276sh, 317, 335 275, 305, 319, 344, 361, 384 268, 275, 314sh, 373 253, 261, 316, 334

392, 411sh, 530, 576/532, 570, 582 433/498, 527, 540, 555 383sh, 402, 438, 473/440, 470, 485 417, 438, 450, 539/548, 587, 625, 642 504/504, 545, 563 405, 425, 455, 495, 534/456, 492, 507, 527 412, 432, 449sh, 542, 584/548, 586, 619 504/503, 529, 603 402, 421, 438, 455, 503/457, 493, 507

0.51 0.69 0.12 1.03 0.62 0.27 0.66 1.69 0.3

10.0 ns (392 nm)/478.2 ms (532 nm) 1.8 ns (433 nm)/527.7 ms (498 nm) 120.0 ns (402 nm)/398.6 ms (440 nm) 0.71 ns (438 nm)/181.63 ms (548 nm) 10.37 ms (504 nm)c /130.80 ms (504 nm) 0.99 ns (455 nm)/44.31 ms (456 nm) – – –

a

Measured in dichloromethane at a concentration of ca. 10−5 M. Measured using quinine sulfate in 1.0 M H2SO4 (FF ¼ 55–56%) as the standard. c The numbers in parentheses are the emission wavelengths of the S1 and T1 states. b

Fig. 14 Absorption (A) spectra of Au-4 and Au-5. (B) Emission spectra of Au-4 and Au-5 (Flu). (C) Emission spectra of Au-4 and Au-5 (Tpa). (D) Emission spectra for Au-4 and Au-5 (Cbz). Reproduced with permission from ref. Tian, Z.; Yang, X.; Liu, B.; Zhong, D.; Zhou, G.; Wong, W.-Y. J. Mater. Chem. C 2018, 6(42), 11416–11426.

luminescence profile and transparency than homometallic Pt(II) poly-ynes. In general, poly-ynes with Flu and Cbz spacers have had more effect on emission than the Tpa (Fig. 14C and D). For instance. Au-4 (Ar ¼ Flu) exhibited stronger triplet emissions compared to Au-5 (Ar ¼ Flu), which can be associated with its lower FF (1.03%). As expected, Au-4 (Ar ¼ Cbz) exhibits stronger triplet emissions than the fluorene counterpart (Au-4, Ar ¼ Flu). On the other hand, PL spectra of Au-4 (Ar ¼ Tpa) and Au-5

Conjugated Poly(metalla-ynes)

Chart 15

41

Some examples of mercury-containing poly(metalla-ynes).

(Ar ¼ Tpa) show one emission band at 298 K (overlapped singlet and triplet emission). The reported bimetallic complexes have shown OPL features, that are much better than state-of-the-art OPL materials. In summary, this work shows that the Au(I) precursors with tetrahedral diphosphine ligands into the backbone of Pt(II) poly-ynes are promising materials for OPL applications (Table 3). Despite the fact that the Hg(II) poly-ynes were first discovered around 60 years ago,162 the first example of phosphorescence in Hg(II) poly-ynes was demonstrated by Wong and co-workers in 2003.163 Hg(II) poly-ynes such as (Hg-1 and Hg-2, Chart 15)163,164 have shown interesting opto-electronic properties; this type of metalla-yne has not been exploited significantly in materials applications. Very recently, Hg(II) metalla-yne Hg-3 (Chart 15) with an ultrahigh number-average molecular weight (Mn ¼ 328,000 and a polydispersity (PDI ¼ 1.94) has been reported.165 This visible absorbing high band gap polymer showed strong intermolecular interactions in solid, as well as excellent mechanical properties.

14.03.5 Applications 14.03.5.1 Optoelectronics 14.03.5.1.1

Bulk heterojunction polymer solar cells (PSCs)

The energy crisis is one of the most important global issues confronting humanity at present. The solution to the problem is to utilize renewable, clean and abundant energy resources. Perhaps the most readily available energy resource is solar energy. Sunlight can be converted into electrical energy using solar cells. Inorganic semiconductors have been used widely, with significant success, as the main materials used to fabricate these solar cells. However, scientists are searching for materials which are more efficient, lightweight, low-cost and allow for large area applications. Therefore, organic solar cells (OSCs) are of considerable current interest. For example, Lin’s group have recently developed an OSC in which the power conversion efficiency is over 17%.166 This result indicates that OSCs have great potential as the next generation of solar cells. Common OSCs involve the use of both donor and acceptor materials. Innovations in the development of these materials are among the critical factors that have led to the evolution of effective solar cells. Currently, the most frequently used donor materials for OSCs are poly(3-hexylthiophene) (P3HT), poly((4,8-bis[(2-ethylhexyl)oxy]benzo[1,2-b:4,5-b’]di-thiophene-2,6-diyl) (3-fluoro-2-[(2-ethylhexyl)-carbonyl]thieno[3,4-b]thiophenediyl)) (PTB7), and poly(4,8-bis(5–2-ethylhexyl)thiophen-2-yl)benzo [1,2-b:4,5-b’]dithiophene-2,6-diyl-alt-(4-(2-ethylhexyl)-3-flurothieno[3,4-b]thiophene)-2-carboxylate-2-6-diyl)] (PTB7-Th).167 However, there is a challenge to find for more versatile, low bandgap donor materials and conjugated organometallic polymers have been found to be a promising class of materials.168 The structural design can be achieved by the inclusion of various transition metals and employing different construction strategies such as D-p-A or D-p-A-p-D structures (D ¼ donor, A ¼ acceptor).117 The optoelectronic and physical properties of the materials can be tuned by using these structural units.

14.03.5.1.1.1 Development of Pt(II) poly-ynes for solar cell applications Table 4168–183 contains a summary of various metal poly-ynes that have been used in solar cell applications together with the relevant photovoltaic data. Köhler et al. first reported the photovoltaic effect of the platinum poly-yne Pt-3 (Chart 16) in 1994. The maximum quantum efficiency was only around 0.03–0.6%. In addition, the phosphorescence lifetime depended on the ratio of the acceptor C60 used. When the ratio of C60 increased, the lifetime decreased.184 Subsequently, Köhler and co-workers reported further photovoltaic studies on Pt(II) poly-ynes. They reported the Pt(II) poly-ynes (Pt-40 – Pt-42, Chart 16) with various numbers of thienyl group as the spacers. The result showed that the maximum photocurrent quantum yield was around 0.04%. The open-circuit voltages (Voc) of Pt(II) poly-ynes with one, two and three thienyl groups were 0.5, 0.75 and 0.47 V, respectively, while the fill factors (FF) of the poly-ynes were 0.32, 0.35 and 0.30, respectively. At that time, the photovoltaic effect of Pt(II) poly-ynes was not good and did not attract much attention.

42

Conjugated Poly(metalla-ynes)

Table 4

Photovoltaic performance of metal poly-yne solar cells.

Polymer

Device Structure

Voc [V]

Jsc [mA cm−2]

Fill Factor (FF)

PCEmax [%]

References

Pt-40 Pt-41 Pt-42 Pt-80 Pt-81 Pt-82 Pt-83 Pt-84 Pt-60 Pt-62 Pt-63 Pt-65 Pt-64 Pt-109 Pt-110 Pt-111 Pt-112 Pt-113 Pt-114 Pt-85 Pt-86 Pt-89 Pt-90 Pt-91 Pt-92 Pt-93 Pt-94 Pt-95 Pt-96 Pt-97 Pt-98 Pt-100 Pt-101 Pt-102 Pt-103 Pt-66 Pt-104 Pt-105 Pt-106 Pt-107 Pt-108 Pt-115 Pt-116 Pt-117 Pt-118 Pt-70 Pt-71 Pt-60(Et) Pt-70(Et) Pt-119 Pt-120 Pt-121 Pt-122 Pt-123 Hg-3 Zn-1 Zn-2 Zn-3 Zn-4

ITO/PEDOT:PSS/Pt-40:PCBM/LiF/Al ITO/PEDOT:PSS/Pt-41:PCBM/Al ITO/PEDOT:PSS/Pt-42:PCBM/Al ITO/PEDOT:PSS/Pt-80:PC71BM/LiF/Al ITO/PEDOT:PSS/Pt-81:PC71BM/LiF/Al ITO/PEDOT:PSS/Pt-82:PC71BM/LiF/Al ITO/PEDOT:PSS/Pt-83:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-84:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-60:PC71BM/LiF/Al ITO/PEDOT:PSS/Pt-62:PC71BM/LiF/Al ITO/PEDOT:PSS/Pt-63:PC71BM/LiF/Al ITO/PEDOT:PSS/Pt-65:PC71BM/LiF/Al ITO/PEDOT:PSS/Pt-64:PC71BM/LiF/Al ITO/PEDOT:PSS/Pt-109:PC71BM/LiF/Al ITO/PEDOT:PSS/Pt-110:PC71BM/LiF/Al ITO/PEDOT:PSS/Pt-111:PC71BM/LiF/Al ITO/PEDOT:PSS/Pt-112:PC71BM/LiF/Al ITO/PEDOT:PSS/Pt-113:PC71BM/LiF/Al ITO/PEDOT:PSS/Pt-114:PC71BM/LiF/Al ITO/PEDOT:PSS/Pt-85:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-86:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-89:PC70BM (1:4)/LiF/Al ITO/PEDOT:PSS/Pt-90:PC70BM (1:4)/LiF/Al ITO/PEDOT:PSS/Pt-91:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-92:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-93:PCBM (1:5)/Al ITO/PEDOT:PSS/Pt-94:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-95:PCBM (1:5)/Al ITO/PEDOT:PSS/Pt-96:PCBM (1:5)/Al ITO/PEDOT:PSS/Pt-97:PCBM (1:5)/Al ITO/PEDOT:PSS/Pt-98:PCBM (1:5)/Al ITO/PEDOT:PSS/Pt-100:PCBM (1:5)/Al ITO/PEDOT:PSS/Pt-101:PCBM (1:5)/Al ITO/PEDOT:PSS/Pt-102:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-103:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-66:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-104:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-105:PCBM/Al ITO/PEDOT:PSS/Pt-106:PCBM/Al ITO/PEDOT:PSS/Pt-107:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-108:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-115:PCBM (1:3)/MoO3/Al ITO/PEDOT:PSS/Pt-116:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-117:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-118:PCBM (1:4)/Al ITO/PEDOT:PSS/Pt-70:PC71BM (1:4)/PFN/Al ITO/PEDOT:PSS/Pt-71:PC71BM (1:4)/PFN/Al ITO/PEDOT:PSS/Pt-60:PC70BM (1:4)/TiOx/Al ITO/PEDOT:PSS/Pt-70:PC70BM (1:4)/TiOx/Al ITO/PEDOT:PSS/55:PC70BM (1:4)/TiOx/Al ITO/PEDOT:PSS/56:PC70BM (1:4)/TiOx/Al ITO/PEDOT:PSS/Pt-121:PCBM (1:5)/Al ITO/PEDOT:PSS/Pt-122:PCBM (1:5)/Al ITO/PEDOT:PSS/Pt-123:PCBM (1:5)/Al ITO/Hg-3(13 mm)/PTB7:PC71BM/MoO3/Al ITO/PEDOT:PSS/Zn-1:PCBM (1:3)/Al ITO/PEDOT:PSS/Zn-2:PCBM (1:4)/Al ITO/PEDOT:PSS/Zn-3:PCBM (1:4)/Al ITO/PEDOT:PSS/Zn-4:PCBM (1:4)/Al

0.64 0.82 0.55 0.84 0.81 0.79 0.80 0.78 0.77 0.66 0.52 0.39 0.53 0.50 0.50 0.32 0.52 0.64 0.68 0.50 0.52 0.66 0.72 0.73 0.83 0.90 0.88 0.77 0.95 0.95 0.89 0.78 0.79 0.72 0.68 0.58 0.53 0.70 0.78 0.52 0.43 0.86 0.75 0.67 0.54 0.92 0.89 3.97 2.97 1.29 0.68 0.74 0.77 0.82 0.74 0.58 0.72 0.78 0.77

0.99 15.43 1.68 7.33 8.67 9.61 4.00 4.94 9.65 2.99 2.71 0.25 2.14 1.39 0.99 0.17 0.86 2.35 4.21 2.90 2.90 4.95 2.99 0.91 2.33 6.93 6.50 1.21 2.73 4.39 7.56 3.71 4.06 1.28 2.88 1.44 2.67 0.14 1.40 1.00 2.40 2.23 5.45 1.13 2.23 12.96 16.24 0.59 0.51 0.26 0.59 3.10 5.10 5.88 13.37 1.52 2.74 3.02 3.42

0.43 0.39 0.32 0.39 0.51 0.49 0.34 0.42 0.32 0.34 0.26 0.17 0.28 0.23 0.23 0.18 0.25 0.20 0.25 0.38 0.31 0.31 0.36 0.32 0.39 0.46 0.44 0.38 0.59 0.60 0.43 0.37 0.41 0.31 0.33 0.33 0.28 0.2 0.32 0.45 0.36 0.48 0.34 0.37 0.30 0.597 0.667 0.40 0.35 0.29 0.27 0.38 0.43 0.56 0.712 0.34 0.34 0.30 0.39

0.27 4.93 0.29 2.69 3.76 4.13 1.09 1.61 2.41 0.68 0.36 0.016 0.32 0.16 0.11 0.009 0.11 0.31 0.71 0.56 0.46 1.02 0.78 0.21 0.76 2.66 2.50 0.36 1.54 2.47 2.88 1.06 1.29 0.32 0.63 0.28 0.39 0.03 0.35 0.23 0.37 0.91 1.40 0.28 0.36 7.62 9.54 0.94 0.53 0.098 0.11 0.88 1.67 2.24 9.11 0.30 0.68 0.71 1.04

169 107 170 171 171 171 117 117 112 112 112 112 112 112 112 112 112 112 112 172 172 173 173 115 115 115 115 174 174 174 174 175 175 176 176 113 113 177 177 178 178 168 179 179 179 118 118 180 180 180 180 181 181 181 165 182 183 183 183

Conjugated Poly(metalla-ynes)

43

In 2007, Wong et al. reported a Pt(II) poly-yne Pt-60 (Chart 16) with the benzodithiazole-thienyl group as the spacer.107 They discovered that the energy bandgap of the Pt poly-yne with this spacer group (Eg ¼ 1.85 eV) was lower than that with the bithienyl (2.55 eV) or the benzodithiazole group (2.20 eV). The efficient photoinduced charge separation was achieved through a charge-transfer excited state rather than the triplet state. The maximum power conversion efficiency of the bulk heterojunction (BHJ) solar cell based on Pt-60:PCBM blend was 4.1%. The Voc, short-circuit current (Jsc) and FF were 0.82 V, 13.1 mA cm−2 and 0.37, respectively. There are many parameters that can be tuned to affect the efficiency of the BHJ solar cell, for instance, the blending ratio, the thickness of active layer, choice of the of casting solvent and the type of cathode used. In order to control the morphology and phase separation of the thin film, the blending ratio is one of the essential factors. In this case, 1:4 and 1:1 Pt-60:PCBM ratios were used and their morphologies were characterized. The result showed that 1:4 was the best blending ratio because of the better phase separation. The phase separation gave the PCBM-rich domain which improved the charge transfer and carrier collection efficiency. As a result, the recombination losses were reduced and Jsc was increased, leading to a larger PCE. To optimize the efficiency of the device, the solvent effect was also studied. The degree of roughness of the film, PCBM-rich domains, morphology, and phase separation were greatly affected by the choice of casting solvent. When toluene or chlorobenzene were used, a long evaporation time was required, leading to a tough and free-standing thin film. The phase separation of the film was also dictated by the different solubility of PCBM in various solvents. The performance of BHJ solar cell using different casting solvent was toluene > chlorobenzene > chloroform  xylene. As we have mentioned before, the thickness of active layer has a great influence on the performance of the solar cell. More light can be absorbed when the thickness increases, but at the same time, it will result in a low charge carrier mobility which leads to a low charge collection. Therefore, it is important to optimize the thickness of the active layer in order to maximize the PCE of the solar cell. The Voc will be affected if the cathode is changed because the Fermi level is pinned to the reduction potential of PCBM.185 In the Pt-60:PCBM case, the difference of Voc was 0.15 V when the cathode was changed from Al to Mg:Ag. Following these successes, attempts were made to improve the performance of Pt(II) poly-ynes on organic solar cells through the structural modifications. There are two approaches for the development. The first one is to provide a uniform path for charge transport in amorphous polymers.186 This can facilitate the intramolecular charge transfer (ICT) on a p-conjugated polymer backbone and lower the Eg value for higher electron-hole mobility. Another approach is to design polymers with enhanced crystallinity to form ordered film through self-organization.187 This enables fast in-plane charge transport through the strong p-p interactions in these polymers. However, post-treatments and the structural regioregularity have a strong influence on the performance of the practical solar cell devices.171

44

Conjugated Poly(metalla-ynes)

Chart 16

Selected platinum(II) poly-ynes used in optoelectronic applications.

Conjugated Poly(metalla-ynes)

45

14.03.5.1.1.2 Structural modifications on donor-acceptor unit Baek et al. have reported three Pt(II) poly-ynes with rigid spacers Pt-80 to Pt-82 in Chart 16.171 These poly(platina-ynes) have strong charge transport properties due to the electron coupling between D and A units and also the modification of alkyl chains along the polymer backbones. According to the results, Pt-81:PCBM and Pt-82:PCBM blends exhibited higher PCE efficiency than Pt-80:PCBM. Pt-81:PCBM resulted in a PCE of 3.76% with Voc of 0.813 V, Jsc of 8.67 mA cm−2 and FF of 0.506, whereas Pt-82: PCBM exhibited a better performance (PCE ¼ 4.13%, Voc ¼ 0.787 V, Jsc ¼ 9.61 mA cm−2 and FF of 0.493). Wu et al. have synthesized six random Pt-containing D-A copolymers Pt-109 to Pt-114 (Chart 17) with two different electron acceptors.112 Varying the D-A-D pairs and composition of comonomer can manipulate the Eg and relative intensity of two ICT bands. Copolymer Pt-109 (Chart 17) with thieno[3,4-b]pyrazine and pyrido[3,4-b]pyrazine resulted in a PCE of 0.16% with Voc of 0.50 V, Jsc of 1.39 mA cm−2 and FF of 0.23. The very low FF and PCE values were due to the poor charge separation and charge transport properties. By employing different strong acceptors or donors, such as 2,3-bis(4-octyloxyphenyl)thieno[3,4-b]pyrazine, triphenylaminebenzothiadiazole, [1,2,5]thia-diazolo[3,4-i]dibenzo[a,c]phenazine, 3,4-ethylenedioxythiophene-benzothiadiazole, diketopyrrolopyrrole (DPP) and 2,1,3-benzothiadiazole-(3,4-ethylenedioxy)-2,5-thienyl in Pt-62, Pt-65 and Pt-85 to Pt-90 (Chart 16), the ICT strength of polymers and absorption coverage in the visible and near-infrared region can be enhanced.112,117,172,173,188 Qin et al. have reported a Pt(II) poly-yne Pt-89 (Chart 16) with [1,2,5]thia-diazolo[3,4-i]dibenzo[a,c]phenazine group.173 The value of PCE was 1.02% with Voc of 0.66 V, Jsc of 4.95 mA cm−2 and FF of 0.31. Wang et al. have also reported two Pt(II) poly-ynes Pt-83 and Pt-84 in Chart 16 with the triphenylamine-benzothiadiazole spacer.117 The Voc of Pt-84 was 0.78 V, Jsc was 4.94 mA cm−2, FF was 0.42 and PCE was 1.61%. He et al. reported a low bandgap BODIPYdPt conjugated polymer Pt-115 (Chart 18) for OPV devices.168 A high Voc of 0.86 V was obtained by thermally evaporated MoO3 in the device fabrication to form ohmic contact. In 2019 Pt(II) poly-yne based solar cells using Pt-116 to Pt-118 (Chart 18) were reported. This work involves the use of the DPP-based ligand which is a strong electron-withdrawing group.179 Strong p-p interactions are present in the DPP molecules due to their planar and conjugated bicyclic

Chart 17

Pt(II) copoly-ynes incorporating thienyl spacer groups.

46

Conjugated Poly(metalla-ynes)

Chart 18

Pt(II) poly-ynes incorporating BODIPY- and DPP-based ligands.

structure. The strong electron-withdrawing effect is attributed to the lactam part in the DPP unit. In addition, the 2,5-position of the DPP moiety can be modified to tune the solubility of the DPP-based polymer, leading to a solution-processable material. The most recent Pt(II)-DPP-based poly-ynes Pt-70 and Pt-71 in Chart 18 were designed and characterized by Nos et al.118 The maximum PCE of Pt-71 was up to 9.54%, with Voc of 0.89 V, Jsc of 16.24 mA cm−2 and FF of 0.667. The Pt(II) poly-ynes described above are mainly p-type semiconductors. However, in order to have a more comprehensive development of solar cell materials, n-type semiconductors are also essential. Strong acceptor molecules such as tetracyanoethylene (TCNE) have already been employed in Pt(II) acetylide small molecules and fully studied.189 Yuan et al. reported Pt(II) co-polymers Pt-60(Et), Pt-70(Et), Pt-119 and Pt-120 in Chart 19 functionalized with TCNE to study their potential as n-type semiconductors.180 They found that it was an effective way to decrease the Eg and expand the absorption coverage by cyanate functionalization on the triple bond of poly-ynes. Disappointingly, the performance of the devices was lower than expected, in which Pt(II) co-polymer Pt-119 exhibited a PCE of 0.11%, with Voc of 0.59 V, Jsc of 0.68 mA cm−2 and FF of 0.27. For the future development on chemical structure, it is necessary to realize an efficient phase separation with the TCNE-adducted polymers.

14.03.5.1.1.2.1 Change of the number of thienyl rings The absorption, charge transport and efficiency of BHJ solar cells are generally influenced by the number of thienyl rings in the organic spacers along the polymer backbone. The effect on solar cell efficiency by increasing the number of thienyl rings in polymers Pt-91 to Pt-108 (Chart 16) were studied. All the results showed that when the number of thienyl rings along the main backbone was increased, the Eg of the Pt(II) poly-ynes was lowered efficiently. For example, in Wong’s work, the Eg of Pt-94 (2.06 eV) was lower than that of Pt-92 (2.18 eV) and Pt-93 (2.22 eV), which improved its photovoltaic response and PCE.115 The Voc obtained for the best cell was 0.88 V, while the Jsc was 6.50 mA cm−2 and the FF was 0.44, resulting in the PCE of 2.50%. In Liu’s work, similar trends were observed in which Pt-98 with the highest number of thienyl rings had the smallest Eg (2.33 eV) among the polymers.174 The photovoltaic cells of Pt-98 gave the maximum PCE of 2.88% with Voc of 0.89 V, Jsc of 7.56 mA cm−2 and FF of 0.43. As mentioned above, the performance of solar cell is improved when the number of thienyl rings increases. It results in high absorption coefficients of the polymers and, at the same time, retains high voltages and charge carrier mobilities. On the other hand, the more extended p-conjugation can lead to an increase of intrachain mobilities for both electrons and holes.

Conjugated Poly(metalla-ynes)

Chart 19

47

Pt(II) copoly-ynes incorporating TCNE-functionalized ligands.

14.03.5.1.1.2.2 Pt(II) poly-ynes with multi-dimensionality The Pt(II) poly-ynes above are usually associated with the rigid rod one-dimensional (1D) structure. The high PCE values were recorded in Pt-121 to Pt-123 showed in Chart 20 (around 5–7.4%). In order to reach higher PCE of the solar cell, control of the photoactive layer morphology is one of the crucial factors. The anisotropy of the optical and charge transport properties is caused by the low dimensionality of the materials. As a result, device fabrication of 1D p-conjugated polymers becomes complicated as it requires a specific control of the molecular interaction and orientation. Therefore, a new approach is required to solve the difficulties of device fabrication. Wang et al. designed Pt(II) poly-ynes with triphenylamine-based ligands and various molecular dimensions.181 It was found that the three-dimensional (3D) polymers Pt-122 and Pt-123 showed a more significant phase separation and larger PCBM domain, while the film of 1D polymer was smooth. Moreover, when the dimensionality of polymers increased, the absorption coefficient was also increased since the polymers absorbed more solar radiation. The values of Jsc and PCE of Pt-122 and Pt-123 were two to four times larger than those of Pt-121. The maximum PCE of Pt-123 was 2.24 with Voc of 0.82 V, Jsc of 5.88 mA cm−2 and FF of 0.56. 14.03.5.1.1.3 Development of Hg(II) poly-ynes for solar cells In the previous section, most of the solar cell devices introduced are normal BHJ solar cells. However, considering the practical and commercial factors, the more useful approach is to use the inverted BHJ solar cell, which consists of air-stable metals with high work function as the anode and a photoactive layer between it and the bottom ITO cathode. But, if the photoactive layer is fabricated on the bare ITO cathode, the performance of the devices will be poor as the ohmic contact with the lowest unoccupied molecular orbital (LUMO) of the photoactive layer is blocked from the ITO cathode with high work function and hence reduces the Voc. Inserting an interlayer between the photoactive layer and ITO can decrease the work function of ITO, enhance the electron collecting properties and hence improve the performance of the device. Liu et al. reported the Hg(II) poly-ynes Hg-3 and Hg-2 in Chart 21 as the interlayer.165 There are four advantages of introducing Hg-3 as the interlayer. Multilayer devices can be constructed through all solution processes because Hg-3 was processable in orthogonal solvents and hence had a good film formation ability.

48

Conjugated Poly(metalla-ynes)

Chart 20

Examples of multi-dimensional Pt(II) poly-ynes.

Chart 21

Derivatized fluorene-incorporated Hg(II) poly-ynes for inverted BHJ solar cells.

The amino-functionalized side chain of Hg-3 can form the desired dipole with the ITO substrate which enhanced the efficiency of extraction of electron and hence reduced the work function of ITO. Polymer Hg-3 has a large Eg, so it can confine the excitons in the active layer and reduce the quenching of excitons. There is no light absorption in the visible and near-infrared region of this interlayer, which can maximize the light harvesting of the active layer. There was strong stacking between polymer chains which improved the charge transport through the Hg–Hg interaction. The maximum PCE of Hg-3 was 9.11% with Voc of 0.74, Jsc of 17.37 mA cm−2 and FF of 0.712. 14.03.5.1.1.4 Development of metalloporphyrin-based poly-ynes The use of metalloporphyrins in the solar cells has attracted much attention due to their largely planar p-conjugated structure, special electronic properties, high photochemical and thermal stability, etc. But, at the same time, these materials have low PCE values due to the limitation of light absorption of the porphyrin. Huang et al. synthesized a soluble main chain zinc porphyrin-diethynylthienothiophene (DTT) co-polymer Zn-1 (Chart 22).182 The DTT unit along the zinc porphyrin polymer backbone led to a reduction in the steric hinderance, enhancement of the light absorption, extension of p-conjugation and improvement of the charge transport property. The maximum PCE of Zn-1 was 0.30% with Voc of 0.58 V, Jsc of 1.52 mA cm−2 and FF of 0.34.

Conjugated Poly(metalla-ynes)

Chart 22

49

Zn(II) porphyrin based copoly-ynes.

As mentioned above, porphyrin-containing molecules when used as the solar cell materials, usually give low PCEs, even though they have numerous advantages. Pt(II) poly-ynes have an efficient electronic p-conjugation and delocalization along the polymer backbone as the p-orbitals of two alkyne units interact with the dxy and dyz orbitals of the platinum ion. The presence of the heavy metal atom causes strong spin-orbit coupling (SOC) and induces the intersystem crossing (ISC) which results in spin forbidden triplet emission and extension of the diffusion length of excitons. Zhan et al. tried to combine these two materials to form the Pt(II) poly-ynes Zn-2 to Zn-4 (Chart 22) with the zinc porphyrin as the building block.183 In order to extend the p-conjugation and cover the missing absorption region (430–530 nm), thiophene units were introduced into the polymer backbone. In addition, the phenyl ring on the porphyrin building block increased the molecular stacking and resulted in a better interpenetration with fullerene derivatives. The maximum PCE obtained by Zn-4 was 1.04% with Voc of 0.77 V, Jsc of 3.42 mA cm−2 and FF of 0.39. In the previous sections, D-A polymers have been discussed. The values of EHOMO, ELUMO and Eg can be tuned by using different D and A units. However, when the optimization of BHJ solar cell is obtained through the structural tuning of D and A groups, Voc, Jsc and FF will show an undesired trade off. For instance, when the EHOMO of the D-A polymer is low, high Voc value can be achieved, but the value of Jsc is limited as well. On the other hand, higher Jsc value can be attained when the Eg is small, but it also results in the elevation of EHOMO and decrease of Voc. Therefore, Chao et al. reported two D-A polymers Zn-5 and Zn-6 (Chart 23)

50

Conjugated Poly(metalla-ynes)

Chart 23

Zn(II) porphyrin based D-A and block poly-ynes.

with the porphyrin-pyrene group in the peripheral position which enhanced the light harvesting ability without endowing its original properties and solar cell performance, while the panchromatic absorber was the main chain polymer.190 The pendant porphyrin-pyrene group was the additional light harvesting unit. The strong aggregation of light harvesting unit was prevented by the substituent 2,6-bis(dodecyloxy)phenyl on the porphyrin moiety. In this case, the maximum PCE achieved by Zn-6 was 8.6% with Voc of 0.77 V, Jsc of 16.1 mA cm−2 and FF of 0.7. Achieving panchromatic absorption is difficult for organic semiconductors as the optical absorption range is often too narrow. Liu et al. tried to introduce the third component to the common D-A conjugated polymers in order to improve the properties like the energy level and Eg of the polymer and understand the relationship between current and voltage. Consequently, they optimized the molecular structures for higher device performance.191 Two block copolymers Zn-7 and Zn-8 in Chart 23 with porphyrin as the pendant unit were synthesized. The maximum PCE of Zn-8 was 7.14% with Voc of 0.84 V, Jsc of 15.2 mA cm−2 and FF of 0.559.

14.03.5.1.2

Organic light-emitting diodes (OLEDs)

Display technology has been applied in many areas, for instance in television displays, smartphones, projectors, desktop monitors, virtual reality devices, etc. Organic light-emitting diodes (OLEDs) play a significant role in these applications.192 They have attracted great interest of scientists because they have low processing temperatures, it is easy to prepare thin film for the devices, they are lightweight and can be prepared by various synthetic routes. Izumi’s group reported a thermally-activated delayed fluorescence (TADF) OLED device with external quantum efficiency (EQE) up to 11.6%, which was larger than the theoretical maximum EQE of common fluorescent emitters (5.0%) and also their analog (6.9%).193 However, when we look deeper into the mechanism of OLEDs, only 25% of the singlet excitons efficiency can be utilized and hence the maximum device efficiency can only reach 5%.194

Conjugated Poly(metalla-ynes)

51

For organic materials, only the singlet excitons can be excited and decay back to the ground state through the radiative route. This problem can be solved by the inclusion of heavy metal atoms since it can induce the SOC and hence promote the ISC. Therefore, organometallic materials can utilize not only the singlet excitons, but also the triplet excitons. The theoretical internal quantum efficiency can be up to 100%.195 In addition, in order to reduce the cost of processing and be applicable in large area applications, organometallic polymers represent one of the best choices for use in OLED devices. In the paragraphs below, different metal-containing poly-ynes for OLED devices will be discussed.116,196,197 There are numerous advantages of exploiting polymer light-emitting diodes (PLEDs) from both academic and commercial aspects. For instance, low operating voltage and power consumption, fast response time, the capability to tune the complete visible color region, ease of processing into flexible film by simple spin-coating techniques and good mechanical properties are all significant advantages.198,199 A successful strategy to obtain narrow-bandgap polymers is the construction of D-A type systems. Based on this concept, the electron-deficient groups such as benzothiadiazole and quinoxaline can be used to prepare low-bandgap Pt(II) poly-yne polymers. Wong et al. designed two soluble, luminescent conjugated Pt(II) poly-ynes Pt-68 and Pt-67 (Chart 24) with bithiazole derivatives as the functional components.116 They were found to possess good solubility in organic solvents favorable for a wide range of device processing and fabrication. Green light emission at 497 nm was emitted via optical and electrical excitation in PLEDs made from these two polymers, and for Pt-67, the maximum luminance efficiency of the device was 0.11 cd A−1 at 9 V. Poly(3,6-carbazole) was well studied compared to poly(2,7-carbazole) in light-emitting devices. However, poly(2,7-carbazole) was more suitable to prevent the aggregation of the carbazole-based polymers, which can increase the fluorescent quantum yield. It has a better solubility, processability and lower steric hindrance to fulfill a better structural organization. The hole-accepting and transporting properties can also be enhanced by the rigid biphenyl unit of carbazole and substituents at the N-position. Therefore, Ho et al. reported a Pt(II) poly-yne Pt-34 (Chart 24) with 2,7-carbazole as the spacer.196 As shown in Table 5, the PLED device had

Chart 24

Some Pt(II) poly-ynes for OLEDs.

52

Conjugated Poly(metalla-ynes)

OLED performance of metal poly-ynes.

Table 5 Polymer

Emission wavelength at 293 K [nm]

Quantum yield [%]

Current efficiency (CE) [cd A−1]

Power efficiency (PE) [lm W−1]

External quantum efficiency (EQE) [%]

References

Pt-68 Pt-67 Pt-34 Pt-125 Pt-126

494a 497a 408a, 428a, 527b, 567b 425a, 555b, 602b 390a, 427a, 555b, 599b

5.4 6.4 2.5c 0.013, 0.025c 0.013, 0.080c

– 0.11 4.66 – 0.24

– – 1.59 – 0.07

– – 1.45 – 0.11

116 116 196 197 197

a

Emission wavelength of fluorescence. Emission wavelength of phosphorescence. c Phosphorescence quantum yield. b

Table 6

OFET performance of metal poly-ynes.

Polymer

Type of semiconductor

Hole mobility [cm2 V−1 s−1]

Pt-61 Pt-80 Pt-81 Pt-82 Pt-121 Pt-122 Pt-123 Zn-1a Zn-1b Pt-127 Pt-128 Pt-129 Pt-130

p-type p-type p-type p-type p-type p-type p-type p-type p-type p-type p-type p-type p-type

1.0  10 2.5  10−3 8.9  10−3 1.0  10−3 1.90  10−5 7.86  10−5 2.77  10−5 1.1  10−4 5.4  10−5 2.1  10−2 2.8  10−2 3.84  10−4 2.89  10−4 −4

On/off ratio 2

4.8  10 5.2  104 1.5  106 1.5  106 – – – 2.0  103 1.0  103 1.0  105 8.0  104 3.3  103 2.5  102

Threshold voltage (V)

References

– – – – – – – −3.3 −1.4 – – −25 −21

171 171 171 171 181 181 181 182 182 200 200 204 204

a

No annealing. Annealed at 80  C.

b

CE of 4.66 cd A−1, PE of 1.59 lm W−1 and EQE of 1.45% when Pt-34 was used as the phosphor dopant at 5 wt%. It was also the first example that a neat triplet emission was observed in PLED under electrical excitation of the metallopolymer. The 1,3,4-oxadiazole (Ox) unit is a popular molecular chromophore in OLEDs, PLEDs and photovoltaic cells. The derivatives of Ox are excellent because of their high electron affinity and electron-transport properties in organic electroluminescent diodes. By combining Ox with the fluorene-based compound which provides a rigidly planar biphenyl unit on the backbone, the new functional Ox-based material possesses good thermal and chemical stability and high emission quantum yield. Goudreault et al. reported two Ox-fluorene-based Pt(II) poly-ynes Pt-125 and Pt-126 (Chart 24) and explored their photophysical properties and performance in PLED devices.197 White light was emitted which was relatively rare for Pt-based polymer, and the device based on polymer Pt-126 at 5 wt% phosphor dopant level had CE of 0.24 cd A−1, PE of 0.07 lm W−1 and EQE of 0.11% shown in Table 5. This work illustrated the prospect of using Pt-containing polymer materials in the design of white light PLEDs. To investigate the possibility of Pt poly-ynes as white OLEDs, Sheng et al. designed Pt-126 (Chart 24) in which Pt atoms are separated by three organic spacer units.198 They concluded that the effective SOC and ISC rate can be controlled through structural modulation. As a result, Pt-126 had a relatively larger fluorescence emission and ultrafast ISC rate, which made it a potential candidate for white OLED applications. Compared to monomer, Pt poly-ynes have a higher fraction of emissive singlet states, which is larger than the maximum efficiency of monomer as OLEDs. By investigating Pt-54 (Chart 24), Wilson et al. found an average singlet generation fraction of 57% for the polymer and 22% for the monomer.199 They suggested a spin-dependent process was favorite for singlet formation, which was a result of exchange interaction on overlapping electron and hole wavefunction.

14.03.5.1.3

Organic field-effect transistors (OFETs)

Organic field-effect transistors (OFETs) have already drawn great attention for use in electronic applications. They are one of the essential components in organic integrated circuits of various electronic devices, for instance, the display drivers, smart cards, electronic identification tags (such as radio frequency identification RFID tags), organic active-matrix displays, electronic paper, etc.200,201 OFETs are electronic devices which consist of three parts – electrodes, dielectric layers and organic semiconductors.202

Conjugated Poly(metalla-ynes)

53

The organic part of OFETs can exhibit numerous benefits, which include lower cost and lower temperature for the device fabrication, light weight and high flexibility, etc. Shinokubo’s research group reported that by substituting the sulfur atom into the perylene bisimide, the electron mobility of the OFETs can be raised to 0.4 cm2 V−1 s−1.203 However, finding the stable n-type organic materials with high electron mobility is still a challenge and hence limits the development of high performance OFETs for organic electronic applications.204 Scientists need to find a new path to solve this problem. Metal-containing polymeric materials can be one of the choices. Organometallic polymers can use the same D-A structures and try to achieve satisfactory results. In view of that, p-type polymer materials as the active layer of OFET devices were introduced, and the relative OFET performance data are summarized in Table 6. OFET is also an effective device to measure the mobility of holes or electrons, and hence to deduce the performance of solar cells. Baek et al. used the OFET to determine the hole mobility of the synthesized polymer Pt-80.171 The hole mobility and on/off ratio were greatly increased for Pt-80 (Chart 16) which had thieno[3,2-b]thiophene (TT) as a more structurally-rigid spacer, as compared to Pt-61 (Chart 16) with a thiophene ring as the spacer. On the other hand, it was found that the alkyl chains in the polymer backbone can also affect the hole mobility. The mobility of Pt-81 and Pt-82 in Chart 16 with alkyl chains on the TT units reached up to 1.0  10−3 cm2 V−1 s−1. The on/off ratio of Pt-81 (1.5  106) and Pt-82 (1.5  106) were also significantly superior to that of Pt-61 (4.8  102) and 7 (5.2  104). Huang et al. used Zn-1 in Chart 22 not only in the solar cell applications, but also as a p-type semiconductor in the OFET applications.182 In that report, they compared the OFET performance at different states: before and after annealing. Before the annealing, the hole mobility was 1.1  10−4 cm2 V−1 s−1 and the on/off ratio was 2  103. But, after annealing at 80  C, the mobility decreased to 5.4  10−5 cm2 V−1 s−1 and the on/off ratio decreased to 1  103 as well. As previously described, the mobility of charge carrier can govern the performance of solar cell devices. Therefore, Wang et al. fabricated three polymers Pt-121 to Pt-123 (Chart 21) in OFET devices for studying their charge transfer properties.181 The p-channel output characteristics were shown for all three polymers. Among them, Pt-122 showed the highest hole mobility which was 7.86  10−5 cm2 V−1 s−1, while Pt-121 and Pt-123 exhibited the hole mobility of 1.90  10−5 cm2 V−1 s−1 and 2.77  10−5 cm2 V−1 s−1, respectively. Thiazolothiazole has a rigid and planar fused ring. Therefore, its p-electron system is highly extended and has a strong p-p stacking. Those small organic molecules and polymers with thiazolothiazole unit possess high charge carrier mobilities. However, there were few reports to study metallopoly-ynes with thiazolothiazole in OFET applications. In view of that, Yan et al. reported two Pt(II) thiazolothiazole poly-ynes Pt-127 and Pt-128 shown in Chart 25.200 It was observed that when the number of thienyl rings along the polymer backbone was increased and the conjugation length was extended, the hole mobility was also increased. The hole mobility of Pt-127 was found to be 2.1  10−2 cm2 V−1 s−1 with the on/off ratio of 1  105, while for Pt-128, the corresponding value was 2.8  10−2 cm2 V−1 s−1 with the on/off ratio of 8  104. Yan et al. synthesized the naphthalene diimide (NDI)-based Pt-poly-ynes Pt-129 and Pt-130 in Chart 25 for OFET studies.204 NDI-based polymers are expected to be classical n-type semiconductors. However, the results of OFET properties showed that those Pt-poly-ynes exhibited p-type semiconductor behavior. This may be due to the increase of electron coupling between the D and A units caused by the presence of rigid heteroaromatic rings. The hole mobility of Pt-129 (3.84  10−4 cm2 V−1 s−1) was higher than that of Pt-130 (2.89  10−4 cm2 V−1 s−1). In addition, the on/off ratio of Pt-129 was significantly higher than that of Pt-130, and the threshold voltage of Pt-129 was much lower than that of Pt-130. It is likely that a rougher film surface on Pt-130 than that on Pt-129 reduced the charge transport distance. As a result, the OFET property of Pt-130 was poorer than that of Pt-129.

14.03.5.1.4

Non-linear optics (NLO)

Laser technology continues to develop quickly. The exposure to intense, sudden laser pulses may cause the damage of human eyes, optical sensors and sensitive optical components. Therefore, a good optical limiter with fast response speed and high linear transmission is required. Optical limiters are devices that are able to reduce the high intensities of optical beams to perform a high transmittance at low intensities.205,206 Nowadays, small molecules such as porphyrins,207 diacetylenes, phthalocyanines,208

Chart 25

Pt(II) poly-ynes for OFETs.

54

Conjugated Poly(metalla-ynes)

nanotubes,209 fullerenes (e.g. C60)210 and other organometallic compounds can be used as the optical limiters. Nevertheless, these materials are far from being used in practical applications because of their poor solubility and technical difficulties in fabricating devices. The largely conjugated structures of fullerenes, porphyrins and phthalocyanines result in intense colors of the materials and hence they have absorption bands in the visible light spectral region (400–700 nm). This reduces the transparency in this region which is one of the severe limitations on practical devices. In order to solve these problems, new types of materials with high transparency are needed to be developed for commercial applications. In the section below, the solution-processable homo- and heterometallic poly-ynes of the type [−M(PBu3)2dC^CArC^Cd]n for some of the group 10–12 transition metals will be discussed. These poly-ynes allow easy tailoring of their physical, optoelectronic and chemical parameters via modification of the aryl (Ar) groups. In addition, they display strong reverse saturable absorption (RSA) for nanosecond laser pulses. The mechanism of RSA is that the molecules in the ground state (S0) are excited to the first singlet excited state (S1). The molecules in S1 state can go to the first triplet excited state (T1) via ISC because of the strong SOC effect caused by the heavy metal center. The T1 state has a longer lifetime. The molecules are easily accumulated in the T1 state and absorb more laser energy to go to the higher triplet energy states. This causes the optical power limiting (OPL) effect of the metallopoly-ynes. In addition, there is another mechanism that explains the non-linear optical process which is called two-photon absorption (TPA). In this case, the molecules not only absorb one photon but two photons that are to be excited to the first singlet state. This induces the OPL process. The Z-scan method is the most common method to characterize the RSA, TPA and hence the OPL properties. The figure of merit (seff/so) parameter is one of the factors determining the OPL performance.211 The equation is shown below. seff ln T sat ¼ so ln T 0 where so represents the ground-state absorption cross-section, seff represents the effective excited-state absorption cross-section, T0 represents the linear transmittance and Tsat represents the transmittance at the saturation fluence. 14.03.5.1.4.1 Homometallic poly-ynes In order to increase the OPL efficiency, a D-p-A structure is incorporated in the conjugation path of metal poly-ynes. This facilitates the formation of ISC as described above. Zhou et al. reported the OPL properties of different platinum poly(aryleneethnylene)s Pt-18, Pt-20, Pt-22, Pt-35, Pt-41 and Pt-68 (Chart 26).83 Different spacers were employed in the polymers, for example, carbazole, bithiophene, bithiazole, fluorene, etc. Here, the linear transmittance (T0) was set to 82% upon excitation at 532 nm. The poly-ynes with strong OPL responses were Pt-35, Pt-41 and Pt-20, which outperformed C60. The other three Pt poly-ynes also showed excellent OPL properties which were similar to fullerene in performance. The optical limiting thresholds (Fth) of all six Pt(II) poly-ynes are shown in Table 7. The polymer Pt-35 has the largest Fth value which was 0.13 J cm−2. The corresponding values for Pt-18, Pt-68, Pt-41, Pt-22 and Pt-20 were 0.056, 0.12, 0.085, 0.13 and 0.063 J cm−2, respectively. High transparency in the visible light region and excellent optical power limiting ability are demonstrated in conjugated platinum(II) poly-ynes incorporating fluorene-based ethynyl ligands. In this spacer, the ethynyl unit can easily be changed from the 2,7-position to the 3,6-position which also tune the photophysical properties of Pt poly-ynes. The absorption bands are blue-shifted in fluorene with the ethynyl units at the 3,6-position compared to that at the 2,7-positions. Therefore, the Pt(II) poly-ynes incorporating fluorene with fluorene-based ethynyl ligands at the 3,6-positions will have higher transparency in the visible light region. Tian et al. reported two series of Pt poly-ynes Pt-24 to Pt-31 (Chart 26) with various substituents at the peripheral 9-position of the fluorene and the ethynyl ligands at the 2,7- and 3,6-positions.87 According to the Z-scan results, the seff/so values of Pt-24, Pt-26, Pt-28 and Pt-30 were 16.0, 14.5, 12.9 and 14.7, respectively, while for Pt-25, Pt-27, Pt-29, and Pt-31, they were 4.9, 6.0, 4.8 and 4.9, respectively. It is obvious that the OPL performance of Pt(II) poly-ynes with fluorene-based ethynyl ligands at the 2,7-position was better than that at the 3,6-position. However, according to the RSA mechanism, the one at the 3,6-position should have better OPL properties than the one at the 2,7-position. This result was ascribed to the too weak ground state absorption at 532 nm of the ethynyl ligands at the 3,6-position to causing the strong OPL response. Apart from Pt, Hg can also be used as OPL metal poly-yne materials. In 2006, Zhou et al. first reported the Hg-fluorene poly-yne Hg-2 (Chart 26) which displayed superior OPL properties with high transparency.212 As with the presence of heavy Pt atoms, Hg contributes strong SOC between the ligand p orbitals and the metal d orbitals in the poly-ynes, which led to an impressive seff/so value (20.81) in Hg-2. It was higher than that of Pt-fluorene poly-ynes Pt-131 (19.07) and Pt-132 (18.62) in the same work in Chart 26, and exceeded by five times the value in C60, a benchmark optical limiter. Moreover, these metal-fluorene poly-ynes possessed very low Fth values in the range of 0.07–0.11 J cm−2 at 92% linear transmittance, which were essential for a practical device. In 2007, Zhou et al. further explored the OPL properties of the metal-fluorene poly-yne Pd-1 (Chart 26) with Pd.84 However, because of a weaker SOC effect of Pd than that of Pt and the absence of triplet emission at room temperature, the ISC efficiency of the Pd fluorene-based poly-yne Pd-1 was reduced which resulted in a poor OPL effect (seff/so ¼ 3.62). At the same time, they investigated the effect of metal ion content on the OPL properties of metal poly-ynes. When the metal content of poly-ynes was increased and the ligand conjugation length was decreased, the enhancement of SOC effect induced more contribution of the S0 ! S1 electronic transitions on the organic ligands to cause a higher ISC. This increased the triplet yield of the poly-ynes, giving a larger OPL response. The value of seff/so for Hg-2, Pt-131 and Pt-132 were 20.81, 19.07, 18.62, respectively, which were higher than that of Hg-4 (13.72) and Pt-133 (9.76) in Chart 26.

Conjugated Poly(metalla-ynes)

55

56

Conjugated Poly(metalla-ynes)

Chart 26

Metal poly-ynes for OPL.

Small molecules containing Au(I) acetylides has shown excellent OPL performance with high transparency.84, 213 Some of the Au(I) molecules are even superior to fullerene as OPL materials. However, there are few reports on the OPL properties of Au(I) polyynes. Tian et al. reported three novel Au(I) acetylide polymers Au-6, Au-7 and Au-8 (Chart 27).160 From the Z-scan data of the solution samples, the T0 was set at 86% and the Fth of Au-6 was over 0.4 J cm−2. In order to apply these OPL materials in practice, testing the performance of the materials in the solid state is more preferable than that in the solution state. Therefore, in this research, Au-6 was doped into polystyrene to produce a model device. The OPL ability of the model complex in the device was studied under 532 nm laser irradiation. The linear transmittance was 90% and the Fth value was 0.14 J cm−2. 14.03.5.1.4.2 Heterometallic poly-ynes Despite the excellent OPL property of the Pt-fluorene poly-ynes, a narrow transparency window in the visible region limits their use in practical devices. Insertion of Hg(II) ions and Pd(II) ions in the backbone of the poly-ynes induces a blue shift in the absorption bands, which expand the transparency region of the polymer. Zhou et al. synthesized a heterobimetallic fluorene-based poly-yne Hg-5 (Chart 28) which optimized the transparency/nonlinearity trade-off and exhibited almost the same OPL effectiveness as that of the homometallic polymer Hg-2 or Pt-131 (Chart 27), in which the value of seff/so was 18.32.212 Nevertheless, introducing the Hg(II) or Pd(II) ions into the backbone of the Pt(II) poly-ynes is not the best method. The reason is that the OPL performance and transparency cannot be greatly enhanced by inserting the Pd(II) ions into the Pt(II) poly-ynes. Zhou et al. synthesized Pd-based heterobimetallic poly-ynes Pd-2 and Pd-3 (Chart 28), with the values of seff/so much lower than those of the homometallic poly-ynes.84 Due to the weak interaction between the p-orbitals of organic ligand and the 5d orbitals of Au(I) center, compounds with Au(I) acetylide motifs commonly have good transparency in the visible region of the electromagnetic spectrum. Tian et al. reported two series of Au(I)dPt(II) heterobimetallic poly-ynes and determined their OPL properties.161 From the Z-scan results, the seff/so values of Au-4(Tpa), Au-5(Tpa) and Au-6(Tpa) (Chart 29) were 9.98, 8.39 and 4.01, respectively; while for Au-4(Cbz),

Conjugated Poly(metalla-ynes)

Table 7

57

The optical limiting parameters of metal poly-ynes.

Polymer

Fth [i] [J cm−2]

T0 [ii] [%]

Sample thickness (mm)

Figure of merit value seff/so

References

Pt-35 Pt-18 Pt-68 Pt-41 Pt-22 Pt-20 Pt-24 Pt-26 Pt-28 Pt-30 Pt-25 Pt-27 Pt-29 Pt-31 Hg-2 Pt-131 Pt-132 Pd-1 Hg-4 Pt-133

0.15 0.056 0.12 0.085 0.13 0.063 – – – – – – – – 0.11 0.07 0.08 0.81 0.19 0.14

82 82 82 82 82 82 94 94 94 94 94 94 94 94 92 92 92 92 92 92

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

– – – – – – 16.0 14.5 12.9 14.7 4.9 6.0 4.8 4.9 20.81 19.07 18.62 3.20 13.72 9.76

83 83 83 83 83 83 87 87 87 87 87 87 87 87 212 213 213 84 84 84

Polymer

Fth [i] [J cm−2]

T0 [ii] [%]

Sample thickness (mm)

Figure of merit value sex/so [iii]

References

Au-6 (film) Au-6 Pd-2 Pd-3 Hg-5 Au-4(Tpa) Au-5(Tpa) Au-6(Tpa) Au-4(Cbz) Au-5(Cbz) Au-6(Cbz) Au-4(Tpa) (film) Au-4(Cbz) (film)

0.14 0.40 0.35 0.75 0.083 – – – – – – 0.18 0.20

90 78 92 92 92 89 89 89 89 89 89 81 82

0.8 1 1 1 1 1 1 1 1 1 1 0.2 0.2

– – 3.90 3.40 18.32 9.98 8.39 4.01 7.92 6.66 3.72 – –

160 160 84 84 212 161 161 161 161 161 161 161 161

i. Optical-limiting threshold, Fth, defined as the input light fluence at which the output light fluence is 50% of that predicted by linear transmittance. ii. Linear transmittance.

Au-5(Cbz) and Au-6(Cbz) (Chart 29), their values were 7.92, 6.66 and 3.72, respectively. All the heterobimetallic poly-ynes showed a better OPL performance than C60. Obviously, the seff/so values of Au-4(Tpa), Au-5(Tpa) and Au-6(Tpa) were higher than those of Au-4(Cbz), Au-5(Cbz) and Au-6(Cbz). The lower fluorescence quantum yields of Au-4(Tpa), Au-5(Tpa) and Au-6(Tpa) resulted in higher T1 quantum yields and hence enhanced the OPL performance. As mentioned previously, Au-4(Tpa) and Au-4(Cbz) were used in the prototype devices. They were doped into the polystyrene and the thickness of the prototype device was 0.2 mm. High transparency with 81% and 82% linear transmittance for Au-4(Tpa) and Au-4(Cbz) were obtained. From the Z-scan results, the Fth values of Au-4(Tpa) and Au-4(Cbz) were 0.18 and 0.21 J cm−2, respectively.

14.03.5.2 Sensors On-site continuous monitoring of numerous analytes, such as heavy metal ions, volatile organic compounds (VOCs), etc., are important to the environmental analysis and human health. Chemical sensing is one of the applications in environmental analysis.214,215 Organic materials as the chemical sensors have different merits, for example, high molar absorption coefficient, ability for structural and property modification and relatively high stability toward the environmental condition, etc.216–219 However, as we mentioned above, the organometallic materials have unique photophysical properties since the heavy metal atoms can cause the SOC and result in ISC which enables the emission lifetime to become longer than that for the pure organic materials. This phenomenon can give a great temporal resolution and the fluorescent background can be easily separated from the luminescence.220,221 This makes them potential candidates for chemical sensors. In the section below, the use of metallo(poly-ynes) as sensors will be described.

58

Conjugated Poly(metalla-ynes)

Chart 27

Au(I) poly-ynes for OPL.

Chart 28

Mixed-metal poly-ynes for OPL.

Conjugated Poly(metalla-ynes)

Chart 29

59

Au(I)dPt(II) poly-ynes for OPL.

As early as 1959, it was found that the absorption on the electrode of a quartz resonator can be used for the detection of changes in oscillation frequency. This is because the square of oscillation frequency is directly proportional to the frequency change by absorption. Hence, in 1979, the first surface acoustic wave (SAW) device was used as the gas sensor. It was also discovered that changes in optical, electrical or mass properties of Pt(II) poly-ynes upon interaction with certain chemicals makes them potential gas sensors. Caliendo et al. designed a Pt(II) poly-yne Pt-134 (Chart 30) using an arene ring as the spacer.222 The spin coating method was used to deposit Pt-134 onto the substrate. Si(001)/SiO2/Zn substrate was implemented in the SAW between two interdigital transducers (IDT) delay line. This device had a fast response of few seconds at a very low relative humidity (< 10%) and provided reliable results. It also showed a high repeatability, reproducibility and stability. Another Pt(II) poly-yne Pt-135 (Chart 30) was synthesized for the gas sensor applications.223 The spin-coating method was used to deposit Pt-135 on the quartz crystal microbalance (QCM). There were different analytes for the SAW sensors, such as trichloromethane, tetrahydrofuran, acetone, isopropyl alcohol, ethanol and methanol. For sensing the organic vapors, Pt-135 was found to be highly reversible, reproducible and sensitive at room temperature. Among these analytes, Pt-135 showed a higher sensitivity toward chloroform vapor and to alcohols with high molecular weights. The detection limits of the device toward different organic vapors were between 1.0 and 3.56 ppm at room temperature. This shows that it is a promising material for the environmental and industrial applications. The latest research on gas sensors is related to an artificial dual self-controlling system using a conjugated bimetallic polymer, in which transition metals were connected with insulated conjugations. Masai et al. designed Pt-136 (Chart 30) with Pt(II)-acetylide, Ru(II) porphyrin-pyridyl moieties, and an oligo(phenylene ethynylene) backbone insulated by permethylated a-cyclodextrins (PM a-CDs).224 This insulated conjugated bimetallic polymer acted as a CO gas sensor through self-activating and self-regulating processes. The self-activation was induced by electron transfer after excitation on the metal moieties, affording a threshold concentration by monomer-induced emission. The self-regulation was driven by ligand substitution on the metal complex, which afforded a second threshold concentration by the change of the reaction rate constant. The incorporation of the self-modulating system into irreversible reactions provides a novel molecular design for controlling the detection of chemical analytes. The p-conjugated polymers (CPs) can also be applied in chemical sensing due to the “amplified luminescence quenching” property by neutral or ionic quenchers. This property is caused by the delocalization and rapid diffusion of singlet excitons along the main chain of the polymer. Based on this feature, the CPs can become much more sensitive chemical sensors than small molecular

60

Conjugated Poly(metalla-ynes)

Chart 30

Pt(II) poly-ynes as sensors.

luminophores. In the CP, the one with bipyridine or terpyridine moieties shows the amplified luminescence quenching when it binds with metal ions.225,226 Up to now, there are not many studies on the chemical sensors based on the phosphorescence quenching compared to the fluorescence quenching. However, they are potential chemical sensors with high sensitivity due to their long lifetime and migration of triplet excitons through intra- or inter-chain pathway.227 The triplet exciton delocalization and the migration capability along the main chain can be understood to a certain extent by studying the interaction between quencher ions and Pt(II) acetylide polymers. Ogawa et al. designed the Pt(II) poly-yne Pt-137 (Chart 31) with the 2,20 -bipyridine unit which acted as the binding site of metal ions.228 The studies of absorption and phosphorescence were carried out by the addition of six different transition metal ions – Fe3+, Co2+, Ni2+, Cu2+, Zn2+ and Pd2+. The difference in quenching efficiency of the various metal ions was caused by the change of the stability constant of bipyridine metal complexes. The result showed that the most efficient process of phosphorescence quenching was through the addition of Cu2+ and Ni2+ ions. The environmental pollutants which have negative biological effects on health have been of great concern to the public. One of them is Ag+. There are various methods to determine the concentration of Ag+, and among them the fluorimetric method which is often used for the chemical sensing due to the high sensitivity, pure optical detection and fast response. There are many fluorescent chemical sensors that have been developed to detect the Ag+ ions in inorganic or mixed organic/aqueous medium. However, there are few reports on the accurate and rapid detection of Ag+ ions in pure water by utilizing water-soluble conjugated polymers. Conjugated polyelectrolytes (CPEs) are polymers that have excellent water solubility and processibility together with amplified luminescence quenching, which make them suitable for the applications of chemical and biological sensing. Qin et al. reported a water soluble Pt(II) acetylide CPE.227 The polymer Pt-138 (Chart 31) showed high selectivity and sensitivity to Ag+ ions. The Ag+ ions induced the ISC of Pt-138 from singlet to triplet states. Therefore, there was obvious color change when the

Conjugated Poly(metalla-ynes)

Chart 31

61

MSP as sensors.

Ag+ interacted with Pt-138 from colorless to yellow. Visible absorption spectroscopic method was used to quantify the concentration of the Ag+ ion colorimetrically. At low concentrations (1–5 mM), a linear relationship was found in the Stern-Volmer plot, and the detection limit was at 0.5 mM which was very low. This detection method is simpler and quicker than other phosphorimetric analytical methods since it is not limited by unfavorable oxygen quenching on the phosphorescence emission. There is an essential type of supramolecular hybrid organic/inorganic material, which is called metallo-supramolecular polymers (MSPs). The structure of polymer is formed by polymeric assemblies of reversible metal-ligand coordination. As this kind of polymers contain metal complexes on the backbone, they not only exhibit conventional organic polymer properties but also possess the properties of metallic complexes, such as redox, optical, electrochromic, catalytic and magnetic properties, etc. Stimuli-responsive character is found in MSPs because in the main chain of the polymer, they incorporate reversible and weak recognition units. One of the most popular chelating end-groups for synthesizing MSPs is terpyridine and its structural derivatives, as they are able to chelate with different transition metals. Nowadays, there are few research on the linkage of p-conjugated organometallic moieties, but they are able to affect physicochemical properties and structural arrangement of polymers. Recently, Yuan et al. designed a MSP Zn-9 (Chart 31) in which the linkage was a rigid benzothiadiazole-functionalized Pt(II) acetylide, while both end groups of the monomer consist of a terpyridine unit.228 The UV–Vis absorption spectrum showed there were two absorption peaks for the monomer located at 358 nm and 477 nm. The peak at 358 nm represented the p-p intra-ligand transition, and the peak at 477 nm revealed inter-charge transport between Pt(II) ions and benzothiadiazole units. When the Zn2+ ion was added into the monomer, the peak height at 370 nm decreased. Moreover, when the concentration of Zn2+ ion showed a stepwise increase, the fluorescence intensity decreased, and the intensity reached the minimum when the ratio of monomer and Zn2+ ion equaled unity. The supramolecular polymerization process was studied by various methods, and it was found that this MSP could be used as a fluorescent security material due to the presence of a fluorescence “ON/OFF” switch process.

14.03.5.3 Other devices 14.03.5.3.1

Memory devices

Nowadays, the development of the economy, computational science and other technologies is very fast. This causes an enormous amount of data to be generated each year. It is estimated that over 40 trillion gigabytes of data will be produced in 2020.229 Therefore, a memory device with excellent performance, and high density characteristics is required for this big data era.233 The conventional inorganic materials used in memory devices have faced the downscaling problem and physical limitation; while the organic materials can provide various advantages, for example, the high diversity of structural modifications, the low cost of device fabrication and the ability to form three-dimensional stacking in date storage applications. However, the organic memory devices have their own imperfection, such as the short retention time and the low device yield due to electrical shorting. Scientists need to find a new path to solve these problems and provide better memory device performance. Insertion of organometallic moieties can provide synergistic properties, for example, low power consumption as the maintenance of memory effect does not require an applied voltage, ease of construction of memory device as it only requires sandwiching the active materials between two electrodes, excellent redox properties of transition metals such as ferrocene (Fe2+) to ferrocenium ion (Fe3+).230 In the section below, the use of metallopolymers as memory devices will be discussed.

62

Conjugated Poly(metalla-ynes)

Chart 32

Fc-containing poly-ynes as memory devices.

Ferrocene is a substance with a unique three-dimensional structure, good solution processability and high thermal stability. In addition, the excellent redox properties of ferrocene (Fe2+) and its oxidation to the ferrocenium ion (Fe3+) makes ferrocene-containing polymers suitable for the memory device applications. Xiang et al. reported four conjugated ferrocene-containing poly(fluorenylethynylene)s Fe-9 to Fe-12 (Chart 32) with triphenylene, 9-substitiuted carbazole or thiophene units along the main chain of the polymers.230 Spin-coating methods were used to fabricate the four polymers on the memory devices. Two terminal single layer devices (ITO/polymer/Al) were used to study the electrical bistability. It was found that Fe-9, Fe10 and Fe-11 showed the flash memory effect, while Fe-12 had a “WORM” memory effect. Therefore, a change of chemical structure can easily tune the memory properties. The memory device of ITO/Fe-9/Al exhibited a low switch-on voltage (1.0 V), long retention time (1000 s), large read cycles (up to 105) and high ON/OFF current ratio (103 to 104). Altering the chemical structures can easily tune the memory properties of metallopolymers. Therefore, Cheng et al. reported two new conjugated ferrocene-containing polymers Fe-13 and Fe-14 (Chart 32) and studied their characteristics of electrical bistability.229 The polymers were spin coated on the ITO-coated glass substrate in the memory device. The memory device structure was Au/ polymer/ITO/glass. Both devices fabricated by Fe-13 and Fe-14 exhibited the WORM memory effect. The ON/OFF ratio of Fe-13 and Fe-14 was 27.29 and 55.73, respectively. The structures of material and memory devices still needed to be modified and optimized. The chemistry of ferrocene-poly-ynes remains under-developed because of the lack of exploration, and the working mechanism of ferrocene is still unclear. Therefore, more effort is needed to study in this aspect. Ferrocene-containing polyimides (PIs) are also potential candidates in memory device applications. The observation of excellent thermal properties, chemical resistance and good mechanical properties make PIs advantageous for the fabrication of memory devices. Tsai et al. reported a zinc porphyrin-based polyimide Zn-10 (Chart 33).231 In this complex, 3,30 ,4,40 -diphenylsulfone tetracarboxylic dianhydride (DSDA) acted as the acceptor unit, while the zinc porphyrin acted as the donor unit. The memory device used to characterize the memory properties was employed with the sandwich structure (ITO-coated glass/PI/Al or Au). The zinc

Conjugated Poly(metalla-ynes)

Chart 33

63

Zn-porphyrin based PI for memory device.

porphyrin metallopolymer exhibited the DRAM memory effect, which had a short retention time (30 s). There was an interesting finding that zinc metal was the key point to trigger the memory properties. The result showed that the polyimides with the zinc porphyrin unit had the memory effect, while the one with the non-metal porphyrin exhibited no memory effect.

14.03.5.3.2

Catalysts

When a sub-stoichiometric amount of an organic compound is added to the chemical reaction to accelerate the rate of the reaction, the process is called “organocatalysis.” Organic molecules acting as the catalysts have attracted huge attention. There is evidence that these types of catalysts play a critical role in the formation of a prebiotic key building block, for example, sugar.232 In addition, it is found that the organic catalysts can be applied in photocatalytic active systems. Xiang’s research group discovered that when the graphitic carbon nitride (g-C3N4) reached 1 wt% in graphene, the photocatalytic hydrogen production rate reached 451 mmol h−1 g−1.233 Moreover, it is found that the organometallic polymeric materials have a better catalytic activity compared with the organic one. In the paragraphs below, the organometallic poly-ynes as the catalytic materials will be discussed. Nowadays, people are facing serious environmental pollution problem. The growing demand for decentralized water treatment, air purification and warfare-agent removal drive researchers to develop new materials and technologies. In this area, decontamination through decentralized systems with low cost and energy demand is called advanced oxidation processes (AOPs).234 In the AOPs, photocatalysis which requires the sun as the energy source is a much cost-effective and efficient approach. In order to remove the organic and inorganic contaminants in the environment, porous materials such as conjugated microporous polymers (CMPs) 235 which act as the physical adsorbents are necessary. High surface area, ease of altering the energy level by changing porosity, and excellent chemical and thermal stability make CMPs potential candidates for photocatalysis. Nevertheless, their photocatalytic efficiency is suppressed by the fast recombination of electrons/holes that are photogenerated. Ferrocene (Fc) has a sandwich-type structure in which the Fe2+ center is coordinated to two cyclopentadienyl rings. Ferrocene-based CMPs can capture the contaminant molecules within its cavities which have a large aperture. Special photocatalytic property can be provided due to its unique electron distribution and structure. Ma et al. reported the Fc-based CMPs Fe-15 to Fe-19 (Chart 34) with both micro- and mesopores.236 These synthesized polymers were found to show superior photocatalytic performance. 99% of methylene blue (MB), a typical organic pollutant, was degraded within 120 min by using Fe-15 as the photocatalyst. It was superior to TiO2 in which only 67% of MB was degraded under the same condition. 100% of 2-chloroethyl ethyl sulfide, known as half mustard, was converted to a nontoxic product within 75 min. This CMP provides a platform for preparing a potential photocatalyst for its sustainable development. Porous organic polymers (POPs) are promising catalysts in heterogeneous catalysis with excellent catalytic efficiency, high catalyst loading and large surface area. The main mechanism of mass transfer in porous material is molecule diffusion. The rate of overall reaction process is determined by the mobility of molecules, which further decides the catalytic efficiency. POPs may encounter a problem in catalysis when the molecule size is similar to the pore dimension, in which the molecules will be confined, and the speed of diffusion will be limited. The POPs cannot fully maximize the effectiveness as the access of molecules is restricted and the transport becomes slow to/from the active site. CMP is a kind of POP which is most frequently used in heterogenous catalysis. There are various forms of CMPs, for instance, fibers, films, gels, etc. Wu et al. tried to increase the dimension of CMPs by inserting the macroporous solid to the microporous CMPs. They reported a polymer Zn-11 (Chart 35) of 5,10,15,20-tetra(4-ethynylphenyl)porphyrin in melamine foam. The result showed that the maximum surface area and pore volume were up to 684 m2 g−1 and 0.23 cm3 g−1, respectively.237 Acyl transfer reactions between N-acetylimidazole (NAI) and 3-pyridylcarbinol (3-pc) were used to determine the catalytic capability. The pure CMP powder with a high surface area could concentrate the reactant in micropores and increase the rate of reaction without altering the activation energy. For Zn-11-foam, the drawback of ordinary POPs was overcome – mass transfer was facilitated, the catalytic process was intensified and finally a high catalytic efficiency was achieved. Hydrogen should be an ideal energy carrier in the future to solve environmental issues and the energy crisis as it acts as a clean and sustainable energy fuel. Nowadays, conjugated organic polymers as organic semiconductors have received much attention on the conversion of water to hydrogen fuel. Chen et al. synthesized a series of porphyrin-based conjugated polymers Zn-12, Co-1 and

64

Conjugated Poly(metalla-ynes)

Chart 34

Fc-based CMPs for photocatalysis.

Chart 35

Zn porphyrin poly-yne in melamine foam for heterogeneous catalysis.

Conjugated Poly(metalla-ynes)

Chart 36

65

Porphyrin based metal poly-ynes as photocatalysts for H2 generation.

Co-2 (Chart 36) which acted as photocatalysts in H2 generation driven by l  400 nm visible light irradiation.238 Among them, 135 exhibited the highest photoactivity (43 mmol h−1) and the best apparent quantum yield (AQY) value (7.36%) at l ¼ 400 nm light illumination. It can be attributed to the metal-to-metal charge transfer (MMCT) from the zinc porphyrin moiety to the cobalt porphyrin moiety, which results in a more effective and fast charge separation.

14.03.5.3.3

Biologically active compounds

Biologically active compounds can be the materials used in bio-imaging, etc. There are many existing challenges in bio-imaging, for example, off-target effects in common organic dyes, the interference of the signal from the target which result in low signal-to-background responses, auto-fluorescence during the imaging, etc.239 New types of materials are required for better bio-imaging techniques. Heavy metal compounds have longer luminescence lifetimes because the heavy metals cause the SOC and hence induce the ISC which allows the molecules in the singlet excited state to go to the triplet excited state. As a result, the emitted photons always have lower energy than the excited sources and a larger Stoke shift occurs.240 Using the time-gated photoluminescence imaging to collect the long-lived signal selectively and the decay details of each pixel are reported from the luminescence lifetime imaging. Therefore, the autofluorescence signal will be separated from the long-lived signal which can greatly improve the sensing sensitivity and contrast of imaging241. Because of that, the organometallic polymers as biologically active compounds will be of interest. Due to special aromatic macrocyclic structures and optical properties of porphyrin derivatives, they have great potential from the biomedical application perspective, for example, fluorescent bioimaging, MRT, etc. The high phosphorescence quantum yield, long-lived triplet states, large Stoke shift and high photostability are observed in specific metalloporphyrins. For example, Pt(II) porphyrin and Pd(II) porphyrin have drawn great interest of scientists. In addition, if the porphyrin units are introduced into the main chain of conjugated polymers, the molar absorbance and photostability can be increased. Therefore, Bian et al. reported two metal-containing polymers Pt-139 and Pd-4 (Chart 37) with Pt(II) and Pd(II) porphyrin unit, respectively.242 The Pt(II) porphyrin-containing polymer Pt-139 was treated with the HeLa cells. Blue and red fluorescence was observed in the HeLa cell due to the presence of fluorene and porphyrin units in the backbone of the conjugated polymer. It was also

66

Conjugated Poly(metalla-ynes)

Chart 37

Porphyrin-based metal poly-ynes for treating HeLa cells.

observed that the fluorescence intensity of HeLa cell was relatively low when they were treated with Pd-4. This is because the large average size of the polymer leads to a small amount of compound that can go into the cancer cell. As Pt-139 exhibited a more favorable particle size, they had a better biocompatibility and were more specific to the HeLa cancer cells.

14.03.6 Summary The chemistry of conjugated poly(metalla-ynes) and their di-yne precursors has developed extensively over the last decade because of the driver to find new materials for use in the electronics industry, as catalysts or as biologically active materials.12,13 The greatest attention has been paid to compounds of platinum and the other heavy transition metals on the right-hand side of the d-block. In particular, compounds of Pt(II) have been attractive because of the square planar geometry of the d8 Pt(II) metal center, that favors the formation of poly(metalla-ynes) with the acetylene units occupying trans sites on the Pt(II) resulting in a linear arrangement of the extended molecules and having the general formula [dPt(L)ndC^CdRdC^Cd]1 (R ¼ organic group).24 The very slow kinetics exhibited by Pt(II) compounds also means that geometry changes around the metal center are unlikely. Additionally, the late heavy transition metals exhibit high spin-orbit coupling that leads to relatively efficient intersystem crossing and give access to excited triplet states providing a route to efficient phosphorescence for the poly(metalla-ynes),22 and the chemistry of Ir, Au25 and Hg-containing26 poly(metalla-ynes) has been developed partially to take advantage of this feature. The access to the triplet states is a distinct advantage for some applications over the purely organic counterparts where only excited singlet states can be accessed and only fluorescence is possible. The most significant developments in the design and synthesis of poly(metalla-ynes) have involved the expansion of the organic and organometallic groups that have been incorporated into the spacer group between the acetylene groups bonded to the heavy transition metal centers. The careful choice of these R groups (in the formula [dM(L)ndC^CdRdC^Cd]1 has the largest effect on tuning the properties of the resultant materials. For example, for the Pt(II) systems, the spacer groups have included phenylenes, fluorenes, carbazoles, thiophene-based ligands, pyridine-based ligands and hybrid spacers.24 The variations in delocalization of the bonding (conjugation length) within these spacer groups, through the acetylene groups to the metal centers alters the electronic properties of the materials. Key features involved in tuning the electronic properties is the planarity or non-planarity of the spacer groups and the position of substitution when aromatic groups are involved. The advances in the synthesis and tuning of the properties of poly(metalla-ynes) has led to their use in a range of applications. The most advanced area where these materials have found use is that of optoelectronics. Both Pt and Hg-based systems have been used in solar cells,107, 165 and metalloporphyrin-based poly-yne systems have also been developed.183 The research has been further expanded to include the use of platinum poly-ynes in OLEDs and organic filed-effect transistors (OFETs). The 3rd order non-linear optical responses of platinum poly-ynes has also been identified and their optical limiting properties studied.87, 212 Poly(metallaynes) have also found use in sensor applications monitoring the presence of volatile organic compounds and gases using the phosphorescent properties of these materials. Recently, poly-ynes with ferrocene derivatives in the polymer side chains have shown potential as a component of memory devices, utilizing the ferrocene/ferrocenium couple.230 In other areas poly-(metalla-ynes) have shown promise as catalysts. The incorporation of ferrocene-based poly-ynes in conjugated microporous polymers has shown promise in photocatalytic processes for the removal of contaminants.240 Related porous organic polymers have been used as heterogeneous catalysts.240 Finally, porphyrin-based metal poly-ynes have been used successfully in imaging studies of biologically active compounds, the imaging being based on the fluorescence properties of the metal poly-ynes.242 With the expansion of the design and use in a range of “real world” applications over the last decade it is likely that the chemistry and properties of poly(metalla-ynes) will remain at the forefront of active research for decades to come and contribute to the development of a range of advanced functional materials.

Conjugated Poly(metalla-ynes)

67

Acknowledgments The authors (AH, MSK and WYW) extend their appreciation to the Deputy for Research & Innovation, Ministry of Education in Saudi Arabia for funding this research work through the project number RDO-2001. MSK also acknowledges the Ministry of Education, Research and Innovation (MoHERI0, Oman for financial support (Grant no. RC/RG-SCI/CHEM/20/01). WYW also thanks the financial support from the Hong Kong Research Grants Council (PolyU 153061/18P), RGC Senior Research Fellowship Scheme (SRFS2021-5S01). Hong Kong Polytechnic University (1-ZE1C), the Research Institute for Smart Energy (CDA2) and Ms. Clarea Au for the Endowed Professorship in Energy (847S). PRR is grateful to the Engineering and Physical Sciences Research Council (UK) for continued support (Grant no. EP/K004956/1).

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53.

Chatt, J.; Shaw, B. J. Chem. Soc. (Resumed) 1962, 5075–5084. Sonogashira, K.; Ohga, K.; Takahashi, S.; Hagihara, N. J. Organomet. Chem. 1980, 188 (2), 237–243. Sonogashira, K.; Yatake, T.; Tohda, Y.; Takahashi, S.; Hagihara, N. J. Chem. Soc., Chem. Commun. 1977, 9, 291–292. Takahashi, S.; Kariya, M.; Yatake, T.; Sonogashira, K.; Hagihara, N. Macromolecules 1978, 11 (6), 1063–1066. Jura, M.; Raithby, P. R.; Wilson, P. J. Chapter 3 Metal Acetylide Complexes, Oligomers and Polymers in Photofunctional Materials Chemistry. In Molecular Design and Applications of Photofunctional Polymers and Materials, The Royal Society of Chemistry, 2012; pp 56–84. Long, N. J. Angew. Chem. Int. Ed. 1995, 34 (1), 21–38. Yam, V. W. W.; Wong, K. M. C. Luminescent Molecular Rods - Transition Metal Alkynyl Complexes. In Molecular Wires: From Design to Properties; DeCola, L., Ed., 2005 pp 1–32. Blau, W. J.; Byrne, H. J.; Cardin, D. J.; Davey, A. P. J. Mater. Chem. 1991, 1 (2), 245–249. Chawdhury, N.; Köhler, A.; Friend, R. H.; Younus, M.; Long, N. J.; Raithby, P. R.; Lewis, J. Macromolecules 1998, 31 (3), 722–727. Chawdhury, N.; Younus, M.; Raithby, P. R.; Lewis, J.; Friend, R. H. Opt. Mater. 1998, 9 (1-4), 498–501. Lhost, O.; Toussaint, J. M.; Bredas, J. L.; Wittmann, H. F.; Fuhrmann, K.; Friend, R. H.; Khan, M. S.; Lewis, J. Synth. Met. 1993, 57 (2-3), 4525–4530. Haque, A.; Xu, L.; Al-Balushi, R. A.; Al-Suti, M. K.; Ilmi, R.; Guo, Z.; Khan, M. S.; Wong, W. Y.; Raithby, P. R. Chem. Soc. Rev. 2019, 48 (23), 5547–5563. Xu, L. L.; Ho, C. L.; Liu, L.; Wong, W. Y. Coord. Chem. Rev. 2018, 373, 233–257. Long, N. J.; Williams, C. K. Angew. Chem. 2003, 42, 2586–2617. Nguyen, P.; Gomez-Elipe, P.; Manners, I. Chem. Rev. 1999, 99 (6), 1515–1548. Abel, E. W.; Stone, F. G. A.; Wilkinson, G. Comprehensive Organometallic Chemistry. Cumulative Indexes Pergamon: Oxford, 1995; vol. 14. Wilkinson, G.; Stone, F. G. A.; Abel, E. W. Comprehensive Organometallic Chemistry: The Synthesis, Reactions and Structures of Organometallic Compounds; Pergamon: Oxford, 1982. Crabtree, R. H.; Mingos, D. M. P. Comprehensive Organometallic Chemistry III; Pergamon: Oxford, 2007. Yam, V. W. W.; Lo, K. K. W.; Wong, K. M. C. J. Organomet. Chem. 1999, 578 (1-2), 3–30. Manna, J.; Kuk, R. J.; Dallinger, R. F.; Hopkins, M. D. J. Am. Chem. Soc. 1994, 116 (21), 9793–9794. Manna, J.; John, K. J.; Hopkins, M. D. The Bonding of Metal Alkynyl Complexes; Elsevier Science: California, USA, 1995; vol. 38. Zhou, G.-J.; Wong, W.-Y. Chem. Soc. Rev. 2011, 40 (5), 2541–2566. Brooks, J.; Babayan, Y.; Lamansky, S.; Djurovich, P. I.; Tsyba, I.; Bau, R.; Thompson, M. E. Inorg. Chem. 2002, 41 (12), 3055–3066. Haque, A.; Al-Balushi, R. A.; Al-Busaidi, I. J.; Khan, M. S.; Raithby, P. R. Chem. Rev. 2018, 118 (18), 8474–8597. Yam, V. W. W.; Law, A. S. Y. Coord. Chem. Rev. 2020, 414, 213298–213307. Liu, L.; Wong, W. Y.; Ho, C. L. Aust. J. Chem. 2006, 59 (7), 434–438. Frapper, G.; Kertesz, M. Inorg. Chem. 1993, 32 (5), 732–740. Wong, W.-Y.; Ho, C.-L. Coord. Chem. Rev. 2006, 250, 2627–2690. Ho, C.-L.; Yu, Z.-Q.; Wong, W.-Y. Chem. Soc. Rev. 2016, 45 (19), 5264–5295. Sonogashira, K.; Fujikura, Y.; Yatake, T.; Toyoshima, N.; Takahashi, S.; Hagihara, N. J. Organomet. Chem. 1978, 145 (1), 101–108. Fratoddi, I.; Battocchio, C.; Groia, A. L.; Russo, M. V. J. Polym. Sci. A Polym. Chem. 2007, 45 (15), 3311–3329. Polzonetti, G.; Iucci, G.; Altamura, P.; Ferri, A.; Paolucci, G.; Goldoni, A. Surf. Interface Anal. 2002, 34, 588. Russo, M.; Iucci, G.; Ferro, D. Polymer 1993, 34 (2), 257–261. Zheng, Q.; Schneider, J. F.; Amini, H.; Hampel, F.; Gladysz, J. A. Dalton Trans. 2019, 48 (17), 5800–5816. Atherton, Z.; Faulkner, C. W.; Ingham, S. L.; Kakkar, A. K.; Khan, M. S.; Lewis, J.; Long, N. J.; Raithby, P. R. J. Organomet. Chem. 1993, 462 (1-2), 265–270. Davies, S. J.; Johnson, B. F. G.; Khan, M. S.; Lewis, J. J. Chem. Soc., Chem. Commun. 1991, 3, 187–188. Fyfe, H. B.; Mlekuz, M.; Zargarian, D.; Taylor, N. J.; Marder, T. B. J. Chem. Soc., Chem. Commun. 1991, (3), 188–190. Haquette, P.; Pirio, N.; Touchard, D.; Toupet, L.; Dixneuf, P. H. J. Chem. Soc., Chem. Commun. 1993, (2), 163–165. Field, L. D.; George, A. V. J. Organomet. Chem. 1993, 454 (1-2), 217–219. Hackenberg, J. D.; Kundu, S.; Emge, T. J.; Krogh-Jespersen, K.; Goldman, A. S. J. Am. Chem. Soc. 2014, 136 (25), 8891–8894. Kumar, A.; Feller, M.; Ben-David, Y.; Diskin-Posner, Y.; Milstein, D. Chem. Commun. 2018, 54 (42), 5365–5368. Mazzacano, T. J.; Leon, N. J.; Waldhart, G. W.; Mankad, N. P. Dalton Trans. 2017, 46 (17), 5518–5521. Wilson, G. L.; Abraha, M.; Krause, J. A.; Guan, H. Dalton Trans. 2015, 44 (27), 12128–12136. Pigulski, B.; Jarszak, A.; Szafert, S. Dalton Trans. 2018, 47 (47), 17046–17054. Clair, S.; de Oteyza, D. G. Chem. Rev. 2019, 119 (7), 4717–4776. Zhang, Y.-Q.; Paintner, T.; Hellwig, R.; Haag, F.; Allegretti, F.; Feulner, P.; Klyatskaya, S.; Ruben, M.; Seitsonen, A. P.; Barth, J. V. J. Am. Chem. Soc. 2019, 141 (13), 5087–5091. Shu, C.-H.; He, Y.; Zhang, R.-X.; Chen, J.-L.; Wang, A.; Liu, P.-N. J. Am. Chem. Soc. 2020, 142 (39), 16579–16586. Kokado, K.; Tokoro, Y.; Chujo, Y. Macromolecules 2009, 42 (8), 2925–2930. Mishra, A.; Ma, C. Q.; Bauerle, P. Chem. Rev. 2009, 109 (3), 1141–1276. Chronopoulou, L.; Scaramuzzo, F. A.; Fioravanti, R.; di Nitto, A.; Cerra, S.; Palocci, C.; Fratoddi, I. Int. J. Biol. Macromol. 2020, 146, 790–797. Ho, C.-L.; Wong, W.-Y. Coord. Chem. Rev. 2011, 255 (21), 2469–2502. Ho, C.-L.; Wong, W.-Y. Coord. Chem. Rev. 2013, 257 (9), 1614–1649. Wong, W.-Y.; Ho, C.-L. Acc. Chem. Res. 2010, 43, 1246–1256.

68

54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116.

Conjugated Poly(metalla-ynes)

Ren, T. Chem. Commun. 2016, 52 (16), 3271–3279. Albrecht, M.; van Koten, G. Angew. Chem. Int. Ed. 2001, 40 (20), 3750–3781. Cooper, T. M.; Krein, D. M.; Burke, A. R.; McLean, D. G.; Rogers, J. E.; Slagle, J. E. J. Phys. Chem. A 2006, 110 (50), 13370–13378. Ramakrishna, G.; Goodson, T., III; Rogers-Haley, J. E.; Cooper, T. M.; McLean, D. G.; Urbas, A. J. Phys. Chem. C 2008, 113 (3), 1060–1066. Rogers, J. E.; Cooper, T. M.; Fleitz, P. A.; Glass, D. J.; McLean, D. G. J. Phys. Chem. A 2002, 106 (43), 10108–10115. Rogers, J. E.; Slagle, J. E.; Krein, D. M.; Burke, A. R.; Hall, B. C.; Fratini, A.; McLean, D. G.; Fleitz, P. A.; Cooper, T. M.; Drobizhev, M. Inorg. Chem. 2007, 46 (16), 6483–6494. Wong, W.-Y. Dalton Trans. 2007, 40, 4495–4510. Mauro, M.; Aliprandi, A.; Septiadi, D.; Kehr, N. S.; De Cola, L. Chem. Soc. Rev. 2014, 43 (12), 4144–4166. Williams, K. A.; Boydston, A. J.; Bielawski, C. W. Chem. Soc. Rev. 2007, 36 (5), 729–744. Yam, V. W. Nat. Chem. 2010, 2 (9), 790. Silverman, E. E.; Cardolaccia, T.; Zhao, X.; Kim, K.-Y.; Haskins-Glusac, K.; Schanze, K. S. Coord. Chem. Rev. 2005, 249 (13-14), 1491–1500. Ahmed, M. K.; Rahman, M. M.; Naher, M.; Mehdi, S. S.; Khan, A. R.; Khan, M. M. R.; Islam, S. S.; Younus, M.; Wedler, S.; Bagnich, S. Macromol. Chem. Phys. 2019, 220 (8), 1800494. Markwell, R. D.; Butler, I. S.; Kakkar, A. K.; Khan, M. S.; Al-Zakwani, Z. H.; Lewis, J. Organometallics 1996, 15 (9), 2331–2337. Wilson, J. S.; Chawdhury, N.; Al-Mandhary, M. R.; Younus, M.; Khan, M. S.; Raithby, P. R.; Köhler, A.; Friend, R. H. J. Am. Chem. Soc. 2001, 123 (38), 9412–9417. Köhler, A.; Wilson, J. S.; Friend, R. H.; Al-Suti, M. K.; Khan, M. S.; Gerhard, A.; Bässler, H. J. Chem. Phys. 2002, 116 (21), 9457–9463. Khan, M. S.; Al-Mandhary, M. R. A.; Al-Suti, M. K.; Corcoran, T. C.; Al-Mahrooqi, Y.; Attfield, J. P.; Feeder, N.; David, W. I. F.; Shankland, K.; Friend, R. H.; Köhler, A.; Marseglia, E. A.; Tedesco, E.; Tang, C. C.; Raithby, P. R.; Collings, J. C.; Roscoe, K. P.; Batsanov, A. S.; Stimson, L. M.; Marder, T. B. New J. Chem. 2003, 27 (1), 140–149. Poon, S. Y.; Wong, W. Y.; Cheah, K. W.; Shi, J. X. Chem. Eur. J. 2006, 12 (9), 2550–2563. Juvenal, F.; Lei, H.; Karsenti, P.-L.; Harvey, P. D. J. Organomet. Chem. 2019, 896, 24–31. Juvenal, F.; Lei, H.; Schlachter, A.; Karsenti, P.-L.; Harvey, P. D. J. Phys. Chem. C 2019, 123 (9), 5289–5302. Dhindsa, J. S.; Maar, R. R.; Barbon, S. M.; Avilés, M. O.; Powell, Z. K.; Lagugné-Labarthet, F.; Gilroy, J. B. Chem. Commun. 2018, 54 (50), 6899–6902. Dhindsa, J. S.; Melenbacher, A.; Barbon, S. M.; Stillman, M. J.; Gilroy, J. B. Dalton Trans. 2020, 49, 16133–16142. Yu, W.-L.; Pei, J.; Huang, W.; Cao, Y.; Heeger, A. J. Chem. Commun. 1999, 18, 1837–1838. Pschirer, N. G.; Bunz, U. H. F. Macromolecules 2000, 33 (11), 3961–3963. Klärner, G.; Davey, M. H.; Chen, W. D.; Scott, J. C.; Miller, R. D. Adv. Mater. 1998, 10 (13), 993–997. Aly, S. M.; Ho, C.-L.; Wong, W.-Y.; Fortin, D.; Harvey, P. D. Macromolecules 2009, 42 (18), 6902–6916. Lewis, J.; Raithby, P. R.; Wong, W.-Y. J. Organomet. Chem. 1998, 556 (1–2), 219–228. Wong, W. J. Chem. Soc, Dalton Trans. 1998, 2761. Wong, W. Y.; Choi, K. H.; Lu, G. L.; Shi, J. X. Macromol. Rapid Commun. 2001, 22 (6), 461–465. Wong, W.-Y.; Lu, G.-L.; Choi, K.-H.; Shi, J.-X. Macromolecules 2002, 35 (9), 3506–3513. Zhou, G.-J.; Wong, W.-Y.; Cui, D.; Ye, C. Chem. Mater. 2005, 17 (20), 5209–5217. Zhou, G. J.; Wong, W. Y.; Ye, C.; Lin, Z. Adv. Funct. Mater. 2007, 17 (6), 963–975. Liu, L.; Wong, W.-Y.; Poon, S.-Y.; Shi, J.-X.; Cheah, K.-W.; Lin, Z. Chem. Mater. 2006, 18 (5), 1369–1378. Wong, W.-Y.; Liu, L.; Poon, S.-Y.; Choi, K.-H.; Cheah, K.-W.; Shi, J.-X. Macromolecules 2004, 37 (12), 4496–4504. Tian, Z.; Yang, X.; Liu, B.; Zhong, D.; Zhou, G. J. Organomet. Chem. 2019, 895, 28–36. Du, Y.; Dong, N.; Zhang, M.; Jiang, B.; Sun, N.; Luo, Y.; Zhang, S.; Wang, G.; Wang, J. J. Mater. Chem. C 2018, 6 (27), 7317–7325. Zhang, N.; Hayer, A.; Al-Suti, M. K.; Al-Belushi, R. A.; Khan, M. S.; Köhler, A. J. Chem. Phys. 2006, 124 (24), 244701. Liu, L.; Wong, W.-Y.; Shi, J.-X.; Cheah, K.-W.; Lee, T.-H.; Leung, L. M. J. Organomet. Chem. 2006, 691 (19), 4028–4041. Liu, L.; Wong, W. Y.; Poon, S.-Y.; Cheah, K.-W. J. Inorg. Orgnomet. Polym. Mater. 2006, 15, 555. Ho, C.-L.; Poon, S.-Y.; Lo, P.-K.; Wong, M.-S.; Wong, W.-Y. J. Inorg. Organomet. Polym. Mater. 2013, 23 (1), 206–215. Aly, S. M.; Ho, C. L.; Fortin, D.; Wong, W. Y.; Abd-El-Aziz, A. S.; Harvey, P. D. Chemistry 2008, 14 (27), 8341–8352. Lisa, G.; Yoshitake, Y.; Michinobu, T. J. Appl. Polym. Sci. 2019, 136 (24), 47639. Chawdhury, N.; Köhler, A.; Friend, R.; Wong, W.-Y.; Lewis, J.; Younus, M.; Raithby, P.; Corcoran, T.; Al-Mandhary, M.; Khan, M. J. Chem. Phys. 1999, 110 (10), 4963–4970. Lewis, J.; Long, N. J.; Raithby, P. R.; Shields, G. P.; Wong, W.-Y.; Younus, M. J. Chem. Soc., Dalton Trans. 1997, (22), 4283–4288. Khan, M. S.; Al-Suti, M. K.; Shah, H. H.; Al-Humaimi, S.; Al-Battashi, F. R.; Bjernemose, J. K.; Male, L.; Raithby, P. R.; Zhang, N.; Köhler, A. Dalton Trans. 2011, 40 (39), 10174–10183. Wilson, J.; Köhler, A.; Friend, R.; Al-Suti, M.; Al-Mandhary, M.; Khan, M.; Raithby, P. J. Chem. Phys. 2000, 113 (17), 7627–7634. Sudha Devi, L.; Al-Suti, M. K.; Zhang, N.; Teat, S. J.; Male, L.; Sparkes, H. A.; Raithby, P. R.; Khan, M. S.; Köhler, A. Macromolecules 2009, 42 (4), 1131–1141. Li, P.; Ahrens, B.; Feeder, N.; Raithby, P. R.; Teat, S. J.; Khan, M. S. Dalton Trans. 2005, (5), 874–883. Mahrok, A. K.; Carrera, E. I.; Tilley, A. J.; Ye, S.; Seferos, D. S. Chem. Commun. 2015, 51 (25), 5475–5478. Khan, M. S.; Al-Mandhary, M. R.; Al-Suti, M. K.; Hisahm, A. K.; Raithby, P. R.; Ahrens, B.; Mahon, M. F.; Male, L.; Marseglia, E. A.; Tedesco, E. J. Chem. Soc., Dalton Trans. 2002, 7, 1358–1368. Al-Balushi, R. A.; Haque, A.; Jayapal, M.; Al-Suti, M. K.; Husband, J.; Khan, M. S.; Koentjoro, O. F.; Molloy, K. C.; Skelton, J. M.; Raithby, P. R. Inorg. Chem. 2016, 55, 6465–6480. Miyagi, Y.; Shibutani, Y.; Otaki, Y.; Sanda, F. Polym. Chem. 2016, 7 (5), 1070–1078. Köhler, A.; Younus, M.; Al-Mandhary, M. R. A.; Raithby, P. R.; Khan, M. S.; Friend, R. H. Synth. Met. 1999, 101 (1), 246–247. Younus, M.; Köhler, A.; Cron, S.; Chawdhury, N.; Al-Mandhary, M. R. A.; Khan, M. S.; Lewis, J.; Long, N. J.; Friend, R. H.; Raithby, P. R. Angew. Chem. Int. Ed. 1998, 37 (21), 3036–3039. Wong, W.-Y.; Wang, X.-Z.; He, Z.; Djurišic, A. B.; Yip, C.-T.; Cheung, K.-Y.; Wang, H.; Mak, C. S.; Chan, W.-K. Nat. Mater. 2007, 6 (7), 521–527. Khan, M. S.; Al-Suti, M. K.; Al-Mandhary, M. R. A.; Ahrens, B.; Bjernemose, J. K.; Mahon, M. F.; Male, L.; Raithby, P. R.; Friend, R. H.; Köhler, A. Dalton Trans. 2003, (1), 65–73. Khan, M. S.; Al-Mandhary, M. R.; Al-Suti, M. K.; Feeder, N.; Nahar, S.; Köhler, A.; Friend, R. H.; Wilson, P. J.; Raithby, P. R. J. Chem. Soc., Dalton Trans. 2002, (12), 2441–2448. Cinar, M. E.; Ozturk, T. Chem. Rev. 2015, 115 (9), 3036–3140. Mei, J.; Ogawa, K.; Kim, Y.-G.; Heston, N. C.; Arenas, D. J.; Nasrollahi, Z.; McCarley, T. D.; Tanner, D. B.; Reynolds, J. R.; Schanze, K. S. ACS Appl. Mater. Interfaces 2009, 1, 1150. Wu, P.-T.; Bull, T.; Kim, F. S.; Luscombe, C. K.; Jenekhe, S. A. Macromolecules 2009, 42 (3), 671–681. Zhan, H.; Wong, W.-Y.; Ng, A.; Djurišic, A. B.; Chan, W.-K. J. Organomet. Chem. 2011, 696 (25), 4112–4120. Wang, X.-Z.; Wong, W.-Y. Polym. Prepr. 2009, 50 (2), 307. Wong, W. Y.; Wang, X. Z.; He, Z.; Chan, K. K.; Djurisic, A. B.; Cheung, K. Y.; Yip, C. T.; Ng, A. M.; Xi, Y. Y.; Mak, C. S.; Chan, W. K. J. Am. Chem. Soc. 2007, 129 (46), 14372–14380. Wong, W. Y.; Zhou, G. J.; He, Z.; Cheung, K. Y.; Ng, A. M. C.; Djurišic, A. B.; Chan, W. K. Macromol. Chem. Phys. 2008, 209 (13), 1319–1332.

Conjugated Poly(metalla-ynes)

69

117. Wang, Q.; Wong, W.-Y. Polym. Chem. 2011, 2 (2), 432–440. 118. Nos, M.; Marineau-Plante, G.; Gao, D.; Durandetti, M.; Hardouin, J.; Karsenti, P.-L.; Gupta, G.; Sharma, G. D.; Harvey, P. D.; Lemouchi, C. J. Mater. Chem. C 2020, 8 (7), 2363–2380. 119. Masai, H.; Terao, J.; Tsuji, Y. Tetrahedron Lett. 2014, 55 (30), 4035–4043. 120. Hosomi, T.; Masai, H.; Fujihara, T.; Tsuji, Y.; Terao, J. Angew. Chem. Int. Ed. 2016, 55 (43), 13427–13431. 121. Miyagishi, H. V.; Tamaki, T.; Masai, H.; Terao, J. Molecules 2019, 24 (7), 1301. 122. Russell, G. M.; Inamori, D.; Masai, H.; Tamaki, T.; Terao, J. Polym. Chem. 2019, 10 (39), 5280–5284. 123. Masai, H.; Yokoyama, T.; Miyagishi, H. V.; Liu, M.; Tachibana, Y.; Fujihara, T.; Tsuji, Y.; Terao, J. Nat. Commun. 2020, 11 (1), 1–8. 124. Zhao, X.; Cardolaccia, T.; Farley, R. T.; Abboud, K. A.; Schanze, K. S. Inorg. Chem. 2005, 44 (8), 2619–2627. 125. Kaneko, S.; Masai, H.; Yokoyama, T.; Liu, M.; Tachibana, Y.; Fujihara, T.; Tsuji, Y.; Terao, J. Polymers 2020, 12 (1), 244. 126. Mihaly, J. J.; Phillips, A. T.; Stewart, D. J.; Marsh, Z. M.; McCleese, C. L.; Haley, J. E.; Zeller, M.; Grusenmeyer, T. A.; Gray, T. G. Phys. Chem. Chem. Phys. 2020, 22, 11915–11927. 127. Gao, Z.; Zhu, J.; Han, Y.; Lv, X.; Zhang, X.; Wang, F. Polym. Chem. 2016, 7 (37), 5763–5767. 128. Schull, T. L.; Kushmerick, J. G.; Patterson, C. H.; George, C.; Moore, M. H.; Pollack, S. K.; Shashidhar, R. J. Am. Chem. Soc. 2003, 125 (11), 3202–3203. 129. Miyagi, Y.; Sotani, T.; Yajima, T.; Sano, N.; Sanda, F. Polym. Chem. 2018, 9 (14), 1772–1779. 130. Zhou, G.; He, Y.; Yao, B.; Dang, J.; Wong, W. Y.; Xie, Z.; Zhao, X.; Wang, L. Chem. Asian J. 2010, 5 (11), 2405–2414. 131. Huang, Z.; Liu, B.; Zhao, J.; He, Y.; Yan, X.; Xu, X.; Zhou, G.; Yang, X.; Wu, Z. RSC Adv. 2015, 5 (46), 36507–36519. 132. Haque, A.; Al-Busaidi, I. J.; Al-Rasbi, N. K.; Al-Bahri, S.; Al-Suti, M. K.; Abou-Zied, O. K.; Khan, M. S.; Skelton’, J. M.; Raithby, P. R. Inorg. Chem. 2021, 60, 746–760. 133. Al-Busaidi, I. J.; Ilmi, R.; Dutra, J. D. L.; Oliveira, W. F.; Haque, A.; Al-Rasbi, N. K.; Raithby, P. R.; Khan, M. S. Dalton Trans. 2021, 50 (5), 1465–1477. 134. Ilmi, R.; Haque, A.; Al-Busaidi, I. J.; Al Rasbi, N. K.; Khan, M. S. Dyes Pigments 2019, 162, 59–66. 135. Li, L.; Ho, C.-L.; Wong, W.-Y. J. Organomet. Chem. 2012, 703, 43–50. 136. Sahoo, P. K.; Swain, S. K. J. Polym. Sci. A Polym. Chem. 1999, 37 (21), 3899–3902. 137. Lissel, F.; Fox, T.; Blacque, O.; Polit, W.; Winter, R. F.; Venkatesan, K.; Berke, H. J. Am. Chem. Soc. 2013, 135 (10), 4051–4060. 138. Semenov, S. N.; Taghipourian, S. F.; Blacque, O.; Fox, T.; Venkatesan, K.; Berke, H. J. Am. Chem. Soc. 2010, 132 (22), 7584–7585. 139. Dong, Q.; Meng, Z.; Ho, C.-L.; Guo, H.; Yang, W.; Manners, I.; Xu, L.; Wong, W.-Y. Chem. Soc. Rev. 2018, 47 (13), 4934–4953. 140. Ingham, S. L.; Khan, M. S.; Lewis, J.; Long, N. J.; Raithby, P. R. J. Organomet. Chem. 1994, 470 (1-2), 153–159. 141. Long, N. J.; Martin, A. J.; Vilar, R.; White, A. J.; Williams, D. J.; Younus, M. Organometallics 1999, 18 (21), 4261–4269. 142. Lavastre, O.; Even, M.; Dixneuf, P. H.; Pacreau, A.; Vairon, J.-P. Organometallics 1996, 15 (6), 1530–1531. 143. Meng, Z.; Sato, K.; Sukegawa, T.; Oyaizu, K.; Ho, C.-L.; Xiang, J.; Feng, Y.-H.; Lo, Y. H.; Nishide, H.; Wong, W.-Y. J. Organomet. Chem. 2016, 812, 51–55. 144. Liu, K.; Ho, C. L.; Aouba, S.; Zhao, Y. Q.; Lu, Z. H.; Petrov, S.; Coombs, N.; Dube, P.; Ruda, H. E.; Wong, W. Y. Angew. Chem. Int. Ed. 2008, 47 (7), 1255–1259. 145. Khairul, W. M.; Fox, M. A.; Zaitseva, N. N.; Gaudio, M.; Yufit, D. S.; Skelton, B. W.; White, A. H.; Howard, J. A.; Bruce, M. I.; Low, P. J. Dalton Trans. 2009, 4, 610–620. 146. Ohkubo, A.; Aramaki, K.; Nishihara, H. Chem. Lett. 1993, 22 (2), 271–274. 147. Irwin, M. J.; Vittal, J. J.; Puddephatt, R. J. Organometallics 1997, 16 (15), 3541–3547. 148. Vicente, J.; Chicote, M.-T.; Alvarez-Falcón, M. M.; Jones, P. G. Organometallics 2005, 24 (11), 2764–2772. 149. Liu, J.; Chen, Q.; Xiao, L.; Shang, J.; Zhou, X.; Zhang, Y.; Wang, Y.; Shao, X.; Li, J.; Chen, W.; Xu, G. Q.; Tang, H.; Zhao, D.; Wu, K. ACS Nano 2015, 9 (6), 6305–6314. 150. Ho, P.-Y.; Komber, H.; Horatz, K.; Tsuda, T.; Mannsfeld, S. C.; Dmitrieva, E.; Blacque, O.; Kraft, U.; Sirringhaus, H.; Lissel, F. Polym. Chem. 2020, 11 (2), 472–479. 151. Matsumi, N.; Chujo, Y.; Lavastre, O.; Dixneuf, P. H. Organometallics 2001, 20 (12), 2425–2427. 152. Zhu, Y.; Millet, D. B.; Wolf, M. O.; Rettig, S. J. Organometallics 1999, 18 (10), 1930–1938. 153. Zhang, S.; Sheng, Y.; Wei, G.; Quan, Y.; Cheng, Y.; Zhu, C. J. Polym. Sci. A Polym. Chem. 2014, 52 (12), 1686–1692. 154. Obara, S.; Itabashi, M.; Okuda, F.; Tamaki, S.; Tanabe, Y.; Ishii, Y.; Nozaki, K.; Haga, M.-A. Inorg. Chem. 2006, 45 (22), 8907–8921. 155. Fu, J.-Z.; Zhang, X.; Wang, J.-Y.; Zhang, L.-Y.; Chen, Z.-N. Inorg. Chem. Commun. 2012, 22, 123–125. 156. del Rio, M. P.; Lopez, J. A.; Ciriano, M. A.; Tejel, C. Chemistry 2013, 19 (15), 4707–4711. 157. Puddephatt, R. Chem. Commun. 1998, 10, 1055–1062. 158. Lima, J. C.; Rodriguez, L. Chem. Soc. Rev. 2011, 40 (11), 5442–5456. 159. Jia, G.; Puddephatt, R. J.; Scott, J. D.; Vittal, J. J. Organometallics 1993, 12 (9), 3565–3574. 160. Tian, Z.; Yang, X.; Liu, B.; Zhao, J.; Zhong, D.; Wu, Y.; Zhou, G.; Wong, W.-Y. J. Mater. Chem. C 2018, 6 (22), 6023–6032. 161. Tian, Z.; Yang, X.; Liu, B.; Zhong, D.; Zhou, G.; Wong, W.-Y. J. Mater. Chem. C 2018, 6 (42), 11416–11426. 162. Hay, A. S. J. Polym. Sci. A Polym. Chem. 1969, 7 (7), 1625–1634. 163. Wong, W.-Y.; Liu, L.; Shi, J. X. Angew. Chem. 2003, 115 (34), 4198–4202. 164. Xu, J.; Meng, B.; Liu, J.; Wang, L. Chem. Commun. 2020, 56 (42), 5701–5704. 165. Liu, S.; Zhang, K.; Lu, J.; Zhang, J.; Yip, H.-L.; Huang, F.; Cao, Y. J. Am. Chem. Soc. 2013, 135 (41), 15326–15329. 166. Lin, F.; Jiang, K.; Kaminsky, W.; Zhu, Z. L.; Jen, A. K. Y. J. Am. Chem. Soc. 2020, 142 (36), 15246–15251. 167. Benten, H.; Mori, D.; Ohkita, H.; Ito, S. J. Mater. Chem. A 2016, 4 (15), 5340–5365. 168. He, W. H.; Jiang, Y. B.; Qin, Y. Polym. Chem. 2014, 5 (4), 1298–1304. 169. Guo, F.; Kim, Y.-G.; Reynolds, J. R.; Schanze, K. S. Chem. Commun. 2006, 17, 1887–1889. 170. Wong, W.-Y.; Wang, X. Z.; Zhang, H. L.; Cheung, K.-Y.; Fung, M.-K.; Djurisic, A. B.; Chan, W.-K. J. Organomet. Chem. 2008, 693 (24), 3603–3612. 171. Baek, N. S.; Hau, S. K.; Yip, H.-L.; Acton, O.; Chen, K.-S.; Jen, A. K. Y. Chem. Mater. 2008, 20 (18), 5734–5736. 172. Wang, X.-Z.; Ho, C.-L.; Yan, L.; Chen, X.; Chen, X.; Cheung, K.-Y.; Wong, W.-Y. J. Inorg. Organomet. Polym. Mater. 2010, 20 (3), 478–487. 173. Qin, C.; Fu, Y.; Chui, C.-H.; Kan, C.-W.; Xie, Z.; Wang, L.; Wong, W.-Y. Macromol. Rapid Commun. 2011, 32 (18), 1472–1477. 174. Liu, L.; Ho, C.-L.; Wong, W.-Y.; Cheung, K.-Y.; Fung, M.-K.; Lam, W.-T.; Djurišic, A. B.; Chan, W.-K. Adv. Funct. Mater. 2008, 18 (18), 2824–2833. 175. Wong, W.-Y.; Chow, W.-C.; Cheung, K.-Y.; Fung, M.-K.; Djurisic, A. B.; Chan, W.-K. J. Organomet. Chem. 2009, 694 (17), 2717–2726. 176. Wang, X. Z.; Wang, Q.; Yan, L.; Wong, W.-Y.; Cheung, K.-Y.; Ng, A.; Djurisic, A. B.; Chan, W.-K. Macromol. Rapid Commun. 2010, 31 (9-10), 861–867. 177. Li, L.; Chow, W.-C.; Wong, W.-Y.; Chui, C.-H.; Wong, R. S.-M. J. Organomet. Chem. 2011, 696 (6), 1189–1197. 178. Wong, W.-Y.; Li, L.; Wang, L. J. Organomet. Chem. 2019, 897, 192–199. 179. H. Zhen, Q. Liu, S.-K. So and W.-Y. Wong, J. Organomet. Chem. 894, 2019, 1–9. 180. Yuan, Y.; Michinobu, T.; Oguma, J.; Kato, T.; Miyake, K. Macromol. Chem. Phys. 2013, 214 (13), 1465–1472. 181. Wang, Q.; He, Z.; Wild, A.; Wu, H.; Cao, Y.; Schubert, U. S.; Chui, C.-H.; Wong, W.-Y. Chem. Asian J. 2011, 6 (7), 1766–1777. 182. Huang, X.; Zhu, C.; Zhang, S.; Li, W.; Guo, Y.; Zhan, X.; Liu, Y.; Bo, Z. Macromolecules 2008, 41 (19), 6895–6902. 183. Zhan, H.; Lamare, S.; Ng, A.; Kenny, T.; Guernon, H.; Chan, W.-K.; Djurišic, A. B.; Harvey, P. D.; Wong, W.-Y. Macromolecules 2011, 44 (13), 5155–5167. 184. Köhler, A.; Wittmann, H. F.; Friend, R. H.; Khan, M. S.; Lewis, J. Synth. Met. 1994, 67 (1-3), 245–249. 185. Brabec, C. J.; Cravino, A.; Meissner, D.; Sariciftci, N. S.; Fromherz, T.; Rispens, M. T.; Sanchez, L.; Hummelen, J. C. Adv. Funct. Mater. 2001, 11 (5), 374–380. 186. Sirringhaus, H. Adv. Mater. 2005, 17 (20), 2411–2425.

70

Conjugated Poly(metalla-ynes)

187. McCulloch, I.; Heeney, M.; Bailey, C.; Genevicius, K.; Macdonald, I.; Shkunov, M.; Sparrowe, D.; Tierney, S.; Wagner, R.; Zhang, W.; Chabinyc, M. L.; Kline, R. J.; McGehee, M. D.; Toney, M. F. Nat. Mater. 2006, 5 (4), 328–333. 188. Liu, Q.; Ho, C.-L.; Lo, Y.-H.; Li, H.; Wong, W.-Y. J. Inorg. Organomet. Polym. Mater. 2014, 25 (1), 159–168. 189. Onitsuka, K.; Takahashi, S. J. Chem. Soc., Chem. Commun. 1995, 20, 2095–2096. 190. Chao, Y. H.; Jheng, J. F.; Wu, J. S.; Wu, K. Y.; Peng, H. H.; Tsai, M. C.; Wang, C. L.; Hsiao, Y. N.; Wang, C. L.; Lin, C. Y.; Hsu, C. S. Adv. Mater. 2014, 26 (30), 5205–5210. 191. Liu, Z.; Huang, Z.; Chen, Y.; Xu, T.; Yu, H.; Guo, X.; Yan, L.; Zhang, M.; Wong, W.-Y.; Wang, X. Macromol. Chem. Phys. 2019, 221 (1), 1900446. 192. Huang, Y.; Hsiang, E. L.; Deng, M. Y.; Wu, S. T. Light Sci. Appl. 2020, 9, 105. 193. Izumi, S.; Higginbotham, H. F.; Nyga, A.; Stachelek, P.; Tohnai, N.; Silva, P.; Data, P.; Takeda, Y.; Minakata, S. J. Am. Chem. Soc. 2020, 142 (3), 1482–1491. 194. Li, T.-Y.; Wu, J.; Wu, Z.-G.; Zheng, Y.-X.; Zuo, J.-L.; Pan, Y. Coord. Chem. Rev. 2018, 374, 55–92. 195. Hu, Y.-X.; Xia, X.; He, W.-Z.; Tang, Z.-J.; Lv, Y.-L.; Li, X.; Zhang, D.-Y. Org. Electron. 2019, 66, 126–135. 196. Ho, C.-L.; Chui, C.-H.; Wong, W.-Y.; Aly, S. M.; Fortin, D.; Harvey, P. D.; Yao, B.; Xie, Z.; Wang, L. Macromol. Chem. Phys. 2009, 210 (21), 1786–1798. 197. Goudreault, T.; He, Z.; Guo, Y.; Ho, C.-L.; Zhan, H.; Wang, Q.; Ho, K. Y.-F.; Wong, K.-L.; Fortin, D.; Yao, B.; Xie, Z.; Wang, L.; Kwok, W.-M.; Harvey, P. D.; Wong, W.-Y. Macromolecules 2010, 43 (19), 7936–7949. 198. Sheng, C. X.; Singh, S.; Gambetta, A.; Drori, T.; Tong, M.; Tretiak, S.; Vardeny, Z. V. Sci. Rep. 2013, 3, 2653. 199. Wilson, J. S.; Dhoot, A. S.; Seeley, A. J.; Khan, M. S.; Köhler, A.; Friend, R. H. Nature 2001, 413 (6858), 828–831. 200. Yan, L.; Zhao, Y.; Wang, X.; Wang, X. Z.; Wong, W. Y.; Liu, Y.; Wu, W.; Xiao, Q.; Wang, G.; Zhou, X.; Zeng, W.; Li, C.; Wang, X.; Wu, H. Macromol. Rapid Commun. 2012, 33 (6-7), 603–609. 201. Guo, Y.; Yu, G.; Liu, Y. Adv. Mater. 2010, 22 (40), 4427–4447. 202. Di, C. A.; Yu, G.; Liu, Y. Zhu, D. J. Phys. Chem. B 2007, 111 (51), 14083–14096. 203. Hayakawa, S.; Matsuo, K.; Yamada, H.; Fukui, N.; Shinokubo, H. J. Am. Chem. Soc. 2020, 142 (27), 11663–11668. 204. Yan, L.; Li, C.; Cai, L.; Shi, K.; Tang, W.; Qu, W.; Ho, C.-L.; Yu, G.; Li, J.; Wang, X. J. Organomet. Chem. 2017, 846, 269–276. 205. Spangler, W.; C., J. Mater. Chem. 1999, 9 (9), 2013–2020. 206. McKay, T. J.; Staromlynska, J.; Wilson, P.; Davy, J. J. Appl. Phys. 1999, 85 (3), 1337–1341. 207. McEwan, K.; Lewis, K.; Yang, G.-Y.; Chng, L.-L.; Lee, Y.-W.; Lau, W.-P.; Lai, K.-S. Adv. Funct. Mater. 2003, 13 (11), 863–867. 208. Perry, J. W.; Mansour, K.; Marder, S. R.; Perry, K. J.; Alvarez, D.; Choong, I. Opt. Lett. 1994, 19 (9), 625–627. 209. Tang, B. Z.; Xu, H. Macromolecules 1999, 32 (8), 2569–2576. 210. Tutt, L. W.; Kost, A. Nature 1992, 356 (6366), 225–226. 211. Zhou, G.; Wong, W.-Y.; Poon, S.-Y.; Ye, C.; Lin, Z. Adv. Funct. Mater. 2009, 19 (4), 531–544. 212. Zhou, G. J.; Wong, W.-Y.; Lin, Z.; Ye, C. Angew. Chem. Int. Ed. 2006, 45 (37), 6189–6193. 213. Liu, B.; Tian, Z.; Dang, F.; Zhao, J.; Yan, X.; Xu, X.; Yang, X.; Zhou, G.; Wu, Y. J. Organomet. Chem. 2016, 804, 80–86. 214. Lieberzeit, P. A.; Dickert, F. L. Anal. Bioanal. Chem. 2007, 387 (1), 237–247. 215. Szabo, N. F.; Du, H.; Akbar, S. A.; Soliman, A.; Dutta, P. K. Sens. Actuators, B 2002, 82 (2-3), 142–149. 216. Saleem, M.; Rafiq, M.; Hanif, M. J. Fluoresc. 2017, 27 (1), 31–58. 217. Madhu, S.; Sharma, D. K.; Basu, S. K.; Jadhav, S.; Chowdhury, A.; Ravikanth, M. Inorg. Chem. 2013, 52 (19), 11136–11145. 218. Zhao, C.; Li, K.; Xie, N.; Zhao, M.; Peng, S. J. Photochem. Photobiol. A 2014, 290, 72–76. 219. Zhang, T.; She, G.; Qi, X.; Mu, L. Tetrahedron 2013, 69 (34), 7102–7106. 220. Zhao, Q.; Cao, T.; Li, F.; Li, X.; Jing, H.; Yi, T.; Huang, C. Organometallics 2007, 26 (8), 2077–2081. 221. Zhao, Q.; Li, F.; Liu, S.; Yu, M.; Liu, Z.; Yi, T.; Huang, C. Inorg. Chem. 2008, 47 (20), 9256–9264. 222. Caliendo, C.; Fratoddi, I.; Russo, M. V. Appl. Phys. Lett. 2002, 80 (25), 4849–4851. 223. Sun, P.; Jiang, Y.; Xie, G.; Yu, J.; Du, X.; Hu, J. J. Appl. Polym. Sci. 2010, 116 (1), 562–567. 224. Zhang, Y.; Murphy, C. B.; Jones, W. E. Macromolecules 2002, 35 (3), 630–636. 225. Kimura, M.; Horai, T.; Hanabusa, K.; Shirai, H. Adv. Mater. 1998, 10 (6), 459–462. 226. Ogawa, K.; Guo, F.; Schanze, K. S. J. Photochem. Photobiol. A 2009, 207 (1), 79–85. 227. Qin, C.; Wong, W.-Y.; Wang, L. Macromolecules 2011, 44 (3), 483–489. 228. Yuan, M.; Wang, F.; Tian, Y.-K. RSC Adv. 2018, 8 (71), 40794–40797. 229. Cheng, X.; Md, A.; Lian, H.; Zhong, Z.; Guo, H.; Dong, Q.; Roy, V. A. L. J. Organomet. Chem. 2019, 892, 34–40. 230. Xiang, J.; Wang, T. K.; Zhao, Q.; Huang, W.; Ho, C.-L.; Wong, W.-Y. J. Mater. Chem. C. 2016, 4 (5), 921–928. 231. Tsai, M.-C.; Wang, C.-L.; Lin, C.-Y.; Tsai, C.-L.; Yen, H.-J.; You, H.-C.; Liou, G.-S. Polym. Chem. 2016, 7 (16), 2780–2784. 232. Dalko, P. I.; Moisan, L. Angew. Chem. Int. Ed. Engl. 2004, 43 (39), 5138–5175. 233. Xiang, Q.; Yu, J. Jaroniec, M. J. Phys. Chem. C 2011, 115 (15), 7355–7363. 234. Hodges, B. C.; Cates, E. L.; Kim, J. H. Nat. Nanotechnol. 2018, 13 (8), 642–650. 235. Jiang, J. X.; Su, F.; Trewin, A.; Wood, C. D.; Campbell, N. L.; Niu, H.; Dickinson, C.; Ganin, A. Y.; Rosseinsky, M. J.; Khimyak, Y. Z.; Cooper, A. I. Angew. Chem. Int. Ed. Engl. 2007, 46 (45), 8574–8578. 236. Ma, L.; Liu, Y.; Liu, Y.; Jiang, S.; Li, P.; Hao, Y.; Shao, P.; Yin, A.; Feng, X.; Wang, B. Angew. Chem. Int. Ed. Engl. 2019, 58 (13), 4221–4226. 237. Wu, K.; Guo, J.; Wang, C. Angew. Chem. Int. Ed. Engl. 2016, 55 (20), 6013–6017. 238. Chen, Z.; Wang, J.; Zhang, S.; Zhang, Y.; Zhang, J.; Li, R.; Peng, T. ACS Appl. Energy Mater. 2019, 2 (8), 5665–5676. 239. Cheng, H. B.; Li, Y.; Tang, B. Z.; Yoon, J. Chem. Soc. Rev. 2020, 49 (1), 21–31. 240. Thorp-Greenwood, F. L. Organometallics 2012, 31 (16), 5686–5692. 241. Zhang, K. Y.; Yu, Q.; Wei, H.; Liu, S.; Zhao, Q.; Huang, W. Chem. Rev. 2018, 118 (4), 1770–1839. 242. Bian, W.; Lian, H.; Zhang, Y.; Tai, F.; Wang, H.; Dong, Q.; Yu, B.; Wei, X.; Zhao, Q. J. Organomet. Chem. 2017, 835, 25–30.

14.04

Organoboron and Related Group 13 Polymers

Holger Helten, Institute of Inorganic Chemistry & Institute for Sustainable Chemistry and Catalysis With Boron (ICB), Julius-Maximilians-Universität Würzburg, Würzburg, Germany © 2022 Elsevier Ltd. All rights reserved.

14.04.1 Introduction 14.04.2 Boron in the main chain: Inorganic–organic hybrid polymers 14.04.2.1 Tricoordinate boron in the backbone of p-conjugated polymers 14.04.2.1.1 Polymers with boron incorporated in the main chain exclusively via B–C linkages 14.04.2.1.2 Polymers with B–N linkages in the main chain 14.04.2.1.3 Polymers with B–O linkages in the main chain 14.04.2.2 Tetracoordinate boron in the polymer backbone 14.04.2.2.1 Coordination polymers with linear dative B N bonds in the main chain 14.04.2.2.2 Polymers with boron chelate complexes in the main chain 14.04.3 Boron in polymer side chains: Organic–inorganic hybrid polymers 14.04.3.1 Triorganoborane groups in side chains of p-conjugated polymers 14.04.3.2 Tetracoordinate boron chelate complexes in polymer side chains 14.04.4 Inorganic polymers comprising (BN)n or (BP)n chains 14.04.5 Heavier group 13 element-containing hybrid polymers 14.04.5.1 Tetra- and pentacoordinate gallium in the main chain of conjugated polymers 14.04.5.2 Organoaluminum-, organogallium-, and organoindium-bridged polymetallocenes 14.04.6 Conclusions Acknowledgment References

71 72 72 72 85 93 96 96 100 114 114 117 119 124 124 126 127 127 127

14.04.1 Introduction Organometallic main group polymers1–3 containing boron have been the subject of tremendous research activity in recent years.4–7 Historically, boron has been known as a component of glass and other thermodynamically very stable inorganic solid state materials. In such compounds the boron atoms are embedded in extended network structures that feature strong BdE bonds with significant covalent character, thus already indicating the potential of incorporating this element into macromolecular frameworks. Indeed, inorganic polymers comprising boron atoms, such as the polysiloxane-based substance termed “silly putty,” have been known for quite some time. Boron forms considerably strong bonds with the other 2nd row elements of the p-block, including carbon. Boron and carbon atoms are of comparable size, and, due to their relatively small difference in electronegativity, BdC bonds are of comparatively low polarity. The inherent electron deficiency of the group 13 elements often poses significant synthetic challenges, but it forms also the basis for the unique properties and functions of the resulting materials that contain these elements. Recent advances in synthetic chemistry have enabled access to various organoboron compounds—including tricoordinate organoborane species8–10—that are perfectly stable towards air and moisture, which is often a key requirement for the applications they were intended for. The rapid development of organoboron macromolecular materials in recent years has been driven by their applications, which includes their use as flame retardants, precursors to high-performance ceramics,11–13 photo- and electro-active materials for electronic and optoelectronic devices, sensory materials,6–8,14–21 electrolytes for lithium ion batteries,18 polymersupported reagents and catalysts,22,23 stimuli-responsive materials,24 and for biomedical applications.25–27 Promising results have recently been obtained also with macromolecules of the heavier group 13 elements, however, this field of research is to date still in its infancy. In this chapter, the developments in the macromolecular chemistry of the group 13 elements over the last about 15 years are discussed. Since this topic was not included in the previous edition of Comprehensive Organometallic Chemistry, some earlier works that have significantly contributed to the development of the field are additionally included. Purely theoretical (computational) studies are not discussed here. The focus of the majority of recent research activities, and therefore, the major focus of this chapter, is on conjugated organoboron polymers and on polymers that exploit the optical/electronic properties of boron-containing chromophores. We herein discuss substances that are commonly regarded as polymers. These soft materials basically have linear/ cyclolinear chain or ladder structures, although they may be branched or even crosslinked resulting in more complex architectures. Well-defined molecular entities, on the other hand, even if they may have high molecular weights, such as macrocycles,28–35 are the topic of this chapter. Similarly, boron-containing dendrimers6,36–40 are excluded; such species are the subject of another chapter of this edition of Comprehensive Organometallic Chemistry. Also, covalent organic frameworks (COFs) and cage compounds41–49 are beyond the scope of this chapter. The same is true for borane and carborane cluster compounds in materials and polymers.4,50 The main part of this chapter is on polymers featuring boron atoms in the main chain (Section 14.04.2). Such species are, according to the IUPAC recommendation, denoted as inorganic–organic hybrid polymers. The section is subdivided into subsections discussing

Comprehensive Organometallic Chemistry IV

https://doi.org/10.1016/B978-0-12-820206-7.00155-4

71

72

Organoboron and Related Group 13 Polymers

polymers with the boron atoms in the tricoordinate and in the tetracoordinate state. This discussion is, however, restricted to compounds containing electrically neutral boron moieties; borate and boronium-functionalized polymers (polyelectrolytes) are not included. If the polymer backbone constitutes a purely organic polymer chain, and the boron atoms reside in a side chain, these compounds are denoted as organic–inorganic hybrid polymers (Section 14.04.3). Here, we focus on some selected examples, which are either p-conjugated polymers featuring triorganoborane groups in its side chains or polymers functionalized with tetracoordinate boron chelate complexes in the side chains. Nonconjugated polymers that feature borane groups in side chains5,22,23 are not discussed here. The same is true for the classes of polymers that are functionalized with boronic or borinic acid or ester groups, which are particular interesting due to of their capability of pH-dependent, reversible binding of saccharides.25,26,51,52 In Section 14.04.4, polymers comprising linear (BN)n or (BP)n chains are discussed. Such species are denoted as inorganic polymers as they have no carbon in the backbone; though they may have organic side groups. Polyborazylenes are inorganic polymers with a network structure of partially fused borazine units, which are versatile precursors to ceramic boron nitrides.11–13 These materials, however are beyond the scope of this chapter. Finally, the known polymers containing the heavier elements of group 13 are presented in Section 14.04.5.

14.04.2 Boron in the main chain: Inorganic–organic hybrid polymers The largest part of recent research activities towards main chain-functionalized organoboron polymers has focused on p-conjugated polymers and/or polymers with desirable luminescent or other useful (opto)electronic properties. Various examples by now have demonstrated that tricoordinate boron centers can be embedded in the main chain of organic polymers. If the polymer backbone is composed of a conjugated p-electron system, this may be extended via the vacant p-orbital of the boron atom. Thereby, novel materials—often with intriguing properties and novel functions—are accessed. As the incorporation of boron adds an empty orbital to the p system, this may lead to n-type semiconducting materials, which can be reductively doped. If strongly electron-accepting borane groups are brought in communication via a p system with electron-releasing groups, donor–acceptor-type materials are produced, which potentially exhibit pronounced nonlinear optical (NLO) properties.53 Furthermore, the Lewis acidity that is principally inherent to trivalent boron can be exploited for sensory applications.54–56 Depending on different influencing factors, such as the electron-withdrawing character and the sterics of the boron’s substituents, certain anions or neutral donor compounds may bind to the boron centers. This causes a change of the optical and/or conductive properties of the polymers, which is exploited for signal generation. Compared to molecular sensory materials, polymers usually offer the advantage of showing a signal amplifying effect.57 The instability of many tricoordinate organoboron compounds towards moisture or oxygen, which is associated with the Lewis acidity of boron, can often be overcome either by sterically shielding the boron center or by incorporating it in a cyclic system. In the following section, conjugated polymers having tricoordinate boron centers in the main chain are discussed. This section is further subdivided according to how the boron atoms are linked in the backbone. In Section 14.04.2.1.1, polymers are discussed in which the boron is exclusively bound to carbon atoms. Polymers that feature one BdN or one BdO bond in the main chain are discussed in Sections 14.04.2.1.2 and 14.04.2.1.3, respectively. Various dyes are known that contain boron in the tetracoordinate state. Incorporation of such dyes into the backbone of conjugated polymers has led to numerous materials with intriguing properties. In some cases, evidence for electronic communication over tetracoordinate boron centers has be received. Such species, which are discussed in Section 14.04.2.2.2, have been used, for example, for lighting or bioimaging purposes.

14.04.2.1 Tricoordinate boron in the backbone of p-conjugated polymers 14.04.2.1.1

Polymers with boron incorporated in the main chain exclusively via B–C linkages

Chujo and co-workers were the first to demonstrate the use of the hydroboration reaction for polymer synthesis. In the early 1990s, they initially applied this “hydroboration polymerization” approach to access a series of non-conjugated organoborane polymers by the combination of dihydroboranes with various difunctional co-monomers.58–60 Apart from that, they also used related haloboration and phenylboration processes for polyaddition reactions.61 In 1998, Siebert, Corriu, and Douglas reported the synthesis of organoborane polymers featuring a fully unsaturated backbone, 2a-c, by hydroboration of 2,5-dialkynylthiophenes 1 with the combination of BCl3 and Et3SiH (Scheme 1).62 Due to the high

Scheme 1 Synthesis of unsaturated thiophene-containing organoborane polymers 2a-d by hydroboration polymerization.

Organoboron and Related Group 13 Polymers

73

sensitivity of the polymers 2a-c to oxygen and water, with their unprotected boron center featuring a reactive BdCl bond, attempts to characterize them by size exclusion chromatography (SEC) or UV-vis spectroscopy were unsuccessful. However, their color was found to be strongly dependent on the substituent R groups. The 11B NMR resonance of the polymers occurred at dB ¼ 6–7, that is, about 50 ppm upfield from that of related molecular model systems, which was interpreted as being a result of electronic interactions between the boron and thiophene groups along the polymer chain. The reaction of 2b with 3-dimethylaminopropan-1-ol yielded 2d, which was characterized by SEC as an oligomer of about 8 repeat units, suggesting that polymer degradation has occurred to some extent. In the same year, Chujo and co-workers presented the successful synthesis of a series of stable p-conjugated organoborane polymers, 3a,g-i, by hydroboration polymerization of mesitylborane with a number of p system-bridged diynes (Scheme 2).63 Subsequently, they added further derivatives,64 including those with heterocyclic building blocks, 3j-l,65 as well as crosslinked polyboranes.66 The bulky mesityl (2,4,6-trimethylphenyl; Mes) group on boron provides sufficient steric shielding, thus rendering the polymers remarkably stable to air and moisture. More recently, Chujo and co-workers presented analogous polymers 4, which feature the even bulkier tripyl (2,4,6-triisopropylphenyl; Tip) group that very effectively stabilizes the boron center.67,68 The synthesis thereof was achieved by similar hydroboration polymerizations of the respective diynes with tripylborane. All polymers obtained proved to be readily soluble in common organic solvents, hence, an estimation of their molecular weights by gel permeation chromatography (GPC) versus polystyrene (PS) standards was feasible. The values varied significantly with the diethynylarene building block employed. For polymer 3a, for example, after optimization of the polymerization conditions and the purification procedure, a sample with a number average molecular weight of up to Mn ¼ 10.5 kDa was obtained.64 Chujo and co-workers additionally prepared a series of related poly(vinylenearylenevinylene-boranes) via haloboration-phenylboration polymerization of aromatic diynes with Ph2BBr,69 though the boron center is less well kinetically stabilized in this case.

Scheme 2 Synthesis of polymers 3 and 4 by hydroboration polymerization.

Evidence of significant p-electron delocalization along the main chain of these polymers via the boron’s vacant p orbital was gained from optical spectroscopy. The lowest-energy absorption band in their UV-vis spectra was assigned to a p–p excitation process in the polymer backbone. Its position in the spectrum strongly depends on the arylene group incorporated (lmax ¼ 317–450 nm, in CHCl3). Furthermore, it is significantly bathochromic shifted compared to that of molecular model compounds for the polymers. The latter are also strongly fluorescent in the blue to green region, some of which showed quantum

74

Organoboron and Related Group 13 Polymers

yields, FF, between 30% and 50%. The 11B NMR resonances of all these polymers appeared at about 31 ppm, that is, significantly upfield from that of the molecular compound (PhCH]CH)2BTip (dB ¼ 64.5).67 The molecular structure of this model compound in the solid state was determined by single-crystal X-ray diffractometry. This revealed a largely coplanar arrangement of the vinylene and the phenyl groups—a prerequisite for extended p conjugation over the boron center. The Tip substituent is oriented almost perpendicularly to the molecular p plane. It causes effective shielding of the boron center through its ortho-iPr groups, which is assumed to be the reason for the air stability of this compound as well as the structurally related polymers. The structure of the model compound further confirmed the regio- and syn-selectivity of the hydroboration reaction in this case, which results in an all-trans configuration of the vinylene groups in the microstructure of the polymers. Poly(vinylenearylenevinylene-boranes) 3 studied by Chujo and co-workers by degenerate four wave mixing showed exceptionally large third-order nonlinear susceptibilities of up to w(3) ¼ 6.87  10−6 esu.16 Conductivity measurements on tripyl-substituted poly(vinylenearylenevinylene-borane) 4e revealed a dramatic increase of the conductivity from 126b > 126a shows that the HOMO energy level decreases with increasing density of BN units in the polythiophene backbone. Therefore, the OFET devices fabricated from 126a,b with a top-gate/bottom-contact (TG/BC) configuration proved to be air-stable. Polymer 126a showed a hole mobility of 0.01 cm2 V−1 s−1, while that of 126b was much higher—up to 0.38 cm2 V−1 s−1. This effect was assumed to be a result of the strong intermolecular interactions of the latter. By tapping-mode atomic force microscopy (AFM)aus fiber-like intercalating networks with crystalline zones were detected. Both polymers adopt an edge-on orientation in the solid state as revealed by grazing incidence X-ray diffraction (GI-XRD), which is favorable for in-plane charge transport. He and co-workers recently reported the incorporation of 9,10-azaboraphenanthrene moieties into conjugated copolymers 128 and 129 by Suzuki–Miyaura- or Stille-type C–C coupling polycondensation, respectively (Scheme 31).159 They prepared the 9,10-azaboraphenanthrene monomer 127 via an aromaticity-driven ring expansion reaction of the respective borafluorene derivative with mesityl azide. Such type of reaction was first demonstrated for the borole system by Braunschweig and co-workers,160,161 and was independently studied at about the same time by Martin and co-workers.162 The polymers 128 (Mn ¼ 6.17 kDa, by GPC) and 129 (Mn ¼ 6.08 kDa) showed good air- and moisture-stability.159 Polymer 129 is regiorandom in terms the position of the hexyl group at the thiophene unit, and both 128 and 129 have a regiorandom distribution of B]N units along the main chain. Compared to structurally related small molecules, both polymers show significant red-shifts of their absorption and fluorescence spectra. They fluoresce brightly blue [lem,max ¼ 414 (128) and 458 nm (129)] with quantum yields of FF ¼ 88% (128) and 57% (129).

Scheme 31 Synthesis of 9,10-azaboraphenanthrene-containing copolymers 128 and 129 by Suzuki–Miyaura or Stille-type cross-coupling polycondensation, respectively.

Recently, Feng, Huang, Liu, and co-workers reported the on-surface bottom-up synthesis of NBN-doped zigzag-edged graphene nanoribbons (GNRs) 132a,b (Scheme 32).163 They used U-shaped monomers 130a,b having NBN units preinstalled at the zigzag edge and subjected them to surface-assisted polymerization at elevated temperatures on Au (111) to give 131a,b in the first step. Subsequent intramolecular cyclodehydrogenation at higher temperature (450  C) yielded the NBN-doped GNRs 132a,b with zigzag-rich edges. Their structures were elucidated by scanning tunneling microscopy (STM) and noncontact atomic force microscopy (nc-AFM). Scanning tunneling spectroscopy (STS) and DFT calculations revealed that the NBN-doping effectively modulates the electronic structure of such GNRs. The band gaps of 132a and 132b were estimated as 1.50 and 0.90 eV, respectively, which is substantially higher than those of the parent pristine carbon-based GNRs (0.52 and 0.27 eV). Furthermore, DFT calculations predict that the band structures of such NBN-GNRs can be further modified to be gapless or metallic by oxidation of the NBN units.

Scheme 32 Fabrication of NBN-doped zigzag-edged graphene nanoribbons 132a,b by on-surface polymerization of 116a,b and subsequent thermal cyclodehydrogenation of the respective intermediate 131a,b.

Sánchez-Sánchez et al. described the bottom-up fabrication of BN-substituted heteroaromatic 2D networks 135 that consist of hexa-peri-hexabenzocoronene units with a central borazine core (Scheme 33).164 Deposition of the specifically designed hexaarylborazine monomers 133 and 134 under ultrahigh vacuum (UHV) conditions on an Ag (111) surface led to the formation of a

Scheme 33 Fabrication of BN-doped heteroaromatic 2D networks 135 by on-surface polymerization of 133 or 134 and subsequent thermal cyclodehydrogenation.

92

Organoboron and Related Group 13 Polymers

self-assembled flower-like superstructure as shown by high-resolution STM images. At elevated temperatures, surface-assisted polymerization of both 133 and 134 and subsequent cyclodehydrogenation occurred to give the network structure 135. In agreement with the experimental observations, DFT calculations revealed that this proceeds via a significantly lower energy barrier when the conformationally more rigid precursor 133 is used. Borazine moieties have also been incorporated in 3D inorganic–organic hybrid polymer network structures. El-Kaderi and co-workers prepared a series of highly porous borazine-linked polymers (BLPs), 136–140, by condensation of di-, tri- or tetraamines with either BH3 or BCl3 (Scheme 34).165–167 These materials exhibit high surface areas, which enabled their use for gas (H2, CO2, CH4) storage and separation applications. Notably, the halogen-decorated derivative 140 exhibited a very high CO2/CH4 selectivity.167

Scheme 34 Synthesis of borazine-linked polymer networks 136 and 137 by condensation of para-phenylenediamine with BH3 and BCl3, respectively, and structures of further network polymers, 138–140, synthesized via analogous routes.

Collaborative work between the groups of Pich and Helten led to the synthesis of a series of borazine-based hybrid cyclomatrix polymers 144a-c (Scheme 35).168 When the reaction of trichloroborazine 141 with diamine 142b was performed under conventional dispersion polycondensation conditions (Route A) in the presence of polyvinyl pyrrolidone (PVP) as a steric stabilizer, field-emission scanning electron microscopy (FESEM) revealed the formation of spherical polymer particles of 144b with a mean diameter of d ¼ 184  21 nm. The reaction of 141 with diamine 142c gave microspheres of comparable size (d ¼ 183  20 nm) even in the absence of any stabilizer or surfactant. The reaction of 141 with N,N0 -disilylated diamine 143c under precipitation polymerization conditions (Route B), on the other hand, led to clean Si/B exchange polycondensation with formation of microspheres of 144c of much larger size (d ¼ 905  150 nm for the major fraction). The obtained materials proved to be remarkably thermally stable. Major mass loss did not occur until ca. 450  C. Even at 880  C under N2 atmosphere the residual mass still amounted to 56%.

Organoboron and Related Group 13 Polymers

93

Scheme 35 Synthesis of borazine-based cyclomatrix hybrid polymers 144a-c via salt elimination (Route A), and synthesis of 144c via silicon/boron exchange polycondensation (Route B).

Bonifazi and co-workers explored the incorporation of borazine rings into well-defined dendritic polyphenylenes.39 They elucidated the effect of borazino-doping dosage and orientation on the photophysical properties of the resulting materials. For the synthesis thereof they employed the decarbonylative Diels −Alder cycloaddition reaction between ethynyl groups and tetraphenylcyclopentadienone derivatives. Borazine rings are also the constituents of polyborazylenes (PBZ), which are highly valuable polymeric precursors to ceramic hexagonal boron nitride (h-BN). However, as these polymers contain no (or only a very low content of ) carbon, they are rather purely inorganic boron than organoboron polymers, and hence, they are not discussed here. The interested reader is therefore referred to the excellent reviews on this particular topic.11–13

14.04.2.1.3

Polymers with B–O linkages in the main chain

Boron–oxygen bonds are commonly employed as dynamic covalent linkages in various covalent organic frameworks (COFs) or cage compounds.41–49,52 The discussion of such species, however, is beyond the scope of this chapter. Furthermore, boronic acid or ester groups installed in the side groups of polymers can serve as (dynamic) cross-linking sites through (reversible) interstrand boronic ester formation; or they are introduced in order to make use of their saccharide binding properties.18,25–27 This section, however, is limited to virtually linear polymers (including cyclolinear and ladder polymers etc.) that feature B–O linkages with tricoordinate boron in the main chain. The B–O linked sulfoximine-containing alternating copolymer 146 was recently presented by Bolm and Helten and co-workers (Scheme 36).136 The synthesis thereof was achieved by condensation of bisphenol 145 with bisborane 106 via salt elimination in the presence of triethylamine as an auxiliary base. The product proved to be sufficiently stable to allow for its characterization by GPC under ambient conditions. When the synthesis was carried out in CH2Cl2 at room temperature, only short oligomers of about three repeat units on average (Mn ¼ 2.5 kDa, by GPC) were obtained. Performing the reaction in 1,2-difluorobenzene (o-DFB) at 80  C yielded a sample of 146 of significantly increased molecular weight (Mn ¼ 5.3 kDa).

Scheme 36 Synthesis of sulfoximine-containing alternating copolymer 146 featuring linear B–O linkages.

While previously described polymers with unprotected linear B–O linkages along the backbone were of rather poor hydrolytic stability,169,170 poly(dioxaborolanes), in which the BdO bonds are incorporated in a cyclic system, proved to be significantly more stable. Such polymers were first described by Lavigne and co-workers in 2005 (Scheme 37).171 Poly(dioxaborolanes) 149 and 151a, b were formed by condensation of 9,9-dihexylfluorene-2,7-diboronic acid (147a) with pentaerythritol (148) and of 147a,b with

94

Organoboron and Related Group 13 Polymers

Scheme 37 Synthesis of poly(dioxaborolanes) 149 and 151a,b by polycondensation of bisboronic acid 147a,b with tetraols 148 or 150 under Dean–Stark conditions.

1,2,4,5-tetrahydroxybenzene (150), respectively, in toluene with azeotropic removal of water (Dean–Stark conditions).171–173 For the former, self-healing properties were demonstrated.171 The molecular weight of 149 could be controlled during the polymerization reaction or by post-polymerization processing. When a slight excess of pentaerythritol was used, GPC analysis suggested a molecular weight average of Mw ¼ 28 kDa (PDI ¼ 2.60), consistent with the result of a 1H NMR end group analysis. Higher polymers were formed with equimolar amounts of the two monomers or when longer reaction times were applied. Interestingly, even with a relatively large excess of 148, polymer of reasonably high molecular weight (Mw ¼ 10 kDa) was obtained. This was attributed to the limited solubility of pentaerythritol in toluene, gradually supplying the bis-diol into the solution during formation of the well soluble polymer, thus ensuring a low, constant concentration of this monomer throughout the process. Upon keeping the as synthesized polymer under reduced pressure, its molecular weight further increased due to condensation occurring in the solid state. In this process, major amounts of insoluble material were formed, presumably higher-molecular-weight polymer. 1H NMR end group analysis of the soluble fraction indicated an Mn of about 76.9 kDa. On the other hand, the polymer chains could effectively be shortened through storing it in wet solvents. The hydrolysis process could be reversed, as the degraded polymers were repaired by storage under reduced pressure. Studies of polymers 151a,b provided convincing evidence of extended p-electron delocalization along its backbone.172,173 Single-crystal X-ray diffraction of a molecular model for these polymers revealed a largely coplanar arrangement of the aryl rings. Unlike nonemissive 149, the conjugated polymer 151a showed blue-light emission. Its absorption maximum in the UV-vis spectrum appeared 15 nm red-shifted from that of its small molecular model compound, and that of the latter appeared 12 nm red-shifted from that of 149. Semi-empirical calculations (ZINDO/1) suggested an effective conjugation length (nECL) for this class of p conjugated poly(dioxaborolanes) of over 3–5 bridging dioxaborolane units. A more recent study by Lavigne and co-workers on the sensing optical abilities of 151b and related molecular bis(dioxaborolanes) revealed that these materials show a selective optical response to hard Lewis bases (F−, CN−, and H2PO−4), but virtually no response to softer anions, including the higher halides.173 Binding of the analytes leads to blue-shifted absorptions. For fluoride binding, it was demonstrated that it can be reversed through the addition of Me3SiCl. Polymer 151b exhibited a 13-fold enhancement in relative response compared with a simple bis(dioxaborolane), thus exemplifying the signal amplification effect of conjugated polymers. Kubo and co-workers demonstrated that pyridine-assisted sequential boronate esterification of benzene-1,4-diboronic acid and 1,2,4,5-tetrahydroxybenzene gives rise to the formation of well-defined submicrospheres by hierarchical layer-by-layer selfassembly.174 The soft particles proved to be responsive to chemical stimuli including a switch in pH value and the presence of certain polyols such as pentaerythritol. Exchange reactions with saccharides induced a morphology change, which is visually detectable through a color change in the solution. Trogler and co-worker exploited a double transesterification polymerization approach for the synthesis of poly-30 ,60 -bis(1,3,2dioxaborinane)fluoran 153 (Mw ¼ 10 kDa, PDI ¼ 1.5, by GPC; Scheme 38).175 As suggested by a model reaction, the driving force for polymer formation in this case is likely to be the formation of energetically favored six-membered di-ester rings in 153 from a monomer, 152, that features five-membered di-ester rings. This thermodynamic stabilization also improves the overall stability of the polymer and makes it resistant to oxidation under ambient conditions and under incidence of UV light. However, degradation of the polymer occurs upon treating it with hydrogen peroxide. Exploiting this reactivity enabled the use of 153 as an efficient vapor-phase sensor for H2O2. Oxidative backbone cleavage of the non-luminescent polymer 153 concomitantly leads to the formation of the green fluorphore fluorescein (154) from the former fluoran moiety, thus generating a fluorescence turn-on response. Detection limits as low as 3 ppb were observed for H2O2 over a period of 8 h.

Organoboron and Related Group 13 Polymers

95

Scheme 38 Synthesis of main-chain boronate-based polymer 153 by a double transesterification polymerization approach and subsequent degradation thereof through the action of H2O2.

A 2 + 2 cyclocondensation approach was used by Lee and co-workers to first assemble the borosiloxane cage compounds 157a,b from dihydroxysilane 155 and arylboronic acids 156a,b (Scheme 39).176 Compounds 157a,b contain two electropolymerizable bithiophene units. Anodic polymerization thereof furnished thin films of 158a,b on polymer-coated indium tin oxide (ITO) electrodes, which were initially green due to oxidative doping. Electrochemical dedoping led to a color change to orange. UV-vis spectroelectrochemical investigations provided evidence for the evolution of polaronic/bipolaronic states with increasing anodic doping level. Polymer 158b, furthermore, underwent a rapid and reversible color change from green to orange upon exposure to volatile amines.

Scheme 39 Synthesis of borasiloxane-containing polymers 158a,b by electrochemical polymerization.

Aldridge and Fallis and co-workers investigated reactions of 1,10 -ferrocenediboronic acid (159) with tetraols (Scheme 40).177 They found that the reaction with pentaerythritol (148) selectively gives the macrocycle 160, while the reaction with (1R,2S,5R,6S)tetrahydroxycyclooctane (161) leads to the selective formation of a polymeric material (162). The product was insoluble in

Scheme 40 Synthesis of ferrocene-containing B–O linked macrocycle 160 and polymer 162.

96

Organoboron and Related Group 13 Polymers

common solvents, which prevented the determination of its molecular weight distribution by solution-based methods. However, MALDI mass spectrometry evidenced the presence of oligomers of up to 17 repeat units. Narita, Müllen, Fasel, and co-workers reported the surface-assisted bottom-up synthesis of chiral OBO-doped (4,1)-graphene nanoribbon (GNR) 165 (Scheme 41).178 The rationally designed monomer 163 was first sublimed on an Au (111) surface, where it self-assembled under vacuum (UHV) at room temperature into linear chains. Annealing at 200  C caused aryl–aryl coupling to give the linear polymer 164. Further annealing at 400  C and then at 450  C led to cyclodehydrogenation with formation of the GNR 165. Its structure has been characterized by STM, nc-AFM, and Raman spectroscopy. STS studies combined with computations revealed that the OBO-doped (4,1)-GNRs exhibit a larger bandgap than the pristine (4,1)-GNRs, but a lowered bandgap than GNRs without the OBO segments.

Scheme 41 Synthesis of a chiral OBO-doped (4,1)-graphene nanoribbon (GNR) 165.

14.04.2.2 Tetracoordinate boron in the polymer backbone Polymers that contain boron in the tetracoordinate state have undergone rapid development in recent years. Embedding the boron atom in a cyclic system by chelate formation often imparts great robustness to the resulting materials. Various efficient dyes with such structural motifs are known, and many of them have been incorporated into polymeric frameworks. Although the coordinatively saturated boron atom itself cannot participate in p conjugation, some of these dyes allow the formation of fully conjugated polymers through the dye’s backbone (e.g., boron dipyrromethene, BODIPY). In other cases, where there are no alternative pathways for conjugation (e.g., in cyclodiborazanes, pyrazaboles, etc.), nevertheless, interesting photophysical properties have been realized. This may, for example, result from the formation of donor–acceptor a structure. Several newly emerged classes of boron N,N- and N,C-chelate complexes have recently attracted particular attention as building blocks in donor–acceptor copolymers.179–181 Among the major applications for boron chelate complex-based macromolecular materials are uses as emissive and charge-transporting components in organic light-emitting diodes (OLEDs), in organic photovoltaics (OPVs), for bioimaging, as lasing materials, and as stimuli-responsive materials. Several valuable macromolecules comprising exclusively linearly linked tetracoordinate boron centers, which are usually described as coordination polymers, have been developed as well. Such species often show dynamic behavior due to reversible formation of dative bonds. Polymeric materials featuring charged boron moieties such as borates (polyelectrolytes), which have been used, for example, as organogels for ion conduction,182–184 are not discussed herein.

14.04.2.2.1

Coordination polymers with linear dative B

N bonds in the main chain

Early examples of such polymers reported by Itsuno et al. have been prepared by self-polyaddition of amine–borane adducts featuring vinyl groups at the amine moiety (Scheme 42).185 Reactions of 4-vinylpyridine (166) with boranes 167a-c gave coordination polymers 168a-c through adduct formation followed by spontaneous intermolecular hydroboration in THF at

Scheme 42 Synthesis of coordination polymers 168a-c by self-polyaddition and structures of further vinylamine monomers.

Organoboron and Related Group 13 Polymers

97

ambient temperature. The highest molecular weight was obtained for 168a with R ¼ H (Mn ¼ 22.4 kDa, PDI ¼ 1.8, by GPC). The polymer was soluble in polar solvents such as DMF and DMSO. It was found to be inert to alcohols, water, acetic acid, and 2N hydrochloric acid, which was attributed to its insolubility in these media. When suspended in acetone or DMF solution, it readily reacted with hydrochloric acid and degraded into boric acid derivatives under H2 evolution. An attempt to synthesize the corresponding polymer with R ¼ Br resulted in the formation of insoluble material. Further vinylamines were additionally tested as monomers, and the best results were obtained with derivatives containing the 4-pyridyl moiety, 169–174; compounds 175 and 176 proved to be suitable monomers as well. Poly(aminoboranes) are inorganic polymers that comprise a linear chain of alternating tetracoordinate B and N atoms. These materials will be discussed in Section 14.04.4. They are most efficiently prepared by catalytic dehydrogenation/dehydrocoupling of certain amine–borane adducts. The latter have been used, for example, as reductants and hydrogen transfer reagents and, particularly the parent ammonia–borane, NH3 ∙ BH3, has been discussed as a possible molecular hydrogen storage material for onboard applications. Certain difficulties in practical implementation thereof, however, have led researchers to search for viable alternatives for that purpose. In this context, Lacôte, Raynaud, and co-workers investigated a series of poly(amine–boranes) having p-phenylene groups bridging adjacent boron centers and aliphatic linkers between the nitrogen centers, 179a-d (Scheme 43).186 The synthesis of these Lewis pair-assembled polymers was achieved by combining diammonium salts 177a-d and bisboronic acid 178 in 1:1 ratio in THF (heterogeneous mixture) and in situ reduction of the latter with LiAlH4 concomitant with deprotonation of the diammonium species. Size-exclusion chromatography (SEC) analysis with THF as the eluent and in the presence of 1 wt% of [nBu4N]Br suggested molar masses of >105 Da. The amorphous polymers 179a,b showed glass transition temperatures of Tg  60  C. Investigation of the thermal dehydrogenation behavior of 179a,b by temperature-programmed desorption measurements supported by thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) revealed that the polymers liberate H2 at 96  C and 95  C, respectively; that is, at lower temperatures than other, molecular amine–boranes used for hydrogen storage and also at lower temperatures than isolated molecular analogs of the polymers, without catalysis. Interestingly, the dehydrogenation process of the polymers was found to be endothermic, while it was exothermic in the case of their molecular analogs, as it is also usually the case for regular amine–boranes. Moreover, H2 release was not associated with a reduction of the molecular weight of the polymers; rather, some degree of mass increase was observed for the partially THF-soluble solid products. They underwent a second dehydrogenation process at about 160  C. The final products were found to be soluble in LiBr-containing DMF (0.5 wt%) at 60  C, thus enabling their SEC analysis (>105 Da). The observed mass gain was suggested to be due to intermolecular interchain interactions in the dehydrogenated products.

Scheme 43 Synthesis of poly(amine–boranes) 179a-d by polycondensation reactions and transfer hydrogenation of imines, aldehydes, and ketones using 179b.

Additionally, it was demonstrated that polymer 179b can serve as an efficient hydrogen transfer reagent for the reduction of imines, ketones, and aldehydes under mild conditions. Also in this respect, the polymers showed a better performance than their molecular analogs. Wagner and co-workers reported one-dimensional reversible coordination polymers through N ! B dative bond formation containing ferrocene units and 4,40 -bipyridine (181a), 1,2-bis(4-pyridyl)ethane (181b), or pyrazine units (183), respectively (Scheme 44).187–190 Reactions of 1,10 -bis(dimethylboryl)ferrocene with the respective ditopic nitrogen base in solution led to precipitation of the insoluble polymers. The macromolecular nature of 182 and 181a was confirmed by single-crystal X-ray188 and powder diffraction189 studies, respectively. Polymer 181a was found to be thermally stable in the in the solid state up to 240  C, but in the presence of toluene, its formation was fully reversed at 85  C.187 The polymers with the fully conjugated, redox-active

Scheme 44

Synthesis of ferrocene-containing reversible coordination polymers 181a,b and 182, and supramolecular polymer 183.

bridging ligands 181a and 182 had intense dark colors, while 181b was yellow.187,188 This was attributed to a charge transfer from the electron-rich ferrocene unit to the electron-deficient heterocyclic ligand. Jäkle and co-workers reported the self-assembly of polymers having borane end groups through donor-acceptor interactions with 4,40 -bipyridine resulting in polymers such as 183 with elongated chains (Scheme 44).191 Severin and co-workers explored the assembly of various boronate esters and nitrogen donor molecules via dative N !B bonds.42,192–195 For example, the three-component reaction of aryl boronic acids with 1,2,4,5-tetrahydroxybenzene and 1,2-bis(4-pyridyl)ethylene or 4,40 -bipyridine afforded the dark purple colored boronate ester coordination polymers 184a-c (Scheme 45).192 Their polymeric nature in the solid state was confirmed by single-crystal X-ray diffraction studies on 184a and

Scheme 45 Synthesis of coordination polymers 184a-c by three-component reactions and structures of coordination polymers 186–189.

Organoboron and Related Group 13 Polymers

99

184b, revealing that the polymer strands show a zig-zag arrangement. Upon heating 184a,b in chloroform, dissolution occurred concomitant with loss of the purple color. NMR spectroscopic investigations for 184a revealed that the BdN bonds were mostly disrupted in solution, suggesting that 184a is in equilibrium with bis(dioxaborole) 185a and 1,2-bis(4-pyridyl)ethylene. Upon cooling, again purple crystals of 184a,b precipitated from the mixture, thus proving the reversibility of the formation of the coordination polymers. Computational studies revealed that the intense colors of 184a-c comes from intrastrand charge transfer excitations from the tetraoxobenzene moiety to the dipyridyl linker. In a similar way, 1,4-benzenediboronic acid, catechol, and the same bipyridyl ligands assembled to the crystalline linear coordination polymers 186a,b.194 Polymer networks 187a,b were formed when the corresponding triboronic acid was employed.193 The effect of higher polypyridyl ligands in three-component reactions with 1,4-benzenediboronic acid and catechol was also investigated.194 With tetra(4-pyridylphenyl)ethylene, the 2D polymer network 188 was obtained. The reaction of 1,4-benzenediboronic acid, catechol, and 2,4,6-tri(4-pyridyl)-1,3,5-triazine (tpt)—initially performed in 3:6:2 ratio—did not lead to the expected prismatic cage; it rather gave the macrocycle 190, in which only two of the three pyridyl donor sites of tpt are utilized (Scheme 46). Interestingly, when the reaction was performed in the presence of conene—with a view to incorporate the PAH molecule as a guest in the center of the macrocycle—the linear coordination polymer 191 was formed, wherein the coronene is intercalated between the polymer strands, as shown by single-crystal X-ray diffraction. The same product was also formed upon addition of coronene to the preformed macrocycle 190. This is a rare example of a template-induced ring opening of a macrocycle. It should be noted that templates are typically used to favor the formation of macrocyclic over polymeric structures.

Scheme 46 Assembly of 1,4-benzenediboronic acid, catechol, and 2,4,6-tri(4-pyridyl)-1,3,5-triazine (tpt) to macrocycle 190 and, in the presence of coronene, to linear coordination polymer 191.

One-dimensional coordination polymers 189a,b (Scheme 45) were formed via self-assembly of arylboronate esters having a 4-pyridyl side chain, formed by condensation of the respective arylboronic acid with 2-(4-pyridyl)propane-1,3-diol under Dean–Stark conditions. Their structures were confirmed by X-ray crystallography. It turned out that electron-withdrawing substituents at boron were crucial for polymer formation. When the substituent on boron was 4-(tert-butyl)phenyl, on the other hand, the respective monomer was present both in solution and in the solid state. Höpfl and co-workers reported 2D and 3D molecular networks with macrocyclic structures formed by the self-assembly of in situ generated bisboronate esters with two intramolecular nitrogen donor sites.196 Recently, Drover et al. presented a new diphosphine ligand with four dicyclohexylboryl groups, which they coordinated to Ni(O) to give an octaboraneyl complex.197 Reaction thereof with 4,40 -bipyridine yielded a coordination polymer through dative N !B bonds. Solid state NMR and EPR spectroscopy revealed that the polymer has charge transfer character due to intramolecular electron transfer (ET) from nickel to the 4,40 -bipy ligands. Iwasawa and co-workers reported on the reversible formation of an organogel containing tetracoordinate boron atoms from a macrocyclic boronic ester having organic guest molecules encapsulated (Scheme 47).198 Esterification of 2,5-difluoro-1,4phenylenediboronic acid (193) with indacene-type bis(1,2-diol) 192 in the presence of azulene (or toluene or naphthalene or benzothiophene) resulted in guest-induced self-assembly to the guest-included macrocycle 194. Addition of 1,3-diaminopropane to a suspension of 194 led to formation of the borate-based gel 195, wherein the boronic ester moieties are cross-linked by the diamine, concomitant with release of the guest molecule. The dynamic nature of both boronic ester formation and amine binding allowed for the observation of stimuli-responsive properties. Heating of the gel triggered its reversible transition to a clear solution, while treating it with fluoride (through TBAF) or aqueous NaOH led to irreversible gel–sol transitions. When the gel was treated with 1 M aqueous HCl, it collapsed and the macrocyclic complex 194 could be recovered in 84% yield.

100

Organoboron and Related Group 13 Polymers

Scheme 47 Assembly of guest-included macrocycle 194, and formation of borate-based gel 195 upon addition of 1,3-diaminopropane.

14.04.2.2.2

Polymers with boron chelate complexes in the main chain

In an early effort, Chujo and co-workers prepared a series of poly(cyclodiborazanes)199 by hydroboration polymerization of various dicyano compounds with dihydroboranes.200–208 The most stable products were obtained with mesityl- and tripylborane;200–206 a chiral B-substituent was also incorporated.207 In Scheme 48 only the derivatives with Ar ¼ Mes or Tip and with conjugated or p-electron system-containing linking groups are depicted. Although extended p conjugation cannot occur over the boron centers, where the conjugation is disrupted by tetracoordination, nonetheless, these species showed some interesting photophysical properties depending on the nature of the linking group. Some derivatives were moderately fluorescent. With electron-rich building blocks, such as dialkoxybenzene and oligothiophenes, the UV-vis absorption and fluorescence emission bands of the polymers showed bathochromic shifts, thus pointing to some intramolecular charge transfer character. Within the series of oligothiophene-bridged

Scheme 48 Synthesis of poly(cyclodiborazanes) 196 and 197 by hydroboration polymerization of dicyano compounds, and synthesis of 199 by Sonogashira–Hagihara cross-coupling polycondensation.

Organoboron and Related Group 13 Polymers

101

polymers 196/197e-i, progressive bathochromic shifts occur with elongation of the bridge, thus allowing some color tuning.203 Poly (cyclodiborazane) 197j, featuring electron-donating dithiafulvene units, showed semiconducting behavior with an electrical conductivity of 2  10−5 S cm−1. A cast film of a charge transfer complex of the 197j with the oxidant 7,7,8,8-tetracyanoquinodimethane (TCNQ) had a significantly increased conductivity of 1  10−4 S cm−1.204 Derivatives of poly(cyclodiborazanes) with disilane linking groups205 and transition metal alkynyl complexes206 in the main chain were also explored. As an alternative route to poly(cyclodiborazanes), Chujo and co-workers investigated Sonogashira–Hagihara cross-coupling polycondensation of a preformed, dibromofunctionalized cyclodiborazane, 198, with 2,5-didodecyloxy-1,4-diethynylbenzene. Four-coordinate organoboron compounds are often sufficiently stable to allow for such transformations. Compared to the results from hydroboration polymerization, the product, 199, obtained from the organometallic polycondensation procedure had a somewhat increased molecular weight (8.5 kDa) and less structural defects; the formed contained some iminoborane moieties as defects, presumably end groups, as evidenced by 11B NMR spectroscopy. Chujo and co-workers also investigated a series of polymers that contain pyrazabole moieties in the backbone, 202a-h (Scheme 49).209–212 These poly(pyrazaboles) were synthesized by Sonogashira–Hagihara polycondensation either from diiodo-functionalized pyrazabole 200 and the respective diethynyl compounds (for 202a-c) or from diethynyl-functionalized pyrazabole 201 and dibromo-functionalized building blocks (for 202d-h). Polymer 202a was obtained in molecular weights up to Mn ¼ 34 kDa. Due to the lack of conjugation over the boron heterocycle, no significant red-shifts were observed compared to the molecular constituents of the polymers. However, most of these poly(pyrazaboles) showed strong fluorescence emission in the near-UV region—presumably coming from the pyrazabole moieties. The nature of the organic linker had some influence on the wavelength of the emission maximum: groups having extended delocalized p systems (e.g., h) gave bathochromic shifts while p-acceptor groups (g) caused hypsochromic shifts. Fluorescence emission could also be stimulated by exposure to neutrons or a-particles, which is of potential interest for the use as scintillating materials for neutron detection.211 Chujo and co-workers, furthermore, incorporated transition metal (Ni, Pd, Pt) alkynyl complexes into poly(pyrazaboles).212

Scheme 49 Synthesis of poly(pyrazaboles) 202a-h by Sonogashira–Hagihara cross-coupling polycondensation (Hex ¼ n-hexyl).

Following the discovery by Wang and co-workers that organoboron quinolato groups (R2Bq, q ¼ 8-hydroxyquinolate) can be advantageously applied as active components of electron conduction and emission layers in OLED devices,213,214 as an alternative to the established aluminum tris(8-hydroxyquinolate) (Alq3) and its derivatives, these chromophores have attracted considerable attention. Chujo and co-workers reported the first linear polymers having R2Bq moieties incorporated in the main chain. They used Sonogashira–Hagihara polycondensation of preformed diiodo-functionalized diphenylboron quinolate monomers 203 with diethynylbenzene derivatives 204 to synthesize copolymers 205 (Mn ¼ 3.2–8.8 kDa, PDI ¼ 1.7–2.4, by GPC; Scheme 50).215,216 The long alkyl side chains at the conjugated organic linker groups ensure good solubility. The polymers proved to be luminescent with emission maxima between 484 and 489 nm (in CH2Cl2) for 205a-c, while methyl substitution in the 2- and 4-position of the quinolato group led to a slight red-shift and to an increased quantum efficiency (for 205e: lmax ¼ 472 nm, FF ¼ 29%). Absorption of the diylkyne chromophore leads to an antenna effect, with energy transfer to the organoboron quinolate. Compound 205e, for instance, showed intense blue emission both in solution and as a thin film.

102

Organoboron and Related Group 13 Polymers

Scheme 50 Synthesis of polymers 205a-e containing organoboron quinolato groups in the main chain, and structures of related polymers 206–208.

Subsequently, Chujo and co-workers presented related polymers 206–208,217–219 which were prepared in a similar way (Scheme 50). Exchanging the oxygen atom of the quinolato ligand by heavier elements of the group 16 (206a-f) had a pronounced effect on the photophysical properties of these materials.217 Upon descending the group, significant bathochromic shifts of the emission bands were observed, while the quantum yields decreased. Interestingly, the polymers showed dual emissions, being most pronounced for the selenium-containing derivative 206c. This was attributed to competing emission pathways from the dialkyne and the quinolato moiety. Polymers 207a-c featuring aminoquinolate chelated boron centers show slightly red-shifted emission bands compared to their quinolate congeners (lmax,em ¼ 503 nm for 207c).218 Calculations suggested that the exchange of oxygen by the N-acyl group causes a lowering of the LUMO energy level. Polymers 208a,b contain the 10-hydroxybenzo[h]quinolinato ligand, which can be regarded as a p-extended 8-hydroxyquinolinato ligand.219 DFT calculations on molecular model compounds suggested that the fluorescence emission of these materials comes from an intraligand charge transfer (ILCT) excited state. Experimental studies revealed an increase of the fluorescence quantum efficiency when going from molecular diphenylboron 10-hydroxybenzo[h]quinolate (FF ¼ 10%) to the respective polymers 208a,b (FF ¼ 16% and 18%). Additionally, further synthetic routes to organoboron quinolate-containing polymers have been developed that proceed via BdO and BdN bond formation in one step either at the postpolymerization stage (i and ii) or during the polymerization (iii) (Scheme 51). Chujo and co-workers reported the synthesis of polymers 210a-c by postmodification of 209 through condensation of the 8-hydroxyquinoline group in the main chain with BPh3 or (C6F5)2BF ∙ Et2O, respectively (i).220,221 Also for 210a, it was suggested that the conjugated main chain behaves as a light harvesting antenna for the organoboron quinolate units. Polymers 210b,c proved to have n-type semiconducting properties. In an electron-only device with the structure ITO/Ca/polymer/LiF/Al, they showed electron mobilities of 3.9 and 2.0  10−5 cm2 V−1 s−1, respectively.

Organoboron and Related Group 13 Polymers

103

Scheme 51 Synthesis of R2Bq-containing polymers 210a-c, 211a,b, and 213a,b by BdO/BdN bond formation routes.

Jäkle and co-worker synthesized the main chain-R2Bq-containing polymers 211a,b (Scheme 51(ii)) via macromolecular substitution of 15a through ether cleavage by the B–Br groups (cf. Scheme 7 in Section 14.04.2.1.1).82 The molecular weights of 211a,b (Mn ¼ 7.9 and 10.8 kDa, respectively) are predetermined by the chain length of precursor polymer 15a. It is noteworthy that 4-dimethylamino group at the quinolato ligand in 211b causes a bathochromic shift resulting in a very weak red emission. Jäkle and co-workers developed a further route to polymers with quinolato coordinated boron atoms in the main chain, 213a,b, in which both the BdO and the BdN bonds are formed in the polymerization process (Scheme 51(iii)).222 Reaction of the methyl-protected bifunctional hydroxyquinolate species 212a,b with a bifunctional bromoborane brought about polycondensation to give 213a,b with concomitant cleavage of volatile CH3Br as the by-product. The photophysical properties of 213a,b showed a strong dependence from the nature of the linking p systems. DFT calculations revealed that the lowest-energy absorption of 213a is assigned to an intramolecular charge transfer (ICT) process from the Thi-C6H4-Thi group to the pyridyl moiety, while in the case of 213b, ICT occurs from the fluorene moiety to the pyridyl groups. Polymer 213a showed a concentration-dependent dual emission, which was suggested to result from excimer formation. Boron dipyrromethene (BODIPY) dyes are certainly among the most widely used organic fluorophores (Scheme 52). They are noted for their strong and sharp absorption and emission features in the visible region (>500 nm), with small Stokes shifts and high quantum efficiencies, combined with excellent chemical and photostability.223–225 Consequently, they have been used for various applications, ranging from charge transporting and emitting layers in organic electronic devices (OLEDs, OFETs, OPVs),226–228 sensors,229 and fluorescent probes for biomedical imaging,230 to therapeutic applications such as photodynamic therapy (PDT).231–233 Derivatives of aza-BODIPY, the meso-nitrogen-doped analogs of BODIPY, have attracted considerable attention as well.234

104

Organoboron and Related Group 13 Polymers

Scheme 52 Generic structures of the BODIPY and the aza-BODIPY framework, synthesis of BODIPY-containing alternating copolymers 215a-d by Sonogashira–Hagihara polycondensation, generic structure of BODIPY–arylene alternating copolymers 216, BODIPY homopolymer 217, and copolymers 216a-d and 218.

The first BODIPY-containing polymers were reported in 2008.235 Since then, these fluorophores have been incorporated into polymers in various ways, and the chemistry of the resulting materials was recently reviewed.236,237 The most common linking mode for BODIPY in the main chain of conjugated polymers is via the positions 2 and 6 (see Scheme 52 for numbering of the positions). In this way, extended p conjugation along the chain and over the BODIPY moiety is possible. Li and co-workers made use of Sonogashira–Hagihara cross-coupling between 2,6-diiodo BODIPY 214 and different diethynyl compounds to synthesize alternating copolymers 215a-d.235 The ethynyl bridges prevent possible steric interactions between the aromatic p moieties and the methyl substituents on the pyrrole rings of the BODIPY unit that could lead to backbone twisting and thus diminish extended p conjugation. Most BODIPY/aza-BODIPY-based materials carry peripheral substituents on the pyrrolic backbone (often methyl groups), as the parent unsubstituted ring systems are difficult to access due to the instability of the intermediates during their

Organoboron and Related Group 13 Polymers

105

synthesis. Polymers 215a-d showed moderate solubility. The THF-soluble fractions consisted of polymeric material of moderate molecular weight (Mn ¼ 1.7–6.0 kDa, by GPC), suggesting that probably higher-molecular-weight material was additionally present. The UV-vis spectra of 215a-d displayed absorption bands due to p–p transitions centered at the BODIPY moiety at lmax ¼ 570–640 nm, and the polymers showed fluorescence at lmax ¼ 615–664 nm (FF up to 25%). These bands are significantly bathochromic shifted compared to those of molecular congeners thereof, and the shift shows a pronounced dependence of the nature of the co-monomer, which clearly demonstrates effective conjugation over the BODIPY unit through the 2,6-linkages. Shortly afterwards, Liu and co-workers reported further BODIPY copolymers by Sonogashira–Hagihara polycondensation that are structurally related to 215, but they introduced long alkoxy groups at the aryl substituent (R) in the meso-position of the BODIPY unit.238,239 This had a solubility-enhancing effect and, consequently, led to higher-molecular-weight polymers (Mn ¼ 14.4–26.7 kDa). Subsequently, the same group reported BODIPY alternating copolymers of type 216 (Scheme 52, bottom row), in which the arylene moieties of the co-monomer units are directly connected to the BODIPY core via its positions 2 and 6 (without ethynyl bridges).240 They were prepared by Suzuki–Miyaura cross-coupling of derivatives of 214 (with alkyl or alkoxy substituents at the meso substituent) and the appropriate diboronic acids. In 2009, Herrmann and Börsch and co-workers added the first BODIPY homopolymer 217, prepared by Ni-catalyzed Yamamoto coupling.241 Until today, a considerable number of copolymers of type 215235,238,239,241–247 (via Sonogashira–Hagihara coupling) or 216240,241,248–259 (mostly via Stille or Suzuki–Miyaura coupling) have been prepared with a view to implement them for various applications. Qin and co-workers reported the synthesis of a BODIPY–Pt-diacetylenide copolymer.260 The effects of substituents at the BODIPY core on the electronic and photophysical properties of copolymers of type 216 shall be exemplified by comparison of 216a with 216b and of 216c with 216d. Exchange of the methyl group in the meso position in 216a by a CF3 group to give 216b results in a red-shift of the peak absorption by ca. 160 nm (from lmax ¼ 486 nm for 216a to lmax ¼ 645 nm for 216b).255 The CF3 group has a strong stabilizing effect on the LUMO, which has a large coefficient at C8, while it has a negligible effect on the HOMO, which has a nodal plane in this position. This leads to an overall reduction of the band gap. The same effect is responsible for the band gap reduction when going from the BODIPY to the aza-BODIPY framework.234 As noted above, substituents R0 /R00 at the pyrrolic rings may sterically interfere with the arene groups directly bound to the BODIPY system in copolymers 216, which may cause twisting of the polymer backbone with detrimental effects on the conjugation along the polymer chain. Facchetti, Stoddart, Usta, and co-workers synthesized copolymers of substituted bithiophene and BODIPY building blocks with (216c) and without (216d) methyl groups in positions 1 and 7 (R0 ).249 They found that omitting the CH3 groups leads to a reduction of the optical band gap from 1.67 (216c) to 1.23 eV (216d) in the solid state (lmax,abs ¼ 595 nm for 216c vs. lmax,abs ¼ 850 nm for 216d). This effect was attributed to planarization of the backbone and, related with that, improved p–p stacking in the thin film. In alternating copolymers with donor–acceptor (D–A) structure the BODIPY unit can serve as either the donor or the acceptor component. Thayumanavan and co-workers performed a systematic study, which showed that the combination of the same BODIPY moiety with different donor and acceptor moieties leads to a reduction of the band gap relative to that of a control polymer in both cases.242,243 Chochos and colleagues recently presented the D–A conjugated copolymer poly(quarterthiophene-altthienylBODIPY) 218, which exhibited an optical energy gap as low as 0.85 eV.254 This example benefits from a pronounced D–A structure and the absence of substituents at the pyrrolic rings. Furthermore, due to the absence of substituents at C1 and C7, the hetaryl (thienyl) substituent in the meso position can adopt a coplanar arrangement with respect to the BODIPY system, which leads to a further reduction of the band gap.240,242–244,251–254,256,258,259 Compound 218 showed p-type semiconductivity in an OFET device, with a modest hole mobility of 5  10−4 cm2 V−1 s−1.254 Fréchet and co-workers were the first to investigate the use of BODIPY-containing polymers in bulk heterojunction (BHJ) solar cells.248 Subsequently, various copolymers have been examined for that purpose,246,247,250,251,253,256,257,259,260 some of which achieved considerable power conversion efficiencies (PCEs). Li, Li, and co-workers fabricated an organic solar cell (OSC) with a thienylBODIPY-alt-thienyl benzodithiophene polymer as the donor blended with a non-fullerene small molecule acceptor, which gave a record PCE (for BODIPY-based solar cells) of 9.86%.259 Hong and Lee and co-workers incorporated their mesoCF3 substituted copolymer 216b as a dopant-free hole transporting material into a perovskite solar cell (PSC), which gave a PCE of 16.02% (i.e., much higher than that using 216a, PCE ¼ 3.85%).255 A series of chiral BODIPY-containing conjugated polymer enantiomers, in which axial chiral 1,10 -binaphthol (BINOL) moieties were incorporated in the main chain, were presented by Cheng, Li, and co-workers.245 The polymers exhibited strong Cotton effects, and they showed circularly polarized luminescence in THF solution, which could be regulated by the dihedral angles of the BINOL moieties. Xie and co-workers prepared nanoparticle formulations of a series of conjugated 2,6-linked BODIPY–diketopyrrolopyrrole copolymers, and evaluated their use as photothermal agents for photothermal therapy (PPT) via both in vitro and in vivo experiments.258 The new materials exhibited good cytocompatibility and photostability, high photothermal conversion efficacy, and efficient photothermal activity towards cancer cells. Valiyaveettil and co-workers reported hyperbranched BODIPY-containing polymers 219–221, synthesized via Sonogashira–Hagihara cross-coupling (Scheme 53).244 The polymers were found to be soluble in common organic solvents, which allowed for their molecular weight determination by GPC in THF (Mn ¼ 10–25 kDa). Upon coating them on a quartz crystal microbalance, they were evaluated for their use for the detection of organic vapors. Particularly, 220 showed a high sensitivity towards benzene and toluene vapors, and a higher sensitivity for toluene over benzene compared to linear BODIPY copolymers. Liras et al. reported the synthesis of microporous polymer network 222 via Sonogashira–Hagihara cross-coupling reactions (Scheme 53).261 The luminescent polymer proved to be applicable as an efficient heterogeneous photocatalyst for the oxidation of thioanisole to methylphenyl sulfoxide, via the generation of singlet oxygen.

106

Organoboron and Related Group 13 Polymers

Scheme 53 Structures of BODIPY-containing hyperbranched polymers 219–221, microporous polymer networks 222 and 223a-c.

Organoboron and Related Group 13 Polymers

107

Conjugated polymers containing 2,6-linked aza-BODIPY fluorophores in the main chain were first developed by Sauvé and co-workers.262,263 The procedure they chose involved the synthesis of azadipyrromethene-alt-p-phenylene ethynylene polymers 224a-f via Sonogashira–Hagihara polycondensation in the first step (Scheme 54). The boron centers were subsequently introduced by coordination of the azadipyrromethene units with BF2 moieties through the reaction of the polymers with BF3 ∙Et2O in the presence of Hünig’s base to give 225a-f. Derivatives 225a-c were of modest molecular weight (Mn ¼ 2.2–11 kDa), due to solubility issues.262 However, when the branched alkyl group 2-ethylhexyl was attached as a side chain, for 225d-f, molecular weights of Mn ¼ 5.0–15.3 kDa were achieved.263 Polymers 225a-c emitted light in the NIR range. They exhibited optical band gaps of about 1.3 eV, i.e., significantly reduced compared to those of the precursor polymers 224a-f (Eg ¼ 1.40–1.49 eV). The authors additionally prepared analogous metallopolymers by coordination of 224d-f with Cu(I)PEt3 and Ag(I)PEt3, respectively.263 Zhu, Cheng, and co-workers reported the synthesis of related D–p–A copolymers 226a-c via Sonogashira–Hagihara cross-coupling from preformed aza-BODIPY monomers.264

Scheme 54 Synthesis of copolymers containing 2,6-linked aza-BODIPY units 225a-f and 226a-c (Oct ¼ n-octyl).

Hupp, Stoddart, Deria, and co-workers recently designed and elaborated BODIPY-based porous organic polymers 223a-c (Scheme 53) for the photochemical detoxification of a sulfur mustard simulant.265 The synthesis of the parent polymer 223a was achieved by Sonogashira–Hagihara cross-coupling via the substituents on the boron center and that in the meso-position of the BODIPY unit. Bromination or iodination of the 2,6-positions was performed at the post-polymerization stage. This led to quenching of the fluorescence and enhancement of the phosphorescence of the material due to efficient intersystem crossing through the heavy atom effect. Polymers 223b,c proved to work as efficient photosensitizers for the generation of singlet oxygen (1O2), which was successfully applied for the selective catalytic photo-oxidation of the sulfur mustard simulant 2-chloroethyl ethyl sulfide to 2-chloroethyl ethyl sulfoxide. Besides linkage through the 2- and 6-positions, various other linking modes for the incorporation of BODIPY or aza-BODIPY units into the main chain or side chains (Section 14.04.3.2) of polymers have been exploited in addition (Scheme 55). Chujo and co-workers reported highly luminescent polymers 227, in which BODIPY moieties are embedded into the polymer’s main chain via

108

Organoboron and Related Group 13 Polymers

Scheme 55 BODIPY- and aza-BODIPY-containing copolymers or classes of polymers with various linking modes.

boron-alkynyl groups.266 These materials underwent supramolecular self-assembly due to their rod–coil type structure. Chujo’s group also explored the incorporation of aza-BODIPY chromophores into conjugated polymers via the 3,5-(compounds 228) or the 1,7-positions (compounds 229) of the pyrrolic rings, respectively.267 In 2014, they presented both 3,5-linked BODIPY- and azaBODIPY-based polymers 230a,b, which showed high electron mobility [(1.5–3.6)  10−4 cm2 V−1 s−1] and low threshold voltage (5–7 V) in electron-only devices.268 Samuel, Skabara, and co-workers also used the 3,5-positions of BODIPY to embed it into copolymers combined with bis(3,4-ethylenedioxythiophene) (bis-EDOT) (231a) or its all-sulfur analog bis(3,4-ethylenedithiathiophene) (bis-EDTT) (231b) via Stille coupling reactions.269 These polymers exhibited ambipolar charge transport behavior. Polyfluorenes doped with 3,5-linked aza-BODIPY moieties, 232, were presented by Burgess and co-workers in 2011.270 A few reports described the use of aromatic ring-fused BODIPY-based as building blocks for conjugated polymers (Scheme 56). Chujo and co-workers presented polymers 233a-c, 234a-c, and 235a-c, featuring boron di(iso)indomethene (BODIN) units in combination with further conjugated co-monomer units that are bound either via an etynyl group (233a-c)271 or directly via CAr–CAr bonds (234a-c and 235a-c)272 to the annelated benzene ring of the BODIN moiety. These polymers emitted in the deep red to NIR region of the spectrum with high quantum yields (FF ¼ 33–79%). In a postmodification process, 234b was transformed into 236b via a cyclization reaction utilizing the ortho-methoxy groups at the attached arenes upon reacting them with BBr3. This led to a further red-shifted and sharper photoluminescence. Copolymers containing BODIN units were recently synthesized and used by Chiu, Wu, and co-workers as long-wavelength-excitable near-IR polymer dots with narrow and strong absorption features for in vivo tumor imaging giving a high contrast.273 Yu and co-workers reported donor–acceptor-type copolymers with thieno-fused BODIPY moieties, 237a-c (Scheme 56).274 In organic thin-film transistors they exhibited ambipolar charge mobilities up to 10−3 cm2 V−1 s−1.

Organoboron and Related Group 13 Polymers

109

Scheme 56 Examples of polymers featuring ring-fused BODIPY moieties in the main chain.

Certain, appropriately substituted BODIPY dyes undergo photocleavage in response to visible (blue-green or green) light.275–277 Ambade and co-workers demonstrated the use of this feature for the visible light-triggered disruption of micelles of an amphiphilic block copolymer having a BODIPY unit at the junction as a photocage (Scheme 57).278 Block copolymer 239 was synthesized from its monomer 238, which contains both an atom transfer radical polymerization (ATRP) initiating group and an ethynyl group. ATRP with styrene and azide–alkyne click reaction with N3-terminated polyethylene glycol (PEG) were accomplished by copper(I) catalysis simultaneously in a one-pot reaction. Self-assembled micelles of 239 were prepared by dialysis of a THF solution of 239 against water. Irradiation with blue light caused selective photocleavage of the BdO bonds at the junction, as evidenced by TEM, DLS, and the observation of precipitation of the hydrophobic polystyrene block and detection of 4-hydroxybenzyl alcohol by HPLC.

Scheme 57 Synthesis and degradation of block copolymer 239.

110

Organoboron and Related Group 13 Polymers

Truong and co-workers utilized the propensity of BODIPY dyes having a pendent thioether attached at the meso-position to undergo photolysis of the CdS bond under visible (green) light irradiation, producing an ion pair intermediate that can react specifically with a propiolate group, for the fabrication of hydrogels and for the photo-immobilization of biomacromolecules.279 Boron difluoride formazanates,280 which comprise a tetracoordinate boron atom embedded in a nitrogen-rich six-membered ring system, have been recently introduced as a feasible alternative to BODIPYs and related compounds. In recent years, the potential of this emerging class of molecular materials has been extensively explored by the Gilroy group.281,282 Their work included the synthesis of the first conjugated polymers containing this chromophore in the main chain (Scheme 58). They prepared the copolymer 241 (Mn ¼ 17.0 kDa, PDI ¼ 2.14) by an azide–alkyne click reaction as well as a number of molecular model compounds.283 These species exhibited high molar absorptivities (25,700–54,900 M−1 cm−1) in the visible region, large Stokes shifts (3590–3880 cm−1), and tunable electrochemical behavior (E red1 ca. −0.75 V and E red2 ca. −1.86 V, vs. ferrocene/ferrocenium). Their studies further revealed that the p conjugation along the backbone does not extend beyond the triazole groups. Nevertheless, the thin-film optical band gap of the polymers was estimated as 1.67 eV. Subsequently, Gilroy and co-workers reported on polymer 242, which was obtained by copper(I) catalyzed coupling of 240 with trans-[Pt(PBu3)2Cl2].284 This metallopolymer exhibited an optical band gap of ca. 1.4 eV. Comparison of the polymer to model compounds confirmed that its unique optoelectronic properties can be directly attributed to the presence of the boron difluoride formazanate repeat unit, and that the platinum complex unit must also be present to achieve the narrow band gaps observed.

Scheme 58 Synthesis of polymers containing boron difluoride formazanate groups 241 and 242, and further BF2 formazanate-containing polymers 243a,b and 244.

Banewar and Zade and co-workers employed electrochemical polymerization of BF2 formazanate complexes having thiophene or 3,4-ethylenedioxythiophene (EDOT) termini to obtain polymers 243a,b, respectively (Scheme 58).285 Recently, Tanaka and colleagues reported a series of conjugated BF2 formazanate-containing copolymers, which they prepared by Stille cross-coupling polycondensation reactions using dibrominated formazanate complexes and stannylated fluorene and bithiophene building blocks.286 They also investigated boron complex formation on the post-polymerization stage, which was used to obtain the random copolymer 244 with varying ratio of boron incorporation. Their studies revealed that these polymers show NIR emission upon charge transfer to the strongly electron-accepting BF2 formazanate group. Polymer 244 proved to be applicable as a wavelengthconversion material. By varying the complexation ratio, broad absorption bands covering the whole visible region were obtained, while a single emission band was generated in the NIR. Boron complexes of b-diketonates, ketoiminates, and diiminates are further interesting boron heterocycles that deserve mentioning in the macromolecules’ context. The first polymers containing such chromophores were polylactides (PLA) 246a,b that bear BF2 diketonate moieties as end groups, reported by Fraser and co-workers (Scheme 59).287,288 Their synthesis was accomplished from boron diketonates 245a,b having a pending OH functionality to initiate the Sn(II)-catalyzed ROP of DLlactide;287–289 analogous polymers derived from caprolactone were prepared as well.290 Fraser and co-workers discovered that polymers 246a,b display intriguing dual emission at room temperature, i.e., strong blue fluorescence besides green phosphorescence at the same time. The emission properties were, furthermore, found to be dependent of the polymers’ molecular weight.

Organoboron and Related Group 13 Polymers

111

Scheme 59 Synthesis of polymers 246a,b having boron difluoride diketonate end groups (oct ¼ octoate), an example of a conjugated polymer containing BF2 b-diketonate moieties in the main chain, 247, and modes of linking BF2 b-ketoiminate or b-diiminate moieties in the backbone of conjugated polymers, 248 and 249.

By nanoprecipitation, these polymers could be formulated into spherical nanoparticles, which were stable and retained their two-photon absorption and room temperature phosphorescence (RTP) properties in aqueous suspension.289 Significantly, the phosphorescence but not the fluorescence was effectively quenched by oxygen. This property was exploited for tumor hypoxia imaging.288,291 To this end, an iodine atom was introduced to the difluoroboron dibenzoylmethane moiety in order to further enhance the phosphorescence intensity due to the heavy atom effect. Furthermore, it was demonstrated that such nanoparticles were internalized into cultured HeLa cells by endocytosis.292 The incorporation of boron diketonate groups in the backbone of main-chain p-conjugated polymers has been investigated by Chujo and colleagues.293,294 For example, polymers 247 were obtained by modification of polymers composed of diketonate and fluorene groups via complexation with varying feed ratios of BF3 ∙Et2O (Scheme 59).294 Thereby the optical properties of the polymers, such as color and emission intensity, were effectively modulated. The same group295–297 and Cheng and Zhu and Wang and colleagues298,299 also investigated conjugated polymers comprising b-ketoiminate or b-diiminate units (Scheme 59). In these cases, the substituent on nitrogen offers an additional linking opportunity (cf. 249 vs. 248), whereby the heterocycle can be directly incorporated in the conjugation path along the polymer backbone. Either polymer modification routes296 or cross-coupling reactions of preformed boron complexes295,297–299 were applied for polymer formation. The substitution of O by NR in these polymers allows for aggregation-induced emission (AIE) and aggregation-induced emission enhancement (AIEE) properties to be observed, as a result of the introduction of a further substituent through the nitrogen atom. Ketoiminate-based conjugated polymers have also been utilized as fluorescence probes for HeLa cell imaging.299 Tanaka and colleagues recently presented the novel polymer 251, which is a derivative of poly(p-phenylene vinylene) (PPV) that contains azobenzene − boron complexes in the main chain (Scheme 60).300 It is formally derived from N2-doped PPV 252, and its monomer 250 is synthesized from the monomer to 252 having OH instead of OMe groups. Polymer 251 showed intense absorption and emission in the NIR (labs,max ¼ 702 nm, lem,max ¼ 760 nm, FPL ¼ 2.0% in diluted toluene) and high stability towards photodegradation.

Scheme 60 Synthesis of poly(p-phenylene vinylene) (PPV) derivative 251 containing azobenzene−boron complexes in the backbone, and N2-doped PPV 252.

112

Organoboron and Related Group 13 Polymers

N,C-chelate complexes of boron have recently emerged as attractive components for (opto)electronic materials.179–181 Yamaguchi and co-workers demonstrated that the borylation of thienylthiazoles with BMes2 groups, which is accompanied by the formation of an intramolecular dative B N bond (i.e., a borylative fusion), is an effective tool to significantly lower the LUMO energy, thus generating electron-transporting materials.301 Recently, Liu and co-workers used such a boracyclic thienylthiazole, namely (3-diphenylboryl-2-thienyl)-2-thiazole, as a building block to produce a series of p-conjugated copolymers of type 254 (Scheme 61).302–306 The boron complex moiety therein can be regarded as a BN/CC isostere of cyclopenta-[2,1-b:3,4b0 ]-dithiophene (CPDT), a widely-used electron-rich building block (donor) for organic optoelectronic materials. Liu’s group prepared copolymer 254a and its CPDT analog, 255, for comparison, both having the weak electron acceptor unit thieno[3,4-c] pyrrole-4,6-dione-1,3-diyl (TPD) as the co-monomer unit.302 Their studies revealed that the substitution of the C–C unit by the B N unit turns the electron donor moiety (CPDT) into an electron acceptor group. Quantitatively, this causes a decrease of the LUMO energy by 0.65 eV and the HOMO level by 0.53 eV. Several copolymers have been prepared via Stille polycondensation of the distannylated monomer 253 with the respective dibrominated co-monomers. Combinations with further electron-poor units were successfully used as polymer acceptors in efficient all-polymer solar cell devices.303–306

Scheme 61 Synthesis of copolymers 254 containing boracyclic (3-diphenylboryl-2-thienyl)-2-thiazole moieties, and comparison of polymer 254a with its CPDT analog 255.

Bazan and co-workers have demonstrated that the optical and electronic properties of D–A polymers can be effectively modified by the interaction with Lewis acids such as B(C6F5)3 (BCF).307–309 Coordination to Lewis basic sites of the polymers causes a decrease of their frontier orbital energies to a different extent, which leads to a reduction of the band gap. This concept has subsequently been used also by other researchers for different purposes.310–315 When applied to polymers with strongly electrondonating moieties, it gives rise to oxidative dopoing.316 Ingleson and Turner and co-workers demonstrated that the reaction of boranes with D–A materials containing the relatively weakly nucleophilic benzothiadiazole (BT) moiety, which is a well-established strong acceptor group in organic electronics, can lead to borylative fusion via electrophilic aromatic C–H borylation.317 Using this concept, they synthesized random copolymers of type 257 of varying borylation ratio via post-polymerization modification of 256 with BCl3 and subsequent arylation with ZnPh2 (Scheme 62).318 The resulting materials proved to be efficient solid-state NIR emitters for NIR-OLEDs, and formulation thereof into polymer nanoparticles (CPNPs) produced effective biological imaging agents.

Organoboron and Related Group 13 Polymers

113

Scheme 62 Synthesis of random copolymers 257 by postmodification via intramolecular electrophilic C–H borylation of 256, synthesis of ladderized BN-fused fluorene copolymers 260 and 261, and synthesis of copolymers 264 containing the BN-indacenodithiophene (BNIDT) building block.

Jäkle’s group prepared a series of luminescent organoboron ladder compounds via directed electrophilic aromatic C–H borylation of ortho-pyridyl-substituted fluorenes.319 With this concept they synthesized the dibrominated derivative 259, which was subjected to Yamamoto coupling to give the homopolymer 260, whereas Stille coupling afforded the vinylene-bridged copolymer 261 (Scheme 62).320 Very recently, Huang and colleagues developed BN-fused indacenodithiophene (abbreviated as BNIDT) as a further strongly electron-accepting building block for conjugated polymers of type 264 (Scheme 62).321–324 Several copolymers have been prepared, some of which showed high efficiencies (PCE up to 8.78%) as polymer acceptors in all-polymer solar cells.321,324 Liu and co-workers recently developed the double B N bridged bipyridine 266a (abbreviated as BNBP), as a novel electron-accepting building block for n-type conjugated polymers (Scheme 63).325 The parent polycyclic molecule 266a and the brominated monomer 266b are prepared by boron complexation of 265a,b, respectively. This leads to a significant decrease of the LUMO energy level from −1.03 eV for 265a (or −1.34 eV for 2,20 -bipyridine) to −2.52 eV for 266a.325 Consequently, the BNBP building block served as an effective electron-acceptor in various D–A type copolymers, 267, which have been developed and successfully used for different applications such as high-performance organic solar cells,180,325–333 organic indoor photovoltaics,334 OFETs,335 and organic thermoelectrics.336,337 Significantly, for all-polymer solar cells incorporating this building block, PCEs of up to 10.1% were reached.332 All-polymer indoor photovoltaic devices showed a PCE as high as 27.4% under fluorescent lamp illumination at 2000 lx.334

114

Organoboron and Related Group 13 Polymers

Scheme 63 Synthesis of BNBP monomers 266a,b and D–A copolymers 267 of this building block.

Pammer and co-workers investigated postfunctionalization by the hydroboration reaction as an alternative approach to laddertype or ladderized p-conjugated polymers with intramolecular B N coordination.338,339 In collaboration with von Hauff, they recently reported the synthesis of two types of such polymers, 269a,b and 271a-c (Scheme 64).339 First, they prepared poly(biphenylene −pyrazinylene) 268 and the head-to-tail regioregular polypyridine 270, both of them equipped with 1-alkenyl side chains. These polymers were then subjected to hydroboration with different hydroboranes (9H-BBN, (C6F5)2BH, and Cl2BH) to give 269a,b and 271a-c, respectively. Studies thereof reveled that the incorporation of the boron atom results in a significant decrease of the LUMO energy levels (down to −4.35 eV for 269b and to −4.15 eV for 271c). Investigation of the semiconducting properties of 269a,b revealed that they are capable of ambipolar charge transport with hole and electron mobilities in order of 2  10−5 cm2 V−1 s−1. Employing the polymers as acceptors in all-polymer solar cells yielded functioning devices, though with relatively low efficiency, which was assumed to be due to the nonoptimized morphology and blend layer thickness.

Scheme 64 Synthesis of ladderized polymer 269a,b and ladder polymers 271a-c with internal B hydroboration reactions.

N coordination by post-polymerization functionalization via

14.04.3 Boron in polymer side chains: Organic–inorganic hybrid polymers Various organoboron moieties have also been installed in polymer side chains, and this has been done in various ways. Herein, two types of such polymers are discussed: (1) p-conjugated polymers with triorganoborane groups that are directly bound to the polymer backbone or, alternatively, attached to a p-system that is bound to the backbone, so that the boron center is crossconjugated with the conjugation path along the main chain. The discussion in this section is restricted to triorganoborane groups, hence, polymers with tricoordinate boron bound to heteroatoms are not included here. (2) Polymers with tetracoordinate boron chelate complexes (such as the ones discussed in the previous section) in polymer side chains. Here, also examples are included that feature a nonconjugated polymer backbone. Such polymers, for instance, polyolefins,15,16 can be obtained by various sophisticated methods that lead to high-molecular-weight materials, including controlled/living radical polymerization protocols,22,23,340 which additionally provide access to block copolymers and other complex architectures.

14.04.3.1 Triorganoborane groups in side chains of p-conjugated polymers Linking of triorganoborane groups to conjugated polymers in lateral positions has been realized either by polymerization of boron-containing substrates via by polymer-modification reactions.341 Yamaguchi and co-workers used the former approach for the

Organoboron and Related Group 13 Polymers

115

synthesis of boryl-substituted poly(arylene ethynylenes) (PAEs) 272a-e by Sonogashira–Hagihara coupling (Scheme 65).342 They introduced and newly designed the boryl bis(4-hexyl-2,6-dimethylphenyl)boryl group, B(HDMP)2, wherein the HDMP substituent assumes the function of kinetically protecting the boron center—comparable with mesityl—while imparting good solubility through the n-hexyl group in the para-position. Polymers 272a-e proved to be highly emissive in solution and in the solid state.

Scheme 65 Synthesis of boryl-substituted poly(arylene ethynylenes) 272a-e and boryl-substituted poly(carbazoles) 274 and 275.

Lambert and co-workers investigated the combination of triarylborane with carbazole moieties with a view to develop ambipolar materials. Their studies involved the electrochemical polymerization of tris(carbazolyl)-substituted triarylboranes343 and the polymerization of boryl-substituted carbazoles 273a,b by Yamamoto coupling (Scheme 65).344 The latter was performed with both the 2,7-chlorinated monomer 273a and its 3,6-substituted isomer 273b to give 2,7- and 3,6-main-chain linked polycarbazoles 274 and 275, respectively. By comparison with a nonborylated polymer analog, they found that boryl-substitution had little influence on the photophysical properties of the 2,7-linked polymer, whereas it had a pronounced impact in the case of the 3,6-linked derivative, as it gave rise to a low-lying intramolecular charge transfer (CT) state, resulting in enhanced fluorescence quantum yield. In the 2,7-isomer, effective conjugation along the polymer backbone results in a polymer state that is lower in energy than an intramolecular CT state involving the boron and nitrogen centers. Jäkle and co-workers demonstrated the introduction of boryl groups in lateral positions of polythiophenes by postpolymerization functionalization (Scheme 66).345,346 In the first step, they synthesized the trimethylsilyl-substituted polymer 276 by Stille-type polycondensation. Then, BBr2 groups were introduced by quantitative Si/B exchange with BBr3, followed by nucleophilic displacement of the bromides on boron with MesCu to give 277a.345 The CV trace of 277a displayed two quasi-reversible reduction events. With respect to the precursor polymer 276, the reduction potential of 277a is significantly lower, due to the electron-withdrawing character of the BMes2 groups. Overall, these studies revealed strong electronic coupling between boron and the conjugated main chain, though considerable twisting in the polythiophene backbone was also detected. The Jäkle group also introduced redox-active ferrocenyl groups at the boron centers of such polymers (277b)346 and synthesized a series of laterally boryl-substituted p-conjugated copolymers 279a-c via the post-modification approach.347

Scheme 66 Synthesis of laterally boryl-substituted polythiophenes 277a,b related copolymers 279a-c.

With a view to diminish backbone twisting in such polythiophene-based polymers, Jäkle et al. subsequently targeted a series of polymers in which the boryl groups are further remote from the main chain. They explored alkenyl and alkynyl groups as spacers, which should still allow for conjugation between the boron center and the polythiophene backbone (Scheme 67). First they synthesized poly(3-alkynylthiophene) 280 via Ni catalyzed Grignard metathesis (GRIM) polymerization (Kumada coupling), though the polymer was formed with no significant degree of regioregularity.348 Partial hydroboration of 280 with Mes2BH

Scheme 67 Synthesis of polythiophenes conjugated with triorganoborane groups in side chains.

Organoboron and Related Group 13 Polymers

117

afforded random copolymer 281 featuring both a- and b-borylated pendent groups. The hydroboration was accompanied with a color change from almost black to bright red. Polymer modification by hydroboration of pendent ethynyl moieties was also used by Bazan and co-workers to introduce Mes2B groups at a conjugated fluorenylenephenylene copolymer, 282 (Scheme 67); here the borane is not in conjugation with the main chain.349 This polymer was used for the development of p/n junctions via fluoride anion complexation. Jäkle et al. presented an alternative approach to access polythiophenes with alkenylborane side groups, as a follow-up study to their aforementioned work, wherein they used the reverse order of reaction steps: hydroboration of the 3-alkynylthiophene starting material 283 and subsequent polymerization of the resulting vinylborane monomers 284a,b (Scheme 67).350 The hydroboration reaction gave a mixture of regioisomers 284a and 284b, which were separated by reverse phase column chromatography and were then subjected to GRIM polymerization to give 285a,b, respectively. The photophysical properties of both products differed significantly, thus demonstrating the effect of the position of the boryl group on the extent of cross-conjugation with the polythiophene main chain. Jäkle et al. also explored the polymerization of 3-alkynylphenylborane-functionalized mono- and bithiophene building blocks (Scheme 67).351 Due to a lack of a vinyl linker between the main chain and the borane side groups, which is present in 281 and 285a,b, backbone twisting in these polythiophene derivatives should be further reduced. The authors succeeded in the synthesis of AB type monomer 286 (and the corresponding AB-monomer to 288), which exhibits both functional groups, Br and SnMe3, required for Stille-type polycondensation. This led to regioregular polymers 287 and 288 of modest molecular weights. Both of them showed low-energy absorptions and emissions (labs,max ¼ 520 nm, lem,max ¼ 620 nm for 287), significantly bathochromic shifted compared to those of rr-P3HT (regioregular poly(3-hexylthiophene)). Experimental and computational studies revealed that these processes are assigned as p–p transitions in the main chain. A higher-energy absorption (at about 340 nm) was assigned to a charge transfer to the borane groups.

14.04.3.2 Tetracoordinate boron chelate complexes in polymer side chains The first examples of organoboron quinolate complex-containing polymers, reported by Jäkle and co-workers, carried this chromophore in side chains. These materials were initially synthesized via a polymer modification approach. Jäkle et al. developed a method that allows to obtain different side chain organoborane-functionalized polymers from a single polymeric precursor, namely poly(4-trimethylsilylstyrene) (289)352 (Scheme 68). This is transformed into the dibromoboryl species 290 via a facile Si/B

Scheme 68 Synthesis of polymers featuring organoboron quinolato groups in side chains (CTA ¼ chain transfer agent, small molecule or macro-CTA).

118

Organoboron and Related Group 13 Polymers

exchange reaction. The latter has also been used to access polystyrenes functionalized with different tricoordinate organoborane groups.150,352–356 The initial route towards the synthesis of boron quinolate complex polymers involved the formation of di (2-thienyl)boryl-substituted polystyrene from 290 and subsequent cleavage of one thienyl–boron bond by the selective reaction with 8-hydroxyquinoline.357 Later, they found a modified route to be more convenient that involves stepwise substitution of the bromine atoms, first by 4-tert-butylphenyl to give 291 and then by 8-hydroxyquinoline derivatives affording 292a-f.358 This allowed to tune the material’s properties at a late stage through variation of substituents at the latter reagent. A complementary approach was followed by Weck and co-worker.359 They chose to synthesize a polystyrene-based copolymer, 293, via free radical polymerization that features the 8-hydroxyquinoline ligand as a side group (Scheme 68). In this case, the boron moiety (as BPh2) was introduced at the post-polymerization stage to yield 294. An aluminum complex (an Alq3 derivative) was obtained via this route as well. In 2010, Jäkle and co-worker introduced a new route to organoboron quinolate polymers and various block copolymers 296 via reversible addition −fragmentation chain transfer (RAFT) polymerization of vinyl-substituted organoboron quinolate complexes 295 (Scheme 68).360–362 Block copolymers of this type having pyridine functionalities served as building blocks for assembled nanostructures.362 The Jäkle group, furthermore, prepared amphiphilic star polymers with a boron quinolate moiety as a luminescent core.363 These species underwent self-assembly in water with formation of super-aggregates that exhibit strong green fluorescence. BODIPY chromophores have been incorporated into polymers in various ways.236,237 For example, Burgess and co-workers investigated conjugated copolymers comprised of polyfluorene doped with BODIPY-based fluors that are cross-conjugated with the main chain (i.e., compounds 297 and 298, Scheme 69, among others).270 The idea was that light energy should be harvested by the strongly absorbing major component, fluorene, and upon internal energy transfer, emission should come primarily from the BODIPY moieties at much higher wavelengths. This concept proved feasible; the most intense luminescence was observed for copolymers featuring a fluorene-to-BODIPY ratio of approximately 1:4. Upon excitation at 358 nm, polymer 297 showed emission at lem,max ¼ 517 nm, while 298 emitted at about lem,max ¼ 677 nm. The larger Stokes shift was attributed to enhanced planarity and partial p electron delocalization into the pendent BODIPY moiety in 298. These polymers were applied for bioimaging by casting them into nanoparticles, which were uptaken by clone 9 rat liver cells. Further related materials comprising BODIPY dyes have proven particularly useful for biomedical imaging applications.273,364

Scheme 69 Polyfluorene copolymers with cross-conjugated pendent BODIPY moieties.

Manners and co-workers incorporated BODIPY units into block copolymers containing the crystallizable poly(ferrocenylsilane) (PFS) block, 299a-c,365 which allows for crystallization-driven living self-assembly (CDSA) processes366–368 (Scheme 70). The three dyes have distinct emission colors, red (a), green (b), and blue (c), and are situated in the corona of the via CDSA self-assembled nanostructures. They synthesized multicompartment micelles in which the emission of each segment can be controlled to produce colors throughout the visible spectrum. The unique morphologies produced were visualized and the process leading to them was tracked using confocal fluorescence microscopy.

Scheme 70 Polyferrocenylsilane–polysiloxane block copolymers comprising BODIPY dyes in side chains.

Organoboron and Related Group 13 Polymers

119

Polymers comprising boron difluoride formazanate groups in side chains, 301, were reported by Gilroy and co-workers (Scheme 71).369 They used the controlled ring-opening metathesis polymerization (ROMP) of norbornene-based BF2 formazanate monomer 300. Their studies revealed that the unique properties of the monomers were largely retained in the polymers. They showed absorption and emission maxima at labs,max ¼ 518 nm and lem,max ¼ 645 nm, respectively, with large Stokes shifts (3800 cm−1). Furthermore, they showed the ability to act as electron reservoirs to form poly(radical anions). In a subsequent study, Gilroy and co-workers reported polymers and copolymers bearing an asymmetrically substituted boron difluoride 3-cyanoformazanate complex.370

Scheme 71 Synthesis of polymers featuring boron difluoride formazanate groups in side chains.

Various further boron chelate complexes have been incorporated in polymer side chains as well. For example, Wang and co-workers developed photochromic random copolymers 302 that feature the photochromic N,C-chelate complex B(ppy) Mes2 (ppy ¼ 2-phenylpyridyl) (Scheme 72).371 They had previously discovered that this unit undergoes a highly efficient and thermally reversible photoisomerization upon UV irradiation, accompanied with a color change from colorless to dark blue. The color of the dark isomer was furthermore found to be highly tunable by variation of the substituents in the complex.179,372–377 In polymers 302, this group is installed in side chains, separated from the polymer backbone through a long alkyl linker. This was meant to minimize p-conjugation between the boron units as this could diminish the photoisomerization quantum efficiency of the boron unit.378–380 Indeed, polymers 302 demonstrated thermally reversible photochromism, with a color switch from colorless to deep blue, and fluorescence from bright sky blue to deep blue. Their photoisomerization quantum efficiencies were be effectively tuned by controlling the monomer ratio in the ATRP process, which was used for their synthesis. The utility of these materials as switchable/erasable ink on glass or paper substrate and for creating switchable/erasable patterns as neat polymer films has been impressively demonstrated.

Scheme 72 Reversible photoisomerization of B(ppy)Mes2-containing polymers 302.

14.04.4 Inorganic polymers comprising (BN)n or (BP)n chains Polymers comprising linear chains of concatenated BN or BP units are classified as “inorganic (main group) polymers” as they contain no carbon atoms in their backbone. Poly(aminoboranes) and poly(phosphinoboranes) feature the respective main group

120

Organoboron and Related Group 13 Polymers

elements in the tetracoordinate state and exclusively connected to each other via single bonds along the polymer chain. Such materials are most conveniently synthesized by dehydrocoupling381–383 of amine– and phosphine–borane adducts, respectively.384–390 Amine–boranes, particularly the parent ammonia–borane, have been discussed as potential hydrogen storage materials for on-board fuel cell applications, which has stimulated considerable research activity in this field in recent times.391–395 We begin with the discussion of poly(phosphinoboranes) because the first assured synthesis of these species has been achieved earlier than that of their BN analogs, and the strategies for their formation could be in part adopted or they formed the basis for approaches to synthesize the latter. The first convincingly characterized poly(phosphinoboranes) 304 were reported by Manners and co-workers in 1999 (Scheme 73A).396,397 While the thermolysis of phosphine–borane adducts at higher temperatures (150–200  C) results in the formation of cyclic phosphinoborane oligomers besides low-molecular-weight polymers, Manners et al. discovered that the dehydrocoupling of primary phosphine–boranes 303 (R ¼ Ph or iBu) can be effectively catalyzed in the presence of 0.3 mol% of the Rh(I) catalyst [{Rh(m-Cl)(1,5-cod)}2] (1,5-cod ¼ 1,5-cyclooctadiene) at 90–130  C in the melt. Applied to the 303 derivative having R ¼ Ph, this afforded polymer 304a in high molecular weight (Mw ¼ 31 kDa), which is formally a polystyrene analog. Polymers with para-substituted aryl groups on phosphorus have also been prepared.398 When activated substrates featuring fluorinated aryl groups were used, the process temperature could be lowered to about 60  C.399 The alkylphosphine-borane adduct iBuPH2 ∙BH3 required more forcing conditions (120  C, in the melt, 13 h) with the same catalyst to give 304 with R ¼ iBu in an average molecular weight of Mw ¼ 10–20 kDa.397,398 Hey-Hawkins and co-workers demonstrated the use of ferrocenylphosphine–borane adducts, FcPH2 ∙ BH3 and FcCH2PH2 ∙BH3 (Fc ¼ ferrocenyl), in such solvent-free dehydropolymerization reactions, the latter of which constitutes a primary alkylphosphine–borane.400 A planar-chiral derivative was also incorporated. Dossi and co-workers added further poly(alkylphosphinoboranes) via the same route.401 The molecular weights of these species were always lower (Mn < 10 kDa) than those of the poly(arylphosphinoboranes). Model compound studies by Weller and Manners et al. provided valuable insight into the Rh-catalyzed process, suggesting the intermediacy of Rh-coordinated oligo(alkylphosphinoborane) species.402–404

Scheme 73 Synthesis of poly(phosphinoboranes).

In 2015, the Manners group reported an iron-catalyzed dehydropolymerization route to high-molecular-weight poly(phosphinoboranes) 304.405,406 For example, polymer 304a (R ¼ Ph) was prepared via dehydrocoupling of 303a in toluene at 100  C in the presence of 1 mol% of [CpFe(CO)2OTf] with Mn ¼ 59 kDa and PDI ¼ 1.6.405 This method allowed for molecular weight control, that is, higher catalyst loading resulted in a decrease of the molecular weight obtained. The studies further suggested that the reaction proceeds through a chain-growth coordination–insertion mechanism. Other transition metal-based catalysts have also

Organoboron and Related Group 13 Polymers

121

been applied to the synthesis of poly(arylphosphinoboranes). For instance, Weller and Macgregor and colleagues employed a cationic Rh complex as a precatalyst,407 Webster and co-workers reported the use of a three-coordinate iron(II) b-diketiminate precatalyst,408 and Braunschweig and Radius and et al. demonstrated that bis(phosphinite) Ir pincer complexes are effective catalysts for the dehydropolymerization of arylphosphine–boranes as well.409 Very recently, Manners and co-workers reported that [CpFe(CO)2OTf] can also be applied as a precatalyst for the dehydropolymerization of primary alkylphosphine–boranes.410 In a collaborative effort, the groups of Manners and Scheer developed a conceptionally different, metal-free approach towards poly(phosphinoboranes) (Scheme 73B).411,412 Thermolysis of the NMe3–phosphinoborane adducts 305a-c, which can be viewed as base-stabilized or “masked” phosphinoboranes, brought about dissociation of the amine and subsequent spontaneous oligo- or polymerization of the transient phosphinoborane to give 306a-c. More precisely, this afforded oligomers in the case of 305a,b, while the reaction of the alkylphosphinoborane adduct 305c in toluene at 22  C resulted in the formation of a polymer of considerable molecular weight (Mn ¼ 27.8–35.0 kDa). Recently, Manners et al. demonstrated that poly(phosphinoborane) 304a can be transformed into partially P-di(organosubstituted) poly(phosphinoboranes) via post-polymerization modification (Scheme 73C).413 They developed a photolytic, radical-mediated hydrophosphination reaction that proceeds under UV light in the presence of the photoinitiator 2,2-dimethoxy-2-phenylacetophenone (DMPAP) and (2,2,6,6-tetramethylpiperidin-1-yl)oxyl (TEMPO). This method was also used for the synthesis of water-soluble bottlebrush copolymers with PEG side chains. The Manners group, furthermore, demonstrated that phosphine −boranes can also be effectively dehydrogenated using cyclic(alkyl)(amino)carbenes as hydrogen acceptors to yield poly(phosphinoboranes) (Scheme 73D).414 Significantly, this method was also applicable to secondary phosphine–borane adducts, thus affording P-disubstituted polymers 309b,c directly. Unlike most polyolefins, which are highly flammable, poly(phosphinoboranes) display flame retardant properties.406,415 Pyrolysis of these inorganic polymers afforded high ceramic yields,398,400 thus suggesting prospective applications for them as macromolecular precursors to ceramic boron phosphide-based solid-state materials. Manners et al., furthermore, have demonstrated their use as lithographic resists.399,405,406 Poly(aminoboranes) (311) are isoelectronic and the direct isosteres of polyolefins (312) (Scheme 74). They are of interest within the context of hydrogen storage applications, as piezoelectric materials416,417 and as precursors to boron-based ceramic solid state materials.382,395,418–421 Manners and co-workers demonstrated that the same Rh(I) catalysts that brought about dehydropolymerization of phosphine–boranes were capable of catalyzing the dehydrocoupling of amine–borane adducts as well, however, this resulted in the formation of small molecular cyclic oligomers in this case; linear polymers were not obtained.422 When they applied Brookhart’s Ir(III) pincer complex [IrH2(POCOP)] (POCOP ¼ [m3-1,3-(OPtBu2)2C6H3]) to the dehydrocoupling of primary alkylamine–boranes (R ¼ Me or nBu) and ammonia–borane (R ¼ H) 310, high-molecular-weight poly(aminoboranes) 311 were produced (e.g., R ¼ Me: Mw ¼ 160 kDa, PDI ¼ 2.9) (Scheme 74A).421,423 The product obtained from the catalytic dehydrocoupling of the parent ammonia–borane turned out to be insoluble in common organic solvents. It was, therefore, characterized by solid-state analysis techniques, including 11B magic-angle spinning (MAS) NMR spectroscopy.423 This revealed that it is an essentially linear material and it allowed to estimate its degree of polymerization as DPn  20 via end group analysis. Applying the same catalytic conditions to mixtures of amine–boranes MeNH2 ∙ BH3/NH3 ∙ BH3 and nBuNH2 ∙ BH3/NH3 ∙ BH3 yielded the respective random copolymers (R ¼ Me/H and R ¼ nBu/H), which proved to be well soluble and were shown to be of high molecular weight. These species were fully characterized using solution-based techniques. The incorporation of monomeric aminoborane (NH2BH2) groups in the polymer chains was impressively demonstrated via electrospray ionization mass spectrometry (ESI-MS).423 Subsequently, the groups of Manners, Weller, Schneider, Beweries, Grützmacher, and Webster and others have

Scheme 74 Synthesis of poly(aminoboranes) 311, isosteres of polyolefins (312).

122

Organoboron and Related Group 13 Polymers

reported the development of various further catalyst systems for the dehydropolymerization of amine–boranes, most of them are based on precious mid to late transition metals (Ir, Ru, Rh, Pd), but also some iron-, titanium-, and zirconium-based catalysts have been shown to be applicable.408,423–444 The mechanisms involved in these polymerizations proved to be highly dependent on the specific (pre)catalyst system; the co-ligand(s) are often not innocent. For example, in the case of the Ir and Rh complexes [IrH2(POCOP)], [Ir(PCy3)2(H)2]+, [Rh(Ph2P(CH2)4PPh2)]+, and [Rh(k2-P,P-xantphos){Z2-H2B(CH2CH2tBu)∙NMe3}]+ detailed mechanistic studies by Weller, Manners, Lloyd-Jones, Helten, Macgregor, and colleagues provided evidence for a chain-growth coordination-insertion mechanism, wherein the metal is involved in both the dehydrogenation step and the BN coupling step.423,425,426,428,431,436 On the other hand, for their Ti-catalyzed system, Manners et al. showed that it proceeds via a stepgrowth mechanism.437 Manners also reported that skeletal nickel can also be used as a heterogeneous catalyst system for the dehydropolymerization of methylamine–borane (310a).445 Overall, the substrate scope of all amine–borane dehydropolymerizations reported so far seems to be limited to primary alkylamine–boranes, in which the alkyl substituent is not branched in the a-position. Attempts to prepare poly(N-phenylaminoborane), which would be a polystyrene analog, yielded only cyclic oligomers rather than polymers.134 However, Manners et al. recently succeeded in the synthesis of poly[(B-aryl)aminoboranes], which are BN-polystyrene analogs as well.441 Attempts to obtain poly[(B-methyl)aminoborane] derivatives using skeletal Ni evidenced the formation of the desired polymers, however, they were prone to spontaneous depolymerization under ambient conditions.446 Recently, Manners and co-workers reported a series of poly(aminoborane) derivatives with aryl groups installed at N-bound alkyl chains.442 Beweries and co-workers recently also incorporated silyl groups in this way.444 A few metal-free routes to linear poly- or oligo(aminoboranes) have also been reported.447–450 Manners and Wass et al. showed that methylamine–borane 310a can be transformed into the amine–boronium salt 313 by protonation with the strong Bronsted acid [H(OEt2)2][B(C6F5)4] with concomitant release of H2 (Scheme 74B).447 Subsequent deprotonation with 2,6-di-t-butylpyridine (DTBP) afforded poly(N-methylaminoborane) 311a, though of much lower molecular weight ( 0.67) and vice versa for a syndiotactic polymer (Pr > 0.67). Zeigler78 and West79 reported that the 29Si NMR spectra of asymmetrically substituted dialkyl polysilanes arising from Wurtz polymerization such as poly[SiMe(n-C6H13)] show broad multiplets consistent with tacticity, in contrast to the sharp singlet of poly [Si(n-C6H13)2]. Zeigler also reported successfully fitting the 29Si multiplet of poly[SiMe(n-C6H13)] to Bernoullian propagation statistics with a probability of finding a meso dyad (Pm) of 0.5, supporting an atactic polymer.78 A major challenge in understanding polysilane tacticity is the lack of methods for preparing stereoisomerically pure small molecule model compounds, which complicates assignment of NMR peaks to distinct triads. This has been particularly noted in the case of poly(SiMePh), where several conflicting reports can be found in the literature. In 1988, West et al. reported that the 29Si NMR

Fig. 4 Polysilane tacticity: representations of stereochemical dyads and triads in poly(SiMePh).

Organosilicon and Related Group 14 Polymers

139

spectrum of poly(SiMePh) arising from Wurtz polymerization of Cl2SiMePh showed three distinct peaks with relative intensity of 3:3:4, inconsistent with the 1:2:1 ratio expected for an atactic polymer.57 In this first report, West did not assign the three peaks to different triads. In 1991, based on cyclo-(SiMePh)6 model compounds, West et al. assigned the three lines of the poly(SiMePh) spectrum (going from highest to lowest frequency) to mm, rr, and mr triads (mm ¼ 0.30; mr ¼ 0.40; rr ¼ 0.30).80 The indicated ratios were inconsistent with Bernouillian statistics, which West attributed to long-range stereoinduction. In 1995, Matyjaszewski reported that poly(SiMePh) arising from anionic ROP of all trans-(SiMePh)4 (initiator: organocuprate) had a two line 29Si NMR spectrum in a 3:1 ratio (mr ¼ 0.75, mm ¼ 0.25), consistent with an isotactic polysilane. In 1996, Jones et al. used peak deconvolution software to reanalyze prior results and reported revised peak ratios for poly(SiMePh) arising from Wurtz polymerization.81 Jones found mm ¼ 0.48, mr ¼ 0.37, rr ¼ 0.15 which is consistent with Bernoullian statistics and suggest a moderately isotactic structure for poly(SiMePh). Group 4 metallocene catalysts, which are remarkably effective at iso- or syndioselective propylene polymerization,82 are also effective for silane polymerization, suggesting the tantalizing possibility of catalyst-controlled stereoselective silane dehydropolymerization. Waymouth83 and Tanaka84 separately investigated this possibility and reported preliminary evidence in favor of a syndiotactic poly(SiPhH) microstructure. However, subsequent studies by Harrod68 and Corey61 using peak deconvolution was more consistent with assignment of poly(SiHPh) to an atactic structure. This overview clearly highlights that control of polysilane tacticity remains a significant outstanding challenge.

14.05.2.3 Geometry and conformation Tetracoordinate molecular silanes, germanes, and stannanes are typically tetrahedral, like saturated alkanes. Backbone atoms in group 14 organometallic polymer are also generally tetrahedral. Pentacoordinate (trigonal bipyramidal) and hexacoordinate (octahedral) complexes are known to form upon coordination of nucleophiles to Si, Ge, and Sn Lewis acids.85 Lewis acid-base interactions have been explored in a polymer context, but remain somewhat unusual. Fujiki reported that C–F !Si side chain to main chain interactions controlled polysilane global conformation.13,76,86 Foucher has extensively investigated hypercoordinate polystannanes based on intramolecular O!Si coordination (see Section 14.05.5.2).12 An expanded nomenclature is required to describe poly(ER2) conformation, as conformation has critical effects on polymer properties. While n-butane’s conformational profile with energetic minima at gauche (torsional angle, o ¼  60 ) and anti (o ¼ 180 ) positions is taught in introductory organic chemistry textbooks, more conformations have been experimentally and computationally found for silanes, germanes, and stannanes. Michl and West suggested the following terms for linear EnR2n+2 chains (Fig. 5): cisoid (o   40 ), gauche (o   60 ), ortho (o   90), deviant (o   150), transoid (o   165 ), and anti (o  180 ).87 Global conformation in solution and in the solid-state depends on the identity of the side chains. For example, on the basis of X-ray diffraction studies, poly[Si(n-C4H9)]2 and poly[Si(n-C5H11)2] were found to adopt 7/3 helical conformations in the solid-state, while poly[Si(n-C6H13)2] has an all-anti structure.88,89 Optically active dialkylpolysilanes are helical in solution and the solid-state.72

Fig. 5 Favored dihedral angles and their proposed labels. Republished from Ref. Michl, J.; West, R. Conformations of Linear Chains. Systematics and Suggestions for Nomenclature. Acc. Chem. Res. 2000, 33(12), 821–823 (Pub. Date 12/01/2000), Copyright 2000 American Chemical Society, with permission.

140

Organosilicon and Related Group 14 Polymers

14.05.3 Polysilanes The most common synthetic methods for polysilanes were introduced in the prior section. Recent advances will be discussed, with a focus on mechanism-driven insights, novel monomers, and novel reaction methodology.

14.05.3.1 Wurtz polymerization An enduring challenge of the Wurtz polymerization is the low functional group compatibility arising from the strongly reducing conditions. Functional groups tolerated as side-chain substituents under standard Wurtz conditions include dialkyl groups, mixed alkyl/aryl substitution, and organic groups with heteroatoms not directly bonded to Si. Among the groups that are not tolerated include vinyl groups, halides, and substituents with heteroatoms directly bonded to Si (alkoxy, amido, etc.). Jones has investigated the mechanism of the Wurtz polymerization in hopes that mechanistic insights would afford rational approaches to increased functional group compatibility and molecular weight control. The complexity of the Wurtz reaction has been discussed in COMC III (2007)3 and in reviews by Jones and Holder35 and Koe and Fujiki.23 We summarize here those mechanistic insights as well as post-2005 advances.

14.05.3.1.1

Mechanism

The typical reaction conditions for the Wurtz polymerization of dichlorodialkylsilanes employ a slight excess of sodium in a high-boiling solvent such as toluene under reflux. The products compose three fractions: cyclic oligomers of the general structure cyclo-(SiR2)5–6 (fraction I), intermediate molecular weight polymers (DP  40, fraction II), and high molecular weight polymers (DP  1000 , fraction III).90,91 Initiation and propagation involves the successive intermediacy of radical anions, radicals, and anions formed via single electron transfer (Fig. 6). Termination can occur via multiple mechanisms, including dissociation from sodium or end-biting to form the cyclosilanes of fraction I. Hydrogen-atom abstraction from solvent or radical-radical recombination can also contribute to termination.67 The high molecular weight fraction III is observed even at low conversion, indicating a chain-growth mechanism,52 as high molecular weight polymer is only formed at high conversion in a step-growth polymerization. A chain-growth mechanism must occur at the sodium surface. Dissociation of intermediate length chains from the sodium surface is thought to account for the intermediate molecular weight fraction II, while chains that remain anchored to the surface grow to high molecular weight fraction III. Jones et al. also suggested the possibility that two non-intersecting mechanisms may occur simultaneously, e.g. a surface-associated chain-growth polymerization mechanism and a solution-phase step-growth polycondensation.90 The underlying mechanisms behind the trimodal molecular weight distribution, especially early termination contributing to the formation of fraction II, has been a matter of significant study.52,92,93 In 1998, Jones proposed a role for conformational defects.90 Poly(SiMePh) in solution contains both M- and P-helices that are in dynamic equilibrium, leading to a net optically inactive macromolecule (Fig. 7A). Helical reversal sites, conformational defects that induce a switch in helix direction, are “weak spots” especially prone to end-biting via nucleophilic attack. The probability of a conformational defect reaching the chain end is highest for polymers with a degree of polymerization ca. 35–40, contributing to the formation of fraction II.91

Fig. 6 Proposed intermediates in Wurtz polymerization.

Organosilicon and Related Group 14 Polymers

141

Fig. 7 (A) Schematic depiction of polysilane helicity. (B) Molecular weight distributions observed for Cl2SiMePh Wurtz polymerization in optically active and inactive limonene at (B) 70  C and (C) 90  C. Modified from Ref. Holder, S. J.; Achilleos, M.; Jones, R. G. Increasing Molecular Weight Parameters of a Helical Polymer through Polymerization in a Chiral Solvent. J. Am. Chem. Soc. 2006, 128(38), 12418–12419 (Pub. Date 09/01/2006), Copyright 2006 American Chemical Society, with permission.

Based on this hypothesis, Holder et al. made the remarkable discovery that chiral solvents reduced the concentration of helical reversal sites, allowing increased yields of high molecular weight fraction III.94 The authors investigated the sodium-mediated Wurtz polymerization of Cl2SiMePh in (R)-limonene, (S)-limonene, and ()-limonene. At 70  C, the relative proportions of fractions I, II, and III did not change significantly with solvent enantiopurity (Fig. 7B). At 90  C, when either (R)- or (S)-limonene were used, the proportion of fraction III was higher than fraction II, whereas ()-limonene yielded a larger proportion of fraction II (Fig. 7C). In this second case, the use of optically active solvent also increased the molecular weight of fraction III (Mw ¼ 80,000 g mol−1, Mw/Mn ¼ 50.2) compared to racemic solvent (Mw ¼ 39,200 g mol−1, Mw/Mn ¼ 22.9). In an achiral solvent, helical reversals are required to obtain a net optically inactive polymer and these regions are particularly prone to termination. In a chiral solvent, a single helical screw sense can dominate and reduce the number of helical reversal sites, thereby reducing termination events. In a second mechanism-driven insight, Holder reported the utility of 1-chloro-4-methyloctaphenyltetrasilane (Me(SiPh2)4Cl) as a Wurtz polymerization initiator.95 Electron affinity increases with increasing silane length in oligosilanes of the general structure SinMe2n+2 (n ¼ 1–4).96 Based on this trend, Holder et al. hypothesized that Me(SiPh2)4Cl is reduced more rapidly than Cl2SiMePh, allowing it to serve as an initiator (Fig. 8A). Me(SiPh2)4Cl-initiated MePhSiCl2 polymerization in a toluene suspension of Na at 65  C provided poly(SiMePh) in increased yield compared to reactions without initiator (35% vs. 19%). Additionally, the proportion of the fraction III relative to fractions I and II was higher (Fig. 8B) and Mn was 5.5 times higher (Mn ¼ 379,600 g mol−1, Mw/Mn ¼ 3.20) than without initiator (Mn ¼ 84,600 g mol−1, Mw/Mn ¼ 2.83). Holder et al. suggested that the perphenyl end group suppressed end-biting at early-stage polymerization (Fig. 8A). The 29Si NMR spectrum showed resonances consistent with Me(SiPh2)4 end groups (Fig. 8C).95 In 2008, Fujiki and coworkers reported a mechanistic study of the binary copolymerization of i-butyl-n-decyldichlorosilane (Cl2Si(i-Bu)(n-decyl), M) and methyl-3,3,3-trifluoropropyldichlorosilane (Cl2SiMe(CH2CH2CF3), MF) (Fig. 9A), which yielded a multiblock copolymer with a slight excess of the alkyl monomer M (ca. 0.55:0.45).97 The authors proposed a mechanism that is conceptually similar to a composition drift scenario in a binary copolymerization of vinyl monomers:47 because reaction rate depends on both concentration and rate constant, in a scenario where one monomer is more reactive than the other, but both are present in the same concentration, the more reactive monomer will polymerize first until its concentration is sufficiently depleted that the less reactive monomer begins to polymerize. In this scenario, the fluorinated monomer MF reacts first with sodium (Fig. 9B, Stage I). Some proportion of oligomers (Mn  3000) detach from the sodium surface to solution and are capped with reactive end groups. When MF is about 50% consumed, the less reactive alkyl monomer M reacts with Na, forming both poly(M) and diblock poly(MF-b-M) chains (Fig. 9B, Stage II). As the concentration of monomer M is depleted, the more reactive MF again preferentially reacts with the sodium surface, leading to multiblock polymers. Multiblock polymers can also be formed in solution or at the gel stage by a condensation mechanism (Fig. 9B, Stages II and III).

142

Organosilicon and Related Group 14 Polymers

Fig. 8 (A) Suppressed end-biting at early stage Cl2SiMePh Wurtz polymerization. (B) Molecular weight distribution of poly(SiMePh) with and without initiator. (C) Cropped 29Si NMR spectrum of poly(SiMePh) showing peak assignments consistent with Me(SiPh2)4 end group. Modified from Ref. Holder, S. J.; Achilleos, M. An Oligosilane Initiator for the Wurtz-Type Polymerisation of Dichloromethylphenylsilane. J. Organomet. Chem. 2008, 693(10), 1938–1944 (Pub. Date 05/01/2008), Copyright 2008 Elsevier, with permission.

14.05.3.1.2

Variations

In this section, we address variations of the standard Wurtz polymerization conditions. Milder alternatives to Na for chlorosilane polymerization appeal due to the potential for greater functional group compatibility. Ogawa reported the reductive polymerization of MePhSiCl2 with Sm/SmI2.42 Modest molecular weight samples were typical e.g. Mn ¼ 1450 g mol−1, Mw/Mn ¼ 2.60. In 2014, Ogawa et al. reported that in the presence of UV/vis light, higher degrees of polymerization could be obtained with quite narrow dispersities (up to Mn ¼ 2400 g mol−1, Mw/Mn ¼ 1.30). Additionally, Shen98 and Jiang99 separately studied reductive polymerization with an in situ prepared low valent Ti species (Fig. 10). This method was evaluated for the polymerization of monomers poorly compatible with sodium, such as dichloromethylvinylsilane (Cl2SiMe(CHCH2)) and dichlorophenylsilane (Cl2SiHPh). Jiang also reported the copolymerization of Cl2SiHPh with dichlorodialkylsilanes (R1R2SiCl2). However, neither manuscript rigorously investigated potential branching or cross-linking from the reactive vinyl or hydride side chains. We also note Sacarescu’s success at achieving SidH moieties in a polysilane via Wurtz copolymerization of Ph2SiCl2 and Cl2SiHPh.100 Sacarescu separately investigated the influence of microwave irradiation on the Na-mediated Wurtz polymerization of Cl2SiMePh.101 Advantages of this approach include (i) “instant heating” that rapidly disperses molten sodium in the liquid monomer, (ii) rapid monomer consumption (ca. 3 min), and (iii) suppression of pathways contributing to low-molecular weight polymers. One main poly(SiMePh) fraction was isolated in high molecular weight (Mn ¼ 41,200, Đ ¼ 1.21). This method showed better performance than an analogous reaction carried out under standard conditions (e.g. in boiling toluene), from which two main fractions A and B (A: Mw ¼ 39,256, Đ ¼ 1.52; B: Mw ¼ 12,346, Đ ¼ 1.38) were identified. Sacarescu et al. later reported Cl2SiPh2 microwave-assisted polymerization.102

14.05.3.1.3

Novel monomers

The most widely used monomers in Wurtz polymerization are commercially available dichlorodiorganosilanes (R1R2SiCl2). The use of more complex monomers has been addressed in COMC III (2007), including 1,3-dibromotrisilanes103 and silole derivatives.104 In this current edition, we add new examples of bi- and trifunctional monomers. The products of these reactions differ in architecture from standard linear polydiorganosilanes. Polysilynes, which have the general formula [RSi]n and are network polysilanes, were first reported by Bianconi and Weidman in 1988 via Na/K reduction of (n-C6H13)SiCl3.45 In 2012, Fujiki reported the synthesis of a polysilyne bearing both n-C4H9 and 3,3,3-trifluopropyl side chains.13 The fluorinated polysilynes were air-stable and solution processable, which was attributed to a LUMO-lowering effect of fluorination. In 2020, Gates and Rosenberg reported the synthesis of polyphenylsilyne by the thermolysis of poly(SiPhH) arising from dehydropolymerization of PhSiH3.105 The thermolysis route avoids the use of hazardous, pyrophoric Na/K.

Organosilicon and Related Group 14 Polymers

143

Fig. 9 (A) Reaction scheme for binary copolymerization. (B) Mechanism proposed for the formation of [(MF)n-co-(M)m], with MF ¼ methyl-3-trifluoropropylsilyl and M ¼ n-decyl-i-butylsilyl. Republished from Ref. Kawabe, T.; Naito, M.; Fujiki, M. Multiblock Polysilane Copolymers: One-Pot Wurtz Synthesis, Fluoride Anion-Induced Block-Selective Scission Experiments, and Spectroscopic Characterization. Macromolecules 2008, 41(6), 1952–1960 (Pub. Date 03/01/2008), Copyright 2008 American Chemical Society, with permission.

A trifunctionalized oligosilane monomer 1 was reported by Rieger et al. in 2015 (Fig. 11).106 Depending on exact reaction conditions, poly(1) samples between Mn ¼ 1700–2000 g mol−1 (Mw/Mn  2.0) were obtained. A 1H-29Si{1H} HMBC (heteronuclear multiple bond coherence) spectrum showed resonances between d ¼ − 60 and −85 that were consistent with MeSi(Si)3 units, as well as Me2Si and Me3Si sites. The branched polymer is miscible in organic solvents, in contrast to insoluble linear poly(SiMe2).66 The authors also demonstrated pyrolysis to SiC via intermediate formation of the polycarbosilane. In 2011, Chernyavskii et al. investigated the polymerization of 1,3- and 1,4-dichlorocyclohexasilanes 2 and 3 (synthesized from cyclo-(SiMe2)6) via hydrolysis to polysiloxanes and Wurtz polymerization to polysilanes (Fig. 12).107 Each monomer was advanced as an approximately 1:1 mixture of cis- and trans-diastereomers. The degree of polymerization was modest (ca. 10) for the Wurtz polymerization products poly(2) and poly(3). The low solubility of poly(3) was particularly noted. UV–vis spectroscopy indicated absorption maxima for both poly(2) and poly(3) at ca. 250 nm and 313 nm. The 313 nm absorption maximum is similar to poly [Si(n-C4H9)2] at room temperature (lmax ¼ 316 nm).9,108 The similarity of the low energy ca. 313 nm transition to a linear polysilane may suggest some ring-opening of the cyclohexasilane repeat units, as the absorbance spectrum of cyclo-(SiMe2)6109 is hypsochromically shifted relative to linear Me(SiMe2)6Me (lmax ¼ 230 vs. 260 nm).110

144

Organosilicon and Related Group 14 Polymers

Fig. 10 Dichlorosilane polymerization by low-valent Ti.

Fig. 11 Synthesis of a branched polymethylsilane.

Fig. 12 Wurtz polymerization of 1,3- and 1,4-dichlorodecamethylcyclosilane.

In 2020, Klausen et al. reported the Wurtz polymerization of a 70:30 trans:cis diastereomeric mixture of 1,3-dichlorocyclosilazane 4 (Fig. 13).111 The Wurtz polymerization of silanes bearing directly attached heteroatoms is unusual. Both Li and Na were investigated, and both yielded low molecular weight samples (Mn < 2000 g mol−1, Mw/Mn  2.0). Spectroscopic characterization (1H, 13C, and 29Si NMR) supported retention of the allyl side chain and indicated that samples arising from Li and Na-mediated polymerization had different backbone structures, although detailed structural assignment was not possible. The low degree of polymerization observed by both Klausen and Chernyavskii suggests that cyclosilanes do not access the fast sodium surface-mediated chain-growth polymerization (perhaps due to steric effects) and may only react via a slower solution-phase step-growth polycondensation. In the case of cyclosilazane polymerization, Klausen et al. reported that extended reaction times resulted in degradation to even lower molecular weight oligomers.

Organosilicon and Related Group 14 Polymers

145

Fig. 13 Diastereomeric 1,3-dichlorocyclosilazane monomers for Wurtz polymerization.

14.05.3.2 Dehydropolymerization Transition metal-catalyzed silane dehydropolymerization complements the scope of Wurtz polymerization. The typical monomer is a primary silane like PhSiH3 and the resulting polymers are functionalized with SidH bonds that are not typically tolerated in a Wurtz polymerization. PhSiH3 may be used neat or with a minimal amount of solvent and is generally more reactive than alkylsilanes.112 The typical degree of polymerization is modest (ca. 10–20), significantly lower than Wurtz polymerization. Like in Wurtz polymerization, small cyclosilanes are commonly observed in addition to linear polymers and are a thermodynamic sink. Secondary silanes (e.g. Ph2SiH2) typically stop reacting after dimerization and tertiary silanes (e.g. Ph3SiH) are entirely unreactive.63,64,113 Early dehydropolymerization catalysts were d0 group 4 metallocenes like Cp2MMe2114 (M ¼ Ti, Zr, Hf ), which remain widely in use.114 Another common precatalyst is Cp2ZrCl2/n-BuLi.112 The commonly accepted mechanism for silane polymerization by d0 group 4 metallocenes is a condensation polymerization that produces H2 and polysilane via sequential s-bond metathesis steps (Fig. 14).36,37,115 The use of an electron-deficient d0 transition metals precludes a catalytic cycle involving changes in the transition metal oxidation state. While a metal hydride is proposed as the active catalyst, the mechanism by which a precatalyst is converted to a metal hydride is complex.116 An alternative mechanism proposed by Harrod involves the intermediacy of Zr(III)-species capable of redox cycles or formation of silylene (Zr ¼ Si) intermediates. The likely low concentration of active catalyst and preponderance of Cp2Zr-containing byproducts is a major challenge to catalyst efficiency. A question of long-standing interest is why silane dehydropolymerization rarely gives high molecular weight polymers. As a step-growth polycondensation reaction, high molecular weight polysilanes are only be obtained at very high conversion (Fig. 15). Tilley suggested that hydrogen removal is the limiting factor, as residual hydrogen contributes to depolymerization.36 Tilley and others have shown that the elemental steps in the s-bond metathesis mechanism are reversible.115 As detailed below, several other research groups have found hydrogen removal to influence product distribution and reactivity trends.

Fig. 14 . s-Bond metathesis mechanism for silane dehydrocoupling polymerization.

Fig. 15 Step-growth polycondensation showing progression from monomer to dimer to oligomer and high polymer only at late stage polymerization. Modified from Ref. Marro, E. A.; Klausen, R. S. Conjugated Polymers Inspired by Crystalline Silicon. Chem. Mater. 2019, 31(7), 2202–2211 (Pub. Date March 8th 2019), Copyright 2019 American Chemical Society (link: https://pubs.acs.org/doi/10.1021/acs.chemmater.9b00131, additional permissions related to the material excerpted should be directed to the ACS), with permission.

146

Organosilicon and Related Group 14 Polymers

We will discuss advances in silane dehydrocoupling in order of position in the periodic table. With respect to early transition metal catalysis, the focus is on new monomers and mechanistic insights. The identification of improved catalysts is a high priority area and in the latter portion of this section we describe examples of Ni and Pd-catalysis. In recent years, two reports of catalytic silane dehydrocoupling by late transition metals (Fe117–119 and Ir120,121) were later corrected to reflect incorrect product assignment.

14.05.3.2.1

Early transition metals

Shankar et al. introduced trihydroalkylsilane monomers with pendant heteroaromatic rings (Fig. 16).122,123 Polymerization with Cp2TiCl2/n-BuLi for 2 days at 50  C yielded very low molecular weight products (Mn < 1750 g mol−1, Mw/Mn  1.5). The authors attributed the low degree of polymerization to the lower reactivity of alkylsilanes compared to arylsilanes. However, the low reactivity of 2-thienylsilane, which Shankar reported failed to polymerize, points to a potential role for thienyl-coordination in suppressing catalyst activity. Tanaka also reported that 2-thienylsilane does not polymerize with a zirconocene-derived catalyst.124 Building on prior work on disilane polymerization,125,126 in 2010 Lunzer and Marschner investigated dehydropolymerization of the disilane H2MeSi–SiPhH2 with Cp2ZrCl2/n-BuLi, which yielded a polymer with an unusually broad molecular weight distribution (Fig. 17).127 This was attributed to cross-linking, as fractionation also indicated the presence of both a soluble and an insoluble portion. Clear evidence of cross-linking, e.g. spectroscopic signatures of fully-substituted Si atoms, was not observed by the 29Si NMR spectroscopy. The final cross-linked polymer is expected to be enriched in phenylsilane units, as GCMS analysis of early-stage polymerization showed oligomers enriched in phenyl groups, which was attributed to some SidSi bond fragmentation. Interestingly, the copolymerization of 1,2-dimethyldisilane (H2MeSi–SiMeH2) and PhSiH3 produced a linear and soluble copolymer with a more typical molecular weight distributions (Mn  2100 g mol−1, Mw/Mn ¼ 2.8). In 2017, Klausen reported cyclosilane 1,4Si6 as a new type of bifunctional monomer.14 The previously unreported monomer was synthesized in five steps from commercially available starting materials via a key dianion intermediate (Fig. 18A). A crystal structure showed that 1,4Si6 adopted a chair conformation in the solid state. Polymerization with Cp2ZrCl2/n-BuLi in toluene at either room temperature or 90  C provided low molecular weight samples of poly(1,4Si6) (Mn  3000 g mol−1, Mw/Mn < 1.5, Fig. 18B). The average degree of polymerization was 10, based both on GPC analysis and integration of 1H NMR resonances corresponding to end groups and internal resonances. Due to the step growth mechanism of silane dehydropolymerization, a bifunctional monomer posed unique challenges. While the 1,4Si6 monomer has only the SiH2 functional groups, dimers and higher oligomers have both SiH2 end groups and SiH internal sites (Fig. 18C). Chain extension from internal sites would give rise to branched polymers. Based on the lack of reactivity seen with sterically encumbered monohydrosilanes (R3SiH) in dehydrocoupling, chain extension from only the SiH2 end groups was expected. 1H, 29Si {1H} DEPT, and 29Si INEPT+ NMR spectroscopy, as well as 1H-29Si correlated spectroscopy, supported assignment to a linear structure as distinct resonances assigned to SiH2 end groups and SiH internal sites were identified and no evidence of SiSi4 sites was found (Fig. 18D).

Fig. 16 Thienyl- and furyl-functionalized monomers for silane dehydrocoupling.

Fig. 17 Disilane polymerization.

Organosilicon and Related Group 14 Polymers

147

Fig. 18 (A) Chemical and X-ray crystal structure of bifunctional cyclosilane monomer 1,4Si6. (B) Catalytic 1,4Si6 polymerization. (C) Schematic representation of dimer reactivity. (D) 29Si INEPT+ spectra of 1,4Si6 and poly(1,4Si6) showing peak assignments to SiH2 end groups (apparent triplet) and SiH internal sites (apparent doublets). Modified from Refs. Press, E. M.; Marro, E. A.; Surampudi, S. K.; Siegler, M. A.; Tang, J. A.; Klausen, R. S. Synthesis of a Fragment of Crystalline Silicon: Poly(Cyclosilane). Angew. Chem. Int. Ed. 2017, 56(2), 568–572 (Pub. Date 11/29/206), Copyright 2016 John Wiley and Sons Ltd.; and from Marro, E. A.; Klausen, R. S. Conjugated Polymers Inspired by Crystalline Silicon. Chem. Mater. 2019, 31(7), 2202–2211 (Pub. Date 03/08/2019), Copyright 2019 American Chemical Society (link: https://pubs.acs.org/doi/10.1021/acs.chemmater.9b00131, additional permissions related to the material excerpted should be directed to the ACS), with permission.

In a subsequent publication, Klausen et al. investigated other group 4 metallocene catalysts for 1,4Si6 polymerization.128 In the Cp2MCl2/n-BuLi series where M ¼ Ti, Zr, or Hf, Zr yielded the highest molecular weight polymer. While the NMR spectra of polymers arising from Zr or Hf catalysis were similar, when M ¼ Ti, the 29Si{1H} DEPT spectrum (DEPT ¼ distortionless enhancement of polarization transfer) suggested a branched polymer. The use of more sterically hindered precatalysts (e.g. (Cp∗)2ZrCl2 where Cp ¼ pentamethylcyclopentadiene) inhibited polymerization and no poly(1,4Si6) was isolated. Additionally, when Cp∗CpZrCl2/2nBuLi or Cp2ZrMe2 were used as precatalysts, the reaction yielded lower molecular weight poly(1,4Si6) (Mn ¼ 2100 g mol−1 vs. 2770 g mol−1). Poly(1,4Si6) possesses a different kind of main chain tacticity than linear polysilanes of the general formula poly(SiR2). Each internal cyclosilane ring can exist as either a cis- or a trans-diastereomer, although the overall polymer is achiral due to a mirror symmetry (Fig. 19A). For examples of investigation into the effect of relative stereochemical configuration on the properties of molecular cyclosilanes, the reader is referred to several recent studies.129–131 Each diastereomer also prefers a different conformation at the central ring, with the trans adopting a chair conformation and the cis preferring a twist conformation, a conclusion based on DFT calculations.132 A solid state NMR (SSNMR) study by Klausen and Rossini identified that the 1H!29Si CPMAS spectrum of poly(1,4Si6) showed two sets of three peaks, labeled a-c and a’-c’ (Fig. 19B). The two peak sets were assigned to the major trans and minor cis diastereomers on the basis of computational modeling and simulated NMR spectra. An additional insight arising from SSNMR spectroscopy included an alternative to GPC for estimation of molecular weight (degree of polymerization ¼ 20) based on integration of end group and internal 29Si NMR resonances. In 2018, the Klausen group reported the synthesis and polymerization of the isomeric cyclosilane monomer 1,3Si6 (Fig. 20A).133 Under conditions used for 1,4Si6 (Cp2ZrCl2/n-BuLi), poly(1,3Si6) was isolated in a similar degree of polymerization (Mn ¼ 3260 g mol−1, Mw/Mn ¼ 1.47). While the 1H NMR spectra of the isomeric monomers 1,3Si6 and 1,4Si6 were similar, the polymer spectra were significantly different. Most notably, in the region where SiH resonances appear, the spectrum of poly(1,3Si6) was dominated by a broad singlet (Fig. 20B). This suggested that poly(1,3Si6) was both highly symmetric and without end groups.

148

Organosilicon and Related Group 14 Polymers

Fig. 19 (A) Representation of trans and cis diastereomers in a 1,4Si6 trimer. (B) 1H!29Si CPMAS spectra of poly(1,4Si6) with the molecular structures (methyl groups omitted and only the chair conformer of the all-trans stereoisomer shown for clarity). The 29Si spectrum was fit to six distinct isotropic peaks. Modified from Ref. Dorn, R. W.; Marro, E. A.; Hanrahan, M. P.; Klausen, R. S.; Rossini, A. J. Investigating the Microstructure of Poly(Cyclosilane) by 29Si Solid-State NMR Spectroscopy and DFT Calculations. Chem. Mater. 2019, 31(21), 9168–9178 (Pub. Date 11/01/2019), Copyright 2019 American Chemical Society, with permission.

Fig. 20 (A) Catalytic polymerization of 1,3Si6 to poly(1,3Si6), an ensemble of largely cyclic polymers. (B) 1H NMR spectra showing poly(1,3Si6) is a high symmetry material without SiH2 end groups. (C) DFT-optimized conformation of a cyclic hexamer of 1,3Si6 (methyl groups omitted for clarity). B3LYP/6-31G(d). Modified from Ref. Marro, E. A.; Press, E. M.; Siegler, M. A.; Klausen, R. S. Directional Building Blocks Determine Linear and Cyclic Silicon Architectures. J. Am. Chem. Soc. 2018, 140(8), 5976–5986 (Pub. Date 05/01/2018), Copyright 2018 American Chemical Society, with permission.

These data, as well as additional 29Si NMR and IR spectra, suggested that poly(1,3Si6) was predominantly cyclic. A minor low-molecular weight fraction apparent by GPC analysis was assigned to low-molecular weight oligomers too short to cyclize. DFT calculations of a six-membered macrocycle supported the high symmetry suggested by the 1H NMR spectrum (Fig. 20C), as the hexamer possessed both mirror symmetry and two rotational axes (C2 and C3).48 A subsequent 2021 study showed that a change in catalyst to Cp2ZrMe2 afforded linear oligomers of 1,3Si6.134

Organosilicon and Related Group 14 Polymers

149

Fig. 21 Poly(cyclosilane) optical properties. (A) Absorbance spectra of 1,3Si6 (dashed) and poly(1,3Si6) (solid) in tetrahydrofuran. THF spectral cutoff is 220 nm. [1,3Si6] ¼ 3.0  10−5 M, [poly(1,3Si6)] ¼ 1.0  10−2 mg mL−1. (B) Calculated absorbance spectra of 1,3Si6 (dashed) and c-(1,3Si6)6 (solid). TD–PBE0/6311G(d)//B3YLP/6-31G(d). (C) Calculated frontier molecular orbitals of cyclo-(1,3Si6)6. Orbital density is concentrated on the silicon atoms lining the pore and is not delocalized over the entire framework. TD–PBE0/6-311G(d)//B3YLP/6-31G(d). Modified from Ref. Marro, E. A.; Press, E. M.; Siegler, M. A.; Klausen, R. S. Directional Building Blocks Determine Linear and Cyclic Silicon Architectures. J. Am. Chem. Soc. 2018, 140 (8), 5976–5986 (Pub. Date 05/01/2018), Copyright 2018 American Chemical Society, with permission.

Polysilanes bearing cyclic subunits pose interesting questions for the nature of s-conjugation. Cyclohexasilanes enforce gauche “turns,” preventing adoption of the all-anti conformation that maximizes s-conjugation. UV–vis spectroscopy comparing cyclosilane monomers and polymers (e.g. 1,3Si6 vs. poly(1,3Si6)) showed a ca. 50 nm bathochromic shift in the onset of absorption after polymerization, with poly(1,3Si6)’s lmax falling at 296 nm at room temperature (Fig. 21A).133 Time-dependent density functional theory (TD-DFT) calculations reproduced this trend (Fig. 21B) using the cyclic hexamer cyclo-(1,3Si6)6 as a model for the cyclic polymer. The experimentally observed lmax of 296 nm is hypsochromically shifted relative to poly[Si(n-C4H9)2] prepared by Wurtz polymerization (lmax ¼ 316 nm at room temperature).9,108 The higher energy optical transition seen in poly(1,3Si6) likely reflected both the lower degree of polymerization and the distinct conformational profile of cyclosilanes. In support of the latter, DFT calculations of cyclo-(1,3Si6)6 showed that in the highest occupied and lowest unoccupied molecular orbitals (HOMO and LUMO) s-conjugation was not delocalized across the entire molecule, but instead focused on the atoms lining the central pore (Fig. 21C). The HOMO-1 to LUMO transition was found to be the major contributor to the calculated absorption spectrum. On the whole, these results point to opportunities in designing polysilane optical properties through manipulation of the silane backbone.

14.05.3.2.2

Late transition metals

Since COMC IV, several groups have investigated silane dehydrocoupling with late transition metal catalysts, especially Rh, Ir, Ni, and Pd. Late transition metals appeal for their well-established functional group tolerance in organic synthesis, but the challenge and opportunity in late transition metal catalysis is access to a greater range of potential mechanisms, including redox cycles and silylenoid intermediates. However, to date very low degrees of polymerization and significant side reactivity (e.g. redistribution) are common.

150

Organosilicon and Related Group 14 Polymers

14.05.3.2.2.1 Group 9 catalysts (Rh, Ir) The polymerization of hydrosilanes by Wilkinson’s catalyst (Ph3P)3RhCl was first reported in 1973 by Nagai et al.135 Significant redistribution was observed: the reaction of Ph2SiH2 with (Ph3P)3RhCl (0.1 mol%) yielded the dimer Ph2HSi–SiHPh2 (38%), recovered monomer (45%), and the redistribution products Ph3SiH, Ph4Si, and PhSiH3 (8%). Rosenberg later demonstrated that hydrogen removal is critical for favoring clean Ph2SiH2 dimerization with (Ph3P)3RhCl.136 In 2006, Rosenberg reported a study of ligand effects on the dimerization and trimerization of di-n-hexylsilane ((n-C6H13)2SiH2), including the inhibitory effect of cis-chelating ligands on the rate of dimerization.137 In 2011, McIndoe and Rosenberg presented a detailed mechanism for silane dehydrocoupling by (Ph3P)3RhCl (Fig. 22) based on ESI-MS and NMR spectroscopic evidence of key Rh-containing intermediates.138 The proposed active catalyst is a Rh(I) hydride which is thought to arise from oxidative addition of silane to (Ph3P)3RhCl, followed by reductive elimination of a chlorosilane (“1” !“3”). Polysilanes were proposed to arise from silylene (R2Si) polymerization and the authors suggested a mechanism for R2Si formation via a Rh(I)/Rh(III) cycle and a-elimination (dotted box, Fig. 22). An alternative pathway involving on-catalyst SidSi bond formation is also consistent with the available evidence. Additionally, the authors provided strong evidence for catalyst degradation arising from adventitious water. Waterman has investigated the reactivity of iridium complexes bearing pincer bis(phosphinite) ligands with phenylsilane (Fig. 23), catalysts previously reported for the dehydropolymerization of methylamine-borane (MeNH2-BH3) to poly(aminoboranes).139 While the product was initially characterized as cyclo-(PhHSi)10,120 reinvestigation showed that PhSiH3 redistribution to Ph2SiH2 was the major pathway.121 The reaction of Ph2SiH2 with adventitious water during work-up yielded siloxane products that were misidentified as cyclo-(PhHSi)10. Subsequent work investigating ligand design demonstrated partial suppression of PhSiH3 redistribution in favor of oligomerization with hydrogen removal under dynamic vacuum.140

14.05.3.2.2.2 Group 10 catalysts (Ni, Pd) The 1990s yielded the first reports of homogeneous group 10 catalysis of silane dehydrocoupling, including both Ni141 and Pt complexes.142 Heterogeneous Ni catalysis of PhSiH3 oligomerization was reported in 1991.143 Since 2007, significant additional development of new ligand scaffolds for group 10 complexes has been made, especially with respect to low valent Ni complexes.

Fig. 22 McIndoe and Rosenberg: proposed mechanism for H2SiR2 dehydrocoupling by Wilkinson’s catalyst. Modified from Ref. Jackson, S. M.; Chisholm, D. M.; McIndoe, J. S.; Rosenberg, L. Using NMR and ESI-MS to Probe the Mechanism of Silane Dehydrocoupling Catalyzed by Wilkinson’s Catalyst. Eur. J. Inorg. Chem. 2011, 2011(3), 327–330 (Pub. Date 12/09/2010), Copyright 2011 John Wiley and Sons Ltd., with permission.

Fig. 23 Ir complexes with pincer bis(phosphinite) ligands for phenylsilane redistribution and oligomerization.

Organosilicon and Related Group 14 Polymers

151

Fig. 24 Nickel indenyl complexes for PhSiH3 polymerization.

In 1998, Zargarian reported that indenyl complex Ni1 activated by Lewis acids or lithium aluminum hydride reduction catalyzed silane dehydropolymerization. In several subsequent publications, Zargarian investigated fine-tuning of the Ni coordination environment to obtain single-component precatalysts for PhSiH3 polymerization (Fig. 24).144,145 The parent system Ni1 activated by methylaluminoxane (MAO) resulted in the highest molecular weight poly(SiPhH) (Mn ¼ 5900 g mol−1, Mw/Mn ¼ 1.20). While the methyl-functionalized complexes Ni2-Ni4 largely provided low molecular weight cyclosilanes (e.g. cyclo-(SiHPh)6), much higher turnover numbers were observed, especially for bifunctional complex Ni3. Zargarian also recently reported a pincer bis(phosphinite) Ni complex similar to Waterman’s Ir complex that promoted PhSiH3 redistribution and dimerization.146 In 2010, Abu-Omar reported that the dimeric nickel complex [(dippe)Ni(m-H)]2 (dippe ¼ 1,2-bis(diisopropylphosphino) ethane) catalyzed both PhSiH3 and PhMeSiH2 dehydrocoupling.147 Analysis of molecular weight distribution by GPC indicated a higher degree of polymerization for PhSiH3 than PhMeSiH2 (Mn ¼ 1750 g mol−1 vs. 951 g mol−1), as also seen for group 4 metallocene catalysis. In both cases, cyclic oligomers were also observed. The reaction outcomes, in terms of substrate relative rates and overall degrees of polymerization, were very similar for [(dippe)Ni(m-H)]2 and group 4 metallocenes. Abu-Omar suggested that a similar s-bond metathesis mechanism might be operative (Fig. 25A). However, [(dippe)Ni(m-H)]2 is far more electron-rich than the d0 Zr(IV) species implicated in PhSiH3 polymerization by s-bond metathesis and a Ni(0)/Ni(II) redox cycle with a Ni]Si intermediate cannot be excluded (Fig. 25B). This example highlights the opportunities arising from the greater mechanistic possibilities available to late transition metal catalysts, as well as the challenges in discerning between those possibilities. In 2013, Osakada reported an investigation of the reactivity of Ni(dmpe)2 where dmpe ¼ 1,2-bis(dimethylphosphino)ethane with PhSiH3, n-H13C6SiH3, and 9,9-dihydrosilafluorene.148 A mixture of cyclic and acyclic oligomers were obtained (Mn < 1500 g mol−1) at room temperature and only cyclopolymer at elevated temperature (Fig. 26A), as characterized by 29Si NMR spectroscopy. FAB-MS supported assignment to an ensemble of different cycles, from octamer to undecamer. The cyclic

Fig. 25 Mechanistic dichotomies in PhSiH3 polymerization by [(dippe)Ni(m-H)]2 (dippe ¼ 1,2-bis(diisopropylphosphino)ethane): (A) s-bond metathesis and (B) Ni-silylene intermediate.

152

Organosilicon and Related Group 14 Polymers

Fig. 26 (A) Cyclopolymerization by [Ni(dmpe)2] (dmpe ¼ 1,2-bis(dimethylphosphino)ethane). (B) X-ray crystallographic characterization of silapalladacycle. Modified from Ref. with permission, Tanabe, M.; Takahashi, A.; Fukuta, T.; Osakada, K. Nickel-Catalyzed Cyclopolymerization of Hexyl- and Phenylsilanes. Organometallics 2013, 32(4), 1037–1043 (Pub. Date 02/01/2013), Copyright 2013 American Chemical Society, with permission.

Fig. 27 Phensilane dehydrocoupling catalyzed by [Ni2(iPr2Im)4(COD)] 5.

oligomers are likely a thermodynamic sink, as a mixture of cyclic and acyclic materials could be converted into exclusively cyclopolymers upon extended reaction times with additional monomer. Osakada suggested that reductive elimination from a bis-silyl intermediate could account for SidSi bond formation. In support, the authors found that 9,9-dihydrosilafluorene reacted with (dmpe)PdMe2 to afford the palladacycle (dmpe)Pd(SiR2)4 in 44% yield, which was crystallographically characterized (Fig. 26B). A bis-silyl precursor was also observed in solution. In 2016, Osakada published a follow-up study that closely investigated the polymerization of a substituted silafluorene monomer and identified conditions leading to a linear poly(silafluorene) with Mn ¼ 3860 g mol−1.149 In 2014, Radius et al. added catalyst 5 (Fig. 27) to the arsenal of low valent Ni complexes known to promote silane dehydrocoupling.150 A low degree of polymerization was observed (5 < n < 17), based on GPC analysis and MALDI-TOF. Both linear and cyclic oligomers were present in a 38:62 ratio. The authors also reported preliminary work on the reverse reaction, the hydrogenolysis of disilanes to hydrosilanes. In 2020, Trincado, Grützmacher and coworkers reported the synthesis and X-ray crystallographic characterization of a series of nickel complexes stabilized by a tropylamine ligand, such as Li[NiH(trop2NMe)] (Fig. 28), which catalyzed both the dehydrogenative oligomerization of Ph2SiH2 and the hydrogenolysis of polysilanes to monomeric hydrosilanes.151 The reversibility of silane dehydropolymerization is well-established115 and thermodynamic considerations are important in understanding reaction outcome. Indeed, while the Me3Si–SiMe3 bond is somewhat weaker than the Me3Si–H bond, the very strong H–H bond (BDE ¼ 436 kJ mol−1) is a significant enthalpic driving force for silane dehydrocoupling. Effective hydrogen removal has also proven critical to influencing reaction outcome, as in Rosenberg’s work on Rh-catalyzed H2SiPh2 dimerization.136 Trincado and

Organosilicon and Related Group 14 Polymers

153

Fig. 28 X-ray crystallographic structure of Li[NiH(trop2NMe)]•3THF. A simplified molecular structure of the Ni coordination environment is provided for clarity. Modified from Ref. Pribanic, B.; Trincado, M.; Eiler, F.; Vogt, M.; Comas-Vives, A.; Grützmacher, H. Hydrogenolysis of Polysilanes Catalyzed by Low-Valent Nickel Complexes. Angew. Chem. Int. Ed. 2020, 132(36), 15733–15739 (Pub. Date 04/14/2020), Copyright 2020 John Wiley and Sons Ltd., with permission.

Fig. 29 Proposed mechanism for disilane hydrogenation via s-bond metathesis. Modified from Ref. Pribanic, B.; Trincado, M.; Eiler, F.; Vogt, M.; Comas-Vives, A.; Grützmacher, H. Hydrogenolysis of Polysilanes Catalyzed by Low-Valent Nickel Complexes. Angew. Chem. Int. Ed. 2020, 132(36), 15733–15739 (Pub. Date 04/14/ 2020), Copyright 2020 John Wiley and Sons Ltd., with permission.

Grützmacher et al. also performed DFT calculations to understand the mechanism of Ph2HSi–SiHPh2 hydrogenation (Fig. 29). The NidH species I was suggested as the active catalyst, which is formed by rearrangement of the ligand scaffold in Li[NiH(trop2NMe)]. The catalytic cycle involves a s-bond metathesis reaction in which the Si − Si bond is cleaved (TSII,III). An anionic silyl intermediate IV is formed after release of one Ph2SiH2 molecule; then, H2 oxidative addition (TSV,VI) generates the dihydride VI that decomposes to form a second molecule of Ph2SiH2 and an isomer of I, VIII, which interconverts to I, allowing the catalytic cycle to begin again. These theoretical results have three interesting implications. First, given microscopic reversibility, an analogous mechanism might apply to dehydropolymerization. Second, that both s-bond metathesis and oxidative addition-reductive elimination steps are crucial in the catalytic cycle of a late transition metal-catalyzed polysilanes hydrogenolysis speaks to diversity of catalysis in this regime. Third, the use of chelating ligands that offer steric protection and flexibility allows both stabilization of low-valent complexes and open space to support coupling reactions. While significant effort has focused on low valent Ni complexes, less work has investigated Pd-catalysis. Marschner and Kempe reported in 2006 the synthesis of 2-aminopyridine complexes of Ni and Pd and showed that the L2Pd0 (L ¼ 2-trimethylsilylaminopyridine) complex was effective for the conversion of 1,2-dimethyldisilane (H2MeSi–SiMeH2) to high molecular weight polymers (Fig. 30).152 Significantly shorter reaction times were observed than required for Pt(PEt3)3 and (Ph3P)3RhCl (30 min vs. 3 days). However, instead of H2 evolution, methylsilane (MeSiH3) evolution was detected by GCMS, suggesting a mechanism involving concomitant redistribution and chain extension.

154

Organosilicon and Related Group 14 Polymers

Fig. 30 1,2-Dimethylsilane polymerization by a L2Pd0 complex (L ¼ 2-trimehtylsilylaminopyridine).

The results presented in this section showcase the mechanistic diversity available to late transition metals in reactions with hydrosilanes, but also indicate the challenges in controlling product distribution.

14.05.3.3 Other synthetic methods In COMC III (2007), examples of polysilane synthesis via ring opening polymerization (ROP) and electropolymerization were also discussed. While the challenging monomer synthesis has somewhat limited further investigation of cyclosilane anionic ring-opening polymerization, the method offers some advantages. In 2005, Sakurai and coworkers reported the polymerization of the amino-functionalized masked disilene 6 (Fig. 31A). Poly(6) can be obtained in yields as high as 89%, with Mn between 15,000–26,000 and Mw/Mn between 1.5 and 1.8, depending on reaction conditions.10 The isolation of poly(6) offered a convenient approach to linear polysilanes in which the backbone has direct bonds with heteroatoms. An analogous synthesis carried out following a Wurtz-type procedure resulted in the formation of a branched polymer (Mn ¼ 5100–8000, Mw/Mn ¼ 1.8–2.5), in lower yields (9–17%).153 Poly(6) was also found to be syndiotactic (enriched in racemo dyads, r relative to meso dyads, m) (Fig. 30B). Depending on the initiator, the % r was as high as 89%. Since COMC III, two new polysilanes were prepared by electropolymerization. Kulandainathan reported the synthesis of poly [SiMe(CHCH2) (Mn ¼ 16,400 g mol−1, Mw/Mn ¼ 1.03) in 76% yield, using a water-free cell equipped with a 1,2-dimethoxyethane (DME) solution of dichloromethylvinylsilane and [nBu4N][BF4], and aluminum electrodes (current density ¼ 4.5 mA cm−2).154 Okano and Yamada carried out electrochemical reduction of 1,1-dichlorosilacyclopentane.155 The reaction was performed using dried DME as solvent, [nBu4N][ClO4] as supportive electrolyte, a platinum cathode and a silver anode. Low to moderate yields (14–28%), relatively low Mn (104 g mol−1) and broad dispersity (Mw/Mn ¼ 2.2–3.1) was observed. Ishifune and coworkers have also studied the use of disilanes containing the triphenylsilyl group (−SiPh3), i.e. Me3Si–SiPh3, as an additive in the electropolymerization of MePhSiCl2 using Mg electrodes.156

Fig. 31 (A) Anionic ring-opening polymerization of amino-functionalized masked disilene 6. (B) Cropped 1H NMR spectrum of poly(6). Modified from Ref. Sakurai, H.; Honbori, R.; Sanji, T. Stereoselective Anionic Polymerization of Amino-Substituted Masked Disilenes. Organometallics 2005, 24(17), 4119–4121 (Pub. Date 08/01/2005), Copyright 2005 American Chemical Society, with permission.

Organosilicon and Related Group 14 Polymers

155

In 2013, Rieger reported PhSiH3 polymerization by tris(pentafluorophenyl)borane (B(C6F5)3).157 Heating PhSiH3 to 90–120  C with B(C6F5)3 (0.2–6 mol%) yielded the branched polysilane [SiHxPhy]n, along with SiH4 and benzene. Yields were circa 15%, with Mn between 1000 and 2000 g mol−1 and Mw/Mn  1.6. Consistent with a branched structure, the 1H–29Si HMBC spectrum showed cross-peaks consistent with terminal −SiPh3, −SiPh2H, −SiPhH2 and –SiH3 moieties. While B(C6F5)3 did not catalyze (n-C6H13) SiH3 homopolymerization, the branched copolymer [SiHxPhy(n-C6H13)z]n (Mn ¼ 1100 g mol−1, Mw/Mn ¼ 1.2) was obtained by heating at 100  C PhSiH3/(n-C6H13)SiH3/B(C6F5)3 mixture in 100/75/1 M ratio. Control experiments using TEMPO and dark conditions excluded a radical mechanism or photolytic initiation. To explore the novel possibility of a cationic polymerization, the authors carried out DFT calculations. Initiation is proposed to proceed through a concerted hydride abstraction-arene coordination that forms a silylium-borate contact ion pair ([PhH2Si]+[HB(C6F5)3]−) that is engaged in a cation-p interaction with the aromatic group of a second PhSiH3 molecule (Fig. 32). Propagation proceeds either via chain growth (loss of benzene), substitution (loss of silane), or branching (loss of benzene). In all cases, silylium intermediates were stabilized by coordination to an aromatic group. We conclude this section with an interesting novel oligomerization reported by Iwamoto in 2020 (Fig. 33).158 The bicyclo[1.1.1] pentasilane 7 self-condensed on treatment with dimethylaminopyridine (DMAP) to give mostly dimer and trace trimer. An additional higher molecular weight fragment was observed by GPC analysis. The structure of the dimer and trimer were confirmed by HRMS and by comparison of NMR spectra to authentic samples synthesized by step-wise oligomerization.159 The degree of polymerization in this one-pot condensation reaction is likely limited by precipitation of DMAP•TMSI, which removes the initiator from solution. Identification of a soluble initiator that reversibly complexes TMSI would potentially allow for higher degrees of polycondensation.

Fig. 32 DFT-calculated initiation and propagation in the cationic polymerization of PhSiH3 by B(C6F5)3.

Fig. 33 One-pot oligomerization of cyclosilane 7 by elimination of iodotrimethylsilane.

156

Organosilicon and Related Group 14 Polymers

14.05.3.4 Postpolymerization functionalization 14.05.3.4.1

SidH functionalization

Some substituents such as heteroatoms, metals or unsaturated hydrocarbons are challenging to incorporate into polysilanes by direct polymerization. In this respect, polyhydrosilanes (poly(SiHR)) are promising for postpolymerization functionalization by SidH functionalization chemistries. However, SidSi bond fragmentation and skeletal rearrangement with Pt-based hydrosilation catalysts,160 titanocene SiH/NH cross-dehydrocoupling catalysts,161 and free radical initiators162,163 are a significant concern, as documented by Rosenberg et al.164 Simionescu studied the H2PtCl6-catalyzed hydrosilylation of poly(SiPh2-co-SiMeH)100 with allyl bromide and allyl iodide, both with and without microwave irradiation.165,166 Microwave irradiation reduced reaction time from 50 h to 2 min without sacrificing yield. Shankar and Shahi reported the use of 2,20 -azo(bisisobutyronitrile) (AIBN) as free radical initiator in the hydrosilylation reaction between poly[SiH(n-C6H13)] and allyl/vinylsilanes containing one or two thienyl substituents,167 a reaction inspired by Waymouth’s original reports of AIBN-functionalization of polyhydrosilanes.162,163 Rosenberg has reported the most comprehensive studies of poly(SiPhH) postpolymerization functionalization.11,164 These studies build off prior work on small molecules in which Rosenberg et al. found that B(C6F5)3 uniquely catalyzed chemoselective SidH bond functionalization without SidSi bond cleavage.168 Rosenberg et al. has demonstrated derivatization of poly(SiPhH) by hydrosilation of alkenes, imines, ketones and thioketones, the dehydrocoupling of phenols, alkanethiols and phenylthiols, and the demethanative coupling of ethers (Fig. 34).11,164 Conversions were quantitatively determined by 1H NMR spectroscopy and were typically between 10% and 50% (Table 1). GPC analysis and thermogravimetric analysis (TGA) supported the conclusion that the polysilane backbone remained intact.

14.05.3.4.2

Polysilane deprotection and functionalization

The cleavage of the SidPh bond by strong acids (HX) to generate silane electrophiles (Si-X) and benzene is a well-known transformation extensively developed in organic synthesis for the Tamao-Fleming oxidation.169,170 Application of this reaction to the cleavage of SidPh bonds in polysilanes dates to the 1990s.171,172 We summarize here recent examples. Qin et al. introduced azobenzene side-chains via chlorination and substitution of poly(SiMe2-co-SiMePh) (Fig. 35).173 The known phenyl-to-chloride conversion was induced using AcCl/AlCl3. Addition of alcohol 8 and related structures, followed by butanol workup, provided the functionalized polysilane 9 as a red solid in 46% yield. The UV spectrum of 8 shows an absorption band at l ¼ 498 nm consistent with the azobenzene chromophore. Characterization including FTIR, 1H NMR and elemental analysis supported the structural assignment. Frey introduced acetal-protected diol side chains to polysilanes (Fig. 36).174 Wurtz copolymerization of Cl2Si(n-C6H13)2 and the acetal-functionalized monomer IMS yielded copolymers with between 6% and 40% IMS incorporation. Acetal deprotection with trifluoroacetic acid (TFA) in water yielded a polyol-functionalized polysilane. Over 90% protecting group removal was accomplished. The authors concluded that no polysilane fragmentation occurred, as the polydispersity was unchanged after acid hydrolysis.

14.05.3.4.3

Miscellaneous postfunctionalization

In 2014, Lai et al. described a novel polysilane postpolymerization functionalization to induce control of helical chirality.175 Wurtz polymerization of Cl2SiMe(4-MeC6H4) gave rise to poly[SiMe(4-MeC6H4)] (Mn ¼ 4040 g mol−1, Mw/Mn ¼ 2.66) as an optically inactive polymer. By analogy to the deprotonation of poly(4-methylstyrene), the authors hypothesized that the benzylic position could be deprotonated by chiral amine/organolithium complexes such as (−)-sparteine/tert-butyllithium. The preferential screw sense of poly[SiMe(4-MeC6H4)] could be switched between predominantly the M- and P-helices by changing the enantiomer of chiral amine, as shown by circular dichroism (CD).

B(C6F5)3

Fig. 34 Synthesis of functionalized polysilanes via B(C6F5)3 catalysis.

Organosilicon and Related Group 14 Polymers

Table 1 Substrate

Functionalization of poly(SiPhH) with B(C6F5)3. X

Method

%X

A

20

A

40

A

10

A

25

A

n. d.

A

n. d.

A

n. d.

B

10

B

40

B

20

C

25

C

25

n. d. ¼ not determined. Modified from Ref. Lee, P. T. K.; Rosenberg, L. Borane-Catalysed Postpolymerisation Modification of the SidH Bonds in Poly(Phenylsilane). Dalt. Trans. 2017, 46(27), 8818–8826 (Pub. Date 03/24/2017), Copyright 2017 Royal Society of Chemistry, with permission.

Fig. 35 Synthesis of azobenzene-functionalized polysilane 9.173

157

158

Organosilicon and Related Group 14 Polymers

Fig. 36 Acetal hydrolysis yielding a polyol-functionalized polysilane.174

14.05.4 Polygermanes In COMC III (2007), the synthesis of polygermanes was described to proceed mainly by Wurtz-type reactions, demethanative coupling, and electropolymerization. In this section, we highlight three manuscripts reporting new aspects to polygermane synthesis and an interesting computational study of poly(GePh2) electronic structure.176 Mochida and coworkers investigated the use of SmI2 as one-electron reducing agent in the dehalogenative coupling of R2GeCl2 (R ¼ Me, Et).177,178 For this Wurtz-type reaction, they found that refluxing conditions slightly improved the product properties. For example, when the reaction is run at room temperature, the product poly(GeEt2) was isolated in 19% yield (Mn ¼ 2000 g mol−1, Mw/Mn ¼ 1.17) but when the reaction mixture is refluxed, the yield was 25% (Mn ¼ 4300 g mol−1, Mw/Mn ¼ 1.13). Yagci et al. reported photopolymerization of dibenzoyldiethylgermane (Et2Ge(OBz)2) to poly(GeEt2) in 2009.179 The authors proposed that Et2Ge(OBz)2 photodecomposed to generate diethylgermylene (Et2Ge), which homopolymerized before trapping with benzoyl radical. In support, 1H NMR spectroscopy showed signatures of benzoyl end groups. Ohshita reported the Wurtz polymerization of 4,4-dichlorodithienogermoles 10a and 10b, respectively (Fig. 37).15 The analogous monomer without alkyl substituents (R ¼ H) yielded an insoluble polymer. The UV–vis spectra of poly(10a-b) showed strong absorption out to ca. 450 nm, bathochromically shifted by 75 nm relative to a molecular digermane. The UV–vis spectrum of a copolymer with interspersed dibutyl and dithienyl side chains (poly[10b-co-Ge(n-C4H9)2]) showed a hypsochromic shift in lmax, suggesting a cooperative effect between the s-conjugated main chain and the p-conjugated backbone. Since COMC III (2007), in addition to the polygermane studies summarized here, there have been several reports regarding the synthesis of linear and cyclic oligogermanes,180–187 but these are not the subject of this chapter. Nevertheless, the authors are optimistic that the advances in molecular synthesis and understanding of structure-property relationships will motivate future work on innovative polygermanes.

Fig. 37 Wurtz polymerization of dichlorodithienogermoles.

Organosilicon and Related Group 14 Polymers

159

14.05.5 Polystannanes Polystannanes are unique relative to polysilanes and polygermanes in that the parent element is a metal and not a semiconductor. Polystannanes are more sensitive in solution to moisture and visible light than polysilanes and polygermanes, to the point that syntheses must be carried out in the dark.188 Caseri documented that the inclusion of radical scavengers and visible-light absorbing dyes enhances polystannane stability, pointing to photoinduced SndSn homolysis as the mechanism of decomposition.188 This section will present new developments in polystannane synthesis, including structural changes towards stabilizing polystannanes. As for polysilanes, the two most widely used synthetic methods yielding polystannanes are Wurtz polymerization and dehydropolymerization.27 A new addition to the arsenal of polystannane syntheses is the condensation polymerization of dihydrostannanes and diaminostannanes, which yields alternating copolymers.189

14.05.5.1 Wurtz polymerization The Wurtz reduction of dihalostannanes dates to the 1850s,190 as documented by Caseri in comprehensive historical overviews.28,48 In 1993, Price demonstrated that Wurtz polymerization could yield high molecular weight polystannanes that degraded over time to smaller cyclic oligomers.191 Recent results in this area include the introduction of fluorinated substituents to dichlorodiarylstannane monomers 11a-b (Fig. 38).192 Molecular weight analysis by GPC proved challenging due to rapid (90% conversions and isolated yields. Only the very large and hydrophobic 1,2-epoxydodecane could only be converted in trace amounts.214 Deoxygenation reactions pose important challenges in the efficient implementation of biomass derived feedstocks. An example is the decarbonylation of hydroxymethylfurfural to give furfuryl alcohol (FFA), a valuable bioderived feedstock chemical. A Xantphos containing polymer (P-131(0), where 0 indicates no monophosphine) loaded with 5 wt% nanoparticulate Pd was

Fig. 65 Phosphonium polymers (POPs) as catalytic hosts for the formation of cyclic carbonates. Decarbonylation of hydroxymethylfurfural using Pd-NPs on polymer support. Additional details: i. Ph3SiH, CO2, P-47(Ru). ii. ZnBr2, DMF. iii. K2CO3.

Organophosphorus and Related Group 15 Polymers

223

successfully used in the decarbonylation of hydroxymethylfurfural. Selectivity toward the target FFA was significantly higher compared to homogeneous Pd(II) and Pd(0) catalyst systems, as well as a Pd@PS-TPP polymer system. Comparable selectivity was only observed for Pd/C, although at significantly lower catalytic efficiency. The selectivity for and yield of FFA was maintained over 10 reuse cycles above 95% and 90%, respectively. An upscaled test reaction at 30 mmol gave 77% isolated yield with a TON of up to 1540, making this system very interesting for larger scale applications.215

14.06.4.5.2

Carbon-carbon cross coupling reactions

Cross-coupling reactions are very important catalytic transformation, and the use of heterogenized catalysts is an important area of research. This section doesn’t attempt to provide a comprehensive overview, but rather focuses on a few selected reports exemplifying the design principles and advantages of polymeric and heterogeneous catalyst systems. Defined dendritic systems have been investigated in many different cross-coupling reactions. Among others, early examples by Reetz et al. showed that terminal dN(CH2PPh2)2 groups of P-41 can be easily coordinated to [PdMe2], giving an octanuclear Pd-dendrimer with potential applications in Heck coupling reactions, as illustrated by stilbene synthesis from bromobenzene and vinylbenzene.69 TPP-styrene-like polymers such as P-50(P) are well explored, and are known to typically contain monodentate phosphine binding sites uniformly distributed within the material. In a detailed study, Sawamura and Miura and coworkers have studied the catalytic behavior of monolithic porous PS-TPP materials for the coupling of aryl boronic acids with aryl chlorides. By strictly controlling the feed ratio of crosslinker (DVB) to tris(vinylphenyl)phosphane, and the use of a porogenic solvent, a microporous structure was fabricated that should not suffer limitations due to mass transport phenomena. Moreover, a controlled isolation of the phosphine sites gave mono-P-ligated palladium fragments, i.e., the TPP-[PdCl2L] motif (where L ¼ Ph-CN or solvent). At 50% Pd loading using [PdCl2(PhCN)2] (with respect to P-ligands), solid state 31P-NMR analysis allowed the authors to distinguish this mono-P-ligation (d31P ¼ 33 ppm), from di-P-ligation (d31P ¼ 24 ppm) and uncoordinated TPP units (d31P ¼ −6 ppm). The material with the highest fraction of mono-P-ligated Pd centers (prepared with only 0.5% DVB crosslinker) showed the highest catalytic activity in the coupling of para-tolylchloride with phenyl boronic acid. The apparent diffusion coefficient was determined for the reagents and the product, indicating no limitations arising from mass transport, in comparison with a non-porous P-50(P), where apparent diffusion coefficients are one to two orders of magnitude smaller. The scope of this catalyst was tested including electron deficient and electron rich arylchlorides, giving satisfactory results in all cases.216 Lau et al. explored the performance of a triphenylarsine-based polymer P-50(As) for the same type of Suzuki catalytic cross-coupling reactions (coupling of aryl iodides, 2-iodo-cyclohexenone). Solubility of the polymer, cross-linking (using JandaJel©, P-51), and density of the ligands (using styrene as comonomer) was investigated. Notably, oxygen sensitivity of the ligand played no role, illustrated by the harsh conditions (50 wt% H2O2) required to obtain the corresponding arsine oxide polymers P-50(As).83 This polymer catalyst, using the same polymer matrix with a different Pd-source, [Pd(OAc)2], was also tested in the epoxidation of alkenes showing very good activity, illustrating the tolerance to oxidizing conditions (Fig. 66).84

Fig. 66 Various Suzuki cross-coupling reactions using mononuclear Pd and nanorticular Pd. Ru-catalyzed borylation. Pd-NPs on polymer support catalyzing addition-cyclizations.

224

Organophosphorus and Related Group 15 Polymers

Suzuki-type catalysis is however not limited to polymeric framework materials with single metal sites, i.e., mimicking homogeneous catalysts. As described earlier, these materials can be loaded with metal salts, which upon reduction form stabilized nanoparticles. Zhang et al. used Pd-NPs stabilized in a phosphonium bromide containing polymeric network P-31(PdNP) as Suzuki catalysts. Standard coupling of phenylboronic acid with arylhalides, including the less reactive chloride and fluoride derivatives, showed excellent performance with excellent yields of 99.8% (I), 97 (Br), 96% (Cl) and 87% (F) and good re-usability of the catalyst dropping from 97.4% to 95.2% in five runs.59 Besides the almost ubiquitous polystyrene polymer supports, other materials have shifted into the focus of attention. Polymer swelling behavior, wettability, and various synthetic aspects are important when designing support materials. Other important factors concern green chemistry approaches, which prompted several groups to investigate natural polymers as supports. Although metal nanoparticles can be stabilized by a variety of biopolymers, metal-support interactions are typically weak and of a diverse nature. Wei and coworkers reported an example of a TPP ligand anchored onto cellulose via an ether linkage at the C6 position, which should increase the stability of the system by the strong donor ability of the TPP site. The phosphine was reacted with [Pd(OAc)2] giving in situ reduction to form Pd(0)-NPs. The catalytic activity of this material was studied in Suzuki-Miyaura coupling for various aryl bromides with a large variety of aryl boronic acids. The polymer showed good swelling properties in ethanol benefiting the catalysis. Except for strongly electron withdrawing boronic acid substrates, yields were good to excellent, however it has to be noted that recycling experiments showed quite dramatic decrease of activity already after three runs.217 The elaborate BINAP containing MOF P-84 was successfully coordinated to Rh and Ru. The Rh(ndb) decorated material P-84(Rh) was tested in the asymmetric addition of aryl boronic acids (2-cyclohexenone) and AlMe3 (2-cyclohexenone, and 2-cycloheptenone) to a,b-unsaturated ketones. Isolated yields for the aryl boronic acid addition ranged from 80% to 95% with ee’s of 99% or higher, outperforming the homogeneous reference system and/or requiring lower catalyst loadings. Studies showed that the improved performance is solely due to better stabilization and site isolation of the catalyst in the framework as the local environment was confirmed to be identical in the MOF and homogeneous catalyst. Similar results were obtained in the synthesis of asymmetric a,b-unsaturated alcohols by addition of AlMe3, with good conversions and excellent enantioselectivity (99% or higher).127 The Rh(ndb) coordinated derivative P-84(Rh) was studied in the asymmetric hydrogenation of substituted alkenes. While quantitative conversions could be reached the enantioselectivity was lower, with ee’s ranging from 70% to 91%.127 Early on, polymer bound catalysts with chiral P-substituents were studied for asymmetric hydrogenation. The polystyrene-P(Menth)2 ligand framework P-52(PMenth2) was reacted with [Rh(C2H4)2Cl]2, giving an active catalyst for the reduction of a-acetamidocinnamic acid. While the conditions could be optimized to reach full conversion, the enantiomeric excess was limited to max ca. 60% ee.94 These examples clearly illustrate the previous and current weaknesses in asymmetric hydrogenation reactions. An interesting heterocycle was synthesized from isonitriles and thiobenzamides using a Pd ligated bis(diphenylphosphanyl) methane-PS polymer from monomer M-70(1). Optimizing the Pd-source and reaction conditions ([Pd(OAc)2], PhCl, 60  C, 1 h) and in situ conversion with K2CO3 gave a series of imidazo[4,5-d]thiazoles in 45% to 87% yield over both steps. Recycling of the catalyst was tested for the first step without significant changes of activity over 10 runs (varying between 90% and 94%).218

14.06.4.5.3

Solid-state reactions using immobilized phosphine derivatives and catalyst sequestration, scavenging

The direct use of immobilized phosphines spans many typical applications, including Witting reactions, Appel reaction, etc. Herein we want to give reference to a few selected examples illustrating this approach. Despite having its possible applications, the consumption of the phosphine decorated polymer makes these alternatives often less attractive than catalytic variants thereof. In 1983, Harrison et al. studied in detail the polymer supported Appel reaction converting alcohols into alkyl chlorides using CCl4 and a PS-TPP support P-50(P). After the reaction, they identified phosphine oxide P-50(PO) as well as ylidic P-centers P-50(P+ C−). Mechanistic considerations led to the conclusion that the reaction is accelerated within the polymer support, due to locally increased polarity increasing the reactivity of catalytic intermediates (Fig. 67).89 Similarly, the PS-TPP P-50(P) could be loaded with a Wittig reagent by reaction with alkyl/benzyl bromides. Already in 1972, Heitz and Michels reported that the subsequent reaction with aldehydes and ketones affords the targeted alkene and the polymeric phosphine oxide P-50(PO). Interestingly the crosslinking of the PS-TPP has a significant impact on the reactions. Higher crosslinked polymers are less efficient for larger substrates, presumably due to smaller pore sizes and inaccessibility of reactive sites.219 In continuation of this early work, others have refined the approach by, e.g., making bi-functional polymer reagents P-137, combining the TPP unit with a diethylamine/di-iso-propylamine substituent, colocalizing the Wittig reagent and the base. Such an approach greatly facilitates the workup, allowing to remove the phosphine oxide and the amine hydrohalide by simple removal of the polymer P-134(PO,N+). Among the tested Wittig reactions, high E selectivity and good to excellent yields have been obtained. Attempts to reuse the polymer were successful using a two-step procedure, i.e., a silane reduction (HSiCl3Si), followed by deprotonation with HNEt2. The regenerated polymer could be isolated and reused, albeit with a slight loss of activity (e.g., a drop of conversion from 98% to 94% was seen over five runs).220 An interesting turn on scavenging applications was demonstrated by Lipshutz and Bloomgren. The scavenging process is used to capture and remove homogeneous triphenylphosphane ligands from the reaction mixture. In this approach a Merrifield resin (P-138) with high density of benzylchloride units was used to trap triphenylphosphane (and its oxide). In order to increase the efficacy of the process, NaI is added in situ, modifying the resin. The system was successfully tested for Pd(0) and Ni(0) mediated cross-coupling reactions. Treatment of the phosphonium iodide polymer with LAH in THF released approx. 80% PPh3. Selectivity toward triphenyl phosphine oxide in the presence of other nucleophiles, e.g., amines, was also confirmed illustrating the broader applicability of this approach.96

Organophosphorus and Related Group 15 Polymers

225

Fig. 67 Capture and or release process using PS-TPP polymers. Additional details: i. CCl4. ii. NaI, acetone. iii. LAH, THF. iv. (1) Cl3SiH, (2) HNEt2.

An interesting aspect is pursued by Cuthbert et al. using a phosphine and sulfur rich soft polymer network P-139. The high density of soft P- and S-donor sites was used in the removal of transition metals from the reaction mixture, facilitating purification. The CapturePhos polymer was tested for the removal of Wilkinson’s catalyst after hydrogenation of styrene, and Grubb’s catalyst after a ring closing metathesis reaction, showing effectiveness in removal of Rh (99.9%) and Ru (98.8%), respectively. The Ru-loaded scavenger can be partially regenerated releasing up to 42% of the metal, although requires higher temperature and both PEt3 and tmeda.221

14.06.5 Concluding remarks The field of organophosphorus polymers has greatly developed over the past decades including many fascinating novel synthetic strategies and applications. Many new studies embraced heavier analogs and study possibilities to impart unprecedented polymer properties and open new avenues for new chemical transformations. Seeing new developments for molecular systems containing in particular heavier pnictogens in the area of small molecule activation, redox catalysis, and sensing paves the way for subsequent exploration of polymeric materials incorporating these molecular motifs. While many organophosphorus polymers exist, polymers of the heavier pnictogens are still vastly under represented, despite some very interesting examples demonstrating their value in many applications. With this prospect the field of organopnictogen polymers is flourishing and many new discoveries are expected.

Acknowledgments The authors would like to thank the Swedish research council (Vetenskapsrådet), the European COST action Smart Inorganic Polymers (SIPs, CM1302) for continuous support. Additional support from private foundations (Carl-Trygger stiftelse, Stiftelse Olle-Engkvist byggmästare, Wenner-Gren foundation) is greatly acknowledged.

226

Organophosphorus and Related Group 15 Polymers

References 1. 2. 3. 4. 5. 6. 7. 8. 9.

10. 11. 12. 13. 14. 15. 16. 17.

18.

19. 20. 21. 22. 23. 24.

25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36.

Kolos, W. Some Remarks on the Structure of NH3 and PH3 Molecules. J. Chem. Phys. 1955, 23 (8), 1554. https://doi.org/10.1063/1.1742363. Bartell, L. S.; Brockway, L. O. Electron Diffraction Study of the Structure of Trimethylphosphine. J. Chem. Phys. 1960, 32 (2), 512–515. https://doi.org/10.1063/1.1730726. Bent, H. A. Distribution of Atomic s Character in Molecules and its Chemical Implications. J. Chem. Educ. 1960, 37 (12), 616. https://doi.org/10.1021/ed037p616. Montgomery, C. D. Factors Affecting Energy Barriers for Pyramidal Inversion in Amines and Phosphines: A Computational Chemistry Lab Exercise. J. Chem. Educ. 2013, 90 (5), 661–664. https://doi.org/10.1021/ed300527t. Pauling, L. The Nature of the Chemical Bond. IV. The Energy of Single Bonds and the Relative Electronegativity of Atoms. J. Am. Chem. Soc. 1932, 54 (9), 3570–3582. https:// doi.org/10.1021/ja01348a011. Gilheany, D. G. No d Orbitals but Walsh Diagrams and Maybe Banana Bonds: Chemical Bonding in Phosphines, Phosphine Oxides, and Phosphonium Ylides. Chem. Rev. 1994, 94 (5), 1339–1374. https://doi.org/10.1021/cr00029a008. Chesnut, D. B.; Savin, A. The Electron Localization Function (ELF) Description of the PO Bond in Phosphine Oxide. J. Am. Chem. Soc. 1999, 121 (10), 2335–2336. https://doi. org/10.1021/ja984314m. Bryce, D. L.; Eichele, K.; Wasylishen, R. E. An 17O NMR and Quantum Chemical Study of Monoclinic and Orthorhombic Polymorphs of Triphenylphosphine Oxide. Inorg. Chem. 2003, 42 (17), 5085–5096. https://doi.org/10.1021/ic020706p. Honeyman, C. H.; Foucher, D. A.; Dahmen, F. Y.; Rulkens, R.; Lough, A. J.; Manners, I. Thermal Ring-Opening Polymerization (ROP) of Strained, Ring-Tilted Phosphorus-Bridged [1]Ferrocenophanes: Synthesis of Poly(Ferrocenylphosphines) and Poly(Ferrocenylphosphine Sulfides). Organometallics 1995, 14 (12), 5503–5512. https://doi.org/10.1021/ om00012a017. Ling, I.; Alias, Y.; Sobolev, A. N.; Raston, C. L. Hirshfeld Surface Analysis of Phosphonium Salts. CrystEngComm 2010, 12 (12), 4321. https://doi.org/10.1039/c002832k. Larrañaga, O.; Romero-Nieto, C.; de Cózar, A. Dismantling the Hyperconjugation of P-Conjugated Phosphorus Heterocycles. Chem. Eur. J. 2019, 25 (38), 9035–9044. https:// doi.org/10.1002/chem.201900225. Nyulászi, L. Aromaticity of Phosphorus Heterocycles. Chem. Rev. 2001, 101 (5), 1229–1246. https://doi.org/10.1021/cr990321x. Nyulászi, L.; von Ragué Schleyer, P. Hyperconjugative p-Aromaticity: How to Make Cyclopentadiene Aromatic. J. Am. Chem. Soc. 1999, 121 (29), 6872–6875. https://doi.org/ 10.1021/ja983113f. Dienes, Y.; Eggenstein, M.; Kárpáti, T.; Sutherland, T. C.; Nyulászi, L.; Baumgartner, T. Phosphorus-Based Heteropentacenes: Efficiently Tunable Materials for Organic n-Type Semiconductors. Chem. Eur. J. 2008, 14 (32), 9878–9889. https://doi.org/10.1002/chem.200801549. Jutzi, P. New Element-Carbon (p-p)p Bonds. Angew. Chem. Int. Ed. Eng. 1975, 14 (4), 232–245. https://doi.org/10.1002/anie.197502321. Nauroozi, D.; Orthaber, A. The Heavier Analogues of Alkenes: A Theoretical Comparison of Unsaturated Group 15/14 Systems: Heavier Analogues of Alkenes. Eur. J. Inorg. Chem. 2016, 2016 (5), 709–717. https://doi.org/10.1002/ejic.201500848. Green, J. P.; Gupta, A. K.; Orthaber, A. Effect of Arsenic Coordination State on the Structure, Aromaticity, and Optical Properties of Dithieno[3,2- b:20 ,30 - d]Arsoles: Effect of Arsenic Coordination State on the Structure, Aromaticity, and Optical Properties of Dithieno[3,2- b:20 ,30 - d]Arsoles. Eur. J. Inorg. Chem. 2019, 2019 (11–12), 1539–1543. https://doi.org/10.1002/ejic.201801169. Chalmers, B. A.; Meigh, C. B. E.; Nejman, P. S.; Bühl, M.; Lébl, T.; Woollins, J. D.; Slawin, A. M. Z.; Kilian, P. Geminally Substituted Tris(Acenaphthyl) and Bis(Acenaphthyl) Arsines, Stibines, and Bismuthine: A Structural and Nuclear Magnetic Resonance Investigation. Inorg. Chem. 2016, 55 (14), 7117–7125. https://doi.org/10.1021/acs. inorgchem.6b01079. Yang, M.; Hirai, M.; Gabbaï, F. P. Phosphonium–Stibonium and Bis-Stibonium Cations as Pnictogen-Bonding Catalysts for the Transfer Hydrogenation of Quinolines. Dalton Trans. 2019, 48 (20), 6685–6689. https://doi.org/10.1039/C9DT01357A. Marczenko, K. M.; Jee, S.; Chitnis, S. S. High Lewis Acidity at Planar, Trivalent, and Neutral Bismuth Centers. Organometallics 2020https://doi.org/10.1021/acs. organomet.0c00378. Wei, Y.; Chan, C. C.; Tian, J.; Jang, G. W.; Hsueh, K. F. Electrochemical Polymerization of Thiophenes in the Presence of Bithiophene or Terthiophene: Kinetics and Mechanism of the Polymerization. Chem. Mater. 1991, 3 (5), 888–897. https://doi.org/10.1021/cm00017a026. Morales Salazar, D.; Mijangos, E.; Pullen, S.; Gao, M.; Orthaber, A. Functional Small-Molecules & Polymers Containing P¼ C and As ¼ C Bonds as Hybrid p-Conjugated Materials. Chem. Commun. 2017, 53 (6), 1120–1123. https://doi.org/10.1039/C6CC08736A. Wells, J. A. L.; Shameem, M. A.; Gupta, A. K. K.; Orthaber, A. Core and Double Bond Functionalisation of Cyclopentadithiophene-Phosphaalkenes. Inorg. Chem. Front. 2020, 7, 4052–4061. Hay, C.; Fischmeister, C.; Hissler, M.; Toupet, L.; Réau, R. Electropolymerization of p-Conjugated Oligomers Containing Phosphole Cores and Terminal Thienyl Moieties: Optical and Electronic Properties. Angew. Chem. Int. Ed. Eng. 2000, 39 (10), 1812–1815. https://doi.org/10.1002/(sici)1521-3773(20000515)39:103.0.co;2-d. Lemau de Talance, V.; Hissler, M.; Zhang, L.-Z.; Karpati, T.; Nyulaszi, L.; Caras-Quintero, D.; Bäuerle, P.; Reau, R. Synthesis, Electronic Properties and Electropolymerisation of EDOT-Capped s3-Phospholes. Chem. Commun. 2008, 19, 2200–2202. https://doi.org/10.1039/b801335g. Hay, C.; Hissler, M.; Fischmeister, C.; Rault-Berthelot, J.; Toupet, L.; Nyulászi, L.; Réau, R. Phosphole-Containing p-Conjugated Systems: From Model Molecules to Polymer Films on Electrodes. Chem. Eur. J. 2001, 7 (19), 4222–4236. https://doi.org/10.1002/1521-3765(20011001)7:193.0.co;2-3. Hay, C.; Fave, C.; Hissler, M.; Rault-Berthelot, J.; Réau, R. Synthesis and Electronic Properties of Alternating a,a’-Thiophene-Phosphole Oligomers. Org. Lett. 2003, 5 (19), 3467–3470. https://doi.org/10.1021/ol0352658. Fadhel, O.; Benkö, Z.; Gras, M.; Deborde, V.; Joly, D.; Lescop, C.; Nyulászi, L.; Hissler, M.; Réau, R. 3,4-Dithiaphosphole and 3,30 ,4,40 -Tetrathia-1,10 -Biphosphole p-Conjugated Systems: S Makes the Impact. Chem. Eur. J. 2010, 16 (37), 11340–11356. https://doi.org/10.1002/chem.201001463. Fave, C.; Hissler, M.; Kárpáti, T.; Rault-Berthelot, J.; Deborde, V.; Toupet, L.; Nyulászi, L.; Réau, R. Connecting p-Chromophores by s-P–P Bonds: New Type of Assemblies Exhibiting s–p-Conjugation. J. Am. Chem. Soc. 2004, 126 (19), 6058–6063. https://doi.org/10.1021/ja0317067. Imoto, H.; Yamazawa, C.; Hayashi, S.; Aono, M.; Naka, K. Electropolymerization of Dithieno[3,2- b:20 ,30 - d]Arsole. ChemElectroChem 2018, 5 (22), 3357–3360. https://doi.org/ 10.1002/celc.201801069. Chen, R.-F.; Zhu, R.; Fan, Q.-L.; Huang, W. Synthesis, Structure, and Optoelectronic Properties of Phosphafluorene Copolymers. Org. Lett. 2008, 10 (13), 2913–2916. https:// doi.org/10.1021/ol801068m. Durben, S.; Nickel, D.; Krüger, R. A.; Sutherland, T. C.; Baumgartner, T. Toward Low-Band Gap Dithienophosphole Copolymers for an Application in Organic Solar Cells. J. Polym. Sci. Part A Polym. Chem. 2008, 46 (24), 8179–8190. https://doi.org/10.1002/pola.23115. Stolar, M.; Baumgartner, T. Synthesis and Unexpected Halochromism of Carbazole-Functionalized Dithienophospholes. New J. Chem. 2012, 36 (5), 1153–1160. https://doi. org/10.1039/c2nj40022g. Durben, S.; Dienes, Y.; Baumgartner, T. Cationic Dithieno[3,2-b:20 ,30 -d]Phospholes: A New Building Block for Luminescent, Conjugated Polyelectrolytes. Org. Lett. 2006, 8 (25), 5893–5896. https://doi.org/10.1021/ol062580w. Kato, T.; Imoto, H.; Tanaka, S.; Ishidoshiro, M.; Naka, K. Facile Synthesis and Properties of Dithieno[3,2-b:20 ,30 -d]Arsoles. Dalton Trans. 2016, 45 (28), 11338–11345. https:// doi.org/10.1039/c6dt02034h. Fell, V. H. K.; Mikosch, A.; Steppert, A.-K.; Ogieglo, W.; Senol, E.; Canneson, D.; Bayer, M.; Schoenebeck, F.; Greilich, A.; Kuehne, A. J. C. Synthesis and Optical Characterization of Hybrid Organic–Inorganic Heterofluorene Polymers. Macromolecules 2017, 50 (6), 2338–2343. https://doi.org/10.1021/acs.macromol.6b02611.

Organophosphorus and Related Group 15 Polymers

227

37. Wang, J.; Xu, X.; Tian, Y.; Yao, C.; Liu, R.; Li, L. A Novel Blue Fluorescent Polymer for Solution-Processed Fluorescent-Phosphorescent Hybrid WOLEDs. J. Mater. Chem. C 2015, 3, 2856–2864. https://doi.org/10.1039/c4tc02899f. 38. Wang, J.; Xu, X.; Peng, J.; Yao, C. High Triplet, Bipolar Polymeric Hosts for Highly Efficient Solution-Processed Blue Phosphorescent Polymer Light-Emitting Diodes. Org. Electron. 2017, 43, 1–8. https://doi.org/10.1016/j.orgel.2017.01.002. 39. Baumgartner, T.; Bergmans, W.; Kárpáti, T.; Neumann, T.; Nieger, M.; Nyulászi, L. From Model Compounds to Extended P-Conjugated Systems: Synthesis and Properties of Dithieno[3,2-b:20 ,30 -d]Phospholes. Chem. Eur. J. 2005, 11 (16), 4687–4699. https://doi.org/10.1002/chem.200500152. 40. Green, J. P.; Han, Y.; Kilmurray, R.; McLachlan, M. A.; Anthopoulos, T. D.; Heeney, M. An Air-Stable Semiconducting Polymer Containing Dithieno[3,2- b:20 ,30 - d]Arsole. Angew. Chem. Int. Ed. 2016, 55 (25), 7148–7151. https://doi.org/10.1002/anie.201602491. 41. Green, J. P.; Cha, H.; Shahid, M.; Creamer, A.; Durrant, J. R.; Heeney, M. Dithieno[3,2-b:20 ,30 -d]Arsole-Containing Conjugated Polymers in Organic Photovoltaic Devices. Dalton Trans. 2019, 48 (20), 6676–6679. https://doi.org/10.1039/C9DT01496A. 42. Cao, H.; Brettell-Adams, I. A.; Qu, F.; Rupar, P. A. Bridged Difurans: Stabilizing Furan with p-Block Elements. Organometallics 2017, 36 (14), 2565–2572. https://doi.org/ 10.1021/acs.organomet.7b00135. 43. Matano, Y.; Ohkubo, H.; Miyata, T.; Watanabe, Y.; Hayashi, Y.; Umeyama, T.; Imahori, H. Phosphole- and Benzodithiophene-Based Copolymers: Synthesis and Application in Organic Photovoltaics: Phosphole- and Benzodithiophene-Based Copolymers. Eur. J. Inorg. Chem. 2014, 2014 (10), 1620–1624. https://doi.org/10.1002/ejic.201301132. 44. Bui, T. T. T.; Park, S.; Hoang, Q. V.; Song, C. E.; Lee, S. K.; Lee, J.-C.; Moon, S.-J.; Shin, W. S. Synthesis and Characterization of Medium Band Gap Polymers with Phosphole [3,2-b:4,5-B0 ]Dithiophene Oxide as Acceptor Unit and Their Application for Polymer Photovoltaic Devices. Synth. Met. 2016, 215, 235–242. https://doi.org/10.1016/j. synthmet.2016.03.001. 45. Park, K. H.; Kim, Y. J.; Lee, G. B.; An, T. K.; Park, C. E.; Kwon, S.-K.; Kim, Y.-H. Recently Advanced Polymer Materials Containing Dithieno[3,2-b:20 ,30 -d]Phosphole Oxide for Efficient Charge Transfer in High-Performance Solar Cells. Adv. Funct. Mater. 2015, 25 (26), 3991–3997. https://doi.org/10.1002/adfm.201500934. 46. Kim, Y. J.; Kim, M.-J.; An, T. K.; Kim, Y.-H.; Park, C. E. A New Multi-Functional Conjugated Polymer for Use in High-Performance Bulk Heterojunction Solar Cells. Chem. Commun. 2015, 51 (58), 11572–11575. https://doi.org/10.1039/C5CC03815D. 47. Cao, H.; Bauer, N.; Bi, S.; Li, D.; You, W.; Rupar, P. A. Post-Polymerization Modification of Phosphorus Containing Conjugated Copolymers. Eur. Polym. J. 2018, 104, 157–163. https://doi.org/10.1016/j.eurpolymj.2018.05.009. 48. Takeda, Y.; Hatanaka, K.; Nishida, T.; Minakata, S. Thieno[3,4-c]Phosphole-4,6-Dione: A Versatile Building Block for Phosphorus-Containing Functional p-Conjugated Systems. Chem. Eur. J. 2016, 22 (30), 10360–10364. https://doi.org/10.1002/chem.201602392. 49. Jiang, X.; Zhao, W.; Wang, W.; Zhang, F.; Zhuang, X.; Han, S.; Feng, X. One-Pot Approach to Pd-Loaded Porous Polymers with Properties Tunable by the Oxidation State of the Phosphorus Core. Polym. Chem. 2015, 6 (35), 6351–6357. https://doi.org/10.1039/C5PY00576K. 50. Ding, Z.-C.; Li, C.-Y.; Chen, J.-J.; Zeng, J.-H.; Tang, H.-T.; Ding, Y.-J.; Zhan, Z.-P. Palladium/Phosphorus-Doped Porous Organic Polymer as Recyclable Chemoselective and Efficient Hydrogenation Catalyst under Ambient Conditions. Adv. Synth. Catal. 2017, 359 (13), 2280–2287. https://doi.org/10.1002/adsc.201700374. 51. Geng, X.-L.; Ott, S. Acetylene-Expanded Dendralene Segments with Exotopic Phosphaalkene Units. Chem. Eur. J. 2011, 17 (43), 12153–12162. https://doi.org/10.1002/ chem.201101358. 52. Geng, X.-L.; Hu, Q.; Schäfer, B.; Ott, S. Phosphaalkenes in p-Conjugation With Acetylenic Arenes. Org. Lett. 2010, 12 (4), 692–695. https://doi.org/10.1021/ol902688n. 53. Morisaki, Y.; Aiki, Y.; Chujo, Y. Synthesis and Properties of First Well-Defined Phosphole-Containing p-Conjugated Polymers. Macromolecules 2003, 36 (8), 2594–2597. https://doi.org/10.1021/ma021095n. 54. Morisaki, Y.; Ouchi, Y.; Tsurui, K.; Chujo, Y. Synthesis of the Optically Active Polymer Consisting of Chiral Phosphorus Atoms and P-Phenylene-Ethynylene Units. Polym. Bull. 2007, 58 (4), 665–671. https://doi.org/10.1007/s00289-006-0703-3. 55. Morisaki, Y.; Suzuki, K.; Imoto, H.; Chujo, Y. Synthesis of Optically Active Polymer with P-Stereogenic Phosphine Units. Macromol. Rapid Commun. 2010, 31 (19), 1719–1724. https://doi.org/10.1002/marc.201000237. 56. Qiao, S.; Huang, W.; Du, Z.; Chen, X.; Shieh, F.-K.; Yang, R. Phosphine Oxide-Based Conjugated Microporous Polymers with Excellent CO2 Capture Properties. New J. Chem. 2015, 39 (1), 136–141. https://doi.org/10.1039/C4NJ01477D. 57. Ghassemi, H.; McGrath, J. E. Synthesis of Poly(Arylene Phosphine Oxide) by Nickel-Catalysed Coupling Polymerization. Polymer 1997, 38 (12), 3139–3143. https://doi.org/ 10.1016/S0032-3861(96)01049-X. 58. Rusch-Salazar, L. A.; Sheares, V. V. A Series of New High-Performance Materials Based on Poly[4’-Fluorophenyl-Bis(4-Phenyl)Phosphine Oxide]. J. Polym. Sci. Part A Polym. Chem. 2003, 41 (14), 2277–2287. https://doi.org/10.1002/pola.10745. 59. Zhang, Q.; Zhang, S.; Li, S. Novel Functional Organic Network Containing Quaternary Phosphonium and Tertiary Phosphorus. Macromolecules 2012, 45 (7), 2981–2988. https://doi.org/10.1021/ma300278d. 60. Xing, G.; Yan, T.; Das, S.; Ye, L.; Ye, K. Charged Porous Organic Frameworks Bearing Heteroatoms with Enhanced Isosteric Enthalpies of Gas Adsorption. RSC Adv. 2018, 8 (36), 20434–20439. https://doi.org/10.1039/C7RA12860F. 61. Hladysh, S.; Bondarev, D.; Svoboda, J.; Vohlídal, J.; Vrbata, D.; Zedník, J. Novel Conjugated Polyelectrolytes Based on Polythiophene Bearing Phosphonium Side Groups. Eur. Polym. J. 2016, 83, 367–376. https://doi.org/10.1016/j.eurpolymj.2016.07.019. 62. Hladysh, S.; Murmiliuk, A.; Vohlídal, J.; Havlícek, D.; Sedlarˇík, V.; Štepánek, M.; Zedník, J. Combination of Phosphonium and Ammonium Pendant Groups in Cationic Conjugated Polyelectrolytes Based on Regioregular Poly(3-Hexylthiophene) Polymer Chains. Eur. Polym. J. 2018, 100, 200–208. https://doi.org/10.1016/j.eurpolymj.2018.01.029. 63. Parke, S. M.; Tanaka, S.; Yu, H.; Hupf, E.; Ferguson, M. J.; Zhou, Y.; Naka, K.; Rivard, E. Highly Fluorescent Benzophosphole Oxide Block-Copolymer Micelles. Macromolecules 2019, 52 (19), 7477–7488. https://doi.org/10.1021/acs.macromol.9b01661. 64. Yang, Y.; Wang, T.; Jing, X.; Zhu, G. Phosphine-Based Porous Aromatic Frameworks for Gold Nanoparticle Immobilization With Superior Catalytic Activities. J. Mater. Chem. A 2019, 7 (16), 10004–10009. https://doi.org/10.1039/C8TA12099D. 65. Yang, X.; Wen, Y.; Chumanov, G.; Smith, R. C. A New Route to Phosphonium Polymer Network Solids via Cyclotrimerization. J. Polym. Sci. Part A Polym. Chem. 2017, 55 (9), 1620–1625. https://doi.org/10.1002/pola.28535. 66. Shao, S.; Ding, J.; Ye, T.; Xie, Z.; Wang, L.; Jing, X.; Wang, F. A Novel, Bipolar Polymeric Host for Highly Efficient Blue Electrophosphorescence: A Non-Conjugated Poly(Aryl Ether) Containing Triphenylphosphine Oxide Units in the Electron-Transporting Main Chain and Carbazole Units in Hole-Transporting Side Chains. Adv. Mater. 2011, 23 (31), 3570–3574. https://doi.org/10.1002/adma.201101074. 67. Yang, Z.; Chen, H.; Li, B.; Guo, W.; Jie, K.; Sun, Y.; Jiang, D.; Popovs, I.; Dai, S. Topotactic Synthesis of Phosphabenzene-Functionalized Porous Organic Polymers: Efficient Ligands in CO 2 Conversion. Angew. Chem. Int. Ed. 2019, 58 (39), 13763–13767. https://doi.org/10.1002/anie.201907015. 68. Petrucci-Samija, M.; Guillemette, V.; Dasgupta, M.; Kakkar, A. K. A New Divergent Route to the Synthesis of Organophosphine and Metallodendrimers via Simple Acid–Base Hydrolytic Chemistry. J. Am. Chem. Soc. 1999, 121 (9), 1968–1969. https://doi.org/10.1021/ja983132p. 69. Reetz, M. T.; Lohmer, G.; Schwickardi, R. Systhesis and Catalytic Activity of Dendritic Diphosphane Metal Complexes. Angew. Chem. Int. Ed. Eng. 1997, 36 (1314), 1526–1529. https://doi.org/10.1002/anie.199715261. 70. Alonso, E.; Astruc, D. Introduction of the Cluster Fragment Ru3 (CO)11 at the Periphery of Phosphine Dendrimers Catalyzed by the Electron-Reservoir Complex [FeICp(C6 Me6)]. J. Am. Chem. Soc. 2000, 122 (13), 3222–3223. https://doi.org/10.1021/ja994332j. 71. Mao, S. S. H.; Tilley, T. D. A Versatile, Transition-Metal Mediated Route to Blue-Light-Emitting Polymers with Chemically Tunable Luminescent Properties. Macromolecules 1997, 30 (18), 5566–5569. https://doi.org/10.1021/ma9701402. 72. Matsumura, Y.; Ueda, M.; Fukuda, K.; Fukui, K.; Takase, I.; Nishiyama, H.; Inagi, S.; Tomita, I. Synthesis of p-Conjugated Polymers Containing Phosphole Units in the Main Chain by Reaction of an Organometallic Polymer Having a Titanacyclopentadiene Unit. ACS Macro Lett. 2015, 4 (1), 124–127. https://doi.org/10.1021/mz500693r.

228

Organophosphorus and Related Group 15 Polymers

73. Matsumura, Y.; Nishiyama, H.; Inagi, S.; Tomita, I. Synthesis of P-Conjugated Polymer Containing Both Thiophene and Phosphole Sulfide in the Main Chain by Simultaneous Post-Element Transformation of Organotitanium Polymer. J. Polym. Sci. Part A Polym. Chem. 2019, 57 (24), 2519–2525. https://doi.org/10.1002/pola.29522. 74. Matsumura, Y.; Ishidoshiro, M.; Irie, Y.; Imoto, H.; Naka, K.; Tanaka, K.; Inagi, S.; Tomita, I. Arsole-Containing p-Conjugated Polymer by the Post-Element-Transformation Technique. Angew. Chem. Int. Ed. 2016, 55 (48), 15040–15043. https://doi.org/10.1002/anie.201608404. 75. Morisaki, Y.; Ohashi, K.; Na, H.-S.; Chujo, Y. First Synthesis of the Bismole-Containing Conjugated Polymer. J. Polym. Sci. Part A Polym. Chem. 2006, 44 (16), 4857–4863. https://doi.org/10.1002/pola.21590. 76. Stolar, M.; Reus, C.; Baumgartner, T. Xylene-Bridged Phosphaviologen Oligomers and Polymers as High-Performance Electrode-Modifiers for Li-Ion Batteries. Adv. Energy Mater. 2016, 6 (20), 1600944. https://doi.org/10.1002/aenm.201600944. 77. Striepe, L.; Vespa, M.; Baumgartner, T. Synthesis and Properties of Electron Accepting Star-Shaped Phosphaviologen Oligomers. Org. Chem. Front. 2017, 4 (5), 717–723. https://doi.org/10.1039/C6QO00872K. 78. Yang, Z.; Yu, B.; Zhang, H.; Zhao, Y.; Chen, Y.; Ma, Z.; Ji, G.; Gao, X.; Han, B.; Liu, Z. Metalated Mesoporous Poly(Triphenylphosphine) with Azo Functionality: Efficient Catalysts for CO2 Conversion. ACS Catal. 2016, 6 (2), 1268–1273. https://doi.org/10.1021/acscatal.5b02583. 79. Tao, R.; Shen, X.; Hu, Y.; Kang, K.; Zheng, Y.; Luo, S.; Yang, S.; Li, W.; Lu, S.; Jin, Y.; Qiu, L.; Zhang, W. Phosphine-Based Covalent Organic Framework for the Controlled Synthesis of Broad-Scope Ultrafine Nanoparticles. Small 2020, 16 (8), 1906005. https://doi.org/10.1002/smll.201906005. 80. Imoto, H.; Morisaki, Y.; Chujo, Y. Synthesis and Coordination Behaviors of P-Stereogenic Polymers. Chem. Commun. 2010, 46 (40), 7542. https://doi.org/10.1039/ c0cc03120h. 81. Orthaber, A.; Soto, A. P. Synthesis of Polymers Containing Group 15 Elements. In Smart Inorganic Polymers; Hey-Hawkins, E., Hissler, M., Eds.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2019;; pp 85–114. https://doi.org/10.1002/9783527819140.ch4. 82. Sun, Q.; Dai, Z.; Liu, X.; Sheng, N.; Deng, F.; Meng, X.; Xiao, F.-S. Highly Efficient Heterogeneous Hydroformylation over Rh-Metalated Porous Organic Polymers: Synergistic Effect of High Ligand Concentration and Flexible Framework. J. Am. Chem. Soc. 2015, 137 (15), 5204–5209. https://doi.org/10.1021/jacs.5b02122. 83. Lau, K. C. Y.; He, H. S.; Chiu, P.; Toy, P. H. Polystyrene-Supported Triphenylarsine Reagents and their Use in Suzuki Cross-Coupling Reactions. J. Comb. Chem. 2004, 6 (6), 955–960. https://doi.org/10.1021/cc049907+. 84. Song He, H.; Zhang, C.; Ng, C. K.-W.; Toy, P. H. Polystyrene-Supported Triphenylarsines: Useful Ligands in Palladium-Catalyzed Aryl Halide Homocoupling Reactions and a Catalyst for Alkene Epoxidation Using Hydrogen Peroxide. Tetrahedron 2005, 61 (51), 12053–12057. https://doi.org/10.1016/j.tet.2005.07.108. 85. Choi, M. K. W.; He, H. S.; Toy, P. H. Direct Radical Polymerization of 4-Styryldiphenylphosphine: Preparation of Cross-Linked and Non-Cross-Linked Triphenylphosphine-Containing Polystyrene Polymers. J. Organomet. Chem. 2003, 68 (25), 9831–9834. https://doi.org/10.1021/jo035226+. 86. Licea-Claveríe, A.; Rivero, I. A.; Morales, M. S.; Moreno, M. G. Synthesis, Characterization and Use of Reactive Polymers With Diisopropylphenylphosphine Groups. Polym. Bull. 1997, 39 (5), 551–557. https://doi.org/10.1007/s002890050185. 87. Farrall, M. J.; Frechet, J. M. J. Bromination and Lithiation: Two Important Steps in the Functionalization of Polystyrene Resins. J. Organomet. Chem. 1976, 41 (24), 3877–3882. https://doi.org/10.1021/jo00886a023. 88. Bernard, M.; Ford, W. T. Wittig Reagents Bound to Crosslinked Polystyrenes. J. Organomet. Chem. 1983, 48 (3), 326–332. https://doi.org/10.1021/jo00151a010. 89. Harrison, C. R.; Hodge, P.; Hunt, B. J.; Khoshdel, E.; Richardson, G. Preparation of Alkyl Chlorides, Acid Chlorides, and Amides Using Polymer-Supported Phosphines and Carbon Tetrachloride: Mechanism of these Reactions. J. Organomet. Chem. 1983, 48 (21), 3721–3728. https://doi.org/10.1021/jo00169a022. 90. Relles, H. M.; Schluenz, R. W. Chemical Transformations with Regenerable, Polymer-Supported Trisubstituted Phosphine Dichlorides. Efficacious Incorporation of Phosphorus Reagents on Polymer Supports. J. Am. Chem. Soc. 1974, 96 (20), 6469–6475. https://doi.org/10.1021/ja00827a034. 91. Schweizer, S.; Becht, J.-M.; Le Drian, C. Development of Efficient and Reusable Diarylphosphinopolystyrene-Supported Palladium Catalysts for C–C Bond Forming Cross-Coupling Reactions. Adv. Synth. Catal. 2007, 349 (7), 1150–1158. https://doi.org/10.1002/adsc.200600503. 92. Schweizer, S.; Becht, J.-M.; Le Drian, C. Highly Efficient Reusable Polymer-Supported Pd Catalysts of General Use for the Suzuki Reaction. Tetrahedron 2010, 66 (3), 765–772. https://doi.org/10.1016/j.tet.2009.11.050. 93. Schweizer, S.; Becht, J.-M.; Le Drian, C. Highly Efficient and Reusable Supported Pd Catalysts for Suzuki–Miyaura Reactions of Aryl Chlorides {. Org. Lett. 2007, 9 (19), 3777–3780. https://doi.org/10.1021/ol701460z. 94. Krause, H. W. Chiral Rhodium Complexes of the Dimenthylphosphine Group Attached to Polystyrene for Asymmetric Hydrogenation of Z-a-Acetamidocinnamic Acid. React. Kinet. Catal. Lett. 1979, 10 (3), 243–246. https://doi.org/10.1007/BF02068991. 95. Yu, H. W.; Tong, Q. S.; Peng, Y. R.; Jia, L.; Shi, J. C.; Jin, Z. L. Preparation of Polymer-Supported Phosphine From Ferrocene for Palladium-Catalyzed Suzuki–Miyaura Cross-Coupling Reactions. Chin. Chem. Lett. 2007, 18 (1), 37–40. https://doi.org/10.1016/j.cclet.2006.11.030. 96. Lipshutz, B. H.; Blomgren, P. A. Efficient Scavenging of Ph3P and Ph3PO with High-Loading Merrifield Resin. Org. Lett. 2001, 3 (12), 1869–1871. https://doi.org/10.1021/ ol0159219. 97. Garrett, C. E.; Prasad, K. The Art of Meeting Palladium Specifications in Active Pharmaceutical Ingredients Produced by Pd-Catalyzed Reactions. Adv. Synth. Catal. 2004, 346 (8), 889–900. https://doi.org/10.1002/adsc.200404071. 98. Rabiee Kenaree, A.; Gilroy, J. B. A Phosphine-Based Heterotrimetallic (M ¼ Fe, Ru, W) Homopolymer. Organometallics 2017, 36 (14), 2483–2486. https://doi.org/10.1021/ acs.organomet.7b00085. 99. Romero-Nieto, C.; Durben, S.; Kormos, I. M.; Baumgartner, T. Simple and Efficient Generation of White Light Emission From Organophosphorus Building Blocks. Adv. Funct. Mater. 2009, 19 (22), 3625–3631. 100. Chiefari, J.; Chong, Y. K.(.B.).; Ercole, F.; Krstina, J.; Jeffery, J.; Le, T. P. T.; Mayadunne, R. T. A.; Meijs, G. F.; Moad, C. L.; Moad, G.; Rizzardo, E.; Thang, S. H. Living Free-Radical Polymerization by Reversible Addition–Fragmentation Chain Transfer: The RAFT Process. Macromolecules 1998, 31 (16), 5559–5562. https://doi.org/10.1021/ ma9804951. 101. Perrier, S. 50th Anniversary Perspective: RAFT Polymerization—A User Guide. Macromolecules 2017, 50 (19), 7433–7447. https://doi.org/10.1021/acs.macromol.7b00767. 102. Keddie, D. J.; Moad, G.; Rizzardo, E.; Thang, S. H. RAFT Agent Design and Synthesis. Macromolecules 2012, 45 (13), 5321–5342. https://doi.org/10.1021/ma300410v. 103. Wang, R.; Lowe, A. B. RAFT Polymerization of Styrenic-Based Phosphonium Monomers and a New Family of Well-Defined Statistical and Block Polyampholytes. J. Polym. Sci. Part A Polym. Chem. 2007, 45 (12), 2468–2483. https://doi.org/10.1002/pola.22009. 104. Cheng, S.; Beyer, F. L.; Mather, B. D.; Moore, R. B.; Long, T. E. Phosphonium-Containing ABA Triblock Copolymers: Controlled Free Radical Polymerization of Phosphonium Ionic Liquids. Macromolecules 2011, 44 (16), 6509–6517. https://doi.org/10.1021/ma200829h. 105. Schultz, A. R.; Fahs, G. B.; Jangu, C.; Chen, M.; Moore, R. B.; Long, T. E. Phosphonium-Containing Diblock Copolymers From Living Anionic Polymerization of 4-Diphenylphosphino Styrene. Chem. Commun. 2016, 52 (5), 950–953. https://doi.org/10.1039/C5CC08699J. 106. Schultz, A. R.; Chen, M.; Fahs, G. B.; Moore, R. B.; Long, T. E. Living Anionic Polymerization of 4-Diphenylphosphino Styrene for ABC Triblock Copolymers: Synthesis of Poly(Sb-I-b-DPPS) Triblock Copolymers. Polym. Int. 2017, 66 (1), 52–58. https://doi.org/10.1002/pi.5253. 107. Leca, F.; Lescop, C.; Rodriguez-Sanz, E.; Costuas, K.; Halet, J.-F.; Réau, R. Bridging Phosphanes: Exotic or Versatile Binucleating Ligands?Angew. Chem. Int. Ed. 2005, 44 (28), 4362–4365. https://doi.org/10.1002/anie.200500529. 108. Nohra, B.; Graule, S.; Lescop, C.; Réau, R. Mimicking [2,2]Paracyclophane Topology: Molecular Clips for the Coordination-Driven Cofacial Assembly of p-Conjugated Systems. J. Am. Chem. Soc. 2006, 128 (11), 3520–3521. https://doi.org/10.1021/ja058617+. 109. Agou, T.; Sébastian, M.; Lescop, C.; Réau, R. Folding of a Supramolecular Framework Based on a Tetrametallic Clip Driven by P–p Interactions. Inorg. Chem. 2011, 50 (8), 3183–3185. https://doi.org/10.1021/ic200268c.

Organophosphorus and Related Group 15 Polymers

229

110. Vreshch, V.; Shen, W.; Nohra, B.; Yip, S.-K.; Yam, V. W.-W.; Lescop, C.; Réau, R. Aurophilicity Versus Mercurophilicity: Impact of D10-D10 Metallophilic Interactions on the Structure of Metal-Rich Supramolecular Assemblies. Chem. Eur. J. 2012, 18 (2), 466–477. https://doi.org/10.1002/chem.201102389. 111. Vitvarová, T.; Svoboda, J.; Hissler, M.; Vohlídal, J. Conjugated Metallo-Supramolecular Polymers Containing a Phosphole Unit. Organometallics 2017, 36 (4), 777–786. https:// doi.org/10.1021/acs.organomet.6b00822. 112. Fleischmann, M.; Dütsch, L.; Elsayed Moussa, M.; Balázs, G.; Kremer, W.; Lescop, C.; Scheer, M. Self-Assembly of Reactive Linear Cu3 Building Blocks for Supramolecular Coordination Chemistry and Their Reactivity toward En Ligand Complexes. Inorg. Chem. 2016, 55 (6), 2840–2854. https://doi.org/10.1021/acs.inorgchem.5b02641. 113. Fleischmann, M.; Dütsch, L.; Moussa, M. E.; Schindler, A.; Balázs, G.; Lescop, C.; Scheer, M. Organometallic Polyphosphorus and -Arsenic Ligands as Linkers between Pre-Assembled Linear Cu I Fragments. Chem. Commun. 2015, 51 (14), 2893–2895. https://doi.org/10.1039/C4CC09845E. 114. Heindl, C.; Peresypkina, E. V.; Virovets, A. V.; Komarov, V. Y.; Scheer, M. 1,2,4-Triphospholyl Anions—Versatile Building Blocks for the Formation of 1D, 2D and 3D Assemblies. Dalton Trans. 2015, 44 (22), 10245–10252. https://doi.org/10.1039/C5DT01230A. 115. Heindl, C.; Kuntz, A.; Peresypkina, E. V.; Virovets, A. V.; Zabel, M.; Lüdeker, D.; Brunklaus, G.; Scheer, M. Unexpected Fragmentations of Triphosphaferrocene—Formation of Supramolecular Assemblies Containing the (1,2,4-P3 C2 Mes2)− Ligand. Dalton Trans. 2015, 44 (14), 6502–6509. https://doi.org/10.1039/C5DT00303B. 116. Rummel, E.-M.; Eckhardt, M.; Bodensteiner, M.; Peresypkina, E. V.; Kremer, W.; Gröger, C.; Scheer, M. Formation of 1,3-Diphosphacyclobutadiene Complexes From Phosphaalkynes and Their Coordination Behavior: 1,3-Diphosphacyclobutadiene Complexes. Eur. J. Inorg. Chem. 2014, 2014 (10), 1625–1637. https://doi.org/10.1002/ ejic.201301299. 117. Heindl, C.; Reisinger, S.; Schwarzmaier, C.; Rummel, L.; Virovets, A. V.; Peresypkina, E. V.; Scheer, M. The First Coordination Polymers Based on 1,3-Diphosphaferrocenes and 1,10 ,2,30 ,4-Pentaphosphaferrocenes: Coordination Polymers Based on Di- and Pentaphosphaferrocenes. Eur. J. Inorg. Chem. 2016, 2016 (5), 743–753. https://doi.org/ 10.1002/ejic.201500873. 118. Vogel, U.; Nixon, J. F.; Scheer, M. Copper(I) Mediated Oligomerisation of a Phosphaalkyne. Chem. Commun. 2007, 47, 5055. https://doi.org/10.1039/b712161j. 119. Humphrey, S. M.; Allan, P. K.; Oungoulian, S. E.; Ironside, M. S.; Wise, E. R. Metal–Organophosphine and Metal–Organophosphonium Frameworks with Layered Honeycomb-like Structures. Dalton Trans. 2009, 13, 2298. https://doi.org/10.1039/b820038f. 120. Bezrukov, A. A.; Törnroos, K. W.; Dietzel, P. D. C. Modification of Network and Pore Dimensionality in Metal–Organic Frameworks Containing a Secondary Phosphine Functionality. Cryst. Growth Des. 2017, 17 (6), 3257–3266. https://doi.org/10.1021/acs.cgd.7b00243. 121. Dunning, S. G.; Reynolds, J. E.; Walsh, K. M.; Kristek, D. J.; Lynch, V. M.; Kunal, P.; Humphrey, S. M. Direct, One-Pot Syntheses of MOFs Decorated with Low-Valent Metal-Phosphine Complexes. Organometallics 2019, 38 (18), 3406–3411. https://doi.org/10.1021/acs.organomet.9b00319. 122. Bezrukov, A. A.; Dietzel, P. D. C. A Permanently Porous Yttrium–Organic Framework Based on an Extended Tridentate Phosphine Containing Linker. Inorg. Chem. 2017, 56 (21), 12830–12838. https://doi.org/10.1021/acs.inorgchem.7b01574. 123. Yan, W.; Li, S.; Yang, T.; Xia, Y.; Zhang, X.; Wang, C.; Yan, Z.; Deng, F.; Zhou, Q.; Deng, H. Molecular Vises for Precisely Positioning Ligands near Catalytic Metal Centers in Metal–Organic Frameworks. J. Am. Chem. Soc. 2020, 142 (38), 16182–16187. https://doi.org/10.1021/jacs.0c07450. 124. Sikma, R. E.; Kunal, P.; Dunning, S. G.; Reynolds, J. E.; Lee, J. S.; Chang, J.-S.; Humphrey, S. M. Organoarsine Metal–Organic Framework with Cis-Diarsine Pockets for the Installation of Uniquely Confined Metal Complexes. J. Am. Chem. Soc. 2018, 140 (31), 9806–9809. https://doi.org/10.1021/jacs.8b05644. 125. He, J.; Bohnsack, A. M.; Waggoner, N. W.; Dunning, S. G.; Lynch, V. M.; Kaska, W. C.; Humphrey, S. M. 1-D and 2-D Phosphine Coordination Materials Based on a Palladium(II) PCP Pincer Metalloligand. Polyhedron 2018, 143, 149–156. https://doi.org/10.1016/j.poly.2017.09.025. 126. Bohnsack, A. M.; Ibarra, I. A.; Bakhmutov, V. I.; Lynch, V. M.; Humphrey, S. M. Rational Design of Porous Coordination Polymers Based on Bis(Phosphine)MCl 2 Complexes that Exhibit High-Temperature H2 Sorption and Chemical Reactivity. J. Am. Chem. Soc. 2013, 135 (43), 16038–16041. https://doi.org/10.1021/ja408508m. 127. Falkowski, J. M.; Sawano, T.; Zhang, T.; Tsun, G.; Chen, Y.; Lockard, J. V.; Lin, W. Privileged Phosphine-Based Metal–Organic Frameworks for Broad-Scope Asymmetric Catalysis. J. Am. Chem. Soc. 2014, 136 (14), 5213–5216. https://doi.org/10.1021/ja500090y. 128. Lucht, B. L.; St. Onge, N. O. Synthesis and Characterization of Poly(p-Phenylenephosphine)s. Chem. Commun. 2000, 21, 2097–2098. https://doi.org/10.1039/b003989f. 129. Jin, Z.; Lucht, B. L. Transition Metal Mediated Routes to Poly(Arylphosphine)s: Investigation of Novel Phosphorus Containing Conjugated Polymers. J. Organomet. Chem. 2002, 653 (1–2), 167–176. https://doi.org/10.1016/S0022-328X(02)01171-3. 130. Jin, Z.; Lucht, B. L. Poly-p-Phenylene Phosphine/Polyaniline Alternating Copolymers: Electronic Delocalization through Phosphorus. J. Am. Chem. Soc. 2005, 127 (15), 5586–5595. https://doi.org/10.1021/ja043358u. 131. Naka, K.; Umeyama, T.; Nakahashi, A.; Chujo, Y. Synthesis of Poly(Vinylene–Phosphine)s: Ring-Collapsed Radical Alternating Copolymerization of Methyl-Substituted Cyclooligophosphine With Acetylenic Compounds. Macromolecules 2007, 40 (14), 4854–4858. https://doi.org/10.1021/ma0702795. 132. Naka, K.; Umeyama, T.; Chujo, Y. Synthesis of Poly(Vinylene-Arsine)s: Alternating Radical Copolymerization of Arsenic Atomic Biradical Equivalent and Phenylacetylene. J. Am. Chem. Soc. 2002, 124 (23), 6600–6603. https://doi.org/10.1021/ja026578l. 133. Naka, K.; Nakahashi, A.; Chujo, Y. Ring-Collapsed Radical Alternating Copolymerization of Phenyl-Substituted Cyclooligostibine and Acetylenic Compounds. Macromolecules 2006, 39 (24), 8257–8262. https://doi.org/10.1021/ma061220l. 134. Naka, K.; Nakahashi, A.; Chujo, Y. Periodic Terpolymerization of Cyclooligoarsine, Cyclooligostibine, and Acetylenic Compound. Macromolecules 2007, 40 (5), 1372–1376. https://doi.org/10.1021/ma0622332. 135. Umeyama, T.; Naka, K.; Nakahashi, A.; Chujo, Y. Radical Copolymerization of Acetylenic Compounds With Phenyl-Substituted Cyclooligoarsine: Substituent Effect and Optical Properties. Macromolecules 2004, 37 (4), 1271–1275. https://doi.org/10.1021/ma035282c. 136. Rawe, B. W.; Scott, M. R.; Brown, C. M.; MacKenzie, H. K.; Gates, D. P. Poly(p-Phenylenediethynylene Phosphine)s and Related p-Conjugated Phosphine–Diyne Polymers: Synthesis, Characterization and Photophysical Properties. Macromolecules 2017, 50 (22), 8916–8927. https://doi.org/10.1021/acs.macromol.7b01933. 137. Béland, V. A.; Wang, Z.; Sham, T.; Ragogna, P. J. Antimony-Functionalized Phosphine-Based Photopolymer Networks. Angew. Chem. Int. Ed. 2018, 57 (40), 13252–13256. https://doi.org/10.1002/anie.201806235. 138. Béland, V. A.; Ragogna, P. J. Orthogonally Bimetallized Phosphane-ene Photopolymer Networks. Chem. Eur. J. 2020, 26 (56), 12751–12757. https://doi.org/10.1002/ chem.202001179. 139. Béland, V. A.; Ragogna, P. J. Metallized Phosphane–Ene Polymer Networks as Precursors for Ceramics With Excellent Shape Retention. ACS Appl. Mater. Interfaces 2020, 12 (24), 27640–27650. https://doi.org/10.1021/acsami.0c09044. 140. Wright, V. A.; Gates, D. P. Poly(p-Phenylenephosphaalkene): A Pi-Conjugated Macromolecule Containing P¼ C Bonds in the Main Chain. Angew. Chem. Int. Ed. Eng. 2002, 41 (13), 2389–2392. https://doi.org/10.1002/1521-3773(20020703)41:133.0.CO;2-6. 141. Smith, R. C.; Chen, X.; Protasiewicz, J. D. A Fluorescent (E)-Poly(p-Phenylenephosphaalkene) Prepared by a Phospha-Wittig Reaction. Inorg. Chem. 2003, 42 (18), 5468–5470. https://doi.org/10.1021/ic0345471. 142. Smith, R. C.; Protasiewicz, J. D. Conjugated Polymers Featuring Heavier Main Group Element Multiple Bonds: A Diphosphene-PPV. J. Am. Chem. Soc. 2004, 126 (8), 2268–2269. https://doi.org/10.1021/ja0394683. 143. Smith, R. C.; Protasiewicz, J. D. Systematic Investigation of PPV Analogue Oligomers Incorporating Low-Coordinate Phosphorus Centres. Eur. J. Inorg. Chem. 2004, 2004 (5), 998–1006. https://doi.org/10.1002/ejic.200300551. 144. Tennyson, E. G.; He, S.; Osti, N. C.; Perahia, D.; Smith, R. C. Luminescent Phosphonium Polyelectrolyte Prepared from a Diphosphine Chromophore: Synthesis, Photophysics, and Layer-by-Layer Assembly. J. Mater. Chem. 2010, 20 (37), 7984. https://doi.org/10.1039/c0jm01430c. 145. Yang, X.; Bedford, M. S.; Wan, W.; Conrad, C. A.; Colter, E. F.; Freeman, E. H.; Wilson, C.; Hu, L.; Chumanov, G.; Whitehead, K. J.; Smith, R. C. Intercation Spacing and Side Chain Effects on Phosphonium Polymers: Thermal, Supramolecular, and Bactericidal Properties. J. Polym. Sci. Part A Polym. Chem. 2019, 57 (1), 24–34. https://doi.org/ 10.1002/pola.29274.

230

Organophosphorus and Related Group 15 Polymers

146. Tsang, C.-W.; Yam, M.; Gates, D. P. The Addition Polymerization of a PC Bond: A Route to New Phosphine Polymers. J. Am. Chem. Soc. 2003, 125 (6), 1480–1481. https://doi. org/10.1021/ja029120s. 147. Noonan, K. J. T.; Gates, D. P. Ambient-Temperature Living Anionic Polymerization of Phosphaalkenes: Homopolymers and Block Copolymers With Controlled Chain Lengths. Angew. Chem. Int. Ed. 2006, 45 (43), 7271–7274. https://doi.org/10.1002/anie.200602955. 148. Rawe, B. W.; Chun, C. P.; Gates, D. P. Anionic Polymerisation of Phosphaalkenes Bearing Polyaromatic Chromophores: Phosphine Polymers Showing “Turn-on” Emission Selectively with Peroxide. Chem. Sci. 2014, 5 (12), 4928–4938. https://doi.org/10.1039/C4SC02133A. 149. Siu, P. W.; Serin, S. C.; Krummenacher, I.; Hey, T. W.; Gates, D. P. Isomerization Polymerization of the Phosphaalkene MesP¼ CPh2: An Alternative Microstructure for Poly(Methylenephosphine)s. Angew. Chem. Int. Ed. 2013, 52 (27), 6967–6970. https://doi.org/10.1002/anie.201301881. 150. Rawe, B. W.; Priegert, A. M.; Wang, S.; Schiller, C.; Gerke, S.; Gates, D. P. An Addition–Isomerization Mechanism for the Anionic Polymerization of MesP]CPh2 and m-XylP] CPh2. Macromolecules 2018, 51 (7), 2621–2629. https://doi.org/10.1021/acs.macromol.8b00100. 151. Noonan, K. J. T.; Gates, D. P. Studying a Slow Polymerization: A Kinetic Investigation of the Living Anionic Polymerization of PC Bonds. Macromolecules 2008, 41 (6), 1961–1965. https://doi.org/10.1021/ma702136k. 152. Tsang, C.-W.; Baharloo, B.; Riendl, D.; Yam, M.; Gates, D. P. Radical Copolymerization of a Phosphaalkene with Styrene: New Phosphine-Containing Macromolecules and Their Use in Polymer-Supported Catalysis. Angew. Chem. Int. Ed. 2004, 43 (42), 5682–5685. https://doi.org/10.1002/anie.200460939. 153. Serin, S. C.; Dake, G. R.; Gates, D. P. Addition-Isomerization Polymerization of Chiral Phosphaalkenes: Observation of Styrene–Phosphaalkene Linkages in a Random Copolymer. Macromolecules 2016, 49 (11), 4067–4075. https://doi.org/10.1021/acs.macromol.6b00667. 154. Kobayashi, S.; Suzuki, M.; Saegusa, T. Cationic Ring-Opening Polymerization of 2-Phenyl-1,2-Oxaphospholane (Deoxophostone): Synthesis of Poly(Phosphine Oxide) and Polyphosphine. Polym. Bull. 1981, 4 (6), 315–321. https://doi.org/10.1007/BF00257694. 155. Kobayashi, S.; Suzuki, M.; Saegusa, T. Kinetic Studies on Cationic Ring-Opening Polymerization of 2-Phenyl-1,2-Oxaphospholane. Macromolecules 1984, 17 (1), 107–110. https://doi.org/10.1021/ma00131a018. 156. Withers, H. P.; Seyferth, D.; Fellmann, J. D.; Garrou, P. E.; Martin, S. Phosphorus- and Arsenic-Bridged [1]Ferrocenophanes. 2. Synthesis of Poly((1,1’-Ferrocenediyl) Phenylphosphine) Oligomers and Polymers. Organometallics 1982, 1 (10), 1283–1288. https://doi.org/10.1021/om00070a005. 157. Neuse, E. W.; Chris, G. J. Metallocene Polymers. XVIII. Poly(Ferrocenylphosphine Oxides). J. Macromol. Sci. Part A Chem. 1967, 1 (3), 371–386. https://doi.org/ 10.1080/10601326708053977. 158. Pittman, C. U. Organometallic Polymers. I. Synthesis of Ferrocene-Containing Poly(Phosphine Oxides) and Poly(Phosphine Sulfides). J. Polym. Sci. A1 1967, 5 (11), 2927–2937. https://doi.org/10.1002/pol.1967.150051120. 159. Peckham, T. J.; Massey, J. A.; Honeyman, C. H.; Manners, I. Living Anionic Polymerization of Phosphorus-Bridged [1]Ferrocenophanes: Synthesis and Characterization of Well-Defined Poly(Ferrocenylphosphine) Homopolymers and Block Copolymers. Macromolecules 1999, 32 (9), 2830–2837. https://doi.org/10.1021/ma981329w. 160. Vanderark, L. A.; Clark, T. J.; Rivard, E.; Manners, I.; Slootweg, J. C.; Lammertsma, K. Anionic Ring-Opening Polymerization of a Strained Phosphirene: A Route to Polyvinylenephosphines. Chem. Commun. 2006, No. 31, 3332. https://doi.org/10.1039/b606311j. 161. Chen, X.; Qiao, S.; Du, Z.; Zhou, Y.; Yang, R. Synthesis and Characterization of Functional Thienyl-Phosphine Microporous Polymers for Carbon Dioxide Capture. Macromol. Rapid Commun. 2013, 34 (14), 1181–1185. https://doi.org/10.1002/marc.201300328. 162. Qiao, S.; Du, Z.; Yang, C.; Zhou, Y.; Zhu, D.; Wang, J.; Chen, X.; Yang, R. Phosphine-Containing Microporous Networks: High Selectivity toward Carbon Dioxide to Methane. Polymer 2014, 55 (5), 1177–1182. https://doi.org/10.1016/j.polymer.2014.01.029. 163. Xu, J.; Zhang, C.; Qiu, Z.; Lei, Z.; Chen, B.; Jiang, J.-X.; Wang, F. Synthesis and Characterization of Functional Triphenylphosphine-Containing Microporous Organic Polymers for Gas Storage and Separation. Macromol. Chem. Phys. 2017, 218 (22), 1700275. https://doi.org/10.1002/macp.201700275. 164. Carvalho, P. J.; Ventura, S. P. M.; Batista, M. L. S.; Schröder, B.; Gonçalves, F.; Esperança, J.; Mutelet, F.; Coutinho, J. A. P. Understanding the Impact of the Central Atom on the Ionic Liquid Behavior: Phosphonium vs Ammonium Cations. J. Chem. Phys. 2014, 140 (6), 064505https://doi.org/10.1063/1.4864182. 165. Ran, J.; Wu, L.; He, Y.; Yang, Z.; Wang, Y.; Jiang, C.; Ge, L.; Bakangura, E.; Xu, T. Ion Exchange Membranes: New Developments and Applications. J. Membr. Sci. 2017, 522, 267–291. https://doi.org/10.1016/j.memsci.2016.09.033. 166. Varcoe, J. R.; Atanassov, P.; Dekel, D. R.; Herring, A. M.; Hickner, M. A.; Kohl, P. A.; Kucernak, A. R.; Mustain, W. E.; Nijmeijer, K.; Scott, K.; Xu, T.; Zhuang, L. Anion-Exchange Membranes in Electrochemical Energy Systems. Energy Environ. Sci. 2014, 7 (10), 3135–3191. https://doi.org/10.1039/C4EE01303D. 167. Ye, Y.; Stokes, K. K.; Beyer, F. L.; Elabd, Y. A. Development of Phosphonium-Based Bicarbonate Anion Exchange Polymer Membranes. J. Membr. Sci. 2013, 443, 93–99. https://doi.org/10.1016/j.memsci.2013.04.053. 168. Yan, X.; Gu, S.; He, G.; Wu, X.; Zheng, W.; Ruan, X. Quaternary Phosphonium-Functionalized Poly(Ether Ether Ketone) as Highly Conductive and Alkali-Stable Hydroxide Exchange Membrane for Fuel Cells. J. Membr. Sci. 2014, 466, 220–228. https://doi.org/10.1016/j.memsci.2014.04.056. 169. Tang, H.; Li, D.; Li, N.; Zhang, Z.; Zhang, Z. Anion Conductive Poly(2,6-Dimethyl Phenylene Oxide)S with Clicked Bulky Quaternary Phosphonium Groups. J. Membr. Sci. 2018, 558, 9–16. https://doi.org/10.1016/j.memsci.2018.04.027. 170. Sebastian, M.; Hissler, M.; Fave, C.; Rault-Berthelot, J.; Odin, C.; Réau, R. Phosphole-Modified Poly(Thiophene)s: Unique Postfunctionalizable Conjugated Polymers that Sense Elemental Chalcogenides. Angew. Chem. Int. Ed. Eng. 2006, 45 (37), 6152–6155. https://doi.org/10.1002/anie.200601844. 171. Carrasco, P.-M.; Pozo-Gonzalo, C.; Grande, H.; Pomposo, J. A.; Cortázar, M.; Deborde, V.; Hissler, M.; Reau, R. Synthesis and Spectroelectrochemical Characterization of an Electrochromic Phosphole-EDOT Copolymer: Poly([1-Phenyl-2,5-Bis(2-Thienyl)Thioxophosphole]0.14-Co-[3,4-Ethylendioxythiophene]0.86). Polym. Bull. 2008, 61 (6), 713–724. https://doi.org/10.1007/s00289-008-0998-3. 172. Na, H.-S.; Morisaki, Y.; Aiki, Y.; Chujo, Y. Synthesis of Conjugated Polymers Containing Phosphole With the 5-Member Fused Carbocycle. Polym. Bull. 2007, 58 (4), 645–652. https://doi.org/10.1007/s00289-006-0702-4. 173. Na, H.-S.; Morisaki, Y.; Aiki, Y.; Chujo, Y. Synthesis and Characterization of Novel p-Conjugated Polymers with Phosphole Ring Derivatives. J. Polym. Sci. Part A Polym. Chem. 2007, 45 (14), 2867–2875. https://doi.org/10.1002/pola.22043. 174. Rawe, B. W.; Gates, D. P. Poly(p-Phenylenediethynylene Phosphane): A Phosphorus-Containing Macromolecule That Displays Blue Fluorescence Upon Oxidation. Angew. Chem. Int. Ed. 2015, 54 (39), 11438–11442. https://doi.org/10.1002/anie.201504464. 175. Yamazawa, C.; Imoto, H.; Naka, K. Syntheses of Dithienoarsole-Containing Polymers via Suzuki-Miyaura and Sonogashira-Hagihara Coupling Reactions. Chem. Lett. 2018, 47 (7), 887–890. https://doi.org/10.1246/cl.180298. 176. Mackenzie, H. K.; Rawe, B. W.; Samedov, K.; Walsgrove, H. T. G.; Uva, A.; Han, Z.; Gates, D. P. A Smart Phosphine–Diyne Polymer Displays “Turn-On” Emission With a High Selectivity for Gold(I/III) Ions. J. Am. Chem. Soc. 2020, 142 (23), 10319–10324. https://doi.org/10.1021/jacs.0c04330. 177. Dienes, Y.; Eggenstein, M.; Neumann, T.; Englert, U.; Baumgartner, T. Synthesis and Optoelectronic Properties of Transition Metal Complexes Incorporating Dithieno [3,2-b:20 ,30 -d]Phosphole Ligands. Dalton Trans. 2006, 11, 1424–1433. https://doi.org/10.1039/B509277A. 178. Oki, O.; Kushida, S.; Mikosch, A.; Hatanaka, K.; Takeda, Y.; Minakata, S.; Kuwabara, J.; Kanbara, T.; Dao, T. D.; Ishii, S.; Nagao, T.; Kuehne, A. J. C.; Deschler, F.; Friend, R. H.; Yamamoto, Y. FRET-Mediated Near Infrared Whispering Gallery Modes: Studies on the Relevance of Intracavity Energy Transfer with Q-Factors. Mater. Chem. Front. 2018, 2 (2), 270–274. https://doi.org/10.1039/C7QM00498B. 179. Morisaki, Y.; Ouchi, Y.; Tsurui, K.; Chujo, Y. Synthesis of Optically Active Polymers Containing Chiral Phosphorus Atoms in the Main Chain. J. Polym. Sci. Part A Polym. Chem. 2007, 45 (5), 866–872. https://doi.org/10.1002/pola.21849. 180. Morisaki, Y.; Suzuki, K.; Imoto, H.; Chujo, Y. P-Stereogenic Optically Active Polymer and the Complexation Behavior. Macromol. Chem. Phys. 2011, 212 (24), 2603–2611. https://doi.org/10.1002/macp.201100432.

Organophosphorus and Related Group 15 Polymers

231

181. Ouchi, Y.; Morisaki, Y.; Ogoshi, T.; Chujo, Y. Synthesis of a Stimuli-Responsive P-Chiral Polymer with Chiral Phosphorus Atoms and Azobenzene Moieties in the Main Chain. Chem. Asian J. 2007, 2 (3), 397–402. https://doi.org/10.1002/asia.200600378. 182. Shao, S.; Wang, S.; Xu, X.; Yang, Y.; Lv, J.; Ding, J.; Wang, L.; Jing, X.; Wang, F. Bipolar Poly(Arylene Phosphine Oxide) Hosts With Widely Tunable Triplet Energy Levels for High-Efficiency Blue, Green, and Red Thermally Activated Delayed Fluorescence Polymer Light-Emitting Diodes. Macromolecules 2019, 52 (9), 3394–3403. https://doi.org/ 10.1021/acs.macromol.9b00077. 183. Ma, Z.; Lü, J.; Dong, W.; Ding, J.; Wang, L. Efficient Red Phosphorescent Polymers With Trap-Assisted Charge Balance: Molecular Design, Synthesis, and Electroluminescent Properties. ACS Appl. Mater. Interfaces 2019, 11 (20), 18730–18738. https://doi.org/10.1021/acsami.9b03618. 184. Shao, S.; Ding, J.; Wang, L.; Jing, X.; Wang, F. White Electroluminescence From All-Phosphorescent Single Polymers on a Fluorinated Poly(Arylene Ether Phosphine Oxide) Backbone Simultaneously Grafted With Blue and Yellow Phosphors. J. Am. Chem. Soc. 2012, 134 (50), 20290–20293. https://doi.org/10.1021/ja310158j. 185. Guterman, R.; Rabiee Kenaree, A.; Gilroy, J. B.; Gillies, E. R.; Ragogna, P. J. Polymer Network Formation Using the Phosphane–Ene Reaction: A Thiol–Ene Analogue with Diverse Postpolymerization Chemistry. Chem. Mater. 2015, 27 (4), 1412–1419. https://doi.org/10.1021/cm504784e. 186. Dorn, H.; Rodezno, J. M.; Brunnhöfer, B.; Rivard, E.; Massey, J. A.; Manners, I. Synthesis, Characterization, and Properties of the Polyphosphinoboranes [RPH–BH2]n (R ¼ Ph, i Bu, p - n BuC6 H4, p-DodecylC6 H4): Inorganic Polymers with a Phosphorus–Boron Backbone. Macromolecules 2003, 36 (2), 291–297. https://doi.org/10.1021/ma021447q. 187. Lu, S.-Y.; Hamerton, I. Recent Developments in the Chemistry of Halogen-Free Flame Retardant Polymers. Prog. Polym. Sci. 2002, 27 (8), 1661–1712. https://doi.org/ 10.1016/S0079-6700(02)00018-7. 188. Mayer-Gall, T.; Knittel, D.; Gutmann, J. S.; Opwis, K. Permanent Flame Retardant Finishing of Textiles by Allyl-Functionalized Polyphosphazenes. ACS Appl. Mater. Interfaces 2015, 7 (18), 9349–9363. https://doi.org/10.1021/acsami.5b02141. 189. Priegert, A. M.; Siu, P. W.; Hu, T. Q.; Gates, D. P. Flammability Properties of Paper Coated with Poly (Methylenephosphine), an Organophosphorus Polymer: Flammability Properties of Paper Coated With Poly(Methylenephosphine). Fire Mater. 2015, 39 (7), 647–657. https://doi.org/10.1002/fam.2264. 190. Iturmendi, A.; Teasdale, I. Water Soluble (Bio)Degradable Poly(Organo)Phosphazenes. In Polyphosphazenes in Biomedicine, Engineering, and Pioneering Synthesis, ACS Symposium Series American Chemical Society, 2018; vol. 1298; pp 183–209. https://doi.org/10.1021/bk-2018-1298.ch009. 191. Teasdale, I. Stimuli-Responsive Phosphorus-Based Polymers: Stimuli-Responsive Phosphorus-Based Polymers. Eur. J. Inorg. Chem. 2019, 2019 (11–12), 1445–1456. https:// doi.org/10.1002/ejic.201801077. 192. Valente, A.; Precker, R. L. M.; Hey-Hawkins, E. Inorganic Polymers for Potential Medicinal Applications. In Smart Inorganic Polymers; Hey-Hawkins, E., Hissler, M., Eds.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2019;; pp 243–276. https://doi.org/10.1002/9783527819140.ch9. 193. Peng, H.; Zhang, W.; Hung, W.; Wang, N.; Sun, J.; Lee, K.; An, Q.; Liu, C.; Zhao, Q. Phosphonium Modification Leads to Ultrapermeable Antibacterial Polyamide Composite Membranes With Unreduced Thickness. Adv. Mater. 2020, 32 (23), 2001383. https://doi.org/10.1002/adma.202001383. 194. Süer, N. C.; Demir, C.; Ünübol, N. A.; Yalçın, Ö.; Kocagöz, T.; Eren, T. Antimicrobial Activities of Phosphonium Containing Polynorbornenes. RSC Adv. 2016, 6 (89), 86151–86157. https://doi.org/10.1039/C6RA15545F. 195. Demir, C.; Ceren Süer, N.; Yapaöz, M. A.; Kébir, N.; Okullu, S.Ö.; Kocagöz, T.; Eren, T. Biocidal Activity of ROMP- Polymer Coatings Containing Quaternary Phosphonium Groups. Prog. Org. Coat. 2019, 135, 299–305. https://doi.org/10.1016/j.porgcoat.2019.06.008. 196. Hisey, B.; Ragogna, P. J.; Gillies, E. R. Phosphonium-Functionalized Polymer Micelles With Intrinsic Antibacterial Activity. Biomacromolecules 2017, 18 (3), 914–923. https:// doi.org/10.1021/acs.biomac.6b01785. 197. Loczenski Rose, V.; Mastrotto, F.; Mantovani, G. Phosphonium Polymers for Gene Delivery. Polym. Chem. 2017, 8 (2), 353–360. https://doi.org/10.1039/C6PY01855F. 198. Loczenski Rose, V.; Shubber, S.; Sajeesh, S.; Spain, S. G.; Puri, S.; Allen, S.; Lee, D.-K.; Winkler, G. S.; Mantovani, G. Phosphonium Polymethacrylates for Short Interfering RNA Delivery: Effect of Polymer and RNA Structural Parameters on Polyplex Assembly and Gene Knockdown. Biomacromolecules 2015, 16 (11), 3480–3490. https://doi.org/ 10.1021/acs.biomac.5b00898. 199. Herma, R.; Wrobel, D.; Liegertová, M.; Müllerová, M.; Strašák, T.; Maly, M.; Semerádtová, A.; Štofik, M.; Appelhans, D.; Maly, J. Carbosilane Dendrimers With Phosphonium Terminal Groups Are Low Toxic Non-Viral Transfection Vectors for siRNA Cell Delivery. Int. J. Pharm. 2019, 562, 51–65. https://doi.org/10.1016/j.ijpharm.2019.03.018. 200. Hemp, S. T.; Smith, A. E.; Bryson, J. M.; Allen, M. H.; Long, T. E. Phosphonium-Containing Diblock Copolymers for Enhanced Colloidal Stability and Efficient Nucleic Acid Delivery. Biomacromolecules 2012, 13 (8), 2439–2445. https://doi.org/10.1021/bm300689f. 201. Zhang, C.; Du, K.; Zhang, X.; Zheng, X.; Cao, G.; Zhang, F.; Feng, F. Optical Properties of Phosphonium-, Quaternary Ammonium- and Imidazolium-Substituted Regioregular Polythiophenes and Application for Imaging Live Cells. Dyes Pigments 2019, 170, 107581https://doi.org/10.1016/j.dyepig.2019.107581. 202. Samuels, M. S.; Swennenhius, B. H.; Kamer, P. C. J. Solid-Phase Synthesis of Ligands. In Phosphorus(III)Ligands in Homogeneous Catalysis: Design and Synthesis, John Wiley & Sons, Incorporated: New York, United Kingdom, 2012;; pp 463–480. 203. Webb, P. B.; Cole Hamilton, D. J. The Design of Ligand Systems for Immobilisation in Novel Reaction Media. In Phosphorus(III)Ligands in Homogeneous Catalysis: Design and Synthesis, John Wiley & Sons, Incorporated: New York, United Kingdom, 2012;; pp 497–532. 204. Moussa, Z.; Judeh, Z. M. A.; Ahmed, S. A. Polymer-Supported Triphenylphosphine: Application in Organic Synthesis and Organometallic Reactions. RSC Adv. 2019, 9 (60), 35217–35272. https://doi.org/10.1039/C9RA07094J. 205. Guinó, M.; Hii, K. K.; (Mimi)., Applications of Phosphine-Functionalised Polymers in Organic Synthesis. Chem. Soc. Rev. 2007, 36 (4), 608–617. https://doi.org/10.1039/ B603851B. 206. Astruc, D.; Chardac, F. Dendritic Catalysts and Dendrimers in Catalysis. Chem. Rev. 2001, 101 (9), 2991–3024. https://doi.org/10.1021/cr010323t. 207. Jia, X.; Liang, Z.; Chen, J.; Lv, J.; Zhang, K.; Gao, M.; Zong, L.; Xie, C. Porous Organic Polymer Supported Rhodium as a Reusable Heterogeneous Catalyst for Hydroformylation of Olefins. Org. Lett. 2019, 21 (7), 2147–2150. https://doi.org/10.1021/acs.orglett.9b00459. 208. Li, R.-H.; Zhang, G.-L.; Dong, J.-X.; Li, D.-C.; Yang, Y.; Pan, Y.-M.; Tang, H.-T.; Chen, L.; Zhan, Z.-P. Xantphos Doped POPs-PPh3 as Heterogeneous Ligand for Cobalt-Catalyzed Highly Regio- and Stereoselective Hydrosilylation of Alkynes. Chem. Asian J. 2019, 14 (1), 149–154. https://doi.org/10.1002/asia.201801241. 209. Sawano, T.; Lin, Z.; Boures, D.; An, B.; Wang, C.; Lin, W. Metal–Organic Frameworks Stabilize Mono(Phosphine)–Metal Complexes for Broad-Scope Catalytic Reactions. J. Am. Chem. Soc. 2016, 138 (31), 9783–9786. https://doi.org/10.1021/jacs.6b06239. 210. Xu, Y.; Wang, T.; He, Z.; Zhou, M.; Yu, W.; Shi, B.; Huang, K. Honeycomb-Like Bicontinuous P-Doped Porous Polymers from Hyper-Cross-Linking of Diblock Copolymers for Heterogeneous Catalysis. Macromolecules 2017, 50 (24), 9626–9635. https://doi.org/10.1021/acs.macromol.7b02222. 211. Kann, A.; Hartmann, H.; Besmehn, A.; Hausoul, P. J. C.; Palkovits, R. Hydrogenation of CO2 to Formate Over Ruthenium Immobilized on Solid Molecular Phosphines. ChemSusChem 2018, 11 (11), 1857–1865. https://doi.org/10.1002/cssc.201800413. 212. Bai, Y.; Woods, D. J.; Wilbraham, L.; Aitchison, C. M.; Zwijnenburg, M. A.; Sprick, R. S.; Cooper, A. I. Hydrogen Evolution From Water Using Heteroatom Substituted Fluorene Conjugated Co-Polymers. J. Mater. Chem. A 2020, 8 (17), 8700–8705. https://doi.org/10.1039/D0TA02599B. 213. Hu, K.; Tang, Y.; Cui, J.; Gong, Q.; Hu, C.; Wang, S.; Dong, K.; Meng, X.; Sun, Q.; Xiao, F.-S. Location Matters: Cooperativity of Catalytic Partners in Porous Organic Polymers for Enhanced CO2 Transformation. Chem. Commun. 2019, 55 (62), 9180–9183. https://doi.org/10.1039/C9CC05051E. 214. Wang, J.; Wei Yang, J. G.; Yi, G.; Zhang, Y. Phosphonium Salt Incorporated Hypercrosslinked Porous Polymers for CO2 Capture and Conversion. Chem. Commun. 2015, 51 (86), 15708–15711. https://doi.org/10.1039/C5CC06295K. 215. Li, W.-H.; Li, C.-Y.; Li, Y.; Tang, H.-T.; Wang, H.-S.; Pan, Y.-M.; Ding, Y.-J. Palladium-Metalated Porous Organic Polymers as Recyclable Catalysts for Chemoselective Decarbonylation of Aldehydes. Chem. Commun. 2018, 54 (61), 8446–8449. https://doi.org/10.1039/C8CC03109F. 216. Matsumoto, H.; Hoshino, Y.; Iwai, T.; Sawamura, M.; Miura, Y. Polystyrene-Supported PPh3 in Monolithic Porous Material: Effect of Cross-Linking Degree on Coordination Mode and Catalytic Activity in Pd-Catalyzed C–C Cross-Coupling of Aryl Chlorides. ChemCatChem 2020, 12 (16), 4034–4037. https://doi.org/10.1002/cctc.202000651.

232

Organophosphorus and Related Group 15 Polymers

217. Wang, X.; Xu, Y.; Wang, F.; Wei, Y. Functionalized Cellulose-Supported Triphenylphosphine and Its Application in Suzuki Cross-Coupling Reactions. J. Appl. Polym. Sci. 2015, 132 (6)https://doi.org/10.1002/app.41427. 218. Tong, W.; Li, W.-H.; He, Y.; Mo, Z.-Y.; Tang, H.-T.; Wang, H.-S.; Pan, Y.-M. Palladium-Metalated Porous Organic Polymers as Recyclable Catalysts for the Chemioselective Synthesis of Thiazoles From Thiobenzamides and Isonitriles. Org. Lett. 2018, 20 (8), 2494–2498. https://doi.org/10.1021/acs.orglett.8b00886. 219. Heitz, W.; Michels, R. Polymeric Wittig Reagents. Angew. Chem. Int. Ed. Eng. 1972, 11 (4), 298–299. https://doi.org/10.1002/anie.197202981. 220. Teng, Y.; Lu, J.; Toy, P. H. Rasta Resin-PPh3-NBniPr2 and Its Use in One-Pot Wittig Reaction Cascades. Chem. Asian J. 2012, 7 (2), 351–359. https://doi.org/10.1002/ asia.201100721. 221. Cuthbert, T. J.; Evoy, E.; Bow, J. P. J.; Guterman, R.; Stubbs, J. M.; Gillies, E. R.; Ragogna, P. J.; Blacquiere, J. M. CapturePhos—A Phosphorus-Rich Polymer as a Homogeneous Catalyst Scavenger. Catal. Sci. Technol. 2017, 7 (13), 2685–2688. https://doi.org/10.1039/C7CY00391A.

14.07

Organometallic Dendrimers

Xu-Qing Wang, Xiao-Qin Xu, Wei Wang, and Hai-Bo Yang, Shanghai Key Laboratory of Green Chemistry and Chemical Processes, Chang-Kung Chuang Institute, School of Chemistry and Molecular Engineering, East China Normal University, Shanghai, China © 2022 Elsevier Ltd. All rights reserved.

14.07.1 Introduction 14.07.2 Metallodendrimers with organometallic cores 14.07.3 Metallodendrimers with organometallic branches 14.07.4 Metallodendrimers with organometallic peripheries 14.07.5 Concluding remarks Acknowledgments References

233 233 237 247 255 255 255

14.07.1 Introduction Since the pioneering work by Vögtle,1 Tomalia,2 Newkome3 and many others,4 dendrimers have evolved as attractive macromolecules with extensive applications in various fields such as supramolecular chemistry, materials science and nanoscience. Relying on the rapid development of synthetic strategies ranging from the classical divergent and convergent synthesis to the recently-developed orthogonal synthesis or double-stage convergent approach, diverse dendrimers have been designed and synthesized, thereby laying a foundation for the in-depth investigations of dendrimer chemistry. In particular, taking advantage of their highly branched and star-shaped nanoscale architectures, dendrimers have proven to be versatile platforms for the introduction of functional groups, which further imparts them with desirable properties and specific functions.5–8 Among the most diverse dendrimers ever synthesized, metallodendrimers which contain metallic components within the dendritic skeletons have attracted more and more attention. The existence of metallic moieties endows the resultant metallodendrimers with not only interesting photo-, electro-, or redox- properties, etc. derived from the metallic centers but also attractive mechanical and processable features such as self-healing behaviors. Thus, metallodendrimers are promising platforms for the construction of novel functional materials. Since the first example of metallodendrimers was reported in the early 1990s, widespread applications of metallodendrimers in catalysis, luminescence, sensors, molecular electronics, magnetic resonance imaging (MRI), etc. have been witnessed during past few decades.9,10 In particular, depending on the forms of metallic moieties, metallodendrimers can be generally divided into coordination dendrimers and organometallic dendrimers. For coordination dendrimers, the metallic moieties are introduced by metal-ligand coordination interactions. Representative examples of such metallodendrimers are the transition metal-coordination complexes of organic dendrimers with coordination sites, which have been widely explored as efficient catalysts for diverse chemical transformations. In the case of organometallic dendrimers, organometallic moieties which act as key building blocks take part in the growth of the targeted metallodendrimers. In contrast to coordination dendrimers based on relatively labile metal-ligand coordination interactions, organometallic dendrimers are synthesized from stable and robust organometallic building blocks, whose numbers and locations within the dendritic skeleton could be well controlled, thus resulting in the precise synthesis of organometallic dendrimers with diverse architectures. Such diversity in both the organometallic building blocks and the architectures of the resultant organometallic dendrimers endows them with attractive properties and wide applications especially in catalysis, luminescent materials, sensors and electronic devices.11–13 In this chapter, the progress in organometallic dendrimers in the new millennium are summarized. Classified by the location of the organometallic moieties within the dendritic skeletons, the selected examples of metallodendrimers with organometallic cores, branches and peripheries are highlighted, with an emphasis on their structural features as well as properties and applications.

14.07.2 Metallodendrimers with organometallic cores As one of the most classic organometallic building blocks, the incorporation of metalloporphyrins into dendrimers resulted in the construction of dendritic porphyrins has been widely explored since the 1990s pioneered by Aida, Diederich and Fréchet et al. The existence of three-dimensional (3D) dendritic structures around the metalloporphyrin cores has proven to well-tune their properties and functions. For instance, aiming at the construction of the regulated nanospace, instead of the commonly-used flexible dendrons, the rigid 1,3,5-phenylene dendron units were attached with the iron porphyrins by Kimura, Shirai and coworkers, resulting in the successful synthesis of a series of dendritic iron porphyrins (Fig. 1). Moreover, in order to mimic the enzymatic activity of cytochrome P-450, the epoxidation of olefins using the resultant dendritic iron porphyrins with the confined and rigid nanospace as catalysts was then tested. By choosing olefins with different shapes and sizes as substrates, a remarkable selectivity of the rigid dendritic iron porphyrins compared with unsubstituted Fe(TPP)(Cl) in olefin epoxidation was observed, indicating an interesting confinement effect of the regulated nanospace around the reaction centre.14

Comprehensive Organometallic Chemistry IV

https://doi.org/10.1016/B978-0-12-820206-7.00158-X

233

234

Organometallic Dendrimers

R

N Cl N Fe N N

R

[G-1] =

R

[G-2.5] =

[G-2] =

R

Fe(P3)Cl: R = [G-1] Fe(P4)Cl: R = [G-2] Fe(P6)Cl: R = [G-2.5]

Fig. 1 Chemical structures of dendritic iron porphyrins Fe(P3)Cl, Fe(P4)Cl, and Fe(P6)Cl.

With the aim to construct dendritic porphyrins with a tunable nanospace around the porphyrin center, Yamamoto et al. demonstrated the attachment of the dendritic phenylazozmethine (DPA) with metalloporphyrin cores, leading to the construction of organometallic dendrimers PnM (n ¼ 1, 2, 3, 4) (Fig. 2). Due to the existence of multiple coordination sites on the branches of the DPA dendrons, the progressive addition of SnCl2 to the resultant dendritic cobalt porphyrins led to the stepwise complexation of DPA dendrons with SnCl2 from the core imines to the terminal ones as monitored by UV-vis spectroscopy.15 Furthermore, by using zinc porphyrin as the core, the same group demonstrated the preparation of the carbazolephenylazomethine dendrimer ZnPG2–2 with double layer-type dendrons (Fig. 3). It was found that this fourth-generation organometallic dendrimer could serve as a new kind of host for fullerenes (C60, C70 and C84) due to the interaction of metalloporphyrin core and dendrons with the fullerene as well as the entire hydrophobic cavity. Moreover, as revealed by the titration experiments, the higher order fullerenes displayed a higher binding constant with ZnPG2–2 mainly because of the size matching.16 Considering the wide applications of metalloporphyrins in the construction of light harvesting antenna complexes (LHCs), Jang, Kim et al. reported a new artificial light harvesting dendrimer system PTz in which a triazole-bearing focal zinc porphyrin was

R

R N

N M N

P1M: R = G1 P2M: R = G2 P3M: R = G3 P4M: R = G4

N R

R

N N N

M = H2, CoII, CoIII

N

N

N N R=

N

N

N

R=

N

N

G1 R=

N

N

N N N

N

N

N R=

N

N N N

N

N

G2

G3

Fig. 2 Chemical structures of organometallic dendrimers PnM (n ¼ 1, 2, 3, 4) with porphyrin cores.

G4

Organometallic Dendrimers

N

N

N

N

R

235

R

N N

N

N N

Zn

N

R=

N

N

N

N N

R

R

ZnPG2-2 N

N

N

N

Fig. 3 Chemical structure of carbazole-phenylazomethine dendrimer ZnPG2–2.

surrounded by different number of freebase porphyrin wings (Fig. 4). As suggested by detailed photophysical investigations, especially time-resolved fluorescence measurements, the addition of anionic guests that could bind with the triazole-bearing zinc porphyrin led to the decrease of the HOMO-LUMO gap of the zinc porphyrin core, which further modulated the energy transfer pathway in the resultant organometallic dendrimers. For instance, through the addition or removal of CN−, a reversible regulation of the energy flow was successfully achieved, thus offering a new approach towards the construction of molecular-scale photonic switches.17 C8H17O

OC8H17

OC8H17

C8H17O

N

N

NH

NH

HN

N

HN N

C8H17O

OC8H17 NH

C8H17O

O

O

N

HN

N

N

HN

NH

Tz O

N

OC8H17

O

O

O

N

N Zn

N

C8H17O

N NH

N

O

O

O

O

Tz

HN

NH O

O

N

N

C8H17O N NH

Tz= HN

NH OC8H17

N

N

C8H17O OC8H17

Fig. 4 Chemical structure of organometallic dendrimer PTz.

PTz

OC8H17

HN OC8H17

OC8H17 N N N

N

N HN

OC8H17 C8H17O

236

Organometallic Dendrimers

In addition to the aforementioned dendritic metalloporphyrins, Bunz et al. demonstrated the synthesis of new organometallic polyphenylene dendrimers bearing (tetraphenylcyclobutadiene)cyclopentadienylcobalt as the core modules. The CpCo(CO)2mediated dimerization of di- or tetraethynyltolanes and the sequential deprotection of the tri(isopropyl)silyl (TIPS) protecting groups led to the synthesis of the corresponding organometallic cores with 4 or 8 alkyne moieties. The further divergent core extension by using tetraphenylcyclopentadienone resulted in the successful synthesis of the targeted organometallic dendrimers with 24 or 44 phenyl rings, which were both air and water stable (Fig. 5). The scanning pulse voltammetry studies indicated that

O

CoCp 160 oC

CoCp

1

O

CoCp CoCp

2

O

CoCp CoCp

3 Fig. 5 Synthesis of organometallic dendrimers 1–3 through the divergent core extension approach.

Organometallic Dendrimers

O

O

O

O

O

O

O

O

O

O O NH C O

O

O

O O

O

O

O

H O

O

O

O

O

O

O

O O

O

O

O C HN

O

O O

O

O O

O

O

O

O

O

4

O

O

O

O O

O O

O O

O

O

O

N H

O

O C NH

H O

Fe C O

O

O

HN C O

O C

H N

O

O

O O

O

O

O

O

O

O

OC NH

O

O

O

Fe O

O

O

H O NH CO

O

O

H

O

O

O

O

O

O

O

O

237

O

5

O O

O O

Fig. 6 Chemical structures of ferrocene-cored organometallic dendrimers 4 and 5.

all the resultant organometallic dendrimers were electroactive but with different oxidation potentials (for 1, 0.80 V; for 2, 0.82 V; for 3, 0.83 V). These results indicated that along with the increase in steric hindrance around the cores of these organometallic dendrimers, higher oxidation potentials were revealed.18 As another well-investigated organometallic building block, ferrocene has also been utilized as a core for the construction of organometallic dendrimers. For instance, ferrocene-cored organometallic dendrimers 4 and 5 bearing 1,2-isopro-pylidenefuranose capped furanoside as branches were synthesized through the convergent approach by Ray et al. Similar to a previous report, the dendritic environment in the resultant organometallic dendrimers was proven to have a great impact on the redox properties of the ferrocene core. The increase in steric bulk of the surrounding dendrons led to a higher E1/2 value. The observed result was reasonable since the more remarkable dendritic encapsulation with the higher-generation dendrons made oxidation of the ferrocene core more difficult (Fig. 6).19

14.07.3 Metallodendrimers with organometallic branches In addition to the aforementioned stable organometallic building blocks such as metalloporphyrins and metallocenes that usually served as cores of organometallic dendrimers, as discussed in the previous section, some other stable organometallic moieties such as metal-acetylides have also been used for the construction of organometallic dendrimers. In these cases, due to the linear feature and the feasibility for stepwise functionalization, the metal-acetylide units are usually inserted into the branches, thus resulting in metallodendrimers with organometallic branches. In 1999, Takahashi and coworkers reported the construction of platinum-acetylide organometallic dendrimers up to the third-generation through a convergent approach. As shown in Fig. 7, the artful introduction of two kinds of trialkylsilyl protecting groups, trimethylsilyl (TMS) and tri(isopropyl)silyl (TIPS), allowed for the preparation of first-, second- and third-generation dendrons, 6, 7 and 8. All these dendrons possessed a terminal acetylene unit at the vertical point, which served as the reactive site. The first-generation dendrimer 10 was prepared through the reaction of dendron 6 with the core compound 9 in a 3:1 M ratio. By applying the similar method, the second- and third-generation dendrimers, 11 and 12, were obtained successively. Notably, the resultant third-generation dendrimer 12 contained up to 45 platinum atoms within the dendritic scaffold.20 Moreover, starting from the same dendrons, the use of a tetraplatinum core led to the successful formation of the corresponding organometallic dendrimers with 60 platinum atoms.21 In a follow-up study, by introducing the pyridine moiety into these platinum-acetylide dendrons as the coordination site, the same group further prepared a family of hybrid organometallic dendrimers with both organometallic and coordination moieties through the ligand exchange reaction. Notably, due to the labile nature of the PtdN coordination bonds, after treating with Cl−, the transformation of the resultant organometallic dendrimers back to the core and the dendrons was easily achieved.22 For the convergent synthesis of the platinum-acetylide dendrimers, the third generation proved to be the synthetic limit. To deal with this limitation, Takahashi et al. developed a divergent strategy that could successfully realize the synthesis of a sixth-generation platinum-acetylide dendrimer. As shown in Fig. 8, mononuclear platinum-acetylide complex 13 with TIPS as the protecting group of the terminal acetylene was used as the key building block for dendrimer growth and 2,4,6-triethynylmesitylene was selected as the core. The facile and efficient formation of platinum-acetylide bond allowed for the generation of a series of organometallic dendrimers up to sixth-generation, which possessed 189 Pt atoms. Impressively, the diameter of PAD-G6 is larger than 10 nm and its molecular weight is as high as 139, 750.23

238

Organometallic Dendrimers

OMe

OMe Et3P

Pt

OMe

PEt3

H Et3P Pt PEt3

Et3P Pt PEt3

Et3P

Pt

PEt3

6

Et3P Pt PEt3

CuI OMe

Et2NH, rt

+

PEt3 Pt Et3P

MeO

OMe

Et3P Cl Pt PEt3

PEt3 Cl Pt Et3P

Et3P Pt PEt3

PEt3 Pt Et3P

Et3P Pt PEt3

Et3P Pt PEt3

Et3P Pt PEt3 Cl

MeO

PEt3 Pt Et3P OMe

10

9

PT PT

PT PT

PT PT PT PT PT PT

PT PT

PT

PT PT PT

7

PT PT

PT PT

PT PT PT PT

PT PT

PT PT

PT PT PT

PT

PT PT

PT PT PT PT

PT PT

PT PT

PT

PT PT

PT PT

PT PT

PT PT PT PT

8

PT PT

PT

PT

PT PT PT

PT PT

PT

H PT

PT

PT PT

PT PT

PT PT

PT

PT

H

PT

PT PT

PT PT

11

PT PT

PT PT

PT PT

PT PT

12 Fig. 7 Synthesis of organometallic dendrimers 10, 11 and 12 via a convergent approach.

By using the similar platinum-acetylide dendrimers as the main scaffold, which were synthesized through the same divergent approach as the previous report, Yang, Wang, Chen et al. successfully realized the facile synthesis of a series of organometallic rotaxane-terminated dendrimers. In their study, the [2]rotaxane 14 with the platinum-acetylide unit as a stopper was employed as the terminal rotaxane moiety. By employing the CuI-catalyzed coupling reaction of [2]rotaxane 14 with organometallic dendrimers PAD-Gn-H (n ¼ 1, 2, 3, 4) with different numbers of alkyne terminals, up to the fourth-generation rotaxane-terminated dendrimers PAD-G4, which contained 48 platinum-acetylide units as well as 24 rotaxane moieties at the termini, was successfully prepared (Fig. 9).24 Taking advantage of the thermally robust and stable nature as well as the facile formation of platinum-acetylide moieties in mild conditions, during the past few years, by the introduction of neutral platinum-acetylide moiety into the dendritic skeleton to serve as the key linker, Yang et al. reported the efficient and facile synthesis of a series of high-generation organometallic rotaxane dendrimers. In 2015, the first synthesis of organometallic dendrimers with rotaxane branches up to the fourth-generation was demonstrated. As shown in Fig. 10, [2]rotaxane 15 was used as the key building block, in which the platinum-acetylide unit served not only as the stopper but also provided the reactive sites. By taking advantage of the formation of platinum-acetylide bonds as the key dendrimer growth step, organometallic rotaxane dendrimers RDG1-RDG4 have been successfully constructed through a controllable divergent strategy. It should be mentioned that the fourth-generation rotaxane dendrimer RDG4 was a highly branched [46]rotaxane system with 45 platinum-acetylide linkers as well as 45 individual rotaxane units dispersed within the dendritic

Organometallic Dendrimers

239

G

SiPri3

divergent strategy

PEt3 Cl Pt PEt3

G

13

SiPri3

PAD-G6

G

Si Si Si Si Si Si Si Si Si Si Si Si Si Si Si PTPT PT Si Si PT Si PTPT Si Si PT PT Si Si PT PT PT PT Si Si PT PT Si PT Si PT PT PT Si PT PT Si PT PT PT PT Si PT PT PT PT Si PT PT Si Si PT Si Si PT Si PT PT Si PT Si PT PT PT PT Si PT PT PT Si Si PT PT PT Si PT PT PT Si PT PT Si PT PT PT PT PT Si PT Si Si PTPT Si PT Si Si Si PTPT PT Si Si Si Si Si SiSi Si Si Si SiSi SiSi G=

PT

PT = Pt(PEt3)2 SI = SiPri3

=

Fig. 8 Synthesis of platinum-acetylide dendrimers up to the sixth-generation PAD-G6.

skeleton with a monodispersed distribution. Notably, the resultant rotaxane dendrimers could be further modified with ferrocenes at the peripheries, leading to the successful construction of heterometallic dendrimers with both organometallic branches and terminals.25 Subsequently, the same group further prepared a series of dynamic rotaxane dendrimers via the same divergent approach. As shown in Fig. 11, the urea moiety was introduced into the [2]rotaxane 16 to serve as stimuli-responsive site, which endowed [2] rotaxane 16 with solvent- and anion-responsiveness. Similar with the aforementioned [2]rotaxane 15, the platinum-acetylide unit in [2]rotaxane 16 was used both as a stopper and as a key growth site. Starting from organometallic [2]rotaxane 16, the stepwise formation of platinum-acetylide bonds allowed for the synthesis of dynamic rotaxane dendrimers DS-Gn (n ¼ 1, 2, 3) up to the third-generation. Interestingly, with the addition or removal of the external stimuli such as DMSO molecules or acetate anions, the obvious and reversible size modulation of the integrated organometallic rotaxane dendrimers was achieved.26 By using the same design strategy but with newly designed [2]rotaxane building block, a series of brand-new platinum-acetylide organometallic rotaxane dendrimers in which the wheel components of the rotaxane units are located both on the branches and at the branching points have been further prepared by the same group, which also displayed anion-induced dimension modulation feature.27 Encouraged by the successful synthesis of organometallic rotaxane dendrimers through such controllable divergent strategy based on the formation of platinum-acetylide bonds, Yang et al. further proposed and realized the first synthesis of daisy chain dendrimers in which the molecular [c2]daisy chain units, the most classic mechanically interlocked molecules (MIMs) for the construction of molecular muscles, served as the branches (Fig. 12). Similarly, due to the introduction of urea moieties in the key building block 17, the resultant daisy chain dendrimers DC-Gn (n ¼ 1, 2, 3) revealed solvent- and anion-responsiveness, which were further employed for the construction of composite polymer films that showed fast, reversible, and controllable shape transformations upon adding or removing DMSO molecules as external stimuli.28

240 Organometallic Dendrimers

Et3P Pt PEt3

Et3P Pt PEt3

PEt3 Pt PEt3

PEt3 Pt PEt3 Pt

PEt3

Et3P

Et3P

Pt PEt3

Et3P Pt PEt3

Et3P

Pt

Et3P Pt PEt3

Et3P

PEt3

Et3P Pt PEt3

PEt3 Pt Et3P

Et3P Pt PEt3

PEt3 Pt Et3P

Et3P Pt PEt3

Pt

PEt3

CuI, Et2NH rt, 12 h, 40%

PEt3 Pt Et3P

PEt3 Et3P Pt

PAD-G4-H Et3P

Pt

PEt3

Et3P

Pt

PAD-G4

PEt3 Pt Et3P

5

Et3P Pt PEt3

O

PEt3

O

=

O

O O

O H2C

14 Fig. 9 Synthesis of organometallic rotaxane-terminated dendrimer PAD-G4 though the coupling reaction of [2]rotaxane 14 with platinum-acetylide dendrimer PAD-G4-H.

PEt3 Pt I PEt3

TIPS

TIPS

O

O H2 C 5

O

5

O

TIPS O

O

O

O H 2C

PEt3 Pt I PEt3

CuI, Et2NH, rt, 12h

Et3P Pt PEt3

79%

TIPS Et3P

O

PEt3 Pt Et3P

Pt PEt3

5

15

O

O

O O

CH2

H2 C

TIPS TIPS

O

5

O

O

RDG1 TIPS TIPS

I

II

III

RDG4 Reaction conditions: (I) (a) TBAF, THF, rt. 12h, 87%; (b) CuI, 15, Et2NH, rt, 12h, 58%; (II) (a) TBAF, THF, rt. 12h, 67%; (b) CuI, 15, Et2NH, rt, 12h, 49%; (II) (a) TBAF, THF, rt. 12h, 63%; (b) CuI, 15, Et2NH, rt, 12h, 83%.

241

Fig. 10 Synthesis of organometallic rotaxane dendrimers RDG1-RDG4 through the controllable divergent approach.

Organometallic Dendrimers

RDG2 RDG3

242

Organometallic Dendrimers

Fig. 11 (a) The chemical structure of [2]rotaxane 16 derived from 15 with a urea moiety (shown in pink) as the stimuli-responsive site, which was employed as the precursor for the synthesis of rotaxane dendrimers DS-Gn (n ¼ 1, 2, 3); (b) Cartoon representations of the reversible transformations between the stretched state and contracted state of rotaxane dendrimers Gn (n ¼ 1, 2, 3) triggered by the addition or removal of external stimuli.

Fig. 12 The chemical structure of molecular [c2]daisy chain 17 for the first successful synthesis of daisy chain dendrimers DC-Gn (n ¼ 1, 2, 3) with up to 21 molecular [c2]daisy chain units.

Organometallic Dendrimers

243

On the basis of the aforementioned study, the same group further attempted to prepare functional organometallic rotaxane dendrimers by introducing additional functional groups. For instance, the employment of 9,10-distyrylanthrance (DSA) unit, a typical aggregation-induced emission luminogen (AIEgen), as core module and the aforementioned platinum-acetylide-based [2] rotaxane 16 as branch module gave rise to a series of organometallic rotaxane dendrimers with AIE behaviours.29 In addition, by introducing another classic AIEgen moiety, tetraphenylethene (TPE), into the organometallic [2]rotaxane 18 as both stopper and functional group, the first synthesis of AIEgen-branched rotaxane dendrimers TPE-RD-Gn (n ¼ 1, 2, 3) was realized. Moreover, a new family of artificial light-harvesting systems (LHSs) were further developed based on these TPE-branched organometallic rotaxane dendrimers, which could serve as efficient photocatalysts for both photooxidation reaction and aerobic cross dehydrogenative coupling (CDC) reaction (Fig. 13).30 Recently, starting from a new organometallic [2]rotaxane 19 with two anthracene moieties (energy donor) modified on the wheels and platinum-acetylide unit used as stopper and reactive site, the employment of a zinc(II) porphyrin moiety as both core module and energy acceptor resulted in the synthesis of a new family of rotaxane dendrimers LGn (n ¼ 1, 2, 3) through the similar controllable divergent approach. Interestingly, the resultant organometallic rotaxane dendrimers were proven to serve as artificial light-harvesting systems, whose energy transfer efficiency and light harvesting antenna effect could be reversibly regulated through the addition or removal of acetate anions since such anions could control the distances between the energy donor and acceptor units within the dendritic skeleton through the anion-induced motions of macrocyclic components in each branch (Fig. 14).31 For all aforementioned reports, platinum-acetylide units only served as excellent links for dendrimer growths. To further explore the possible heavy atom effect of the platinum atoms inserted into the dendrimer, starting from the [2]rotaxane 19, new organometallic rotaxane dendrimers PGn (n ¼ 1, 2, 3) containing up to 21 platinum atoms and 42 anthracene units as photosensitizer moieties were designed and prepared by Yang et al. (Fig. 15). It was found that compared with the corresponding small-molecule photosensitizers, the resultant organometallic dendrimer-based photosensitizers showed a higher 1O2 generation efficiency. Interestingly, along with the increase of dendrimer generation, a gradually enhanced photosensitization efficiency was also revealed. This enhanced photosensitization feature was mainly caused by the enhancement of intersystem crossing (ISC),

Fig. 13 The chemical structure of [2]rotaxane 18 for the synthesis of AIEgen-branched rotaxane dendrimers TPE-RD-Gn (n ¼ 1, 2, 3), which were further employed for the construction of novel artificial LHSs as photocatalysts for both photooxidation and aerobic cross-dehydrogenative coupling (CDC) reactions.

244

Organometallic Dendrimers

Fig. 14 The chemical structure of [2]rotaxane 19 for the synthesis of rotaxane dendrimers LGn (n ¼ 1, 2, 3), and the dynamic regulation of antenna effect in rotaxane dendrimer LG3 upon the addition or removal of acetate anions as external stimuli.

which was attributed to the integration of multiple anthracene moieties and Pt atoms, thus highlighting the privileged role of the platinum-acetylide units for the construction of functional organometallic dendrimers.32 In addition to the platinum-acetylide units, the ruthenium-acetylide units have also been proven to be excellent links for the construction of metallodendrimers with organometallic branches. In particular, the resultant organometallic dendrimers usually display attractive multiphoton-absorption (MPA) properties. For instance, Humphrey et al. demonstrated the synthesis of organometallic dendrimer 20 with nine ruthenium-acetylide units incorporated into the branches through the convergent approach (Fig. 16),33 whose nonlinear absorption properties were then evaluated by Z-scan experiments. Impressively, organometallic dendrimer 20 revealed a record three-photon absorption (3PA) coefficient (s3 ¼ 1.5  10−77 cm6s2), thus highlighting the potential applications of organometallic dendrimers in multiphoton absorption.34 Encouraged by such promising results, more efforts were put on the construction of ruthenium-acetylide dendrimers for nonlinear optics (NLO). In a following study, both the influence of p-system lengthening and the generation effect of the organometallic dendrimer on their NLO properties were evaluated. Z-scan experiments indicated that the newly-designed N-cored organometallic dendrimers 21–23 (Fig. 17) displayed two-photon absorption (2PA) behavior when the wavelengths are below 1000 nm with an interesting positive dendrimer effect. Along with the increase of dendrimer generation, the increase in nonlinearity

Organometallic Dendrimers

245

Fig. 15 Cartoon representation of rotaxane dendrimer PG3 and the enhanced intersystem crossing within the rotaxane dendrimer skeleton.

was observed. Beyond 1000 nm, the dendrimers were likely to go through 3PA-induced photochemistry. Notably, compared with the similar organic dendrimers, these organometallic dendrimers revealed better NLO performances, again suggesting the superiority of the rational incorporation of organometallic units within the dendritic architectures.35 To gain a more in-depth understanding of such a unique “dendrimer organometallation” effect, the same group then compared the NLO performances of three dendrimers 24–26 (Fig. 18) which differed in the linkages of branches. The results indicated that, upon increasing the content of ruthenium metal centers, the substantial increase in the multiphoton absorption properties of organometallic dendrimers was achieved. More impressively, for organometallic dendrimer 26 with the maximum number of ruthenium-acetylide units, a record four-photon absorption (4PA) cross-section was observed, and the value was an order of magnitude greater than the previous record one.36

246

Organometallic Dendrimers

NO2

O 2N

[Ru]

[Ru]

[Ru]

[Ru] = trans-[Ru(dppe)2 NO2

O 2N

[Ru]

[Ru] [Ru]

[Ru]

[Ru]

[Ru]

NO2

20

NO2

Fig. 16 Chemical structure of organometallic dendrimer 20 with nine ruthenium-acetylide units distributed on the branches.

M Cl

M

M

M

Cl

N

M

M

M

M

M

Cl

21

M

N

M

M

N

M = trans-[Ru(dppe)2] M

22 M

M

23

Fig. 17 Chemical structures of N-cored organometallic dendrimers 21–23.

Organometallic Dendrimers

247

O

O

Y

Y

Y

X

Compound

24 Ph2P

X

Ph2P

PPh2

Ph2P

PPh2

26

X

Ph2P

PPh2 Ru

Ru Ph2P

O

PPh2 Ru

25

PPh2

Ph2P

PPh2

O

X

Y

Y

Y

Y

O

O

Fig. 18 Chemical structures of dendrimers 24–26 containing ruthenium metal centers.

Attributed to their unique redox and reactive features, 6-arene-5-cyclopentadienyliron(II) complexes have been successfully employed by Abd-El-Aziz et al. as key building blocks for the construction of diverse organoiron dendrimers. For instance, in a very recent report, by introducing such organoiron units at the branches as well as the well-known drug ibuprofen at the peripheries, the synthesis of a series of organoiron dendrimers up to the fourth generation was successfully achieved (Fig. 19). Furthermore, both the antimicrobial and anti-inflammatory activities of these novel organometallic dendrimers were then tested. The results indicated that some of these metallodendrimers revealed a better antimicrobial performance than the reference drugs. Moreover, all metallodendrimers exhibited considerable anti-inflammatory activity, making them quite attractive for practical biomedical uses.37

14.07.4 Metallodendrimers with organometallic peripheries As the most classic redox building blocks, ferrocenes have been widely explored for the construction of metallodendrimers with organometallic peripheries. In particular, the Astruc group has made great contributions to such types of unique organometallic dendrimers.38 For instance, in 2003, they demonstrated the synthesis of metallodendrimers Gn-Fc (n ¼ 1–5) up to the fifth generation with pentamethylamidoferrocenyl termini. As shown in Fig. 20, the reactions of DSM dendrimers Gn-DAB-dend-(NH2) x (n ¼ 1–5, x ¼ 4, 8, 16, 32, 64) with pentamethylferrocenoyl chloride resulted in the formation of the corresponding Gn-Fc (n ¼ 1–5). Impressively, these resultant permethylated metallodendrimers showed excellent electrochemical recognition ability of oxoanions such as H2PO−4, HSO−4 and Adenosine 50 -Triphosphate (ATP) anions, which was mainly due to the stereoelectronic effect of the stabilized 17-electron ferrocenium form and hydrophobic effect of the many methyl groups of the Cp moieties at the peripheries of the organometallic dendrimers.39 In a following report, aiming at the construction of new metallodendrimers for catalytic applications, the same group further prepared a family of triazolylferrocenyl dendrimers (Fig. 21) based on the “click chemistry.” The generated 1,2,3-triazolyl moieties could coordinate to PdII(OAc)2, thus resulting in the formation of the corresponding PdII–triazole dendrimers containing both organometallic and coordination bonds. Notably, the ferrocenyl termini was introduced for effectively monitoring the number of

248

Organometallic Dendrimers

R M R R

M

O

M M M

O

O

R

R

R

M M

M M

O

O

R

M

O

R M M

O

O

M

CO O

O

OC

R

CO

O

O

M M

R M

O

OC

O

O

O O M

R M

CO O

O

R M

M

M

O

OC O

O

O

M

O

O

O

R

O

O M

M O

O R

OC O M

O

CO

O CO

O M

O

R

O

O

M O

M R

M

M C O O

O

O

O

O

OC O

M O

O CO O

M O

O

R OC O

M

M O

M

R

M

R

O M M

O CO O

O CO O

O C O R

M

O

O

O

M

M

O

M O

M

C O

M

O

O

O OC

M O

O

O

O

O C

O

O CO

O O

O

M

O C

O

O M

M

O

O

M

M

O

O

O

O

O

OC

M

M

O C O

O M

O

R

O O

O R

O CO

M O

M

O O CO

M R M

CO

O

O

M

O

O

OC

R O

O

O OC

O M

M

O

O

M

O

O

O CO

R

O

O

R

M M

O

M M

O

O

M

O

M R

O

O O

O

R

M

O

M

OC O

M

O

O

M

O

O

O OC

M

O CO

R

R

O O OC

M

M

R

O O

O C

M O

M M

M O

O O C O

O

O

O

O

M O

O

M

R

O C

O C

M

O

M M

O

M

O

R

O

O

M

O

C O

M M

O C O

M

O C O

O

O

R

R

O

O

R

O

M

CO

O M

O OC

O

O

R O

CO

O OC

M M

M M

O O

R

R

M M

O

M

O M M

O

R

M

M

M R

R

R

R

R R

M

R M

O

O

O

CH3 OH R = Cl

M

=

R=

H3C O

R=

CH3 CO

O

Fe

O

G4-D10

G4-D11

G4-D12

Fig. 19 Chemical structures of the fourth-generation organoiron dendrimers G4-D10, G4-D11 and G4-D12.

incorporated PdII by cyclic voltammetry. Moreover, the reduction of PdII-triazole dendrimers produced Pd nanoparticles (PdNPs) that were stabilized by the triazolylferrocenyl dendrimers, which revealed not only remarkable stability but also impressive catalytic efficiency for olefin hydrogenation.40 In addition, based on giant organometallic dendrimers terminated by first-row late-transition-metal metallocenes, permethyl metallocenes and other sandwich complexes, Astruc et al. further coined the concept of dendritic molecular electrochromic batteries.41 For instance, through the tether-lengthening strategy, a series of giant redox dendrimers were constructed with ferrocenyl and pentamethylferrocenyl termini up to the seventh-generation (Fig. 22), which were well-characterized by various techniques such as NMR, MS, dynamic light scattering (DLS), atomic force microscopy (AFM) and electron-force microscopy (EFM). In particular, according to cyclic voltammetry studies, full chemical and electrochemical reversibility of the resultant organometallic dendrimers was revealed with even up to 14,000 redox termini. Such redox stability of these new electrochromic organometallic

H H

H

H

N

N

N

N

H H

H N

N H

H H

N

H N H

N

N

H

N

N

H

N

N

H N H

N H H

N

H

N

N

N

N H

H H N H

FcCOCl NET3 CH2Cl2

Fe

H

N H

H

N H

N

N

N

N

N H N H H H

H

Fc* COCl NET3 CH2Cl2

G3-DAB-dend-(NH2)16

Fe

Fe

Fe C

O

C O

H

H

N

C N

Fe O

C

H

H

Fe

O

O

H

N

N

N

N

O N

N

C

N

N

N H

N

C

N

Fe

N

N

N

N

N

O C Fe

N H

Fe

O

C N H

N H O

N

N C

H

O C H

O

Fe

N N

N

N

O H

N

Fe

O

O C

Fe

N

N

N

N

N

Fe

O C

H

O

C N H

C O H N

N

N

N

N H O

N

N C

H

O C H

H

N

H

Fe

O

C Fe

O C Fe

Fe

Fe

G3-DAB-dend-(NHCOFc*)16

Fe

249

Fig. 20 Synthesis of the selected third-generation organometallic dendrimers G3-Fc .

C

N

C N

O C

G3-DAB-dend-(NHCOFc)16

Fe

H N

N

Fe

Fe

O

N

Fe Fe

C

N

N

H

Fe

C

H

N

H

C O N H

N

N

Fe O

Organometallic Dendrimers

H

C N

N

C

O N

H

N

O Fe

O C

O H

O H N C

C O

H

C

Fe

H

O

H

Fe

C

N

N

C

Fe

Fe

Fe

Organometallic Dendrimers

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe Fe

Fe

Si

Si

Si

Si

Si Si

Si

Fe N

N

N

Si

N N

Si

Si

Si

N

Si

N

Si Si Si

Si

Si

N N

Si

Si Si

Si

O

O

Si Si

N N N N N N N N N N N Fe N N N N Fe Si

N N

N

N N

N

Fe

N N

N

N N

N

N N

N N

Si

N

Si

N N

Si

Fe

Si

Si N N N

Si

Si

Si

Si

Si

N

N N

N N

Si

O

Si

Fe

Si

N

N

N N N

N N N

N N N

N N

N N

Fe N

N

Fe

N N N Fe

N N N

Si

N N N

Fe

Fe

N N N

Fe

N N N

Fe

N N N

Fe

N N N

Fe

N N N N

Fe

N

N N N

Si Si

N

Si

Fe

N

Si O

O

Si

N

N N N

N

Si

Fe

O

N

N

Si O

Si

N

N N N

N N N

N N

Si

Si

Si

Si

N

N N

Si N N N

O

Si

Fe

N N N

N N N

Si

Si

N N

Si

N

N N

Fe

N

Si

O

N

N

O

Si

Si

N

O

Si

O

Fe

O

N N N

Si

N N N

N

N

N

O

Si

O

N

N N N

O

N

N

Si

N

N

N N N

N N N

Si

Si

N N

N

N

N

Si

O

Si

N

N

O Si

N

N

Si

Si

N N N

N

N

N Si

Si

Fe

N

Si

N

Fe

N

N

N

O

N

N

Si N N N

O

N

N

N N

Fe

N

N

Si

Si

N

N

Fe

N

N

N

O

Si

Si

N

Fe

N

O N

N N N

Si

Si

N

N

N

Fe N

N

Si

Si N O

N

N N

N

Fe

Si Si

Si

N N

N

N N

Si

N

N

Fe

N N N

O

Si

Fe

N

N N N

O

Si

N N

N

Si

N

Fe

Fe

O

Si

N

N N

Si

N N N

Fe

N N N

O N N N N

Si

N N Fe

Fe N N N

Si

Si

N N N

Fe

N N N Si

O

N N N

Fe

N N

O

N N N

O

N N N

Fe

N

Si

N N N

Fe

Si

Si

Fe

N Si

N N N

O Si

N N N N N N

Fe N N N

O

N N N

Si

Fe

N N

Si

N

N N

O

Si

Fe

N N

N N N

Si

N

N N

N

Si

Fe

Fe

N N N

O

O

O

Si

N N

Si

N

Si

N N N

Si

N N

Fe

Fe

N N N

Si

N

Si

N N

Fe

N

N N

N

N

N

N

Fe

Si

N

N

N N

Si

Fe

N

Si

Si

N

N

N

N N

Si

N N N N N N

Fe Fe

N N N

N N Si

Fe

N N N

Si

250

Fe Fe

N Fe

N Fe Fe

Fe

N Fe

N N

Fe Fe Fe

Fe Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fig. 21 Chemical structure of selected triazolylferrocenyl dendrimer synthesized by click chemistry.

dendrimers endowed them with a battery behavior and enabled them to be isolated in oxidized, reduced and mixed-valence forms. In addition, as revealed by AFM studies, along with the redox switching process, the reversible size changes of these organometallic dendrimers were observed, indicating a breathing mechanism.42 Based on the “click chemistry,” the same group further prepared five generations of triazolylbiferrocenyl dendrimers with different numbers of biferrocene termini (Fig. 23). Attributed to the different oxidizability of the inner and the outer Fc groups, the first mixed-valence organometallic dendrimers were isolated and characterized. Moreover, the resultant triazolylbiferrocenyl organometallic dendrimers displayed interesting dendritic effects upon the redox recognition of ATP2− and Pd2+. ATP2− interacted with the outer Fc+ based on the ion-pairing interaction, while Pd2+ attached to the inner Fc group through the coordination with the triazolyl ligand, which induced the splitting of the corresponding CV waves at 0.43 V and 0.74 V, respectively.43 To achieve organometallic dendrimers with multiple oxidation state, Astruc et al. demonstrated the preparation a series of organometallic dendrimers Cp FeII(dppe)-alkynyl units as termini through either click reaction or Sonogashira reaction. Due to the existence of electronical communicating between the two metal centers of the dendrons, organometallic dendrimers in three oxidation states (FeIIFeII, FeIIFeIII and FeIIIFeIII) were successfully achieved. Moreover, it was found that small Au0 nanoparticles

Organometallic Dendrimers

Fe

251

Fe

Fe Fe

Fe

Fe

O

Fe

O

O

Fe

O

O

O

O

O O

O

Si

O

Si

Si

O O

Fe

O

Si

Si

Fe

O Si

Si

O O Si

O

Fe O

O O

O

Si

Si

Fe

O

O

Si Si

Si

O Si

O

O

O Si

O

O

Si

Fe

Si O

Si

Si O O

Fe

O

O

Si

Si

Si

Si

O

O

O Si O

Fe

Si

Si

O

O

O

Si

O

O

Si Fe

Si O

O

Si

O

O

Si

Fe

O

Si Si

O

O

O

O

O

O

O

O

Si

Si

O

Fe

O

Si

Fe O

O O

O

Fe Fe

Fe Fe

Fe Fe Fe

Fe

G1-27-Fc Fig. 22 Chemical structure of the first-generation redox dendrimer G1–27-Fc.

(AuNP) were intra-dendritically stabilized by the organometallic dendrimers that were obtained through click reactions based on the AuNP-triazole interactions, while the ones synthesized by Sonogashira reactions could stabilize large Au0 nanoparticles through inter-dendritic interactions (Fig. 24).44 As mentioned above, due to their wide application for the construction of artificial light harvesting systems, dendritic porphyrins with metalloporphyrins as cores, branches or peripheries have been extensively explored, as summarized by some excellent reviews.45,46 In particular, the development of click chemistry has promoted the synthesis of diverse dendritic porphyrins for the investigation of some key photophysical process such as energy transfer. For instance, Gros, Harvey et al. synthesized a series of dendritic porphyrins with zinc porphyrin dendrons and either bis(zinc porphyrin) or bis(copper porphyrin) as cores (Fig. 25). The photophysical investigation on these organometallic dendrimers indicated the existence of paramagnetic d9 copper(II) in the core module promoted the singlet-triplet energy transfer from the dendrons to the core and slowed the reverse triplet-triplet energy transfer from the core to the dendrons. In the case of bis(zinc porphyrin) as cores, chain folding was revealed as evidenced by the observation of triplet-triplet energy transfer in the heterobimetallic systems.47

Organometallic Dendrimers

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe Fe

Fe

Si

Si

Si

Si

Si

Si

Si

Si

Si

Si

Fe

Fe

N N N

Si

N N N Si

Fe

Si

Si

N

N N

Si

Si

Si

Fe

Si

Si

Fe

N N N

N N N

N N N N

Fe

Fe

Fe

G2-81 Fig. 23 Chemical structure of the second-generation triazolylbiferrocenyl dendrimer G2–81.

Fe

Si Si

Fe Fe Fe

Fe

Fe Fe Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe

N

Fe

Fe Fe

N

N

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe

N

Fe

N

N N N

N N N N N N

N

Fe

Fe

Fe

Fe

Fe Fe

N N

Si

Si

N N N

N

Si

N N N N N Si N N N Si Si Si Si N N N N N N Fe N N N N N N N N N N N N N N N Fe N N N N N N N Fe Fe Fe Fe Fe Fe Fe Fe Fe Fe Fe Si

Fe Fe

N

Si

Si

N N

Fe Fe

Si

O

O

Si

N

Fe

O O

Si

N Fe

Si

Si

Fe

Fe

N N

Si

Si Si

Si

O Si

N N N

N

Si

N

Fe

N

Si

O

N

Si

N

Fe Fe

Si

Si

N

Fe

N

N N

N N

O

Si

Fe

N

Si

N N N N N N

N N N N N N

O

O

N N

Si

Si Si

Si

N

Fe

Si

O

O

Si

Si

N

Si

O

N

Fe Fe

Si

O

Si

N

Si Si

Si

Si

Fe

O O

N

N

Fe

Si

O Si

Fe

N N N

O

Si

N N N

Fe

Si

Si Si

N N N Fe

Si

Fe

Fe

N N N

O

Si

Si

Fe

Si O

Si

N N N

Fe

Fe N

Si

Si

N

Si

N N

Fe

N

N

Si

Si

O Si

Fe

N

O

O O

Si

Fe

N

N

N

Si

Si

Fe

N

Fe

N Si

O

Fe

N

N

Si

N

Fe N

N

O

Fe

N

Si

Si

Fe

N

N

Si

Si

O

Si

N N

Fe

Fe

N

O

Si

Fe Fe

N

Si

Si

N N

Fe

Fe

N N N

Si

Si

Fe

Si

Fe

Fe

Fe

N N N

Si

Si

Fe

Fe N N N

O

O

O

N N N

N N N

Si

N N N Fe

Si

Si

Si

Si

Fe

Fe Fe

N N N

Si

N N N

Fe

Fe

O

Si

Fe

N N N

O O

Si

Fe

Fe

O

Si

Fe

Fe

Fe N N N

Si

Fe

N N N

O O

Si

N N N N N N N N N

Fe

N N

Si

Fe

N N

N N

N N N N N N

N N

Si

N Fe

N N

N

Si

Fe

Fe

N N

N

N N

Si

Fe

N

Si

Fe

Si

N

N N N N N N

N N N

Si

Fe Fe

N N N N N N

Fe

Fe

Fe

Fe

N

Fe Fe

Fe

N

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe Fe

Fe

Fe

Fe

Si

252

Organometallic Dendrimers

Fig. 24 The chemical structures of organometallic dendrimers obtained by click reactions (top) and Sonogashira reactions (down).

253

254

Organometallic Dendrimers

N N N

N N M N N

N N N

O N N M N N

Gx

Gx

N

M= Cu, Gx= G1, G2, G3 M= Zn, Gx= G1

N Zn

N

N

O N

NH N

N

N

Zn

HN

N

N

O

O

Zn

N

N NH NH

HN

H N

O

N

G1

O O

N O

N N H

NH

N

N Zn

HN

O

NH N

O N

N

N

O

Zn N

HN

O O

NH

N

N

O

N

HN O

Zn

N

N N

HN

N Zn

N

N

G3 G2 Fig. 25 Chemical structures of the dendritic porphyrins.

N

Organometallic Dendrimers

255

14.07.5 Concluding remarks During the past two decades, along with the rapid developments of both organometallic chemistry and dendrimer chemistry, the blossoming of organometallic dendrimers as new types of dendrimers has been witnessed. The introduction of organometallic moieties into the dendritic skeletons not only gave rise to diverse dendrimers with impressive architectures, but also endowed intriguing properties and attractive functions upon the resultant organometallic dendrimers, thereby making them promising and versatile platforms for potential applications in wide fields such as functional materials, catalysis, sensing, nonlinear optics, or biomedicines. Although great achievements have been made in this field, the construction of novel organometallic dendrimers and the further exploration and extension of their applications remains to be an attractive topic. In particular, more attentions should be paid on the precise arrangements of the organometallic units within the dendritic skeletons. To achieve this goal, the development of both novel organometallic building blocks and new synthetic strategies will be of great importance. Moreover, the more in-depth understanding of the structure-property relationships is still great needed, which will be helpful for the function-oriented construction of novel organometallic dendrimers.

Acknowledgments H.-B.Y. acknowledges the financial support sponsored by the National Natural Science Foundation of China (21625202 and 92056203), the Innovation Program of Shanghai Municipal Education Commission (2019-01-07-00-05-E00012) and the Fundamental Research Funds for the Central Universities. X.-Q. W. is grateful to the Guanghua Excellent Postdoctoral Scholarship and China Postdoctoral Science Foundation (No. 2019 M660084 and 2020T130198) for support.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41.

Buhleir, E.; Wehner, W.; Vögtle, F. Synthesis 1978, 155–158. Tomalia, D. A.; Baker, H.; Dewald, J.; Hall, M.; Kallos, G.; Martin, S.; Roeck, J.; Ryder, J.; Smith, P. Polym. J. 1985, 17, 117–132. Newkome, G. R.; Yao, Z.; Baker, G. R.; Gupta, V. K. J. Org. Chem. 1985, 50, 2003–2004. Jansen, J. F. G. A.; de Brabander-van den Berg, E. M. M.; Meijer, E. W. Science 1994, 266, 1226–1229. Newkome, G. R.; Moorefield, C. N.; Vögtle, F. Dendrimers and Dendrons. Concepts, Syntheses, Applications; Wiley-VCH: Weinheim, Germany, 2001. Tomalia, D. A., Fréchet, J. M. J., Eds.; In Dendrimers and Other Dendritic Polymers; Wiley: Amsterdam, The Netherlands, 2001. Vögtle, F.; Richardt, G.; Werber, N. Dendrimer Chemistry: Concepts, Syntheses, Properties; Applications; Wiley: Weinheim, 2009. Astruc, D.; Boisselier, E.; Ornelas, C. Chem. Rev. 2010, 110, 1857–1959. Hearshaw, M. A.; Moss, J. R. Chem. Commun. 1999, 1–8. Onitsuka, K.; Takahashi, S. Top. Curr. Chem. 2003, 228, 39–63. Chase, P. A.; Gebbink, R. J. M.; van Koten, K. G. J. Organomet. Chem. 2004, 689, 4016–4054. Hwang, S.-H.; Shreiner, C. D.; Moorefield, C. N.; Newkome, G. R. New J. Chem. 2007, 31, 1192–1217. Govender, P.; Therrien, B.; Smith, G. S. Eur. J. Inorg. Chem. 2012, 2853–2862. Kimura, M.; Shiba, T.; Yamazaki, M.; Hanabusa, K.; Shirai, H.; Kobayashi, N. J. Am. Chem. Soc. 2001, 123, 5636–5642. Imaoka, T.; Horiguchi, H.; Yamamoto, K. J. Am. Chem. Soc. 2003, 125, 340–341. Albrecht, K.; Kasai, Y.; Kuramoto, Y.; Yamamoto, K. Chem. Commun. 2013, 49, 865–867. Yim, D.; Sung, J.; Kim, S.; Oh, J.; Yoon, H.; Sung, Y. M.; Kim, D.; Jang, W.-D. J. Am. Chem. Soc. 2017, 139, 993–1002. Waybright, S. M.; McAlpine, K.; Laskoski, M.; Smith, M. D.; Bunz, U. H. F. J. Am. Chem. Soc. 2002, 124, 8661–8666. Ray, A.; Khan, S. Synlett 2018, 29, 1367–1372. Onitsuka, K.; Fujimoto, M.; Ohshiro, N.; Takahashi, S. Angew. Chem. Int. Ed. 1999, 38, 689–692. Onitsuka, K.; Fujimoto, M.; Kitajima, H.; Ohshiro, N.; Takei, F.; Takahashi, S. Chem. A Eur. J. 2004, 10, 6433–6446. Onitsuka, K.; Iuchi, A.; Fujimoto, M.; Takahashi, S. Chem. Commun. 2001, 741–742. Onitsuka, K.; Shimizu, A.; Takahashi, S. Chem. Commun. 2003, 280–281. Wang, Y.-X.; Zhou, Q.-F.; Chen, L.-J.; Xu, L.; Wang, C.-H.; Li, X.; Yang, H.-B. Chem. Commun. 2018, 54, 2224–2227. Wang, W.; Chen, L.-J.; Wang, X.-Q.; Sun, B.; Li, X.; Zhang, Y.; Shi, J.; Yu, Y.; Zhang, L.; Liu, M.; Yang, H.-B. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 5597–5601. Wang, X.-Q.; Wang, W.; Li, W.-J.; Chen, L.-J.; Yao, R.; Yin, G.-Q.; Wang, Y.-X.; Zhang, Y.; Huang, J.; Tan, H.; Yu, Y.; Li, X.; Xu, L.; Yang, H.-B. Nat. Commun. 2018, 9, 3190. Wang, X.-Q.; Li, W.-J.; Wang, W.; Wen, J.; Zhang, Y.; Tan, H.; Yang, H.-B. J. Am. Chem. Soc. 2019, 141, 13923–13930. Li, W.-J.; Wang, W.; Wang, X.-Q.; Li, M.; Ke, Y.; Yao, R.; Wen, J.; Yin, G.-Q.; Jiang, B.; Li, X.; Yin, P.; Yang, H.-B. J. Am. Chem. Soc. 2020, 142, 8473–8482. Wang, X.-Q.; Wang, W.; Li, W.-J.; Qin, Y.; Yin, G.-Q.; Jiang, W.-L.; Li, X.; Wu, S.; Yang, H.-B. Org. Chem. Front. 2019, 6, 1686–1691. Li, W.-J.; Wang, X.-Q.; Zhang, D.-Y.; Hu, Y.-X.; Xu, W.-T.; Xu, L.; Wang, W.; Yang, H.-B. Angew. Chem. Int. Ed. 2021. https://doi.org/10.1002/anie.202106035. Li, W.-J.; Wang, X.-Q.; Wang, W.; Hu, Z.; Ke, Y.; Jiang, H.; He, C.; Wang, X.; Hu, Y.-X.; Jia, P.-P.; Yin, P.; Chen, J.; Sun, H.; Sun, Z.; Xu, L.; Yang, H.-B. Giant 2020, 2, 100020. Li, W.-J.; Hu, Z.; Xu, L.; Wang, X.-Q.; Wang, W.; Yin, G.-Q.; Zhang, D.-Y.; Sun, Z.; Li, X.; Sun, H.; Yang, H.-B. J. Am. Chem. Soc. 2020, 142, 16748–16756. McDonagh, A. M.; Powell, C. E.; Morrall, J. P.; Cifuentes, M. P.; Humphrey, M. G. Organometallics 2003, 22, 1402–1413. Samoc, M.; Morrall, J. P.; Dalton, G. T.; Cifuentes, M. P.; Humphrey, M. G. Angew. Chem. Int. Ed. 2007, 46, 731–733. Roberts, R. L.; Schwich, T.; Corkery, T. C.; Cifuentes, M. P.; Green, K. A.; Farmer, J. D.; Low, P. J.; Marder, T. B.; Samoc, M.; Humphrey, M. G. Adv. Mater. 2009, 21, 2318–2322. Simpson, P. V.; Watson, L. A.; Barlow, A.; Wang, G.; Cifuentes, M. P.; Humphrey, M. G. Angew. Chem. Int. Ed. 2016, 55, 2387–2391. Abd-El-Aziz, A. S.; Abdelghani, A. A.; El-Ghezlani, E. G.; El-Ezz, D. A.; Abdel-Rahman, L. H. Macromol. Biosci. 2021, 21, 2000242. Astruc, D.; Ornelas, C.; Ruiz, J. Acc. Chem. Res. 2008, 41, 841–856. Daniel, M.-C.; Ruiz, J.; Blais, J.-C.; Daro, N.; Astruc, D. Chem. A Eur. J. 2003, 9, 4371–4379. Ornelas, C.; Aranzaes, J. R.; Salmon, L.; Astruc, D. Chem. A Eur. J. 2008, 14, 50–64. Astruc, D.; Ornelas, C.; Ruiz, J. Chem. A Eur. J. 2009, 15, 8936–8944.

256

42. 43. 44. 45. 46. 47.

Organometallic Dendrimers

Ornelas, C.; Ruiz, J.; Belin, C.; Astruc, D. J. Am. Chem. Soc. 2009, 131, 590–601. Djeda, R.; Rapakousiou, A.; Liang, L.; Guidolin, N.; Ruiz, J.; Astruc, D. Angew. Chem. Int. Ed. 2010, 49, 8152–8156. Wang, Y.; Salmon, L.; Ruiz, J.; Astruc, D. Nat. Commun. 2014, 5, 3489. Li, W.-S.; Aida, T. Chem. Rev. 2009, 109, 6047–6076. Maes, W.; Dehaen, W. Eur. J. Org. Chem. 2009, 4719–4752. Brégier, F.; Aly, S. M.; Gros, C. P.; Barbe, J.-M.; Rousselin, Y.; Harvey, P. D. Chem. A Eur. J. 2011, 17, 14643–14662.

14.08

Organometallic Functionalized MOFs - Reactivity and Catalysis

Thomas M Rayder and Casey R Wade, Department of Chemistry and Biochemistry, The Ohio State University, Columbus, OH, United States © 2022 Elsevier Ltd. All rights reserved.

14.08.1 Introduction 14.08.2 Linker-based organometallic complexes 14.08.2.1 Metal carbonyl complexes 14.08.2.2 NHC complexes 14.08.2.3 Metal-phosphine complexes 14.08.2.4 Cyclometalated linkers 14.08.2.5 N-donor-based linkers 14.08.2.6 Organometallic complexes tethered to linkers 14.08.3 Organometallic species bound at MOF nodes 14.08.4 Non-covalently encapsulated organometallic species 14.08.5 Organometallic metal nodes Acknowledgement References

258 258 258 261 264 268 269 271 273 275 278 281 281

Nomenclature acac bdc BET BINAP bpdc BPHV BPV bpy bpydc btc btdd BTTri cdpm cod Cp h5 Cp h5 dcbdt dccpy DMF DMSO dppe dobdc edb EDS eth EXAFS FTIR ICP IRMOF LSK MFU MIL MMAO MOF MONT MS nbd

Acetylacetonate 1,4-Benzenedicarboxylate Brunauer-Emmett-Teller 2,20 -Bis(diphenylphosphino)-1,10 -binaphthyl Biphenyl-4,40 -dicarboxylate 4,40 -Bis(carboxyethenyl)-1,10 -biphenyl 5,50 -Bis(carboxyethenyl)-2,20 -bipyridyl 2,20 -Bipyridyl 2,20 -Bipyridyl-5,50 -dicarboxylate 1,3,5-Benzenetricarboxylate Bis(1H-1,2,3-triazolo-[4,5-b][40 ,50 -i])dibenzo-[1,4]-dioxin 1,3,5-Tris-(1,2,3-triazole)-benzene Bis[4-(4-carboxyphenyl)-3,5-dimethylpyrazolyl]methane 1,5-Cyclooctadiene cyclopentadienyl pentamethylcyclopentadienyl 3,6-Dicarboxybenzene-1,2-dithiolate 2-Phenylpyridine-5,40 -dicarboxylate N,N0 -dimethylformamide Dimethyl sulfoxide 1,2bis(diphenylphosphino)ethane 2,5-Dioxido-1,4-benzenedicarboxylate 4,40 -Ethynylenedibenzoate Energy-dispersive x-ray spectroscopy Ethylene Extended x-ray absorption fine structure Fourier-transform infrared Inductively coupled plasma Isoreticular metal-organic framework Laboratory for Catalysis and Sustainable Chemistry, Paul Scherrer Institute Metal-Organic Framework Ulm University Materials Institute Lavoisier Modified methylaluminoxane Metal-organic framework Metal-organic nanotube Mass spectrometry 2,5-Norbornadiene (bicyclo[2.2.1]hepta-2,5-diene)

Comprehensive Organometallic Chemistry IV

https://doi.org/10.1016/B978-0-12-820206-7.00120-7

257

258

Organometallic Functionalized MOFs - Reactivity and Catalysis

NHC NHDC NMR NU OES PCM PCN py SBU TEM TFA TGA THF TOF TON tpdc tpp UiO XPS ZJU

N-heterocyclic carbene N-heterocyclic dicarbene Nuclear magnetic resonance Northwestern University Optical emission spectroscopy Phase changed materials Porous coordination network Pyridine Secondary building unit Transmission electron microscopy Trifluoroacetic acid Thermal gravimetric analysis Tetrahydrofuran Turnover frequency (moles product moles catalyst−1 unit time−1) Turnover number (moles product moles catalyst−1) Para-terphenyl-4,40 -dicarboxylate 4,40 ,400 ,4000 -Tetracarboxyphenylpyrene University of Oslo x-Ray photoelectron spectroscopy Zhejiang University

14.08.1 Introduction Metal-organic frameworks (MOFs) have become the subject of intense study over the past two decades.1,2 Their remarkable porosity has resulted in a heavy focus on gas separation and storage applications.3,4 However, MOFs have attracted interest from researchers in all areas of the chemical sciences, and their tunability and ease of synthesis has opened the door to new applications in areas such as sensing,5 drug delivery,6 and catalysis.7,8 MOF synthesis conditions are often prohibitively harsh, precluding the use of reactive organometallic species as building blocks for self-assembly. However, the development of post-synthetic modification methods has allowed for the introduction of sensitive chemical functionalities that are not compatible with direct synthesis.9 Consequently, these synthetic advances have enabled a boom in the design of MOFs containing reactive organometallic functional groups for catalysis and other applications.10 This chapter describes recent discoveries and advances at the intersection of organometallic and MOF chemistry, with a particular focus on catalysis. The scope is generally restricted to MOFs containing well-defined organometallic species with at least one metalcarbon bond.11 However, some pertinent examples, particularly pre-catalyst systems, that may not rigorously conform to the definition of an organometallic compound are also discussed. The advances presented in this chapter are categorized by the strategy used for immobilization or generation of organometallic complexes as follows: (1) linker-based functionalization, (2) grafting at inorganic secondary building units (SBUs) or nodes, (3) noncovalent encapsulation, and (4) organometallic SBUs. In many cases, organometallic species are appended to or immobilized in well-known MOFs. Graphical descriptions of the structures and building units of some commonly discussed MOFs are included in Table 1 below to aid the reader. In addition, specific labels are used in conjunction with compound numbering to help distinguish the various strategies employed for incorporating organometallic complexes within MOFs. Organic linkers are labelled as L while linkers metalated via post-synthetic reactions are indicated by an appended metal group (i.e. L-MX). Organometallic complexes, or metallolinkers, used directly for MOF self-assembly are distinguished with the label ML while complexes grafted at metal nodes are given with a G prefix. Organometallic molecules encapsulated within the pores of a MOF are denoted as E.

14.08.2 Linker-based organometallic complexes Organic linkers offer well-defined and precisely tunable building blocks for incorporating organometallic functionality within MOFs. Functionalized linkers are often used as ligands to support organometallic species introduced via post-synthetic metalation or as sites for appending metal complexes via covalent bond formation with pendent functional groups. In some cases, MOFs have been assembled directly from organometallic compounds decorated with framework-forming functional groups. The following section describes examples of linker-supported organometallic complexes organized by identity of the linker-based donor group.

Organometallic Functionalized MOFs - Reactivity and Catalysis

Table 1

259

Graphical descriptions of the structures and building units of selected MOFs.12–16

14.08.2.1 Metal carbonyl complexes Despite the foundational role of metal carbonyl complexes in organometallic chemistry, there are relatively few examples of MOFs containing linker-supported metal-carbonyl complexes. Some of the first reported examples involved the generation of (Z6-arene)M(CO)3 (M ¼ Cr, Mo) piano stool complexes supported at the central arene group of 1,4-benzenedicarboxylate (bdc-M(CO)3) or biphenyl-4,40 -dicarboxylate (bpdc-M(CO)3) linkers (Fig. 1). Early computational studies explored the possibility of using MOF-immobilized (Z6-arene)Cr(CO)3 complexes as strong H2 adsorption sites via formation of Kubas-type Z2-H2 species.17,18 Around the same time, an experimental study by Long and Kaye showed that high temperature reaction of MOF-5 with Cr(CO)6 facilitates formation of (Z6-bdc)Cr(CO)3 groups at the terephthalate linker sites.19 Subsequent photolysis experiments

260

Organometallic Functionalized MOFs - Reactivity and Catalysis

Fig. 1 (Z6-Arene)M(CO)3 piano stool complexes bound at the aryl group of MOF linkers.17–23

under N2 and H2 atmospheres revealed the formation of (Z6-bdc)Cr(CO)2(N2) and (Z6-bdc)Cr(CO)2(Z2-H2) complexes. Later reports by Groppo, Bordiga, and co-workers used similar methods to support (Z6-arene)Cr(CO)3 species at the bdc and 2,5-dioxido-1,4-benzenedicarboxylate (dobdc-M(CO)3) linkers of UiO-66 and CPO-27-Ni (MOF-74-Ni), respectively.20,21 Horiuchi, Matsuoka, and co-workers have also used chemical vapor deposition to generate (Z6-arene)M(CO)3 complexes at the bdc and bpdc linkers of UiO-66 and UiO-67.22 The resulting materials exhibited good catalytic activity and selectivity for epoxidation of cyclooctene. Zadehahmadi and co-workers used UiO-66 functionalized with (Z6-bdc)Mo(CO)3 groups as precatalysts for alkene epoxidation and sulfide oxidation.23 In 2012, Burrows and co-workers reported the post-synthetic reaction of a Co MOF 1 ([Co3(edb)3(DMF)4]
2.6DMF) containing 4,40 -ethynylenedibenzoate (edb2−) (L1) linkers with Co2(CO)8 (Fig. 2).24 FT-IR characterization of the product was consistent with Z2 coordination of the linker-based alkyne groups to Co2(CO)6 fragments. However, the post-synthetic metalation appeared to be limited to the surface of the MOF crystals, and strain resulting from alkyne coordination to the organometallic framework was believed to result in partial framework collapse. More recent examples of MOF-supported metal carbonyl complexes have focused on the design of heterogeneous catalysts with site-isolated organometallic species. Inspired by [FeFe]-hydrogenase active sites, Cohen, Ott, and co-workers used post-synthetic linker exchange to prepare an analog of UiO-66 (2) containing [FeFe](dcbdt)(CO)6 (dcbdt ¼ 3,6-dicarboxybenzene-1,2-dithiolate) (L2) metallolinkers (Fig. 2).25 This bio-inspired heterogeneous catalyst was found to be more active for photocatalytic hydrogen evolution than its homogeneous analog when used in conjunction with [Ru(bpy)3]2+ as a photosensitizer. MOF-immobilization of [FeFe](dcbdt)(CO)6 was believed to improve its stability under the photocatalytic conditions, inhibit nonproductive charge recombination, and promote disproportionation reactions that generate catalytically active species. In a subsequent report, Ott and co-workers prepared 3 by post-synthetic amide coupling of [FeFe](cbdt)(CO)6] at the amine-functionalized bdc linkers (L3) of (Cr)MIL-101-NH2 (Fig. 2).26 The large mesopores of MIL-101-NH2 (pore diameters: 29 & 34 A˚ ) allowed chemical reductants such as cobaltocene (CoCp2) to access a greater fraction of diiron sites by circumventing the mass transport limitations of 2 (UiO-66 pore diameter: 6 & 10 A˚ ). However, IR characterization indicated that ion pairing between [CoCp2]+ and reduced [FeFe]2− species at high catalyst loadings resulted in pore clogging and inhibited substrate access to the active site. 3 also showed good activity for photocatalytic hydrogen evolution, and IR investigations indicated that 80%–85% of the diiron sites were involved in catalysis compared to 95% isotactic poly(1-hexene) in the first report of a MOF-catalyzed stereoregular polymerization. Lin and co-workers used a straightforward grafting approach with MgMe2 to generate O–MgMe species at the Zr oxo nodes of the extended UiO-type MOF (G63).117 The single-site MOF catalyst, 63-MgMe, catalyzed the hydroboration of ketones and imines with TONs as high as 84,000. It also facilitated intramolecular hydroamination of aminoalkenes.

Fig. 10 Heterobimetallic Rh-Ga and monometallic Rh complexes grafted at the nodes of NU-1000.118

Organometallic Functionalized MOFs - Reactivity and Catalysis

275

Lu, Farha and co-workers grafted a heterobimetallic Rh-Ga complex (G64-RhGa) and a monometallic Rh analog (64-Rh), at the nodes of NU-1000 (Fig. 10).118 The products were characterized using a combination of synchrotron X-ray powder diffraction, X-ray absorption spectroscopy, and density functional theory, revealing formation of Rh-OH-Zr linkages with minimal alteration in the structure of the original complexes. 64-RhGa catalyzed semihydrogenation of diphenylacetylene to E-stilbene in 93% yield while 64-Rh fully hydrogenated the substrate, generating bibenzyl in 99% yield. The difference in selectivity persisted over a range of substrates and revealed the role of Rh-Ga cooperative interactions in influencing activity and selectivity. MOF-808, composed of tritopic 1,3,5-benzenetricarboxylate linkers and 6-connected Zr or Hf oxo clusters (Table 1), has also been used as platform for grafting organometallic species. Wright, Clarke, and co-workers functionalized the nodes of Hf-MOF-808 with sulfonated triarylphosphine ligands as anchor points for organometallic catalyst species.119 3-(diphenylphosphino)benzene sulfonic acid (G65a) and bis(p-sulfonatophenyl) phenylphosphine (G65b) were used to prepare phosphine-functionalized MOFs. Subsequent metalation using [IrCl(cod)]2 or Rh(acac)(CO)2 generated MOF catalysts 65a-Ir, 65a-Rh, 65b-Ir, and 65b-Rh (Scheme 22). 65a-Ir and 65b-Ir catalyzed reductive amination of 4-fluoroacetophenone with benzylamine to the corresponding secondary amine. A series of control reactions indicated that the tandem reaction relies on Lewis acidic Hf framework sites to facilitate the imine condensation step while the Ir-phosphine species catalyze imine hydrogenation. The Ir catalysts were reusable through five cycles, and 65a-Ir also exhibited compatibility with a small library of substrates. However, 65a-Rh and 65b-Rh were not competent catalysts for the tandem transformation, a finding that was consistent with homogeneous control reactions. Nevertheless, all four metalated MOFs catalyzed the more easily accessible reductive amination of benzaldehyde with aniline as well as the hydroaminomethylation of cyclohexene and cyclopentene with substituted anilines, demonstrating the utility of 65a and 65b as insoluble “porous phosphine ligands.”

Scheme 22

Many other notable MOF catalysts have been developed by grafting transition metal species at metal oxide nodes, but are not strictly defined as organometallic. Solvothermal deposition and atomic layer deposition methods have afforded heterogeneous MOF catalysts that are active for the dehydrogenation of propane to propene,120 hydrogenation of ethylene to ethane,121 vapor-phase epoxidation of cyclohexene,122 hydrosilylation of esters,123 homocoupling of o-xylene,124 oxidative alkenylation of arenes,125 borylation and silylation of benzylic C-H bonds,126 amination of alkenes,126 and hydrogenation of carbon dioxide to ethanol.127 Grafting approaches have also yielded MOF catalysts for the hydroboration of alkynes,123 nitriles,123 aldehydes,126 ketones,126 and alkenes,126 as well as hydrogenation of alkenes,126,128 imines,128 ketones,128 aldehydes,128 and arenes.129

14.08.4 Non-covalently encapsulated organometallic species In the cases discussed above, organometallic species were directly bound to a MOF host via covalent or dative bonds. However, there are numerous examples in which a guest is instead trapped within the pores of a MOF by noncovalent interactions. This approach generally removes the need for elaborate linker design and synthesis, making it compatible with a wide range of materials and catalytic guests. However, it also gives rise to concerns regarding catalyst leaching, decreased porosity, and unfavorable guestframework interactions. This section describes recent efforts toward the encapsulation of organometallic species within MOFs for the design of heterogeneous catalysts. In an early example of noncovalent encapsulation, Fischer and co-workers compared the guest adsorption, thermal, and electrochemical properties of several metallocene complexes encapsulated within MIL-53(Al).130 Subsequent reports have expanded upon this strategy, with the most commonly employed method relying on electrostatic interactions to retain guest species within a MOF. For example, Sanford and co-workers used the anionic indium MOF ZJU-28 (66) to encapsulate a series of cationic transition metal complexes, including [Pd(MeCN)4][BF4]2, [FeCp(CO)2(THF)]BF4, [Ir(cod)(PCy3)(py)]PF6, [Rh(dppe) (cod)]BF4, and [Ru(Cp )(MeCN)3]OTf.131 The ion exchange reactions resulted in displacement of 24%–35% of internal dimethylammonium cations in the MOF pores. Attempts to incorporate neutral complexes or those larger than the MOF pores were

276

Organometallic Functionalized MOFs - Reactivity and Catalysis

unsuccessful, reflecting the electrostatic driving force for encapsulation. The MOF-encapsulated [Rh(dppe)(cod)]+ (E66-Rh-a) exhibited similar activity as the homogeneous complex, [Rh(dppe)(cod)]BF4, for catalytic hydrogenation of 1-octene but was recyclable through five cycles (Fig. 11). A follow-up study examined the catalytic activity of the cationic complexes [(dppe)Rh(cod)]+ (Rh-a) and [(MeCN)2Rh(cod)]+ (Rh-b) encapsulated in both ZJU-28 (66) and MIL-101-SO3 (67).132 The latter framework contains large mesopores, exhibits good thermal stability, and bears localized anionic charges on the –SO−3 linker groups. E66-Rh-a and E67-Rh-a exhibited good activity (4000 TON) for hydrogenation of 1-octene to n-octane and could be recycled up to four times without any significant decrease in activity. ICP-OES analysis showed that E66-Rh-a was stable against leaching, but a small amount of the Rh complex was leached from E67-Rh-a after four cycles. However, control experiments indicated that the leached species were not responsible for the observed hydrogenation activity. A kinetic study revealed that E67-Rh-a was considerably faster than E66-Rh-a for the hydrogenation of 1-octene, which can be attributed to the larger pore size allowing more facile substrate and product transport through the framework. Comparison of the MOF catalysts using a more challenging substrate, 2,3-dimethylbutene, confirmed E67-Rh-a as the more active catalyst. E66-Rh-b and E67-Rh-b both exhibited significantly lower activity and poor recyclability.

Fig. 11 Cationic Rh complexes encapsulated in ZJU-28 and MIL-101-SO3.131,132

Weller, Rosseinsky, and coworkers investigated the encapsulation of [CpFe(CO)2(L)]+ (L ¼ tetrahydrofuran or acetone) in a different anionic In-based MOF (68, [Et4N]3[In3(btc)4], btc ¼ 1,3,5-benzenetricarboxylate) (Fig. 12).133 Preliminary studies used [Cp2Co]+ to probe the accessibility and stability of the framework toward substitution of the charge-balancing Et4N+ cations. Approximately 53% of the Et4N+ cations could be exchanged, and a single crystal X-ray diffraction study revealed that the [Cp2Co]+ guests interact weakly with the framework, leaving guest-accessible porosity. Cation exchange with [CpFe(CO)2(L)]+ resulted in substitution of 17%–24% of the Et4N+ cations (E68-Fe) and was supported by observation of the corresponding CO stretching bands in the IR spectrum of the product. E68-Fe catalyzed the Diels-Alder reaction between isoprene and methyl vinyl ketone to form isomers of methylacetylcyclohexene in modest yield. The parent MOF 68 did not catalyze the Diels-Alder reaction, although the homogeneous complex [CpFe(CO)2(THF)][BF4] was found to be active. Nevertheless, the encapsulated catalyst could be recycled two times with only a modest decrease in activity and no observed iron leaching.

Fig. 12 Cationic complexes encapsulated in anionic MOFs.133,134

Ion exchange has also been employed by several groups to incorporate multinuclear palladium clusters into MOFs.135,136 In one such example, Ma and co-workers introduced a cationic trinuclear palladium cluster into the anionic bio-MOF-100 framework (E69-Pd), replacing dimethylammonium cations in the pores (Fig. 12).134 The resulting composite catalyzed semihydrogenation of 1-phenylpropyne and diphenylethyne with activity similar to the homogeneous cluster, but the rate of catalytic turnover was significantly slower for the MOF-encapsulated catalyst. E69-Pd could also be reused for up to three catalytic runs, although the product yields decreased significantly from 95% in the first cycle to 69% in the third.

Organometallic Functionalized MOFs - Reactivity and Catalysis

277

Many subsequent advances in electrostatically-driven, noncovalent encapsulation have been described for variants of MIL-101. For example, Grela, Chmielewski, and co-workers investigated the adsorption of second-generation Hoveyda-Grubbs olefin metathesis catalysts (HGa-g) in (Cr)MIL-101 (70) and (Al)MIL-101-NH2 (71) (Fig. 13).137 While HGa-c were readily adsorbed into the mesoporous frameworks, they readily desorbed in the presence of solvents such as toluene and CH2Cl2. On the other hand, the cationic complexes HGd-g could only be desorbed with alcoholic solvents such as isopropanol.

Fig. 13 Second-generation Hoveyda-Grubbs metathesis catalysts encapsulated in MIL-101 variants.137–139

HGd-f showed modest activity for ring-closing olefin metathesis when adsorbed into 71, but were completely inactive when 70 was used as the host material. The latter observation suggested that the catalysts were being deactivated by interactions with the framework. However, a dramatic increase in catalytic activity was observed when the MOFs were treated with HCl either before or after catalyst adsorption. This phenomenon was attributed to quenching of Lewis acidic framework metal sites by Cl− or inhibition of deactivation mechanisms related to the presence of OH− or F−. Consequently, the HCl-treated composites E71-HGd and E71HGg catalyzed ring-closing metathesis of a small library of olefins with over 75% conversion. Strong electrostatic interactions between the MOF and catalyst complexes prevented leaching over the course of 24 h, leading to a TON approaching 9000 in batch reactions and the ability to run ring-closing metathesis in flow over 8 h for a TON of 4700. Unfortunately, leaching occurred in green solvents such as dimethyl carbonate or isopropanol. This drawback was later addressed by encapsulating HGc into the Brønsted acidic MOF (Cr)MIL-101-SO3H (72).138 This composite exhibited excellent resistance to catalyst leaching in polar solvents, presumably due to an acid-base reaction between the framework –SO3H groups and the piperidine substituent of HGc. E72-HGc catalyzed the ring-closing metathesis of diethyl diallylmalonate to 100% conversion after just 4 h. Moreover, the reaction could be carried out in ethyl acetate solvent, which caused catalyst leaching for E71-HGd and E71-HGg. Using a similar electrostatic encapsulation approach, Grela and co-workers prepared a robust MOF catalyst by exchanging sodium cations for HGg in (Cr)MIL101-SO3Na (73).139 The resulting framework, E73-HGg, catalyzed the ring-closing metathesis of various substrates, including several intermediates relevant to the synthesis of pharmaceutical products. Rosseinsky and co-workers employed a similar strategy to encapsulate Crabtree’s catalyst ([Ir(cod)(PCy3)(py)]+) in (Cr)MIL-101-SO3Na, forming E73-Ir (Fig. 14).140 This MOF catalyst was highly active for alkene hydrogenation, reaching TONs similar to its homogeneous counterpart with the benefit of extended catalyst lifetime. In addition, the hydrophilic pore environment of the MOF significantly improved selectivity for the hydrogenation of unsaturated alcohols over isomerization.

Fig. 14 Hydrogenation pre-catalysts immobilized by encapsulation in MOFs.140–144

278

Organometallic Functionalized MOFs - Reactivity and Catalysis

Although electrostatic encapsulation offers a means of robust catalyst immobilization, it is still dependent on specific host-guest interactions and can require that additional functional groups be added to the molecular catalyst. Byers, Tsung, and co-workers developed a robust guest encapsulation method that takes advantage of dissociative “aperture-opening” linker exchange reactions to internalize neutral guests larger than the original pore aperture size of a MOF.141 The aperture-opening strategy eschews the need for strong electrostatic attractions by trapping the catalyst in large pores with small windows that prevent it from diffusing out. This method was first employed to encapsulate a (PNP)Ru pincer complex in UiO-66 to form E74-PNP-Ru (Fig. 14). The MOF composite exhibited similar activity to its homogeneous counterpart for the hydrogenation of CO2 to formate with no decrease in TON over five cycles.142 Active site isolation within the pores inhibited bimolecular decomposition pathways and poisoning of the catalyst by bulky thiols. This isolation was leveraged in subsequent work to address incompatibility between (PNP)Ru and a (PNN)Ru catalyst known to catalyze ester hydrogenation.143 The mixture of E74-PNP-Ru and homogeneous (PNN)Ru facilitated the cascade hydrogenation of CO2 to methanol with activity comparable to the most active homogeneous complexes. (PNN)Ru was also encapsulated in UiO-66 to form E74-PNN-Ru and mixed with E74-PNP-Ru to afford a dual heterogeneous catalyst system that was recyclable over five cycles. In a subsequent report, an ammonium-functionalized MOF, UiO-66-NH+3, was used to encapsulate (PNP)Ru, forming E75-PNP-Ru. This heterogeneous composite in combination with homogeneous (PNN)Ru was found to hydrogenate CO2 to methanol with an initial TOF of 9100 h−1.144 The Brønsted acidity of the linker-based –NH+3 groups served to improve the efficiency of the CO2 to formic acid hydrogenation step. When E75-PNP-Ru was combined with E74-PNN-Ru, the resulting tandem heterogeneous system could be recycled at least 10 times with a cumulative TON of 100,000.

14.08.5 Organometallic metal nodes While significant advances have been made toward the design of MOFs containing organometallic catalysts, materials with organometallic metal nodes remain relatively rare. This section describes recent examples of adsorption processes and post-synthetic reactions that generate well-characterized metal-carbon bonds at framework metal sites as well as MOF catalysts for which metal-carbon bond formation at the nodes is implied. Finally, it details emerging examples of MOFs assembled with metal-carbon bonds providing structural support. Long and co-workers studied Fe-MOF-74 (76) for adsorptive separation of various saturated and unsaturated hydrocarbons (Scheme 23).145 The MOF exhibits a strong preference for the adsorption of unsaturated hydrocarbons over saturated species such as methane, ethane, and propane. This selectivity originates from interaction of the organic adsorbates with coordinatively unsaturated Fe(II) sites at the chain-type metal nodes of the framework. Neutron powder diffraction studies revealed Z2 coordination of acetylene, ethylene, and propylene at the Fe(II) centers and much weaker interactions between the metal sites and C–H groups of ethane and propane. Reports from the same group have detailed spin state changes that occur upon CO adsorption in a series of Fe(II) triazolate MOFs (77, 78, 79, Scheme 24).146,147 The high-spin Fe(II) centers in the tetranuclear metal node of 77 (Fe-BTTri) assembled from 1,3,5-tris(1,2,3-triazol-5-yl)benzene (L77) linkers, switch to the low spin state upon CO binding at the open coordination site. This phenomenon was characterized using a suite of spectroscopic and magnetic techniques. The spin state change makes 77 selective for CO adsorption over H2, N2, CO2, CH4, and hydrocarbons and results in a large low-pressure CO capacity (1.45 mmol CO per g MOF at 100 mbar pressure). A high- to low-spin state transition was also observed upon CO adsorption in the chain-type benzotriazolate MOFs 78 and 79. Both MOFs exhibited step-shaped CO isotherms that were not consistent with their rigid structures and the odd adsorption behavior was attributed to cooperative spin-transition effects between adjacent iron centers.

Scheme 23

Organometallic Functionalized MOFs - Reactivity and Catalysis

279

Scheme 24

There have also been recent developments in the design of MOF catalysts with organometallic nodes. Lin and co-workers used a series of post-synthetic reactions to produce Zr-alkyl species at the 6-connected Zr oxo nodes of MOF-808 (80, Scheme 25). An organolithium reagent facilitated the formation of Zr-CH2SiMe3 groups (80-CH2SiMe3) while MMAO-12 generated Zr-Me species (80-Me) at the nodes. The products were characterized by solid-state 13C NMR and EXAFS, both of which supported the formation of Zr-alkyl groups. In situ generated 80-Me catalyzed ethylene polymerization to afford high-molecular-weight polyethylene (M.W. > 300,000 g/mol) with well-controlled chain length (dispersity 99% selectivity and high molecular weight (700,000 g/mol).154 Separately, a Ni(II)-substituted analog, 81-Ni, showed better than 97% selectivity for the dimerization of ethylene to 1-butene at a TOF of 41,500 h−1.155 There have been many other reports of olefin oligomerization and polymerization using single-site MOF catalysts, but few examples in which the active site has been shown to contain well-defined organometallic groups.116,148,156 Figueroa and co-workers have pioneered the use of ditopic m-terphenyl isocyanide linkers for the assembly of Cu- and Ni-based MOFs.157–159 These structures are analogous to homogeneous isocyanide complexes prepared and studied by the same group.160,161 The 3D framework 82 was prepared by solvothermal reaction of L82 and [Cu(MeCN)4]PF6 in THF solvent (Scheme 27). The sterically encumbering mesityl substituents on the linkers help direct framework assembly and offer steric protection to the monometallic nodes. Single crystal X-ray diffraction showed that the MOF adopts a 2-fold interpenetrated diamondoid framework with monometallic Cu(CNAr)4 nodes. 82 undergoes a solid-state transformation from a 3D to a 2D coordination network upon heating at 125  C. The new phase, 83, was characterized by single crystal X-ray diffraction and found to contain Cu(I) tris-isocyanide nodes via linker loss from the framework. A MOF containing Cu(I) tris-isocyanide nodes could also be prepared by direct solvothermal assembly. Unfortunately, these materials were not stable to thermal activation, precluding gas adsorption and porosity measurements. 84, a non-interpenetrated Ni0-analog of 82, proved to be stable to thermal activation and exhibited N2 uptake and a BET surface area of 580 m2/g, which was commensurate with the framework structure and predicted surface area.158

Scheme 27

Organometallic Functionalized MOFs - Reactivity and Catalysis

281

Scheme 28

An extended m-terphenyl isocyanide linker L85 has also been used for MOF assembly with [Cu(MeCN)4]PF6 (Scheme 28).159 The resulting material, 85, is comprised of mononuclear Cu(CNAr)3(THF) nodes with a fourfold interpenetrated layered structure. Ligand exchange studies revealed that the Cu-bound THF ligand could be readily exchanged with pyridine. Although catalytic applications have yet to be developed for these uniquely organometallic MOFs, their synthesis and characterization demonstrates the feasibility of using low valent metal species and metal-carbon bond formation for MOF self-assembly. For example, in recent report, Dong and co-workers prepared the Pd-based MOF 86 using tritopic isocyanide linkers L86 (Scheme 28).162 Single crystals suitable for X-ray diffraction could not be obtained, but PXRD data was used to infer the formation of a layered structure built upon trans-PdI2(CNAr)2 metal nodes. The organometallic framework was shown to catalyze the Suzuki coupling of a small library of aryl bromides and aryl iodides with phenylboronic acid.

Acknowledgement The authors thank the US National Science Foundation (CHE-2044904) for support of this work.

References 1. Yuan, S.; Feng, L.; Wang, K.; Pang, J.; Bosch, M.; Lollar, C. T.; Sun, Y.; Qin, J.-S.; Yang, X.; Zhang, P.; Wang, Q.; Zou, L.; Zhang, Y.; Zhang, L.; Fang, Y.; Li, J.; Zhou, H.-C. Adv. Mater. 2018, 30, 1704303. 2. Furukawa, H.; Cordova, K. E.; O’Keeffe, M.; Yaghi, O. M. Science 2013, 341, 1230444. 3. Redfern, L. R.; Farha, O. K. Chem. Sci. 2019, 10, 10666–10679. 4. Li, H.; Wang, K.; Sun, Y.; Lollar, C. T.; Li, J.; Zhou, H.-C. Mater. Today 2018, 21, 108–121. 5. Li, H.-Y.; Zhao, S.-N.; Zang, S.-Q.; Li, J. Chem. Soc. Rev. 2020, 49, 6364–6401. 6. Sun, Y.; Zheng, L.; Yang, Y.; Qian, X.; Fu, T.; Li, X.; Yang, Z.; Yan, H.; Cui, C.; Tan, W. Nano-Micro Lett. 2020, 12, 103. 7. Yang, D.; Gates, B. C. ACS Catal. 2019, 9, 1779–1798. 8. Rogge, S. M. J.; Bavykina, A.; Hajek, J.; Garcia, H.; Olivos-Suarez, A. I.; Sepúlveda-Escribano, A.; Vimont, A.; Clet, G.; Bazin, P.; Kapteijn, F.; Daturi, M.; RamosFernandez, E. V.; Xamena, F. X.; Van Speybroeck, V.; Gascon, J. Chem. Soc. Rev. 2017, 46, 3134–3184. 9. Kalaj, M.; Cohen, S. M. ACS Cent. Sci. 2020, 6, 1046–1057. 10. Dinca, M.; Gabbaï, F. P.; Long, J. R. Organometallics 2019, 38, 3389–3391. 11. Crabtree, R. H. The Organometallic Chemistry of the Transition Metals; John Wiley & Sons, Ltd, 2014; pp 1–39.

282

Organometallic Functionalized MOFs - Reactivity and Catalysis

12. Hafizovic Cavka, J.; Jakobsen, S.; Olsbye, U.; Guillou, N.; Lamberti, C.; Bordiga, S.; Petter Lillerud, K.; Cavka, J. H.; Lillerud, K.-P. J. Am. Chem. Soc. 2008, 130, 13850–13851. 13. Mondloch, J. E.; Bury, W.; Fairen-Jimenez, D.; Kwon, S.; DeMarco, E. J.; Weston, M. H.; Sarjeant, A. A.; Nguyen, S. T.; Stair, P. C.; Snurr, R. Q.; Farha, O. K.; Hupp, J. T. J. Am. Chem. Soc. 2013, 135, 10294–10297. 14. Furukawa, H.; Gándara, F.; Zhang, Y.-B.; Jiang, J.; Queen, W. L.; Hudson, M. R.; Yaghi, O. M. J. Am. Chem. Soc. 2014, 136, 4369–4381. 15. Férey, G.; Mellot-Draznieks, C.; Serre, C.; Millange, F.; Dutour, J.; Surblé, S.; Margiolaki, I. Science 2005, 309, 2040–2042. 16. Denysenko, D.; Grzywa, M.; Tonigold, M.; Streppel, B.; Krkljus, I.; Hirscher, M.; Mugnaioli, E.; Kolb, U.; Hanss, J.; Volkmer, D. Chem. Eur. J. 2011, 17, 1837–1848. 17. Lochan, R. C.; Khaliullin, R. Z.; Head-Gordon, M. Inorg. Chem. 2008, 47, 4032–4044. 18. Vitillo, J. G.; Groppo, E.; Bordiga, S.; Chavan, S. M.; Ricchiardi, G.; Zecchina, A. Inorg. Chem. 2009, 48, 5439–5448. 19. Kaye, S. S.; Long, J. R. J. Am. Chem. Soc. 2007, 130, 806–807. 20. Chavan, S. M.; Vitillo, J. G.; Uddin, M. J.; Bonino, F.; Lamberti, C.; Groppo, E.; Lillerud, K.-P.; Bordiga, S. Chem. Mater. 2010, 22, 4602–4611. 21. Chavan, S. M.; Vitillo, J. G.; Larabi, C.; Alessandra Quadrelli, E.; Dietzel, P. D. C.; Bordiga, S. Micropor. Mesopor. Mat. 2012, 157, 56–61. 22. Saito, M.; Toyao, T.; Ueda, K.; Kamegawa, T.; Horiuchi, Y.; Matsuoka, M. Dalton Trans. 2013, 42, 9444–9447. 23. Afzali, N.; Tangestaninejad, S.; Moghadam, M.; Mirkhani, V.; Mechler, A.; Mohammadpoor-Baltork, I.; Kardanpour, R.; Zadehahmadi, F. Appl. Organomet. Chem. 2018, 32, e3958. 24. Burrows, A. D.; Fisher, L. C.; Hodgson, D.; Mahon, M. F.; Cessford, N. F.; Düren, T.; Richardson, C.; Rigby, S. P. CrystEngComm 2012, 14, 188–192. 25. Pullen, S.; Fei, H.; Orthaber, A.; Cohen, S. M.; Ott, S. J. Am. Chem. Soc. 2013, 135, 16997–17003. 26. Roy, S.; Pascanu, V.; Pullen, S.; González Miera, G.; Martín-Matute, B.; Ott, S. Chem. Commun. 2017, 53, 3257–3260. 27. Blake, A. J.; Champness, N. R.; Easun, T. L.; Allan, D. R.; Nowell, H.; George, M. W.; Jia, J.; Sun, X.-Z. Nat. Chem. 2010, 2, 688–694. 28. Easun, T. L.; Jia, J.; Calladine, J. A.; Blackmore, D. L.; Stapleton, C. S.; Vuong, K. Q.; Champness, N. R.; George, M. W. Inorg. Chem. 2014, 53, 2606–2612. 29. Min Choi, K.; Kim, D.; Rungtaweevoranit, B.; Trickett, C. A.; Trese Deniz Barmanbek, J.; Alshammari, A. S.; Yang, P.; Yaghi, O. M. J. Am. Chem. Soc. 2016, 139, 356–362. 30. Deming, D. A.; Hurlock, M. J.; Li, X.; Kriegsman, K. W.; Ding, G.; Guo, X.; Zhang, Q. J. Organomet. Chem. 2019, 897, 114–119. 31. Bloch, W. M.; Burgun, A.; Coghlan, C. J.; Lee, R.; Coote, M. L.; Doonan, C. J.; Sumby, C. J. Nat. Chem. 2014, 6, 906–912. 32. Burgun, A.; Coghlan, C. J.; Huang, D. M.; Chen, W.; Horike, S.; Kitagawa, S.; Alvino, J. F.; Metha, G. F.; Sumby, C. J.; Doonan, C. J. Angew. Chem. Int. Ed. 2017, 56, 8412–8416. 33. Huxley, M. T.; Young, R. J.; Bloch, W. M.; Champness, N. R.; Sumby, C. J.; Doonan, C. J. Organometallics 2019, 38, 3412–3418. 34. Huxley, M. T.; Burgun, A.; Ghodrati, H.; Coghlan, C. J.; Lemieux, A.; Champness, N. R.; Huang, D. M.; Doonan, C. J.; Sumby, C. J. J. Am. Chem. Soc. 2018, 140, 6416–6425. 35. Peralta, R. A.; Huxley, M. T.; Evans, J. D.; Fallon, T.; Cao, H.; He, M.; Song Zhao, X.; Agnoli, S.; Sumby, C. J.; Doonan, C. J. J. Am. Chem. Soc. 2020, 142, 13533–13543. 36. Hopkinson, M. N.; Richter, C.; Schedler, M.; Glorius, F. Nature 2014, 510, 485–496. 37. Ezugwu, C. I.; Kabir, N. A.; Yusubov, M.; Verpoort, F. Coord. Chem. Rev. 2016, 307, 188–210. 38. Crees, R. S.; Cole, M. L.; Hanton, L. R.; Sumby, C. J. Inorg. Chem. 2010, 49, 1712–1719. 39. Chun, J.; Gu Jung, I.; Jin Kim, H.; Park, M.; Soo Lah, M.; Uk Son, S. Inorg. Chem. 2009, 48, 6353–6355. 40. Chun, J.; Sol Lee, H.; Gu Jung, I.; Won Lee, S.; Jin Kim, H.; Uk Son, S. Organometallics 2010, 29, 1518–1521. 41. Capon, P. K.; Burgun, A.; Coghlan, C. J.; Crees, R. S.; Doonan, C. J.; Sumby, C. J. CrystEngComm 2016, 18, 7003–7010. 42. Lee, H.-J.; Kwon, H.; Sim, J.; Song, D.; Kim, Y.; Kim, J.; Kim, K.; Lee, E. CrystEngComm 2017, 19, 1528–1534. 43. Burgun, A.; Crees, R. S.; Cole, M. L.; Doonan, C. J.; Sumby, C. J. Chem. Commun. 2014, 50, 11760–11763. 44. Yao, Q.; Lu, X.; Liu, K.; Ma, C.; Su, J.; Lin, C.; Li, D.; Dou, J.; Sun, J.; Duan, W. Dalton Trans. 2019, 48, 5144–5148. 45. Oisaki, K.; Li, Q.; Furukawa, H.; Czaja, A. U.; Yaghi, O. M. J. Am. Chem. Soc. 2010, 132, 9262–9264. 46. Carson, F.; Martínez-Castro, E.; Marcos, R.; Miera, G. G.; Jansson, K.; Zou, X.; Martín-Matute, B. Chem. Commun. 2015, 51, 10864–10867. 47. Kong, G.-Q.; Xu, X.; Zou, C.; Wu, C.-D. Chem. Commun. 2011, 47, 11005–11007. 48. Kong, G.-Q.; Ou, S.; Zou, C.; Wu, C.-D. J. Am. Chem. Soc. 2012, 134, 19851–19857. 49. Dong, Y.; Li, Y.; Wei, Y.-L.; Wang, J.-C.; Ma, J.-P.; Ji, J.; Yao, B.-J.; Dong, Y.-B. Chem. Commun. 2016, 52, 10505–10508. 50. Wei, Y.-L.; Li, Y.; Chen, Y.-Q.; Dong, Y.; Yao, J.-J.; Han, X.-Y.; Dong, Y.-B. Inorg. Chem. 2018, 57, 4379–4386. 51. Dong, Y.; Li, W.-H.; Dong, Y.-B. J. Org. Chem. 2021, 86, 1818–1826. 52. Niknam, E.; Panahi, F.; Khalafi-Nezhad, A. Appl. Organomet. Chem. 2020, 34, e5470. 53. Niknam, E.; Panahi, F.; Khalafi-Nezhad, A. J. Organomet. Chem. 2021, 935, 121676. 54. Downing, J. H.; Smith, M. B. In Comprehensive Coordination Chemistry II; McCleverty, J. A., Meyer, T. J., Eds.; Pergamon: Oxford, 2003; pp 253–296. 55. Falkowski, J. M.; Sawano, T.; Zhang, T.; Tsun, G.; Chen, Y.; Lockard, J. V.; Lin, W. J. Am. Chem. Soc. 2014, 136, 5213–5216. 56. Sawano, T.; Thacker, N. C.; Lin, Z.; McIsaac, A. R.; Lin, W. J. Am. Chem. Soc. 2015, 137, 12241–12248. 57. Bohnsack, A. M.; Ibarra, I. A.; Bakhmutov, V. I.; Lynch, V. M.; Humphrey, S. M. J. Am. Chem. Soc. 2013, 135, 16038–16041. 58. Nuñez, A. J.; Chang, M. S.; Ibarra, I. A.; Humphrey, S. M. Inorg. Chem. 2013, 53, 282–288. 59. Nuñez, A. J.; Shear, L. N.; Dahal, N.; Ibarra, I. A.; Yoon, J.; Hwang, Y. K.; Chang, J.-S.; Humphrey, S. M. Chem. Commun. 2011, 47, 11855–11857. 60. Václavík, J.; Servalli, M.; Lothschütz, C.; Szlachetko, J.; Ranocchiari, M.; van Bokhoven, J. A. ChemCatChem 2013, 5, 692–696. 61. Dunning, S. G.; Nandra, G.; Conn, A. D.; Chai, W.; Sikma, R. E.; Lee, J. S.; Kunal, P.; Reynolds, J. E., III; Chang, J.-S.; Steiner, A.; Henkelman, G.; Humphrey, S. M. Angew. Chem. Int. Ed. 2018, 57, 9295–9299. 62. Dunning, S. G.; Reynolds, J. E., III; Walsh, K. M.; Kristek, D. J.; Lynch, V. M.; Kunal, P.; Humphrey, S. M. Organometallics 2019, 38, 3406–3411. 63. Sawano, T.; Lin, Z.; Boures, D.; An, B.; Wang, C.; Lin, W. J. Am. Chem. Soc. 2016, 138, 9783–9786. 64. Feng, X.; Song, Y.; Li, Z.; Kaufmann, M.; Pi, Y.; Chen, J. S.; Xu, Z.; Li, Z.; Wang, C.; Lin, W. J. Am. Chem. Soc. 2019, 141, 11196–11203. 65. Morel, F. L.; Ranocchiari, M.; van Bokhoven, J. A. Ind. Eng. Chem. Res. 2014, 53, 9120–9127. 66. Morel, F. L.; Pin, S.; Huthwelker, T.; Ranocchiari, M.; van Bokhoven, J. A. Phys. Chem. Chem. Phys. 2015, 17, 3326–3331. 67. Beloqui Redondo, A.; Morel, F. L.; Ranocchiari, M.; van Bokhoven, J. A. ACS Catal. 2015, 5, 7099–7103. 68. Jiang, H.; Zhang, W.; Kang, X.; Cao, Z.; Chen, X.; Liu, Y.; Cui, Y. J. Am. Chem. Soc. 2020, 142, 9642–9652. 69. He, J.; Waggoner, N. W.; Dunning, S. G.; Steiner, A.; Lynch, V. M.; Humphrey, S. M. Angew. Chem. Int. Ed. 2016, 55, 12351–12355. 70. He, J.; Bohnsack, A. M.; Waggoner, N. W.; Dunning, S. G.; Lynch, V. M.; Kaska, W. C.; Humphrey, S. M. Polyhedron 2018, 143, 149–156. 71. Burgess, S. A.; Kassie, A. A.; Baranowski, S. A.; Fritzsching, K. J.; Schmidt-Rohr, K.; Brown, C. M.; Wade, C. R. J. Am. Chem. Soc. 2016, 138, 1780–1783. 72. Kassie, A. A.; Wade, C. R. Organometallics 2020, 39, 2214–2221. 73. Reiner, B. R.; Mucha, N. T.; Rothstein, A.; Sebastian Temme, J.; Duan, P.; Schmidt-Rohr, K.; Foxman, B. M.; Wade, C. R. Inorg. Chem. 2018, 57, 2663–2672. 74. Reiner, B. R.; Kassie, A. A.; Wade, C. R. Dalton Trans. 2019, 48, 9588–9595. 75. Kassie, A. A.; Duan, P.; McClure, E. T.; Schmidt-Rohr, K.; Woodward, P. M.; Wade, C. R. Inorg. Chem. 2019, 58, 3227–3236. 76. Kassie, A. A.; Duan, P.; Gray, M. B.; Schmidt-Rohr, K.; Woodward, P. M.; Wade, C. R. Organometallics 2019, 38, 3419–3428. 77. Bruce, M. I. Angew. Chem. Int. Ed. 1977, 16, 73–86. 78. Albrecht, M. Chem. Rev. 2009, 110, 576–623. 79. Wang, C.; Xie, Z.; DeKrafft, K. E.; Lin, W. J. Am. Chem. Soc. 2011, 133, 13445–13454.

Organometallic Functionalized MOFs - Reactivity and Catalysis 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140.

283

Barrett, S. M.; Wang, C.; Lin, W. J. Mater. Chem. 2012, 22, 10329–10334. Wang, C.; Lin, W. J. Am. Chem. Soc. 2011, 133, 4232–4235. Dau, P. V.; Kim, M.; Garibay, S. J.; Münch, F. H. L.; Moore, C. E.; Cohen, S. M. Inorg. Chem. 2012, 51, 5671–5676. Dau, P. V.; Kim, M.; Cohen, S. M. Chem. Sci. 2013, 4, 601–605. Quan, Y.; Song, Y.; Shi, W.; Xu, Z.; Chen, J. S.; Jiang, X.; Wang, C.; Lin, W. J. Am. Chem. Soc. 2020, 142, 8602–8607. Lee, J. S.; Kapustin, E. A.; Pei, X.; Llopis, S.; Yaghi, O. M.; Toste, F. D. Chem 2020, 6, 142–152. Li, L.; Tang, S.; Wang, C.; Lv, X.; Jiang, M.; Wu, H.; Zhao, X. Chem. Commun. 2014, 50, 2304–2307. Fei, H.; Cohen, S. M. Chem. Commun. 2014, 50, 4810–4812. Øien, S.; Agostini, G.; Svelle, S.; Borfecchia, E.; Lomachenko, K. A.; Mino, L.; Gallo, E.; Bordiga, S.; Olsbye, U.; Petter Lillerud, K.; Lamberti, C. Chem. Mater. 2015, 27, 1042–1056. Manna, K.; Zhang, T.; Lin, W. J. Am. Chem. Soc. 2014, 136, 6566–6569. Gonzalez, M. I.; Oktawiec, J.; Long, J. R. Faraday Discuss. 2017, 201, 351–367. Ryu, U. J.; Kim, S. J.; Lim, H.-K.; Kim, H.; Choi, K. M.; Kang, J. K. Sci. Rep. 2017, 7, 612. Wang, C.; Wang, J.-L.; Lin, W. J. Am. Chem. Soc. 2012, 134, 19895–19908. Zhu, Y.-Y.; Lan, G.; Fan, Y.; Veroneau, S. S.; Song, Y.; Micheroni, D.; Lin, W. Angew. Chem. Int. Ed. 2018, 57, 14090–14094. Manna, K.; Zhang, T.; Greene, F. X.; Lin, W. J. Am. Chem. Soc. 2015, 137, 2665–2673. Thacker, N. C.; Ji, P.; Lin, Z.; Urban, A.; Lin, W. Faraday Discuss. 2017, 201, 303–315. Zhang, X.; Huang, Z.; Ferrandon, M. S.; Yang, D.; Robison, L.; Li, P.; Wang, T. C.; Delferro, M.; Farha, O. K. Nat. Catal. 2018, 1, 356–362. Feng, L.; Wang, K.-Y.; Joseph, E.; Zhou, H.-C. Trends Chem. 2020, 2, 555–568. Rajasree, S. S.; Li, X.; Deria, P. Commun. Chem. 2021, 4, 47. Fang, Z.-B.; Liu, T.-T.; Liu, J.; Jin, S.; Wu, X.-P.; Gong, X.-Q.; Wang, K.; Yin, Q.; Liu, T.-F.; Cao, R.; Zhou, H.-C. J. Am. Chem. Soc. 2020, 142, 12515–12523. Jin, J. New J. Chem. 2020, 44, 15362–15368. Zhang, L.; Yuan, S.; Feng, L.; Guo, B.; Qin, J.-S.; Xu, B.; Lollar, C. T.; Sun, D.; Zhou, H.-C. Angew. Chem. Int. Ed. 2018, 57, 5095–5099. Carrasco, S.; Sanz-Marco, A.; Martín-Matute, B. Organometallics 2019, 38, 3429–3435. Gallagher, A. T.; Malliakas, C. D.; Harris, T. D. Inorg. Chem. 2017, 56, 4654–4661. Sawano, T.; Ji, P.; McIsaac, A. R.; Lin, Z.; Abney, C. W.; Lin, W. Chem. Sci. 2015, 6, 7163–7168. Thacker, N. C.; Lin, Z.; Zhang, T.; Gilhula, J. C.; Abney, C. W.; Lin, W. J. Am. Chem. Soc. 2016, 138, 3501–3509. Elumalai, P.; Mamlouk, H.; Yiming, W.; Feng, L.; Yuan, S.; Zhou, H.-C.; Madrahimov, S. T. ACS Appl. Mater. Inter. 2018, 10, 41431–41438. Madrahimov, S. T.; Atesin, T. A.; Karagiaridi, O.; Sarjeant, A. A.; Farha, O. K.; Hupp, J. T.; Nguyen, S. T. Cryst. Growth Des. 2014, 14, 6320–6324. Kökçam-Demir, Ü.; Goldman, A.; Esrafili, L.; Gharib, M.; Morsali, A.; Weingart, O.; Janiak, C. Chem. Soc. Rev. 2020, 49, 2751–2798. Han, Y.; Li, J.-R.; Xie, Y.; Guo, G. Chem. Soc. Rev. 2014, 43, 5952–5981. Meilikhov, M.; Yusenko, K.; Fischer, R. A. J. Am. Chem. Soc. 2009, 131, 9644–9645. Islamoglu, T.; Goswami, S.; Li, Z.; Howarth, A. J.; Farha, O. K.; Hupp, J. T. Acc. Chem. Res. 2017, 50, 805–813. Madrahimov, S. T.; Gallagher, J. R.; Zhang, G.; Meinhart, Z.; Garibay, S. J.; Delferro, M.; Miller, J. T.; Farha, O. K.; Hupp, J. T.; Nguyen, S. T. ACS Catal. 2015, 5, 6713–6718. Rimoldi, M.; Nakamura, A.; Vermeulen, N. A.; Henkelis, J. J.; Blackburn, A. K.; Hupp, J. T.; Stoddart, J. F.; Farha, O. K. Chem. Sci. 2016, 7, 4980–4984. Chen, Y.; Li, P.; Zhou, J.; Buru, C. T.; Đorđevic, L.; Li, P.; Zhang, X.; Mustafa Cetin, M.; Fraser Stoddart, J.; Stupp, S. I.; Wasielewski, M. R.; Farha, O. K. J. Am. Chem. Soc. 2020, 142, 1768–1773. Thiam, Z.; Abou-Hamad, E.; Dereli, B.; Liu, L.; Emwas, A.-H.; Ahmad, R.; Jiang, H.; Adamu Isah, A.; Birame Ndiaye, P.; Taoufik, M.; Han, Y.; Cavallo, L.; Basset, J.-M.; Eddaoudi, M. J. Am. Chem. Soc. 2020, 142, 16690–16703. Klet, R. C.; Tussupbayev, S.; Borycz, J.; Gallagher, J. R.; Stalzer, M. M.; Miller, J. T.; Gagliardi, L.; Hupp, J. T.; Marks, T. J.; Cramer, C. J.; Delferro, M.; Farha, O. K. J. Am. Chem. Soc. 2015, 137, 15680–15683. Manna, K.; Ji, P.; Greene, F. X.; Lin, W. J. Am. Chem. Soc. 2016, 138, 7488–7491. Puneet Desai, S.; Ye, J.; Zheng, J.; Ferrandon, M. S.; Webber, T. E.; Platero-Prats, A. E.; Duan, J.; Garcia-Holley, P.; Camaioni, D. M.; Chapman, K. W.; Delferro, M.; Farha, O. K.; Fulton, J. L.; Gagliardi, L.; Lercher, J. A.; Lee Penn, R.; Stein, A.; Lu, C. C. J. Am. Chem. Soc. 2018, 140, 15309–15318. Prasad, R. R. R.; Dawson, D. M.; Cox, P. A.; Ashbrook, S. E.; Wright, P. A.; Clarke, M. L. Chem. Eur. J. 2018, 24, 15309–15318. Li, Z.; Peters, A. W.; Bernales, V.; Ortuño, M. A.; Schweitzer, N. M.; DeStefano, M. R.; Gallington, L. C.; Platero-Prats, A. E.; Chapman, K. W.; Cramer, C. J.; Gagliardi, L.; Hupp, J. T.; Farha, O. K. ACS Cent. Sci. 2016, 3, 31–38. Shabbir, H.; Pellizzeri, S.; Ferrandon, M. S.; Kim, I. S.; Vermeulen, N. A.; Farha, O. K.; Delferro, M.; Martinson, A. B. F.; Getman, R. B. Catal. Sci. Technol. 2020, 10, 3594–3602. Otake, K.; Ahn, S.; Knapp, J.; Hupp, J. T.; Notestein, J. M.; Farha, O. K. Inorg. Chem. 2021, 60, 2457–2463. Feng, X.; Ji, P.; Li, Z.; Drake, T.; Oliveres, P.; Chen, E. Y.; Song, Y.; Wang, C.; Lin, W. ACS Catal. 2019, 9, 3327–3337. Van Velthoven, N.; Waitschat, S.; Chavan, S. M.; Liu, P.; Smolders, S.; Vercammen, J.; Bueken, B.; Bals, S.; Lillerud, K.-P.; Stock, N.; De Vos, D. E. Chem. Sci. 2019, 10, 3616–3622. Van Velthoven, N.; Henrion, M.; Dallenes, J.; Krajnc, A.; Bugaev, A. L.; Liu, P.; Bals, S.; Soldatov, A. V.; Mali, G.; De Vos, D. E. ACS Catal. 2020, 10, 5077–5085. Manna, K.; Ji, P.; Lin, Z.; Greene, F. X.; Urban, A.; Thacker, N. C.; Lin, W. Nat. Commun. 2016, 7, 12610. An, B.; Li, Z.; Song, Y.; Zhang, J.; Zeng, L.; Wang, C.; Lin, W. Nat. Catal. 2019, 2, 709–717. Ji, P.; Manna, K.; Lin, Z.; Urban, A.; Greene, F. X.; Lan, G.; Lin, W. J. Am. Chem. Soc. 2016, 138, 12234–12242. Ji, P.; Song, Y.; Drake, T.; Veroneau, S. S.; Lin, Z.; Pan, X.; Lin, W. J. Am. Chem. Soc. 2017, 140, 433–440. Meilikhov, M.; Yusenko, K.; Fischer, R. A. Dalton Trans. 2010, 39, 10990–10999. Genna, D. T.; Wong-Foy, A. G.; Matzger, A. J.; Sanford, M. S. J. Am. Chem. Soc. 2013, 135, 10586–10589. Genna, D. T.; Pfund, L. Y.; Samblanet, D. C.; Wong-Foy, A. G.; Matzger, A. J.; Sanford, M. S. ACS Catal. 2016, 6, 3569–3574. Grigoropoulos, A.; Whitehead, G. F. S.; Perret, N.; Katsoulidis, A. P.; Chadwick, F. M.; Davies, R. P.; Haynes, A.; Brammer, L.; Weller, A. S.; Xiao, J.; Rosseinsky, M. J. Chem. Sci. 2016, 7, 2037–2050. Ren, J.; Ching Lan, P.; Chen, M.; Zhang, W.; Ma, S. Organometallics 2019, 38, 3460–3465. Fortea-Pérez, F. R.; Mon, M.; Ferrando-Soria, J.; Boronat, M.; Leyva-Pérez, A.; Corma, A.; Herrera, J. M.; Osadchii, D.; Gascon, J.; Armentano, D.; Pardo, E. Nat. Mater. 2017, 16, 760–766. Mon, M.; Rivero-Crespo, M. A.; Ferrando-Soria, J.; Vidal-Moya, A.; Boronat, M.; Leyva-Pérez, A.; Corma, A.; Hernández-Garrido, J. C.; López-Haro, M.; Calvino, J. J.; Ragazzon, G.; Credi, A.; Armentano, D.; Pardo, E. Angew. Chem. Int. Ed. 2018, 57, 6186–6191. Chołuj, A.; Zielinski, A.; Grela, K.; Chmielewski, M. J. ACS Catal. 2016, 6, 6343–6349. Chołuj, A.; Karczykowski, R.; Chmielewski, M. J. Organometallics 2019, 38, 3392–3396. Chołuj, A.; Krzesinski, P.; Ruszczynska, A.; Bulska, E.; Kajetanowicz, A.; Grela, K. Organometallics 2019, 38, 3397–3405. Grigoropoulos, A.; McKay, A. I.; Katsoulidis, A. P.; Davies, R. P.; Haynes, A.; Brammer, L.; Xiao, J.; Weller, A. S.; Rosseinsky, M. J. Angew. Chem. Int. Ed. 2018, 57, 4532–4537.

284 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162.

Organometallic Functionalized MOFs - Reactivity and Catalysis Morabito, J. V.; Chou, L.-Y.; Li, Z.; Manna, C. M.; Petroff, C. A.; Kyada, R. J.; Palomba, J. M.; Byers, J. A.; Tsung, C.-K. J. Am. Chem. Soc. 2014, 136, 12540–12543. Li, Z.; Rayder, T. M.; Luo, L.; Byers, J. A.; Tsung, C.-K. J. Am. Chem. Soc. 2018, 140, 8082–8085. Rayder, T. M.; Adillon, E. H.; Byers, J. A.; Tsung, C.-K. Chem 2020, 6, 1742–1754. Rayder, T. M.; Bensalah, A. T.; Li, B.; Byers, J. A.; Tsung, C.-K. J. Am. Chem. Soc. 2021, 143, 1630–1640. Bloch, E. D.; Queen, W. L.; Krishna, R.; Zadrozny, J. M.; Brown, C. M.; Long, J.; Science, R. Science 2012, 335, 1606–1610. Reed, D. A.; Keitz, B. K.; Oktawiec, J.; Mason, J. A.; Runcevski, T.; Xiao, D. J.; Darago, L. E.; Crocellà, V.; Bordiga, S.; Long, J. R. Nature 2017, 550, 96–100. Reed, D. A.; Xiao, D. J.; Gonzalez, M. I.; Darago, L. E.; Herm, Z. R.; Grandjean, F.; Long, J. R. J. Am. Chem. Soc. 2016, 138, 5594–5602. Ji, P.; Solomon, J. B.; Lin, Z.; Wilders, A. M.; Jordan, R. F.; Lin, W. J. Am. Chem. Soc. 2017, 139, 11325–11328. Denysenko, D.; Jelic, J.; Reuter, K.; Volkmer, D. Chem. Eur. J. 2015, 21, 8188–8199. Bien, C. E.; Liu, Q.; Wade, C. R. Chem. Mater. 2019, 32, 489–497. Röß-Ohlenroth, R.; Bredenkötter, B.; Volkmer, D. Organometallics 2019, 38, 3444–3452. Comito, R. J.; Fritzsching, K. J.; Sundell, B. J.; Schmidt-Rohr, K.; Dinca, M. J. Am. Chem. Soc. 2016, 138, 10232–10237. Comito, R. J.; Wu, Z.; Zhang, G.; Lawrence, J. A., III; Korzynski, M. D.; Kehl, J. A.; Miller, J. T.; Dinca, M. Angew. Chem. Int. Ed. 2018, 57, 8135–8139. Dubey, R. J.-C.; Comito, R. J.; Wu, Z.; Zhang, G.; Rieth, A. J.; Hendon, C. H.; Miller, J. T.; Dinca, M. J. Am. Chem. Soc. 2017, 139, 12664–12669. Metzger, E. D.; Brozek, C. K.; Comito, R. J.; Dinca, M. ACS Cent. Sci. 2016, 2, 148–153. Goetjen, T. A.; Liu, J.; Wu, Y.; Sui, J.; Zhang, X.; Hupp, J. T.; Farha, O. K. Chem. Commun. 2020, 56, 10409–10418. Arroyave, A.; Gembicky, M.; Rheingold, A. L.; Figueroa, J. S. Inorg. Chem. 2020, 59, 11868–11878. Agnew, D. W.; DiMucci, I. M.; Arroyave, A.; Gembicky, M.; Moore, C. E.; MacMillan, S. N.; Rheingold, A. L.; Lancaster, K. M.; Figueroa, J. S. J. Am. Chem. Soc. 2017, 139, 17257–17260. Agnew, D. W.; Gembicky, M.; Moore, C. E.; Rheingold, A. L.; Figueroa, J. S. J. Am. Chem. Soc. 2016, 138, 15138–15141. Fox, B. J.; Sun, Q. Y.; DiPasquale, A. G.; Fox, A. R.; Rheingold, A. L.; Figueroa, J. S. Inorg. Chem. 2008, 47, 9010–9020. Fox, B. J.; Millard, M. D.; DiPasquale, A. G.; Rheingold, A. L.; Figueroa, J. S. Angew. Chem. Int. Ed. 2009, 48, 3473–3477. Dong, Y.; Jv, J.-J.; Wu, X.-W.; Kan, J.-L.; Lin, T.; Dong, Y.-B. Chem. Commun. 2019, 55, 14414–14417.

14.09

Organometallic Mesogens

Manuel Bardají and Silverio Coco, IU CINQUIMA/Química Inorgánica, Facultad de Ciencias, Universidad de Valladolid, Valladolid, Spain © 2022 Elsevier Ltd. All rights reserved.

14.09.1 Preamble 14.09.2 General introduction 14.09.2.1 Terminology in liquid crystals 14.09.2.2 Liquid crystals characterization 14.09.2.3 Physical properties and applications 14.09.3 Organometallic liquid crystals of the main group elements 14.09.4 Organometallic liquid crystals of the group 6 elements 14.09.5 Organometallic liquid crystals of the group 7 elements 14.09.6 Organometallic liquid crystals of the group 8 elements 14.09.6.1 Complexes of iron without Cp 14.09.6.2 Ruthenium organometallic liquid crystals 14.09.6.3 Ferrocene-containing liquid crystals 14.09.6.3.1 Introduction 14.09.6.3.2 Monosubstituted ferrocenes 14.09.6.3.3 Disubstituted ferrocenes 14.09.6.3.4 Heteronuclear complexes with ferrocene as ligand 14.09.6.3.5 Ferrocene-containing fullerenes 14.09.6.3.6 Conclusions 14.09.7 Organometallic liquid crystals of the group 9 elements 14.09.7.1 Rhodium carbonyl complexes 14.09.7.2 Cyclometalated iridium complexes 14.09.8 Organometallic liquid crystals of the group 10 elements 14.09.8.1 Isocyanide complexes 14.09.8.2 Carbene complexes 14.09.8.3 s-Acetylide complexes 14.09.8.4 Allyl and olefin complexes of palladium 14.09.8.5 Ortho-metalated palladium(II) and platinum(II) complexes 14.09.8.5.1 Ortho-metalated azo and azoxy complexes 14.09.8.5.2 Ortho-metalated imine complexes 14.09.8.5.3 Ortho-metalated pyrimidine and pyridine complexes 14.09.8.5.4 Other ortho-metalated complexes, and related systems 14.09.9 Organometallic liquid crystals of the group 11 elements 14.09.9.1 Isocyanide complexes 14.09.9.1.1 Mixed isocyanide acetylide complexes 14.09.9.1.2 Mixed isocyanide halides complexes 14.09.9.1.3 Mixed isocyanide haloaryl complexes 14.09.9.1.4 Isocyanide dendrimers 14.09.9.1.5 Hydrogen-bonded isocyanide derivatives 14.09.9.1.6 Isocyanide for discotic mesogens 14.09.9.1.7 Isocyanide as a colorant 14.09.9.2 Carbene complexes 14.09.10 Concluding comments Acknowledgments References

285 286 286 288 288 289 292 293 293 293 294 295 295 295 299 301 302 303 304 304 304 307 307 310 311 311 312 312 313 317 322 325 325 325 325 326 326 328 330 331 332 334 334 335

14.09.1 Preamble This chapter can be considered as an update of the chapter entitled Metallomesogens In Comprehensive Organometallic Chemistry III,1 therefore it deals with metallomesogens described from 2007 to 2020. This chapter focuses primarily on small molecules or ions of an organometallic nature, which self-organize as liquid crystals. Therefore, as in the previous reference, the objective of this chapter is to provide a specific review to the area of organometallic mesogens. It does not cover liquid crystals in polymeric systems2,3 or combined with nanoparticles,4–6 which have been recently reviewed.

Comprehensive Organometallic Chemistry IV

https://doi.org/10.1016/B978-0-12-820206-7.00119-0

285

286

Organometallic Mesogens

The chapter begins by introducing a basic terminology, the types of mesophases and their characterization, as well as their physical properties and main applications. The subject matter in this review is organized according to groups from the periodic table and then by ligand or compound type. There has been a series of previous reviews about metallomesogens or metal-containing liquid crystals, although not specifically about organometallics, the latest and most extensive are collected here.7–10

14.09.2 General introduction 14.09.2.1 Terminology in liquid crystals We must start by remembering some concepts in this subject. The terminology relating to low molar mass liquid crystals follows the recommendations made by the IUPAC and the International Liquid Crystal Society in 2001.11 The liquid crystal state (LC) is a phase of matter which is intermediate between the crystalline solid and the isotropic liquid states, where a combination of properties is shown: mobility and fluidity as a liquid, some positional and/or orientational order, self-assembly and anisotropism as a solid crystal. Mesogen or mesomorphic compound is a compound which behaves as a liquid crystal and displays mesophases (liquid crystal phases). The liquid crystal state depends on the intermolecular forces, such as dipole–dipole, van der Waals interactions, p–p stacking; if they are too strong or too weak the mesophase is lost. The self-assembly associated with liquid crystals may be generated through the use of temperature (thermotropic LC), a solvent (lyotropic LC) or both (amphitropic LC). Therefore, the mesophases exist in a range of temperature or solvent concentration. In these materials, it can be distinguished: (a) the melting point, the temperature at which a thermotropic mesogen goes from the solid to a fluid mesophase; (b) the transition temperatures, to pass between different mesophases; (c) the clearing or isotropization point: the temperature at which the mesophase transforms into an isotropic liquid. Two types of thermotropic mesophases can be distinguished. On the one hand, enantiotropic mesophases, which are thermodynamically stable over a definite temperature, and are observed both upon heating and on cooling. On the other hand, monotropic mesophases, which are metastable in nature and only appear in the cooling process. Thermotropic mesogens can be classified into four principal types, by taking into account structural factors of the units (molecules) forming the mesophase (Fig. 1): calamitic (formed by rod-like molecules), discotic (formed by disk-like molecules), polycatenar mesogens (based on molecules with a rather extended calamitic central core containing several long substituents), and banana-shaped liquid crystals (formed by bent-core molecules). These types of structures determine the molecular self-assembly process and, consequently, the structure of the mesophase. The mesophases of calamitic mesogens are classified in two groups: nematic and smectic. The nematic mesophase (N) is characterized by an orientational order of the molecules that are aligned along a preferred direction (defined by a director n)

Fig. 1 Schematic representation of (A) calamitic (rod), (B) discotic (disk), (C) polycatenar, and (D) banana liquid crystals.

Organometallic Mesogens

287

Fig. 2 Schematic representation of N, SmA and SmC calamitic mesophases with the corresponding director vector.

(Fig. 2). The molecules can slide and move in the nematic mesophase (while roughly keeping their molecular orientation) and rotate around their main axis. This is the less ordered mesophase and usually it is very fluid. In the smectic mesophases the molecules are oriented, as in a nematic mesophase, with their principal axis roughly parallel to the director, but they are also defining layers. These layers can be perpendicular to the director, as in the smetic A mesophase (SmA), or tilted, as in the smectic C (SmC). The SmA and SmC mesophases are the less ordered and more common smectic mesophases. Other less common types of smectic mesophases are known, which differ in the degree or kind of molecular ordering within and between the layers. A more precise description of the smectic phase structure can be found in the following references.12, 13 The discotic mesophases are classified in two types: columnar, and nematic discotic. The structure of the nematic discotic mesophase (ND, Fig. 3, left) is similar to that of rod-like molecules, but constituted by disk-like units. In columnar mesophases, the molecules are staking in a columnar disposition and depending on the type of columnar arrangement several columnar mesophases are known. Some common lattices of the columnar phases are columnar nematic (NCol), columnar hexagonal (Colh), columnar rectangular (Colr), and columnar oblique (Colob) mesophases. Polycatenar mesogens display mesomorphism characteristic of rod-like or disk-like mesogens depending on the number, length and arrangement of the chains. Detailed descriptions of these materials may be found elsewhere.14,15 In the case of banana liquid crystals, the bent-shaped molecules can adopt a compact packing arrangement that restricts rotational freedom, thus allowing the molecules to organize into novel types of liquid crystalline phases. These new mesophases are denoted by the letter B, which refers to the characteristic banana or bent molecular shape. For example, B1, B2, B3, etc. have been used to designate different phases, chronologically as they have been discovered.16 When the mesomorphic compounds are chiral (or when chiral molecules are added as dopants) chiral mesophases can be produced, characterized by helical ordering of the constituent molecules in the mesophase. The simplest and most common is the chiral nematic phase (N ), which is also called cholesteric, taken from its first observation in a cholesteryl derivative. These chiral structures have reduced symmetry, which can lead to a variety of interesting physical properties such as thermochromism, ferroelectricity, etc. Finally, there are lyotropic mesophases, which are typically formed by surfactant molecules in water. These molecules consist of a polar hydrophilic head and one or several hydrophobic aliphatic chains, which aggregates to form micelles in water. Depending on temperature, concentration and solvent different mesophases are found by aggregation of such micelles. Three representative types of micelles are presented in Fig. 4.

Fig. 3 Schematic representation of N, Colh and Colr discotic mesophases.

Fig. 4 Schematic representation of micelle shapes: spherical, cylindrical and plate.

288

Organometallic Mesogens

Fig. 5 First organometallic mesogens.

A book containing the essentials of liquid crystal science from the perspectives of chemistry and physics has recently been published, updating earlier versions by one of the authors.17–19 From a historical point of view, liquid crystals started with Reinitzer’s pioneering work in the late 19th century. He found that cholesteryl benzoate displays two different melting points.20 It is generally accepted that the first report on metal-containing liquid crystals was made by Vorländer in 1923, who reported on a series of mesomorphic mercury complexes: actually, organometallic compounds (Fig. 5).21 Heintz in 1855 had reported carboxylate magnesium soaps.22 The term metallomesogens was proposed by Giroud-Godquin and Maitlis to refer to liquid crystalline materials containing metals.23 They combine the variety of coordination chemistry with the exceptional properties of liquid crystals to form metal complexes organized in fluid phases. The incorporation of metallic fragments into mesophases enhances useful characteristics such as polarizability and birefringence, modifies other physical properties, such as color, electrical conductivity, luminescence and magnetism, which broadens the preparation of new materials with new applications.

14.09.2.2 Liquid crystals characterization Mesophase characterization of liquid crystalline materials can be achieved by a combination of different techniques. The three basic procedures are briefly described. Liquid crystals are characterized by Polarized Optical Microscopy (POM), an optical microscope equipped with a heating stage and two polarizing filters. Mesogens possess birefringence due their anisotropic refractive index and each mesophase appears to have a characteristic texture resulting from the different domains, which can often be used to assign the type of mesophase. In contrast, an isotropic liquid is dark under crossed polarizers. This technique allows the determination of transition temperatures and temperature ranges. Liquid crystals are always characterized by Differential Scanning Calorimetry (DSC), which is complementary to POM. It is a technique that measures changes in heat capacity as a function of temperature, and it is used to detect phase transitions. Obviously, transition temperatures, energy involved and thermal stability (reproducibility of the heating–cooling cycles) can be determined more accurately with DSC. A review with data for 3000 organic liquid crystals was published.24 Finally, small-angle X-ray diffraction can determine the structure of a liquid crystal phase. It is important to note that X-ray diffraction experiments on the mesophase offer much less information than single crystal experiments. The number of observed reflections is low because of the disordered nature of the phases, and sample alignment is strongly recommended. The technique provides data such as the interplanar distance, the relative orientation of different sets of planes, the thickness of the layer, or the number of columns. In addition, the wide-angle X-ray region displays a broad peak corresponding to short range correlations between the molten hydrocarbon chains.

14.09.2.3 Physical properties and applications As stated above, liquid crystals are anisotropic fluids, which implies that their physical properties are also anisotropic and this is the basis for the widespread application of the materials. For instance, the anisotropic refractive indices lead to birefringence and it is the primary way to characterize a mesophase. Other physical properties such as linear polarizability, dielectric permittivity, and diamagnetism can show anisotropy. Furthermore, these properties in thermotropic liquid crystals will depend on the temperature. The physical properties will determine the applications. The first application reported for mesogens was as temperature sensors, based on the thermochromic behavior of the chiral nematic liquid crystals. This technology is still in use. The development of very stable room temperature liquid crystals such as cyanobiphenyls25 and the formulation of mixtures with temperature ranges sufficiently broad to be used in display applications (around −20 to 70  C) led to the main application of this material: Liquid Crystal Displays (LCD) for flat panel displays. First, twisted nematic phase mode LCDs were common, then, TFT–LCDs (thin film transistor) enabled a large number of segments (for example 640  1024) and improve image qualities such as addressability and contrast. LCDs exhibit advantages as low weight, low space requirement, low power consumption, emits almost no undesirable electromagnetic radiation and can be made in almost any size or shape (up to 82-in. diagonal). That is why they are used in notebooks, desktop computers, television monitors, mobile phones, handheld devices, video game systems, navigation systems, and car dashboards. The display size and performance have been dramatically improved, by the introduction of fast switching, better viewing angle dependency, good color quality, high brightness and contrast, to such an extent that LCD screens

Organometallic Mesogens

289

have ousted the traditional cathode ray tube (CRT) monitors. It is important to note that LCDs use the light-modulating properties of liquid crystals combined with polarizers, and need a backlight as source of illumination, lately light-emitting diodes (LEDs) have been used for this purpose. Despite the dominance of LCDs, other flat panel display technologies continue to be developed to address their shortcomings. The main competitor is OLED (organic light-emitting diode) displays: an OLED, organic LED, or organic electroluminescent diode, consists of a film of organic compound that emits light in response to an electrical current. This technology is already in the market and is competing strongly with LCD technology. A book has been written about liquid crystal devices, from fundamentals to the last advancements within LCD technology.26 Liquid crystals have become a major, multidisciplinary field of research, for LCD applications and beyond. For instance a theme issue beyond display applications,27 reviews on applications for discotic liquid crystals,28,29 ionic liquid crystals,30 light responsive liquid crystals,31 functional films for display applications,32 liquid crystals in industry,33,34 have been published. A book has appeared that deals specifically with semiconductor liquid crystals.35 Many common fluids are in fact liquid crystals, as for instance soap, which forms a variety of lyotropic liquid crystal phases depending on its concentration in water. There are also biological liquid crystals, which show typical lyotropic liquid crystalline phases (e.g., peptides, protein assemblies, cell membranes, and so on).36 Metallomesogens continue to be compounds of great interest in materials science due to their optical, mechanical, electrical and magnetic properties, obtained by combining the properties of metallic fragments with the fluidity and self-organization of liquid crystals. Therefore, in recent years many efforts have been carried out to add physical properties to mesogens in order to synthesize polyfunctional materials. These properties can be tuned with temperature by taking advantage of phase changes related to the liquid crystal behavior. To date, a considerable number of luminescent, especially phosphorescent, metallomesogens have been produced mainly based on lanthanides, platinum, iridium, palladium, and gold. The most effective strategy for designing them has been the modification of known chromophores in order to introduce the anisotropy necessary to be mesomorphic. Luminescence has been obtained not only in solution or in the solid state but in the mesophase and can be tuned by temperature and the aggregation state (e.g., aggregation-induced phosphorescence leads to on–off systems). Metallomesogen-based emitting materials have been proposed for applications in LEDs (Light Emitting Diodes), information storage, sensors and optoelectronic devices. Many studies have been conducted on spin crossover metallomesogens, which contain d-metal fragments with different magnetic and electronic properties due to their accessible high and low spin states. Some relationships between transition temperatures and these magnetic and electronic properties have been reported. Electrical and magnetic properties of metallomesogens containing d- or f-metal fragments as well as changes with temperature and phase state have also been studied in detail. The more relevant results for these physical properties on metallomesogens are collected in various reviews and many will be explained later in this chapter.9,37–43 There are many lanthanide liquid crystals and a specific name to refer to them has been proposed: lanthanidomesogens. They have been recently reviewed but unfortunately no organometallic compounds have been reported.44–46 In summary, liquid crystal research is an ever-expanding field that especially attracts synthetic chemists and physicists with new materials and new phases for new or improved applications.

14.09.3 Organometallic liquid crystals of the main group elements Organometallic liquid crystals of the Main Group Elements remain scarce. The examples known are practically limited to some boron compounds, a number of silicon derivatives and a series of alkylgermanes. Most of the mesomorphic boron compounds are based on clusters such as those shown as 1, which display mainly smectic A mesophases.47,48 Discotic borolane49 (2) and boroxine derivatives (3)50 showing columnar hexagonal mesophases have also been reported. It is worth noting that the mesophases of the borolane derivatives show good charge carrier mobilities (up to 8  10−2 cm2 V−1 s−1, Time-of-flight method).

290

Organometallic Mesogens

N-methyliminodiacetic acid (MIDA) boronates with different degree of substitution on the phenyl ring have also been studied (4). Their mesophase type is determined by the number and length of alkoxy chains.51 Compounds bearing one or two short tails (n ¼ 6, 8) display smectic A phases while the dialkoxy derivatives with longer substituents, and the trialkoxy series led to hexagonal columnar mesophases.

Recently, a few examples of N-heterocyclic carbene boranes (5) displaying smectic mesophases have also been prepared.52 Some of them show polymesomorphism, but a complete characterization of the mesophases has not been done.

The organometallic liquid crystals of silicon are mainly based on oligomeric silsesquioxane molecules, in which a number of conventional mesogens are linked to the inorganic silsesquioxane cube through flexible spacers (6). Their mesomorphic properties can be modulated trough the shape of the mesogens, the spacer length, as well as the number of substituted corner chains.53,54

Organometallic Mesogens

291

A variety of mesomorphic systems displaying different mesophases have been reported. For instance, 40 -o-alkyl-4cyanobiphenyls derivatives 7 displaying SmA mesophases,55 derivatives bearing perfluorinated tails (8) showing lamellar mesophases,56 or triphenylene compounds 9 and 10 that display columnar57 or lamellar58 phases respectively.

Concerning organometallic germanium liquid crystals, the examples reported are trialkylgermanes of the type 11, in which the mesomorphic behavior is tuned by the number, nature and distribution of both the aromatic rings, and the tails.59 Most of these complexes exhibit smectic and nematic mesophases. Ferroelectric behavior has also been found in complexes with chiral chains (SmC ).

292

Organometallic Mesogens

14.09.4 Organometallic liquid crystals of the group 6 elements Previously, several liquid crystals of formula M(CO)3(arene) were reported, but none during the period of time covered in this review.1 However, some mesogens have been described by using the fragments M(CO)4–5. For instance, the Mo(0) surfactants 12 (n ¼ 2, 6, 10) contain a hydrophobic metal carbonyl fragment and are easily prepared from surfactant phosphine ligands. The supramolecular arrangement of these surfactants leads to molybdenum vesicles in water.60

CO OC

Mo

OC

CO PPh2(CH2)nSO3Na CO

OC

PPh2(CH2)nSO3Na

Mo

OC

CO

CO

PPh2(CH2)nSO3Na

12 H-bonds are not quite strong, but are thermally stable at low to moderate temperatures and have been widely used to prepare liquid crystals, as can be seen in Section 14.09.9.1.5. This approach was carried out using 2,4,6-triarylamino-1,3,5-triazines equipped by 9 peripheral alkoxy chains, which behave as discotic liquid crystals at room temperature. These compounds are combined with para-isocyanobenzoic metal complexes, namely [M(CNC6H4CO2H)(CO)5] (M ¼ Cr, Mo, W) to give H-bonded aggregates 13, and room temperature hexagonal columnar mesophases.61 The clearing temperatures are in the range 77–84  C, higher than for the free triazine (57  C) and following the order: Cr > Mo > W. The supramolecular aggregates display high thermal stability, even in the isotropic state as confirmed by the infrared spectra (CN and CO vibrations). All the metal complexes (molecular or supramolecular) show a broad emission band at about 375 nm in solution at room temperature. Only the Mo and W complexes (molecular or supramolecular) display an intense yellow-green phosphorescence (maximum around 533 nm), assigned to a metal-to-ligand charge-transfer (MLCT), and with a lifetime of 53.9 ms. The corresponding gold(I) isocyanobenzoic acid complexes does not produce such adducts, likely due to the weak H-bonds with the triazine and the high insolubility of the gold acids. An extension of this work was carried out by synthesizing gold(I) (see Section 14.09.9.1.5) and chromium(0) [Cr(CO)5(zPPh2C6H4COOH)] (z ¼ 2 or 4) ortho and para phosphino metallo-acids.62 X-Ray diffraction shows dimeric structures with the expected double carboxylic H-bonds in [Cr(CO)5(2-PPh2C6H4COOH)]. Reactions with the triazine mesogen lead to hydrogen-bonded chromium(0) supramolecular adducts, although only the 4-diphenylphosphinobenzoic adduct 14 shows a columnar hexagonal mesophase at room temperature with a clearing temperature of 38  C (lower than 57  C found for the free triazine). Therefore, the triazine bearing nine lateral alkoxy chains leads to mesomorphic adducts even for bulky metal − organic fragments, although the lability of the H-bonds limits the thermal stability. CO OC N

OC10H21 C10H21O

C

OC CO

CO

N

C10H21O N H N

H N

N

N H

O

O N

C10H21O

OC10H21 OC10H21

N H N

OC10H21

H H N

N

N H

M = Cr, Mo, W C10H21O

OC10H21 OC10H21

13

C10H21O

Cr

PPh2

OC10H21 C10H21O

O H

CO

CO

M

OC10H21 OC10H21

O

OC10H21 OC10H21 OC10H21

14

CO

CO CO

Organometallic Mesogens

293

14.09.5 Organometallic liquid crystals of the group 7 elements For this group, only a series of Re(I) compounds 15 (n ¼ 12, 18) have been synthesized with dialkyl 2,20 -bipyridine-4,40 dicarboxylate ligands.63 The ligands and 15 (n ¼ 12) are not mesomorphic, whereas 15 (n ¼ 18) shows a short range existence unidentified mesophase (only 5  C; clearing at 117  C). The corresponding coordination compounds with the fragment CdCl2 are not mesomorphic, while with the PtCl2 (n ¼ 16) unit shows a smectic phase from 22 to 129  C.

Bipyridine and phenanthroline ligands have also been used, but carbonyl rhenium (I) compounds were always non-mesomorphic, although the ligand or other coordinating compounds behave like liquid crystals. This has been related to steric effects due to the large volume of this octahedral fragment.64–66

14.09.6 Organometallic liquid crystals of the group 8 elements 14.09.6.1 Complexes of iron without Cp Many efforts have been carried out in order to prepare iron liquid crystals and also to relate the spin transition with the transition temperatures. In some cases there is an apparent synergy between spin-crossover and liquid crystalline transition and this coupling allows dielectric and magnetic properties induced by light, temperature, and pressure. The designed compounds are not properly organometallics and have been reviewed previously.9,10,38,46 During the time covered in this review, only a family of Fe(II) organometallic compounds without cyclopentadienyl have been described: a series of imines containing bis(tricarbollide)Fe(II) 16 (R ¼ CnH2n+1; n ¼ 4, 6, 8, 10, 12, 18), where each vertex of the icosahedron is a BH.67 Each anionic carborane cluster (imine-C3B8H−10) is isolobal with the cyclopentadienyl anion and allows to prepare the analog ferrocene derivatives. These compounds exhibit nematic and smectic phases with clearing temperatures above 200  C.

Furthermore, some Fe(0) carbonyl mesogens have been reported. An alkoxy phenyl isocyanide containing a partially fluorinated hydrocarbon chain has been used to prepare the corresponding Fe(0) complex 17, as part of a study containing this ligand and several metallic fragments.68 The compound melts at 92  C and is not mesomorphic, although most of the close complexes (metal groups 10–11) display SmA phases.

As described in Section 14.09.4 for group 6 metals, the H-bond adduct 18 obtained by reaction of the para-isocyanobenzoic Fe(0) complex [Fe(CNC6H4CO2H)(CO)4] with 2,4,6-triarylamino-1,3,5-triazine lead to room temperature hexagonal columnar

294

Organometallic Mesogens

mesophases.61 This phase exists in the range 8–86  C, with an isotropization temperature higher than for the free triazine and slightly higher than for the analogous Cr, Mo and W complexes. In solution at room temperature the complex and the aggregate display an emission band around 375 nm.

14.09.6.2 Ruthenium organometallic liquid crystals Dinuclear ruthenium(I) dendritic complexes 19, namely Ru2(CO)4(O2CR1)2L2 (generation 1 and 2; L ¼ PPh3, pyridine, 4-picoline; generation 3, L ¼ PPh3) have been prepared for dendrons of generation 1, 2 and 3.69 They consist of a Ru–Ru bonded Ru2(CO)4 sawhorse unit bridged by two dendritic carboxylate ligands containing cyanobiphenyl moieties. All of them behave as quite stable enantiotropic liquid crystals, giving rise to smectic A (glass transition in the range 38–45  C to isotropic from 150 to 204  C) or smectic A (glass transition in the range 55–62  C to nematic from 143 to 170  C) and nematic (only for a short range of 3–12  C before clearing) mesophases.

R1 R1 O O

O O

Ru

Ru

L OC

CO OC

O C O L

R1 = O C T

(CH2)10 O Generation 1

O C O

CO

19

(CH2)10 O Generation 2

O C O

O C T T O C

C T O T=

O C O

CO2R RO2C

R=

T C O

O

(CH2)10 O

O C O

(CH2)10 O Generation 3

O C O

O C O

C O CN

O C T

O O C T

As commented above, the use of carbonyl carboxylato ruthenium is a typical way of producing liquid crystals. Another typical strategy is by means of arene ruthenium fragments. Reactions of the arene ruthenium metallacycle [Ru4(p-cymene)4(bpe)2(donq)2] [DDS]4 (bpe ¼ 1,2-bis(4-pyridyl)-ethylene, donq ¼ 5,8-dioxydo-1,4-naphtoquinonato, DDS ¼ dodecyl sulfate) with mesomorphic dendrimers containing pyrenyl-functionalized poly(arylester) and cyanobiphenyl, lead to host–guest supramolecular liquid crystals 20 (for generation 2 and 3).70 The pyrenyl fragment is located in the central cavity of the tetranuclear ruthenium complex. The mesophases display a multilayered structure due to the tendency of each component to separate in different organized zones by chemical nature (microsegregation). The pyrenyl dendrimer ligands display a multilayered smectic A-like phase, with a glass

295

Organometallic Mesogens

transition at 33 and 66  C, and an isotropization temperature of 165 and 219  C, respectively, for generations 2 and 3. On the other hand, the adducts exhibit a multicontinuous thermotropic cubic phase as deduced by SAXS, which decompose above 100  C due to the weak host–guest interaction. It is a spectacular example of liquid-crystalline behavior observed for such a large complex.

T 4+

N Ru OO

Ru N OO

C O

O T C

OO N Ru

Ru N

T C

C O

O T=

O Gn

(CH2)10 O

C R=

CO2R

O C O T

RO2C

O

20

O C

O C T

O

O

O O C

G3 =

G2 = OO

C O

T

(CH2)10

O C O

O

CN

An extension of the previous work reported 21 [Ru4(p-cymene)4(L)2(donq)2][DDS]4 (donq ¼ 5,8-dioxydo1,4-naphtoquinonato, DDS ¼ dodecyl sulfate), being L the bidentate N-donor ligand 1,4-di(4-pyridinyl)-benzene with poly(arylester) dendrimers.71 L and Ru4-L2 exhibit smectic phases (multilayered) above 50  C, with an isotropization temperature of 180  C for L, while Ru4-L2 decomposes at 160  C.

4+

R1 Ru N OO

R1

R1 = O

N Ru OO

O (CH2)10 O O T=

OO Ru N

R1

OO N Ru R1

21

O C O CO2R

O C T

C T O

RO2C R=

(CH2)10 O

O C O

CN

14.09.6.3 Ferrocene-containing liquid crystals 14.09.6.3.1

Introduction

Ferrocene compounds have attracted much attention due to their well-developed synthetic chemistry, good solubility and high chemical stability.72 Besides, its unique electrochemical properties have been exploited for the synthesis of sensors, biosensors and redox active supramolecular switches. Furthermore, its high thermal stability has allowed the synthesis of a great variety of liquid-crystalline materials.9,10,73,74 Therefore, systems with ferrocene are one of the most common in organometallic liquid crystals. This chapter will show the main results obtained for ferrocene-containing liquid crystals, including mono- and disubstituted ferrocenes, systems based on ferrocenophane, heteronuclear complexes with ferrocene-containing ligands and ferrocene decorated fullerenes. Note that ferrocene-containing liquid-crystalline polymers will not be presented here, as stated in the general introduction, although some specific references are given.75–79

14.09.6.3.2

Monosubstituted ferrocenes

Monosubstituted ferrocenes are the most common ferrocenes with liquid crystal behavior. Moreover, monosubstituted ferrocenophanes are included in this section.

296

Organometallic Mesogens

Aryl monosubstituted ferrocenes 22 (R ¼ C18H37, C(O)-C17H35) and 23 (R ¼ C18H37, C(O)-C17H35) containing a Schiff base and at least three phenyl rings are mesomorphic (with only two rings are non-mesomorphic).80 Compounds display an unidentified phase in the range 133–144  C (22, alkoxy), 126–154  C (22, ester), 114–145  C (23, alkoxy) and 129–155  C (23, ester). O O

N

O

Fe

N

Fe

OR

22

23

OR

O

Close aryl monosubstituted ferrocenes 24 and 25 (R ¼ C18H37, C(O)-C17H35) containing not only imine but azo functional groups have been prepared.81 All the compounds exhibit enantiotropic nematic phases with melting points from 130 to 166  C and clearing points in the range 154–247  C, being the largest mesophase range of 81  C (on heating) and 93  C (on cooling) for compound 25 (R ¼ C18H37). O O

Fe

N

N

24

OR

N O N

O

Fe

N

N OR

25

Aryl monosubstituted ferrocene mesogens 26 (n ¼ 6–9) have been prepared by using ester connections and different chain length for the flexible spacer, but now with a terminal biphenyl group.82 Derivative with n ¼ 6 displays an enantiotropic nematic phase in a short range of temperatures, from 148 to 155  C. The others show only monotropic nematic phases from 113–142  C to 40–76  C. As expected the substitution of phenyl by biphenyl lead to higher phase transitions points as well as wider mesophase ranges. O O

Fe

(CH2)n

O O O

26

OMe

O

Rotationally fixed [3]ferrocenophane leads to a higher variety of molecular geometries in comparison with ferrocenes. Monosubstituted ferrocenophanes 27 (X ¼ H, R ¼ OC12H25; X ¼ OH, R ¼ OC12H25; X ¼ H, OH, R ¼ OOC-C6H4-OC12H25), 28 (R ¼ OC12H25, OOC-C6H4-OC12H25), and the [3]ferrocenophane monosubstituted with a b-enaminoketone 29, have been prepared. Compounds 27 with three phenyl rings display enantiotropic nematic phases as the corresponding ferrocenes. However 27 with only two phenyl rings and 28 derivatives are not mesomorphic or show monotropic nematic phases. Moreover, compound 29 display nematic and SmC phases on cooling. Thus, the introduction of the propylidene bridge at the b-position to the substituent on the cyclopentadienyl ring enhances mesomorphism. In contrast, the presence of the propylidene group in the a-position drastically reduces liquid crystal behavior compared to the corresponding monosusbstituted ferrocenes. N

N R

Fe

27

OC12H25 O

Fe

29

28

X

NH

R

Fe

Organometallic Mesogens

297

Ferrocene-carboxylates 30 (n ¼ 2–12) have also been synthesized.83 Only derivatives with longer spacer (n ¼ 9–12) show monotropic nematic and smectic phases at room temperature, with temperature ranges of 25–37  C. The liquid crystal behavior is much more limited that for the corresponding 1, 10 -disubstituted compounds.84 O O

(CH2)n

O

Fe

O O

30

OMe

With the objective of achieving chiral phases, alkyl monosubstituted ferrocene 31 (n ¼ 6, 11) and 32 (n ¼ 6, X ¼ N, Y ¼ C, R1 ¼ R2 ¼ H; n ¼ 11, X ¼ C, Y ¼ N, R1 ¼ R2 ¼ H; n ¼ 6, X ¼ C, Y ¼ N, R1 ¼ OH, R2 ¼ H; n ¼ 6, X ¼ C, Y ¼ N, R1 ¼ NO2, R2 ¼ H) containing a cholesteryl have been synthesized.85 Compound 31 (n ¼ 6) exhibits a chiral nematic phase at 113  C (clearing at 158  C), while for n ¼ 11 display a SmA phase at 110  C, followed by short range twist grain boundary (TGB) and N phases (clearing at 155  C). All the compounds 32 show only a chiral nematic phase over a wide temperature range (from 90 to 240  C), except for n ¼ 11 where a SmA phase is observed before the N . (CH2)n

(CH2)5

O

Fe

N

O

O

31

O

(CH2)n

O

R2

R1

O

Fe

O

32

X

(CH2)5 Y

O

O O

A series of ferrocene compounds 33 (n ¼ 6, 11; R ¼ H, OH; R1 ¼ C12H25, C14H29) has been synthesized in which the oxadiazole unit is in a terminal position connected to the ferrocene moiety through three phenyl with ester, ether or imine connections.86 All of them exhibit a smectic phase starting at 101–149  C, which converts into a nematic phase in the range of 170–195  C (except the derivative with n ¼ 11, R ¼ OH, R1 ¼ C14H29, that isotropizes directly at 194  C), and finally the clearing point from 177 to 203  C. Similarly, 34 compounds (n ¼ 6, 11; R ¼ H, OH; m ¼ 5, 10) were prepared taking advantage of the same complex skeleton, where a cholesteryl group was placed at the end through an ester linkage.87 All of them exhibit a TGBC twist grain boundary phase having SmC slabs, with melting points from 137 to 186  C, which evolves to a N phase from 155 to 225  C. Finally, the clearing points go from 202 to 250  C. A blue phase I and blue phase II is observed in a short range of 3  C. It is remarkable the presence of a TGBC phase in a wide thermal range in the heating and cooling cycles for these chiral unsymmetrical metallomesogens. (CH2)n Fe

O

R

O

O N

N N

O

33

(CH2)n Fe

34

O

O

R N

O O

N N

S

(CH2)m O O

SR1

298

Organometallic Mesogens

A series of ferrocene compounds 35 (n ¼ 6, 11; m ¼ 14, 16, 18) equipped with a 2-phenylbenzoxazole and variable length alkyl spacer and terminal alkoxy chains have been synthesized.88 All of them exhibit enantiotropic smectic C phases with melting temperatures from 111 to 127  C, and isotropization temperatures around 200  C. Remarkably, the phases are stable over a wide temperature range due to the presence not only of long spacer and terminal chains, but also intramolecular hydrogen bonds. (CH2)n Fe

O

OCmH2m+1

O

OH

O

N N

O

35 Ferrocene liquid crystals 36 and 37 containing 1,4-substituted-1,2,3-triazole and 5-halogen-1,4-substituted-1,2,3-triazole have been synthesized in order to stabilize the nematic phase, diminish the transition temperatures and to get larger mesomorphic ranges.89 Compounds with X ¼ H exhibit the typical nematic phase at high temperatures (around 200  C) and short ranges. Halogen substituted derivatives display a wider mesomorphic temperature range (about 100  C on cooling) but worse thermal stability; the transition temperatures are still higher than 100  C except on cooling. The absorption spectra exhibit a peak at 480 nm for all of them, while the fluorescence emission spectra in solution display the maxima around 400 nm. O X

O

R

O

Fe

N O N

37

36

N

R = n-C12H25, X = H, Br, I

R2

R= R1

X = H, Br; R1 = H, R2 = OC10H21, OC12H25 R1 = R2 = OC6H13,

Monosubstituted ferrocene compounds 38 (n ¼ 8, 14, 16) have been prepared, which contain a heterocyclic unit as 1,2,3triazole directly attached to the ferrocene moiety.90 They show unidentified mesophases with melting points from 74 to 83  C, and clearing points in the range 143–145  C. Energies involved in isotropization transitions are quite high and the textures are unclear. Cl OCnH2n+1 O N Fe

NH

N

38 Most ferrocene mesogens show high transitions phase temperatures and typical ways of decreasing these temperatures is by lowering symmetry of the substituents and by adding cyclohexane fragments. A series of cyclohexane-containing aryl monosubstituted ferrocene 39 (Z ¼ nothing, –N]CH–; n ¼ 5, 10), 40 (without fluor) and [3]ferrocenophane 41 (Z ¼ nothing, –N]CH–; n ¼ 5, 10) have been synthesized.91 Compound 39 biphenyl (Z ¼ nothing) is not liquid crystal but the same 39 with imine display enantiotropic nematic phases in the ranges 119–135  C (n ¼ 5) and 110–117  C (n ¼ 10) for the first heating. Ferrocene-imine 40 and ferrocenophane 41 show monotropic nematic phases on cooling from 134 to 127  C (40), 86–83  C (41, biphenyl, n ¼ 5), 82–61  C (41, biphenyl, n ¼ 10), from 143  C (41, imine, n ¼ 5) and 117–61  C (41, imine, n ¼ 10). In general, lateral fluoro-substituents and [3]ferrocenophane play a positive effect over the nematic state. Complex thermal behavior is observed after several heating–cooling cycles. CnH2n+1 Fe

39

C10H21

N

Z F

Fe

F

40 CnH2n+1 Z Fe

41

F

F

Organometallic Mesogens

299

Combination of ferrocene and naphthalene moieties has also been carried out in order to get low temperature mesomorphism (below 100  C), which can be of interest for many applications in devices. The resulting compounds 42 (n ¼ 8, 10, 12, 14, 16) exhibit monotropic nematic phases in a short range (14–17  C) and isotropization temperatures in the range 69–88  C.92 O O Fe

OCnH2n+1

42

During the last years, various interesting possibilities for ferrocene compounds have been reported, we can highlight lyotropic phases and chiral dopants. The surfactant with ferrocenyl group 43 has been synthesized.93 Surfactants are functional molecules containing a hydrophilic head (as ammonium) and a hydrophobic tail (as alkylferrocene). An ordered lamellar lyotropic liquid crystal is observed when the concentration of ionic derivative 43 in water solution is over 40 wt%. Moreover, this lyotropic liquid-crystalline organization can be controlled by redox reaction (ferrocene/ferrocenium redox couple) and photopolymerization. Ferrocene maybe useful as redox–responsive chiral dopant, as reported for compound 44, capable of reversibly modulating color (from blue to green) in a cholesteric liquid crystal such as 40 -pentyloxy-4-cyanobiphenyl.94 O

(CH2)11 Fe

O +N

Br-

O

43

14.09.6.3.3

O O

O

Fe

44

Disubstituted ferrocenes

Disubstituted ferrocene liquid-crystalline compounds can be classified into two types as a function of the position of the substituents: heteroannularly disubstituted 1, 10 -, and homoannularly 1, 3-. The former type gives place to “S” or “U” shapes, while the latter type leads to a “T” shape, all of them enhance mesomorphism compared to the corresponding monosubstituted compounds. Other possibilities as for instance 1, 2-substitutions are detrimental for this property due to the hairpin shape. Rotationally fixed disubstituted [3]ferrocenophanes 45 (R ¼ OC12H25, OOC-C6H4-OC12H25) and 46 (R ¼ OC12H25, OOC-C6H4-OC12H25) are prepared, but the geometries are detrimental for mesomorphism: 45 derivatives with the two substituents in the same Cp are not mesomorphic, while 46 derivatives (1, 10 -substitution) show monotropic unstable N or Sm phases.95 The liquid crystal properties are worse than those found for the corresponding monosubstituted compounds (reported in the same article but commented in the previous section). R

N

N R

Fe

Fe N

45

R

46

N R

Some relevant results were reported by preparing unsymmetrically 1,10 -bis-substituted ferrocene 47, which contains an extended ferrocene core due to a standard phenylimine and a difluorobiphenyl group.96 Besides, a flexible p-decylciclohexane moiety is bonded to the biphenyl. Non-conventional tetrahedratic smectic C (SmCT) and tetrahedratic nematic (NT) mesophases have been observed, in the short range 203–216  C and from 216 to 275  C, respectively. A reasonable explanation for the macroscopic chiral domains and helical superstructures can be drawn from the tetrahedral liquid crystal order. An extension of this work led to a series of symmetrically and asymmetrically 1,10 -disubstituted ferrocene compounds 48–50, with related ligands and able to display again non-conventional lamellar (SmAT and SmCT), nematic (NT) and even columnar (Colmix) phases.97 The main features of these mesophases include spontaneously developed macroscopic homochiral domains, helical and myelinic supramolecular formations, and some optical biaxiality. Free rotation of the cyclopentadienyl rings in the ferrocene lead to bent conformations and the conformational enantiomers stabilize by assembly into tetrahedral molecular units. The columnar mesophase is formed by a mixture of bent and nonbent conformers. The compounds melt around 200  C (except for 48 with n ¼ 10, Colmix at 163  C), and isotropizations temperatures go from 206 to 287  C.

300

Organometallic Mesogens

C12H25O

C10H21

O Fe

O

F

F

N

47

CnH2n+1 F

F

Fe

F

F

H2n+1Cn

48

n = 5, 10

C10H21 Fe

C12H25O

F

N

F

49

C12H25O

C10H21

O Fe

O OH

F

F

N

50

Some efforts have been carried out to solve X-ray single crystal structures of mesogens in order to establish crystal structure/ mesophase relationships. For instance the single crystal structure of 1,10 -bis[3-[4-(4-methoxyphenoxycarbonyl)phenoxy]propyloxycarbonyl]ferrocene 51 show that the two substituents lie in the same direction (U shape), and not one on each side, which is related to the observed nematic phase.98 O O

(CH2)n Fe

O

O O

O (CH2)n

O

51

OMe

O O OMe

O

1,3-disubstituted ferrocene derivatives 52 (n from 2 to 10) were also prepared.99 They display monotropic nematic mesophases (except for n ¼ 2 and 4) in short ranges of temperature. The results are rationalized by noting that: when n is odd the molecule is more linear but more banana if n is even.

O MeO

O

O

(CH2)n

O

O (CH2)n

O

Fe

52

O

O O O

OMe

Organometallic Mesogens

301

Recently, the disubstituted ferrocene compound, 1-triisopropylsilylethynyl-10 -trimethylammoniummethylferrocene 53 has been synthesized.100 It is a surfactant that forms micelles and vesicles by aggregation at a very low concentration as well as lyotropic liquid crystals, not only with water, but also with a variety of organic solvents. Furthermore, the alkyne group can be deprotected and used as ligand to coordinate to metallic fragments.

+N

Fe

I-

Si

53 14.09.6.3.4

Heteronuclear complexes with ferrocene as ligand

Ferrocene or ferrocenophane compounds have been extensively used as ligands to prepare heteronuclear complexes of various metals, which can combine the optical, magnetic or electric properties of both metallic fragments. Note that these complexes are only described here, but not in the corresponding second metal section. In such a way monosubstituted ferrocenophane equipped with a base Schiff, yield ortho-palladated compounds 54–57.101 54 are non-mesomorphic, whereas 55 display SmA mesophases in the range 198–260  C (mixed with some decomposition at the isotropization) and 56–57 nematic mesophases phases over a broad temperature range from 71 to 205  C. R

R

Fe

R = OC12H25

N

N Fe

Pd

O

Pd

OC12H25

R=

Cl

OAc

O

2

2

55

54

R

R

N

N Fe

Fe

Pd N

Pd Fe

O

Fe

O

N

C12H25O

O C12H25O

O

57

56

An extension of this work has been carried out by functionalization of the ferrocenophane group with salycilidene and additional chelation to copper(II), nickel(II), palladium(II) and oxovanadium(IV), to give complexes 58. Similarly, copper(II) and palladium(II) compounds 59 have also been prepared after functionalization with aminovinylketone.102 The heteropolynuclear complexes 58 exhibit high temperature enantiotropic nematic phases for copper (199–230  C), palladium (257–277  C) and oxovanadium (217–242  C), while complexes 59 display monotropic nematic phases for copper (at 131  C) and palladium (at 188  C), in the last case followed by a SmC phase at 173  C. O OC12H25 O N M

Fe

M Fe

O

N

N

C12H25O

O C12H25O

O

Fe

O

OC12H25

N

O

Fe

O M = Cu, Ni, Pd, VO

58

M = Cu, Pd

59

302

Organometallic Mesogens

The analog monosubstituted ferrocene ligands containing an enaminoketone have been synthesized and then chelated to Cu(II) and Pd(II) ions in a 2:1 M ratio (60).103 Cu(II) complexes show disordered soft crystal phases (on cooling at 107 or 120  C for n ¼ 12 or 16, respectively) while Pd(II) complexes display monotropic smectic C phases (on cooling at 136 or 140  C for n ¼ 12 or 16, respectively). Monosubstituted ferrocene ligands equipped with a diketonato were prepared, as well as the corresponding chelated copper(II) and zinc(II) complexes 61 (n ¼ 8, 9, 10, 12, 14, 16).104 The copper compounds display melting points in the range 76–106  C and clearing points from 106 to 163  C. On the other hand, the zinc compounds show lower melting (from 23 to 49  C) and clearing points (from 97 to 139  C), as well as a wider mesophase range compared with the copper derivative. OCnH2n+1 Fe Br

O Fe

M

O

N

O

O

O

M O

O

Br

N

H2n+1CnO

Fe

H2n+1CnO

Fe

O

14.09.6.3.5

O

O

OCnH2n+1

60

61

Ferrocene-containing fullerenes

Liquid-crystalline materials equipped with fullerenes have attracted much interest for the development of supramolecular switches and in solar cells.105 Monosubstituted nonamethylferrocene compound 62 and the model compound 63 (without a large mesogenic fragment) are prepared by attachment to a fulleropyrrolidine.106 Only the dyad 62 exhibits an enantiotropic smectic A phase in the range 57–155  C.

T O O O

C C

N

C

O O

Fe

O

O C

(CH2)10

O O

T

62

R O

(CH2)5

O

O O

HN

O C

T= O C O

R=

O

(CH2)10

O

C O

R

Fe

O N

O O

HN

(CH2)5

63

CN

Organometallic Mesogens

303

Monosubstituted nonamethylferrocene compound 64 and the corresponding ferrocenium compound 65 (not drawn: as 64 but monocationic, anion is 4-methylbenzenesulfonate), which contain C60 in the middle and ferrocene and a mesogenic unit as terminal ends, are enantiotropic liquid crystals.107 Both exhibit sequentially SmA and SmB phases with melting points around 80–90  C, mesophase transition around 125  C and clearing point about 130  C. These compounds were studied by 13C NMR in isotropic, liquid crystal and crystalline phases to find out that the fullerene–ferrocene dyads rotate fast to form highly dynamic liquid-crystalline phases.

R O

Fe

O

O

O

C

C

O

(CH2)10

O C

O

O

O C

O

(CH2)5

O

O C

64

O R

O R=

(CH2)10

O

C O

OC8H17

Compound 66 (R ¼ n-C12H25, n-C18H37) contains a polar iron–ferrocene fragment, a conical shape, and long alkyl chains, which allow the formation of liquid crystalline phases in a large range of temperatures, from 55 to 230  C for C12, and from room temperature to 186  C for C18.108 The molecules exhibit microphase separation to form a crystalline and a thermotropic liquid crystalline phase. Moreover, the compounds are redox active (C60-ferrocene) and emissive (central cyclophenacene emits at 480–700 nm).

14.09.6.3.6

Conclusions

Ferrocene is still one of the most used systems to prepare organometallic mesogens. A great variety of compounds have been described because of the versatile synthetic chemistry that can be carried out and its high thermal stability. Shape–structure–property relationships have been stablished in order to understand the properties obtained. New carefully designed materials, can add the components properties to the final material.

304

Organometallic Mesogens

14.09.7 Organometallic liquid crystals of the group 9 elements 14.09.7.1 Rhodium carbonyl complexes A series of cis-[RhCl(CO)2(Ln)] complexes derived from polycatenar pyrazole ligands (Ln) have been described (67).109 However, only the complex containing six decyloxy substituents (R1, R2, R3 ¼ OC10H21; R4 ¼ H) is liquid crystal. It shows an enantiotropic columnar mesophase in the range 31–47  C. Although these molecules are not discotic when considered individually, stacked in an alternating antiparallel arrangement, they are able to generate columns with an almost circular cross section. The diameter of the column is less than twice the molecular radius, indicating that there is slight interdigitation between neighboring antiparallel molecules.

14.09.7.2 Cyclometalated iridium complexes Octahedral cyclometalated Ir(III) complexes are interesting systems of high importance as phosphorescent emitters. Their photophysical properties can be fine-modulated by selective functionalization on both the cyclometalated and the ancillary ligands, extending their potentiality beyond usual light emitting application. Perhaps, that versatility to tune their properties together with the fact that cyclometalated Ir(III) complexes constitute a fairly stable system, can explain their great development in recent years.110 The first reported example of liquid crystal based on cyclometalated Ir(III) complexes was the bipyridine complex 68, which displays a monotropic hexagonal columnar phase (Colh).111 The compound displays a phosphorescent emission in solution, in the solid state and in the mesophase, corresponding to a mixed 3LLCT/3MLCT transition (LLCT ¼ ligand-to-ligand charge transfer). The emission bands of the liquid-crystalline and crystalline phases were both blue shifted, with significantly higher quantum yield, compared to those observed in dichloromethane. The dodecyloxy derivative has also been prepared, displaying an enantiotropic columnar mesophase from room temperature until 115  C. This chromophore, inside micelles formed by the poly(ethylene oxide)100-poly(propyleneoxide)70-poly(ethylene oxide)100 copolymer in water, leads to luminescent lyotropic phases.112

Two similar complexes (69), but with a different linker between the bipyridyl moiety and the mesogenic pendent groups, display smectic A mesophases (69a: 216–235  C; 69b: 156–177  C).113 This system constituted the first example of directly polarized electroluminescence based on phosphorescent iridium complexes.

Organometallic Mesogens

305

Cyclometalated Ir(III) complexes bearing anisotropic polycatenar 2,5-diphenylpyridine ligands also lead to liquid-crystalline complexes (70–73).114

The complex 70 was obtained by cleavage of the corresponding di-m-chlorodiiridium-(III) dimer with dimethyl sulfoxide. This compound is not stable over time, but shows a centered rectangular mesophase (c2mm, a ¼ 60.92 A˚ , b ¼ 43.1 A˚ ), which could be indicative of either a columnar rectangular phase or a ribbon phase. In contrast, the bis-acetonitrile cationic complex 71 displays good stability and shows a columnar mesophase between 145  C and 163  C. Both complexes were slightly emissive in CH2Cl2 solution. Looking for luminescent iridium-based LCs, attention was then turned to neutral complexes bearing acetylacetonate (acac) as the ancillary ligand. 72 is liquid crystal, exhibiting an emissive columnar hexagonal phase between 31  C and 66  C (l ¼ 582 nm, F ¼ 9.1%). Similarly, the di-m-chlorodiiridium-(III) dimer 73 also exhibit a columnar liquid crystal phase from room temperature to 75  C, but in this case with a weaker emissive character (l ¼ 570 nm, F ¼0.8%). Using strongly mesogenic groups attached to an acetylacetonate ancillary ligand, mesomorphic iridium complexes have also been prepared (74). All of them display a smectic A mesophase, which in the case of the fluorinated derivatives, is monotropic in nature. These systems show blue- and green- phosphorescence in solution, as well as high hole mobility in solid films.115

306

Organometallic Mesogens

Related dinuclear Ir(III) complexes based on polycatenar diphenylpyridine ligands with 1,1,2,2-tetraacetylethane as the bridging ligand (75),116 exhibit orange luminescence in solution of CH2Cl2, and luminescent columnar hexagonal mesophases. As the two iridium centers are asymmetric, two diastereoisomers are possible (meso form with L-,D-stereochemistry and the racemate form with D, D- and L, L- stereochemistry). All the complexes were isolated as a mixture of these two stereoisomers, which, in the case of complex B, could be separated by column chromatography. However, it did not prove possible to identify their respective absolute configurations. Despite both isomers exhibiting identical emission properties and columnar hexagonal phases, their mesophase temperature ranges differed from 79–126  C for one isomer to 63–95  C for the other.

More recently, related neutral liquid-crystalline iridium(III) complexes with pyridyltetrazolate ligands have been reported (76). They display hexagonal columnar mesophases with an ambipolar carrier mobility behavior along the columns.117

Organometallic Mesogens

307

14.09.8 Organometallic liquid crystals of the group 10 elements 14.09.8.1 Isocyanide complexes Isocyanides (CNR) are versatile ligands which give stable complexes of many metals, including the group 10 elements. A variety of types of mesomorphic palladium and platinum isocyanide complexes are known. All of them are phenyl isocyanide derivatives bearing different substituents in the para position of the aromatic ring. A series of mesomorphic palladium and platinum complexes, trans-[MI2(CNC6H4O(CH2)4(CF2)8F)] (77), containing an isocyanide ligand with a semifluorinated alkoxy chain, have been prepared.68 Complexes of Ag, Au and Cu complexes with this isocyanide have also been studied, and their mesomorphic properties will be described in the corresponding section. The free ligand melts at 62  C and displays a SmA mesophase in a narrow range of temperature (2  C), in contrast to its hydrocarbon analog that is not liquid crystal. The palladium and platinum complexes display a SmA mesophase too, from 56 to 115  C for palladium and in the range 60–92  C for platinum. It is worth noting that the corresponding hydrocarbon analogues show no mesomorphic behavior.

Palladium and platinum complexes with the unusual isocyanide ligand 4-isocyanobenzoic acid, cis-[MCl2(CNC6H4COOH)2] and trans-[MI2(CNC6H4COOH)2] (M ¼ Pd, Pt) have been isolated.118 The carboxylic acid group of the coordinated isocyanide acts as a hydrogen donor for hydrogen bonding and hydrogen-bonded liquid crystalline metal complexes have been prepared with stilbazoles (78). When 4-decyloxystilbazole is used, only the trans hydrogen-bonded decyloxystilbazole complexes, which have a rod-like structure, display enantiotropic nematic mesophases.119 With tris(3,4,5-decyloxy)stilbazole, all the supramolecular palladium and platinum polycatenar aggregates, which have a trans- arrangement of the ligands, display a hexagonal columnar mesophase.120

308

Organometallic Mesogens

However, these aggregates have low thermal stability, most likely due to the thermal lability of the hydrogen bond. This led to the development of systems with higher stability based on isocyano–triphenylene complexes of palladium and platinum as those shown as 79, with a cis 79a or trans 79b configuration.121

Neither the free isocyanide ligand, nor the trans-isocyanide complexes 79b are mesomorphic, but the cis-isocyanide palladium and platinum complexes 79a display an enantiotropic columnar rectangular mesophase close to room temperature. Concerning the different thermal behavior observed in these systems, it is worth noting that both structures, cis and trans, can generate columnar stacking, but only the cis complexes possess a net dipolar moment associated with the metal fragment, whereas the trans isomers are non-polar. Thus, the deeply stabilizing intermolecular antiparallel dipole–dipole interactions seem decisive to support the stacked arrangement after chain melting, resulting in mesophases only for the cis complexes. The structure of the mesophase, determined by X-ray diffraction methods, is uncommon and is formed by the simultaneous p-stacking of the triphenylene discs into one-dimensional columns and the aggregation of the metallic fragments into tortuous threads, running parallel to the triphenylene columns, both segregated from the molten chains merged into an infinite continuum (Fig. 6).

Fig. 6 Proposed packing for the columnar mesophase of 79a.

Organometallic Mesogens

309

Fig. 7 Proposed schematic representation for the columnar mesophase of 80. Reproduced from Tritto, E.; Chico, R.; Ortega, J.; et al. J. Mater. Chem. C 2015, 3, 9385–9392 with permission from The Royal Society of Chemistry.

Interestingly, these inorganic/organic dual columnar mesophases show high hole mobility (above 1 cm2 V−1 s−1) along the columnar stacking. In addition, the dicyanoplatinum complex displays mesophase phosphorescence based on Pt


Pt interactions. Similar results were found for ionic Ortho-metalated benzoquinolate (bzq) complexes [Pt(bzq(CN-C6H4dOd(CH2)6-TriPh)2]A (A ¼ NO−3, BF−4, PF−6) with the triphenylene–isocyanide mentioned above (80).121,122 All of them display a similar semiconducting columnar mesophase with high one-dimensional hole mobility. Again, the structure of the mesophase is hybrid organic/inorganic in nature. It is constituted by a central column formed by the stacking of the organometallic benzoquinoline–platinum fragments, surrounded by six columns in hexagonal disposition formed by stacking the triphenylene groups (Fig. 7). These materials show aggregation-induced phosphorescence based on inter-disk Pt ⋯ Pt interactions.

The structure of the aryl isocyanide ligand has been further modified to introduce more paraffinic chains, and examples of metallodendrimers containing monodendrons with an isocyanide group in the focal point, and its organometallic complexes 81 trans-[MI2(CN-Gn)2] (M ¼ Pd, Pt, n ¼ 1, 2) have been reported.123

310

Organometallic Mesogens

Although the free isocyanide dendrons and firs generation complexes are no mesomorphic, the palladium and platinum compounds for generation 2 display a micellar cubic mesophase Im3m over a wide range of temperatures, which is similar to those described for related gold and copper complexes (see Section 14.09.9.1.4). Small-angle X-ray diffraction data on the mesophase support that each micelle contains 12 conical monodendrons, which in this case, correspond to six molecules of [MI2(CN-G2)2] (M ¼ Pd, Pt) (Fig. 8).

14.09.8.2 Carbene complexes Bis(carbene)nickel(II) complexes 82a have been prepared as potential pre-catalyst systems for olefin dimerization reactions in ionic liquid crystalline media.124 These complexes are not liquid crystals themselves, but their solutions in the liquid crystalline 1,3-didodecylimidazolium tetrafluoroborate, with concentrations of up to 10% weight Ni complex, retain the smectic mesophase of the imidazolium salt (55–71  C).

Fig. 8 Schematic representation of structure for compound 81.

Organometallic Mesogens

311

Analogous carbene platinum complexes 82b containing two or four pentadodecyloxytriphenylene units, have also been prepared.125 They display rectangular columnar mesophases in the temperature range of 50–76  C (R ¼ Me, R’ ¼ Tph) and from 30 to 57  C (R ¼ R’ ¼ Tph). In the mesophases, the triphenylene cores and the metal fragments segregate into different columnar units, leading to multicolumnar structures. In addition, the compounds display emission spectra related to the triphenylene core in solution, in the mesophase, in the isotropic liquid, and in the solid state. Similar results have been reported for related metal carbene of copper, silver and gold (Section 14.09.9.2).

14.09.8.3 s-Acetylide complexes The mesomorphic s-alkynyl platinum(II) terpyridine complex 83 exhibits strong solvatochromism. 83 shows a hexagonal columnar phase (20–200  C), in which the molecular self-assembly is reinforced by Pt ⋯ Pt and Pt ⋯ alkyne interactions.125 Substitution of the branched chains by linear dodecyl chains, suppresses the mesomorphic behavior. However, the dodecyl complex 83b is a good dodecane gelator, and forms gels that exhibit a strong near-infrared emission.

14.09.8.4 Allyl and olefin complexes of palladium Ionic allyl palladium(II) complexes [Pd(3-C3H5)(HLRpy)][PF6] (84a) and [Pd(3-C3H5)(LRpyH)][PF6] (84b) containing b-diketone ligands bearing pyridyl and pyridiniumyl substituents have been reported. Only the palladium complexes with longer alkoxy substituents (n > 16 for 84a and n > 14 for 84b) are liquid crystals displaying a enantioptropic smectic C mesophase.126 Luminescent studies for compound [Pd(3-C3H5)(LRpyH)][PF6] with n ¼ 14 show that it is fluorescent in solution, in the solid state, in the mesophase and even in the isotropic liquid.127

312

Organometallic Mesogens

The mesomorphic behavior of a series of triolefinic palladium(0) complexes 85 bearing a different number of alkoxy substituents, has been studied.128 Curiously, only the palladium complex that contains two dodecyloxy tails on each aromatic ring is liquid crystal, displaying a Colr mesophase (54–72  C).

14.09.8.5 Ortho-metalated palladium(II) and platinum(II) complexes Ortho-metalated complexes constitute an interesting and broadly studied type of metallomesogens.129 Particularly square–planar Pt(II) complexes of this type have received great attention in the last years because of their interesting photophysical properties.

14.09.8.5.1

Ortho-metalated azo and azoxy complexes

With the objective of preparing chiral metallomesogens featuring the stereogenic center as close to the metal ion as possible, mononuclear cyclopalladated complexes 86 based on azo or azoxybenzene ligands containing the chiral chelating ligand (D-(-)a-Phenylglycinol (HPhenylgly) have been synthesized.130 According to their 1H NMR spectra, only one isomer was formed in the case of azoxy complexes, whereas for the azo complexes isomeric mixtures were obtained as a result of a coordination N,N-trans or N,N-cis of the anionic phenylglycinol ligand in the mononuclear complexes. Neither the azo compound with R ¼ C6H13, [(Azo6)Pd(Phenylgly)], nor the azoxy derivative [(Azoxy)Pd(Phenylgly)] are liquid crystals, but the rest of ortho-palladated complexes display a nematic mesophase.

Complexes formed by a cyclopalladated azobenzene fragment bonded to an ancillary Schiff base ligand (87) bearing 11 or 12 peripheral alkyl chains, give rise to photoconducting hexagonal columnar mesophases at room temperature.131 It is worth noting that complex 87b shows a normalized photoconductivity of s/I ¼ 1.8  10−11 S cmW−1 at l¼ 760 nm.

Organometallic Mesogens

313

A new series of acetate and chloro-bridged dinuclear ortho-palladated complexes 88 derived from azobenzene with different polar groups (Me, Cl, F, NO2, CN) have also been studied. All chloro-bridged complexes predominantly exhibit SmA mesophases. However the acetato-bridged derivatives due to their typical open book shape do not show mesomorphism, except the cyano and fluoro substituted complexes, which exhibit a monotropic Smectic A mesophase.132 The melting points decrease in the order NO2 > CN > CH3 > Cl > F for the acetate complexes, and F > NO2 > Cl > CH3 > CN for the chloro-bridged derivatives.

14.09.8.5.2

Ortho-metalated imine complexes

Dinuclear and mononuclear ortho-palladated metallomesogens of crown ether derivatized imines 89, have been reported.133

The imine ligands and the acetato bridged complexes are non-mesomorphic while the chloro-bridged and the diketonato complexes display enantiotropic smectic A mesophases. The treatment of these complexes with sodium perchlorate gives rise to the corresponding sodium adducts [(O2ClO2)Na-89a] and [(O2ClO2)Na-89b] (n ¼ 4). However, none of the sodium adducts is liquid crystal. The acetato bridged complexes consist of a mixture of syn and anti-isomers and a dramatic increase in the syn:anti ratio is produced upon complexation of NaClO4. The crown ether derivatives extract sodium picrate from aqueous solutions and the presence of the palladium centers clearly improves the extraction. The same research group has also reported mesomorphic dinuclear ortho-palladated complexes with substituted dibenzo-18crown-6-ethers (90). In this case, only the anti-isomer is formed. The free imine ligands are not liquid crystals, but the palladium

314

Organometallic Mesogens

complexes, including acetato-bridged derivatives, show SmC mesophases. In contrast to the previously mentioned 15-crown-5 ether derivatives, complexation with potassium produces a significant increase in the mesophase range and stability.134

Other mesomorphic dinuclear and mononuclear ortho-metalated imine complexes containing semiperfluorinated alkyl chains have been synthesized, where the number of chains and the degree of fluorination has been systematically varied. In the chloro-bridged complexes 91, the presence of six semifluorinated chains and two alkyl chains in the system gives rise to hexagonal columnar mesophases with dramatically enhanced mesophase stabilities compared to the homologous hydrocarbon complexes, which show only monotropic discotic nematic mesophases.135 In contrast, compounds bearing four semifluorinated alkyl chains and two alkyl chains exhibit enantiotropic smectic A, as well as a monotropic smectic C mesophase.136 Similarly, mononuclear palladium complexes 92 bearing seven chains exhibit hexagonal columnar phases while the mesophases of the analogous compounds with five peripheral chains are smectic and nematic.137 This change is discontinuous, and the mesophase type is determined by the volume required by the non-polar periphery. Regarding the length of the fluorinated chain, an increase in the number of CF2 groups leads to a stabilization of the mesophases.

Mesomorphic Pd(II) and Pt(II) complexes 93 based on cyclometalated imine ligands and N-benzoylthiourea derivatives as auxiliary ligands, have also been prepared. These complexes display exhibit N or SmA mesophases. The platinum(II) complexes show photoluminescence both in solution and in solid state at room temperature, with the emission band centered around 600 nm.138 Several studies have been done varying the number of alkoxy groups attached to the imine ligand, alkyl chain length or the use of branched alkoxy terminal groups (94). The complexes display SmA and SmC phases, and the introduction of branched alkoxy terminal groups leads to lower transition temperatures and stabilization of SmC mesophases in both the Pd(II) and Pt(II) complexes.139

Organometallic Mesogens

315

Related ortho-palladated imine complexes 95 with di- and tri-alkoxysubstituted N-benzoyl thioureas are also liquid crystals.140 These complexes are thermally stable on a wide temperature range up to 230  C, showing SmA and SmC mesophases. In addition, they display a yellow-orange solid–state emission at room temperature with two emission maxima at lmax around 580 and 650 nm. In these complexes, when the number of alkoxy groups on the N-benzoyl thiourea ligand increases, the mesomorphism is lost and only melting in the isotropic liquid state is observed.141 However, increasing the number of terminal chains produces a transition from lamellar to columnar organization in 96, which contains the Schiff base a-(4-cyanobiphenyl-40 -yloxy)-o-(4-n-decyloxyanilinebenzylidene-40 -oxy) hexane.142 Four alkoxy substituents give rise to a SmA phase, whereas complexes bearing five or six tails show hexagonal columnar mesophases.

With the aim of obtaining photosensitive metallomesogens, palladium complexes derived from an ortho-metalated imine, bearing one or two azocarboxylate bridges (97), have been synthesized.143 The mixed thiolate–azocarboxylate bridges complexes display nematic and smectic A mesophases, whereas the biscarboxylate derivatives give rise to “soft” crystal phases. Electronic spectroscopy and 1H NMR show that all of them undergo a trans–cis isomerization of the azobenzene moiety at l ¼ 365 nm, which is faster and more stable when the trisubstituted azocarboxylate is present and the motion of the azo group is not hindered by the ortho-metalated imine. The photoresponse has also been observed in the condensed phases, which change from the ordered phase to the isotropic liquid upon irradiation.

316

Organometallic Mesogens

The substitution of the azo group by a decyloxy tail in the biscarboxylato bridges, together with the incorporation of a second alkoxy substituent in the metalated aromatic ring, 98, give rise to the appearance mainly of monotropic SmA mesophases.144 The same behavior has been found for related dinuclear acetate complexes.145

Mono and dinuclear triphenylene–imine ortho-palladated complexes 99 have also been prepared showing unusual columnar mesophases near room temperature.146

Organometallic Mesogens

317

Their structures contain columnar packing of triphenylene moieties, so they are organic columns that support an independent stack of metal fragments. Additionally, these palladium complexes exhibit fluorescence at room temperature in dichloromethane solution, associated with the triphenylene core.

14.09.8.5.3

Ortho-metalated pyrimidine and pyridine complexes

Ionic ortho-palladated phenylpyrimidine complexes 100 have been prepared, showing lamellar mesophases.147 The key factor to control the mesomorphism in this system (SmA for R ¼ COC22H45, and SmC for R ¼ CH2OH) is the type of substituent on the bipyridine fragment, whereas no significant differences are observed by changing the anion from BF−4 to ClO−4.

Luminescence and photoconductivity of phenylpyrimidine series of ortho-platinated and -palladated metallomesogens previously reported (101),148 have been studied. Only the platinum compounds show luminescence in the solid state and in solution, as well as photoconductivity caused by Pt ⋯ Pt closed shell interactions. The substitution pattern conditions and the metallophilic interactions have a significant influence on both properties. After photoexcitation, a competition between radiative deactivation (luminescence) and photo-conductivity was observed. Hence, strong luminescent compounds are non-conductive, whereas comparatively less or non-luminescent materials behave as photoconductors.149

A series of liquid crystals have also been synthesized based upon mononuclear ortho-platinated 2-phenylpyridines and 2-thienylpyridines 102, and the effect of the number of peripheral chains was studied systematically in several papers. Increasing the number of tails attached to the 1,3-diketonate units produces, as described previously in ortho-metalated imine complexes, a transition from lamellar (SmA) to hexagonal columnar phases (Colh).150 They all display a luminescence emission in toluene solutions originating from a mixed ligand-centered-MLCT excited state.

318

Organometallic Mesogens

Another series of cycloplatinated complexes based on 2-(4-alkoxyphenyl)-5-(alkoxymethyl)pyridines, but bearing the less bulky acetylacetonate ancillary ligand, has been synthesized (103).151

Only the complexes with longer substituents (n ¼ 12 and 16) exhibit enantiotropic lamellar mesophases with interdigitation of the molten aliphatic chains. In addition, they show intense polarized luminescence (lmax ¼ 532 nm and a polarization ratio as high as 10.5). More recently, platinum complexes 104 containing b-diketone ligands with biphenyl substituents have been studied, searching for phosphorescent metallomesogens with efficient dichroic ratios and high emission efficiency.152

Complex 104a displays an enantiotropic nematic phase between 118 and 183  C, and exhibits intense sky-blue emission in solution and in the neat film. Aligned films of a 104a:polyimide mixture show polarization-dependent photoluminescence with polarized ratio of 5.4. This has allowed the fabrication of polarized organic light-emitting diodes displaying a broad emission spectrum in the range of 450–900 nm with polarized ratio of 1.33 and an external quantum efficiency of 1.1%. The introduction of the tetraphenylethene bulk moiety into the system (104b), leads to a monotropic SmA mesophase, but the emission efficiency is drastically decreased.

Organometallic Mesogens

319

Using modified 2-phenylpyridine derivatives, in which cyanobiphenyl unit is linked on 4th position of pyridine group, new ortho-platinated metallomesogens 105 have been obtained.153

Picolinate derivatives with one cyanobiphenyl unit show a SmC mesophase, whereas acetylacetonate complexes and picolinate derivatives with two cyanobiphenyl units exhibit a nematic phase. Moreover, these metallomesogens show strongly polarized photoluminescence in liquid crystalline phases, with a polarized dichroic ratio of up to 24.6 in nematic phase. Interestingly, the introduction of a chiral 2-butanol chain into the picolinate group leads to an chiroptic smectic C phase, in which the molecules form heliconical arrangements and the chirality is transferred from the molecule to the mesophase.154 The functionalization of the 2-phenylpyridine unit with chiral 2-octanol chain (R and S) has also been studied (106).155 These complexes show chiral smectic (SmA ) and nematic (N ) phases and an intense phosphorescent emission at 504 nm both in solution and in the solid state. Moreover, the neat films annealed at 100  C (the complexes are in the N phase), exhibit an intense circularly polarized luminescence (CPL) spectrum with a large glum value up to 0.02.

Related cycloplatinated complexes with diphenylpyridine groups, [Pt(C^N)Cl(acac)] (107, 108), also display mesomorphism.156 Although the parent ligands exhibit a rich smectic polymorphism, the b-diketonate complexes 107 show a SmA mesophase at high temperatures. However, the complexes 108 with the fused cyclopentene ring display monotropic nematic and/or SmA mesophases. It is worth noting that related series of complexes [Pt(C^N)Cl(dmso)], were also mesomorphic with a similar behavior to the acetylacetonate derivatives.

320

Organometallic Mesogens

All of the b-diketonate complexes are very brightly phosphorescent in dichloromethane solution with emission quantum efficiencies around 0.5. Related derivatives containing hexafluoroacetylacetonate, trifluoroacetylacetonate and 3,5-heptanedionato ligands, have also been synthesized.157 However, neither mesomorphism nor luminescent properties are improved. In fact, for the hexafluoroacetylacetonate complexes, these properties are even lost. By combination of the mesogenic cycloplatinated monomer (monotropic SmA mesophase) based on a 2,5-di(4-alkenyloxyphenyl)pyridine ligand with 1,1,3,3,5,5- tetramethyltrisiloxane, the phosphorescent liquid-crystalline polymer 109 has been obtained. This polymer displays a SmA mesophase from 64 to 150  C, and shows polarized emission.158

Dinuclear platinum complexes containing a 6,12-dihydro-indeno[1,2-b]fluorene bridging unit have also been described (110).159

These complexes display a smectic mesophase in the temperature range 188–220  C for R1 ¼ CH3, and 75–193  C for the dodecyloxyphenyl derivative. In addition, both complexes display fluorescence (532, 574 nm, 1p,p ) and phosphorescence (640 nm, 3p,p ) dual emissions in solution. With the aim of extending these studies to molecules prone to give columnar mesophases, the same research group has developed a new series of platinum complexes bearing triphenylene and carbazole units (111).160

Organometallic Mesogens

321

These carbazol derivatives display monomolecular emission both in solution and in neat film, as well as hole mobilities (annealed films) in the range of 10−5–10−6 cm2 V−1 s−1. However, only the complex with the triphenylene group directly bonded to the pyridine ring and R ¼ –C6H4–O–C12H25 is liquid crystal, showing a columnar mesophase from 66 to 143  C. Cationic ortho-platinated 2-phenylpyridyne complexes containing pyridyl triazole ligands 112 have been found to exhibit columnar hexagonal mesophases over a wide range of temperatures, including room temperature. In addition, they display mechanochromism in the crystal phase.161 These compounds exhibit environmentally sensitive Pt ⋯ Pt intermolecular interactions, which help the formation of columnar liquid crystalline phases, as well as the mechanochromic behavior. Tetrazole ligands have also been used as auxiliary ligand to prepare cycloplatinated 2-phenylpyridine complexes with different degrees of substitution.162 Only the complexes 113 bearing three flexible alkoxy chains are liquid crystals, displaying a complex lamellar mesophase (from 17–32  C to 205–230  C) supported by Pt ⋯ Pt and p – p interactions. Aligned films of these complexes show highly polarized phosphorescence with a maximum peak at 653 nm, and a polarization ratio in the range 4.1–7.1. Moreover, they display ambipolar carrier mobility.

322

Organometallic Mesogens

Phosphorescent cationic ortho-platinated 2-phenylpyridine complexes 114, without long peripheral chains, constitute nonconventional examples of columnar mesophases.163 They show hexagonal columnar mesophase over a wide range of temperatures (96–284  C for R ¼ H and 101–266  C for R ¼ OMe). The phenyl and methoxyphenyl groups appear to act as the characteristic “soft” regions of traditional discotic liquid crystal structures. It is worth noting that with longer alkoxy substituent (OCnH2n+1, n ¼ 2, 4, 6), the mesomorphic behavior is completely lost. This reflects a subtle relationship of the molecular interactions that control the formation of liquid crystals. Another analogous example of mesomorphic systems without long peripheral chains, although not organometallic in nature, is constituted by cyclopalladated 3,5-disubstituted-2-(20 -pyridyl)pyrroles, 115.164 In this case, the mesophase formation is driven by the self-segregation of the trifluoromethyl groups from the rigid organometallic cores, in such a way that the mesophase structure consists of a columnar stacking of the metallic fragments surrounded by the fluoridated region. When these compounds are heated, the fluoridated zones melt at first to form a soft part, which preserves the self-assembly of the central cores so that a columnar mesomorphism is induced.

14.09.8.5.4

Other ortho-metalated complexes, and related systems

Using a simple strategy of combining in the same molecule a luminophore group with moieties able to induce liquid crystal behavior, a red emitting columnar mesophase (35–173  C) containing the cyclopalladated Nile red chromophore has been synthesized (116).165

Related complexes with a polyalkylated Schiff base as an ancillary ligand are also mesomorphic, showing photoconducting columnar rectangular mesophases from UV–Vis to near IR wavelength. The columnar organization is induced by the formation of dimeric pairs through hydrogen-bonded interactions between Nile red fragments (117).166

Organometallic Mesogens

323

The same group has also reported the cationic Nile red cyclopalladated metallomesogen 118, which displays a Colr mesophase. However, in contrast to the related neutral derivatives, the emission is quenched due to the presence of the bipyridine ancillary ligand.167

C,N,N0 -cycloplatinated complexes of 1,3-bis(2-pyridyl)benzene ligands containing a 3,4,5-trialkoxyphenyl group constitute an interesting family of luminescent metallomesogens (119).168 Complexes 119a show rectangular columnar mesophases for n > 6 (from 92–149  C to 151–187  C), whereas compounds 119b display hexagonal columnar phases (n ¼ 10, 12) over a wider range of temperatures (33–220  C). These Pt(II) complexes show an interesting thermochromic behavior, which depend on the thermal history of the sample. Indeed, when the isotropic liquid is slowly cooled into the mesophase, an orange emission from a monomer is observed. In contrast, a fast cooling leads to a glassy state displaying a red excimer-like emission.

324

Organometallic Mesogens

The introduction of fluorine atoms into the central phenyl group changes both the mesophase and photophysical properties of the system.169,170 Compounds 120 display a mesophase with a monoclinic space group, as well as a P6/mm lattice at higher temperatures. For compound 121 a rhombohedral R3m phase has been indexed. These fluorinated compounds display intense fluorescence in solution and in the solid state, but none exhibit temperature-dependent emission.

Recently, a series of tetradentate platinum(II) complexes 122 displaying red phosphorescence in dichloromethane solution (lmax  640 nm, f ¼ 4.3 % – 6.7 % ) have been synthesized.171 However, only the complex with n ¼ 12 is liquid crystal, showing a columnar mesophase from 22.0 to 78  C. The mesophase exhibits ambipolar carrier transport (measured by the space charge limited current SCLC method) with hole and electron mobility values of 4.72  10−4 cm2V−1 s−1 and 3.08  10−4 cm2V−1 s−1 respectively. It is remarkable that the carrier mobility in the mesophase is two orders of magnitudes greater than that in the amorphous state.

Organometallic liquid crystals based on platinum(IV) systems are rare, and for a long time, the only reported examples had been a few cyclometalated azobenzene complexes, showing mainly nematic phases.172,173 Recently, new mesomorphic ortho-metalated 2-phenylpyridine platinum(IV) complexes 123, have been obtained.174 They display a lamellar mesophase (78–265  C), in which a short-range Pt ⋯ Pt correlation is observed. In addition, these complexes are luminescent in deoxygenated dichloromethane solution (lmax ¼ 532 nm, t ¼ 230 ms, f ¼ 10%).

Organometallic Mesogens

325

14.09.9 Organometallic liquid crystals of the group 11 elements The majority of the organometallic liquid crystals of the group 11 elements contain an isonitrile (isocyanide) ligand and gold as metal. Gold(I) tends to give linear 2-coordinated compounds, which favors calamitic liquid crystals; even non-mesomorphic ligands often become mesomorphic after coordinating with a gold center. Some other ligands have been used as carbenes. Some silver and copper mesogens are also known, scarcer than for gold. There have been a number of previous specific reviews on gold175–178 and silver mesogens.179

14.09.9.1 Isocyanide complexes Isocyanide or isonitrile ligands are strongly attached to metal centers, particularly to gold(I) which leads to high thermal stability and liquid crystal behavior. Some isocyanide ligands have been designed in order to add specific functional groups.

14.09.9.1.1

Mixed isocyanide acetylide complexes

Alkyl isocyanide combined with alkoxyphenyl acetylide yielded very simple rod-like gold(I) compounds 124a, with gold–gold distances around 3.6 A in the crystalline state and liquid crystal behavior (R ¼ CnH2n+1, n ¼ 5–8).180 They exhibit nematic or smectic A monotropic mesophases at less than 100  C, although with short mesophase ranges (up to 25  C). Most interesting are the photophysical properties: a blue phosphorescent emission with quantum yields in the range 8%–50% in the solid state at room temperature, and a lifetime about 50 ms. The same emission was observed with a low the quantum yield for the mesophase and the isotropic liquid, but no emission was found in dilute solutions. Changing phenyl acetylene by biphenyl or naphthyl acetylene led to compounds 124b (R ¼ CnH2n+1, n ¼ 5–8) with a wider mesophase range (on heating, from 4 up to 19  C for the analog hexyloxy derivatives) and to green emission. Now, there are not gold–gold interactions and the two-ring cores slightly expanded the LC temperature range and moved the emission color.181,182

14.09.9.1.2

Mixed isocyanide halides complexes

Chloro combined with alkoxyphenyl isocyanide led to very simple rod-like gold(I) compounds 125a (n ¼ 1, R ¼ CnH2n+1, n ¼ 5–7), which arrange into dimers or chains (always in an anti-configuration) in the solid state through short gold–gold distances (around 3.3 A).183 They exhibit smectic enantiotropic mesophases in the range 121 to 172  C with moderate mesophase ranges between 24 and 48  C. Replacement of alkoxyphenyl by an alkoxybiphenyl isocyanide 125b (n ¼ 2, R ¼ CnH2n+1, n ¼ 3–8) kept the solid state arrangement with short gold–gold distances (in acetylide analogues was lost), rises the transition temperatures (melting points from 130 to 218  C) and mesomorphic ranges, although they decompose without clearing around 300  C.184 Most interesting are the photophysical properties: as observed in the analog acetylide isocyanide compounds, they exhibit an intense blue (phenyl) or green (biphenyl) phosphorescent emission in the solid or liquid crystal states. Quantum yields are in the range 6%–66% depending on the core (better biphenyl) and the length chain. In dilute solutions they are non-emissive or weakly emissive. Compounds 125a display a reversible “on–off” switching of the luminescence induced by the phase transition between mesophase and isotropic liquid. Moreover, hexyloxy complex 125a showed thermochromic photoluminescence by the phase transition between crystalline and SmC phases.

The effect of bonding siloxane groups at the end of an alkoxy chain in a isocyanide was studied.185 Compounds 125c (n ¼ 4, 5) showed small changes in the solid state structure, the liquid-crystalline or the emission properties compared to the parent compounds. Again, a reversible “on–off” switching of the luminescence induced by the phase transition was observed. Therefore, these gold mesogens show potential application as materials for light-emitting devices. The emission intensities of the compounds are enhanced in the solid and liquid crystal phases, meaning that they exhibit aggregation–induced emission and can be useful as phosphorescent AIEgens. As part of a report commented before for iron in Section 14.09.6.1, palladium and platinum in Section 14.09.8.1, a hydrofluorocarbon alkoxy isocyanide mesomorphic ligand was used to prepare Ag(I) 126a (X ¼ BF4, NO3), Au(I) and Cu(I) 126b (M ¼ Au, Cu) and dimer Cu(I) 126c complexes.68 126a and 126b compounds showed SmA calamitic mesophases stable over a remarkable wide temperature range (148, 83 and 81  C, respectively for Ag, Au and Cu) and high transition temperatures (up to 274  C for Au) compared to the similar hydrocarbon compounds. On the contrary, Cu(I) 126c complexes were non mesomorphic.

326

Organometallic Mesogens

14.09.9.1.3

Mixed isocyanide haloaryl complexes

The combination of two equal mesogenic alkoxytetrafluoroaryl ligands and several substituted biphenyl diisocyanides led to nematic dinuclear gold(I) complexes 127 (R ¼ CnH2n+1, n ¼ 4, 6, 8, 10, R’ ¼ H, Cl, Me).186 Transition temperatures decrease in the order 4,40 -biphenylene > 2.20 -dichloro-4,40 -biphenylene > 2.2’dimethyl-4,40 -biphenylene, which was related to the twist angle (Me > Cl > > H) of the biphenyl and therefore, the planarity of the molecule: more planar means stronger interactions and therefore higher transition temperatures. Complexes show emission maxima in the range 480–532 nm at room temperature in the solid state and also in solution, from 452 to 524 nm.

Mononuclear gold(I) compounds 128 (Y ¼ H, Z ¼ OC4H9, OC8H17, OC10H21, OC12H25; Y ¼ Z ¼ OC4H9, OC8H17, OC12H25) containing a perfluoroaryl mesogenic unit and a crown ether isocyanide have been prepared.187 Complexes with one alkoxy chain display a smectic C mesophase at temperatures close to room temperature. The trialkoxy-chain complexes show smectic C or unidentified mesophases, more clearly observed on cooling. The increase of the alkoxy chain length diminishes the melting temperature in these complexes. The bulky and flexible crown ether fragment leads to less intermolecular attractions, lower melting points, less ordered mesophases, and shorter mesophase ranges. These mesogens are emissive not only in the solid, and in solution, but also in the mesophase, and even in the isotropic liquid at mild temperatures. The complexes in the solid state display emission maxima from 457 to 511 nm at room temperature and in the range 450–492 nm at 77 K. They also emit in solution at room temperature in the range 426–445 nm, and at 77 K from 425 to 497 nm.

The 15-crown ether moiety is able to coordinate sodium cation from NaClO4, solubilizing it in chlorinated solvents, leading to the corresponding bimetallic complexes 129 (Y ¼ H, Z ¼ OC8H17; Y ¼ Z ¼ OC8H17), although the liquid crystal behavior is lost.

14.09.9.1.4

Isocyanide dendrimers

The synthesis and design of liquid crystal metallodendrimers has attracted much interest during the last years.188 We will describe here the results obtained for organometallic complexes of group 11 metals, and in particular for isocyanide. Two generations of isocyanide dendrimers have been reported, and, also, the corresponding organometallic complexes 130a [MCl(CN-Gn)] (M: Au, Cu) and 130b [{CuCl(CN-Gn)2}2].123 The free ligands and the first-generation complexes are not mesomorphic, but the second-generation complexes exhibit a thermotropic micellar cubic mesophase, over a large temperature range. In particular the coordination to AudCl fragment leads to a mesophase in a wide temperature range of 94–260  C, 110–290  C for CudCl and from room temperature to 268  C for the Cu dimer. The palladium and platinum complexes were described in Section 14.09.8.1, along with a schematic representation of their structure.

Organometallic Mesogens

327

More recently, non-mesomorphic oily dendrimers containing isocyanide as terminal functional groups were designed and prepared.189 Dendritic polyisocyanides are polytopic ligands able to generate a great diversity of metallodendrimers because of their ability to bind to various transition metals. Moreover, they can be used to build liquid-crystalline organometallic dendrimers. Generation zero, one, and two, with 3, 6, or 12 nitro terminal groups were synthesized, and then transformed into the corresponding polyisocyanides. Three types of gold(I) fragments were connected at the periphery of the dendrimers to give metallodendrimers 131.

328

Organometallic Mesogens

Their liquid crystal properties are largely influenced by the nature of the peripheral gold fragments, while increasing the dendritic generation contributes to a strong stabilization of the mesophases. Smectic C mesophases are observed for 131a and 131b, with mesophase temperature ranges 68–173 (G0) and 105–172  C (G1) for AudCl, but only 112–128 (G0) and 110–121  C (G1) for Au–acetylide. A room temperature columnar hexagonal mesophase is displayed by the three generations of dendritic gold complexes bearing hemipolycatenars 131c. The clearing temperatures decrease with the generation from 116 (G0) to 73 (G1) and 36  C (G2). Based on X-ray diffraction studies the columnar phase results from supramolecular aggregation of molecular dendrimers into one-dimensional cylindrical assemblies, which are further organized to give a hexagonal network. On the other hand, the smectic phases form by the lateral two-dimensional registry of the dendrimers in antiparallel head-to-head prolate conformation.

14.09.9.1.5

Hydrogen-bonded isocyanide derivatives

4-isocyanobenzoic acid gold(I) complexes [AuX(CNC6H4COOH)] (X ¼ Cl, fluoroaryl; metallo-organic acids) have been prepared. These complexes act as proton donors face to decyloxystilbazole, to yield supramolecular metal complexes 132 (R ¼ Cl, C6F5, C6F4OC6H13, C6F4C6F4Br) showing liquid crystalline behavior.118,119 132 with the AudCl fragment shows a calamitic SmA mesophase in the short range 170–187  C, while the Au–fluoroaryl fragments show rod-like SmC mesophases with melting points in the range 127–190  C, and clearing points from 170 to 190  C. The fluoroaryl gold(I)–stilbazole aggregates exhibit luminescence in the solid state at room temperature with emission maxima in the range 468–485 nm. This emission has been assigned as an overlapping of the emissions coming from the free stilbazole and the gold(I) moiety. The corresponding H-bond palladium and platinum supramolecular complexes were described in Section 14.09.8.1.

In order to reduce the transition temperatures, supramolecular metal complexes 133 (R ¼ C6F5, C6F4OC10H21) formed through H-bonds between tris(3,4,5-decyloxy)stilbazole and several gold(I) acids [AuR(CNC6H4COOH)] were prepared.120 Unexpectedly, the compounds are not mesomorphic, although the corresponding palladium and platinum trans complexes show hexagonal columnar mesophases (Section 14.09.8.1). Remarkably, the H-bond in the gold aggregate [Au(C6F4OC10H21)(CNC6H4COOH)] survives on the water surface giving rise to Langmuir films where the molecules are parallel to the water surface. Both gold adducts emit intensely at room temperature in the solid state, with emission maxima at 497 and 528 nm, respectively. The emission was related to the luminescent stilbazole modified by the gold fragment.

This strategy can lead not only to calamitic mesogens but also to discotic liquid crystals. As commented in Sections 14.09.4 and 14.09.6.1, supramolecular aggregates were reported by reaction of carboxylic acid, namely para-isocyanobenzoic metal complex (carbonyl compounds of Fe, Cr, Mo and), with 2,4,6-triarylamino-1,3,5-triazine.61 They behave as discotic liquid crystals at room temperature. By using this strategy, mono- and dinuclear thiolatobenzoic gold(I) [Au(PR3)(x-SC6H4COOH)] and [m-(BINAP) {Au(x-SC6H4COOH)}2] (x ¼ 2 or 4; BINAP: 2,2’-Bis(diphenylphosphino)-1,10 -binaphthalene) metalloacids were combined with the same triazine to yield the corresponding adducts 134a (ortho, para substitution; R ¼ C10H21) and 134b (P–P ¼ R-BINAP, S-BINAP; ortho, para substitution; R ¼ C10H21) by means of H-bonds.190 They show Colhex mesophases at room temperature, with clearing temperatures in the short range 43–57  C, close or lower than for the free triazine (57  C), while for the previous metallo-acids (Fe, Cr, Mo, and W) is higher. The diffractograms in the mesophase display a new broad peak, which may be related to gold–gold distances from closest neighbors. Remarkably, hexagonal columnar mesophases found in the free triazine were preserved in the metal-containing supramolecular aggregates. The nine alkoxy chain triazine structure is able to swallow metal-organic fragments, even containing bulky ligands as phosphines.

Organometallic Mesogens

329

As commented in Section 14.09.4 for chromium(0) [Cr(CO)5(z-PPh2C6H4COOH)] (z ¼ 2 or 4), an extension of this work was carried out by synthesizing neutral gold(I) and gold(III) [AuXn(z-PPh2C6H4COOH)] (n ¼ 1, X ¼ Cl, z ¼ 2 or 4; n ¼ 3, X ¼ C6F5, z ¼ 2 or 4) and cationic gold(I) [Au(z-PPh2C6H4COOH)2](CF3SO3) (z ¼ 2 or 4), ortho and para phosphino metallo-acids.62 X-Ray diffraction show dimeric structures with the expected double carboxylic H-bonds in [AuCl(4-PPh2C6H4COOH)] and [Au(C6F5)3 (4-PPh2C6H4COOH)], whereas [Au(C6F5)3(2-PPh2C6H4COOH)] display monomeric species with the carboxylic acid H bonded to cocrystallized solvent molecules. Gold compounds are emissive at 77 K with emission maxima from 404 to 520 nm, and some of them also at room temperature with emission maxima in the range 441–485 nm. Reactions with the 2,4,6-triarylamino-1, 3,5-triazine mesogen leads to hydrogen-bonded gold(I) supramolecular adducts, while the related gold(III) complexes do not form adducts likely by steric reasons. Only the 4-diphenylphosphinobenzoic adducts 135a and 135b show a columnar hexagonal mesophase at room temperature, with clearing temperatures in the range 36–60  C for gold(I). The aggregates are luminescent at 77 K, with emission maxima in the range 419–455 nm. Again is confirmed the triazine bearing nine lateral alkoxy chains is able to accommodate large metal −organic fragments to produce adducts without losing the mesomorphic behavior, although the weakness of the connection of the two components limits the thermal stability. However, the ortho-phosphine metalloacids are clearly detrimental for the formation of these adducts and the liquid crystal behavior.

By this approach 4-isocyanoanilides have been used to prepare Ag(I) neutral complexes 136a [Ag(X)(CN–C6H4–NHCOR)] (R ¼ Me, X ¼ NO−3, CF3SO−3; R ¼ C9H19, X ¼ NO−3) and cationic complexes 136b [Ag(CN–C6H4–NHCOR)2]X (R ¼ Me, X ¼ NO−3, CF3SO−3, BF−4; R ¼ C9H19, X ¼ NO−3, H25C12OSO−3, CF3SO−3, BF−4).191 Single crystal X-ray diffraction studies of methyl derivatives show layered supramolecular structures supported by Ag–O interactions and hydrogen bonds, induced by the amide group and the anionic ligand. Substitution of the methyl group by a nonyl chain in the amide leads to mesomorphic complexes, which display SmA mesophases as expected by the methyl structures, and supported by the presence of H-bonds in the FTIR spectra. Melting

330

Organometallic Mesogens

temperatures go from 91 to 144  C, while clearing temperatures are in the range 138–200  C, with a maximum liquid crystal range of 109  C for compound 136b (R ¼ C9H19, X ¼ NO−3).

14.09.9.1.6

Isocyanide for discotic mesogens

We have described that organometallic discotic mesogens have been synthesized by means of H–bonds around a 2,4,6– triarylamino–1,3,5–triazine core. These bonds are labile and the corresponding liquid crystals are stable over a short temperature range. A better strategy for preparing organometallic discotic mesogens is the use of suitable covalent bonds around an appropriate core. Hexaalkoxy triphenylene derivatives have been thoroughly used to prepare purely organic columnar discotic liquid crystals. The right functionalization allows coordinating metal fragments, which in many cases are mesogens. By this approach, the new ligand 1–isocyano–2,3,6,7,10,11–hexadodecyloxytriphenylene (CN–TriPh) have been synthesized in order to prepare organometallic mononuclear gold(I) 137a (X ¼ Cl, C6F5, C6F4O–C10H21, C6F4O–(R)–2–octyl), dinuclear gold(I) 137b, copper(I) 137c and silver(I) 137d.192 The free isocyanide ligand and the gold(I) 137a–b compounds exhibit typical hexagonal columnar mesophases over a wide range of temperatures, from 5 to 220  C. Melting points are 37  C for the ligand and from 5 to 7  C for 137a; clearing points are 130  C for the ligand and in the range 204–220  C for 137a. Gold compound 137b displays a rectangular columnar mesophase between 102 and 122  C, and a nematic mesophase till 172  C. On the contrary copper and silver complexes 137c–d are not mesomorphic. This may be explained for complex 137d due to the presence of two trans coordinated isocyanide ligands, which cannot be coplanar due to the too short linker, while the longer linker in 137b compound allows planarity. All the compounds show a fluorescent emission in solution centered on the triphenylene core in the range 421–467 nm; the emission is lost in concentrated solutions or in aggregate states. The nature of the metallic fragment is dramatically affecting the liquid crystal behavior, which can be mainly related to the fact than the isocyanide is directly bonded to the triphenylene core.

As commented above in Section 14.09.8.1 (palladium and platinum), and following the same approach, a new isocyano ligand namely 2-(6-(4-isocyanophenoxy)hexyloxy)-3,6,7,10,11-pentakisdodecyloxytriphenylene have been prepared and used to synthesize organometallic gold(I) and copper(I) 138a (M ¼ Au, X ¼ Cl, C6F5, C6F4O-C10H21, CN; M ¼ Cu, X ¼ Cl), dinuclear gold(I) 138b, and dinuclear copper(I) 138c complexes.121 Compounds 138a (only [Au(CN)(CNR)] and [CuCl(CNR)]) and 138c display a rectangular columnar mesophase in a short range: 61–98  C, 46–79  C and 72–79  C, respectively. 138b shows a lamellar rectangular columnar mesophase in a wider range: 22–98  C. Both types of mesophase show at the same time, p–stacking of the triphenylene cores and aggregation of the metallic fragments into segregated columnar zones. Therefore, heterocolumnar arrangements are obtained as a consequence of the nanosegregation induced by the different nature of the molecular components. All these complexes show again fluorescence in solution related to the isocyanide triphenylene ligand. Pd and Pt complexes exhibit semiconducting mesophases as commented above. The previous isocyanide triphenylene was equipped with six equal alkoxy chains and the isocyanide was directly bound to the triphenylene core, therefore the metallic moiety was dramatically affecting the thermal properties. However, in this case the triphenylene was equipped with five equal alkoxy chains and the isocyanide is at the end of the sixth alkoxy chain: the effect is much less intense, but the metals can form their own column inducing a heterocolumnar rearrangement.

Organometallic Mesogens

331

In addition to triphenylene, cyclotriphosphazene has been used as core to support columnar discotic liquid crystals. Aminocyclotriphosphazenes have been reacted with the isocyanide silver(I) complex [Ag(OTf )(CNR)] (OTf ¼ OSO2CF3) in molar ratio 1:3, to give room temperature columnar metallomesogens 139 (R ¼ R’ ¼ Me; R ¼ H, R’ ¼ Cy).193 The same complexes in molar ratios 1:1 or 1:2 are not mesomorphic. Each silver fragment “AgCNR” is bonded to one nitrogen in the phosphazene core. The starting silver complex [Ag(OTf)L] and the trinuclear complexes equipped with nine alkoxy chains display a columnar hexagonal mesophase with isotropization temperatures of 145, 61 and 72  C, respectively, for [Ag(OTf )(CNR)] and the two trinuclear silver phosphazenes. A model consisting of two complementary molecules leads to an overall disk shape, which promotes the supramolecular columnar rearrangement driven by efficient space–filling and nanosegregation of chemically different zones.

14.09.9.1.7

Isocyanide as a colorant

The introduction of dyes into mesomorphic systems is a useful strategy for the preparation of new functional materials. 4,4’–Disubstituted azobenzene compounds with adequate substituents have been used to prepare ortho-metalated mesogens, mainly with group 9–10 metals. Azo derivatives are well known dyes and the color is usually transferred to the new compounds. The coordinated azo fragment is blocked in its trans conformation, and therefore the photosensitivity is lost. New azo isocyanide mesogens were designed, with an azo, an isocyanide, and an alkoxy chain as functional groups. These ligands exhibit nematic and SmA mesophases for n > 4, and they are able to coordinate metals through the isocyanide. The new complexes are intensely orange and the most interesting feature is that possess a free photosensitive azo functional group. Their gold(I) compounds 140 (X ¼ Cl, C6F5; n ¼ 4, 8, 12), also display nematic and SmA mesophases, but with higher transition temperatures and wider mesophase ranges.194 The complexes are photoresponsive in solution due to fast trans to cis isomerization of the azo group under UV light (365 nm lamp), which goes back photochemically or thermally to the trans isomer. Most relevant, the process can be carried out in the mesophase by irradiation with a very intense He–Cd laser, which produces rapid isomerization, and consequent destabilization to an isotropic liquid. The initial mesophase is recovered as soon as illumination stops (Fig. 9). As expected, linear gold complexes exhibit high birefringence values (Dn from 0.51 to 0.59) than the free azo ligand (Dn are 0.32), because of gold increases the molecular anisotropy. In summary, these colored azo gold–containing mesogens are photosensitive, not only in solution but also in the mesophase, remarkably the photoisomerization has been demonstrated in a mesophase showed by a pure metallomesogen.

Fig. 9 Changes in azo isocyanide gold mesogens 140 after irradiation with a laser.

332

Organometallic Mesogens

Following the same approach, azulene, a classical azure–blue chromophore was functionalized with an isocyanide group and coordinated to gold(I) synthons to prepare mononuclear 141a (X ¼ Cl, C6F5), 141b (x ¼ 1, R ¼ C8H17, C10H21; x ¼ 2, R ¼ C10H21), 141c (R ¼ Me, C10H21), and dinuclear 141d complexes.195 Gold compounds 141b (x ¼ 1, 2; R ¼ C10H21) and 141c (R ¼ C10H21) display SmA mesophases being the widest range for the alkynyl ligand from 106 to 150  C. The free isocyanoazulene derivative shows a weak fluorescent emission centered on the azulene group at 380 nm, which diminishes upon coordination to the gold moieties. All the compounds are intensely blue colored. Theoretical studies confirmed that the UV–vis absorption (color) and emission are centered on the azulene core.

14.09.9.2 Carbene complexes N–heterocyclic carbenes may be easily modified to modulate their electronic and steric properties, a strategy commonly used to improve catalyst but can be also useful for the design of liquid crystals. A series of N–heterocyclic carbene (NHC) imidazole gold(I) complexes 142 (n ¼ 12, 18, m ¼ 1, 6, 12, 18; n ¼ 16, m ¼ 10, 12, 14, 16, 18) have been prepared.196 They exhibit a SmA mesophase, typical for ionic liquid crystals. The melting and clearing temperatures can be tuned mainly by the imidazole alkyl chain length: longer chain, higher clearing and melting points. The widest intervals are obtained for the C18 carbene: the melting points are from 57.3 to 101.5  C, and the clearing points from 96.9 to 124.1  C. These complexes were used to obtain gold nanoparticles in organic phase by chemical reduction. The related complexes [AuCl(NHC)] and [Au(NHC)2][NO3] are not mesomorphic although they possess two or four alkyl chains. Self–assembly of N–heterocyclic carbenes containing two different chains (one amide able to form H–bonds and one aliphatic) yields gold(I) and silver(I) complexes 143 (R ¼ CnH2n+1; n ¼ 12, 14, 16, 18. R’ ¼ CH2C(O)NH2. M ¼ Au and X ¼ Br, NO3, BF4. M ¼ Ag and X ¼ NO3, BF4), which behave as enantiotropic liquid crystals.197 They display SmA mesophases with melting points around 150  C for gold, 170  C for silver–BF4 and 110  C for silver–NO3, and clearing points up to 226  C for gold (being the best with n ¼ 16 and bromide as anion) and 200  C for silver with decomposition. Gold(I) compounds form xerogels in DMSO and show oriental lantern-shaped bundles of belts and helical fibers when observed by transmission electron and scanning electron microscopies. These gels display smaller chain motion than in the mesophase, closer to found in the solid state, as studied by infrared spectroscopy.

Imidazolium salts are well known as ionic liquids and also as ionic liquid crystals. Moreover, they can be used as N–heterocyclic carbene (NHC) precursors. A series of bis(imidazolium) salts with appropriate mesogenic groups (cyanobiphenyl or cholesteryl) linked through a flexible alkyl spacer have been designed and prepared in order to get their silver carbene complexes 144.198 The silver complexes and the bis(imidazolium) salts display SmA mesophases with close melting point but wider temperature ranges for silver. The silver complex with cholesteryl group shows higher transition temperatures (51–180  C) than the complexes containing cyanobiphenyl mesogenic groups (36–54  C, 39–76  C). The silver complexes show a blue emission centered around 450 nm in

Organometallic Mesogens

333

solid state, as 10% PMMA (Polymethylmethacrylate) films, in isotropic and LC phase, and in dichloromethane solution (centered in this case at 360 nm), which it is related to the bis(imidazolium) salts.

By the same approach, propylene bridged bis(imidazolium) salts bearing benzyl group functionalized with one or two long alkyl chains have been synthesized. The corresponding N–heterocyclic carbene dinuclear gold(I) complexes 145 (R ¼ CH2–3,4– C6H3–(OC12H25)2; X ¼ Br, PF6, BF4) have been also prepared but only complexes with eight aliphatic chains show lamellar mesophases in a short range.199 Imidazol core can be substituted by benzimidazol, and equipped with an 18 alkyl chain on 1 N atom and an alkyl or substituted benzyl moiety on the other N. The corresponding mononuclear silver(I) N–heterocyclic dicarbene complexes 146 (R ¼ CnH2n+1; n ¼ 10, 12, 14, 16. R ¼ 4–CH2–C6H4–R; R’]H, CH3, CN) have been prepared.200 Only two silver complexes containing two long alkyl chains display monotropic SmA in a short range (the best 21  C) at temperatures under 100  C. The related silver carbene compounds with R ¼ Me, and alkyl chains CnH2n+1 (n ¼ 12, 14, 16, 18) instead of 18 in 146 are not mesomorphic.201 As seen in other reports, the use of bulky hexafluorophosphate is detrimental for liquid crystal behavior and it is the only anion in this work.

N–heterocyclic carbene silver(I) complex [{Ag(NHC)2}2][Ag2Cl4] 147a (R ¼ CH2–CH(OH)–C14H29; R’ ¼ C16H33) behaves as liquid crystal while the related gold(I) complexes 147b and 147c (X ¼ BF4, PF6) are soft materials.202 This carbene combines an alkyl chain with an hydroxy alkyl chain in order to facilitate not only hydrophobic interactions but also hydrogen bonding. A typical SmA mesophase is observed in a narrow 5  C temperature range of 92.6–97.6  C, expanding to 50  C upon cooling. Complexes 147a and 147b exhibit yellow (569 nm) and orange (607 nm) emissions.

334

Organometallic Mesogens

Gold(I) complexes 148 combining alkoxy acetylide and a carbene derived from benzimidazolium equipped with two alkoxy chains in the phenyl ring have been reported. Only the trialkoxy compound behaves as monotropic liquid crystal and displays a SmA mesophase in a short range (190–203  C). This complex exhibit an intense blue phosphorescence at room temperature in the solid state (quantum yield 0.33; lifetime 36 ms).203

A series of N–heterocyclic carbene complexes 149a [MX(NHC)] (R ¼ R0 ¼Tph; M ¼ Cu, X ¼ Br; M ¼ Au, X ¼ Cl, CCPh), and silver 149b (R ¼ Me, R0 ¼ Tph; R ¼ R0 ¼Tph), with the carbene bearing one or two triphenylene fragments have been reported.125 Copper(I) 149a and silver(I) 149b complexes align into columnar liquid crystals (rectangular or hexagonal symmetries), driven by peripheral hexaalkoxy triphenylene moieties. Melting points go from 30 to 55  C and clearing points are in the range 57–98  C. Models point to multicolumnar systems obtained by nanosegregation of triphenylene columns and metal carbene moieties, separated by alkoxy chains. The complexes show a blue emission related to the triphenylene core in solution, in the mesophase, in the isotropic liquid, and in the solid state.

Finally, a series of phosphorescent gold(I) N–heterocyclic allenylidene complexes 150 (X ¼ Cl, CF3SO3, PF6) have been reported.204 These organometallic complexes exhibit a strong blue emission (centered 414–450 nm) in acetonitrile solutions with life time in the range 1.4–5.6 micros and quantum yield from 7% to 14%. Complex with imidazole substituents and chloride formed lyotropic chromonic mesophases (N phase) in aqueous solutions.

14.09.10

Concluding comments

The literature on organometallic mesogens is more limited than that on mesomorphic coordination compounds and is dominated by ortho–metalated complexes (mainly Pd(II) and Pt(II)), and ferrocene derivatives. In general, the M–C bond is not very thermally stable and tends to be labile and easily reacts with moisture and oxygen. That is why M–CO and especially M–CNR with the metal in low oxidation state, which are quite stable, are also typical fragments in many metallomesogens. However, ligand design has made it possible to decorate chelates (to metalate), cyclopentadienyls (in ferrocene) and isocyanides with a huge number of chemical motifs to add or enhance the optical, electrical, and magnetic properties of liquid crystal materials.

Acknowledgments We gratefully acknowledge the Ministerio de Economía y Competitividad (Projects CTQ2017–89217–P and PID2020118547GB-I00) and the Junta de Castilla y León (Project VA224P20) for financial support.

Organometallic Mesogens

335

References 1. Donnio, B.; Guillon, D.; Bruce, D. W.; Deschenaux, R. Metallomesogens. In Comprehensive Organometallic Chemistry III: From Fundamentals to Applications; Crabtree, R. H., Mingos, D. M. P., Eds.; Elsevier: Oxford, UK, 2006; vol. 12 pp 195–293. Chapter 12.05. 2. Jonas, U.; Petri, C.; Kunzler, C.; Spiess, H. W. Polymeric Discotic Liquid Crystals, Reference Module in Materials Science and Materials Engineering; Elsevier, 2017. 3. Zhu, L., Li, C. Y., Eds.; In Liquid Crystalline Polymers; Springer, 2020. 4. Prakash, J.; Khan, S.; Chauhan, S.; Biradar, A. M. J. Mol. Liq. 2020, 297, 112052. 5. Ermakov, S. F.; Myshkin, N. K. Liquid-Crystal Nanomaterials: Tribology and Applications; Springer, 2018. 6. Choudhary, A.; Singh, G.; Biradar, A. M. Nanoscale 2014, 6, 7743–7756. 7. Serrano, J. L. Metallomesogens: Synthesis, Properties and Applications; Wiley-VCH: Weinheim, Germany, 1996. 8. Donnio, B.; Guillon, D.; Deschenaux, R.; Bruce, D. W. Metallomesogens. In Comprehensive Coordination Chemistry II; McCleverty, J. A., Meyer, T. J., Eds.; Elsevier: Oxford, UK, 2003; vol. 7 pp 357–627. Chapter 7.9. 9. Torroba, J.; Bruce, D. W. Metallomesogens. In Comprehensive Inorganic Chemistry II (Second Edition): From Elements to Applications; Reedijk, J., Poeppelmeier, K., Eds.; Elsevier, 2013; vol. 8 pp 837–917. Chapter 8.21. 10. Pucci, D.; Donnio, B. Metal-containing liquid crystals. In Handbook of Liquid Crystals; Non-Conventional, Supramolecular, Chromonic and Amphiphilic Liquid Crystals; Goodby, J. W., Collings, P. J., Kato, T., Tschierske, C., Gleeson, H., Raynes, P., Eds.; Wiley-VCH: Weinheim, Germany, 2014; vol. 5 pp 1–67. Chapter 4. 11. Barón, M. Pure Appl. Chem. 2001, 73, 845–895. 12. Leadbetter, A. J. In Thermotropic Liquid Crystals; Gray, G. W., Ed.; Wiley: Chichester, 1987 pp 1–27. Chapter 1. 13. (a) Gray, G. W.; Goodby, J. W. Smectic Liquid Crystals; Textures and Structures; Leonard Hill: Glasgow, 1984; (b) Dierking, I. Textures of Liquid Crystals; Wiley-VCH: Weinheim, 2003. 14. Malthete, J.; Nguyen, H.-T.; Destrade, C. Liq. Cryst. 1993, 13, 171–187. 15. Nguyen, H. L.; Destrade, C.; Malthête, J. Adv. Mater. 1997, 9, 375–388. 16. Ros, M. B.; Serrano, J. L.; de la Fuente, M. R.; Folcia, C. L. J. Mater. Chem. 2005, 15, 5093–5098. 17. Collings, P. J.; Goodby, J. W. Introduction to Liquid Crystals: Chemistry and Physics, 2nd edn.; CRC Press, Taylor & Francis Group, 2019. 18. Collings, P. J.; Hird, M. Introduction to Liquid Crystals: Chemistry and Physics; CRC Press, Taylor & Francis Group, 1997. 19. Collings, P. J. Liquid Crystals; Princeton University Press: New Jersey, NJ, 1990. 20. Reinitzer, F. Monatsch. Chem. 1888, 9, 421–441. 21. Heintz, W. J. Prakt. Chem. 1855, 66, 1–51. 22. Vorländer, D. Z. Phys. Chem. 1923, 105, 211–254. 23. Giroud-Godquin, A.-M.; Maitlis, P. M. Angew. Chem. Int. Ed. Engl. 1991, 30, 375–402. 24. Acree, W. E.; Chickos, J. S. J. Phys. Chem. Ref. Data 2006, 35, 1051–1330. 25. Gray, G. W.; Harrison, K. J.; Nash, J. A. Electron. Lett. 1973, 9, 130–131. 26. Yang, D.-K.; Wu, S.-T. Fundamentals of Liquid Crystal Devices; John Wiley & Sons, 2014. 27. Tschierske, C., Ed.; Theme Issue: Liquid Crystals Beyond Display Applications. J. Mater. Chem. 2008, 18, 2857–3060. 28. Bisoyia, H. K.; Kumar, S. Chem. Soc. Rev. 2010, 39, 264–285. 29. Kaafarani, B. R. Chem. Mater. 2011, 23, 378–396. 30. Goossens, K.; Lava, K.; Bielawski, C. W.; Binnemans, K. Chem. Rev. 2016, 116, 4643–4807. 31. Kim, D.-Y.; Jeong, K.-U. Liq. Cryst. Today 2019, 28, 34–45. 32. Yun, C.-J.; Song, J.-K. J. Inf. Disp. 2017, 18, 119–129. 33. Wei, D. Liq. Cryst. Today 2015, 24, 35–36. 34. Kasch, N. Liq. Cryst. Today 2013, 22, 70–71. 35. Bushby, R. J., Kelly, S. M., O’Neill, M., Eds.; In Liquid Crystalline Semiconductors: Materials, Properties and Applications; Springer, 2012. 36. Zhao, J.; Gulan, U.; Horie, T.; Ohmura, N.; Han, J.; Yang, C.; Kong, J.; Wang, S.; Xu, B. B. Small 2019, 15, 1900019. 37. Binnemans, K. J. Mater. Chem. 2009, 19, 448–453. 38. Gasparb, A. B.; Seredyuka, M.; Gütlich, P. Coord. Chem. Rev. 2009, 253, 2399–2413. 39. Binnemans, K. Physical Properties of Metallomesogens. In Molecular Materials. Inorganic Materials Series. 4; Bruce, D. W., O’Hare, D., Walton, R. I., Eds.; Wiley: Chichester, 2010; pp 61–141. 40. Wang, Y.; Shi, J.; Chen, J.; Zhu, W.; Baranoff, E. J. Mater. Chem. C 2015, 3, 7993–8005. 41. Wöhrle, T.; Wurzbach, I.; Kirres, J.; Kostidou, A.; Kapernaum, N.; Litterscheidt, J.; Haenle, J. C.; Staffeld, P.; Baro, A.; Giesselmann, F.; Laschat, S. Chem. Rev. 2016, 116, 1139–1241. 42. Wu, X.; Zhu, M.; Bruce, D. W.; Zhu, W.; Wang, Y. J. Mater. Chem. C 2018, 6, 9848–9860. 43. Guy, K.; Ehni, P.; Paofai, S.; Roiland, C.; Amela-Cortes, M.; Cordier, S.; Laschat, S.; Molard, Y. Angew. Chem. Int. Ed. 2018, 57, 11692–11696. 44. Binnemans, K. Lanthanidomesogens. In Handbook on the Physics and Chemistry of Rare Earths; Bünzli, J. C. G., Pecharsky, V. K., Eds.; 2013 pp 1–158.Chapter 254. 45. Guerra, S.; Dutronc, T.; Terazzi, E.; Buchwalder, K.-L.; Guénée, L.; Deschenaux, R.; Eliseeva, S. V.; Petoud, S.; Piguet, C. Coord. Chem. Rev. 2017, 340, 79–97. 46. Rajendiran, K.; Yoganandham, S. T.; Arumugam, S.; Arumugam, D.; Thananjeyan, K. J. Mol. Liq. 2021, 321, 114793. 47. Ringstrand, B.; Jankowiak, A.; Johnson, L. E.; Kaszynski, P.; Pociecha, D.; Górecka, E. J. Mater. Chem. 2012, 22, 4874–4880. 48. Pecyna, J.; Sivaramamoorthy, A.; Jankowiak, A.; Kaszynski, P. Liq. Cryst. 2012, 39, 965–971. 49. Tatum, L. A.; Johnson, C. J.; Fernando, A. A. P.; Ruch, B. C.; Barakoti, K. K.; Alpuche-Aviles, M. A.; King, B. T. Chem. Sci. 2012, 3, 3261–3264. 50. Wöhrle, T.; Baro, A.; Laschat, S. Materials 2014, 7, 4045–4056. 51. Wöhrle, T.; Gündemir, R.; Frey, W.; Knecht, F.; Köhn, A.; Laschat, S. Chem. Eur. J. 2017, 23, 4149–4159. 52. Takahashi, K.; Shimoi, M.; Watanabe, T.; Maeda, K.; Geib, S. J.; Curran, D. P.; Taniguchi, T. Org. Lett. 2020, 22, 2054–2059. 53. Cui, L.; Zhu, L. Langmuir 2006, 22, 5982–5985. 54. Karahaliou, P. K.; Kouwer, P. H. J.; Meyer, T.; Mehl, G. H.; Photino, D. J. Soft Matter 2007, 3, 857–865. 55. Wang, X.; Cho, C. M.; Say, W. Y.; Tan, A. Y. X.; He, C.; On Chan, H. S. O.; Xu, J. J. Mater. Chem. 2011, 21, 5248–5257. 56. Shao, Y.; Xu, X.; Yin, G.-Z.; Han, S.-Y.; Han, D.; Fu, Q.; Yang, S.; Zhang, W.-B. Macromolecules 2019, 52, 2361–2370. 57. Miao, J.; Zhu, L. J. Phys. Chem. B 2010, 114, 1879–1887. 58. Cui, L.; Collet, J. P.; Xu, G.; Zhu, L. Chem. Mater. 2006, 18, 3503–3512. 59. Thurmes, W. N.; More, K. M.; Meadows, M. R.; O’Neill, M. B.; Vohra, R. T.; Wand, M. D. Liq. Cryst. 2009, 36, 461–477. 60. Parera, E.; Comelles, F.; Barnadas, R.; Suades, J. Chem. Commun. 2011, 47, 4460–4462. 61. Coco, S.; Cordovilla, C.; Domínguez, C.; Donnio, B.; Espinet, P.; Guillon, D. Chem. Mater. 2009, 21, 3282–3289. 62. Miguel-Coello, A. B.; Bardají, M.; Coco, S.; Donnio, B.; Heinrich, B.; Espinet, P. Inorg. Chem. 2014, 53, 10893–10902. 63. Allenbaugh, R. J.; Aoyama, Y.; Hyatt, M. G.; Reynolds, T. H.; Morgan, M. L.; Adler, A. M.; Schauer, C. K. Liq. Cryst. 2013, 40, 783–786.

336

Organometallic Mesogens

64. Cardinaels, T.; Ramaekers, J.; Nockemann, P.; Driesen, K.; Van Hecke, K.; Van Meervelt, L.; Lei, S.; De Feyter, S.; Guillon, D.; Donnio, B.; Binnemans, K. Chem. Mater. 2008, 20, 1278–1291. 65. Cardinaels, T.; Ramaekers, J.; Nockemann, P.; Driesen, K.; Van Hecke, K.; Van Meervelt, L.; Wang, G.; De Feyter, S.; Iglesias, E. F.; Guillon, D.; Donnio, B.; Binnemans, K.; Bruce, D. W. Soft Matter. 2008, 4, 2172–2185. 66. Cardinaels, T.; Ramaekers, J.; Driesen, K.; Nockemann, P.; Van Hecke, K.; Van Meervelt, L.; Goderis, B.; Binnemans, K. Inorg. Chem. 2009, 48, 2490–2499. 67. Januszko, A.; Kaszynski, P.; Grüner, B. Inorg. Chem. 2007, 46, 6078–6082. 68. Dembinski, R.; Espinet, P.; Lentijo, S.; Markowicz, M. W.; Martín-Alvarez, J. M.; Rheingold, A. L.; Schmidt, D. J.; Sniady, A. Eur. J. Inorg. Chem. 2008, 1565–1572. 69. Frein, S.; Auzias, M.; Sondenecker, A.; Vieille-Petit, L.; Guintchin, B.; Maringa, N.; Süss-Fink, G.; Barberá, J.; Deschenaux, R. Chem. Mater. 2008, 20, 1340–1343. 70. Pitto-Barry, A.; Barry, N. P. E.; Russo, V.; Heinrich, B.; Donnio, B.; Therrien, B.; Deschenaux, R. J. Am. Chem. Soc. 2014, 136, 17616–17625. 71. Alvariño, C.; Heinrich, B.; Donnio, B.; Deschenaux, R.; Therrien, B. Inorg. Chem. 2019, 58, 9505–9512. 72. Astruc, D. Eur. J. Inorg. Chem. 2017, 6–29. 73. Kadkin, O. N.; Yu, G. G. Russ. Chem. Rev. 2012, 81, 675–699. 74. Deschenaux, R. Ferrocene-Containing Thermotropic Liquid Crystals. In Ferrocenes: Ligands, Materials and Biomolecules; Stepnicka, P., Ed.; John Wiley & Sons, 2008; pp 446–463. 75. Kowalski, K. Ferrocene–Containing Polymers and Dendrimers. In Ferrocenes: Ligands, Materials and Biomolecules; Long, N. J., Stepnicka, P., Eds.; John Wiley & Sons, 2008; pp 393–446. 76. Hussein, M. A.; Asiri, A. M. Des. Monomers Polym. 2012, 15, 207–251. 77. Zhai, X.; Yu, H.; Wang, L.; Deng, Z.; Zain-Ul-Abdin, ; Tong, R.; Yang, X.; Chen, Y.; Saleem, M. Appl. Organomet. Chem. 2016, 30, 62–72. 78. Liu, X.; Zhao, L.; Liu, F.; Astruc, D.; Gu, H. Coord. Chem. Rev. 2020, 419, 213406. 79. Wu, J.; Wang, L.; Yu, H.; Zain-Ul-Abdin, ; Khan, R. U.; Haroon, M. J. Organomet. Chem. 2017, 828, 38–51. 80. Carlescu, I.; Scutaru, A. M.; Apreutesei, D.; Alupei, V.; Scutaru, D. Appl. Organomet. Chem. 2007, 21, 661–669. 81. Carlescu, I.; Scutaru, A. M.; Apreutesei, D.; Alupei, V.; Scutaru, D. Liq. Cryst. 2007, 34, 775–785. 82. Nakamura, N.; Ishimizu, M.; Nishikawa, M.; Dogen, N. Mol. Cryst. Liq. Cryst. 2010, 516, 114–121. 83. Nakamura, N.; Hiro, K.; Nishikawa, M.; Okabe, T.; Uno, K. Mol. Cryst. Liq. Cryst. 2010, 516, 122–131. 84. Nakamura, N.; Mizoguchi, R.; Ueda, M.; Hanasaki, T. Mol. Cryst. Liq. Cryst. 1998, 312, 127–136. 85. Majumdar, K. C.; Chakravorty, S.; Pal, N.; Sinha, R. K. Tetrahedron 2009, 65, 7998–8006. 86. Majumdar, K. C.; Shyam, P. K. Mol. Cryst. Liq. Cryst. 2010, 528, 3–9. 87. Majumdar, K. C.; Shyam, P. K.; Rao, D. S. S.; Prasad, K. Liq. Cryst. 2012, 39, 1117–1123. 88. Majumdar, K. C.; Ghosh, T.; Shyam, P. K. Liq. Cryst. 2011, 38, 567–573. 89. Cheng, H.; Ma, M.; Chen, Y.; Ni, H.-L.; Feng, C.; Wang, B.-Q.; Zhao, K.-Q.; Yu, W.-H.; Hu, P. Liq. Cryst. 2017, 44, 1450–1461. 90. Zhao, H. Y.; Guo, L.; Chen, S. F.; Bian, Z. X. J. Mol. Struct. 2013, 1054–1055, 164–169. 91. Kim, S. Y.; Kadkin, O. N.; Kim, E. H.; Choi, M. G. J. Organomet. Chem. 2011, 696, 2429–2437. 92. Navale, D. N.; Zote, S. W.; Ramana, M. M. V. Liq. Cryst. 2013, 40, 1333–1338. 93. Cheng, Z.; Chang, X.; Qiu, Y.; Tan, G.; Cheng, F.; Fan, H.; Ren, B. J. Organomet. Chem. 2018, 869, 81–87. 94. Tokunaga, S.; Itoh, Y.; Tanaka, H.; Araoka, F.; Aida, T. J. Am. Chem. Soc. 2018, 140, 10946–10949. 95. Kadkin, O. N.; Han, H.; Galyametdinov, Y. G. J. Organomet. Chem. 2007, 692, 5571–5582. 96. Kadkin, O. N.; Kim, E. H.; Kim, S. Y.; Rha, Y. J.; Tae, J.; Choi, M.-G. Chem. Eur. J. 2009, 15, 10343–10347. 97. Kim, E. H.; Kadkin, O. N.; Kim, S. Y.; Choi, M.-G. Eur. J. Inorg. Chem. 2011, 2933–2941. 98. Okabe, T.; Nakazaki, K.; Igaue, T.; Nakamura, N.; Donnio, B.; Guillon, D.; Gallani, J. L. J. Appl. Cryst. 2009, 42, 63–68. 99. Nakamura, N.; Kagawa, S. Mol. Cryst. Liq. Cryst. 2011, 549, 150–159. 100. Bitter, S.; Kunkel, M.; Burkart, L.; Mang, A.; Winter, R. F.; Polarz, S. ACS Omega 2018, 3, 8854–8864. 101. Kadkin, O. N.; An, J.; Han, H.; Galyametdinov, Y. G. Eur. J. Inorg. Chem. 2008, 1682–1688. 102. Chae, H. W.; Kadkin, O. N.; Choi, M.-G. Liq. Cryst. 2009, 36, 53–60. 103. Kadkin, O. N.; Kim, E. H.; Kim, S. Y.; Choi, M. G. Polyhedron 2009, 28, 1301–1307. 104. Li, Z.; Liu, J.; Zhao, H.; Li, B.; Bian, Z. Res. Chem. Intermed. 2015, 41, 8545–8556. 105. Guillon, D.; Donnio, B.; Deschenaux, R. [60]Fullerene–Containing Thermotropic Liquid Crystals. In Supramolecular Chemistry of Fullerenes and Carbon Nanotubes; Martín, N., Nierengarten, J.-F., Eds.; Wiley-VCH: Weinheim, 2012; pp 203–235. Chapter 9. 106. Campidelli, S.; Severac, M.; Scanu, D.; Deschenaux, R.; Vazquez, E.; Milic, D.; Prato, M.; Carano, M.; Marcaccio, M.; Paolucci, F.; Aminur Rahman, G. M.; Guldi, D. M. J. Mater. Chem. 2008, 18, 1504–1509. 107. Dvinskikh, S. V.; Yamamoto, K.; Scanu, D.; Deschenaux, R.; Ramamoorthy, A. J. Phys. Chem. B 2008, 112, 12347–12353. 108. Li, C.-Z.; Matsuo, Y.; Nakamura, N. J. Am. Chem. Soc. 2010, 132, 15514–15515. 109. Giménez, R.; Elduque, A.; López, J. A.; Barberá, J.; Cavero, E.; Lantero, I.; Oro, L. A.; Serrano, J. L. Inorg. Chem. 2006, 45, 10363–10370. 110. Rota Martir, D.; Zysman-Colman, E. Coord. Chem. Rev. 2018, 364, 86–117. 111. Szerb, E. I.; Talarico, A. M.; Aiello, I.; Crispini, A.; Godbert, N.; Pucci, D.; Pugliese, T.; Ghedini, M. Eur. J. Inorg. Chem. 2010, 3270–3277. 112. Talarico, A. M.; Ghedini, M.; Cesare Oliverio Rossi, C. O.; Szerb, E. I. Soft Matter. 2012, 8, 11661–11669. 113. Wu, X.; Xie, G.; Cabry, C. P.; Xu, X.; Cowling, S. J.; Bruce, D. W.; Zhu, W.; Baranoff, E.; Wang, E. Y. J. Mater. Chem. C 2018, 6, 3298–3309. 114. Santoro, A.; Prokhorov, A. M.; Kozhevnikov, V. N.; Whitwood, A. C.; Donnio, B.; Williams, J. A. G.; Bruce, D. W. J. Am. Chem. Soc. 2011, 133, 5248–5251. 115. Wang, Y.; Cabry, C. P.; Xiao, M.; Male, L.; Cowling, S. J.; Bruce, D. W.; Shi, J.; Zhu, W.; Baranoff, E. Chem. Eur. J. 2016, 22, 1618–1621. 116. Prokhorov, A. M.; Santoro, A.; Williams, J. A. G.; Bruce, D. W. Angew. Chem. Int. Ed. 2012, 51, 95–98. 117. Zou, G.; Zhao, L.; Zeng, L.; Luo, K.; Ni, H.; Wang, H.; Li, Q.; Yu, W.; Li, X. Inorg. Chem. 2019, 58, 861–869. 118. Coco, S.; Espinet, E.; Espinet, P.; Palape, I. Dalton Trans. 2007, 3267–3272. 119. Coco, S.; Cordovilla, C.; Domínguez, C.; Espinet, P. Dalton Trans. 2008, 6894–6900. 120. Domínguez, C.; Donnio, B.; Coco, S.; Espinet, P. Dalton Trans. 2013, 42, 15774–157984. 121. Chico, R.; de Domingo, E.; Domínguez, C.; Donnio, B.; Heinrich, B.; Termine, R.; Golemme, A.; Coco, S.; Espinet, P. Chem. Mater. 2017, 29, 7587–7595. 122. Tritto, E.; Chico, R.; Ortega, J.; Folcia, C. L.; Etxebarria, J.; Coco, S.; Espinet, P. J. Mater. Chem. C 2015, 3, 9385–9392. 123. Coco, S.; Cordovilla, C.; Donnio, B.; Espinet, P.; García-Casas, M. J.; Guillon, D. Chem. Eur. J. 2008, 14, 3544–3552. 124. Wang, X.; Sobota, M.; Kohler, F. T. U.; Morain, B.; Melcher, B. U.; Laurin, M.; Wasserscheid, P.; Libuda, J.; Meyer, K. J. Mater. Chem. 2012, 22, 1893–1898. 125. (a) Miguel-Coello, A. B.; Bardají, M.; Coco, S.; Donnio, B.; Heinrich, B.; Espinet, P. Inorg. Chem. 2018, 57, 4359–4369; (b) Camerel, F.; Ziessel, R.; Donnio, B.; Bourgogne, C.; Guillon, D.; Schmutz, M.; Iacovita, C.; Bucher, J.-P. Angew. Chem. Int. Ed. 2007, 46, 2659–2662. 126. Mayoral, M. J.; Cornago, P.; Claramunt, R. M.; Cano, M. New J. Chem. 2011, 35, 1020–1030. 127. Mayoral, M. J.; Ovejero, P.; Campo, J. A.; Heras, J. V.; Oliveira, E.; Pedras, B.; Lodeiro, C.; Cano, M. J. Mater. Chem. 2011, 21, 1255–1263. 128. Moreno-Mañas, M.; Reichardt, C.; Sebastián, R. M.; Barberá, J.; Serrano, J. L.; Sierra, T. J. Mater. Chem. 2005, 15, 2210–2219.

Organometallic Mesogens

129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. 180. 181. 182. 183. 184. 185. 186. 187. 188. 189. 190. 191. 192. 193. 194. 195. 196. 197.

337

Donnio, B.; Bruce, D. W. Liquid Crystalline Orthopalladated Complexes. In Palladacycles; Dupont, J., Pfeffer, M., Eds.; Wiley-VCH: Weinheim, 2008; pp 239–283. Ghedini, M.; Pucci, D.; Viñuales, A. Mol. Cryst. Liq. Cryst. 2007, 465, 59–70. Godbert, N.; Dattilo, D.; Termine, R.; Aiello, I.; Bellusci, A.; Crispini, A.; Golemme, A.; Ghedini, M. Chem. Asian J. 2009, 4, 1141–1146. Choudhury, T. D.; Shen, Y.; Rao, N. V. S.; Clark, N. A. J. Organomet. Chem. 2012, 712, 20–28. Arias, J.; Bardají, M.; Espinet, P. J. Organomet. Chem. 2006, 691, 4990–4999. Coco, S.; Cordovilla, C.; Espinet, P.; Gallani, J.-L.; Guillon, D.; Donnio, B. Eur. J. Inorg. Chem. 2008, 1210–1218. Bilgin-Eran, B.; Ocak, H.; Tschierske, C.; Baumeister, U. Liq. Cryst. 2012, 39, 467–476. Bilgin-Eran, B.; Tschierske, C.; Dielec, S.; Baumeister, U. J. Mater. Chem. 2006, 16, 1136–1144. Bilgin-Eran, B.; Tschierske, C.; Dielec, S.; Baumeister, U. J. Mater. Chem. 2006, 16, 1145–1153. Mocanu, A. S.; Ilis, M.; Dumitrascu, F.; Ilie, M.; Cîrcu, V. Inorg. Chim. Acta 2010, 363, 729–736. Ilis, M.; Micutz, M.; Dumitrasçu, F.; Pasuk, I.; Molard, Y.; Roisnel, T.; Cîrcu, V. Polyhedron 2014, 69, 31–39. Ilis, M.; Batalu, D.; Pasuk, I.; Cîrcu, V. J. Mol. Liq. 2017, 233, 45–51. Ilis, M.; Micutz, M.; Cîrcu, V. J. Organomet. Chem. 2017, 836–837, 81–89. Micutz, M.; IPasuk, I.; Ilis¸, M. J. Mol. Liq. 2017, 243, 151–156. Baena, M. J.; Espinet, P.; Folcia, C. L.; Ortega, J.; Etxebarría, J. Inorg. Chem. 2010, 49, 8904–8913. Cîrcu, V.; Simonescu, C. M. Cryst. Res. Technol. 2010, 45, 512–516. Cîrcu, V.; Dumitrascu, F. Mol. Cryst. Liq. Cryst. 2011, 534, 41–49. Tritto, E.; Chico, R.; Sanz-Enguita, G.; Folcia, C. L.; Ortega, J.; Coco, S.; Espinet, P. Inorg. Chem. 2014, 53, 3449–3455. Pucci, D.; Barberio, G.; Bellusci, A.; Crispini, A.; Ghedini, M. J. Organomet. Chem. 2006, 691, 1138–1142. Hegmann, T.; Kain, J.; Diele, S.; Schubert, B.; Bögel, H.; Tschierske, C. J. Mater. Chem. 2003, 13, 991–1003. Damm, C.; Israel, G.; Hegmann, T.; Tschierske, C. J. Mater. Chem. 2006, 16, 1808–1816. Venkatesan, K.; Kouwer, P. H. J.; Yagi, S.; Müller, P.; Swager, T. M. J. Mater. Chem. 2008, 18, 400–407. Wang, Y.; Liu, Y.; Luo, J.; Qi, H.; Li, X.; Nin, M.; Liu, M.; Shi, D.; Zhu, W.; Cao, Y. Dalton Trans. 2011, 40, 5046–5051. Yang, X.; Wu, X.; Zhou, D.; Yu, J.; Xie, G.; Bruce, D. W.; Wang, Y. Dalton Trans. 2018, 47, 13368–13377. Zou, G.; Luo, K.; Zhao, L.; Ni, H.; Wang, H.; Li, Q. Liq. Cryst. 2018, 45, 593–606. Yang, B.; Ni, H.; Wang, H.; Hu, Y.; Luo, K.; Yu, W. J. Phys. Chem. C 2020, 124, 23879–23887. Qian, G.; Yang, X.; Wang, X.; Herod, J. D.; Bruce, D. W.; Wang, S.; Zhu, W.; Duan, P.; Wang, Y. Adv. Optical Mater. 2020, 2000775. Santoro, A.; Whitwood, A. C.; Williams, J. A. G.; Kozhevnikov, V. N.; Bruce, D. W. Chem. Mater. 2009, 21, 3871–3882. Spencer, M.; Santoro, A.; Freeman, G. R.; Díez, A.; Murray, P. R.; Torroba, J.; Whitwood, A. C.; Yellowlees, J. Y.; Williams, J. A. G.; Bruce, D. W. Dalton Trans. 2012, 41, 14244–14256. Díez, A.; Cowling, S. J.; Bruce, D. W. Chem. Commun. 2012, 48, 10298–10300. Wang, Y.; Fan, J.; Li, T.; Wang, O.; Shi, J.; Qu, Z.; Tan, H.; Liu, Y.; Zhu, W. RSC Adv. 2016, 6, 45864–45872. Wang, Y.; JFan, J.; Shi, J.; Qi, H.; Baranoff, E.; Xie, G.; Li, Q.; Tan, H.; Liu, Y.; Zhu, W. Dyes Pigm. 2016, 133, 238–247. Krikorian, M.; Liu, S.; Swager, T. M. J. Am. Chem. Soc. 2014, 136, 2952–2955. Geng, H.; Luo, K.; Zou, G.; Zhao, L.; Wang, H.; Li, Q.; Ni, H. Dyes Pigm. 2018, 149, 82–91. Geng, H.; Luo, K.; Cheng, H.; Zhang, S.; Ni, H.; Wang, H.; Yu, W.; Li, Q. RSC Adv. 2017, 7, 11389–11393. Pucci, D.; Aiello, I.; Aprea, A.; Bellusci, A.; Crispini, A.; Ghedini, M. Chem. Commun. 2009, 1550–1552. Ghedini, M.; Pucci, D.; Crispini, A.; Bellusci, A.; La Deda, M.; Aiello, I.; Pugliese, T. Inorg. Chem. Commun. 2007, 10, 243–246. Ionescu, A.; Godbert, N.; Crispini, A.; Termine, R.; Golemme, A.; Ghedini, M. J. Mater. Chem. 2012, 22, 23617–23626. Ionescu, A.; Godbert, N.; Aiello, I.; Crispini, A.; Ghedini, M. Mol. Cryst. Liq. Cryst. 2012, 558, 84–92. Kozhevnikov, V. N.; Donnio, B.; Bruce, D. W. Angew. Chem. Int. Ed. 2008, 47, 6286–6289. Kozhevnikov, V. N.; Donnio, B.; Heinrich, B.; Bruce, D. W. Chem. Commun. 2014, 50, 14191–14193. Kozhevnikov, V. N.; Donnio, B.; Heinrich, B.; Williams, J. A. G.; Bruce, D. W. J. Mater. Chem. C 2015, 3, 10177–10187. Zhao, L.; Yang, B.; Zeng, L.; Luo, K.; Wang, H.; Ni, H.; Yang, C.; Li, Q. Dyes Pigm. 2019, 164, 398–406. Ghedini, M.; Pucci, D.; Barberio, G. Organometallics 1999, 18, 2116–2124. Ghedini, M.; Pucci, D.; Barberio, G. Liq. Cryst. 2000, 27, 1277–1283. Parker, R. R.; Sarju, J. S.; Whitwood, A. C.; Williams, J. A. G.; Lynam, J. M.; Bruce, D. W. Chem. Eur. J. 2018, 24, 19010–19023. Espinet, P. Gold Bull. 1999, 32, 127–134. Bardají, M. Liquid Crystals. In Modern Supramolecular Gold Chemistry: Gold–Metal Interactions and Applications; Laguna, A., Ed.; Wiley-VCH: Weinheim, Germany, 2008; pp 403–428. Chapter 7. Coco, S.; Espinet, P. Liquid Crystals Based on Gold Compounds. In Gold Chemistry; Mohr, F., Ed.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2009; pp 357–396. Chapter 8. Bardají, M. Inorganics 2014, 2, 433–454. Pucci, D. Liq. Cryst. 2011, 38, 1451–1465. Fujisawa, K.; Kawakami, N.; Onishi, Y.; Izumi, Y.; Tamai, S.; Sugimoto, N.; Tsutsumi, O. J. Mater. Chem. C 2013, 1, 5359–5366. Sugimoto, N.; Tamai, S.; Fujisawa, K.; Tsutsumi, O. Mol. Cryst. Liq. Cryst. 2014, 601, 97–106. Younis, O.; Rokusha, Y.; Sugimoto, N.; Fujisawa, K.; Yamada, S.; Tsutsumi, O. Mol. Cryst. Liq. Cryst. 2015, 617, 21–31. Fujisawa, K.; Okuda, Y.; Izumi, Y.; Nagamatsu, A.; Rokusha, Y.; Sadaike, Y.; Tsutsumi, O. J. Mater. Chem. C 2014, 2, 3549–3555. Yamada, S.; Rokusha, Y.; Kawano, R.; Fujisawa, K.; Tsutsumi, O. Faraday Discuss. 2017, 196, 269–283. Fujisawa, K.; Mitsuhashi, F.; Anukul, P.; Taneki, K.; Younis, O.; Tsutsumi, O. Polym. J. 2018, 50, 761–769. Coco, S.; Cordovilla, C.; Espinet, P.; Martín-Alvarez, J. M.; Muñoz, P. Inorg. Chem. 2006, 45, 10180–10187. Arias, J.; Bardají, M.; Espinet, P. Inorg. Chem. 2008, 47, 3559–3567. Donnio, B. Liquid-Crystalline Metallodendrimers. Inorg. Chim. Acta 2014, 409, 53–67. Cordovilla, C.; Coco, S.; Espinet, P.; Donnio, B. J. Am. Chem. Soc. 2010, 132, 1424–1431. Domínguez, C.; Heinrich, B.; Donnio, B.; Coco, S.; Espinet, P. Chem. Eur. J. 2013, 19, 5988–5995. Baena, M. J.; Coco, S.; Espinet, P. Cryst. Growth Des. 2015, 15, 1611–1618. Chico, R.; Domínguez, C.; Donnio, B.; Heinrich, B.; Coco, S.; Espinet, P. Cryst. Growth Des. 2016, 16, 6984–6991. Jiménez, J.; Sanz, J. A.; Serrano, J. L.; Barberá, J.; Oriol, L. Inorg. Chem. 2020, 59, 4842–4857. Arias, J.; Bardají, M.; Espinet, P.; Folcia, C. L.; Ortega, J.; Etxebarría, J. P. Inorg. Chem. 2009, 48, 6205–6210. de Domingo, E.; Barcenilla, M.; Martín-Alvarez, J. M.; Miguel, J. A.; Coco, S. Dyes Pigm. 2020, 176, 108195–108204. Huang, R. T. W.; Wang, W. C.; Yang, R. Y.; Lu, J. T.; Lin, I. J. B. Dalton Trans. 2009, 7121–7131. Hsu, T. H. T.; Naidu, J. J.; Yang, B.-J.; Jang, M.-Y.; Lin, I. J. B. Inorg. Chem. 2012, 51, 98–108.

338

198. 199. 200. 201. 202. 203. 204.

Organometallic Mesogens

Pana, A.; Ilis, M.; Micutz, M.; Dumitrascu, F.; Pasuk, I.; Circu, V. RSC Adv. 2014, 4, 59491–59497. Baron, M.; Bellemin-Laponnaz, S.; Tubaro, C.; Heinrich, B.; Basato, M.; Accorsi, G. J. Organomet. Chem. 2016, 801, 60–67. Yeap, C. W.; Haque, R. A.; Yam, W. S.; Razali, M. R. J. Mol. Liq. 2019, 277, 341–348. Yeap, C. W.; Haque, R. A.; Yam, W. S.; Razali, M. R. Liq. Cryst. 2018, 45, 1210–1222. Syu, H. J. H.; Chiou, J. Y. Z.; Wang, J.-C.; Lin, I. J. B. RSC Adv. 2017, 7, 14611–14617. Kuroda, Y.; Nakamura, S.; Srinivas, K.; Sathyanarayana, A.; Prabusankar, G.; Hisano, K.; Tsutsumi, O. Crystals 2019, 9, 227–239. Xiao, X.-S.; Zou, C.; Guan, X.; Yang, C.; Lu, W.; Che, C.-M. Chem. Commun. 2016, 52, 4983–4986.

14.10

Organometallic Complexes for Optoelectronic Applications

Zhijun Ruan and Zhen Lib,c, aCollege of Chemistry and Chemical Engineering, Huanggang Normal University, Huanggang, China; bHubei Key Lab on Organic and Polymeric Opto-Electronic Materials, Department of Chemistry, Wuhan University, Wuhan, China; cInstitute of Molecular Aggregation Science, Tianjin University, Tianjin, China a

© 2022 Elsevier Ltd. All rights reserved.

14.10.1 14.10.1.1 14.10.1.1.1 14.10.1.1.2 14.10.1.2 14.10.1.2.1 14.10.1.2.2 14.10.1.2.3 14.10.1.3 14.10.1.4 14.10.1.5 14.10.1.6 14.10.1.7 14.10.2 14.10.2.1 14.10.2.2 14.10.2.3 14.10.2.4 14.10.2.5 14.10.2.6 14.10.2.7 14.10.2.8 14.10.2.9 References

Organometallic complexes for mechanoluminescence Introduction Mechanism of ML Characterization of ML Eu(III) complexes Ionic Eu(III) complexes Neutral Eu(III) complexes Eu(III) coordination polymers Sm(III), Tb(III) and Dy(III) complexes Mn(II) complexes Cu(I) complexes Other complexes Conclusions Organometallic complexes as mechanochromic luminogens Introduction Ir (III) complexes Pt(II) complexes Cu(I) complexes Ag(I) complexes Au(I) complexes Zn(II) complexes Other complexes Conclusions

339 339 339 340 340 340 342 343 345 345 347 348 349 349 349 349 354 360 363 365 373 377 379 379

14.10.1 Organometallic complexes for mechanoluminescence 14.10.1.1 Introduction Mechanoluminescence (ML) also known as triboluminescence (TL), refers to the phenomenon of light being produced, when a substance is subjected to external mechanical stimuli (such as grinding, rubbing, crushing or pressing).1–5 ML is the process of converting mechanical energy into light, which has drawn considerable attention owing to its potential applications in pressure sensors, damage detections, bioimaging, lighting, and display devices.6–17 As early as 1605, Bacon first discovered the ML phenomenon when scraping hard sugar cubes with a knife.1 Until now, various kinds of materials have been found with ML properties, such as organic and inorganic crystals, metal complexes, rare-earth-doped ceramics, semiconductors, alkali halides, polymers and so on.5,18–25 According to their composition, ML luminogens can be mainly classified into inorganic materials, pure organic molecules and organometallic complexes. Since Cotton et al. first discovered that organometallic complexes possess ML properties in the 1960s,26,27 ML organometallic complexes have attracted great attention due to their advantages of low cost, low toxicity, wide luminous range and excellent modifiability. In recent years, the research groups of Fontenot, You, Takata and Chi et al. have done a lot of outstanding work in the synthesis, application and mechanism exploration of ML organometallic complexes.

14.10.1.1.1

Mechanism of ML

ML materials exhibit attractive emissions upon mechanical stimuli without the need for other stimulation or irradiation. Although the production of ML looks quite simple, the detailed investigation of its mechanism is extremely complicated, and there is still no recognized theory that can explain the complete physical process of ML. Under the stimulation of external force, ML materials will produce elastic and plastic deformation, and most of the crystals will even break. Accordingly, ML can be divided into elastico-ML, plastico-ML and fracto-ML.5,28–30 As shown in Fig. 1A, fracto-ML was generally derived from charge separation and a strong electric field formed on the newly formed surfaces.31,32 The strong electric field can excite surrounding gas via electron bombardment or directly excite the luminescent center of the materials to emit light. For the materials without a luminescent center (no photoluminescence), the ML spectrum at this time is the gas-discharge emission in the atmosphere.33,34 If the crystal has a luminescent center, the ML spectrum may be the

Comprehensive Organometallic Chemistry IV

https://doi.org/10.1016/B978-0-12-820206-7.00171-2

339

340

Organometallic Complexes for Optoelectronic Applications

Fig. 1 Proposed ML mechanisms. Charging of mechanoluminescence solids resulting from fracture (A), piezoelectricity (B), and triboelectricity (C). Reproduced from Chen, B.; Zhang, X.; Wang, F. Expanding the Toolbox of Inorganic Mechanoluminescence Materials. Acc. Mater. Res. 2021, 2, 364−373, with permission from the American Chemical Society.

intrinsic emission of the crystal, or the superposition of the above two processes. For non-centrosymmetric crystals, due to their piezoelectric properties, electric field can usually be generated in the process of crystal fracture. For some centrosymmetric crystals, the electric field may be caused via some new created non-central structures and charged defects. For alkali metal halide crystals, after being irradiated by X-rays and g-rays, the plastico-ML is typically derived from the interaction between the moving dislocation segments and filled electron traps (such as F-center), resulting in charges recombination emissions with a broad peak width.35 Elastico-ML is a combination of piezoelectricity and light emission in piezoelectric compounds, in which strain-induced internal electric field triggers electronic transitions between different energy levels in the materials (Fig. 1B). Elastico-ML generally appears in materials with a non-centrosymmetric structure, which is mainly realized by piezoelectricity-induced electroluminescence and piezoelectricity induced carrier de-trapping. Some centrosymmetric materials were also reported to be elastico-ML active, which was attributed to the destruction of the local symmetry caused by impurity doping.36,37 ML materials need to be in direct contact with other types of substances (glass, metal, etc.) in testing and applications. In this process, contact electrification is likely formed even without mechanical deformation (Fig. 1C).5 The triboelectrification can also promote ML through mechanisms such as inducing interfacial discharge and inducing electric fields around the material. In particular, some research indicates that the triboelectric potential might be strong enough to induce electroluminescence in some inorganic sulfide crystals.38,39 In general, organic/organometallic ML materials usually have non-centrosymmetric structures and piezoelectric properties, although non-centrosymmetric structure is not a necessary condition for ML. Mechanically induced luminescence of them mainly belongs to the fracto-ML, which is typically affected by molecular packing and intramolecular/intermolecular interactions.30,40,41 When the material is subjected to mechanical stimulation, the material internally fractures, forming positive and negative charges on the newly formed surfaces, and then the charges recombine to form excitons to trigger emission processes.

14.10.1.1.2

Characterization of ML

The drop tower style for the ML test is shown in Fig. 2A. During the process of measurement, the sample powder is first placed under the glass plate, and the steel ball with certain mass falls from a certain height and collides with the sample. A photodiode detector is placed to detect the ML signal, and a linear amplifier is used to gain the ML signal.42–44 The collected signal is simulated by the computer to obtain the ML curve. Sakai et al. developed a new device based on an atomic force microscope (AFM) that can measure the ML spectrum generated by a single particle (Fig. 2B).45 In addition, thin film style is to stretch the ML film on a sample machine, and the testing machine records the load and displacement data automatically. At the same time, a high-speed camera is used to record the ML behavior (Fig. 2C).46 As shown in Fig. 2D, to place the sample in an open container, then, the ML spectrum can be recorded on the computer by scraping or grinding the sample with a spectrometer probe protected by a quartz tube in a dark environment. By adding a square piezoelectric sensor to the middle of the platform, the ML device could be accurately overlapped with the piezoelectric sensor. Through piezoelectric sensor and charge-coupled device, the mechanical force intensity and ML spectral information can be obtained simultaneously, and the relationship between mechanical force and luminous intensity can be established.

14.10.1.2 Eu(III) complexes Rare earth elements have a special electronic structure (underfilled 4f, 5d electronic configuration) and a wealth of energy levels, which can produce abundant luminescence. In addition, the luminescence of rare earth complexes has excellent performance such as narrow peak width, high color purity and stable physical and chemical properties.47–49 Therefore, rare earth complexes are extremely useful emitters for optoelectronic materials. Among them, Eu(III) complex (ionic complexes, neutral complexes and polymers) are widely used in the ML field.

14.10.1.2.1

Ionic Eu(III) complexes

In 1966, Hurt’s group firstly reported the ionic Eu(III) complexes with ML properties based on the framework structure of dibenzoylmethane.50 By comparing different ionic ligands, it was found that the ionic complex 1a (Fig. 3) formed with triethylamine ions had the best ML effect (ML color was orange). Subsequently, Sweeting’s group explored the ML properties of crystals

Organometallic Complexes for Optoelectronic Applications

341

Fig. 2 Schematic diagram of the equipments for ML characterization. (A) Specially designed drop tower used to measure the ML. Reproduced from Fontenot, R. S.; Hollerman, W. A.; Aggarwal, M. D.; Bhat, K. N.; Goedeke, S. M. A Versatile Low-Cost Laboratory Apparatus for Testing Triboluminescent Materials. Measurement 2012, 45, 431–436, with permission from the Elsevier. (B) Single particle style based on AFM. Reproduced from Sakai, K.; Koga, T.; Imai, Y.; Maehara, S.; Xu, C. N. Observation of Mechanically Induced Luminescence From Microparticles. Phys. Chem. Chem. Phys. 2006, 8, 2819–2822, with permission from the Royal Society of Chemistry. (C) Thin film style. Reproduced from Li, C.; Xu, C. N.; Imai, Y.; Bu, N. Real-Time Visualisation of the Portevin–Le Chatelier Effect With Mechanoluminescent-Sensing Film. Strain 2011, 47, 483–488, with permission from the John Wiley & Sons. (D) Experimental setup for testing the quantitative relationship between pressure force and ML intensity. Reproduced from Wang, C.; Yu, Y.; Yuan, Y.; Ren, C.; Liao, Q.; Wang, J.; Chai, Z.; Li, Q.; Li, Z. Heartbeat-Sensing Mechanoluminescent Device Based on a Quantitative Relationship Between Pressure and Emissive Intensity. Matter 2020, 2, 181−193, with permission from the Elsevier.

Fig. 3 Chemical structures of ionic Eu(III) complexes.

342

Organometallic Complexes for Optoelectronic Applications

prepared via different solvent based on complex 1a.51 It was found that only the crystal grown in methanol has strong ML characteristic, whereas the co-crystal formed by dichloromethane and complex 1a had no ML activity, regardless of their centrosymmetric structures. Interestingly, weak ML activity could be obtained by partially converting the co-crystal to 1a via evaporating dichloromethane. They speculated that the disorder in crystal 1a provided a means of separating charge during fracture. Complex 1a had attracted much attention due to its high ML quantum yield and bright ML effect under daylight. Fontenot’s group had done a lot of work to study the influence of the doping of organic molecules, rare earth elements and polymers on the ML properties of complex 1a. In 2012, they studied the effect of doping piperine, dimethyl methylphosphonate (DMMP) and triethylphosphine sulfide (TEPS) on the ML properties of complex 1a.52 It was found that doping of piperine could partially improve the ML performance, while doping of DMMP could make the ML effect very sensitive, and bright red light could be generated with a small force. In 2015, when dibutyl phosphate (DBP) was added, crystal of complex 1a changed from rod shape to flat shape and joined together, and its ML performance could be improved several times.53 In 2013, Fontenot’s group doped Dy into complex 1a, and obtained bright red emission. When 1 mol% Dy was doped, the ML intensity was 80% higher than that of the un-doped system.54 In the same year, they found that adding 4 mol% of uranium to 1a could significantly increase the ML performance.55 In addition, this research group also studied the influence of poly(methyl methacrylate) (PMMA) doping on the ML properties of complex 1a. It was found that the best enhancement could be obtained when 3.4% PMMA (volume ratio) was added. And the ML will rapidly weaken accompanied by further increasing the content of PMMA. They believed that this was mainly because the formed composite material becomes difficult to break with high PMMA concentration, thereby reducing the ML effect.56 Subsequently, You et al. studied the ML properties of complexes 1b and 1c containing dibenzoylmethane and different aromatic ion ligands (imidazole ions and pyrrole ions).57,58 Crystals of complex 1b were centrosymmetric, and the ML spectrum was similar to that of its photoluminescence (PL). Under mechanical stimulation, the charge could be separated due to the weak bonded interactions between ions and molecules in its lattice, which might be responsible for the ML activity. Complex 1c containing pyrrole ions belongs to the non-centrosymmetric space group and had piezoelectric properties, resulting in a red ML emission. In 2017, non-centrosymmetric complex 1d based on dibenzoylmethane unit was synthesized by Law’s group, which showed bright red ML under the daylight.59 In 1989, Rheingold’s group discovered that centrosymmetric complex 2 with benzoylacetone skeleton had ML effect.60 Just like the situation of complex 1b, weakly bonded interactions between ions and molecules in the lattices in complex 2 with good deformability might be responsible for its ML activity. You’s group also explored the ML properties of a series of ionic Eu(III) complexes 3a–3e with 2-thenoyltrifluoroacetone (TTA) as the skeleton.61,62 Among them, complex 3a had the brightest ML effect under the daylight. Through single crystal analysis, they considered that it was the irregular arrangement of sulfur and fluorine that provided the structural basis for the generation of charge separation on the surface of the cracks. Subsequently, Law’s group synthesized complexes 3f and 3g, which also showed excellent ML performance.59

14.10.1.2.2

Neutral Eu(III) complexes

The neutral Eu(III) complexs with ML effect were mainly composed of the b-diketone ligands and the neutral aryl ligands (always containing N or P]O). And effective energy transfer can occur between b-diketone and lanthanide ions. In 1997, Takada’s group synthesized complexes 4a–4d by combining Eu(TTA)3 core and N-containing aryl ligands, and fully researched their ML properties (Fig. 4).63,64 It was found that the ML spectra of complexes 4a and 4b were almost the same as those of their PL spectra, indicating

Fig. 4 Chemical structures of neutral Eu(III) complexes.

Organometallic Complexes for Optoelectronic Applications

343

that ML was derived from the luminescence of Eu(III) ions. Subsequently, they cut complex 4a in different crystal axis directions, and found that the ML phenomenon could only appear on fracture surfaces that had piezoelectric charge.65 Complexes 4c and 4d do not have ML activity themselves, however, when these complexes were doped with 90 wt% polycarbonate, both could exhibit ML activity. Further research speculated that mechanical induced triboelectricity at the interface was the cause of the ML effect.66 In addition, You’s group had conducted a lot of research on ML materials based on neutral Eu(III) complexes. In 1999, You’s group synthesized complexes 5, 6 and 7, and studied their ML phenomenon.67 Complex 5 showed bright red ML in the daylight, and complexes 6–7 exhibited red ML in dark. After recrystallization, the ML intensity of the latter increased, and red ML could be emitted under daylight. The crystal analysis showed that the irregular arrangement of thiophene rings and trifluoromethyl groups in complex 6, and the local asymmetry caused by water in complex 7 could promote charge separation, resulting in ML. In 2007, this group obtained complexes 8a and 8b with brilliant red ML during the daylight by introducing chiral molecular ligands.68 The utilization of chiral ligand makes the complex crystals highly asymmetry, which was more conducive to the generation of piezoelectric effect. Subsequently, they synthesized complex 9 with a chiral ligand based on Eu(TTA)3.69 This complex had a satisfactory dipole moment and piezoelectric effect, showing strong ML performance. Moreover, if the thiophene in the TTA ligand was replaced with benzene, the dipole moment and the ML activity will be canceled. In 2021, Li et al. synthesized two new Eu(III) complexes 10a and 10b by introducing chiral ligands. The introduction of chiral ligands made the complexes had distorted non-centrosymmetric structures, resulting in good ML effect.70 In 2019, Law et al. synthesized complexes 11 and 12 with a centrosymmetric structure by introducing chlorine atoms at different positions of dibenzoylmethane.71 Single crystal analysis demonstrated that strong interactions (p⋯ p, C–H⋯ p and C–H⋯ Cl) and close packing in the crystal were the possible causes of the ML effect.72 Recently, Bryleva et al. synthesized complexes 13a–13c with ML effect. Although all of the above complexes were centrosymmetric, the introduction of many F atoms caused strong intermolecular interactions in the crystal, which was the dominant factor for their ML activity. In 2014, Edelmann et al. synthesized heterometallic complex 14, which had a brilliant red ML.73 They found that the ML of complex 14 originated from the radiation transition of Eu3+ ion through 5D0 !7F2. Recently, Reddy et al. synthesized complex 15 with bright red ML under daylight. After doping with polymer PMMA, its ML intensity would decrease slightly (Fig. 5). This might be due to the decrease in the short range order observed in hybrid polymer film compared with the solid state.74

14.10.1.2.3

Eu(III) coordination polymers

Polymer materials have attracted great attention from researchers due to their rich structure, excellent processability and a wide range of applications. Since 2011, Hasegawa’s group synthesized a series of Eu(III) coordination polymers and studied their ML properties and mechanism in detail. They first synthesized polymer 16 with non-centrosymmetric structural networks, which exhibited bright red ML in daylight at room temperature (Fig. 6).75 It was found that non-centrosymmetric Cc space group and lowvibrational-frequency structure (suppression of radiationless transition through vibrational relaxation) could achieve effective ML. In 2017, they synthesized polymer 17 based on a rigid triangular spacer ligands (tris(4-diphenylphosphorylphenyl)benzene), and its ML lifetime and efficiency were 0.57 ms and 49%, respectively.76 In the same year, this group successfully prepared a series of ML polymers 18a–18c by introducing polar bridging ligands such as furan and thiophene unit.77 The red ML of 18a and 18b were clearly visible during the daylight, whereas 18c could only be observed in the darkness. Through careful analysis of the crystal structure, they found that the dipole moment of the furan-based bridging ligand was larger than that of the thienyl bridging ligand. The large dipole moment helps to form a face-to-face alternating intermolecular interaction of CF3 in the polymer, and the large stack area of the arrangement also plays an important role in ML activity (Fig. 7A and B). In addition, based on furan ligands, Tb(III) was added to form a Tb/Eu-mixed coordination polymers.

Fig. 5 Chemical structures of Eu(III) complexes 14 and 15. ML spectra of complex 15 and 9% complex 15 doped PMMA film.

344

Organometallic Complexes for Optoelectronic Applications

Fig. 6 Chemical structures of Eu(III)polymers.

Fig. 7 (A) Schematic of Eu(III) coordination polymers with face-to-face and alternate intermolecular packing structures. (B) Images of ML and corresponding space filling models on the intermolecular CF3 arrangement in 18a, 18b and 18c. (C) Normalized ML and PL spectra and images of n(Tb)/n(Eu) ¼ 1 (top) and n(Tb)/ n(Eu) ¼ 10 (bottom). Reproduced from Hirai, Y.; Nakanishi, T.; Kitagawa, Y.; Fushimi, K.; Seki, T.; Ito, H.; Hasegawa, Y. Triboluminescence of Lanthanide Coordination Polymers With Face-to-Face Arranged Substituents. Angew. Chem. Int. Ed. 2017, 56, 7171–7175, with permission from John Wiley & Sons.

Interestingly, unlike previous reports, the observed ML colors of the mixed coordination polymers were obviously different from those of the PL spectra. When n(Tb)/n(Eu) ¼ 1, it exhibited yellow ML and reddish-orange PL, while n(Tb)/n(Eu) ¼ 10, it exhibited green ML and greenish-yellow PL (Fig. 7C). By comparing the PL and ML spectra in Fig. 7C, as to Ln(III) coordination compounds, they deduced that the production of ML included both the ligand-excitation and direct Ln(III)-excitation, and was mainly caused by the direct excitation of rare earth ions. Subsequently, based on the furan ligand, by changing the benzene ring to toluene (18d) and cyclohexane (18e), the effect of the side chain groups on polymer arrangement and ML performance was studied in detail.78 It was found that the rigidity of assembly steric structures was controlled by intermolecular interactions through the side groups in bridging ligands. The ML activity was demonstrated to be inversely proportional to the mechanical stability, and the intermolecular structure was the dominant factor for ML activity.

Organometallic Complexes for Optoelectronic Applications

345

14.10.1.3 Sm(III), Tb(III) and Dy(III) complexes Compared with Eu(III) complexes, the reports of rare earth Sm(III), Tb(III) and Dy(III) complexes in ML materials are relatively scarce. Recently, two Sm(III) complexes 19a and 19b with circularly polarized luminescence (CPL) properties were synthesized by introducing chiral ligands.70 Among them, 19a showed bright orange-red ML in dark and daylight, due to its non-centrosymmetric structures and the optimal energy level match between the organic ligands and the Sm(III) ions. In 2013, Reddy et al. synthesized two complexes 20a and 20b via Tb(III) and Dy(III) with non-centrosymmetric structures and piezoelectric properties, and bright green ML could be observed during the daylight.79 The ML spectra of these two compounds were almost the same as their PL spectra, indicating that ML and PL have the same emission center, that is, the rare earth ions. Lately, Bryleva et al. synthesized three Tb(III) complexes 21a–21c with green ML effect in dark.72 It was found that the generation of ML depends on the proportion of the strong intermolecular interactions in the crystal and the mechanical stability of the compounds, although all of the above compounds were centrosymmetric. In 2015, the borate ligand of hydrotris[3-(20 -pyridyl)-pyrazol-l-yl]borate was used to yield complexes 22–24 with different lanthanide ions (Ln). Among them, only complexes 22 (Ln ¼ Eu) and 24 (Ln ¼ Ty) had bright red and green ML, respectively (Fig. 8A).80 It was considered that mechanical fracturing led to an asymmetric charge distribution between the two sides of a fracture, and this in turn could generate excited states leading to emission. When complex 23 (Ln ¼ Gd) was doped with 1 mol% Eu/Tb (prepared by cocrystallization), it could also exhibit visible ML, indicating that mechanical energy was transformed and transferred to Eu(III) or Tb(III) center. In 2019, based on fluorinated b-diketonate, dinuclear lanthanide-lithium complexes 25–27 with bright ML effect were synthesized by Bazhin et al. (Fig. 8B).81 They found that intermolecular H-bonds between the layers in crystals prevent the mechanical destruction, therefore, suppress ML. It was proposed that weak intermolecular interactions within the layered crystal architectures were preferred for ML compared with hydrogen bonding in a wavy chain network.

14.10.1.4 Mn(II) complexes As early as 1961, the Cotton group first reported the transition metal Mn(II) complexes 28a–28d, and studied their ML properties in detail.26 They found that complexes 28b and 28c could exhibit light green and light yellow ML at room temperature. However,

Fig. 8 Chemical structures of Ln(III) complexes. (A) Structures and ML images of complexes 22–24. (B) Structures and ML images of complexes 25–27.

346

Organometallic Complexes for Optoelectronic Applications

complexes 28a and 28d could only show ML at 80 K. And among these complexes, only complex 28a had no PL property at both room temperature and 80 K. Subsequently, they synthesized complexes 29 and 30a with yellow-green ML.27 They speculated that the high local potential discharge at the fracture surface excited the Mn(II) ions, resulting in the ML. In 1982, Chandra group studied the ML and PL performance of complex 28b in detail.82 They found that the ML spectrum of 28b was composed of both the emission of the molecule itself and the emission of nitrogen, and the spectrum of the molecular-emission ML were similar to its PL spectra. Complex 29 had a strong and robust ML effect, in 1986, the same group was further utilized to investigate the effect of external stimuli on the performance of its ML.83 And the ML intensity increased with the impact velocity and decreased with the temperature of the crystal. In 2014, Balsamy et al. synthesized complexes 30b and 30c with bright and continuous green ML in daylight.84 In comparison with other classic ML materials, their ML intensity was slightly weaker than that of the ZnS/Mn, but stronger than that of the salicylsalicylic acid under identical conditions. More interestingly, ML of 30b and 30c was switched ON and OFF reversibly by vapors of aprotic and protic solvents, respectively. In particular, these two complexes would lose their ML properties when exposed to protic solvents (methanol, ethanol, isoamyl alcohol and water), and the ML activity could be restored when further exposed to aprotic solvents (acetone and glycol). Detailed analyses indicated that solids of the 30b and 30c undergo phase transitions depending on the environment, which regulated the ML activity of these compounds. The authors believed that this solventinduced phase transformations may occur through changes in dihedral angles (between central phosphorus atom and aryl unit) as well as weak C–H ⋯ p and p ⋯ p intermolecular interactions between the aryl units. In 2015, Chen et al. synthesized a series of stable centrosymmetric complexes 31a–31c with strong green ML (Fig. 9A).85 Moreover, the similar ML and PL spectra and emission colors of these compounds indicated that the PL and the ML emissions of these three complexes were originated from the same excited state.

Fig. 9 Chemical structures of Mn(III) complexes. (A) Structures and ML images of complexes 31a–31c. Reproduced from Chen, J.; Zhang, Q.; Zheng, F.; Liu, Z.; Wang, S.; Wu, A.; Guo, G. Intense Photo- and Tribo-Luminescence of Three Tetrahedral Manganese (II) Dihalides With Chelating Bidentate Phosphine Oxide Ligand. Dalton Trans. 2015, 44, 3289–3294, with permission from the Royal Society of Chemistry. (B) Structure and ML images of polymer 33. Reproduced from Artem’ev, A. V.; Davydova, M. P.; Berezin, A. S.; Sukhikh, T. S.; Samsonenko, D. G. Photo- and Triboluminescent Robust 1D Polymers Made of Mn (ii) Halides and Meta-Carborane Based Bis (Phosphine Oxide). Inorg. Chem. Front. 2021, 8, 2261–2270, with permission from the Royal Society of Chemistry.

Organometallic Complexes for Optoelectronic Applications

347

Most recently, Artemev et al. synthesized complexes 32a and 32b with yellow and green ML at room temperature.86 Grinding their crystals could produce bright flashes, and these compounds could retain their structural integrity after the ML-associated grinding. Again, the crystals obtained by recrystallization of the ground powders, exhibited the same ML. Considering the similar ML and PL, they supposed that the ligand-sensitized 4T1 ! 6A1 transitions in Mn(II) ions was responsible for both ML and PL processes. Subsequently, the same group synthesized 1D coordination polymer 33 with green ML based on carborane cage ligands (Fig. 9B).87 Crystals of 33 and 33
CH2Cl2 both had bright ML due to their non-centrosymmetric structures, which were more conducive to the generation of electric field. Subsequently, the released energy was transferred to the emitting excited state for the producing of luminescence.

14.10.1.5 Cu(I) complexes In 1991, Koten group synthesized complex 34 with bright green ML under daylight.88 Subsequently, based on the chiral ligand, a chiral complex 35 with ML properties was synthesized.89 Through analysis of the spectra of these compounds, they believed that the luminescence of these complexes originated from the ligand to metal charge transfer. In 2017, Demir et al. synthesized a blue-green ML complex 36, and its ML spectrum was tested using the drop tower system (Fig. 10A).90 The ML emission characterized with respect to tower height and applied force was shown in Fig. 10D and E, and the ML intensity increased with the increasing of tower height and applied force. They surmised that the ML properties of complex 36 were mainly due to the breakage of the Cu–N bond and the charge separation between positively (Cu cations) and negatively (N anions) charged species to generate an electric field. In order to verify its application as a mechanosensing platform, complex 36 was mixed with poly(methyl methacrylate) (PMMA), polystyrene (PS), polyurethane (PU) and poly(vinylidene fluoride) (PVDF) to form a composite material, respectively. It was found that the ML performance doped in polymer PU had the best effect (Fig. 10B and C). The authors speculated that this situation was

Fig. 10 Chemical structures of Cu(II) complexes and ML test of complex 36. (A) Drop tower system. (B) ML emission of the composites prepared by PMMA, PS, PVDF, and PU as a function of number of drops. (C) ML response of PU composite. (D) ML spectrum of 25 mg crystalline particles as a function of height. (E) ML spectrum with respect to applied force from 0.98 N to 4.98 N. Reproduced from _lncel, A.; Varlikli, C.; McMillen, C. D.; Demir, M. M. Triboluminescent Electrospun Mats With Blue-Green Emission Under Mechanical Force. J. Phys. Chem. C 2017, 121, 11709–11716, with permission from the American Chemical Society.

348

Organometallic Complexes for Optoelectronic Applications

Fig. 11 Chemical structures of Cu(II) complexes 37a–37c. ML images in crystal of 37a (A), 37b (B) and 37c (C) under air and PMMA film containing 10 wt% of 37a (D), 37b (E), and 37c (F) under Ar. Reproduced from Karimata, A.; Patil, P. H.; Fayzullin, R. R.; Khaskin, E.; Lapointe, S.; Khusnutdinov, J. R. Triboluminescence of a New Family of CuI–NHC Complexes in Crystalline Solid and in Amorphous Polymer Films. Chem. Sci. 2020, 11, 10814–10820, with permission from the Royal Society of Chemistry.

attributed to the chemical composition of PU, which contained phenyl rings and carbonyl groups to provide a medium for van der Waals, p-p, and dipole-dipole interactions between the complex and the polymer matrix. Recently, a series of novel and stable Cu(I) complexes 37a–37c based on N4 pyridinophane ligands were synthesized by Khusnutdinova et al.91 These complexes exhibited bright ML in the crystal state in ambient indoor light under air. Moreover, ML activity was still retained when these complexes were blended into amorphous polymer films even at small concentrations (Fig. 11). These complexes were all ionic compounds, and disordered fragments were present in some of the crystal structures. These factors may be the source of their ML effect.

14.10.1.6 Other complexes Except the metal elements mentioned above, other organometallic complexes with ML activity were rarely reported. In 2016, the Chi group discovered that Pt(II) complex 38 with non-centrosymmetric space group had ML properties, which could produce blue-green ML under mechanical stimulation (Fig. 12).92 They found that the introduction of tert-butyl and isopropyl groups in the complex played an important role in building a non-centrosymmetric structure to achieve ML activity. In 2019, Tukhbatullin et al. discovered that Ru(II) complex 39 had ML activity, and its ML spectrum was composed of both the N2-discharge emission and the luminescence of crystal itself in air.93

Fig. 12 Chemical structures of complexes 38 and 39 (left), emission spectra of complex 38 (right). (A) PL spectrum of 38, (B) ML spectrum of 38, (C) PL spectrum of grinded 38.

Organometallic Complexes for Optoelectronic Applications

349

14.10.1.7 Conclusions In recent years, organometallic ML materials have achieved great development, but their species are still limited, and more organometallic ML complexes need to be explored. At the same time, the mechanism of how mechanical force is transformed into light in organometallic compounds is still not fully understood. Therefore, the mechanism of organometallic ML materials needs to be further studied through a large number of experimental verifications and theoretical calculations. Actually, inorganic ML materials, especially like ZnS:Mn etc. have been widely used in display, lighting, sensing, imaging and other fields, due to their excellent recyclability and processability.94–100 In contrast, the emission of most of the organometallic ML materials is caused by crystal crack, very difficult to be reused. Therefore, more robust and bright organometallic ML materials need to be developed. It is best to get organometallic ML materials that do not rely on crystals. And the expansion of the applications will further promote the development of organometallic ML materials.

14.10.2 Organometallic complexes as mechanochromic luminogens 14.10.2.1 Introduction Mechanochromic (MC) luminogens that can change their emission colors or intensity upon external mechanical stimulus, such as grinding, rubbing and pressing, have attracted great attention, due to their fundamental importance and promising applications in security papers, luminescence switches, optoelectronic devices, mechanosensors, and optical storage.101–105 Basically, the luminescent properties of MC materials are heavily dependent on molecular packing in an aggregated state, usually in crystal form.106 For crystalline compounds, mechanical force induces a phase transition via crystal-to-crystal transformation or from the crystalline to the amorphous state. After phase transition, molecular packing modes, intermolecular interactions (such as p–p, metal–metal, or hydrogen-bonding interactions) or conformations in crystals are changed. These would affect the HOMO–LUMO energy levels and alter the luminescent properties.107–110 Moreover, MC molecules caused by phase transition under mechanical force without bond breakage are always reversible. Luminogens can be restored to their original state by other external stimuli, such as thermal treatment or fuming by solvents, sometimes even recover spontaneously because of their low phase transition energy.111–113 Actually, organometallic or coordination compounds containing both metal and organic moieties were also found to exhibit reversible and intriguing MC properties.114,115 And the related literatures on MC organometallic materials have significantly increased in recent years.

14.10.2.2 Ir (III) complexes Iridium (III) complexes are extremely useful phosphorescent emitters with relatively short triplet lifetime and high phosphorescence quantum yields.116–118 Cyclometalated Ir(III) complexes are usually consisted by three bidentate ligands, with six chemical bonds taking an octahedral molecular conformation, due to its d6 electron configuration of Ir(III) ion. Most Ir(III) complexes are highly emissive both in solution and the aggregated state at room temperature, and their luminous properties heavily depend on the ligands (electronic nature and the position of the ligands).119,120 Lots of Ir(III) complexes with various molecular structures have been designed and synthesized for use in organic light-emitting diodes (OLED), chemosensing, and bioimaging. Among them, many Ir(III) complexes were found to exhibit excellent and reversible MC properties (Fig. 13). In 2010, Ghedini et al. reported the first MC Ir(III) complex 40a.121 Its crystalline film obtained via slow cooling showed bright green phosphorescence emission at 520 nm. And yellow phosphorescence at 560 nm of its mesophase film (formed under fast cooling) was also observed. Moreover, an amorphous thin film could be fabricated by spin coating solutions of 40a, resulting in a further red shift in the emission wavelength to 580 nm with an orange-red emission. These thin films could turn their emissions to green upon heating, and then going back to orange-red (580 nm) by just rubbing with a velvet cloth or by exerting slight surface friction. More importantly, this color-tuning process by surface heating and stress application was fully reversible. For comparison, complex 40b has no MC property. Since 2012, Su and Liao group has synthesized a series of Ir(III) complexes, and carefully researched their MC and structure-properties relationships. Two new cationic Ir(III) complexes 41 (B1) and 42 (Y1) with reversible MC properties were synthesized by this group in 2012.122 Upon grinding, the emission colors of these complexes showed reversible interconversion, from blue to blue-green for 41 and from green to orange for 42 (Fig. 14). The single-crystal structure demonstrated weak C–H⋯ p intermolecular interactions in the packing diagram. Thus, the ordered crystalline structures with relatively loose packing could be easily collapsed by external pressure. XRD of the original samples exhibited intensive and sharp reflection peaks, indicating their highly ordered crystalline aggregates. After grinding, the ground samples had very weak and broad diffraction signal, indicating that the original ordered structures were transformed into amorphous states. Heating or recrystallization could convert the amorphous ground samples to the crystalline states, and the reversible MC could be realized. In the same year, this group synthesized two other similar Ir(III) complexes 43 and 44 by introducing an additional phenyl group, which also exhibited reversible MC behaviors.123 Upon grinding, an obvious emission red-shift from sky-blue (471 nm) to blue-green (499 nm) for complex 43 was realized. Similarly, complex 44 exhibited remarkable MC properties with emission color changed from yellow to orange (the emission peak red-shifted from 542 to 563 nm) upon grinding. The XRD results illustrated the transformation of the original samples from the crystalline state to the amorphous state. Meanwhile, the DSC curves also exhibited an exothermic recrystallization peak, which indicates a metastable state of the ground sample. In 2012, three cyclometalated Ir(III)

350

Organometallic Complexes for Optoelectronic Applications

O

R

N

Ir

O

40a 40b

F

N

F F

41 43

N

R N Ir

N N

Ph

N

N N

49

PF6

N

N

N

R=C4H9 R=C6H13 R=C8H17 R=C10H21

R

F F

N N

N Ir

F F N R N 50a R=C2H5 Ph 50b R=C4H9 50 50c R=C6H13 50d R=C8H17

N

N

Ar

N

N N

N

F

N N

R R

Ir

2 N

N

53a 53b

F

PF6 N Ir

N

N

R=H R=F

N N Ph

R

54a 54b

R

N

F

55

N

PF6

N N

N

N N N N

N

N

N N

R=H R=F

56

F F

Ir N

R=H R=F

Fig. 13 Chemical structures of Ir(III) complexes.

N

C2H5OOC

R1 O N

R1 R1

HN

R3

N

N COOH

R2 NH

Ir

R1

59

Ir

R2

F

58a 58b

R R

N

PF6

N

56a R1=Br, R2=Br 56b R1=Br, R2=H 56c R1=H, R2=H

N

OH

N

PF6

N

F

N

PF6 N

Ir

R R

N

R1

Ir

R

Ir

N

N

F F

N N N N

N N

R

N

N

R R

COOC2H5

N N

R R

N O

52

2

NH

N HN

R

N Ir O

N

N

N

2 PF6

R

O

51a Ar= 51b Ar=

F

51

Ir

N

F F

R N

NH

2 PF6

N Ir

R O N Ir

N

N

48

N

N

N

Ir

N

X=Cl, PF6, ClO4

PF6

N

47

PF6

N

45a, 45b, 45c

N

R

N N

N

N

R=Ph

47c R=

H2 N

Ir

N N

R=H

47b R=

N

N

R=C2H5

R

F

N

49a 49b 49c 49d 49e

N

N

N

N

Ir

N

42 44

Ir

N

N

47a R=H N

F

PF6

X

PF6

Ir

N N

R

F

N

N

N

F F

N

N N

F F

R=Ph

N

N

N

R=H

PF6

N

46

R

F

R

N

PF6

N N

R=H

N N

R

N Ir

R=OC8H17

Ir

F

F F

R O

F

N

PF6

R R

N

N

F

R

O N

60

R4

60a 60b 60c 60d 60e 60f

N

N N

N

57

57a R=H 57b R=OCH3 57c R=CF3

R1=H; R2=C6H4Cl-4; R3=H; R4=H R1=H; R2=C6H4Cl-4; R3=CF3; R4=H R1=H; R2=C6H4Cl-4; R3=H; R4=F R1=H; R2=C6H4Cl3-2,4,6; R3=CF3; R4=H R1=H; R2=C6H4Cl3-2,4,6; R3=H; R4=F R1=F; R2=C6H4Cl-4; R3=H; R4=H

Organometallic Complexes for Optoelectronic Applications

351

Fig. 14 Emitting color upon irradiation of UV-light of B1 and G1 (A); and Y1 and O1 (B); (C) and (D) the luminescence spectra of complexes B1, G1, Y1 and O1; (E) and (F) repeated cycles of the MC; B1 (G) and Y1 (H) were cast on the filter paper and the letters “Ir” and “Su” were written with a spatula under UV-light at room temperature. B1 and G1 represent complex 41 before and after grinding; Y1 and O1 represent complex 42 before and after grinding. Reproduced from Shan, G.; Li, H.; Cao, H.; Zhu, D.; Li, P.; Su, Z.; Liao, Y. Reversible Piezochromic Behavior of Two New Cationic Iridium (III) Complexes. Chem. Commun. 2012, 48, 2000–2002, with permission from the Royal Society of Chemistry.

352

Organometallic Complexes for Optoelectronic Applications

complexes 45a–45c with different anions were synthesized by Talarico et al.124 Upon mechanical grinding, the color of all the crystalline solids changes from orange to yellow, while the PL emission was partially (45b and 45c) or completely (45a) converted from orange to green. The transformation of the crystalline state into an amorphous state upon grinding induces the disassembling of chromophores within the crystal lattice and the modifications of their local environment. Thus, grinding-triggered color and luminescence changes had been attributed to a crystal-to-amorphous state conversion for all these crystalline solids. Aggregation-induced emission (AIE), is an unusual phenomenon, which has no emission in the dilute solution but dramatically enhanced emission in the solid state. In 2012, Su group reported the first example of an Ir(III) complex 46 that simultaneously displays MC and AIE, by introducing carbazole based dendrons into triazole-pyridine ligands.125 Complex 46 was non-emissive when dissolved in good solvents but showed enhanced emission in the aggregate state. Furthermore, the as-prepared powder obtained through recrystallization exhibited a bright-yellow emission, which changed to orange upon grinding (Fig. 15). Subsequently, in 2013, based on the same design concept, they synthesized three new Ir(III) complexes (47a–47c) with reversible MC and AIE properties.126 Complexes 47a and 47b exhibited an obvious MC in solid states, and the emission color could be smartly switched by grinding and heating. Their MC behavior was also attributed to crystalline-amorphous phase transformation. In 2014, Zhao et al. reported a novel phosphorescent Ir(III) complex 48 with a N–H bond, and found its emission color was sensitive to mechanical force, solvent vapor and electric field, thus resulting in MC, vapochromic and electrochromic properties.127 Complex 48 emitted green phosphorescence at 543 nm in solid state, then, yellow-emitting powder with the emission peak red shift to 585 nm was obtained upon grounding. After exposing the ground sample to organic solvent vapors (such as CH2Cl2) for several seconds, the emission color restored to green, because of the vapor induced recrystallization. The authors speculated that the MC and vapochromism of this complex were probably caused by the breaking and restoring of hydrogen bonds between PF−6 and the N–H in the ligand. In the same year, Sun and Liao group synthesized two series of Ir(III) complexes (49a–49e and 50a–50d) with different alkyl chain length, and carefully studied the relationship between their MC and the molecular structures in detail.128 In the solid state, all these complexes had MC properties, their emission color could be reversibly and quickly switched by grinding-fuming or grinding-heating processes with high contrast. Differences in n-alkyl chain length had negligible effect on their emission spectra in solution. However, the n-alkyl chain lengths can effectively control their MC properties, showing chain length dependent emission behaviors: longer alkyl chains were shown to produce more marked MC with bigger grinding-induced spectral shift (Fig. 16A and B). Powder XRD and DSC results suggested that the reversible transformation between crystalline and amorphous states upon grounding was responsible for the observed MC behavior. Since 2015, a series of binuclear Ir(III) complexes with AIE and reversible MC properties were synthesized by Su and Zhu group. Complexes 51a and 51b were almost non-emissive in solution, but their solid powders exhibited intense phosphorescence at room temperature with FPL values of 0.31 and 0.24, respectively.129 Upon grinding, complexes 51a and 51b with yellow and orange phosphorescence exhibited the red shift of about 20 nm. Moreover, the emission color could be recovered by solvent fuming. Therefore, phosphorescence re-writable data recording device could be fabricated (Fig. 16C). Subsequently, a neutral binuclear Ir(III) complex 52 with a Schiff base bridging ligand was synthesized by the same group, which exhibited AIE, MC and vapochromism activity.130 In the solid state, this complex displayed a reversible color change between weak red and bright orange with high contrast intensity, triggered by high polarity volatile organic compounds (VOCs) or by mechanical grinding. Specifically, complex 52 exhibited ultrahigh stability, with the orange color remaining unchanged in air for several months at room temperature. A simple and efficient monitoring device had been constructed, in which highly polar VOCs act as a switch to “turn on” the device within 10 s (Fig. 16D). The XRD and DSC results proved that this behavior was attributed to solvent vapors readily inducing recrystallizations accompanied by a change of emission color. Recently, these authors reported two new cationic binuclear Ir(III) complexes 53a and 53b containing acylhydrazone ancillary ligands, which exhibited excellent AIE and high-contrast MC properties.131

Fig. 15 (A) Emission spectra of complex 46 before and after grounding. (B) The original powder was cast on the filter paper and the letters “AIPE” were written with a spatula under UV light at room temperature. Reproduced from Shan, G. G.; Li, H. B.; Qin, J. S.; Zhu, D. X.; Liao, Y.; Su, Z. M. Piezochromic Luminescent (PCL) Behavior and Aggregation-Induced Emission (AIE) Property of a New Cationic Iridium (III) Complex. Dalton Trans. 2012, 41, 9590–9593, with permission from the Royal Society of Chemistry.

Organometallic Complexes for Optoelectronic Applications

353

Fig. 16 (A) Images of complexes 49a–49e in different states under a 365 nm UV lamp. (B) Images of complexes 50a–50d in different states under a 365 nm UV lamp. (C) A phosphorescence re-writable data recording device based on MC and vapochromic phosphorescence of complex 51b. (D) Photographic images under UV light of the monitoring of VOCs by complex 52 after exposure to different VOCs and grinding. (E) A phosphorescence re-writable data recording device based on mechanochromic and vapochromic phosphorescence of complex 53. (A) and (B) Reproduced from Han, Y.; Cao, H. T.; Sun, H. Z.; Wu, Y.; Shan, G. G.; Su, Z. M.; Hou, X. G.; Liao, Y. Effect of Alkyl Chain Length on Piezochromic Luminescence of Iridium (III)-Based Phosphors Adopting 2-Phenyl-1 H-Benzoimidazole Type Ligands. J. Mater. Chem. C 2014, 2, 7648–7655, with permission from the Royal Society of Chemistry. (C) Reproduced from Li, G.; Ren, X.; Shan, G.; Che, W.; Zhu, D.; Yan, L. I.; Su, Z.; Bryce, M. R. New AIE-Active Dinuclear Ir (III) Complexes With Reversible Piezochromic Phosphorescence Behaviour. Chem. Commun. 2015, 51, 13036–13039, with permission from the Royal Society of Chemistry. (D) Reproduced from Jiang, Y.; Li, D.; Zhu, G.; Su, Z.; Bryce, M. R. An AIE-Active Phosphorescent Ir (III) Complex With Piezochromic Luminescence (PCL) and Its Application for Monitoring Volatile Organic Compounds (VOCs). J. Mater. Chem. C 2017, 5, 12189–12193, with permission from the Royal Society of Chemistry. (E) Reproduced from Xie, J.; Li, D.; Duan, Y.; Geng, Y.; Yang, T.; Li, G.; Zhu, D.; Su, Z. Cationic Dinuclear Ir (III) Complexes Based on Acylhydrazine Ligands: Reversible Piezochromic Luminescence and AIE Behaviours. Dyes Pigm. 2020, 172, 107855, with permission from the Elsevier.

Phosphorescence data encryption and re-writable devices were successfully fabricated based on their good MC properties (Fig. 16E). The powder XRD suggested that the color change caused by grinding originated from transformation between crystalline and amorphous state. IR spectra further indicated that MC activity of these complexes could be ascribed to the disorder of hydrogen bonds between NdH and C]O groups. In 2018, two new Ir(III) complexes with relatively rigid ligands were synthesized by Wang et al. The introduction of F atoms (electron withdrawing group) made the emission of the corresponding complex blue shift about 50 nm in the solid-state. Complexes 54a and 54b both exhibited reversibly MC properties, and their emission colors were tunable by a grinding–fuming process with high contrast.132 Upon grinding, initial samples of these two complexes exhibited a red shift of about 40 nm, for example, emission color of 54b changed from yellow-green to orange. Also, reproducible phosphorescence data recording and chemical monitoring devices were constructed by using these complexes. In the next year, tricolor luminescence switching of an Ir(III) complex was synthesized by them for the first time (Fig. 17). Complex 55 could form two kinds of crystals, and the transformation between two crystalline states and an amorphous state led to the tricolor switching behavior of this complex among blue, green and yellow emissions.133 Solvent molecules were shown to play a crucial role in the crystallization and luminescence. Most recently, based on tetrazol-pyridine ligand, a new class of Ir(III) complexes (56a–56c) with AIE and MC activity was also synthesized.134 In 2020, three Ir(III) complexes containing different electron donating or electron withdraw groups were synthesized by Liu et al. All complexes exhibited AIE behavior, and among them, complexes 57b and 57c showed revisable MC activity. And the emission color of 57b and 57c in solid state could be changed from yellow to orange and green to yellow.135 Moreover, complex 57c was used as smart luminescent material in a model of data security protection. In 2020, two new cyclemetalated cationic Ir(III) complexes were synthesized by introducing O–H substituents into Schiff base ligands.136 Through the synergistic effect of O–H and F substituents in complex 58b, p-hydrogen bond was successfully utilized to realize the MC property, while complex 58a had no MC activity. A loosely packed structure constructed from intermolecular hydrogen bond interactions (O–H⋯ p and C–H ⋯F) was obtained in complex 58b, and it was susceptible to mechanical stimulation. Powder XRD study proved that the MC behavior originated from the reversible phase transition from crystalline to amorphous state under grinding and solvent recrystallization. A neutral Ir(III) complex 59 with excellent MC properties was synthesized by introducing carboxyl and F substituents into the ancillary and cyclometalating ligands, respectively.137 The emission colors were tunable by the grinding-fuming or grinding-heating process with good reversibility. A combination of powder XRD, DSC, NMR, X-ray photoelectron and IR spectroscopy confirmed that the production of MC involved destruction of the intermolecular p ⋯ p interactions and hydrogen bond. Subsequently, a new class of MC Ir(III) complexes (60a–60f) with pyridyl acyclic carbene ligands were developed by Ko et al.138 These complexes showed predominant MLCT phosphorescence mixed with 3LC

354

Organometallic Complexes for Optoelectronic Applications

Fig. 17 (A) Chemical structure of complex 55. (B) Normalized emission spectra of complex 55 in CH3CN–H2O mixtures. (C) Photographs and schematic representation of a tricolor luminescence switching process. (D) Normalized emission spectra of B-, G- and Y-forms. (E) Repeated cycles of the tricolor luminescence switching process. Reproduced from Yang, T.; Wang, Y.; Liu, X.; Li, G.; Che, W.; Zhu, D.; Su, Z.; Bryce, M. R. Reversible Tricolour Luminescence Switching Based on a Piezochromic Iridium(iii) Complex. Chem. Commun. 2019, 55, 14582–14585, with permission from the Royal Society of Chemistry.

character. Detailed study based on XRD, X-ray crystallography, Raman spectroscopy and DFT calculation revealed that mechanical stimulation on the solid state of these complexes would lead to conformational change (trans-to-cis isomerization) on the bidentate acyclic carbine ligands, resulting in red shift of the phosphorescence (Fig. 18). This discovery provides new design strategies for mechanoresponsive materials.

14.10.2.3 Pt(II) complexes In recent years, the synthesis and application of Pt(II) complexes, especially their applications in phosphorescent OLEDs, bioimaging and photocatalysis, have received great attention.139–141 Compared with d6 Ir(III) complexes that have an octahedral coordination geometry, d8 Pt(II) complexes usually adopt a square planar coordination geometry. The square-planar geometry endows Pt(II) complexes with fascinating photophysical properties, which can be adjusted systematically by selecting appropriate ligands. Until now, many Pt(II) complexes with MC properties have been developed (Fig. 19).

Organometallic Complexes for Optoelectronic Applications

355

Fig. 18 Chemical structure of complexes 60a–60f and proposed MC process. Reproduced from Han, J.; Tang, K.-M.; Cheng, S.-C.; Ng, C.-O.; Chun, Y.-K.; Chan, S.-L.; Yiu, S.-M.; Tse, M.-K.; Roy, V. A. L.; Ko, C.-C. Mechanochemical Changes on Cyclometalated Ir(iii) Acyclic Carbene Complexes – Design and Tuning of Luminescent Mechanochromic Transition Metal Complexes. Inorg. Chem. Front. 2020, 7, 786–794, with permission from the Royal Society of Chemistry.

Fig. 19 Chemical structures of Pt(II) complexes.

356

Organometallic Complexes for Optoelectronic Applications

In 2008, Kozhevnikov et al. reported two liquid-crystalline complexes of Pt(II) (61 and 62), showing that the emission of the complexes was responsive to heating and grinding.142 The original spin-casting film of 62 displayed only excimer-like emission (660 nm). However, after heating the film to 110 C (followed by cooling to room temperature), the emission color changed from red to yellow due to simultaneous emissions from monomer and excimer. After rubbing the annealed film, the red emission of the excimer could be recovered. A further heat-cool cycle could restore the monomer emission. Thus, the emission was under mechanical control and the initial state could readily be reset. In 2009, a new Pt(II) complex with an interesting MC mechanism was discovered by Shinozaki et al. When the crystal of 63 was ground on a glass substrate, the yellow luminescence of the crystalline complex changed to orange.143 A broad emission band was observed at around 670 nm for the ground powder, which was not detected for the crystal and was similar to the excimer emission observed in solution. NMR test showed that there were no bond-breaking or isomerization reactions under grinding. Furthermore, the XRD results indicated that the crystalline structure of the ground powder did not change and remained similar to that of the initial crystal before grinding. It is considered that molecules in the amorphous state are sufficiently close to each other to produce exciton and charge-transfer interactions. Thus, the MC activity of 63 in the solid state was attributed to excimer state emission, which was induced by the mechanical grinding. Based on the similar ligands, three new Pt(II) complexes containing amide groups were synthesized by Kanbara et al.144 Complex 64a exhibited a weak luminescence due to the carbamoyl group, whereas complexes 64b and 64c exhibited strong luminescence in the solid state. As shown in Fig. 20A, as to 64c, upon grinding, the green luminescent crystals (recrystallization from DMF) changed to orange. Exposure of the ground powder to methanol vapors induced an emission change from orange to yellow. The orange luminescence could recover from the yellow luminescent crystals (recrystallization from MeOH) by mechanical grinding or by heating at 200 C. In 2011, square-planar Pt(II) complexes with a reversible and reproducible mechanical stimuli-responsive color and luminescence switch was reported by Chen et al. When crystalline 65a and 65b were ground, their bright yellow-green emission immediately converted to red luminescence with the red shift of 121 and 53 nm, respectively (Fig. 20C).145 Compared with 65a, the corresponding response shift in 65b was significantly smaller, because a bulky tert-butyl group induces the planar Pt(II) molecules to stack through a longer Pt-Pt distance with less intermetallic contact. The ground powder (in amorphous phase) could be restored to the crystalline state by fuming or heating, accompanied by the reversion of the red luminescence to a yellow-green emission. The researchers proposed that the drastic grinding-triggered red-shift was likely involved in the formation of a dimer or an

Fig. 20 (A) Two-step changes in the PL of complex 64c by external stimuli under UV light. From Choi, S.; Kuwabara, J.; Nishimura, Y.; Arai, T.; Kanbara, T. Two-Step Changes in Luminescence Color of Pt (II) Complex Bearing an Amide Moiety by Mechano- and Vapochromism. Chem. Lett. 2012, 41, 65–67. (B) Photographic images of 67 in a PMMA film in response to CH2Cl2 vapor or mechanical stimuli under UV light. Reproduced from Zhang, X.; Wang, J.; Ni, J.; Zhang, L.; Chen, Z. Vapochromic and Mechanochromic Phosphorescence Materials Based on a Platinum (II) Complex With 4-Trifluoromethylphenylacetylide. Inorg. Chem. 2012, 51, 5569–5579, with permission from the American Chemical Society. (C) Photographic images of crystalline 65a in response to mechanical grinding under ambient light and UV light, showing the color and luminescence switches from yellow-green to red upon grinding, and from red turning back to yellow-green by addition of acetone: crystalline sample, partially and thoroughly ground sample, addition of a drop of acetone (from left to right). Reproduced from Ni, J.; Zhang, X.; Qiu, N.; Wu, Y.; Zhang, L.; Zhang, J.; Chen, Z. Mechanochromic Luminescence Switch of Platinum (II) Complexes With 5-Trimethylsilylethynyl-2, 20 -Bipyridine. Inorg. Chem. 2011, 50, 9090–9096, with permission from the American Chemical Society.

Organometallic Complexes for Optoelectronic Applications

357

aggregate through Pt-Pt interaction. Subsequently, three Pt(II) complexes (66a–66c) with similar chemical structure had been developed by the same group.146 When the solid species were gently ground, the crystalline state was converted into an amorphous phase, along with the change of their phosphorescence under mechanical force stimuli. The bright yellow-orange luminescence of the crystalline sample was converted to dark red under UV light, exhibiting a 100–160 nm red shift response. In 2012, Chen group further reported a similar Pt(II) complex containing 4-trifluoromethylphenylacetylide, which exhibited remarkable luminescence vapochromic and MC properties as well as a thermo-triggered luminescence change.147 Photographic images of 67 in poly(methyl methacrylate) films showed remarkable luminescence color switches under UV light in response to vapor or mechanical stimuli (Fig. 20B). Both X-ray crystallographic and density functional theory (DFT) studies suggested that the variation in the intermolecular Pt-Pt interaction is the key factor in inducing an intriguing luminescence switches. In 2014, by simply introducing bromine atom into the 2,20 -bipyridine ligand, a square-planar Pt(II) complex was synthesized by Ni et al.148 In solid state, 68 exhibited reversible thermo- and mechanical grinding-triggered color and luminescence changes. When crystalline 68
2(CH2Cl2) or 68
2(CHCl3) were heated or ground, the emission possessed a spectral red shift, along with the change of their phosphorescence from bright yellow-green to red (Fig. 21A). Once fuming amorphous ground sample via organic vapors, it could be reverted to the original crystalline state, along with red luminescence turning back to yellow-green emission (Fig. 21B). The authors suggested that thermochromism and MC originate from the conversion of metal-to-ligand charge transfer (3MLCT) and ligand-to-ligand charge transfer (3LLCT) emission states to 3MMLCT (metal-metal to ligand charge transfer) triplet state. In the same year, Konno et al. synthesized two Pt(II) complexes with terpy as the coordination unit, and ClO−4 (69a) or PF−6 (69b) as the anions.149 These two complexes have the similar absorption and emission in solution, but their emission colors are different in crystals. 69a showed orange emission with lem ¼ 630 nm, whereas 69b exhibited a longer emission wavelength with lem ¼ 670 nm. In crystals, two complexes underwent 1D stacking driven by the p–p interaction between terpy ligands and hydrogen bonds (Fig. 21C). When the orange powder of 69a was gently ground, the sample color and phosphorescent luminescence both changed from orange to red with emission peak red shift from 630 to 667 nm. When ClO4− was replaced by PF6−, however, as to 69b, the MC behavior almost disappeared due to its restricted molecular configuration changes.

Fig. 21 (A) Photographic images of color and luminescence changes of 68
2(CH2Cl2) during the heating process (left to right: 0–60 min). (B) Photographic images of 68
2(CH2Cl2) as unground sample, completely ground sample and ground sample with a drop of CH2Cl2 added. (C) 1D structures of 69a and 69b in crystal, and emission spectra in the solid state: (black lines) the samples before grinding and (red lines) the samples after grinding. (A) and (B) Reproduced from Ni, J.; Wang, Y. G.; Wang, H. H.; Xu, L.; Zhao, Y. Q.; Pan, Y. Z.; Zhang, J. J. Thermo- and Mechanical-Grinding-Triggered Color and Luminescence Switches of the Diimine-Platinum (II) Complex With 4-Bromo-2, 20 -Bipyridine. Dalton Trans. 2014, 43, 352–360, with permission from the Royal Society of Chemistry. (C) Reproduced from Kitani, N.; Kuwamura, N.; Tsukuda, T.; Yoshinari, N.; Konno, T. Counteranion-Dependent Mechanochromism of a Photoluminescent Platinum (ii) Complex With Mixed Terpyridine and Thioglucose. Chem. Commun. 2014, 50, 13529–13532, with permission from the Royal Society of Chemistry.

358

Organometallic Complexes for Optoelectronic Applications

Fig. 22 (A) Crystal packing diagrams of 70a-Y (left) and 70a-R (right) depicting the Pt–Pt distances. (B) Reversible MC of 70a-Y by grinding and adding CH2Cl2 under ambient light and 365 nm UV light. (C) Reversible MC of 70c by grinding and adding CH2Cl2 under ambient light and 365 nm UV light. Reproduced from Zhang, X.; Mei, J.; Lai, J.; Li, C.; You, X. Mechano-Induced Luminescent and Chiroptical Switching in Chiral Cyclometalated Platinum (II) Complexes. J. Mater. Chem. C 2015, 3, 2350–2357, with permission from the Royal Society of Chemistry.

Recently, chiral ligands were introduced into Pt(II) complexes to enrich the MC materials, Zhang et al. have done a lot of excellent work in this area. In 2015, four chiral complexes with reversible and red shift MC were reported. Two polymorphs (Form-Y and Form-R) of the complex 70a were obtained.150 70a-Y is a yellow crystal that emits orange phosphorescence, while 70a-R has a red emission, and this red-shifted emission should be caused by its shorter Pt ⋯ Pt distance in 70a-R (Fig. 22A). After grinding, yellow 70a-Y changed into red and emitted red phosphorescence (Fig. 22B). The yellow 70a-Y was found to undergo crystalto-amorphous transformation upon mechanical grinding, resulting in luminescent and chiroptical switching behaviors. The MC process could be reversed repeatedly by the addition of a few drops of dichloromethane. When the counteranion Cl− was replaced by trifluoromethanesulfonate (OTf−), complexes 70c and 70d were obtained, which showed a more pronounced luminescent switching behavior, suggesting that the switching properties could be tuned by the counteranions (Fig. 22C). In the next year, a couple of enantiomeric chiral cyclometalated Pt(II) complexes 71a and 71b were synthesized by the same group, and their reversible luminescent and chiroptical switching were studied in detail.151 Upon mechanical grinding, the crystalline samples underwent a crystal-to-amorphous and ordered-to-disorder transformation, along with the luminescence was distinctly red-shifted (△lem ¼ 160 nm) compared to that of the crystallites. Moreover, electronic circular dichroism signals and nonlinear optical responses became almost silent. Remarkably, the amorphous ground powder could be easily restored to original crystalline state upon treatment with a few drops of dichloromethane. Subsequently, two alkynyl containing chiral complexes 72a and 72b were prepared by the same author with obvious MC properties.152 Mechanical force would induce a remarkable color change from yellow to orange and a distinct luminescence variance from orange to red. Most recently, based on the same molecular skeleton, Zhang’s group have synthesized a series of chiral Pt(II) complexes by introducing different types and numbers of substituents into the ligands, which exhibited similar luminescence and chiroptical switches in response to mechanical force.153–155 In 2016, six novel phosphorescent Pt(II) complexes with a-aminocarboxylato ligands (73a–73f) were synthesized by Fujihara et al.156 The complexes except 73e in the solid state exhibited reversible luminescent MC, in response to the mechanical grinding and treatment with a few drops of solvent (Fig. 23A). The authors suggested that the MC activity were induced by the change in the energy level of the triplet state, due to the change in the extent of intermolecular interactions during the phase transition from its crystal to amorphous state. In the same year, three planar Pt(II) complexes 74a–74c based on the modified 2-phenylpyridine derivatives as the main ligand and picolinic acid as auxiliary ligand were synthesized ,and their MC behavior were investigated by Luo et al.157 They found that the MC properties of these complexes were relative to the number of flexible alkoxy chains in cyclometalated ligand. Ni et al. reported a new square-planar Pt(II) complex based on 1,10-phenanthroline and 3-chlorophenylacetylene ligands.158 Complex 75 crystallized in triclinic, in which, adjacent molecules along the a axis were stacked

Organometallic Complexes for Optoelectronic Applications

359

Fig. 23 (A) Changes in XRD patterns by mechanical grinding and solvent treatment for 73a; emission color and luminescence spectral changes of 73a by grinding. Reproduced from Ohno, K.; Yamaguchi, A. S.; Nagasawaa, A. A.; Fujihara, T. Mechanochromism in the Luminescence of Novel Cyclometalated Platinum (II) Complexes With a-Aminocarboxylates. Dalton Trans. 2016, 45, 5492–5503, with permission from the Royal Society of Chemistry. (B) Schematic drawings for the MC luminescence behavior of 77b and 77d, and variable pentiptycene-pentiptycene packing modes. Reproduced from Lin, C.-J.; Liu, Y.-H.; Peng, S.-M.; Shinmyozu, T.; Yang, J.-S. Excimer–Monomer Photoluminescence Mechanochromism and Vapochromism of Pentiptycene-Containing Cyclometalated Platinum (II) Complexes. Inorg. Chem. 2017, 56, 4978−4989, with permission from the American Chemical Society. (C) Emission color changes of 78a–78c by grinding and fuming. Reproduced from Lien, C.-Y.; Hsu, Y.-F.; Liu, Y.-H.; Peng, S.-M.; Shinmyozu, T.; Yang, J.-S. Steric Engineering of Cyclometalated Pt (II) Complexes Toward High-Contrast Monomer–Excimer-Based Mechanochromic and Vapochromic Luminescence. Inorg. Chem. 2020, 59, 11584−11594, with permission from the American Chemical Society.

in a columnar structure through p⋯ p interactions, and such neighboring columnar structures were connected with each other by C–H⋯ p hydrogen bonds to form a 2D supramolecular network. This complex exhibited reversible luminescence MC property with the luminescence red-shifting in a range of ca. 146–182 nm. It was suggested that such extremely large red-shifts of luminescence spectra were caused by the formation of aggregate via Pt-Pt interaction during the mechanical grinding. Recently, five new Pt(II) complexes 76a–76e bearing different terminal substituents on 1,3-di(2-pyridyl)benzene ligands were designed and synthesized by Liu et al.159 Upon excitation, the solid states of these complexes exhibited tunable emissions from yellow to red. All complexes, except 76e, showed obvious MC and vapochromic behaviors. Based on in-depth experimental studies, the authors suggested that the MC behavior were induced by the changing in the distance of the neighboring Pt complexes, which caused a switching of the emission state from 3p,p /3MLCT to 3MLCT. In 2017, four novel Pt(II) complexes based on the dipyridylbenzene (N^C^N) ligand (77a–77d) were synthesized by Yang group.160 The bulky H-shaped pentiptycene scaffold was introduced to promote their MC and vapochromic luminescence properties. The MC properties were highly dependent on the complex structure: grinding resulted in a blue shift of 19 nm for the excimer emission band of 77a, while a diminution of the monomer-to-excimer emission intensity for 77b, and a blue shift of 14 nm for the excimer emission along with a small increase of the monomer intensity for 77c, and formation of excimer emission at 662 nm for 77d. There were strong p–p interactions in the initial samples of 77a and 77c, thus a minor interruption of such interactions by external forces that increased the interplanar distance or reduced the interplanar stacking area could account for their MC luminescence behaviors. In contrast, 77b and 77d had weak intermolecular p⋯ p interactions in the original samples due to a relatively large pentiptycene steric shielding effect, and a breakdown of the shielding by external forces toward short contacts of the NCNPt cores was viable (Fig. 23B). Therefore, they suggested that interplay of the monomer versus excimer dual emission and the

360

Organometallic Complexes for Optoelectronic Applications

structural variability of pentiptycene-pentiptycene interactions (e.g., the UU, UV, and VV packing in Fig. 23B) were the key factors for the observed MC. To further improve the luminescence color responses of Pt(II) complexes to external stimuli such as mechanical stress and chemical vapors, the same group conducted a steric engineering of the previous systems by introducing two tert-butyl groups to the tridentate ligand to form complexes 78a–78c.161 The combined steric effects of the tert-butyl groups at one side and the pentiptycene group at the other side of the NCNPt core in 78b were “just right” for generating as-prepared powders with pure monomer (green) emission or pure excimer (red) emission, depending on the rate of precipitation from solutions. The synergistic steric effects were also beneficial to the solid-state luminescence quantum efficiency. As a result of the differences in steric interactions and thus in the relative monomer vs excimer emission intensity, each complex of 78a–78c performed a two-step luminescence MC and vapochromism with different color patterns (Fig. 23C). This work provided an intriguing example of steric engineering of Pt(II) complexes toward highly emissive molecular solids with high-contrast MC and vapochromic luminescence. In 2020, Lalinde et al. reported two series of neutral Pt(II) complexes, and carefully investigated the impact of chemical structural on their photophysical properties in detail.162 The phenylquinoline derivatives 80a and 80b did not show self-assembly tendency. However, the PL characteristics of the 2-phenylpyridine-based complexes 79a–79b were largely determined by intermolecular p ⋯ p aggregation, giving emissions significantly red-shifted with respect to the monomer in glasses (also in PMMA). The high degree of molecular aggregation on these complexes induced simultaneously phosphorescent AIE, and reversible MC color and emission changes, with remarkable red shift and decreased quantum yields. Powder XRD studies suggested a reversible crystalline-toamorphous phase transition as the mechanism of the MC process.

14.10.2.4 Cu(I) complexes Cu complexes have rich structural diversity with their emission spectra covering the entire visible light region from blue to red. In addition, these complexes usually have good photochemical and photophysical properties. Thus, Cu complexes are widely used in the field of luminescence. In recent years, many Cu(I) complexes with MC properties have been developed (Fig. 24). In 2010, a molecular Cu(I) iodide cluster 81a with MC and thermochromic luminescence properties was reported by Perruchas et al.163 This cluster exhibited a significant change in its solid-state emission properties from weak green to intense yellow, upon grinding at room temperature (Fig. 25A). Interestingly, the green emission of crystalline cluster turned bright blue, and changed to purple upon grinding at 77 K under UV light. These changes triggered by temperature and mechanical force were reversible upon exposure to solvent or heating. The crystalline structure of 81a was analyzed in detail. However, no intermolecular interactions directly involved the Cu(I) iodide core, also no typical p ⋯ p stacking interactions were observed. Only some short C–H⋯ H contacts involving the allyl groups of ligands were detected. These results suggested that the mechanical constraints induced by grinding modified these

Fig. 24 Chemical structures of Cu(I) complexes.

Organometallic Complexes for Optoelectronic Applications

361

Fig. 25 (A) Photographs showing luminescence changes of 81a upon grinding and at different temperatures. (I) Cluster before grinding and (II–III) cluster upon grinding under UV lamp at room temperature; cluster (IV) before and (VI) after grinding under UV lamp at 77 K; (V) the corresponding photograph of II under ambient light at room temperature. (B) Photos of the polymorphic crystals of 81b under ambient light and under UV irradiation at 312 nm at room temperature. (C) Photos of the ground (left) and intact (right) crystalline powder of 81b-G under ambient light and under UV irradiation at 312 nm at room temperature. (A) Reproduced from Perruchas, S.; Goff, X. L.; Maron, S.; Maurin, I.; Guillen, F.; Garcia, A.; Gacoin, T.; Boilot, J. Mechanochromic and Thermochromic Luminescence of a Copper Iodide Cluster. J. Am. Chem. Soc. 2010, 132, 10967−10969, with permission from the American Chemical Society. (B) and (C) Reproduced from Benito, Q.; Goff, X. F. L.; Maron, S.; Fargues, A.; Garcia, A.; Martineau, C.; Taulelle, F.; Kahlal, S.; Gacoin, T.; Boilot, J.-P.; Perruchas, S. Polymorphic Copper Iodide Clusters: Insights Into the Mechanochromic Luminescence Properties. J. Am. Chem. Soc. 2014, 136, 11311−11320, with permission from the American Chemical Society.

interligand interactions, which led to a different cluster packing in the crushed compound. These local distortions allowed the relaxation of the Cu(I) iodide core and drastically modify the luminescence properties. Subsequently, based on the copper iodide [Cu4I4] cluster, Perruchas group synthesized a series of Cu(I) complexes to study their mechanism of MC in detail. In 2014, an in-depth study of MC and thermochromic luminescent copper iodide clusters exhibiting structural polymorphism was reported and gave new insights into the origin of the luminescence MC properties.164 Two crystalline polymorphs of 81b exhibited different luminescence properties, with one being green emissive (81b-G) and the other one (81b-Y) being yellow emissive (Fig. 25B). Upon mechanical grinding, only 81b-G exhibited remarkable change of its emission from green to yellow (Fig. 25C). Interestingly, the photophysical properties of the resulting partially amorphous crushed compound were close to those of the yellow polymorph. The structure of the [Cu4I4P4] cluster cores was tested by 31P and 65Cu solid-state NMR, and the results indicated that the grinding process changed the local environments of phosphorus and copper atoms with clearly distinct NMR signals. Combined with detailed analysis of the single crystals of the two polymorphs, the MC phenomenon was thus explained by the disruption of the crystal packing, i.e. shortening of the Cu–Cu bond distances in the [Cu4I4] cluster core. These results definitely established the role of cuprophilic interactions in the MC of copper iodide clusters. This study constituted a step further into the understanding of the mechanism involved in the MC luminescent properties of metal-based compounds. In the following years, this group synthesized a series of new complexes with polymorphism through subtle modification of the ligands, and systematically studied their MC process. By replacing allyl with ethoxy, complex 81c with two different polymorphs (A and B) were synthesized.165 Only polymorph A presented MC properties with significant modifications of its luminescence properties upon grinding, with an emission change from weak green to intense yellow. Just like 81b mentioned above, the photophysical properties of this ground compound presented features similar to the yellow polymorph B. However, when the above substituents were changed to propyl one, the synthesized complex 81d lost its MC activity, due to slight differences in the crystal packing controlled by the nature of the ligands.166 In 2014, Li et al. reported that two novel planar trinuclear Cu(I) complexes (82a and 82b) could change their emission from fluorescence to phosphorescence upon grinding.167 Their crystal structures exhibited similar chair like dimer stacking supported by short Cu⋯ Cu contacts, which could facilitate the formation of photoinduced excimers. All crystals possessed similar dual emission, and the dual emission from the organic fluorophore and excimeric copper cluster phosphor was found to undergo mechanically induced intensity switching between their high-energy (HE) and low-energy (LE) bands. Specifically, the relative intensities of crystalline samples were HE > LE, while the ground solid samples showed LE > HE, resulting in the overall emission color interchanging between bluish violet and red (Fig. 26A). This switching could be reversed by application of solvent to the ground samples, and also by heating. TD-DFT calculations revealed that the emissive singlet ligand localized state (S1) and triplet cluster centered state (T8) lied close in energy, suggesting the feasibility of dual emission and the possibility of reverse intersystem crossing. In 2013, a multifunctional 1D Cu(I) coordination polymer (83) based on 5,6-dimethylbenzimidazolate was solvothermally synthesized by Zhang group.168 Complex 83 exhibited unusual MC, sensing of nitrobenzene and photocatalytic properties for the degradation of organic dyes. Upon grinding, crystal of 83 with bright green phosphorescence (ascribed to MLCT) displayed an obvious red shift of about 160 nm (Fig. 26B). The alteration of Cu ⋯ Cu interactions and even p ⋯ p interactions in the structure may be attributed to the luminescent change. Subsequently, they synthesized a series of 2D coordination polymers, showing excellent

362

Organometallic Complexes for Optoelectronic Applications

Fig. 26 (A) Solid state emission spectra of 82a before and after grinding the sample and subsequent soaking in ethanol. Inset: crystal structure of 82a and the letters “STU” written on the sample of 82a under 254 nm UV light. Reproduced from Xiao, Q.; Zheng, J.; Li, M.; Zhan, S.; Wang, J.; Li, D. Mechanically Triggered Fluorescence/Phosphorescence Switching in the Excimers of Planar Trinuclear Copper (I) Pyrazolate Complexes. Inorg. Chem. 2014, 53, 11604–11615, with permission from the American Chemical Society. (B) Normalized solid-state emission spectra for the unground (green line) sample and ground (red line) sample, photographs of sample 83 show luminescence changes with grinding time at room temperature under ambient light and 365 nm UV lamp. Reproduced from Wen, T.; Zhang, D.; Liu, J.; Lin, R.; Zhang, J. A Multifunctional Helical Cu (I) Coordination Polymer With Mechanochromic, Sensing and Photocatalytic Properties. Chem. Commun. 2013, 49, 5660–5662, with permission from the Royal Society of Chemistry. (C) Solid-state emission colors of various samples of 85 and 86 under ambient light and 365 nm UV lamp. Reproduced from Deshmukh, M. S.; Yadav, A.; Pant, R.; Boomishankar, R. Thermochromic and Mechanochromic Luminescence Umpolung in Isostructural Metal–Organic Frameworks Based on Cu6I6 Clusters. Inorg. Chem. 2015, 54, 1337–1345, with permission from the American Chemical Society.

MC performance.169 In 2014, a redox-active Cu(I) boron imidazole framework with a ladder-chain structure was reported by the same group.170 The emission maximum wavelength for 84 was 512 nm, which displayed a blue shift of about 20 nm upon grinding. Interestingly, this complex could directly reduce and entrap Ag nanoparticles into its structure to form Ag@84 in the presence of rich B–H functional bonds, and the Ag@84 samples exhibited high catalytic activities for the reduction of 4-nitrophenol to 4-aminophenol. In 2015 and 2016, the same group reported other similar imidazolate frameworks with obvious MC activity.171,172 These frameworks could be used to directly introduce Au-Ag or Au-Pd bimetallic nanoparticles into their structures for the reduction of 4-nitrophenol to 4-aminophenol. In 2015, two isostructural metal-organic framework (MOF) materials (85 and 86) with Cu6I6 cluster were synthesized by Boomishankar et al. Excitingly, these two isostructural MOFs showed unusual reversible MC luminescent behavior at 298 K.173 Upon grinding, blue-shifted high energy emission for 85 (from orange to yellowish-orange), and a red-shifted low-energy emission for 86 (from green to orange) were obtained (Fig. 26C). This was primarily due to the variations in their cuprophilic interactions, as 85 displayed shorter Cu⋯ Cu distances (2.745 A˚ ) in comparison with those present in 86 (3.148 A˚ ). As the result, the ground sample of 86 exhibited a prominent red shift in luminescence owing to the reduction of its Cu ⋯ Cu distances to an unknown value closer to the sum of van der Waals radii between two Cu(I) atoms (2.80 A˚ ). However, the blue-shifted emission in 85 was presumably attributed to the rise in its lowest unoccupied molecular orbital energy levels caused by changes in the secondary

Organometallic Complexes for Optoelectronic Applications

363

packing forces. In 2017, tetraphosphine-supported tetradecanuclear Ag12Cu2 acetylide cluster complexes (87a–87c) were prepared and characterized by Chen et al.174 They showed reversible phosphorescent MC with obvious emission color change from yellow to orange-red. Interestingly, upon mechanical grinding, Ag12Cu2 cluster complexes in dilute PMMA also exhibited drastic phosphorescent red shift. The MC of these complexes was ascribed to the contraction within Ag12Cu2 cluster and the enhanced intramolecular Cu-Ag and Ag-Ag contacts. For d10 coinage metal complexes, changes of M-M interactions have been proposed as the main mechanism. In 2018, Steffen et al. synthesized three novel dicopper(I) complexes (88a–88c), and proposed a new mechanism for MC phosphorescence, namely, stimulus-triggered anion-cation-exciplex formation.175 For these complexes, the mechanical induced reversible cation-anion exciplex formation based on Cu-F interactions, led to highly efficient MC phosphorescence, and unusual large emission shift from UV-blue to yellow was realized. The low-energy luminescence was thermo- and vapor-responsive, allowing for the generation of white light as well as for recovering the original UV-blue emission. Recently, five MC luminescent tetranuclear Cu(I) complexes (89a–89e) with reversible MC luminescence have been successfully designed and prepared by Chen group (Fig. 27A).176 These complexes owned a large coplanar multinuclear Cu(I) unit showing weak intramolecular p⋯ p interactions with the planar rings of the coordinated ligands in the molecule. Through carefully analyzing the powder XRD and FT-IR results, the MC behavior could be attributed to the change in the rigidity of the molecular structure, resulting from the disruption and restoration of intramolecular p⋯ p interactions between the pyridyl and phenyl rings triggered by grinding and CH2Cl2 vapor. Subsequently, the same group reported a new series of ionic bimetallic Cu(I) complexes (90a–90c) exhibiting reversible MC luminescence and thermally activated delayed fluorescence.177 Most recently, a new Cu(I) complex 91 was reported by Jin et al., and two polymorphs were isolated.178 The blue luminescent crystal (91-B) had evident MC luminescent properties, whereas the yellow-green luminescent crystal (91-Y) had no MC activity (Fig. 27B and C). Based on photophysical and structural analysis, the pore structure in the blue crystal was considered to be the main reason for the MC properties.

14.10.2.5 Ag(I) complexes In 2010, Tsukuda et al. reported the first MC Ag(I) complex 92 (Fig. 28), and two polymorphs were obtained.179 Interestingly, the blue-emitting crystal obtained from CHCl3-ether owned obvious MC luminescent properties (from blue to green upon grinding),

Fig. 27 (A) Photographic images of 89a–89e: (I) Crystalline samples under ambient light and (II) crystalline, (III) ground, (IV) reverted, and (V) partially ground samples under UV lamp. (B) Photoluminescence spectra of the two crystal forms at room temperature. (C) Photos of various states of 91 during the ground-fumed conversion, the letters “C” and “u” were made up of 91-B and 91-G respectively. (A) Reproduced from Peng, D.; He, L.-H.; Ju, P.; Chen, J.-L.; Ye, H.-Y.; Wang, J.-Y.; Liu, S.-J.; Wen, H.-R. Reversible Mechanochromic Luminescence of Tetranuclear Cuprous Complexes. Inorg. Chem. 2020, 59, 17213–17223, with permission from the American Chemical Society. (B) and (C) Reproduced from Yu, X.; Li, X.; Cai, Z.; Sun, L.; Wang, C.; Rao, H.; Wei, C.; Bian, Z.; Jin, Q.; Liu, Z. Mechanochromic Properties in a Mononuclear Cu (i) Complex Without Cuprophilic Interactions. Chem. Commun. 2021, 57, 5082–5085, with permission from the Royal Society of Chemistry.

364

Organometallic Complexes for Optoelectronic Applications

Fig. 28 Chemical structures of Ag(I) complexes.

whereas the green-emitting crystal obtained from CH3CN had no MC activity. When blue-emitting crystal (lem ¼ 458 nm) was ground in a ceramic mortar, the obtained powder emitted green luminescence (lem ¼ 518 nm). Upon treatment of the ground samples with drops of CHCl3-hexane (1:2) or upon heating at 200 C for 10 min, it reverted to the initial state (Fig. 29A). The XRD results showed that reversible phase conversion occurred from crystalline to amorphous upon grinding and heating. From the single

Fig. 29 (A) Emitting color of 92 before (I) and after (II) grinding under UV-light, and the corresponding luminescence spectra (V); emitting color before heating (III) and after heating at 200 C for 10 min (IV). Reproduced from Tsukuda, T.; Kawase, M.; Dairiki, A.; Matsumoto, K.; Tsubomura, T. Brilliant Reversible Luminescent Mechanochromism of Silver (I) Complexes Containing o-Bis (Diphenylphosphino) Benzene and Phosphinesulfide. Chem. Commun. 2010, 46, 1905–1907, with permission from the Royal Society of Chemistry. (B) Photographs of samples of 94 show luminescence changes with grinding time at room temperature under ambient light and a 365 nm UV lamp. Reproduced from Liu, L.; Wen, T.; Fu, W.; Liu, M.; Chen, S.; Zhang, J. Structure-Dependent Mechanochromism of Two Ag (I) Imidazolate Chains. CrystEngComm 2016, 18, 218–221, with permission from the Royal Society of Chemistry. (C) PL spectra, PXRD patterns and luminescent images of complex 97 with various states, such as as-synthesized, ground and fuming samples. From Ma, X.-H.; Wang, J.-Y.; Guo, J.-J.; Wang, Z.-Y.; Zang, S.-Q. Reversible Wide-Range Tuneable Luminescence of a Dual-Stimuli-Responsive Silver Cluster-Assembled Material. Chin. J. Chem. 2019, 37, 1120–1124.

Organometallic Complexes for Optoelectronic Applications

365

crystal data analysis, the researchers proposed that mechanical grinding led to a disruption of the intermolecular interactions between the phenylene rings of adjacent molecules in blue-emitting crystal, resulting in green emission. In 2011, Babashkina et al. reported a new hexanuclear Ag(I) complex 93, which exhibited reversible conversion between the yellow-emitting complex [(Ag3L3)2] and the blue-emitting complex [Ag3L3] upon grinding and recrystallization.180 In 2016, Ag(I) imidazolate coordination polymers with chain-like structures were reported by Wen et al. Helical chain-like complex 94 with 5,6-dimethylbenzimidazole unit exhibited unique luminescence MC, due to short Ag ⋯Ag interactions (2.986 A˚ ).181 In solid state, complex 94 possessed dual emission, with one wavelength at 384 nm (high energy band, HE) and another intense band at lmax ¼ 530 nm (low energy band, LE). The PL intensity of this complex changed from HE < LE to HE > LE after grinding, corresponding to the emission color change from pale yellow to pale blue (Fig. 29B). In the same year, Tsubomura et al. reported three Ag(I) complexes (95a–95c) bearing N-heterocyclic carbene and diphenylphosphinobenzene ligands.182 Among these complexes, the as-synthesized sample of 95a showed a striking reversible blue shift by 30 nm upon grinding, and the blue-green emission was restored by adding a portion of diethyl ether followed by drying under air, or heating the sample at 100 C for 40 min. The powder XRD signals indicated that 95a became amorphous on grinding, and adding some drops of diethylether to ground powder restored the PXRD signal pattern of 95a. The interaction in the crystal of 95a was very weak, the blue shift of 95a on grinding could be caused by the breaking of the intermolecular interactions between the ligands and the counter anions during the mechanical process. Moreover, as to complex 95c, the introduction of bulky tert-butyl groups at 3- and 5-positions of the phenyl groups in the diphosphine ligand played a key role in both the blue-shift and the enhancement of the PL. In 2017, a new Ag(I) iodide cluster 96 with triply stimuli-responsive luminescent chromism (mechano-, thermo- and solvent-responsive chromism), was reported by Zhang et al.183 Upon the mechanical grinding, the relative intensities of HE and LE of 96 varied with a concomitant blue shift. And when the temperature decreased from 300 to 5 K, unprecedented contrary thermo-responsive trend for single crystal and powered samples (blue shift of single crystals and red shift of powdered samples) was observed. Recently, Zang et al. have prepared a new silver cluster-assembled complex 97 by a facile one-pot method, which exhibited reversible luminescent MC upon mechanical stimulation and solvent treatment.184 97 exhibited reversible luminescent MC between blue (lem ¼ 458 nm) and cyan (lem ¼ 491 nm) upon external mechanical stimulation and treatment with ethanol-acetone (Fig. 29C). The PXRD results confirmed that the luminescent MC was caused by crystal-to-amorphous phase transitions. In addition, complex 97 also showed interesting reversible luminescent thermochromism from blue at room temperature to bluish white at 83 K.

14.10.2.6 Au(I) complexes In 2008, a novel Au(I) complex 98 (Fig. 30) with reversible MC properties via external stimuli was reported by the Ito group.185 Upon gentle grinding, the blue luminescence changed to intense yellow luminescence with a 118 nm red shift under a 365 nm UV light, yet the color did not change under ambient light. Upon exposure to solvent or vapor of dichloromethane, the ground powder with yellow luminescence reverted to the original blue color (Fig. 31A). Moreover, the blue to yellow conversion could be reversibly

Fig. 30 Chemical structures of Au(I) complexes 98–109.

366

Organometallic Complexes for Optoelectronic Applications

Fig. 31 (A) Photographs showing 98 on an agate mortar under UV irradiation (365 nm): (i) after grinding the right-half with a pestle, (ii) the same sample under ambient light, (iii) entirely ground powder, (iv) partial reversion to the blue luminescence by dropwise treatment using dichloromethane onto the center of the ground powder, (v) powder after treatment with dichloromethane, (vi) repetition of the yellow emission by scratching the powder with a pestle. (B) Emission spectra of 100c in various states; insets show the fluorescence images of 100c powder under 365 nm UV light: before grinding, after grinding, after treatment with dichloromethane, and repetition of the green yellow emission by scratching the powder with a pestle. (C) Fluorescence imaging of 101e taken under irradiation of a 365 nm UV light: as-synthesized solid sample, ground sample, and sample after treatment with dichloromethane vapor. (D) Photographic images of 101e thin film, upon exposure to vapors of VOCs and away from VOCs vapors under 365 nm UV light. (A) Reproduced from Ito, H.; Saito, T.; Oshima, N.; Kitamura, N.; Ishizaka, S.; Hinatsu, Y.; Wakeshima, M.; Kato, M.; Tsuge, K.; Sawamura, M. Reversible Mechanochromic Luminescence of [(C6F5Au)2(m-1,4-Diisocyanobenzene)]. J. Am. Chem. Soc. 2008, 130, 10044–10045, with permission from the American Chemical Society. (B) Reproduced from Liang, J.; Hu, F.; Lv, X.; Chen, Z.; Chen, Z.; Yin, J.; Yu, G.; Liu, S. Synthesis, Characterization and Mechanochromic Behavior of Binuclear Gold (I) Complexes With Various Diisocyano Bridges. Dyes Pigm. 2012, 95, 485–490, with permission from the Elsevier. (C) and (D) Reproduced from Chen, Z.; Liang, J.; Nie, Y.; Xu, X.; Yu, G.; Yin, J.; Liu, S. H. A Novel Carbazole-Based Gold (I) Complex With Interesting Solid-State, Multistimuli-Responsive Characteristics. Dalton Trans. 2015, 44, 17473–17477, with permission from the Royal Society of Chemistry.

repeated several times without any degradation in the luminescence. The authors proposed a mechanism of the MC process of 98 that involved two phases. (i) The first phase transformed the microcrystalline powder into a metastable amorphous phase by grinding. This phase had aurophilic interactions responsible for the red shifted emission. (ii) The second phase involved the rearrangement of the amorphous phase into the more stable crystalline phase via a partial dissolution and recrystallization process upon solvent treatment of the ground sample. Tetrahedral Au(I) complexes (99) were reported by Osawa et al. in 2010.186 Upon grinding, the complex 99 with BF4− as a counteranion changed its phosphorescence from blue (lmax ¼ 494 nm) to yellow orange (lmax ¼ 575 nm). A reversible transformation from yellow-orange to blue was observed by treating the ground sample with volatile organic compounds. The XRD patterns revealed the change of the crystalline phase to an amorphous phase after grinding. The authors also demonstrated that symmetry reductions were responsible for the changes in the phosphorescence color caused by external stimuli. In recent years, a large number of MC gold complexes have been reported by many researchers. Among them, Liu and Ito group have done a lot of outstanding work. In 2012, a series of binuclear Au(I) complexes with various aromatic bridges instead of phenyl linkages were synthesized by Liu group, and the ligand structure significantly affected their MC activity.187 The gold complexes with methyl-substituted phenyl bridges (100a and 100b) exhibited MC with up to 100 nm red shift after grinding. The complex with a diphenylmethane bridge (100c) exhibited MC through a change in fluorescence from blue (lem ¼ 477 nm) to green (lem ¼ 503 nm) with a 26 nm red shift after grinding (Fig. 31B). However, as to the biphenyl or diphenylethane containing complex, no obvious MC was observed for complexes 100d and 100e. The ground complexes with MC activities reverted to their original states upon CH2Cl2 treatment. XRD was used to further study the MC behavior. The results showed that grinding caused a crystalline to metastable phase conversion in MC complexes. By contrast, for the complexes without MC activity, their XRD patterns were essentially unchanged even via thoroughly ground, and no crystalline to non-crystalline transition occurred. In 2015, the same group first synthesized a

Organometallic Complexes for Optoelectronic Applications

367

carbazole-based mononuclear Au(I) complex 101e with reversible MC, dual-responsive thermochromism and sensitive thin-film vapochromism properties.188 Upon grinding, complex 101e changed its luminescence from white (pristine solid) to green (ground powder), because of the formation of intermolecular aurophilic interactions. White emission could be restored by fuming or thermal annealing because of the reformation of crystals (Fig. 31C). Interesting, when the thin film of 101e was exposed to vapors of volatile organic compounds (VOCs), including benzene, pyridine, acetone, dichloromethane, toluene etc., the emission band centred at 512 nm red-shifted to the yellow-emitting region with lmax at 556 nm, and the bright green luminescence was converted to yellow emission. Moreover, the yellow emission reverted to the initial green emission, when the thin film was removed from the VOC vapors (Fig. 31D). Subsequently, in the same year, seven analogues with different alkyl chains (101a–101g) were synthesized to study the influence of the length of the alkyl chain on the luminescence and MC.189 All of these complexes exhibited outstanding AIE characteristics and reversible MC behaviors. Moreover, their PL was strongly dependent on the alky chain because of different Au ⋯ Au interactions, which was confirmed by their single-crystal structures. Then, novel diisocyano-based dinuclear Au(I) complex 102 and mononuclear Au(I) complexes (103) with a fluorene-based skeleton were synthesized.190,191 The obtained complex 102 exhibited a crystallization-induced emission enhancement (CIEE) effect and reversible MC behavior, with luminescence changing between green and yellow emissions. The initial emission of all above mentioned complexes could be observed again, after solvent fuming or thermal annealing. The possible mechanism for these interesting AIE and MC phenomena involved a variation in weak multiple intermolecular C–H⋯ F and p ⋯ p interactions and the formation or alteration of aurophilic interactions. In the next year, two carbazole-based gold(Ι) complexes 104a and 104b were synthesized, and both of them showed reversible high-contrast MC and long room-temperature phosphorescence. Upon grinding, reversible MC behaviors with luminescence change between yellow and yellow-green emission were realized. Furthermore, 104a also exhibited AIE property and thin-film vapochromic luminescence behaviors.192 More interestingly, 104b exhibited a self-reversible MC behavior: the ground yellow-green luminescence could be restored to the initial yellow color spontaneously without using other stimuli.193 Diisocyano-based Au(I) moieties directly linked with carbazole bearing different alkyl chains were also synthesized to study the influence of the length of the alkyl chain on the MC.194 105a and 105b had strong yellow emissions, which changed to green after grinding. 105c and 106a–106c were weakly emissive and possessed enhanced green luminescence upon mechanical force stimuli. Three trinuclear Au(I) complexes (107a–107c) with different link position were synthesized,195 which exhibited AIE characteristics and irreversible MC phenomena. They showed off-on green fluorescence in response to mechanical grinding, and the emission in ground solid could not recover to initial state by solvent fuming. Under the mechanical force, the MC process is always accompanied by the change from a crystalline state to an amorphous state. Meanwhile, its inverse process involving transition from amorphous state to crystalline state is usually irreversible without any additional treatments. Thus, self-recovered MC luminogens constitute only a small proportion, and relevant systematic explorations are extremely limited. In 2019, a series of new Au(I) complexes (108a–108f) were reported by Liu group.113 Alkyl chains with different length were elaborately introduced to increase the molecular mobility and ultimately realize the self-recovered MC. Upon grinding, the emission colors of these complexes all changed from green to orange. Excitingly, the orange emission could spontaneously return to their initial green state without any treatments (Fig. 32A). The self-recovery was faster and even too fast

Fig. 32 (A) MC luminescence and the corresponding self-recoveries of 108a–108d powders under 365 nm UV light. Reproduced from Dong, Y.; Zhang, J.; Li, A.; Gong, J.; He, B.; Xu, S.; Yin, J.; Liu, S. H.; Tang, B. Z. Structure-Tuned and Thermodynamically Controlled Mechanochromic Self-Recovery of AIE-Active Au (i) Complexes. J. Mater. Chem. C 2020, 8, 894–899, with permission from the Royal Society of Chemistry. (B) Structure of the complex 109, single crystal structures of its polymorphs, and the photographs of polymorph 109-B in various states taken under 365 nm UV irradiation. Reproduced from Wang, X.-Y.; Yin, Y.; Yin, J.; Chen, Z.; Liu, S. H. Persistent Room-Temperature Phosphorescence or High-Contrast Phosphorescent Mechanochromism: Polymorphism-Dependent Different Emission Characteristics From a Single Gold(I) Complex. Dalton Trans. 2021, 50, 7744–7749, with permission from the Royal Society of Chemistry.

368

Organometallic Complexes for Optoelectronic Applications

to be captured at ambient temperature with the increase of chain length. Lowering the temperature allowed the capture of the originally superfast and unobservable self-recoveries of 108d–108f, due to the decreased molecular mobility. Further investigations revealed that these unique self-recovered MC properties were caused by fast intermolecular rearrangement. Increasing the alkyl chain length could speed up the self-recovery due to much faster intermolecular rearrangements. The driving forces for these distinctive self-recovery processes were the regeneration of strong intermolecular p–p interactions and aurophilic interactions. Most recently, the same group reported a new Au(I) complex 109, and two polymorphs of 109-B and 109-YG with blue and yellow-green luminescence were obtained.196 Interestingly, the blue-emitting crystal 109-B exhibited high-contrast phosphorescent MC behavior, while 109-YG exhibited a persistent room-temperature phosphorescence effect (Fig. 32B). After mechanical stimulation at room temperature, a new emission peak appeared at about 500 nm, and the phosphorescent color of 109-B changed from blue to green. Since the first reported of gold complex 98 in 2008, the Ito group has done a lot of beautiful work on MC Au(I) complexes (Fig. 33), especially, MC that induced by single-crystal to single-crystal (SCSC) transformation via mechanical force. In 2013, a novel chiral Au(I) complex 110 was synthesized by replacing two fluorine atoms in 98 with cholesterol moieties.197 Interestingly, the complex could self-organize into distinct microscopic structures, through vapor-diffusion of a poor solvent into its dichloromethane solution. Thus, acetone vapor produced microbelts with blue emission, which should be ascribed to the phosphorescence from the p–p excited state. By contrast, methanol vapor generated globular agglomerates of sheet-like microstructures with green emission (Fig. 34A). Powder XRD patterns confirmed that blue-emitting microbelts were semicrystalline and green-emitting globular agglomerates were amorphous. Upon grinding, the luminescence of the blue-emitting microbelts changed from blue to green, with a broad emission band at 515 nm due to the shortened Au ⋯ Au distance. The ground powder showed no defined XRD peaks, suggesting the occurrence of a mechanically induced semicrystalline to amorphous transformation. In 2013, Ito group reported a simple Au(I) complex 111, and found a noticeable single-crystal to single-crystal transformation by gentle mechanical force.198 Two kinds of crystals, metastable 111B with blue phosphorescence and stable 111Y with yellow phosphorescence, were obtained through rapid and slow crystallization of 111 from hexane/CH2Cl2, respectively (Fig. 34B). As to 111B, large distances (>4.65 A˚ ) between the gold atoms of adjacent molecules, indicated the absence of aurophilic interactions. By contrast, the distance between adjacent gold atoms in 111Y crystal was much shorter (3.177 A˚ ), which allowed aurophilic interactions and induced luminescence with a red-shifted wavelength. Only 111B had MC activity and changed the PL color from blue to yellow after grinding via ball-milling. XRD test found that the ground sample had exactly the same diffraction peak as 111Y, indicated that the ball-milling process induced direct crystal-to-crystal transformation of 111B to 111Y. Excitingly, the gradually single-crystal to single-crystal phase transition was triggered, when a yellow luminescent spot on the surface of the 111B crystal was created by gentle mechanical force using a needle under atmospheric conditions (Fig. 34C). The transformation could also be triggered by contacting a seed crystal of 111Y with a crystal of 111B (Fig. 34D). Inspired by the epitaxial mechanism proposed for the thermal phase transformation, the authors proposed a mechanism for the mechanical stimulus-triggered phase transformation (Fig. 34E). A small portion of the 111Y daughter phase formed initially in the 111B mother phase through mechanical stress or contacted with a seed crystal of 111Y. The molecules in the thermodynamically unstable 111B phase diffused across the narrow gap between phases 111B and 111Y (behave as molecular domino), and rearranged to the 111Y phase through the formation of intermolecular aurophilic interactions. Subsequently, two methyl groups were introduced to 111, and the resultant complex 112 exhibited similar mechano- and seeding-triggered single-crystal to single-crystal phase transition.199,200 Stable crystal 112B owned blue emitting with a low quantum yield, whereas the metastable crystal 112G had strong green phosphorescence with a higher quantum yield. A gentle

Fig. 33 Chemical structures of Au(I) complexes 110–120.

Organometallic Complexes for Optoelectronic Applications

369

Fig. 34 (A) Chemical structure and schematic representation of solvent-assisted self-organization of complex 110 into distinct microstructures. (B) Photographs and crystal structures of two polymorphs (Ib and IIy) of complex 111. (C) The single-crystal to single-crystal transformation through gentle mechanical force. (D) Contacting a seed crystal of IIy with a crystal of Ib. (E) Proposed mechanism for the mechanical stimulus-triggered phase transformation: (i) mechanical stimulation of the metastable Ib phase (blue rectangles); (ii) generation of the IIy phase; (iii, iv) the thermodynamically stable IIy phase extends by absorbing molecules from the metastable Ib phase. (A) Reproduced from Kawaguchi, K.; Seki, T.; Karatsu, T.; Kitamura, A.; Ito, H.; Yagai, S. Cholesterol-Aided Construction of Distinct Self-Organized Materials From a Luminescent Gold (i)–Isocyanide Complex Exhibiting Mechanochromic Luminescence. Chem. Commun. 2013, 49, 11391–11393, with permission from the Royal Society of Chemistry. (B–E) From Ito, H.; Muromoto, M.; Kurenuma, S.; Ishizaka, S.; Kitamura, N.; Sato, H.; Seki, T. Mechanical Stimulation and Solid Seeding Trigger Single-Crystal-to-Single-Crystal Molecular Domino Transformations. Nat. Commun. 2013, 4, 2009.

mechanical force could trigger a single-crystal to single-crystal transformation from 112G to 112B. However, grinding 112G and 112B both produced intensity green phosphorescence. XRD results showed that during the grinding process, there was no crystal phase change in 112B, and 112G was transformed into 112B. The authors suggested the following MC mechanism. Upon grinding, almost all 112G changed to 112B, but a small amount of residual 112G domain still existed, and similar situation existed in the ground sample of 112B. Thus, energy transfer from blue- to green-emitting crystalline domains occurred in ground powder, and green emission was observed. In 2015, the Ito group found that Au(I) isocyanide complex 113 possessed multiple PL forms (four kinds of solids with different emission colors), which could interconvert by treatment with acetone and mechanical stimulation with solvent-induced blue-shifted and mechano-induced red-shifted (Fig. 35A).201 Soaking this complex in acetone yielded a blue-emitting solid 113B. The subsequent removal of acetone yielded green-emitting 113G via crystal-to-crystal phase transition. 113G exhibited stepwise emission color change to yellow (113Y) and then to orange (113O), upon mechanical stimulation by ball-milling. 113B could be recovered upon the addition of acetone to 113G, 113Y and 113O. Thus, four emitting states could be switched between repeatedly by acetone soaking and the application of mechanical stimulation (Fig. 35B and C). Through single crystal, Rietveld refinement technique, and powder XRD analysis, the authors suggested that the weak interaction of 113 with acetone would allow a solvent inclusion/release mode, which was an important structural factor for the multicolor MC luminescence. Subsequently,

370

Organometallic Complexes for Optoelectronic Applications

Fig. 35 (A) Photographs of the powder forms of complex 113 showing different PL under UV light at 365 nm. (B) Schematic representation of the solid state molecular arrangements of 113, in which a molecule of 113 is denoted as a rectangle, with the colors of the corresponding emission. Solvent molecules are denoted as red circles. (C) Specific procedures for the interconversion of the four emitting states of 113. (D) Photographs showing the luminescent color change of 114 upon grinding. The photograph on the left shows a powder sample of 114B. Photographs labeled from “0 s” to “1.0 s” were taken of a thin film of 114B being ground with a spatula to yield the transient generation of 114Y. The photograph on the right shows the formation of 114G after repeated grinding. (A–C) Reproduced from Seki, T.; Ozaki, T.; Okura, T.; Asakura, K.; Sakon, A.; Uekusa, H.; Ito, H. Interconvertible Multiple Photoluminescence Color of a Gold(I) Isocyanide Complex in the Solid State: Solvent-Induced Blue-Shifted and Mechano-Responsive Red-Shifted Photoluminescence. Chem. Sci. 2015, 6, 2187–2195, with permission from the Royal Society of Chemistry. (D) Reproduced from Yagai, S.; Seki, T.; Aonuma, H.; Kawaguchi, K.; Karatsu, T.; Okura, T.; Sakon, A.; Uekusa, H.; Ito, H. Mechanochromic Luminescence Based on Crystal-to-Crystal Transformation Mediated by a Transient Amorphous State. Chem. Mater. 2016, 28, 234–241, with permission from the American Chemical Society.

complex 114 was obtained by introducing flexible triethylene glycol moieties, which displayed an unprecedented crystallineto-crystalline transformation mediated by a transient amorphous phase.202 When blue-emitting solid (114B) was ground with a spatula, the ground area temporarily showed yellow luminescence (114Y), which automatically vanished and turned back to the initial blue luminescence within seconds (Fig. 35D). Upon grinding 114B repeatedly, a green-emitting solid 114G emerged. Furthermore, grinding of 114G led to the transient emergence of 114Y and the recovery of 114G. This result suggested that the conversion between 114B and 114G was realized by a transition state as an amorphous solid (114Y). Single crystal analysis revealed that the different strength of the aurophilic interactions should be responsible for the multicolor MC. In 2016, by introducing different substituents (six R1 and eight R2 substituents), the Seki and Ito groups prepared 48 Au(I) isocyanide complexes 115 (R1–R2).203 Among them, 37 complexes exhibited good solid state PL emission, and 28 complexes were

Organometallic Complexes for Optoelectronic Applications

371

Fig. 36 Photographs of the 48 R1–R2 complexes 115 under UV illumination. In each panel, unground and ground powders are on the left- and right-hand sides, respectively. Reproduced from Seki, T.; Takamatsu, Y.; Ito, H. A Screening Approach for the Discovery of Mechanochromic Gold (I) Isocyanide Complexes With Crystal-to-Crystal Phase Transitions. J. Am. Chem. Soc. 2016, 138, 6252–6260, with permission from the American Chemical Society.

found to be MC activity (Fig. 36). In addition to the previously reported 111, the newly synthesized CF3–CN complex also exhibited a crystal-to-crystal phase transition upon mechanical stimulation. Density functional theory (DFT) calculations indicated that the mechano-induced red-shifted emission of CF3–CN (from green to orange) was caused by the formation of aurophilic interactions. Comparison of the crystal structures of CF3–CN with those of the other complexes suggested that the weaker intermolecular interactions in the as-prepared sample were the important factor for the mechano-induced crystal-to-crystal phase transition. In the same year, a chiral Au(I) complex with luminescent MC was investigated. The racemic (116a) and homochiral (116b) forms of complex 116 owned distinct crystal packing with different emission colors.204 Cubic crystals of 116a exhibited yellowish green emission, and the hexahedral crystals of 116b displayed green emission. Upon grinding, both crystals transformed into amorphous powders that exhibiting similar red-shifted emission color (Fig. 37A). An unprecedented low-temperature-selective MC of a thienyl Au(I) complex was reported in 2017. At room temperature, the as-prepared solid of complex 117 did not show any emission color changes upon grinding.205 When the temperature was below −50 C, a notable MC behavior was observed, and the emission color of 117 changed from blue to green upon grinding, due to the occurrence of the phase transition at low temperatures (Fig. 37B). In 2018, a new Au(I) isocyanide complex 118 with a pendant carboxy group was reported, and two polymorphs with opposite (red- and blue-shifted) MC behaviors were achieved.206 Blue-emitting (118B) and magenta-emitting (118R) crystals were obtained from recrystallization of 118. Interestingly, 118R adopted a unique orthogonal molecular arrangement due to the formation of double hydrogen bonds (from the presence of the pendant carboxy group), resulting in a strong aurophilic interaction. Furthermore, this polymorph showed blue-shifted luminescence MC. Upon grinding, the change of the aurophilic interactions in 118B and 118R occurred in an opposite trend. The aurophilic interactions were formed in the ground sample of 118B with the emission color red-shifted from blue to green, whereas they were weakened in 118R, resulting in blue-shifted emission from magenta to green. In the same year, a bent-shaped meta-diisocyanide benzene-based binuclear Au(I) isocyanide complex 119 exhibiting multiple solid-state molecular arrangements and luminescent MC was also reported.207 Recently, two N-heterocyclic carbene Au(I) complexes 120a and 120b with contrasting luminescent MC behavior were reported.208 Upon grinding, complex 120a displayed blue-shifted emission from green to blue, whereas the emission of complex 120b exhibited a red-shifted emission from blue to green (Fig. 37C). The opposite emission shift upon mechanical grinding could be explained by the aggregate-to-monomer transformation and intensified intermolecular interactions, respectively (Fig. 37D). As mentioned above, grinding-triggered crystalline to amorphous phase transition and grinding-triggered single-crystal to single-crystal phase transformation are two typical morphological changes. There are almost no reports about the transition from amorphous to crystalline that triggered by mechanical force, because it is difficult to realize an entropy-decrease process from a disordered state to another well-ordered state upon mechanical stimulation.110 Recently, two impressive chiral Au(I) complexes 121

372

Organometallic Complexes for Optoelectronic Applications

Fig. 37 (A) Structures of complexes 116a and 116b, and photographs of their crystals and ground powders under UV irradiation. (B) Structure and the photographs of 117 in various states taken under UV irradiation. (C) Photographs and emission spectra of 120a and 120b obtained before and after mechanical stimulation taken under UV light. (D) Schematic representation of the molecular packing of 120a and 120b, which explains the opposite shift of their luminescent MC. (A) Reproduced from Jin, M.; Seki, T.; Ito, H. Luminescent Mechanochromism of a Chiral Complex: Distinct Crystal Structures and Color Changes of Racemic and Homochiral Gold(i) Isocyanide Complexes With a Binaphthyl Moiety. Chem. Commun. 2016, 52, 8083–8086, with permission from the Royal Society of Chemistry. (B) Reproduced from Seki, T.; Kobayashi, K.; Ito, H. Low-Temperature-Selective Luminescent Mechanochromism of a Thienyl Gold Isocyanide Complex. Chem. Commun. 2017, 53, 6700–6703, with permission from the Royal Society of Chemistry. (C) and (D) Reproduced from Seki, T.; Kashiyamaa, K.; Ito, H. Luminescent Mechanochromism of Gold N-Heterocyclic Carbene Complexes With Hypso-and Bathochromic Spectral Shifts. Dalton Trans. 2019, 48, 7105–7109, with permission from the Royal Society of Chemistry.

(Fig. 38) were reported by Tang et al., in which their powders could realize a dramatic transformation from nonemissive isolated crystallites to emissive well-defined microcrystals under the stimulation of mechanical force (Fig. 39).209 Such an unusual crystallization was presumed to be caused by molecular motions driven by the formation of strong aurophilic interactions as well as multiple C− H⋯ F and p ⋯ p interactions.110 Since 2014, several double-stranded digold(I) complexes with phenyl-substituted diphosphine ligands were reported by Deák’s group. Multiple molecular conformations with different packing arrangements and solid-state emission colors were obtained from complex 122a due to its flexible structure.210 Blue (122-B), bluish green (122-G) and yellow (122-Y) emitting crystalline samples could be obtained by slight modification of the crystallization conditions. Upon grinding, the emission color of 122-B changed from blue to red (amorphous state, lmax ¼ 690 nm) with a huge red-shifted about 200 nm. The analogues of 122a with different counteranions also exhibited good MC behavior.211 Another similar complex 122b was reported by the same group.212 In crystal, one-dimensional helical packing was formed by hydrogen bonding between N–H and CF3COO−, and the emission maximum of newly prepared crystals was centered at 621 nm. The luminescence disappeared upon grinding, and turned on after exposure to solvent vapors because of recrystallization. In 2017, a series of donor–acceptor Au(III) complexes (123–125) were synthesized by the Yam group.213 These complexes displayed high PL quantum yields of up to 0.81, and also interesting reversible MC luminescence behaviors. Upon grinding, a dramatic luminescence color change from green to red could be observed in solid state. XRD results indicated that their MC behaviors could be attributed to a crystal-to-amorphous state conversion.

Organometallic Complexes for Optoelectronic Applications

373

Fig. 38 Chemical structures of Au(I) complexes.

Recently, a double salt prepared by co-precipitation of complex 126 and K[Au(CN)2] in methanol with a quantitative yield was report by Chen et al.214 Two polymorphs with distinct phosphorescence (needle-shaped cyan-emitting and block green-emitting crystals) and variations at Au ⋯Au distances and ligand–ligand torsion angles were obtained (Fig. 40A). Upon grinding, both the cyan- and green-luminescent polymorphs produced an orange-emitting (lem ¼ 610 nm) ground powder (Fig. 40B). Upon treatment by an organic solvent, only the green luminescence was restored to the powder. Most recently, cyclohexyl-based Au(I) isocyanide complex 127 with three polymorphs was reported by Balch et al.215 These polymorphs owned totally different colors under ambient light and UV light: yellow crystals with green-emitting (127-1), colorless crystals with blue-emitting (127-2), and colorless crystals with non-luminescent (127-3). The close approach of the cations was responsible for the distinct luminescence, since solutions of either polymorph were colorless and non-luminescent. Interestingly, the colorless non-luminescent 127-3 could be selectively transformed into green luminescent 127-1, by either mechanical grinding or CH2Cl2 vapor (Fig. 40C). A novel N-heterocyclic carbine Au(I)-Cu(I) complex (128) was reported by Catalano et al., which could transform into three crystals with different emission colors upon crystallization from different solvents.216 Upon grinding, the emission color of the as-prepared crystals of 128
2CH3CN changed from blue to yellow (Fig. 40D). After the dropping of acetonitrile to the ground powder, the blue emission recovered. This reversible MC was caused by crystalline-to-amorphous conversion accompanied by partial desolvation. In 2019, two 4-pyridylisocyanide Au(I) complexes (129a and 129b), and their silver incorporated Au(I)-Ag(I) complexes (130a and 130b) were reported by Espinet et al.217 The incorporation of Ag(I) ligand caused a significant red-shifted of the PL spectrum, due to structural changes disturbing stronger p ⋯ p stackings and shorter Au ⋯ Au interactions. Upon grinding, the blue luminescence of 129a changed to an intense yellow luminescence, and the green-emitting 129b also changed to yellow, almost coincident with the ground powder of 129a.

14.10.2.7 Zn(II) complexes In 2005, a helical 3,30 -di-tert-butylsalen-Zn(II) complex 131 (Fig. 41) with unique solid-state optical properties was reported by Mizukami et al.218 In the solid state, crystals of 131 emitted a bluish-green (lmax ¼ 488 nm) fluorescence, however, the emission color changed to light blue (lmax ¼ 474 nm) upon treatment by grinding or solvent fuming (Fig. 42A). Interestingly, two microcrystals that possessed these two fluorescence colors could be obtained by sublimating at the same time. Also, green- and blue-emitting crystals were gotten upon crystallization from different solvents. Thus, the MC property was assumed to be the result of crystal phase transition, which weakened the intermolecular p⋯ p interactions. In 2011, a tetraphenylethylene based AIE-active Zn(II) complex 132 with multi-stimuli-responsive activity was reported by the Chi group.219 The colors and emissions could be smartly switched using various external stimuli, including grinding, heating, solvent-fuming and exposure to acid and base vapors. The as-synthesized complex exhibited strong blue emissions in solid state. Upon grinding, a red-shifted of the PL spectrum about 81 nm was achieved, and the emission color changed from blue (lem ¼ 474) to brilliant yellow (lem ¼ 555). Moreover, a fluorescence off-on switching of this complex could be observed upon exposure to acid and base vapors. Most recently, a novel Schiff base Zn(II) complex 133 containing tetraphenylethylene moieties was reported by Zheng et al.220 Upon exposure to the acid and base vapors, it exhibited remarkable tricolor acidochromic behavior with rapid

374

Organometallic Complexes for Optoelectronic Applications

Fig. 39 (A) PXRD patterns of 121 powder and its different crystals measured by synchrotron X-ray. Inset: Photos of 121 crystal showing luminescence during the scratching process taken under 365 nm UV light at room temperature. (B) PL spectra of 121 crystal (nonemissive) obtained by crystallization from DCM/hexane in different states. (C) Schematic illustration of the proposed force-induced solid-state molecular motion and luminescence mechanism. Reproduced from Zhang, J.; He, B.; Wu, W.; Alam, P.; Zhang, H.; Gong, J.; Song, F.; Wang, Z.; Sung, H. H. Y.; Williams, I. D.; Wang, Z.; Lam, J. W. Y.; Tang, B. Z. Molecular Motions in AIEgen Crystals: Turning on Photoluminescence by Force-Induced Filament Sliding. J. Am. Chem. Soc. 2020, 142, 14608–14618, with permission from the American Chemical Society.

response and high contrast (Fig. 42B). The XPS analyses showed that the acidochromism originated mainly from the adsorption of vapor and the gas-solid reaction equilibrium on the crystal surface. Complex 133 also exhibited reversible luminescence MC behavior between yellow and orange emission, during the grinding–fuming/heating cycles due to the converting between crystalline and amorphous states. In 2018, a new bis(salicylaldiminato)Zn(II) Schiff base complex 134 with reversible MC and organogelation properties was synthesized by Wang group.221 Upon mechanical grinding, the as-prepared crystals of 134 showed a high-contrast emission color change from yellow (lem ¼ 545 nm) to red (lem ¼ 645 nm), and the ground solid could return to yellow via solvent fuming. SEM, powder XRD and thermal analyses demonstrated that mechanical force could induce the transformation from the crystalline to the amorphous state. In the next year, the same group reported a similar Zn(II) complex 135 based on a phthalonitrile-bridging salophen skeleton, which showed an excellent reversible MC properties.222 Reversible emission color change between yellow (lmax ¼ 560 nm) and orange (lmax ¼ 591 nm) could be repeatedly switched by grinding and organic vapor fuming with highly sensitive. Thus, a potential sensor platform for both external pressure and volatile organic compounds based on 135 was constructed (Fig. 42C). Recently, three quinoline-based Zn(II) complexes 136–137 were synthesized by the Chen group, which exhibited aggregation-induced emission enhancement (AIEE) characteristics and reversible MC behaviors.223 Interestingly, the AIEE performance could be obviously promoted by linking the two quinoline units through an alkoxy chain (137). All these

Organometallic Complexes for Optoelectronic Applications

375

Fig. 40 (A) Fluorescence micrographs of polymorphs prepared from the precursors of 126 and K[Au(CN)2], and the reversible transformation between polymorphs and ground powder. (B) Normalized emission spectra of the corresponding polymorphs and ground powder at 297 K. (C) Photographs of 127 under ambient light (left) and ground solid under UV light (right). (D) Photographs of 128 in the pristine crystal (left) and ground solid (right) under UV light. (A) and (B) Reproduced from Liu, Q.; Xie, M.; Chang, X.; Gao, Q.; Chen, Y.; Lu, W. Correlating Thermochromic and Mechanochromic Phosphorescence With Polymorphs of a Complex Gold (i) Double Salt With Infinite Aurophilicity. Chem. Commun. 2018, 54, 12844–12847, with permission from the Royal Society of Chemistry. (C) Reproduced from Luong, L. M. C.; Olmstead, M. M.; Balch, A. L. A Non-Luminescent Polymorph of [(Cyclohexyl Isocyanide) 2 Au] PF 6 That Becomes Luminescent Upon Grinding or Exposure to Dichloromethane Vapor. Chem. Commun. 2021, 57, 793–796, with permission from the Royal Society of Chemistry. (D) Reproduced from Chen, K.; Nenzel, M. M.; Brown, T. M.; Catalano, V. J. Luminescent Mechanochromism in a Gold (I)–Copper (I) N-Heterocyclic Carbene Complex. Inorg. Chem. 2015, 54, 6900–6909, with permission from the American Chemical Society.

Fig. 41 Chemical structures of Zn(II) complexes.

376

Organometallic Complexes for Optoelectronic Applications

Fig. 42 (A) Fluorescence microscope images of (i) single crystals, (ii) crushed crystals (powder), (iii) single crystals after exposure of THF vapor, and (iv) sublimed microcrystals of 131. Reproduced from Mizukami, S.; Houjou, H.; Sugaya, K.; Koyama, E.; Tokuhisa, H.; Sasaki, T.; Kanesato, M. Fluorescence Color Modulation by Intramolecular and Intermolecular p–p Interactions in a Helical Zinc (II) Complex. Chem. Mater. 2005, 17, 50–56, with permission from the American Chemical Society. (B) Photographs of the 133 samples in different states under ambient light and 365 nm UV illumination. Reproduced from Zheng, H.-W.; Wu, M.; Yang, D.; Liang, Q.-F.; Li, J.-B.; Zheng, X.-J. Multistimuli Responsive Solid-State Emission of a Zinc (II) Complex with Multicolour Switching. Inorg. Chem. 2021, 60(15), 11609–11615, with permission from the American Chemical Society. (C) Writing and erasing cycles on 135 coated paper. Reproduced from Yan, X.; Song, X.; Mu, X.; Wang, Y. Mechanochromic Luminescence Based on a Phthalonitrile-Bridging Salophen Zinc (II) Complex. New J. Chem. 2019, 43, 15886–15891, with permission from the Royal Society of Chemistry.

complexes showed similar reversible MC behavior, especially the switching between dark and bright emission states during the grinding–fuming cycles, due to the transformation between amorphous and crystalline states. Taking complex 136a as an example, upon grinding, the emission color of the newly prepared crystals changed from yellow to weakness orange, accompanied by a significant decrease of quantum yields (Fem) from 17.23% to 2.5%. Then, upon fuming the ground powder with CH2Cl2 for 5 min, the initial strong yellow emission was restored. Recently, a novel Zn(II) complex that exhibited mechanical force induced crystal-to-crystal transformation was reported by Zheng et al.224 The as-prepared crystal solids of 138 showed red emission with the maximum peak at 647 nm. After slight grinding, the red-emitting crystal was transformed into orange-emitting mini-crystals (138-SG) with a blue-shifted emission centered at 624 nm. Subsequently, after heavily grinding, yellow-emitting samples with the PL maximum peak were observed to further blue-shift to 608 nm, but no suitable crystals for single crystal X-ray diffraction was obtained (Fig. 43A and B). Carefully analysis

Fig. 43 (A) Top: naked eye visualization under room light and UV light; bottom: the corresponding SEM image. (B) Fluorescence spectra of 138 in different states. (C) Photographs of crystalline powder of 140 at different treatment conditions under UV irradiation at room temperature. (A) and (B) Reproduced from Li, S.; Wu, M.; Kang, Y.; Zheng, H.-W.; Zheng, X.-J.; Fang, D.-C.; Jin, L.-P. Grinding-Triggered Single Crystal-to-Single Crystal Transformation of a Zinc(II) Complex: Mechanochromic Luminescence and Aggregation-Induced Emission Properties. Inorg. Chem. 2019, 58, 4626− 4633, with permission from the American Chemical Society. (C) Reproduced from Zou, R.; Zhang, J.; Hu, S.; Hu, F.; Zhang, H.; Fu, Z. Reversible Mechanochromic and Thermochromic Luminescence Switching via Hydrogen-Bond-Directed Assemblies in a Zinc Coordination Complex. CrystEngComm 2017, 19, 6259–6262, with permission from the Royal Society of Chemistry.

Organometallic Complexes for Optoelectronic Applications

377

of the crystal structures of 138 and 138-SG suggested that it was due to the alteration of molecular conformation of the ligands and a significant sliding of its crystal packing instead of weak intermolecular interaction, which resulted in its unique MC property. Zn(II) complex 139 with two type of coordination frameworks was reported by Tzeng et al., i.e., a 1D double-zigzag framework with two coordinated water (139-B) and a 2D polyrotaxane framework without water (139-N).225,226 Complex 139-B was strongly blue emissive (lmax ¼ 495 nm), whereas complex 139-N was no emissive due to the intermolecular stacking interactions (such as, p ⋯ p interactions). Upon grinding the no emissive powders of 139-N with one drop of water, a strong blue emission was immediately observed, which was caused by a phase transformation from the polyrotaxane framework to the double-zigzag one. Subsequently, Zhang et al. reported a new zinc coordination complex 140 with a planar aromatic core and multiple hydrogen-bonding sites.227 A bifunctional luminescent switch was constructed by introducing a terpyridyl derivative into a hydrogen-bond-assisted layered assembly structure. Under UV irradiation at 365 nm at room temperature, it exhibited interesting MC and thermochromic luminescence behaviors with reversible color changes from pink to blue-purple upon mechanical grinding and from pink to blue upon heating (Fig. 43C).

14.10.2.8 Other complexes Compared with the above mentioned organometallic complex systems, the reports of other metal complexes with MC behaviors were relatively limited (Fig. 44). In 2014, a mechano-responsive luminescent Cd(II) coordination polymer 141 was reported by Zheng et al.228 The single crystal of 141 gave a weak yellowish-green emission, and cyan-emitting defected areas appeared when slight crushing the crystal by a metal spatula (Fig. 45A). Moreover, ground powder with bright cyan emission could be observed when it was heavily ground by a glass rod (Fig. 45B). Powder XRD results indicated that the initial crystal and the ground powder had almost the same diffraction patterns. Thus, crushing or grinding did not cause an obvious crystal phase transition. The authors suggested that the luminescence MC and crystalline maintenance of 141 could be caused by a minor crystalline structural variation during grinding. In 2018, a novel Cd(II) based multichromic coordination polymer 142 was synthesized by Biradha et al., which consisted of two-dimensional layered structure with continuous channels.229 Upon mechanical grinding, the blue-emitting powder of 142 (lmax ¼ 412 nm) changed its emission color to cyan (lmax ¼ 490 nm). This complex also exhibited instant and reversible solvatochromic property. Most recently, a 1D chain co-crystallized Cd(II) coordination polymer was reported by Yan et al. Complex 143 with an ordered arrangements of rigid anion and flexible cation, exhibited reversible luminescence MC, which could be recovered via solvent treatment or a self-recovery process.230 Upon grinding, 143 exhibited an obvious red-shifted emission from blue (lem ¼ 479 nm) to green (lem ¼ 498 nm) with an enhanced PL intensity at room temperature under UV lamp (Fig. 45C). Moreover, the initial blue emission could be restored by a rapid recovery process under CH3CN fuming, or a few hours self-recovery process under natural conditions. During mechanical grinding, the sample suffered a crystal to quasi-amorphous transformation accompanied by the weakening of anionic and cationic interactions. On the other hand, both of the above-mentioned recovery processes were attributed to the self-recognized electrostatic interaction between the anion and cation. Thus, this work demonstrated that developing co-crystallized anionic–cationic 1D chain polymers with rigid frameworks and flexible interactions could be an effective strategy

Fig. 44 Chemical structures of complexes 141–146.

378

Organometallic Complexes for Optoelectronic Applications

Fig. 45 (A) Photographs of a pristine single crystal and crystal fragments after the crushing of 141 under UV irradiation. (B) Photographs of 141 showing the luminescence changes after grinding under a 365 nm UV lamp. (C) Solid-state emission colors of 143 upon grinding and fuming by CH3CN under a UV lamp. (D) Photographs of 144 showing luminescence changes with grinding time under ambient light and UV lamp. (A) and (B) Reproduced from Yan, Y.; Chen, J.; Zhang, N.-N.; Wang, M.-S.; Sun, C.; Xing, X.-S.; Li, R.; Xu, J.-G.; Zheng, F.-K.; Guo, G.-C. Grinding Size-Dependent Mechanoresponsive Luminescent Cd (II) Coordination Polymer. Dalton Trans. 2016, 45, 18074–18078, with permission from the Royal Society of Chemistry. (C) Reproduced from Yang, Y.; Fang, X.; Zhao, S.-S.; Bai, F.; Zhao, Z.; Wang, K.-Z.; Yan, D. One-Dimensional Co-Crystallized Coordination Polymers Showing Reversible Mechanochromic Luminescence: Cation–Anion Interaction Directed Rapid Self-Recovery. Chem. Commun. 2020, 56, 5267−5270, with permission from the Royal Society of Chemistry. (D) Reproduced from Wen, T.; Zheng, Y.; Xu, C.; Zhang, J.; Jaroniec, M.; Qiao, S.-Z. A Boron Imidazolate Framework With Mechanochromic and Electrocatalytic Properties. Mater. Horiz. 2018, 5, 1151−1155, with permission from the Royal Society of Chemistry.

to obtain self-recovery MC materials. A novel alkaline-stable Cd(II) boron imidazolate framework 144 was synthesized by Liao et al. in 2018, and this 3D molecular structure exhibited a unique MC behavior.231 The as-prepared 144 solids emitted bright blue light (lem ¼ 461 nm) attributed to the ligand-to-ligand charge transfer, and its ground sample exhibited an obvious blue shift of about 23 nm (lem ¼ 438 nm) (Fig. 45D). The PXRD, IR and TG analyses proved that 144 still remained its original crystallinity and the almost unchanged framework after grinding. The authors suggested that the alteration of p⋯ p interactions between the imidazole ring and benzene ring in the anionic structure might be responsible for the luminescence change. In 2011, a blue-emitting fac-Alq3 (145a) crystalline powder was obtained by annealing mer-Alq3 (145b) at around 400 C.232 The blue-emitting 145a was converted to a green-emitting amorphous solid after grinding or pressing treatment, and the initial blue emission sample could be restored by heating (Fig. 46A). The powder XRD and solid state NMR data indicated that the ground green

Fig. 46 (A) MC behaviors of complexes 145a (top) and 145b (bottom) under 365 nm UV light. Reproduced from Bi, H.; Chen, D.; Li, D.; Yuan, Y.; Xia, D.; Zhang, Z.; Zhang, H.; Wang, Y. A Green Emissive Amorphous fac-Alq 3 Solid Generated by Grinding Crystalline Blue fac-Alq 3 Powder. Chem. Commun. 2011, 47, 4135–4137, with permission from the Royal Society of Chemistry. (B) Chemical structures, photographs and emission spectra of pristine and ground powders of 146a and 146b under UV light. Reproduced from Calupitan, J. P.; Poirot, A.; Wang, J.; Delavaux-Nicot, B.; Wolff, M.; Jaworska, M. Métivier, R.; Benoist, E.; Allain, C.; Fery-Forgues, S. Mechanical Modulation of the Solid-State Luminescence of Tricarbonyl Rhenium (I) Complexes Through the Interplay Between Two Triplet Excited States. Chem. Eur. J. 2021, 27, 4191–4196, with permission from the John Wiley & Sons.

Organometallic Complexes for Optoelectronic Applications

379

solid still remained its original facial form. However, the emission of the ground sample of 145b did not show a significant color change. The enhanced intermolecular interactions of the quinoline ligands could be the key factor that determined the newly achieved luminescence of 145a. Most recently, two tricarbonyl Re(I) complexes with remarkable MC activity were first reported by Allain et al. in 2021.233 In solid state, pristine microcrystalline powders of 146a and 146b showed yellow and green emission, respectively. Mechanical grinding of these complexes caused a formation of amorphous phase with reduced PL quantum yields, also a red shift of emission by 30 nm for 146a and 58 nm for 146b were obtained (Fig. 46B). Subsequently, the ground samples could be fully restored to their initial states upon exposing to solvent vapors. This reversible process could be repeated for several times without obvious changes. Quantum chemical calculations revealed the existence of two low-lying triplet excited states with very similar energy levels with almost pure intraligand (IL) and metal-to-ligand charge-transfer (MLCT) character. Transition between these states could be promoted by rotation around the pyridyltriazole-phenylbenzoxazole bond. Upon grinding, rotation was facilitated and the transition to the 3MLCT state resulted in a larger proportion of long-wavelength PL.

14.10.2.9 Conclusions MC luminescent organometallic complexes based on different metals (Ir(III), Pt(II), Cu(I), Ag(I), Au(I), Zn(II) and other metal ions), and the structure-property relationships as well as the proposed MC mechanisms have been described in detail. MC behavior occurs due to the change in molecular arrangement and crystalline phase transition, in response to the external stimuli such as mechanical grinding or external pressure. Generally, the induced emission color change can be restored to the initial color through recrystallization via heating, solvent fuming or self-recovery. The phase transformations are always accompanied by the destruction and reconstruction of intermolecular interactions (such as metal–metal, p–p, hydrogen bonding, cation–anion, and dipole–dipole interactions). Among them, mechano-triggered crystalline to amorphous phase transformation and mechano-triggered singlecrystal to single-crystal phase transformation are two typical morphological changes. Most recently, impressive transformation from isolated crystallites to well-defined microcrystals under mechanical force is also reported. For Pt(II), Cu(I), Ag(I) and Au(I) complex systems, most MC properties are obtained by phase transformation accompanied by the alteration of metal–metal interactions, which usually exhibit phosphorescent with a large spectral shift after grinding. As to Ir(III) based complexes, the spectral shift is usually relatively small, due to the absence of Ir–Ir interaction in the crystal and ground solid. In addition, Zn(II) and Cd(II) based MC complexes are usually caused by the changing of the stacking mode of their ligands. For the further and rapid development of MC organometallic complexes, perhaps, some issues should be considered: (1) the quantitative study of the relationship between of the stimulation and the corresponding property changes (emission wavelength and intensity). If the change of MC can be quantified, the mechanism research and application development of MC materials will be explored more deeply. Thus, some targeted equipment might need to be constructed for the required experiments. (2) Most reported MC materials displayed emission switches between two colors in response to mechanical stimulation. Thus, the development of high-contrast multicolor luminescent switch will beneficial to extend their applications. (3) In recent years, AIE-active MC complexes have been widely reported. Twisted AIE moieties with strong solid state emission are useful building blocks for MC compounds, and the corresponding crystals with loose packing can be destroyed by slight mechanical stimuli. Thus, AIE organometallic complexes are expected to become an important source of MC materials.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

Zink, J. I. Acc. Chem. Res. 1978, 11, 289–295. Zink, J. I.; Kaska, W. C. J. Am. Chem. Soc. 1973, 95, 7510–7512. Chandra, B. P.; Zink, J. I. Inorg. Chem. 1980, 19, 3098–3102. Picard, J. Science 1939, 89, 460–461. Xie, Y.; Li, Z. Chem 2018, 4, 943–971. Peng, D.; Chen, B.; Wang, F. ChemPlusChem 2015, 80, 1209–1215. Xiong, P.; Peng, M.; Yang, Z. iScience 2021, 24, 101944. Feng, A.; Smet, P. F. Materials 2018, 11, 484. Wang, X.; Zhang, H.; Yu, R.; Dong, L.; Peng, D.; Zhang, A.; Zhang, Y.; Liu, H.; Pan, C.; Wang, Z. L. Adv. Mater. 2015, 27, 2324–2331. Wang, C.; Yu, Y.; Yuan, Y.; Ren, C.; Liao, Q.; Wang, J.; Chai, Z.; Li, Q.; Li, Z. Matter 2020, 2, 181–193. Xie, Y.; Li, Z. Mater. Chem. Front. 2020, 4, 317–331. Xu, C. N.; Watanabe, T.; Akiyama, M.; Zheng, X. G. Appl. Phys. Lett. 1999, 74, 1236–1238. Jeong, S. M.; Song, S.; Joo, K.-I.; Kim, J.; Hwang, S.-H.; Jeong, J.; Kim, H. Energy Environ. Sci. 2014, 7, 3338–3346. Terasaki, N.; Xu, C. N. J. Colloid Interface Sci. 2014, 427, 62–66. Zhang, J.-C.; Long, Y.-Z.; Yan, X.; Wang, X.; Wang, F. Chem. Mater. 2016, 28, 4052–4057. Chen, Y.; Zhang, Y.; Karnaushenko, D.; Chen, L.; Hao, J.; Ding, F.; Schmidt, O. G. Adv. Mater. 2017, 29, 1605165. Tu, D.; Xu, C. N.; Yoshida, A.; Fujihala, M.; Hirotsu, J.; Zheng, X. G. Adv. Mater. 2017, 29, 1606914. Wang, C.; Xu, B.; Li, M.; Chi, Z.; Xie, Y.; Li, Q.; Li, Z. Mater. Horiz. 2016, 3, 220–225. Dia, B.-H.; Chen, Y.-L. Chin. Chem. Lett. 2018, 29, 245–251. Ubba, E.; Tao, Y.; Yang, Z.; Zhao, J.; Wang, L.; Chi, Z. Chem. Asian J. 2018, 13, 3106–3121. Mukherjee, S.; Thilagar, P. Angew. Chem. Int. Ed. 2019, 58, 7922–7932. Bünzli, J.-C. G.; Wong, K.-L. J. Rare Earth 2018, 36, 1–41. Jha, P.; Chandra, B. P. Luminescence 2014, 29, 977–993. Olawale, D. O.; Dickens, T.; Sullivan, W. G.; Okoli, O. I.; Sobanjo, J. O.; Wang, B. J. Lumin. 2011, 131, 1407–1418.

380 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96.

Organometallic Complexes for Optoelectronic Applications Liu, M.; Wu, Q.; Shi, H.; An, Z.; Huang, W. Acta Chim. Sin. 2018, 76, 246–258. Goodgame, D. M. L.; Cotton, F. A. J. Chem. Soc. 1961, 3735–3741. Cotton, F. A.; Goodgame, D. M. L.; Goodgame, M. J. Am. Chem. Soc. 1962, 84, 167–172. Chen, B.; Zhang, X.; Wang, F. Acc. Mater. Res. 2021, 2, 364–373. Chang, K.; Li, Q.; Li, Z. Chin. J. Org. Chem. 2020, 40, 3656–3671. Li, Q.; Li, Z. Acc. Chem. Res. 2020, 53, 962–973. Chandra, B. P.; Chandra, V. K.; Jha, P.; Patel, R.; Shende, S. K.; Thaker, S.; Baghel, R. N. J. Lumin. 2012, 132, 2012–2022. Chandra, B. P.; Chandra, V. K.; Jha, P. J. Lumin. 2013, 135, 139–153. Zink, J. I.; Hardy, G. E.; Sutton, J. E. J. Phys. Chem. 1976, 80, 248–249. Camara, C. G.; Escobar, J. V.; Hird, J. R.; Putterman, S. J. Nature 2008, 455, 1089–1092. Butler, C. Phys. Rev. 1966, 141, 750–757. Li, L.; Wong, K.-L.; Li, P.; Peng, M. J. Mater. Chem. C 2016, 4, 8166–8170. Chandra, B.; Chandra, V.; Jha, P. Phys. B 2015, 463, 62–67. Wei, X. Y.; Wang, X.; Kuang, S. Y.; Su, L.; Li, H. Y.; Wang, Y.; Pan, C.; Wang, Z. L.; Zhu, G. Adv. Mater. 2016, 28, 6656–6664. Matsui, H.; Xu, C.-N.; Liu, Y.; Tateyama, H. Phys. Rev. B: Condens. Matter Mater. Phys. 2004, 69, 235109. Xu, S.; Liu, T.; Mu, Y.; Wang, Y. F.; Chi, Z.; Lo, C. C.; Liu, S.; Zhang, Y.; Lien, A.; Xu, J. Angew. Chem. Int. Ed. 2015, 54, 874–878. Li, Q.; Tang, Y.; Hu, W.; Li, Z. Small 2018, 14, 1801560. Fontenot, R. S.; Bhat, K. N.; Hollerman, W. A.; Aggarwal, M. D.; Nguyen, K. M. CrystEngComm 2012, 14, 1382–1386. Fontenot, R. S.; Hollerman, W. A.; Aggarwal, M. D.; Bhat, K. N.; Goedeke, S. M. Measurement 2012, 45, 431–436. Hollerman, W. A.; Fontenot, R. S.; Bhat, K. N.; Aggarwal, M. D.; Guidry, C. J.; Nguyen, K. M. Opt. Mater. 2012, 34, 1517–1521. Sakai, K.; Koga, T.; Imai, Y.; Maehara, S.; Xu, C. N. Phys. Chem. Chem. Phys. 2006, 8, 2819–2822. Li, C.; Xu, C. N.; Imai, Y.; Bu, N. Strain 2011, 47, 483–488. Cheng, Z.; Lin, J. Macromol. Rapid Commun. 2015, 36, 790–827. Huang, J.; Hou, B.; Ling, H.; Liu, J.; Yu, X. Inorg. Chem. 2014, 53, 9541–9547. Liu, Z.; Lai, B.; Wen, H.; Robbins, J.; Nei, H. Chin. J. Chem. 2016, 34, 1304–1308. Hurt, C. R.; Mcavoy, N.; Bjorklund, S.; Filipescu, N. N. Nature 1966, 212, 179–180. Sweeting, L. M.; Rheingold, A. L. J. Am. Chem. Soc. 2002, 109, 2652–2658. Fontenot, R. S.; Hollerman, W. A.; Bhat, K. N.; Aggarwal, M. D. J. Lumin. 2012, 132, 1812–1818. Fontenot, R. S.; Bhat, K. N.; Owens, C. A.; Hollerman, W. A.; Aggarwal, M. D. J. Lumin. 2015, 158, 428–434. Fontenot, R. S.; Bhat, K. N.; Hollerman, W. A.; Alapati, T. R.; Aggarwal, M. D. ECS J. Solid State Sci. Technol. 2013, 2, 384–388. Fontenot, R. S.; Hollerman, W. A.; Bhat, K. N.; Aggarwal, M. D. J. Lumin. 2013, 134, 477–482. Fontenot, R. S.; Hollerman, W. A.; Bhat, K. N.; Aggarwal, M. D.; Penn, B. G. Polym. J. 2013, 46, 111–116. Chen, X.; Liu, S.; Yu, Z.; Cheung, K.; Ma, J.; Min, N.; You, X. J. Coord. Chem. 1999, 47, 349–358. Xiong, R.; You, X. Inorg. Chem. Commun. 2002, 5, 677–681. Wong, H. Y.; Lo, W. S.; Chan, W. T. K.; Law, G. L. Inorg. Chem. 2017, 56, 5135–5140. Rheingold, A. L.; King, W. Inorg. Chem. 1989, 28, 1715–1719. Chen, X.; Liu, S.; Duan, C.; Xu, Y.; You, X. Polyhedron 1998, 17, 1883–1889. Chen, X.; Duan, C.; Zhu, X.; You, X.; Raj, S. S. S.; Fun, H.; Wu, J. Mater. Chem. Phys. 2001, 72, 11–15. Takada, N.; Sugiyama, J.; Minami, N.; Hieda, S. Mol. Cryst. Liq. Cryst. 1997, 295, 71–74. Takada, N.; Sugiyama, J.; Katoh, R.; Minami, N.; Hieda, S. Synth. Met. 1997, 91, 351–354. Takada, N.; Hieda, S.; Sugiyama, J.; Katoh, R.; Minami, N. Synth. Met. 2000, 111, 587–590. Takada, N.; Hieda, S.; Sugiyama, J.; Katoh, R.; Minami, N. Mol. Cryst. Liq. Cryst. 1998, 315, 269–274. Chen, X.; Zhu, X.; Xu, Y.; Raj, S. S. S.; Öztürk, S.; Fun, H.; Ma, J.; You, X. J. Mater. Chem. 1999, 9, 2919–2922. Li, X.; Zheng, Y.; Zuo, J.; Song, Y.; You, X. Polyhedron 2007, 26, 5257–5262. Li, D.; Li, C.; Wang, J.; Kang, L.; Wu, T.; Li, Y.; You, X. Eur. J. Inorg. Chem. 2009, 4844–4849. Cui, M.; Wang, A.; Gao, C.; Zhou, L.; Wu, T.; Fang, S.; Xiao, H.; Li, F.; Li, X. Dalton Trans. 2021, 50, 1007–1018. Wong, H.-Y.; Chan, W. T. K.; Law, G.-L. Molecules 2019, 24, 662. Bryleva, Y. A.; Artem’ev, A. V.; Glinskaya, M. I.; Rakhmanova, L. A.; Samsonenko, D. G.; Komarov, V. Y.; Rogovoy, M. I.; Davydova, M. P. New J. Chem. 2021, 45, 13869–13876. Rausch, J.; Lorenz, V.; Hrib, C. G.; Frettloh, V.; Adlung, M.; Wickleder, C.; Hilfert, L.; Jones, P. G.; Edelmann, F. T. Inorg. Chem. 2014, 53, 11662–11674. George, T. M.; Sajan, M. J.; Gopakumar, N.; Reddy, M. L. P. J. Photochem. Photobiol. A 2016, 317, 88–99. Hasegawa, Y.; Hieda, R.; Miyata, K.; Nakagawa, T.; Kawai, T. Eur. J. Inorg. Chem. 2011, 4978–4984. Hasegawa, Y.; Tateno, S.; Yamamoto, M.; Nakanishi, T.; Kitagawa, Y.; Seki, T.; Ito, H.; Fushimi, K. Chem. Eur. J. 2017, 23, 2666–2672. Hirai, Y.; Nakanishi, T.; Kitagawa, Y.; Fushimi, K.; Seki, T.; Ito, H.; Hasegawa, Y. Angew. Chem. Int. Ed. 2017, 56, 7171–7175. Hirai, Y.; Rosa, P. P. F.; Nakanishi, T.; Kitagawa, Y.; Fushimi, K.; Seki, T.; Ito, H.; Hasegawa, Y. Inorg. Chem. 2018, 57, 14653–14659. Biju, S.; Gopakumar, N.; Bunzli, J. C. G.; Scopelliti, R.; Kim, H. K.; Reddy, M. L. P. Inorg. Chem. 2013, 52, 8750–8758. Mikhalyova, E. A.; Yakovenko, A. V.; Zeller, M.; Kiskin, M. A.; Kolomzarov, Y. V.; Eremenko, I. L.; Addison, A. W.; Pavlishchuk, V. V. Inorg. Chem. 2015, 54, 3125–3133. Bazhin, D. N.; Kudyakova, Y. S.; Bogomyakov, A. S.; Slepukhin, P. A.; Kim, G. A.; Burgart, Y. V.; Saloutin, V. I. Inorg. Chem. Front. 2019, 6, 40–49. Chandra, B. P.; Jaiswal, A. K.; Cwandraker, T. R.; Kaza, B. R. J. Lumin. 1982, 27, 101–107. Chandra, B. P.; Deshmukh, N. G.; Sahu, R. B.; Verma, A. K. Cryst. Res. Technol. 1986, 21, 1559–1565. Balsamy, S.; Natarajan, P.; Vedalakshmi, R.; Muralidharan, S. Inorg. Chem. 2014, 53, 6054–6059. Chen, J.; Zhang, Q.; Zheng, F.; Liu, Z.; Wang, S.; Wu, A.; Guo, G. Dalton Trans. 2015, 44, 3289–3294. Artem’ev, A. V.; Davydova, M. P.; Rakhmanova, M. I.; Bagryanskaya, I. Y.; Pishchur, D. P. Inorg. Chem. Front. 2021, 8, 3767–3774. Artem’ev, A. V.; Davydova, M. P.; Berezin, A. S.; Sukhikh, T. S.; Samsonenko, D. G. Inorg. Chem. Front. 2021, 8, 2261–2270. Knotter, D. M.; Maanen, H. L.; Grove, D. M.; Spek, A. L.; Koten, G. V. Inorg. Chem. 1991, 30, 3309–3317. Knotter, D. M.; Blasse, G.; Vliet, J. P. M. V.; Koten, G. V. Inorg. Chem. 1992, 31, 2196–2201. _lncel, A.; Varlikli, C.; McMillen, C. D.; Demir, M. M. J. Phys. Chem. C 2017, 121, 11709–11716. Karimata, A.; Patil, P. H.; Fayzullin, R. R.; Khaskin, E.; Lapointe, S.; Khusnutdinov, J. R. Chem. Sci. 2020, 11, 10814–10820. Hsu, C. W.; Ly, K. T.; Lee, W. K.; Wu, C. C.; Wu, L. C.; Lee, J. J.; Lin, T. C.; Liu, S. H.; Chou, P. T.; Lee, G. H.; Chi, Y. ACS Appl. Mater. Interfaces 2016, 8, 33888–33898. Sharipov, G. L.; Tukhbatullin, A. A. J. Lumin. 2019, 215, 116691. Jeong, S. M.; Song, S.; Kim, H.; Joo, K.-I.; Takezoe, H. Adv. Funct. Mater. 2016, 26, 4848–4858. Patel, D. K.; Cohen, B.-E.; Etgar, L.; Magdassi, S. Mater. Horiz. 2018, 5, 708–714. Park, H. J.; Kim, S.; Lee, J. H.; Kim, H. T.; Seung, W.; Son, Y.; Kim, T. Y.; Khan, U.; Park, N. M.; Kim, S. W. ACS Appl. Mater. Interfaces 2019, 11, 5200–5207.

Organometallic Complexes for Optoelectronic Applications 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167.

381

Zhang, J.; Bao, L.; Lou, H.; Deng, J.; Chen, A.; Hu, Y.; Zhang, Z.; Sun, X.; Peng, H. J. Mater. Chem. C 2017, 5, 8027–8032. Liang, G.; Ruan, Z.; Liu, Z.; Li, H.; Wang, Z.; Tang, Z.; Mo, F.; Yang, Q.; Ma, L.; Wang, D.; Zhi, C. Adv. Electron. Mater. 2019, 5, 1900553. Kim, Y.; Kim, J. S.; Kim, G. W. Sci. Rep. 2018, 8, 12023. Jang, J.; Kim, H.; Ji, S.; Kim, H. J.; Kang, M. S.; Kim, T. S.; Won, J. E.; Lee, J. H.; Cheon, J.; Kang, K.; Im, W. B.; Park, J. U. Nano Lett. 2020, 20, 66–74. McConnell, A. J.; Wood, C. S.; Neelakandan, P. P.; Nitschke, J. R. Chem. Rev. 2015, 115, 7729–7793. Chi, Z. G.; Zhang, X. Q.; Xu, B. J.; Zhou, X.; Ma, C. P.; Zhang, Y.; Liu, S. W.; Xu, J. R. Chem. Soc. Rev. 2012, 41, 3878–3896. Dong, Y. Q.; Lam, J. W.; Tang, B. Z. J. Phys. Chem. Lett. 2015, 6, 3429–3436. Wang, L.; Ye, K. Q.; Zhang, H. Y. Chin. Chem. Lett. 2016, 27, 1367–1375. Wang, C.; Li, Z. Mater. Chem. Front. 2017, 1, 2174–2194. Yang, J.; Fang, M.; Li, Z. Acc. Mater. Res. 2021, 2, 644–654. Anthony, S. P. ChemPlusChem 2012, 77, 518–531. Misra, R.; Jadhav, T.; Dhokale, B.; Mobin, S. M. Chem. Commun. 2014, 50, 9076–9078. Seki, T.; Ito, H. Chem. Eur. J. 2016, 22, 4322–4329. Zhang, J.; He, B.; Hu, Y.; Alam, P.; Zhang, H.; Lam, J. W. Y.; Tang, B. Z. Adv. Mater. 2021, 33, 2008071. Yang, Z.; Chi, Z.; Mao, Z.; Zhang, Y.; Liu, S.; Zhao, J.; Aldred, M. P.; Chi, Z. Mater. Chem. Front. 2018, 2, 861–890. Ruan, Z.; Shan, Y.; Gong, Y.; Wang, C.; Ye, F.; Qiu, Y.; Liang, Z.; Li, Z. J. Mater. Chem. C 2018, 6, 773–780. Dong, Y.; Zhang, J.; Li, A.; Gong, J.; He, B.; Xu, S.; Yin, J.; Liu, S. H.; Tang, B. Z. J. Mater. Chem. C 2020, 8, 894–899. Zhang, X.; Chi, Z.; Zhang, Y.; Liu, S.; Xu, J. J. Mater. Chem. C 2013, 1, 3376–3390. Xue, P.; Ding, J.; Wang, P.; Lu, R. J. Mater. Chem. C 2016, 4, 6688–6706. Yusoff, A. R. B. M.; Huckaba, A. J.; Nazeeruddin, M. K. Top. Curr. Chem. 2017, 375, 39. Dixon, I. M.; Collin, J.-P.; Sauvage, J.-P.; Flamigni, L.; Encinas, S.; Barigelletti, F. Chem. Soc. Rev. 2000, 29, 385–391. Cho, Y. J.; Lee, J. Y. Adv. Mater. 2011, 23, 4568–4572. Fan, C.; Yang, C. Chem. Soc. Rev. 2014, 43, 6439–6469. Zhang, Q.-C.; Xiao, H.; Zhang, X.; Xu, L.-J.; Chen, Z.-N. Coordin. Chem. Rev. 2019, 378, 121–133. Szerb, E.; Talarico, A.; Aiello, I.; Crispini, A.; Godbert, N.; Pucci, D.; Pugliese, T.; Ghedini, M. Eur. J. Inorg. Chem. 2010, 3270–3277. Shan, G.; Li, H.; Cao, H.; Zhu, D.; Li, P.; Su, Z.; Liao, Y. Chem. Commun. 2012, 48, 2000–2002. Shan, G.; Li, H.; Zhu, D.; Su, Z.; Liao, Y. J. Mater. Chem. 2012, 22, 12736–12744. Mastropietro, T. F.; Yadav, Y. J.; Szerb, E. I.; Talarico, A. M.; Ghedini, M.; Crispini, A. Dalton Trans. 2012, 41, 8899–8907. Shan, G. G.; Li, H. B.; Qin, J. S.; Zhu, D. X.; Liao, Y.; Su, Z. M. Dalton Trans. 2012, 41, 9590–9593. Shan, G. G.; Li, H. B.; Sun, H. Z.; Zhu, D. X.; Cao, H. T.; Su, Z. M. J. Mater. Chem. C 2013, 1, 1440–1449. Sun, H.; Liu, S.; Lin, W.; Zhang, K. Y.; Lv, W.; Huang, X.; Huo, F.; Yang, H.; Jenkins, G.; Zhao, Q.; Huang, W. Nat. Commun. 2015, 5, 3601. Han, Y.; Cao, H. T.; Sun, H. Z.; Wu, Y.; Shan, G. G.; Su, Z. M.; Hou, X. G.; Liao, Y. J. Mater. Chem. C 2014, 2, 7648–7655. Li, G.; Ren, X.; Shan, G.; Che, W.; Zhu, D.; Yan, L. I.; Su, Z.; Bryce, M. R. Chem. Commun. 2015, 51, 13036–13039. Jiang, Y.; Li, D.; Zhu, G.; Su, Z.; Bryce, M. R. J. Mater. Chem. C 2017, 5, 12189–12193. Xie, J.; Li, D.; Duan, Y.; Geng, Y.; Yang, T.; Li, G.; Zhu, D.; Su, Z. Dyes Pigments 2020, 172, 107855. Wang, Y.; Yang, T.; Liu, X.; Li, G.; Che, W.; Zhu, D.; Su, Z. J. Mater. Chem. C 2018, 6, 12217–12223. Yang, T.; Wang, Y.; Liu, X.; Li, G.; Che, W.; Zhu, D.; Su, Z.; Bryce, M. R. Chem. Commun. 2019, 55, 14582–14585. Zhao, K.-Y.; Song, W.-L.; Yang, Y.; Duan, Y.-C.; Li, G.-F.; Shan, G.-G.; Fu, Q.; Su, Z.-M. Dyes Pigments 2021, 184, 108817. Li, Y.; Zhang, Q.; Zhang, L.; Ye, Y.; Zhang, R.; Gu, X.; Liu, R.; Zhu, H. J. Organomet. Chem. 2020, 930, 121595. Pei, Y.; Xie, J.; Cui, D.; Liu, S.; Li, G.; Zhu, D.; Su, Z. Dalton Trans. 2020, 49, 13066–13071. Li, D.; Li, G.; Xie, J.; Zhu, D.; Su, Z.; Bryce, M. R. J. Mater. Chem. C 2019, 7, 10876–10880. Han, J.; Tang, K.-M.; Cheng, S.-C.; Ng, C.-O.; Chun, Y.-K.; Chan, S.-L.; Yiu, S.-M.; Tse, M.-K.; Roy, V. A. L.; Ko, C.-C. Inorg. Chem. Front. 2020, 7, 786–794. Li, K.; Tong, G. S. M.; Wan, Q.; Cheng, G.; Tong, W.-Y.; Ang, W.-H.; Kwong, W.-L.; Che, C.-M. Chem. Sci. 2016, 7, 1653–1673. Pashaei, B.; Karimi, S.; Shahroosvand, H.; Abbasi, P.; Pilkington, M.; Bartolotta, A.; Fresta, E.; Fernandez-Cestau, J.; Costa, R. D.; Bonaccorso, F. Chem. Soc. Rev. 2019, 48, 5033–5139. Lo, S.-C.; Burn, P. L. Chem. Rev. 2007, 107, 1097–1116. Kozhevnikov, V.; Donnio, B.; Bruce, D. Angew. Chem. Int. Ed. 2008, 47, 6286–6289. Abe, T.; Itakura, T.; Ikeda, N.; Shinozaki, K. Dalton Trans. 2009, 711–715. Choi, S.; Kuwabara, J.; Nishimura, Y.; Arai, T.; Kanbara, T. Chem. Lett. 2012, 41, 65–67. Ni, J.; Zhang, X.; Qiu, N.; Wu, Y.; Zhang, L.; Zhang, J.; Chen, Z. Inorg. Chem. 2011, 50, 9090–9096. Ni, J.; Zhang, X.; Wu, Y.; Zhang, L.; Chen, Z. Chem. Eur. J. 2011, 17, 1171–1183. Zhang, X.; Wang, J.; Ni, J.; Zhang, L.; Chen, Z. Inorg. Chem. 2012, 51, 5569–5579. Ni, J.; Wang, Y. G.; Wang, H. H.; Xu, L.; Zhao, Y. Q.; Pan, Y. Z.; Zhang, J. J. Dalton Trans. 2014, 43, 352–360. Kitani, N.; Kuwamura, N.; Tsukuda, T.; Yoshinari, N.; Konno, T. Chem. Commun. 2014, 50, 13529–13532. Zhang, X.; Mei, J.; Lai, J.; Li, C.; You, X. J. Mater. Chem. C 2015, 3, 2350–2357. Zhang, X.; Zhu, L.; Wang, X.; Shi, Z.; Lin, Q. Inorg. Chim. Acta 2016, 442, 56–63. Zhang, X.-P.; Zhang, D.-S.; Qi, X.-W.; Zhu, L.-H.; Wang, X.-H.; Sun, W.; Shi, Z.-F.; Lin, Q. Inorg. Chim. Acta 2017, 467, 99–105. Yang, Q.-Y.; Zhang, H.-H.; Qi, X.-W.; Sun, S.-S.; Zhang, D.-S.; Han, L.-Z.; Zhang, X.-P.; Shi, Z.-F. Dalton Trans. 2021, 50, 8938–8946. Zhang, H.-H.; Yang, Q.-Y.; Qi, X.-W.; Sun, S.-S.; Li, B.-S.; Zhang, D.-S.; Zhang, X.-P.; Shi, Z.-F. Inorg. Chim. Acta 2021, 523, 120411. Yang, Q.-Y.; Zhang, H.-H.; Xu, G.; Diao, Z.-X.; Zhou, Z.-M.; Zhang, D.-S.; Chen, X.-H.; Zhang, X.-P.; Shi, Z.-F. Inorg. Chim. Acta 2021, 515, 120072. Ohno, K.; Yamaguchi, A. S.; Nagasawaa, A. A.; Fujihara, T. Dalton Trans. 2016, 45, 5492–5503. Geng, H.; Luo, K.; Zou, G.; Wang, H.; Ni, H.; Yu, W.; Li, Q.; Wang, Y. New J. Chem. 2016, 40, 10371–10377. Kang, J.-J.; Ni, J.; Li, Y.-Q. Chinese J. Struct. Chem. 2020, 39, 140–146. Zhu, S.; Hu, J.; Zhai, S.; Wang, Y.; Xu, Z.; Liu, R.; Zhu, H. Inorg. Chem. Front. 2020, 7, 4677–4686. Lin, C.-J.; Liu, Y.-H.; Peng, S.-M.; Shinmyozu, T.; Yang, J.-S. Inorg. Chem. 2017, 56, 4978–4989. Lien, C.-Y.; Hsu, Y.-F.; Liu, Y.-H.; Peng, S.-M.; Shinmyozu, T.; Yang, J.-S. Inorg. Chem. 2020, 59, 11584–11594. Junquera, M. M.; Lara, R.; Lalinde, E.; Moreno, M. T. J. Mater. Chem. C 2020, 8, 7221–7233. Perruchas, S.; Goff, X. L.; Maron, S.; Maurin, I.; Guillen, F.; Garcia, A.; Gacoin, T.; Boilot, J. J. Am. Chem. Soc. 2010, 132, 10967–10969. Benito, Q.; Goff, X. F. L.; Maron, S.; Fargues, A.; Garcia, A.; Martineau, C.; Taulelle, F.; Kahlal, S.; Gacoin, T.; Boilot, J.-P.; Perruchas, S. J. Am. Chem. Soc. 2014, 136, 11311–11320. Benito, Q.; Maurin, I.; Cheisson, T.; Nocton, G.; Fargues, A.; Garcia, A.; Martineau, C.; Gacoin, T.; Boilot, J.-P.; Perruchas, S. Chem. Eur. J. 2015, 21, 5892–5897. Benito, Q.; Maurin, I.; Poggi, M.; Martineau-Corcos, C.; Gacoin, T.; Boilot, J.-P.; Perruchas, S. J. Mater. Chem. C 2016, 4, 11231–11237. Xiao, Q.; Zheng, J.; Li, M.; Zhan, S.; Wang, J.; Li, D. Inorg. Chem. 2014, 53, 11604–11615.

382 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. 180. 181. 182. 183. 184. 185. 186. 187. 188. 189. 190. 191. 192. 193. 194. 195. 196. 197. 198. 199. 200. 201. 202. 203. 204. 205. 206. 207. 208. 209. 210. 211. 212. 213. 214. 215. 216. 217. 218. 219. 220. 221. 222. 223. 224. 225. 226. 227. 228. 229. 230. 231. 232. 233.

Organometallic Complexes for Optoelectronic Applications Wen, T.; Zhang, D.; Liu, J.; Lin, R.; Zhang, J. Chem. Commun. 2013, 49, 5660–5662. Wen, T.; Zhou, X.-P.; Zhang, D.-X.; Li, D. Chem. Eur. J. 2014, 20, 644–648. Wen, T.; Zhang, D.-X.; Zhang, H.-X.; Zhang, H.-B.; Zhang, J.; Li, D.-S. Chem. Commun. 2014, 50, 8754–8756. Wen, T.; Zhang, D.; Liu, J.; Zhang, H.; Zhang, J. Chem. Commun. 2015, 51, 1353–1355. Zhang, D.-X.; Zhang, H.-X.; Wen, T.; Li, D.-S.; Zhang, J. Inorg. Chem. Front. 2016, 3, 263–267. Deshmukh, M. S.; Yadav, A.; Pant, R.; Boomishankar, R. Inorg. Chem. 2015, 54, 1337–1345. Zhang, X.; Wang, J.-Y.; Qiao, D.; Chen, Z.-N. J. Mater. Chem. C 2017, 5, 8782–8787. Hupp, B.; Nitsch, J.; Schmitt, T.; Bertermann, R.; Edkins, K.; Hirsch, F.; Fischer, I.; Auth, M.; Sperlich, A.; Steffen, A. Angew. Chem. Int. Ed. 2018, 57, 13671–13675. Peng, D.; He, L.-H.; Ju, P.; Chen, J.-L.; Ye, H.-Y.; Wang, J.-Y.; Liu, S.-J.; Wen, H.-R. Inorg. Chem. 2020, 59, 17213–17223. Chen, X.-W.; He, L.-H.; Ju, P.; Chen, J.-L.; Liu, S.-J.; Wen, H.-R. J. Mater. Chem. C 2020, 8, 16160–16167. Yu, X.; Li, X.; Cai, Z.; Sun, L.; Wang, C.; Rao, H.; Wei, C.; Bian, Z.; Jin, Q.; Liu, Z. Chem. Commun. 2021, 57, 5082–5085. Tsukuda, T.; Kawase, M.; Dairiki, A.; Matsumoto, K.; Tsubomura, T. Chem. Commun. 2010, 46, 1905–1907. Babashkina, M.; Safin, D.; Bolte, M.; Garcia, Y. Dalton Trans. 2011, 40, 8523–8526. Liu, L.; Wen, T.; Fu, W.; Liu, M.; Chen, S.; Zhang, J. CrystEngComm 2016, 18, 218–221. Kakizoe, D.; Nishikawa, M.; Degawa, T.; Tsubomura, T. Inorg. Chem. Front. 2016, 3, 1381–1387. Li, S.-L.; Han, M.; Wu, B.; Wang, J.; Zhang, F.-Q.; Zhang, X.-M. Sci. Rep. 2017, 7, 13058. Ma, X.-H.; Wang, J.-Y.; Guo, J.-J.; Wang, Z.-Y.; Zang, S.-Q. Chin. J. Chem. 2019, 37, 1120–1124. Ito, H.; Saito, T.; Oshima, N.; Kitamura, N.; Ishizaka, S.; Hinatsu, Y.; Wakeshima, M.; Kato, M.; Tsuge, K.; Sawamura, M. J. Am. Chem. Soc. 2008, 130, 10044–10045. Osawa, M.; Kawata, I.; Igawa, S.; Hoshino, M.; Fukunaga, T.; Hashizume, D. Chem. Eur. J. 2010, 16, 12114–12126. Liang, J.; Hu, F.; Lv, X.; Chen, Z.; Chen, Z.; Yin, J.; Yu, G.; Liu, S. Dyes Pigments 2012, 95, 485–490. Chen, Z.; Liang, J.; Nie, Y.; Xu, X.; Yu, G.; Yin, J.; Liu, S. H. Dalton Trans. 2015, 44, 17473–17477. Chen, Z.; Yang, L.; Hu, Y.; Wu, D.; Yin, J.; Yu, G.-A.; Liu, S. H. RSC Adv. 2015, 5, 93757–93764. Chen, Z.; Zhang, J.; Song, M.; Yin, J.; Yu, G.-A.; Liu, S. H. Chem. Commun. 2015, 51, 326–329. Chen, Z.; Nie, Y.; Liu, S. H. RSC Adv. 2016, 6, 73933–73938. Chen, Z.; Liu, G.; Pu, S.; Liu, S. H. Dyes Pigments 2017, 143, 409–415. Chen, Z.; Liu, G.; Wang, R.; Pu, S. RSC Adv. 2017, 7, 15112–15115. Chen, Z.; Li, Z.; Hu, F.; Yu, G.-A.; Yin, J.; Liu, S. H. Dyes Pigments 2016, 125, 169–178. Chen, Z.; Huang, P.-S.; Li, Z.; Yin, J.; Yu, G.-A.; Liu, S. H. Inorg. Chim. Acta 2015, 432, 192–197. Wang, X.-Y.; Yin, Y.; Yin, J.; Chen, Z.; Liu, S. H. Dalton Trans. 2021, 50, 7744–7749. Kawaguchi, K.; Seki, T.; Karatsu, T.; Kitamura, A.; Ito, H.; Yagai, S. Chem. Commun. 2013, 49, 11391–11393. Ito, H.; Muromoto, M.; Kurenuma, S.; Ishizaka, S.; Kitamura, N.; Sato, H.; Seki, T. Nat. Commun. 2009, 2013, 4. Seki, T.; Sakurada, K.; Ito, H. Angew. Chem. Int. Ed. 2013, 52, 12828–12832. Seki, T.; Sakurada, K.; Ito, H. Chem. Commun. 2015, 51, 13933–13936. Seki, T.; Ozaki, T.; Okura, T.; Asakura, K.; Sakon, A.; Uekusa, H.; Ito, H. Chem. Sci. 2015, 6, 2187–2195. Yagai, S.; Seki, T.; Aonuma, H.; Kawaguchi, K.; Karatsu, T.; Okura, T.; Sakon, A.; Uekusa, H.; Ito, H. Chem. Mater. 2016, 28, 234–241. Seki, T.; Takamatsu, Y.; Ito, H. J. Am. Chem. Soc. 2016, 138, 6252–6260. Jin, M.; Seki, T.; Ito, H. Chem. Commun. 2016, 52, 8083–8086. Seki, T.; Kobayashi, K.; Ito, H. Chem. Commun. 2017, 53, 6700–6703. Seki, T.; Kobayashi, K.; Mashimo, T.; Ito, H. Chem. Commun. 2018, 54, 11136–11139. Seki, T.; Ida, K.; Ito, H. Mater. Chem. Front. 2018, 2, 1195–1200. Seki, T.; Kashiyamaa, K.; Ito, H. Dalton Trans. 2019, 48, 7105–7109. Zhang, J.; He, B.; Wu, W.; Alam, P.; Zhang, H.; Gong, J.; Song, F.; Wang, Z.; Sung, H. H. Y.; Williams, I. D.; Wang, Z.; Lam, J. W. Y.; Tang, B. Z. J. Am. Chem. Soc. 2020, 142, 14608–14618. Jobbágy, C.; Molnár, M.; Baranyai, P.; Hamza, A.; Pálinkása, G.; Deák, A. CrystEngComm 2014, 16, 3192–3202. Jobbágy, C.; Molnár, M.; Baranyai, P.; Deák, A. Dalton Trans. 2014, 43, 11807–11810. Deák, A.; Jobbágy, C.; Marsi, G.; Molnár, M.; Szakács, Z.; Baranyai, P. Chem. Eur. J. 2015, 21, 11495–11508. Wong, B. Y.-W.; Wong, H.-L.; Wong, Y.-C.; Au, V. K.-M.; Chan, M.-Y.; Yam, V. W.-W. Chem. Sci. 2017, 8, 6936–6946. Liu, Q.; Xie, M.; Chang, X.; Gao, Q.; Chen, Y.; Lu, W. Chem. Commun. 2018, 54, 12844–12847. Luong, L. M. C.; Olmstead, M. M.; Balch, A. L. Chem. Commun. 2021, 57, 793–796. Chen, K.; Nenzel, M. M.; Brown, T. M.; Catalano, V. J. Inorg. Chem. 2015, 54, 6900–6909. Conejo-Rodríguez, V.; Peñas-Defrutos, M. N.; Espinet, P. Dalton Trans. 2019, 48, 10412–10416. Mizukami, S.; Houjou, H.; Sugaya, K.; Koyama, E.; Tokuhisa, H.; Sasaki, T.; Kanesato, M. Chem. Mater. 2005, 17, 50–56. Xu, B.; Chi, Z.; Zhang, X.; Li, H.; Chen, C.; Liu, S.; Zhang, Y.; Xu, J. Chem. Commun. 2011, 47, 11080–11082. Zheng, H.-W.; Wu, M.; Yang, D.; Liang, Q.-F.; Li, J.-B.; Zheng, X.-J. Inorg. Chem. 2021, 60 (15), 11609–11615. Song, X.; Yu, H.; Yan, X.; Zhang, Y.; Miao, Y.; Ye, K.; Wang, Y. Dalton Trans. 2018, 47, 6146–6155. Yan, X.; Song, X.; Mu, X.; Wang, Y. New J. Chem. 2019, 43, 15886–15891. Qiao, D.; Wang, J.-Y.; Zhang, L.-Y.; Dai, F.-R.; Chen, Z.-N. Dalton Trans. 2019, 48, 11045–11051. Li, S.; Wu, M.; Kang, Y.; Zheng, H.-W.; Zheng, X.-J.; Fang, D.-C.; Jin, L.-P. Inorg. Chem. 2019, 58, 4626–4633. Tzeng, B.; Chang, T.; Sheu, H. Chem. Eur. J. 2010, 16, 9990–9993. Tzeng, B.; Chang, T.; Wei, S.; Sheu, H. Chem. Eur. J. 2012, 18, 5105–5112. Zou, R.; Zhang, J.; Hu, S.; Hu, F.; Zhang, H.; Fu, Z. CrystEngComm 2017, 19, 6259–6262. Yan, Y.; Chen, J.; Zhang, N.-N.; Wang, M.-S.; Sun, C.; Xing, X.-S.; Li, R.; Xu, J.-G.; Zheng, F.-K.; Guo, G.-C. Dalton Trans. 2016, 45, 18074–18078. Dey, A.; Garai, A.; Gude, V.; Biradha, K. Cryst. Growth Des. 2018, 18, 6070–6077. Yang, Y.; Fang, X.; Zhao, S.-S.; Bai, F.; Zhao, Z.; Wang, K.-Z.; Yan, D. Chem. Commun. 2020, 56, 5267–5270. Wen, T.; Zheng, Y.; Xu, C.; Zhang, J.; Jaroniec, M.; Qiao, S.-Z. Mater. Horiz. 2018, 5, 1151–1155. Bi, H.; Chen, D.; Li, D.; Yuan, Y.; Xia, D.; Zhang, Z.; Zhang, H.; Wang, Y. Chem. Commun. 2011, 47, 4135–4137. Calupitan, J. P.; Poirot, A.; Wang, J.; Delavaux-Nicot, B.; Wolff, M.; Jaworska, M.; Métivier, R.; Benoist, E.; Allain, C.; Fery-Forgues, S. Chem. Eur. J. 2021, 27, 4191–4196.

14.11

Organometallic Lanthanide Complexes as Single Molecule Magnets

Bijoy Dey and Vadapalli Chandrasekhara,b, aTata Institute of Fundamental Research Hyderabad, Hyderabad, India; bDepartment of Chemistry, Indian Institute of Technology Kanpur, Kanpur, India a

© 2022 Elsevier Ltd. All rights reserved.

14.11.1 Introduction 14.11.2 h4-Cyclobutadienyl ligand based lanthanide complexes as SMMs 14.11.3 h5-Cyclopentadienyl ligand based lanthanide complexes as SMMs 14.11.3.1 General remarks 14.11.3.2 Z5-Cyclopentadienyl ligated lanthanide complexes with competing equatorial ligands 14.11.3.3 Cationic metallocene SMMs 14.11.3.4 Dysprosium metallocene complexes containing bridging borohydride ligands 14.11.3.5 Lanthanide metallocenes containing bridging radical ligands 14.11.4 h5-Dicarbollide dianion based dysprosiacarborane SMMs 14.11.5 h6-Arene ligated dysprosium complexes as SMMs 14.11.6 h7-Cycloheptatrienyl ligand based lanthanide complexes as SMMs 14.11.7 h8-Cyclooctatetraenyl ligand based lanthanide complexes as SMMs 14.11.7.1 [CpLn(COT)] and [Ln(COT)2]− motif-based complexes 14.11.7.2 COT-ligated lanthanide complexes with heteroaromatic ligands 14.11.8 Methanide and bismethane(diide) ligand based lanthanide complexes as SMMs 14.11.9 Conclusions Acknowledgments References

383 388 391 391 391 396 400 401 404 405 407 408 408 413 414 416 416 416

14.11.1 Introduction Single molecule magnets (SMMs) are coordination complexes which when magnetized retain the magnetization indefinitely below a certain critical temperature. This field has its origin to the discovery of slow relaxation of magnetization in the manganese complex, [Mn12O12(OAc)16(H2O)4] (Mn12Ac).1 Mn12Ac showed slow relaxation of magnetization and has a ground state spin of S ¼ 10 and an Ising type magnetic anisotropy characterized by D (axial zero field splitting parameter) value of −0.5 cm−1. Analysis of these results led to a qualitative prescription for the construction of SMMs whereby a large ground state spin and a large value of magnetic anisotropy were deemed to be necessary which would lead to a bistable ground state (Fig. 1) where upon magnetization one of them is effectively populated and after removal of the field, thermal equilibration is achieved (relaxation). The energy barrier in the resulting double-well for a transition metal based SMMs can be qualitatively estimated as Ueff ¼ |D|S2 for even spin systems and Ueff ¼ | D |(S2–1/4) for odd spin systems.2 This prescription led to a frantic search for analogous properties among polynuclear transition metal complexes.3 The hope was that an appropriate structural arrangement of the paramagnetic ions within such complexes would through super exchange result in a large ground state spin and if the anisotropy axes are favorably aligned would also result in an overall magnetic anisotropy. Very soon it was realized that this recipe for preparing SMMs was an oversimplification. Firstly, synthesis of polynuclear transition metal complexes of appropriate size and topology was more by a serendipity-based approach rather than by design.3 Secondly even in such complexes, control of anisotropy was far more challenging than anticipated.4 Thirdly, it was also shown that increase of S is accompanied by a concomitant decrease in D.5 Experimentally, it was verified that possessing a large ground state spin alone was not going to be sufficient. Thus, the compound, [MnIII12MnII7(m4-O)8(m3,Z1-N3)8(HL)12(MeCN)6]2+ (H3L ¼ 2,6-bis(hydroxymethyl)4-methylphenol) which was shown to possess an exceptionally large ground state spin of 83/2 was not a SMM.6 However, in this complex replacement of the central MnIII by DyIII resulted in SMM property which was rationalized as being due to the magnetic anisotropy generated by DyIII.7 These results emphasized the importance of the presence of an Ising type of anisotropy in the complexes for achieving single molecule magnet behavior. While most of these investigations were on polynuclear transition metal complexes, Ishikawa and co-workers in a path breaking paper revealed that phthalocyanine sandwiched complexes,8 [Pc2Ln]−‚TBA+ (Ln ¼ Tb, Dy, TBA ¼ tetrabutylammonium) showed slow relaxation of magnetization even though only one metal center is present in the complex. The coordination environment around the metal ion in these complexes corresponded to an approximate D4d crystal field environment (Fig. 2). This discovery opened the door for exploration of lanthanide complexes in the field of SMMs and single ion magnets (SIMs) which are a subset of the former (In the following we will use the general term SMMs for both of these classes of compounds). During this journey the many desirable properties of lanthanide ions soon became apparent. Firstly, many lanthanide ions have unfilled f-shells and therefore can contribute to the ground-state spin.9 Secondly, because of weak crystal field effects, even in their complexes, lanthanide ions possess a large unquenched orbital angular momentum and this results in strong spin-orbit coupling and hence to magnetic anisotropy.10 Thirdly, among lanthanide ions those such as DyIII, ErIII are the so-called Kramers ions, because

Comprehensive Organometallic Chemistry IV

https://doi.org/10.1016/B978-0-12-820206-7.00153-0

383

384

Organometallic Lanthanide Complexes as Single Molecule Magnets

Fig. 1 Double well potential energy diagram for a system with spin state S with easy axis magnetic anisotropy where the –Ms levels are localized in the left well and + Ms. levels are localized in right well. (A) Equally populated potential wells, (B) With application of magnetic field left well is stabilized preferentially, (C) After removing the field the energy levels achieve thermal equilibrium via various processes, (D) Various relaxation processes.

Fig. 2 Molecular structure of complex [Tb(Pc)2]–, color codes: gray, C; brown, Tb, blue, N. Adapted from Ref. McAdams, S. G.; Ariciu, A. M.; Kostopoulos, A. K.; Walsh, J. P.S.; Tuna, F. Coord. Chem. Rev. 2017, 346, 216–239 with permission from Elsevier.

Table 1

Commonly used LnIII ions in the assembly of SMMs and SIMs.

4fn Spin-orbit ground term Free ion g-value

Tb3+

Dy3+

Ho3+

Er3+

4f8 7 F6 3/2

4f9 H15/2 4/3

4f10 I8 5/4

4f11 I15/2 6/5

6

5

4

of the presence of an odd number of electrons. Such Kramers lanthanide ions have degenerate mJ states even in the absence of a magnetic field11 which is an advantage in the design of single molecule magnets. The electronic properties of some commonly used lanthanide ions in the assembly of SMMs and SIMs are given in Table 1. These factors have led to a renaissance of lanthanide coordination complexes in molecular magnetism.12,13 Many heterometallic complexes based on the combination of 3d and 4f ions

Organometallic Lanthanide Complexes as Single Molecule Magnets

385

also have been synthesized and studied.14–17 The effective energy barriers of these complexes, however, lag behind those of pure lanthanide complex-based SMMs or SIMs.18,19 Research in lanthanide complexes as SMMs/SIMs received a further fillip by thoughtful insights provided by researchers. Thus the f-electron densities of lanthanide ions fall into two categories, oblate and prolate. An insightful review by Long and co-workers20 has suggested that to maximize the anisotropy for oblate ions strong axial crystal fields are important while for prolate ions strong equatorial field must be important. Studies by Ungur and Chibotaru21 on model DyIII complexes have revealed that in ideal two-coordinate linear lanthanide complexes the axial effects are best imposed. Such systems although accessible and amenable for study in silico are difficult to synthesize. An alternate synthetic paradigm could involve the design of sandwich complexes where aromatic planar ligands sandwich the lanthanide ions such that DyIII does not have any equatorial ligands. This concept has led to the field of organometallic lanthanide complexes with applications in molecular magnetism. In fact the two most spectacular examples in this field are [(Cpttt)2Dy][B(C6F5)4] and [(Z5-Cp )Dy(Z5-CpiPr5)][B(C6F5)4] with blocking of magnetization temperatures of 60 and 80 K respectively.22–24 Before we describe the status of this subject in organometallic lanthanide complexes as SMMs and SIMs, in the following we will briefly cover the various aspects of single-molecule magnets including the origin of magnetic anisotropy in lanthanides, the relaxation pathways that dissipate the magnetization, the characterization techniques and the quality check parameters. As mentioned above most lanthanide ions have a large unquenched orbital angular momentum. In fact, in the lanthanide ions, except those that possess the terms 1S0 and 1S7/2 all others possess the ground state electronic terms that have large and unquenched orbital contribution to the magnetic moment. Though the spin-orbit coupling is much stronger than crystal field effects in lanthanide complexes, the latter play a crucial role in determining the extent of magnetic anisotropy (Fig. 3). In lanthanide ions the spin orbit coupled ground term 2S+1LJ is split into (2J +1) microstates while in transition metal based SMMs the coupling of the spin angular momentum with orbital angular momentum is neglected.25 The mJ states of lanthanide ions are generally a linear combinations of mJ wave functions but in complexes possessing high symmetries (C1v, D1h etc.) they can be of pure nature. The fate of the magnetization retention in SMMs/SIMs depends on many factors. In most of the SMMs quantum tunnelling of magnetization (QTM) is one of the major relaxation mechanisms where the wave functions of the (+mJ) and (−mJ) ground state overlap with each other. In other words QTM occurs when there is strong transverse interactions coupling the two mJ levels. The transverse interaction causes the symmetric and anti-symmetric superposition of the two mJ states, where the energy difference (tunnel splitting, Dt) corresponds to the rate at which the system can tunnel from one to the other state (Fig. 4A). These transverse interactions can result because of many factors.

Fig. 3 Electronic interactions in lanthanide ions (example Tb3+) and their typical magnitude. Adapted from Ref. Liddle, S. T.; Slageren, J. V. Chem. Soc. Rev. 2015, 44, 6655–6669 with permission from Royal Society of Chemistry.

386

Organometallic Lanthanide Complexes as Single Molecule Magnets

Fig. 4 (A) Schematic view of how a transverse interaction (CF splitting, hyperfine interaction, magnetic field) can lead to superposition of states with opposite projections of the magnetic moment. (B) Various spin-lattice relaxation processes.

(a) low symmetry of the crystal field. (b) any effective transverse magnetic field. (c) hyperfine interactions with nuclear spins. The transverse field which is described by H⊥ ¼ g⊥mBB⊥Sbx,y will be very small if g⊥ is small where g⊥ is the perpendicular component of the g value. So, with low H⊥ the matrix element of the two opposite projections will be low and the tunnelling probability will be suppressed.25 Thus the requirement of a high axial g value serves as a good criterion for improving the SMM property. QTM is a temperature independent phenomenon but many times QTM can also occur from the excited higher energy state. In this case it is referred to as thermally assisted QTM. In addition to QTM which is generally a temperature independent relaxation process as described above, there are three main temperature dependent relaxation processes, Direct, Raman, and Orbach processes which are together called as spin-lattice or spin-phonon relaxation processes. The direct process involves the emission of a single phonon to flip the spin state. The Orbach process involves thermal excitation to a real state (+2) followed by an emission of phonon while the Raman process involves excitation to a virtual state (2 ) followed by the emission of a phonon (Fig. 4B).26 QTM and Orbach processes can occur in the same system at different temperatures. All of these relaxation processes can be expressed as a function of temperature according to Eq. (1).25   B1 DCF (1) + AHn1 T + CTn2 + t0 −1 exp − t −1 ¼ 2 1 + B2 H KB T Where, t ¼ relaxation time, B1 and B2 ¼ Wentzel-Kramers-Brillouin (WKB) exponent of the tunnelling rate, H ¼ applied magnetic field, A ¼ coefficient for direct relaxation related to the effect of phonons, n1 ¼ exponent for direct relaxation process, T ¼ temperature, n2 ¼ Raman exponent, C ¼ Raman coefficient related to the effect of phonons, t0 ¼ pre-exponential factor for Arrhenius equation, DCF ¼ crystal field splitting parameter. The spin-phonon relaxation between two different states m and n can be indicated by the spin-phonon coupling matrix element < m |H|n>. It turns out that if the anisotropic axes of the two doublets involved (m and n) are collinear then the matrix element of these two opposite projections of magnetic moments are small.27 So, the spin phonon interaction will be small. Thus, an understanding of these phenomena helps in the design of complexes where the relaxation of magnetization can be slow. The slow relaxation of magnetization in SMMs/SIMs is measured typically by alternating current (ac) magnetic susceptibility measurement where ac magnetic susceptibility is the differential dM/dH of the magnetization of a sample in the presence of an oscillating magnetic field. This driving field is kept as low as possible (1–5 Oe) to study the undisturbed ground state of the sample. These measurements are done at various temperature and frequency ranges. The ac magnetic susceptibility is a complex value as defined in Eq. (2). wac ¼ w0

iw00

(2)

The real component w0 relates to the reversible magnetization processes and stays in phase with the oscillating magnetic field (Eq. 3). Ht ¼ H0 + h cos ð2putÞ

(3)

00

The imaginary component w relates to the losses of irreversible magnetization process and incorporates a phase shift (y) in the magnetization M(t) with respect of H(t) according to Eq. (4). MðtÞ ¼ M0 + m cos ð2put −yÞ 28

(4) 00

This irreversible magnetization process indicates relaxation behavior. Thus, the peak dependence in the out of phase (w ) ac data with respect to temperature and frequency characterizes slow magnetic relaxation. Also each out of phase ac data gives a relaxation 1 time t at a particular temperature (t ¼ 2pu , where u is the peak maxima) and the plot of ln t vs T−1 provides the effective energy barrier for spin reversal when fitted to the Arrhenius equation (Eq. 5).   U eff t ¼ t0 exp (5) KBT

Organometallic Lanthanide Complexes as Single Molecule Magnets

387

SMM behaviors characterized by blocking temperature (temperature below which the magnetization of a SMM/SIM is retained indefinitely) which is defined in many ways: (i) The temperature up to which the magnetization (M) vs field (H) shows an open hysteresis loop, (ii) The temperature at which the dc magnetization relaxation time is 100 s, and (iii) The temperature at which the FC (field cooled) and ZFC (zero field cooled) plots show divergence.29 To understand the role of the effective ligand environment around a given lanthanide ion to maximize the magnetic anisotropy, one can examine the electronic ground states. The quadruple moment of the f-electron charge cloud can be used mathematically to describe the basic shapes of the ground J state,30 which is axially expanded (Oblate) for CeIII, PrIII, NdIII, TbIII, DyIII, and HoIII, axially elongated (prolate) as for PmIII, SmIII, ErIII, TmIII, and YbIII and isotropic (spherical) for LuIII, and GdIII (Fig. 5). To maximize the anisotropy of an oblate ion such as DyIII it is necessary to deploy an axial ligand field while for a prolate ion such as ErIII an equatorial ligand field will be favorable. The reason for this is that the f-electron cloud and the ligand electrons try to have minimum electronic repulsion to lower the energy of the system and thus when an oblate ion (DyIII) is placed in an axial ligand field the ground state will have bistable orientations (of mJ parallel and mJ antiparallel) along the molecular axis (large mJ).20 This approximation is supported by various examples and also from theoretical calculations. Chilton calculated the effect of the bending angle [L-Dy-L] of a model complex (Fig. 6) and found that strict linearity is not a perquisite and that near linear angles work just as fine.31 The study of the effect of Dy-L bond length on the effective energy barrier revealed that the stronger the bond the higher is the energy barrier as a strong bond leads to strong crystal field effect which enhances the energy separation between the spin multiplets.31 Also in some other reports it has been found that generally the ground state magnetic anisotropic axis is often aligned with the shortest Dy-L bond.31 In another study, Chibotaru and co-workers have found that among a series of DyFn complexes the complex DyF2 with a linear geometry possessed the highest energy barrier.21 These factors have influenced the design of appropriate organometallic lanthanide complexes as SIMs. Typically, the lanthanide ion in such complexes is sandwiched by p-ligands which generates a strong axial field around the lanthanide ion. In the following we will discuss about such lanthanide complexes. The narrative follows an increase in the ring size of the p-ligands. We will begin the discussion with cyclobutadienyl ligands. We will also discuss other types of organometallic lanthanide complexes toward the end.

Fig. 5 Quadrupole approximations of the 4f-shell electron distribution for the tripositive lanthanides. Values are calculated using the total angular momentum quantum number (J), the Stevens coefficient of second order (a) and the radius of the 4f shell squared .30 Europium is not depicted due to a J ¼ 0 ground state. Adapted from Ref Rinehart, J. D.; Long, J. R. Chem. Sci. 2011, 2, 2078–2085 with permission from Royal Society of Chemistry.

Fig. 6 The relaxation barrier Ueff for model complexes as a function of the bending angle y, averaged for all torsion angles f. Error bars are 1 standard deviation from the mean of the torsion angles f. Inset: Structure of the model complexes. Adapted from Ref. Chilton, N. F. Inorg. Chem. 2015, 54, 2097–2099 with the permission of the American Chemical Society.

388

Organometallic Lanthanide Complexes as Single Molecule Magnets

14.11.2 h4-Cyclobutadienyl ligand based lanthanide complexes as SMMs 1,2,3,4-Tetrakis(trimethylsilyl)cyclobuta-1,3-diene (Cb) based ligands have recently been employed to obtain organometallic SMMs with mainly involving DyIII ion. The use of Cb was thought to be efficient because of its high charge density which may lead to strong covalent binding with the metal center resulting in a high axial anisotropy. Using [K][M{Z4-C4(SiMe3)4}2] as a precursor the complexes, [M{Z4-C4(SiMe3)4}{Z4-C4(SiMe3)3-k-(CH2SiMe2}]2− (M ¼ DyIII (1), YIII (2)),32 were synthesized (Scheme 1A). The metal ion (DyIII) is sandwiched between two cyclobutadienyl anions. However, one of the SiMe3 substituents on one of the Cb rings is deprotonated in situ which results in a M-CH2 linkage. The Cb group further binds with K+ cation in a Z4 fashion and the K+ further binds with the p cloud of the toluene. Thus, overall a coordination polymer is formed due to the K+/p interactions. (Fig. 7). The DydC bond lengths were found to be in the range 2.524(6)–2.621(5) A˚ while the centroid (Cb)Dy-centroid (Cb) angle was found to be 156.42(9)o.

Scheme 1 (A) Schematic structural representation of complexes 1 (M ¼ DyIII) and 2 (M ¼ YIII). (B) Structural representations of complexes 3 (YIII), 4 (DyIII), and 5 (LuIII). (A) Adapted from Ref. Day, B. M.; Guo, F. S.; Giblin, S. R.; Sekiguchi, A.; Mansikkamaki, A.; Layfield, R. A. Chem. A Eur. J. 2018, 24, 16779–16782 with the permission of Wiley-VCH. (B) Adapted from Ref. Chakraborty, A.; Dey, B. M.; Durrant, J. P.; He, M.; Tang, J.; Layfield, R. A. Organometallics 2020, 39, 8–12 with the permission of the American Chemical Society.

Fig. 7 Molecular structure (A) and the extended polymeric structure (B) of 1. Thermal ellipsoids at 30% probability. Color codes: sky blue, K; orange, Si; black, C; green, Dy. Hydrogen atoms are omitted for clarity. Adapted from Ref. Day, B. M.; Guo, F. S.; Giblin, S. R.; Sekiguchi, A.; Mansikkamaki, A.; Layfield, R. A. Chem. A Eur. J. 2018, 24, 16779–16782 with the permission of Wiley-VCH.

Dynamic magnetic studies revealed that 1 showed some peaks in the high frequency region of the out of phase ac magnetic susceptibility (w00 ) at zero applied magnetic field. However, the presence of strong QTM was inferred. To suppress the latter an optimum field of 1000 Oe was applied. This improved the response (Fig. 8A). An analysis of the magnetic data revealed that the Orbach process was operating at higher temperatures and the Raman process at lower temperatures. The higher temperature data −1 with t0 ¼ 1.83  10−9 s. The could be fit to the equation t−1 ¼ t−1 0 exp.(−Ueff/KBT) to afford the values Ueff ¼ 323(22) cm magnetization versus field plot showed a S shaped hysteresis loop with no appreciable coercive field which remains open up to 7 K (Fig. 8B).

389

Organometallic Lanthanide Complexes as Single Molecule Magnets

(A)

0.5

(B)

4 2.1 K 8K

3.0 K 5.0 K 7.0 K

2

40 K

0.3

M / Nb

cMcc / cm3 mol–1

0.4

0.2

0

–2 0.1 –4 0.0 1

100

10 Q/ Hz

1000

–30

–20

–10

0

10

20

30

H / kOe

00

Fig. 8 (A) Frequency dependence of w (n) in an applied field of 1 kOe for complex 1. (B) M(H) hysteresis at T ¼ 2.1–7.0 K using an average sweep rate of 50 Oe s−1 (5 mT s−1). Adapted from Ref. Day, B. M.; Guo, F. S.; Giblin, S. R.; Sekiguchi, A.; Mansikkamaki, A.; Layfield, R. A. Chem. A Eur. J. 2018, 24, 16779–16782 with the permission of Wiley-VCH.

Fig. 9 Principal axis of g tensor of complex 1. Adapted from Ref. Day, B. M.; Guo, F. S.; Giblin, S. R.; Sekiguchi, A.; Mansikkamaki, A.; Layfield, R. A. Chem. A Eur. J. 2018, 24, 16779–16782 with the permission of Wiley-VCH.

Ab initio calculations showed that the principal magnetic axis of the ground Kramers doublet state of the 6H15/2 ground multiplet is aligned along the center of the Cb rings (Fig. 9) and also the calculated g tensors revealed the highly axial nature of the ground doublet. Layfield et al. used [Na2{Z4-C4(SiMe3)4}(THF)]2 as the reagent and synthesized sandwich coordination polymers, [M{Z3-C4(SiMe3)4H}{Z4-C4(SiMe3)3-k-(CH2SiMe2}Na]1 (M ¼ YIII (3), DyIII (4), LuIII (5))33 (Scheme 1B). An elucidation of the molecular structures revealed a double activation of the Cb rings. One of the Cb rings that is attached to DyIII ion in a Z4 fashion loses a proton from the Me group of SiMe3 substituent and is bound to DyIII, as described above. Another Cb ring is protonated and then binds to DyIII ion in a Z3 fashion (Fig. 10A). The sodium ion is involved in interactions with the Cb ring and the side chain to afford a polymeric structure (Fig. 10B). While the distances of Dy to the centroids of the Z3-Cb and Z4-Cb ligands are 2.460(5) A˚ and 2.308(5) A˚ respectively and the Cb-Dy-Cb angle is 159.157(5)o. The ac magnetic susceptibility measurements showed characteristic peaks in the out of phase ac data which indicated slow relaxation of magnetization even in the absence of applied dc field for complex 4.

390

Organometallic Lanthanide Complexes as Single Molecule Magnets

Fig. 10 Thermal ellipsoid representation (30% probability) of the (A) molecular structure of 4. 3 and 5 are isostructural to 4. Color codes: pink, Na; orange, Si; black, C; green, Dy. Hydrogen atoms are omitted for clarity except the protonated hydrogen (color: gray). (B) segment of the extended polymeric structure. For clarity, the hydrogen atoms are not shown. Adapted from Ref. Chakraborty, A.; Dey, B. M.; Durrant, J. P.; He, M.; Tang, J.; Layfield, R. A. Organometallics 2020, 39, 8–12 with the permission of the American Chemical Society.

The plot of ln t vs T−1 showed a thermally activated relaxation process approximately above 12 K. Below this a temperature n −1 independent process is found. The ln t vs T−1 plot was fitted by the equation t−1 ¼ t−1 0 exp(−Ueff/kBT) + CT + tQTM, where n and C −1 −1 are the Raman exponent and Raman coefficient and tQTM is rate of QTM. This fitting gave Ueff ¼ 309(20) cm with t0 ¼ 1.03(4)  10−9 s, C ¼ 0.78(2) s−1 K-n, n ¼ 2.07(8) and tQTM ¼ 3.78(8)  10−3 s. This energy barrier is similar to that found for 1 but in the latter the SMM behavior was revealed only in presence of an applied field. In another work, {Z4-C4(SiMe3)4}Dy was reacted with [Dy(BH4)3(THF)3] along with alkali metal cyclobutadienyl salts [A2{C4(SiMe3)4}] (A ¼ Na, K) resulting in the formation of the coordination polymers, [Dy{Z4-C4(SiMe3)4}(BH4)2(THF)Na]1 (6) and [Dy{Z4-C4(SiMe3)4}(BH4)2(THF)K]1 (7) (Scheme 2).34

Scheme 2 Synthesis of complexes 6 and 7 (M ¼ DyIII, L ¼ THF, R ¼ SiMe3). Adapted from Ref. Durrant, J. P.; Tang, J.; Mansikkamaki, A.; Layfield, R. A. Chem. Commun. 2020, 56, 4708–4711 with the permission of the Royal Society of Chemistry

The Dy-Cb (centroid) distance in 6 is 2.262(4) A˚ and the Dy-Cb-Na angle was found to be 178.4(2)o. The molecular structure of 7 is almost similar with the only noticeable difference being the involvement of two bridging BH−4 units. The Dy-Cb (centroid) distance and Dy-Cb-K angle in this complex are 2.264(3) A˚ and 176.46(13)o respectively. Both 6 and 7 showed SMM behavior in zero applied field with energy barriers of 371(7) cm−1 and 357(4) cm−1 respectively.

Organometallic Lanthanide Complexes as Single Molecule Magnets

391

14.11.3 h5-Cyclopentadienyl ligand based lanthanide complexes as SMMs 14.11.3.1 General remarks Among the lanthanide based SMMs most of the remarkable results are found when DyIII is coupled with cyclopentadienyl (Cp) based ligands. Various types of substituted Cp have been used for this purpose. These include 1,2,3,4,5-pentamethyl cyclopentadienyl (Cp ), tetramethylcyclopentadienyl (CpMe4H), 1,2,4-tri(tert-butyl)cyclopentadienide (Cpttt), pentaisopropylcyclopentadienyl (Cpipr5), and 1,10 ,3,30 -tetra(tert-butyl)dicyclopentadienyl (Fvttt) (Scheme 3). Cp type of ligands form two main classes of compounds with lanthanide ions, one of these with competing equatorial ligands and the other without any equatorial ligands.35 In both the cases, it has been revealed by ab-initio calculations that the principal magnetization axis of the ground state KD mostly passes through the Cp rings.

Scheme 3 Various types of cyclopentadienyl based ligands used in the assembly of organometallic lanthanide SMMs.13,36–52

As discussed above this type of axial binding of the Cp ring is very beneficial to maintain a high axial anisotropy and thus increasing the blocking temperature for oblate type of lanthanide ions such as DyIII, TbIII etc. The angle Cp-Dy-Cp is important as it gives a qualitative sense of the axiality of the system. It is found that with increase in the Cp-Dy-Cp angle the effective energy barrier for slow magnetic relaxation increases. Other important parameters in these complexes are the Dy-Cp distances. It has been found that with decrease in the Dy-Cp distance, i.e. strengthening of the bond, the anisotropy of the complexes increased. The [Dy(Cp)2X2]n type of complexes where X represents the equatorially bound ligands showed a more pronounced tendency for quantum tunnelling as compared to the [Dy(Cp)2]+ units where such equatorial ligands are absent. This is because the equatorial ligands contribute to the transverse component of the crystal field. Dinuclear complexes bridged by the radical ligand, N23− exhibited high effective energy barrier for magnetization reversal. However, this seems to more due to the strong exchange interactions mediated between the lanthanide ions by the radical ligand. Among all the complexes studied those that contained Cpipr5/Cp /Cpttt when combined with DyIII showed the highest blocking temperatures so far.

14.11.3.2 h5-Cyclopentadienyl ligated lanthanide complexes with competing equatorial ligands The first organometallic lanthanide based SMM was synthesized by Winpenny et al. These researchers used the cyclopentadienyl ligand along with benzotriazole (btaH]1H-1,2,3-benzotriazole) and synthesized [{Cp2Dy(m-bta)}2] (8) (Fig. 11A).36 Use of 2-amino-4,6-dimethylpyrimidine (NH2pmMe2) afforded [{Cp2Dy[m-N(H)pmMe2]}2] (9) (Fig. 11B). Static magnetic data of complexes 8 and 9 are almost similar (Fig. 11C) but the dynamic magnetic data showed remarkable difference where complex 8 behaves as an SMM while 9 does not. Complex 8 showed a temperature dependence in the out of phase (w00 ) ac data in the absence of any external magnetic field. The relaxation time was found to nearly saturate around 4.5 K (t ¼ 7.0  10−4 s). Below this temperature the relaxation was found to be dominated by QTM (Fig. 11D). To understand the different dynamic magnetic behavior theoretical calculations were done and it was found that in 8 there is no pathway for communication between the two DyIII centers via the bridging moieties whereas there is a pathway of communication for DyIII ions in complex 9 via the bridging N1 and N2 centers (Fig. 11B). Thus DyIII centers in complex 8 essentially behave as single ion centers and the magnetic property is attributed to this isolated nature of the metal centers while in complex 9 the super exchange interaction mediated by the ligand results in a fast relaxation.

392

Organometallic Lanthanide Complexes as Single Molecule Magnets

Fig. 11 Molecular structure of complex 8 (A) and 9 (B). Color codes: gray, C; blue, N; green, Dy. (C) Temperature dependence of wMT of 8 and 9 for the field strength H ¼ 1000 G. (D) Magnetization relaxation time (t) versus T−1 plot for 8 under zero-dc field (circles) and under a dc field of 1000 G (squares), based on data collected in frequency (shaded symbols) and temperature (open symbols) variation regimes. The solid lines represent the best fits to the Arrhenius law of the thermally activated region with the parameters given in the text. Adapted from Ref. Layfield, R. A.; McDouall, J. J. W.; Sulway, S. A.; Tuna, F.; Collison, D.; Winpenny, R. E. P. Chem. A Eur. J. 2010, 16, 4442–4446 with the permission of Wiley-VCH.

The chloro bridged dinuclear dysprosium complexes, [(Z5-Cp)2Dy(m-Cl)]2 (10a), [(Z5-Cp)2Dy(m-Cl)]1 (10b), and [(Z5-Cp)2(thf )Dy(m-Cl)]2 (11) (Fig. 12) were synthesized and studied.37 10 has been found to contain two species. While 10a is a discrete dimer 10b is a coordination polymer. Complex 11 which is prepared from [(Z5-Cp)2Dy(m-Cl)]n is a dimer. PXRD analysis of 10 showed a distribution of 10a and 10b in a 3:1 ratio. The ac magnetic susceptibility showed two relaxation pathways having 75% and 25% weightage which indicates that both the magnetic species 10a and 10b have their own separate relaxation pathways (Fig. 13A). The micro squid measurement at 40 mK at zero applied field showed hysteresis with no tunnelling steps whereas when 0.1 T field is applied stepped hysteresis is observed which is assignable to the quantum tunnelling of magnetization (Fig. 13B). So, for complex 11 QTM is exchange biased i.e. the exchange between the DyIII ions is sufficient to provide a finite local magnetic field at each DyIII ion, even in the absence of external magnetic field. Layfield et al. reported a sulfur-bridged dinuclear complex, [{Cp0 2Dy(mSSiPh3)}2] (12) (Cp0 ¼ Z5-C5H4Me).38 The rationale behind using a soft bridging ligand was that it can mediate a good exchange interaction between the lanthanide centers and also

Dy1

CI1

CI1

O1A CI1

Dy1

Dy1A

Dy1

Dy1A CI1A

CI1A Dy1A

O1 CI1A

Fig. 12 Structures of 10a (A), 10b (B) and 11 (C) with thermal ellipsoid at the 50% probability level. Color codes: black, C; green, Cl; red, O; blue, Dy. Hydrogen atoms omitted for clarity. Adapted from Ref. Sulway, S. A.; Layfield, R. A.; Tuna, F.; Wernsdorfer, W.; Winpenny, R. E. P. Chem. Commun. 2012, 48, 1508–1510 with the permission of the Royal Society of Chemistry.

Organometallic Lanthanide Complexes as Single Molecule Magnets

393

Fig. 13 (A) Temperature dependence of w00 for complex 10 at zero applied external dc field. (B) Magnetization versus field loops for 11 at different scan rates and a temperature of 40 mK. For 10a, Ueff ¼ 26.3  0.76 cm−1 and t0 ¼ 1.4  10−6 s; 10b, Ueff ¼ 67.8  1.7 cm−1 and t0 ¼ 2.8  10−6 s; 11, 33.84  0.48 cm−1 and a pre-exponential factor (t0) of 4.0  10−7 s. Ref. Sulway, S. A.; Layfield, R. A.; Tuna, F.; Wernsdorfer, W.; Winpenny, R. E. P. Chem. Commun. 2012, 48, 1508–1510 with the permission of the Royal Society of Chemistry.

Fig. 14 (A) Molecular structure for complex 12, ellipsoid set at 50% probability level. For clarity, only the ipso carbons of the phenyl rings are shown, and the hydrogen atoms are omitted. (B) Arrhenius plot of ln t versus (1/T) for 12. The solid line is the best fit of the data in the thermally activated regime (T > 20 K). For 12, Ueff ¼ 133  3.5 cm−1 and t0 ¼ 2.38  10−7 s. Adapted from Ref. Tuna, F.; Smith, C. A.; Bodensteiner, M.; Ungur, L.; Chibotaru, L. F.; McInnes, E. J. L.; Winpenny, R. E. P.; Collison, D.; Layfield, R. A. Angew. Chem. Int. Ed. 2012, 51, 1–6 with the permission of Wiley-VCH.

because of the variation in the crystal field that can be affected. The complex was centrosymmetric and connected by m-SSiPh3 ligands. The DyIII centers reside in a pseudo tetrahedral geometries with respect to the Cp ligand centroids and the sulpfr atoms (Fig. 14A). The ac magnetic susceptibility data showed frequency dependence of the out of phase ac data up to 40 K in the absence of any dc field (Fig. 14B). The anticipation that the sulfur ligands provide strong exchange interaction seemed to be validated by the above results. Long and co-workers synthesized [Cp 2Ln(BPh4)] (Cp ]pentamethylcyclopentadienyl; Ln ¼ TbIII (13) and DyIII (14)) where the lanthanide ions are strongly coordinated to the Cp rings in a bent fashion and weakly coordinated to the BPh−4 anion (Fig. 15A) through agostic interactions.39 In the dc magnetic measurement for complex 14 a sudden drop in the wMT value is observed in the wMT vs T plot at 3.2 K (Fig. 15B) which is indicative of a large blocking for magnetization which is further confirmed from the divergence of the FC and ZFC plot at 3.2 K (energy barriers for complexes 13 and 14 are given in Fig. 15 caption). To further increase the energy barrier Murugesu et al. synthesized a BPh−4 linked Ln-(Cp)2 dimer, [Dy2Cp 4(m-BPh4)][Al(OC(CF3)3)4] (15).40 In accordance with the expectation the energy barrier values increased: Ueff ¼ 330 cm−1 with t0 ¼ 2.75  10−8 s. Ab initio calculations on this complex revealed that the projection of magnetic axis of the ground KD is along the tilted Cp rings. This complex also shows butterfly type hysteresis loop which remains open up to 6.5 K. The effect of equatorial ancillary ligands on the relaxation dynamics was explored using the complexes, [Cp 2DyCl(THF)] (16), [Cp 2DyBr(THF)] (17), [Cp 2DyI(THF)] (18), [Cp 2DyCl2K(THF)]n (19), and [Cp 2DyTp] (Tp ¼ hydrotris(1-pyrazolyl)borate) (20)41 (Scheme 4).

394

Organometallic Lanthanide Complexes as Single Molecule Magnets

Fig. 15 (A) Molecular structure of complex 14, color code: gray, C; green, Dy; purple, B. The red arrow shows the orientation of the ground state anisotropy axis. H atoms are removed for clarity. Complex 13 is isostructural to 14. (B) variable-temperature dc magnetic susceptibility data for restrained polycrystalline samples of 13 and 14 collected under a 1 kOe applied dc field. For 13, Ueff ¼ 221 cm−1 and t0 ¼ 5  10−10 (HDC ¼ 2400 Oe); 14, Ueff ¼ 314 cm−1 and t0 ¼ 2  10−8 (HDC ¼ 1600 Oe). Adapted from Ref. Demir, S.; Zadrozny, J. M.; Long, J. R. Chem. A Eur. J. 2014, 20, 9524–9529 with permission of Wiley-VCH.

Scheme 4 Schematic structures of 16–20.

For isostructural 16–18, the effective energy barrier (Ueff) showed the trend 18 (419 cm−1) > 17 (163 cm−1) > 16 (112 cm−1) which can be correlated to the strength of the Dy-X bond. From 16 to 18 the equatorial Dy-X bond becomes weaker due to the increase in the size of the halide ion leading to a decrease in the transverse component of the crystal field. The QTM also follows the same order 18 (3.7 ms) > 17 (1.4 ms) > 16 (0.28 ms). Complex 19 which is a coordination polymer has a higher energy barrier (379 cm−1) in comparison to 16. This was attributed to the more symmetrical structure and the longer DydCl bond length. 20 shows a lower energy barrier of Ueff ¼ 106 cm−1 which was attributed to the binding of hydrotris(1-pyrazolyl)borate in the equatorial position providing significant transverse component and thus reducing the anisotropy of the system. Layfield et al. also explored low symmetry DyIII complexes, [Cp 2M(m-Fp)]2 (M ¼ Y, 21; M ¼ Dy, 22; Cp ¼ 1,2,3,4,5-pentamethyl cyclopentadiene; Fp ¼ CpFe(CO)2) (Fig. 16A).42 In these complexes, in spite of the low symmetry the presence of weak coordination by Fp ligands results in a high energy barrier: 662(2) cm−1 with t0 ¼ 1.7(1)  10−12 s. A similar complex containing the analogs tungsten carbonyl ligands, [Cp 2Dy(m-OC)W(Cp)(CO)(m-CO)]1 (23) (43) revealed a polymeric structure (Fig. 16B and C). Dy-Cp distances are 2.34534(8) and 2.34805(6) A˚ . The O-Dy-O and Cp -Dy-Cp angles are 85.3(3)o and 138.74(13)o respectively. The energy barrier for this complex was found to be 557(18) cm−1 with t0 ¼ 3  10−12 s which is similar to that observed in 22. The complexes, [(CpMe)2Dy{m-E(H)Mes}]3 with E ¼ P (24), As (25) or Sb (26) and [Li(THF)4]2[{(CpMe)2Dy(m-EMes)}3Li] with E ¼ P (27), As (28)44–46 have been assembled. Pnictogens were used as they have more diffuse electron density which can form good overlap with the diffuse f-orbitals of the lanthanide ions. These complexes also provided an opportunity to understand the effect of the equatorial ligand field strength on the relaxation behavior in [Cp2DyE]n type complexes. Structures of complexes 24–26 are very similar in a C2v site symmetry on the DyIII center (Fig. 17A). All these complexes showed SMM behavior with effective energy barriers of 210(6) cm−1, 256(5) cm−1, and 345 cm−1, respectively. The increase in the effective energy barrier with the heavier pnictogens has been rationalized with the lower overlap of the p and f-orbitals moving from phosphide to stibinie complexes. This is also supported by the increasing Dy-pnictogen bond distances

Fig. 16 (A) Molecular structure of 22 (21 is isostructural to 22). (B) Repeating structural unit in 23 and (C) Polymeric chain like structure of 23 along crystallographic b-axis. Color code: gray, C; green, Dy; orange, Fe; red, O; blue, W. Hydrogen atoms are omitted for clarity.

Fig. 17 Molecular structure of 25 (A) and 28 (B). Grounds Kramers doublet projections for 25 (C) and 28 (D). Dy ¼ green, arsenic ¼ purple, lithium ¼ pink, carbon ¼ gray. 24 and 26 are isostructural to 25 and 27 is isostructural to 28. Adapted from Ref. Pugh, T.; Vieru, V.; Chibotaru, L. F.; Layfield, R. A. Chem. Sci. 2016, 7, 2128–2137 with the permission of the Royal Society of Chemistry.

396

Organometallic Lanthanide Complexes as Single Molecule Magnets

Fig. 18 Molecular structures of 29 (A) and 30 (B) with hydrogen atoms and solvent molecules removed for clarity. Teal ¼ Dy, yellow ¼ S, red ¼ oxygen, blue ¼ N, light green ¼ F, green ¼ Cl, and gray ¼ C. H atoms are removed for clarity. The pink transparent lines indicate gzz directions of the magnetic ground states. Adapted from Ref. Burns, C. P.; Wilkins, B. O.; Dickie, C. M.; Latendresse, T. P.; Vernier, L.; Vignesh, K. R.; Bhuvanesh, N. S.; Nippe, M. Chem. Commun. 2017, 53, 8419–8422 with the permission of the Royal Society of Chemistry.

which for complexes 24–26 are 2.920(6)–2.946(6) A˚ to 2.984(2)–3.012(2) A˚ to 3.118(2)–3.195(2) A˚ , respectively. From ab initio calculations it was found that for 25, the principal magnetic axis of the ground state KD is aligned by making a certain angle with the Dy3 plane (66.9–67.6 ) (Fig. 17C). But for 28 the magnetic axis passes through the center of the Cp − ligand which is almost perpendicular to the Dy3 plane (85.0–86.9 ) (Fig. 17D). For 24–26 the pnictogens lie in the equatorial plane and thus weakening of the Dy-pnictogen bond decreases the transverse component of the ligand field which increases the anisotropy of the metal center and thus an increase in the effective energy barrier is observed from complex 24 to 26. The effective energy barriers for 27 and 28 (Fig. 17B) are 13(1) cm−1 and 23 cm−1 respectively and this can be explained on the basis that in these complexes the equatorial moiety, [m-E-Mes]2− (E ¼ P and As) bears more negative charge and thus contribute to more transverse component of the crystal field as compared to 24–26. Nippe et al. reported a new approach to prepare dinuclear Cp based DyIII complexes. They used a sterically demanding PyCp22− ligand backbone with chloride and trifluoromethanesulfonate anion as the bridging ligand. Employing the rigid ligand 2− PyCp22− (PyCp2− 2 ¼ [2,6-(CH2C5H3)2C5H3N] ) they synthesized [(PyCp2)Dy-(m-O2SOCF3)]2 (29) (Fig. 18A) and [(PyCp2)Dy47 (m-Cl)]2 (30) (Fig. 18B). Both the complexes showed a frequency dependence of the out-of-phase ac data at zero applied field. Using the Arrhenius equation to extract the relaxation parameters they obtained the same effective energy barrier for 29 and 30 viz., 48.7 cm−1 with pre exponential factors of 4.8  10−7 s and 7.2  10−7 s respectively. From the ab initio calculations it was found that the ground state for both the complexes is indeed mJ ¼  15/2 with some contribution from the mJ ¼  11/2 state and both the complexes relax through the first excited KDs. 2− 2− A sterically rigid PyCp2− 2 ligand (PyCp2 ¼ [2,6-(CH2C5H3)2C5H3N] ) was used to obtain, [PyCp2Dy-{FeCp(CO)2}] (31; d(Dy-Fe) ¼ 2.884(2) A˚ ) and [PyCp2Dy-{RuCp(CO)2}] (32; d(Dy-Ru) ¼ 2.9508(5) A˚ ).48 These complexes showed moderate SMM behavior (Table 3). SMMs containing YbIII are rare. The organometallic compound [Cp Yb(DAD)(THF)]
C7H8 (33) (DAD ¼ enediamido[2,6-Me2C6H3NCH]CHNC 6H3Me2–2,6]2−) containing the Cp ligand was found to show a SIM behavior at an optimum dc field of 1500 Oe with an energy barrier of Ueff ¼ 14  2 cm−1 with the value t0 ¼ 1.74  10−6 s (Table 3).49

14.11.3.3 Cationic metallocene SMMs Sandwich complexes of LnIII without any equatorial ligands have the best possibility of a strong axial field with minimum transverse contribution. This goal was addressed in the complex, [(Cpttt)2Dy][B(C6F5)4] (35) (Cpttt ¼ (1,2,4-tri(tert-butyl)cyclopentadienide)) (Scheme 5).22,50

Scheme 5 Synthesis of complex 35. Adapted from Ref. [22] with permission of Wiley-VCH

Organometallic Lanthanide Complexes as Single Molecule Magnets

397

Fig. 19 (A) Molecular structure of the cationic part of 35, (B) w00 vs T plot for complex 35 at HDC ¼ 0 Oe, and (C) DC magnetic relaxation for 35 from 40 to 57 K. Black lines represent the best fits to the exponential decay as M(t) ¼ M(0) exp.(−t/t), where M(t) is the magnetization at a given measurement time, t is the measurement time, t is the magnetic relaxation time. Adapted from Ref. Guo, F. S.; Day, B. M.; Chen, Y. C.; Tong, M. L.; Mansikkamäki, A.; Layfield, R.A. Angew. Chem. Int. Ed. 2017, 56, 11445–11449 with permission of Wiley-VCH. ttt ttt Both 34 and 35 adopt a bent structure. However the Dy-Cpttt c (Cpc ¼ centroid of the Cp ligand) distance (2.324(1) and 2.309 ttt (1) A˚ ) is shorter in the cationic complex 35 in comparison to the neutral complex, 34 (2.413(2) A˚ ). Also the Cpttt c -Dy-Cpc angle in 35 is slightly larger (152.845(2)o) than in 34 (147.59(7)o). (Fig. 19A). Complex 34 does not behave as a SMM whereas 35 showed a remarkable frequency dependence of the w00 of the ac data at temperature 72–110 K in the absence of any external field (Fig. 19B). Plotting relaxation time (t) with reciprocal of temperature (T−1) revealed that at higher temperature the relaxation happens through a thermally activated pathway (Orbach process). This data was then fitted with the Arrhenius equation to provide the relaxation parameters as Ueff ¼ 1277(14) cm−1 with t0 ¼ 8.12  10−12 s. The low temperature relaxation behavior was probed with dc magnetic measurements where the decay of the magnetization was studied as a function of time to extract the relaxation time. The analysis showed that the linear dependence of relaxation time (t) with time (t) persists down to 53 K where the relaxation time turned out to be 100 s (Fig. 19C). Therefore the blocking temperature for this complex was assigned as 53 K. At the extended low temperature range the relaxation develops a slight curvature which may indicate the onset of Raman relaxation at this temperature region. From FC and ZFC data collected at HDC ¼ 1000 Oe the divergence temperature was found to be 60 K which also can be assigned to the blocking temperature (Fig. 20A). Magnetic hysteresis measurements were also performed on this complex with an average sweep rate of 39 Oe s−1 which showed that the hysteresis loop is open to 60 K and having a coercive field of Hc ¼ 0.06 T (Fig. 20B). At lower temperatures 2–55 K the hysteresis loop is wider ranging from 0.19–2.46 T. Detailed ab initio study of 35 showed that the principal axis of the ground state KD is oriented along the center of the Cpttt rings (Fig. 21A) which is in agreement with the predicted ground state anisotropy axis for hypothetical species [Dy(Cp∗)2]+.41 For 35 almost all the KDs except the eighth KD have near axial g values. All excited KDs are roughly parallel to the ground state KD with the maximum deviation of 5.6o with the fifth KD. For complex 35 the QTM is prohibited up to second excited state due to the high axiality. However, from the fourth KD QTM becomes non negligible and in sixth KD it is dominant (Fig. 21B). The energy of the sixth KD is 1156 cm−1 which is close to the experimentally evaluated value of 1277 cm−1.

Fig. 20 (A) Field cooled (FC, black) and zero-field-cooled (ZFC, red) variable-temperature magnetic susceptibility for [35] at 1000 Oe (warm mode, 2 K min−1). (B) M(H) hysteresis for 35 using an average sweep rate of 3.9 mT s−1. Adapted from Ref. Guo, F. S.; Day, B. M.; Chen, Y. C.; Tong, M. L.; Mansikkamäki, A.; Layfield, R. A. Angew. Chem. Int. Ed. 2017, 56, 11445–11449 with permission of Wiley-VCH.

398

Organometallic Lanthanide Complexes as Single Molecule Magnets

Fig. 21 (A) Direction of the principal axis of the g-tensor in the ground Kramers doublet of 35 and (B) Relaxation of magnetization in 35. Dark gray arrows indicate the largest matrix elements between states and therefore map the most probable relaxation route. Light gray arrows indicate less significant but non-negligible matrix elements between states. Adapted from Ref. Guo, F. S.; Day, B. M.; Chen, Y. C.; Tong, M. L.; Mansikkamäki, A.; Layfield, R. A. Angew. Chem. Int. Ed. 2017, 56, 11445–11449 with permission of Wiley-VCH.

Layfield et al. adopted a synthetic approach where they have designed the ligands in a way so that they are bulky enough to provide a wide Cp-Dy-Cp angle but not too bulky so that the close approach of the ligand to the metal center is hindered. The complex [(Z5-Cp )Dy(Z5-CpiPr5)][B(C6F5)4] (38) was synthesized according to Scheme 6 with [Dy(Z5-CpiPr5)(BH4)2(THF)] (36) (CpiPr5, pentaiso-propylcyclopentadienyl) as precursor.23

Scheme 6 Synthesis of 38. Adapted from Ref. Guo, F. S.; Day, B. M.; Chen, Y. C.; Tong, M. L.; Mansikkamäki, A.; Layfield, R. A. Science 2018, 362, 1400–1403 with the permission of the American Association for the Advancement of Science.

From crystallographic data of complex 38 it was found that the Cp -Dy (2.296(1) A˚ ) and Cpipr5-Dy (2.284(1) A˚ ) distances are on an average shorter by 0.026 A˚ as compared to the analogous distance (2.32380(8) and 2.30923(8) A˚ ) in [(Cpttt)2Dy]+ (35). For 38 the Cp -Dy-Cpipr5 angle is 162.507(1)o which is almost 9.7o larger than the analogous angle of 152.845(2)o found in 35. These structural parameters suggest that complex 38 has a more axial crystal field than the analogous complex 35 (Fig. 22A) and can possess better magnetic properties. Accordingly, 38 shows an out-of-phase ac susceptibility (w00 ) peak maxima which is temperature dependent up to 130 K (Fig. 22B). 36 and 37 show energy barriers of 241 cm−1 and 7 cm−1 respectively. From the plot of magnetization decay vs temperature, the relaxation times in the temperature range 2–83 K were evaluated in the interval of 5 K which shows that the magnetization in 38 decays almost to zero at 77 K with 55 s time period and the relaxation time is increased to almost 500 min at 15 K. At 65 K the relaxation time is 100 s (Fig. 22C). Both the low temperature and high temperature relaxation time analysis were combined to get further insight into the relaxation dynamics by plotting relaxation time vs T−1 and the overall data were fitted to Eq. (6). Here, the first term represents the thermally activated relaxation process, the second term represents the Raman process (C, Raman coefficient and n, Raman exponent) and the third term represents the quantum tunnelling of the magnetization.13   DCF + CTn + tQTM −1 (6) t −1 ¼ t0 −1 exp − KB T The fitted plot gives the effective energy barrier of 1541(11) cm−1 with a pre-exponential factor (t0) of 4.2(6)  10−12 s. For the Raman process C ¼ 3.1(1)  10−8 s−1 K-n and n ¼ 3.0(1) and for quantum tunnelling process tQTM ¼ 2.5(2)  104 s were obtained. The effective energy barrier for this complex is about 20% higher than that found in the analogous complex [(Cpttt)2Dy][B(C6F5)4)] (35).

Organometallic Lanthanide Complexes as Single Molecule Magnets

399

Fig. 22 (A) Molecular structure of cationic part of 38, (B) out of phase ac susceptibility (w00 ) vs frequency plot at different temperatures, and (C) temperature dependence of the relaxation time for 38. The red points are from the ac susceptibility data, and the blue points are from measurements of the dc magnetic relaxation time. The solid green line is the best fit to Eq. (6), using the parameters stated in the text. Adapted from Ref. Guo, F. S.; Day, B. M.; Chen, Y. C.; Tong, M. L.; Mansikkamäki, A.; Layfield, R. A. Science 2018, 362, 1400–1403 with the permission of the American Association for the Advancement of Science.

Fig. 23 (A) Magnetization versus field hysteresis loop in the temperature range 2–75 K, (B) principal magnetic axis of ground Kramers doublet, and (C) relaxation dynamics for 38. Blue arrows show the most probable relaxation route, and red arrows show transitions between states with less probable, but non negligible, matrix elements; darker shading indicates a higher probability. Adapted from Ref. Guo, F. S.; Day, B. M.; Chen, Y. C.; Tong, M. L.; Mansikkamäki, A.; Layfield, R. A. Science 2018, 362, 1400–1403 with the permission of the American Association for the Advancement of Science.

Complex 38 showed hysteresis from 2 to 85 K with a sweep rate of 200 Oe s−1 with loops slowly closing as the temperature increases. At 80 K with a sweep rate of 25 Oe s−1 a coercive field of 63 Oe was obtained and the loops were completely closed at higher temperature (Fig. 23A). This observation is consistent with the finding of divergence of the field cooled and zero field cooled magnetization data for 38 at 78 K. Ab initio calculations performed on this complex indicated that the magnetization axis of the ground state doublet is aligned along the centroid of the cyclopentadienyl ligands (Fig. 23B) and that the g tensor of 38 turns out to be perfectly axial with gz ¼ 20 and gx ¼ gy ¼ 0 which justifies the absence of quantum tunnelling of magnetization at zero applied field. On going from the ground state KD to the higher energy KDs the transverse anisotropy slowly increases and in the fifth KD it is non negligible and in the sixth KD it is sufficient to allow QTM. Axiality of the two highest KDs are weak and considerable crystal field mixing happens which may be the reason for the asymmetry of the coordination environment (Fig. 23C). From the crystal field (CF) calculations it was found that due to the low C1 point symmetry of complex 38, the crystal field operator has a non negligible diagonal element. However, the axial second rank parameter B02 is about two orders of magnitude larger than any other parameters. This creates high axial CF environment even in the absence of point group symmetry. This result indicates that it is possible to have a high axial CF system even in the absence of a strict point group symmetry if the axial components of the CF are larger as compared to the other parameters arising from the low symmetry components of the CF. From the frequency calculations it was found that the out of plane vibration of the Cpipr5 ligand closely matches the energy required for excitation from ground state KD to the first excited KD. Thus, it appears that this vibration mode contributes significantly to the relaxation pathway through the Orbach process. A series of analogous complexes such as described above, [Dy(CpiPr4R)2][B(C6F5)4] (R ¼ H (39), Me (40), Et (41), iPr (42))51 were studied (Scheme 7). This revealed that the Cp-Dy-Cp angle and Cp-Dy bonding distance increase from 39 to 42. The ac magnetic susceptibility studies suggest slow magnetic relaxation for all the complexes but the maximum blocking temperature is exhibited by 40. The effective energy barriers calculated for 39–42 are 1285, 1468, 1380, and 1334 cm−1 respectively which are higher than reported for 35.20,50 Thus, a balance between the steric and electronic effects appear to be important. It has been

400

Organometallic Lanthanide Complexes as Single Molecule Magnets

Scheme 7 Synthesis of metallocene 39–42. Adapted from Ref. McClain, K. R.; Gould, C. A.; Chakarawet, K.; Teat, S. J.; Groshens, T. J.; Long, J. R.; Harvey, B. G. Chem. Sci. 2018, 9, 8492–8503 with the permission of the Royal Society of Chemistry.

Table 2

Selected geometric parameters and magnetic properties for 35, 38–42. 35

38

39

40

41

42

Cp-Dy-Cp (o)a Cp-Dy (A˚ )b

152.7 2.316(3)

147.2 2.29(1)

156.6 2.298(5)

161.1 2.302(6)

162.1 2.340(7)

Ueff (cm−1)c TB (K)d

1277 60

162.507 2.296(1) 2.284(1) 1541 80

1285 32

1468 72

1380 66

1334 66

a

Average value for all positions in the crystal structure. Average value for the highest occupancy component in the crystal structure. c Determined with HDC ¼ 0 Oe. d TB is defined as the maximum hysteresis temperature. b

suggested that Cp-Dy distance appears more important than the Cp-Dy-Cp angle (Table 2). This observation is in line with the theoretical study by N. Chilton31 which showed that Ueff increases linearly with decrease in the Dy-Cp bond length for a fixed Cp-Dy-Cp angle. Blocking temperatures for 39–42 are 32, 72, 66, and 66 K respectively considering the maximum temperature up to which the hysteresis loop is open.

14.11.3.4 Dysprosium metallocene complexes containing bridging borohydride ligands The fulvalenyl ligand, [1,10 ,3,30 -(Ct5Bu2H2)2]2− (Fvtttt) along with the Cp ligand was employed to synthesize di- and trinuclear dysprocenium complexes, [{Dy(Cp )(m-BH4)}2(Fvtttt)] (43), [{Dy(Cp )(Fvtttt)}2Dy(m-BH4)3] (44) and [{Dy(Cp )(Fvtttt)}2Dy (m-BH4)3][B(C6F5)4] (45) (Scheme 8).52 44 is composed of a trimetallic Dy3 core where each pair of DyIII are bridged by a borohydride unit (BH4−). Dy1 and Dy2 are connected by the fulvalenyl ligand in a Z5:Z5 bridging mode (Fig. 24A) and so are Dy2 and Dy3. Dy1 and Dy3 additionally are bound to Cp ligands.

Scheme 8 Synthesis of 44–45. Adapted from Ref. He, M.; Guo, F. S.; Tang, J.; Mansikkamaki, A.; Layfield, R. A. Chem. Commun. 2021, 57, 6396–6399 with the permission of the Royal Society of Chemistry.

Organometallic Lanthanide Complexes as Single Molecule Magnets

401

Fig. 24 Molecular structure of complex 44 (A) and 45 (B). Color codes: pink, B; black, C; white, H; green, Dy. The principal magnetic axes of the local ground KDs of the DyIII ions in 44 (C) and 45 (D). Adapted from Ref. He, M.; Guo, F. S.; Tang, J.; Mansikkamaki, A.; Layfield, R. A. Chem. Commun. 2021, 57, 6396–6399 with the permission of the Royal Society of Chemistry.

The intramolecular Dy⋯ Dy distances for Dy1/Dy2, Dy2/Dy3 and Dy3/Dy1 are 4.741(4), 4.715(3) and 5.685(3) A˚ , respectively. In 45 there are only two BH4− units bridging Dy1-Dy2 and Dy2-Dy3 and no bridging between Dy1 and Dy3 (Fig. 24B). In 45, the intramolecular Dy


Dy distances for Dy1/Dy2, Dy2/Dy3 and Dy3/Dy1 are 4.880(4), 4.867(5), and 7.908(1) A˚ respectively. This indicates that the Dy centers in 45 have larger separation than in 44. The effective energy barrier (Ueff) for 44 and 45 are 138 (4) cm−1with t0 ¼ 5.44(7)  10−7 s and 411(23) cm−1 with t0 ¼ 4.16(2)  10−9 s, respectively at zero applied dc field. The higher Ueff for 45 has been rationalized as due to the less triangular shape of the Dy3 core in 45. This causes the DyIII centers in 45 to experience more axial ligand field from {(CpR)2Dy} units than in 44. This is further supported by the ab initio calculations which show the principal magnetic axes for ground state KDs of Dy1, Dy2, and Dy3 in 44 (Fig. 24C) and 45 (Fig. 24D).

14.11.3.5 Lanthanide metallocenes containing bridging radical ligands To understand the role of exchange coupling on the relaxation behavior of SMM, Long et al. prepared a series of (N23−) radical based complexes, [K(18-crown-6)(THF)2]{[(Me3Si)2N]2(THF)Ln}2(m-Z2:Z2-N2) (Ln ¼ Gd (46), Tb (47), Dy (48), Ho (49), Er (50); THF, tetrahydrofuran)53,54 It was shown that due to its large negative charge, the radical ligand N3− 2 is able to effectively interact with the deep seated f-orbitals. The exchange properties were evaluated in the magnetically isotropic GdIII complex 46 for which the intramolecular coupling constant (J) and intermolecular coupling constant (j) were found to be −27 cm−1 and 0.07 cm−1, respectively. Ac magnetic susceptibility measurement revealed a large effective energy barrier for both the complexes 47 (Fig. 25A) and 48 with Ueff ¼ 227 cm−1 (t0 ¼ 8.2(1)  10−9 s) and 123 cm−1 (t0 ¼ 8  10−9 s) respectively at zero applied dc field. 47 showed deviation from Arrhenius behavior at lower temperature and at this region the relaxation behavior was measured through the decay of magnetization which provided a relaxation time of 100 s at 13.9 K. 47 showed hysteresis loop up to 14 K with a maximum coercive field of 5 T below 11 K (Fig. 25B) while 48 showed a hysteresis loop up to 8.3 K with a maximum coercive field of 1.5 T. 49 and 50 also showed SMM behavior with Ueff ¼ 73(6) and 36(1) cm−1 respectively. radical, Long et al. reported a series of complexes, [K(crypt-222)(THF)][(CpMe4H Ln Combining CpMe4H with N3− 2 2 (THF))2(m− N2)] (crypt-222 ¼ 2.2.2-cryptand, THF ¼ tetrahydrofuran, CpMe4H ¼ tetramethylcyclopentadienyl, Ln ¼ GdIII (51), TbIII (52), DyIII (53)) and [K(crypt-222)][(CpMe4H Ln)2(m − N2)], (Ln ¼ TbIII (54), DyIII (55)).55 The blocking of magnetization 2

402

Organometallic Lanthanide Complexes as Single Molecule Magnets

Fig. 25 (A) Molecular structure of 47, Orange, Tb; green, Si; gray, C; blue, N; red, O. H atoms are omitted, and methyl groups are faded for clarity. (B) Magnetization vs field hysteresis plot for complex 47. 46, 48, 49 and 50 are isostructural to 47. Adapted from Ref. Rinehart, J. D.; Fang, M.; Evans, W. J.; Long, J. R. J. Am. Chem. Soc. 2011, 133, 14236–14239 with the permission of the American Chemical Society.

Fig. 26 Molecular structures of 52 (A) and 54 (B). Dark red, blue, and gray spheres represent Tb, N, and C atoms, respectively; H atoms have been omitted for clarity. 51 and 53 are isostructural to 52. 55 is isostructural to 54. (C) Magnetization versus field hysteresis plot for complex 54. Adapted from Ref. Demir, S.; Gonzalez, M. I.; Darago, L. E.; Evans, W. J.; Long, J. R. Nat. Commun. 2017, 8, 2144–2151 with the permission of Nature Publishing Group.

behavior in these complexes was confirmed by the divergence in the FC and ZFC data at 1 kOe, which showed bifurcation at 14.5 K for 52, 20 K for 54 and 7.5 K for 55. All the complexes were shown to be SMMs. The Ueff for 53, 52 and 55 were found to be 110(1), 242(2) and 108.1(2) cm−1 respectively. 53 showed a waist-restricted hysteresis even at 2 K while 52 and 55 showed a wider hysteresis loop with no indication of quantum tunnelling. The blocking temperatures for 52 and 55 were evaluated as 14 K and 6.6 K respectively. 54 (Fig. 26A) showed two relaxation pathways (Ueff values of 276 cm−1 and 564 cm−1). 54 also showed magnetic hysteresis up to 30 K and showed a giant coercive field of 7.9 T at 10 K which remains open down to 2 K (Fig. 26B). This hysteresis behavior is much better than that shown by even commercially available permanent magnets including Nd14Fe80B6 (iHc ¼ 1.39 T at 298 K and 3.90 T at 77 K) and SmCo5 (iHc ¼ 2.9 T at 298 K and iHc ¼ 4.3 T at 4.2 K). The above work simulated further exploration of radical based ligands in the field of SMMs. Bipyrimidine, tppz (2,3,5,6-tetra (2-pyridyl)-pyrazine) and indigo based ligands have been used as radical precursors. Organometallic complexes containing LnIII and these radical ligands, [(Cp 2Ln)2(m-bpym)][BPh4]56 (Cp ¼ pentamethylcyclopentadienyl; Ln ¼ GdIII (56); TbIII (57); DyIII (58)) [(Cp 2Ln)2(m-tppz)][BPh4]57 (Ln ¼ GdIII, 59; TbIII, 60; DyIII, 61; tppz ¼ 2,3,5,6-tetra(2-pyridyl)pyrazine) and [K(crypt-222)] [(Cp 2Ln)2(m-tppz)]58 (Ln ¼ GdIII, 62; TbIII, 63; DyIII, 64). [(Cp 2Dy)2(m-ind)] (65), [K(thf )6][(Cp 2Dy)2(m-ind)]
thf (66), and [{K(thf )3}2{Cp 2Dy}2(m-ind)] (67)58 were assembled and studied (Fig. 27A–C). The energy barriers for these complexes are given in Table 3.

Fig. 27 Molecular structures of 58 (A), 61 (B), and 65 (C). 56 and 57 are isostructural to 58. 59 and 60 are isostructural to 61. Green, blue, red and gray spheres represent Dy, N, O, and C atoms, respectively.

Organometallic Lanthanide Complexes as Single Molecule Magnets

Table 3

403

Magnetic relaxation parameters of the Cp-based lanthanide complexes that are SMMs/SIMs.

Complex

Ueff [cm−1]a (Exp.)

Ueff [cm−1]b (Theoretical)

Hdc [K Oe]c

t0 [s]d

TB (Hys) [K]e

References

[{Cp2Dy(m-bta)}2] (8) [(Z5-Cp)2Dy(m-Cl)]2 (10a) [(Z5-Cp)2Dy(m-Cl)]1 (10b) [(Z5-Cp)2(thf )Dy(m-Cl)]2 (11) [{Cp0 2Dy(mSSiPh3)}2] (12) [Cp 2Tb(BPh4)](13) [Cp 2Dy(BPh4)] (14) [Dy2Cp 4(m-BPh4)][Al(OC(CF3)3)4] (15) [Cp 2DyCl(THF)] (16) [Cp 2DyBr(THF)] (17) [Cp 2DyI(THF)] (18) [Cp 2DyCl2K(THF)]n (19) [Cp 2DyTp] (20) [Cp 2Dy(m-Fp)]2 (22) [Cp 2Dy(m-OC)W(Cp)(CO)(m-CO)]1 (23) [(CpMe)2Dy{m-P(H)Mes}]3 (24) [(CpMe)2Dy{m-As(H)Mes}]3 (25) [(CpMe)2Dy{m-Sb(H)Mes}]3 (26) [Li(THF)4]2[{(CpMe)2Dy(m-PMes)}3Li] (27) [Li(THF)4]2[{(CpMe)2Dy(m-AsMes)}3Li] (28) [(PyCp2)Dy-(m-O2SOCF3)]2 (29) [(PyCp2)Dy-(m-Cl)]2 (30) PyCp2Dy-FeCp(CO)2 (31) PyCp2Dy-RuCp(CO)2 (32) [Cp Yb(DAD)(THF)]
C7H8 (33) [(Cpttt)2Dy][B(C6F5)4] (35) [(Cpttt)2Dy][B(C6F5)4] (35) [Dy(Z5-CpiPr5)(BH4)2(THF)] (36) [(Z5-CpiPr5)Dy(Z5-Cp )(BH4)] (37) [(Z5-Cp )Dy(Z5-CpiPr5)][B(C6F5)4] (38) [Dy(CpiPr4H)2][B(C6F5)4] (39) [Dy(CpiPr4Me)2][B(C6F5)4] (40) [Dy(CpiPr4Et)2][B(C6F5)4] (41) [Dy(CpiPr5)2][B(C6F5)4] (42) [{Dy(Cp )(Fvtttt)}2Dy(m-BH4)3] (44) [{Dy(Cp )(Fvtttt)}2Dy(m-BH4)3][B(C6F5)4] (45) [K(18-crown-6)(THF)2]{[(Me3Si)2N]2(THF) Tb}2(m-Z2:Z2-N2) (47) [K(18-crown-6)(THF)2]{[(Me3Si)2N]2(THF) Dy}2(m-Z2:Z2-N2) (48) [K(18-crown-6)(THF)2]{[(Me3Si)2N]2(THF) Ho}2(m-Z2:Z2-N2) (49) [K(18-crown-6)(THF)2]{[(Me3Si)2N]2(THF) Er}2(m-Z2:Z2-N2) (50) [K(crypt-222)(THF)][(CpMe4H2Tb(THF))2(m–N2)] (52) [K(crypt-222)(THF)][(CpMe4H2Dy(THF))2(m–N2)] (53) [K(crypt-222)][(CpMe4H2Tb)2(m–N2)] (54) [K(crypt-222)][(CpMe4H2Dy)2(m–N2)] (55) [(Cp 2Tb)2(m-bpym•)] (BPh4) (57) [(Cp 2Dy)2(m-bpym•)] (BPh4) (58) [(Cp 2Ln)2(m-tppz•)][BPh4] (60) [(Cp 2Ln)2(m-tppz•)][BPh4] (61) [(Cp 2Dy)2(m-ind)] (65) [K(thf )6][(Cp 2Dy)2(m-ind)]
thf (66) [Cp 6Dy3(m3-HAN)] (70)

32.32  1.67 26.3  0.76 67.8  1.7 33.84  0.48 133  3.5 221 314 330 112 163 419 379 106 662(2) 557(18) 210(6) 256(5) 345 13(1) 23(2) 48.7 48.7 40 36 14  2 1277(14) 1223 241(7) 7(1) 1541(11) 1285 1468 1380 1334 138 (4) 411(23) 227

103 144 134 96 113 – – 460 147.2 165.8 423.1 414.2 153.8 639 386 290.63 (Avg.) 303.5 (Avg.) 416(3) 78 (Avg.) 83.53 (Avg.) 138 134 – – – 1156 – – – 1524 – – – – 151–190 324–643 –

0 0 0 0 0 2.4 1.6 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1.5 1.6 1.5 0 0 0 0 0 0 0 0 0 0 0 0

4.5  10−7 1.4  10−6 2.8  10−6 4.0  10−7 2.38  10−7 5  10−10 2  10−8 2.75  10−8 – – 3.38  10−11 – – 1.7(1)  10−12 3  10−12 6.53  10−9 2.01  10−9 1.57  10−10 7.75  10−7 2.99  10−7 4.8  10−7 7.2  10−7 1.5  10−6 1.41  10−6 1.74  10−6 8.12  10−12 1.98610−11 5.0(1)  10−3 7.6(5)  10−5 4.2(6)  10−12 3.39  10−12 4.01  10−12 7.79  10−12 1.18  10−11 5.44(7)  10−7 4.16(2)  10−9 8.2(1)  10−9

– – – – – – 5.3 6.5 2 3 4 5 2 6.2 – – – – – – – – – – – 60 60 – – 80 32 72 66 66 – – 14

36 37 37 37 38 39 39 40 41 41 41 41 41 42 43 44 45 46 44 45 47 47 48 48 49 22 50 23 23 23 51 51 51 51 52 52 54

123

–

0

8  10−9

8.3

53

73(6)

–

0

3(3)  10−8

–

54

36(1)

–

1

4(2)  10−11

–

54

242(2) 110(1) 276(1) 108.1(2) 44(2) 87.8(3) 5.1(1) 35.9(2) 39(1) 35(1) 51

– – – – – – – – – – –

0 0 0 0 0 0 0 0 0 0 0

1.4(2)  10−9 3.1(1)  10−9 1.3(1)  10−7 1.7(1)  10−8 4(1)  10−8 1.03(4)  10−7 6(1)  10−6 2.1(1)  10−7 5.08  10−5 1.6  10−8 1.2  10−8

15 – 30 8 – 6.5 – 3.25 – – 3.5

55 55 55 55 56 56 57 57 58 58 59

a

Effective energy barrier determined from ac susceptibility measurements. Theoretically calculated effective energy barrier. c Applied dc field used in ac susceptibility measurements. d Pre-exponential factor of Eq. (1) denoting the thermal relaxation rate. e Reported blocking temperature defined as the maximum temperature at which hysteresis was observed. b

404

Organometallic Lanthanide Complexes as Single Molecule Magnets

Fig. 28 (A) Molecular structure of 70, Green, blue, and gray spheres represent Dy, N, and C atoms, respectively; H atoms are omitted for clarity. (B) Magnetization versus field hysteresis loop for 70 from 1.8 to 3.5 K at a sweep rate of 4 mTs−1. 68 and 69 are isostructural to 70. Adapted from Ref. Gould, C. A.; Darago, L. E.; Gonzalez, M. I.; Demir, S.; Long, J. R. Angew. Chem. Int. Ed. 2016, 129, 10237–10241 with the permission of Wiley-VCH.

Trinuclear lanthanide complexes containing radical ligands, [Cp 6Ln3(m3-HAN)] (Ln ¼ Gd, 68; Tb, 69; Dy, 70)59 (HAN] hexaazatrinapthylene) were synthesized and studied. Each lanthanide ion is bound to two Cp and two nitrogens from the bridging HAN moiety (Fig. 28A). From the dc magnetic data analysis it was found that all the complexes have antiferromagnetic interactions. Complex 70 shows SMM behavior with Ueff of 51 cm−1 and t0 ¼ 1.2  10−8 s at no applied dc field and an open hysteresis loop up to 3.5 K (Fig. 28B). Magnetic relaxation parameters for the Cp-based lanthanide complexes that are SMMs/SIMs are presented in Table 3.

14.11.4 h5-Dicarbollide dianion based dysprosiacarborane SMMs − The dicarbollide dianion, nido-C2B9H2− 11 is isoelectronic and isolobal (Fig. 29A) with Cp and hence has been recently employed to obtain four sandwich-type lanthanacarboranes [Na(THF)5][(C2B9H11)2Ln(THF)2] (Ln ¼ DyIII, 71; Ln ¼ HoIII, 72; Ln ¼ ErIII, 73;

Fig. 29 (A) The schematic drawings of cyclopentadiene ion (Cp−) and nido-C2B9H2− 11 ion showing approximately five p orbitals perpendicular to Cp ring and sp3 orbitals directed at the vacant vertex in the C2B9H2− 11 cage, respectively. (B, C) Two views of the molecular structure of 71. The principal magnetic axis of the ground Kramers’ doublet is shown as an arrow (B). Color codes: black, C; orange, B; red, O; green, Dy. Thermal ellipsoids are drawn at the 30% probability level and all the hydrogen atoms are omitted for clarity. 72, 73 and 74 are isostructural to 71. Adapted from Ref. Jin, P. B.; Zhai, Y. Q.; Yu, K. X.; Winpenny, R. E. P.; Zheng, Y. Z. Angew. Chem. Int. Ed. 2020, 59, 9350–9354 with the permission of Wiley-VCH.

Organometallic Lanthanide Complexes as Single Molecule Magnets

405

Fig. 30 (A) The synthetic route from 75 to 76. (B and C) Two views of the molecular structure of 76. Color codes: black, C; orange, B; red, O; green, Dy; purple, Li. Thermal ellipsoids drawn at the 30% probability level and all the hydrogen atoms are removed for clarity except those coordinating to Dy or Li. The principal magnetic axis of the ground Kramers’ doublet is shown as an green arrow. Magnetization versus field hysteresis loop for complexes 710 (D) and 76 (E). Adapted from Ref. Jin, P. B.; Zhai, Y. Q.; Yu, K. X.; Winpenny, R. E. P.; Zheng, Y. Z. Angew. Chem. Int. Ed. 2020, 59, 9350–9354 with the permission of Wiley-VCH.

Ln ¼ YIII, 74).60 All of these four complexes are isostructural. Structural analysis of the DyIII analog revealed DyIII center of 71 is in a 2− ligand distorted tetrahedral coordination environment bound to two C2B9H2− 11 ligands and two THF molecules. The C2B9H11 binds with the metal center with it’s C2B3 pentagonal face in an Z5 fashion (Fig. 29B). The angle at the LnIII between the centroids of 2−  ˚ and 2.291(2) A˚ . the two C2B9H2− 11 is 128 and Dy ⋯ face centroid (C2B9H11) distances are 2.289(2) A Similarly a half-sandwich mononuclear dysprosiacarborane [(THF)3(m-H)3Li]2[{Z5-C6H4(CH2)2C2B9H9}Dy{2:5-C6H4(CH2)2C2B9H9}2Li] (76) was synthesized from [(THF)3(m-H)3Li]2[(THF)(m-H)4Li][{5-C6H4(CH2)2C2B9H9}2Li] (75) which is a III known C2B2− center where it is bound to the pentagonal 9 system (Fig. 30A). 76 forms a half sandwiched structure around Dy 2− 5 C2B3 face of one [o-xylylene-C2B9H9] unit in a Z mode (Fig. 30B). The other coordinating sites are occupied by three BdH- - -Dy interactions and these two interactions come from the two dicarbollide ligands which sandwich the Li+ ion. Fitting of ac magnetic susceptibility data provided effective energy barriers of 298.9(3.5) cm−1 with t0 ¼ 1.2(3)  10−9 s and 558.8(4.9) cm−1 with t0 ¼ 4.0(3)  10−11 s for 71 and 76 respectively. From ab initio calculations it was found that the principal magnetic axis of the 5 2− face (Fig. 30C). Thermal ground state for 71 lies toward C2B9H2− 11 unit whereas for 76 it goes through the center of  -C2B3 relaxation of 71 involves the fourth and fifth KD whereas for 76 fifth, sixth and seventh KDs are involved. The diluted complex 71@74 (710 ) and 76 showed hysteresis loops which remain open up to 4.9 and 6.8 K respectively (Fig. 30D and E).

14.11.5 h6-Arene ligated dysprosium complexes as SMMs Having looked at cyclopentadienyl complexes in the previous section, let us look at complexes involving arenes which bind in a Z6 manner. A neutral complex involving hexamethylbenzene (HMB), [Dy(HMB)(AlCl4)3] (77) was prepared and characterized.61 HMB is coordinated to DyIII in a Z6 fashion. Each Al center is bridged with DyIII by chloride anions (Fig. 31A). The geometry around DyIII is puasi-Cs. The DyIII-centroid(HMB) distance (2.471 A˚ ) is a little larger than the average DyIII-centroid(Cp) distances in various Cp based lanthanide complex SMMs (2.60–2.82 A˚ ). The angle between the HMB and the equatorial plane connecting the five Cl atoms (Cl1, Cl2, Cl5, Cl9, and Cl10) is 3.48o. Ac magnetic susceptibility measurement at zero applied dc field shows characteristic out of phase (w00 ) peak and dependence on frequency (Ueff ¼ 70.2 cm−1 with t0 ¼ 5.1  10−10 s). However quantum tunnelling appeared to be dominant below 7 K. The ab initio calculations show that the easy axis of magnetization lies along the axis passing through Dy-centroid (HMB) with a little tilting angle of 2.5o (Fig. 31B and C).

406

Organometallic Lanthanide Complexes as Single Molecule Magnets

Fig. 31 Molecular structure of 77 (A). Color Codes: Teal, Dy; light green, Cl; sky blue, Al; gray. C and hydrogen atoms are omitted for clarity. The calculated easy axis orientation by the ab initio method (red arrow) viewed perpendicular (B) and parallel (C) to the Dy–centroid (HMB) direction. Adapted from Ref. Liu, S. S.; Ziller, J. W.; Zhang, Y. Q.; Wang, B. W.; Evans, W. J.; Gao, S. Chem. Commun. 2014, 50, 11418–11420 with the permission of Royal Society of Chemistry.

Theoretically calculated difference between the ground and first excited Kramers doublet (79.93 cm−1) matches well with the experimental energy barrier. A similar complex was synthesized by Liu et al. in 2017 where they have used toluene as the arene moiety. Two complexes [(C7H8)Dy(AlCl4)3] (78) and [(C7H8)Dy(AlBr4)3] (79) were reported.62 Both complexes showed similar structural features. The energy barriers for 78 at zero and 2000 Oe applied field are 77.15 cm−1 (t0 ¼ 2.9  10−9 s) and 108.4 cm−1 (t0 ¼ 7.3  10−11 s) respectively whereas the energy barriers for 79 at zero and 2000 Oe applied field are 76.45 cm−1 (t0 ¼ 1.8  10−9 s) and 117.5 cm−1 (t0 ¼ 8.1  10−12 s) respectively (Table 4). Also both the complexes showed butterfly-like hysteresis loop having loop closing temperatures of 4 K (78) and 3.5 K (79).

Table 4

Magnetic relaxation parameters of the COT ligated lanthanide complexes behaving as SMMs/SIMs.

Complex

Ueff [cm−1]a (Exp.)

Ueff [cm−1]b (Theoretical)

Hdc [K Oe]c

t0 [s]d

[(Cp )Dy(COT)] (87) [(Cp )Ho(COT)] (88)

17.65 2.5 5.14 224.5 136.9 147(1) 147(1) 198.8(1.4) 7.65(0.7) 12.5 130(0.7) 17.4 6.3 224.5(2.1) 119.5 208.5 113 95.6(9) 102.9(3.1) 107.1(1.3) 133.6(2.2) 30 26 20 259 300 174 248.8(2.1)

24.3 80.6

0 0

189.4

0

– – 186.7 13.7 – – – 25.3 230.6 178.1 186.4 – 99.8 90.1 89.8 138 56.2 33.6 39.7 201 223.5 458.8 –

0 0 0 0 0 0 0 0 0 0

5  10 1.3  10−4 4  10−5 8.17  10−11 3.13  10−9 8.3(6)  10−8 8.3(6)  10−8 3.7  10−9 2.2  10−5 6  10−6 4  10−8 6  10−6 1.8  10−5 5.7  10−10 3.6  10−10 8.6  10−9 4.14  10−10 9.2(1)  10−10 9.6(2.7)  10−10 6.3(8)  10−10 9(1.4)  10−10 5.2  10−10 1.2  10−9 6.8  10−9 5.3  10−12 5.5  10−12 9.2  10−10 1.6(3)  10−11

[(Cp )Er(COT)] (89) [K(18-c-6)][Er(COT)2]
2THF (92) [K(18-c-6)(THF)2][Er(COT)2] (93) [K(18-crown-6)][Er(COT)2] (94) [K(18-crown-6)][Dy(COT)2] (95) [DyIII(COT00 )2Li(THF)(DME)] (96) [Li(DME)3][ErIII(COT00 )2] (97) [DyIII(COT00 )2][Li(DME)3] (98) [DyIII2(COT00 )3] (99) [ErIII2(COT00 )3] (101) K2(THF)4[ErIII2(COT)4] (103) [Er(m2-Cl)(COT)(THF)]2 (104) Er(COT)I(THF)2 (105) Er(COT)I(Py)2 (106) Er(COT)I(MeCN)2 (107) Er(COT)(Tp ) (108) [(C5H5BH)Dy(COT)] (109) [(C5H5BMe)Dy(COT)] (110) [(C5H5BNMe2)Dy(COT)] (111) [(C5H5BH)Er(COT)] (112) [(C5H5BMe)Er(COT)] (113) [(C5H5BNMe2)Er(COT)] (114) Er(COT)(dsp) (115) a

0 0 0 0 0 1.2 1.5 1.5 0 0 0 0

Effective energy barrier determined from ac susceptibility measurements. Theoretically calculated effective energy barrier. c Applied dc field used in ac susceptibility measurements. d Pre-exponential factor of Eq. (1) denoting the thermal relaxation rate. e Reported blocking temperature defined as the maximum temperature at which hysteresis was observed. b

−7

TB (Hys) [K]e

References

– –

66 66

5

65

10 10 12 – – 8 – 12 12 12

67 67 68 68 69 70 71 71 72 72

4 – – – – – – – 8 6 – 9

73 74 74 74 74 75 75 75 75 75 75 76

Organometallic Lanthanide Complexes as Single Molecule Magnets

407

Fig. 32 Molecular structure (A) and (B) out of phase (w00 ) ac data vs frequency plot for complex 80. Color codes: gray, C; sky blue, Dy; red, O. 81 is isostructural to 80. Adapted from Ref. Meng, Y. S.; Xu, L.; Xiong, J.; Yuan, Q.; Liu, T.; Wang, B. W.; Gao, S. Angew. Chem. Int. Ed. 2018, 57, 4673–4676 with the permission of Wiley-VCH.

Fig. 33 Ab initio calculated easy axis for 80 (A) and 81 (B). The orientation of easy axis for 80 forms the angle of 19.9 and 17.3 with Dy-O1 and Dy-O2 direction, respectively. Adapted from Ref. Meng, Y. S.; Xu, L.; Xiong, J.; Yuan, Q.; Liu, T.; Wang, B. W.; Gao, S. Angew. Chem. Int. Ed. 2018, 57, 4673–4676 with the permission of Wiley-VCH.

A bulky phenolic ligand, ArOH (Ar ¼ 2,6-Dipp2C6H3, Dipp ¼ 2,6-diisopropylphenyl) was employed to assemble [(ArO)Ln(OAr0 )] (Ar0 ¼ 6-Dipp-2-(20 -iPr-60 -CHMe(CH−2)C6H3)C6H3O−; Ln ¼ DyIII (80), ErIII (81)).63 In these complexes the aryl substituent binds in a Z6 manner. In addition, a methyl group of one of the ArOH ligands gets deprotonated and bind with the metal center. (Fig. 32A). 80 showed temperature and frequency dependence of the in and out phase of ac data in the range 2 K to 70 K, at zero dc field (Fig. 32B). The analysis of ac data provided an effective energy barrier of 667.9 cm−1 with t0 ¼ 8.2  10−12 s. Complex 81 showed fast QTM process (tQTM ¼ 0.025 ms) in the absence of applied field and showed characteristic SIM features in the presence of 1000 Oe dc field [provided Ueff ¼ 60.5 cm−1 and t0 ¼ 1.7  10−8 s]. Ab initio calculations revealed that the orientation of the easy axis of magnetization in 80 is almost linear with the O1-Dy-O2 motif (Fig. 33A). For 81 the easy axis is out of the plane making an angle of 77 (Fig. 33B). Consequently, 80 is found to have small but non-negligible transversal components (gx ¼ 0.0002; gy ¼ 0.0003) whereas 81 has a larger transversal components (gx ¼ 0.02; gy ¼ 0.04) which may be responsible for faster QTM process (tQTM ¼ 0.025 ms) for 81. The spin reversal occurs probably from second excited KD for complex 80.

14.11.6 h7-Cycloheptatrienyl ligand based lanthanide complexes as SMMs Cycloheptatrienyl trianion is a versatile ligand which has been used quite widely in organometallic chemistry. Using this ligand, the complexes [KLn2(C7H7)(N(SiMe3)2)4] (Ln ¼ GdIII (82), DyIII(83), ErIII (84)) and [K(THF)2Er2(C7H7)(N(SiMe3)2)4] (85) were

408

Organometallic Lanthanide Complexes as Single Molecule Magnets

Fig. 34 Molecular structures of 84 (A), 85 (B) and ground state magnetic anisotropy axis for complex 83 (C) and 84 (D). Color codes: gray, C; blue, N; light pink, K; green, Si; teal, Er; purple, Dy. Green arrows show the orientation of local magnetic moments on Ln sites in the ground exchange coupled state. For 84, Ueff ¼ 40.3 cm−1 and t0 ¼ 2.9  10−8 s (HDC ¼ 800 Oe). Adapted from Ref. Harriman, K. L. M.; Le Roy, J. J.; Ungur, L.; Holmberg, R. J.; Korobkov, I.; Murugesu, M. Chem. Sci. 2016, 8, 231–240 with the permission of the Royal Society of Chemistry.

prepared.64 Complexes 82–84 are isostructural where the cycloheptatrienyl trianion functions as a bridging unit between two lanthanide centers in a Z7 binding mode. The remaining coordination sites around the lanthanide ions contain two [N(SiMe3)2]− ligands. Also a potassium (K+) ion is bound to the two nitrogen atoms of the two [N(SiMe3)2]− units (Fig. 34A and B). These complexes were found to show SMM behavior. Ab initio calculations showed that for 83 the easy axis is almost lying in the N-Dy-N plane (Fig. 34C) whereas for 84 it is almost perpendicular to the N-Er-N plane (Fig. 34D). This opposite sense of direction of the easy axis of magnetization for complex 83 and 84 in the same axial ligand field environment arises from the fact that they possess opposite sign of Stevens parameters, a and b which are related to the second and fourth rank operators of the ground ionic J15/2 multiplet of DyIII and ErIII.

14.11.7 h8-Cyclooctatetraenyl ligand based lanthanide complexes as SMMs 14.11.7.1 [CpLn(COT)] and [Ln(COT)2]− motif-based complexes Considering the importance of symmetric and low coordinated lanthanide complexes, cyclooctatetraene (COT) ligand also has been used along with the cyclopentadienyl ligand to assemble [(Cp )Ln(COT)] (Ln ¼ TbIII (86), DyIII (87), HoIII (88), ErIII (89), TmIII (90), YIII (91))65,66 In these complexes the LnIII is sandwiched between the two different aryl ligands. In 89 the aromatic rings are not perfectly parallel to each other but make an angle of 8o with respect to each other (Fig. 35A) and due to this tilting the ErIII complex adopts a Cs point group. Among these molecules only ErIII and DyIII analogs show significant blocking of magnetization. AC magnetic studies revealed that the ErIII complex relaxes through QTM below 10 K. Above 10 K two clear peak maxima were found which may be due to the presence of two different conformers present in the crystal. Analysis of this data yielded two sets of relaxation parameters [Ueff ¼ 224.5 cm−1, t0 ¼ 8.17  10−11 s and Ueff ¼ 136.9 cm−1, t0 ¼ 3.13  10−9 s]. Butterfly shaped hysteresis loop is obtained for the ErIII complex (89) with Hc of 100 Oe at 1.8 K, the loop closing at 5 K (Fig. 35B). The butterfly shape of the hysteresis loop is due to the QTM which arises due to the tilting of the two aromatic rings at a certain angle and mixing of the

Organometallic Lanthanide Complexes as Single Molecule Magnets

409

Fig. 35 (A) Structure of 89: pink ball represents terbium, and orange balls represent carbon atoms. Hydrogen atoms are omitted for clarity. (B) Magnetization vs field plot from 0.5–5 K for 86. 86, 87, 88, 90 and 91 are isostructural to 89. (A) Adapted from Ref. Jiang, S. D.; Liu, S. S.; Zhou, L. N.; Wang, B. W.; Wang, Z. M.; Gao, S. Inorg. Chem. 2012, 51, 3079–3087 with the permission of the American Chemical Society. (B) Adapted from Ref. Jiang, S.-D.; Wang, B.-W.; Sun, H.-L.; Wang, Z.-M.; Gao, S. J. Am. Chem. Soc. 2011, 133 (13), 4730–4733 with the permission of the American Chemical Society.

Fig. 36 Molecular structures of complexes 92 (A) and 93 (B). Color Codes: Yellow (Dy), gray (C), red (O) with H atoms omitted for clarity. Adapted from Ref. Meihaus, K. R.; Long, J. R. J. Am. Chem. Soc. 2013, 135, 17952–17957 with the permission of the American Chemical Society.

Kramers doublets. The DyIII analog shows a frequency dependence of the w00 of ac data which is very weak. A diluted sample was studied (5% Dy@Y) for which an energy barrier of 23 cm−1 was detected at 100 Oe applied dc field. A series of lanthanide complexes have been synthesized containing only the COT ligand among which the ErIII complexes67 [K(18-c-6)][Er(COT)2]
2THF (92) and [K(18-c-6)(THF)2][Er(COT)2] (93) (Fig. 36) showed significant SMM behavior. Relaxation parameters for these complexes are given in Table 4. Other examples of this family include [K(18-crown-6)][Ln(COT)2], Ln ¼ ErIII (94), DyIII (95) (Fig. 37).68 Both these complexes showed slow relaxation of magnetization and the ErIII analog showed a very large coercive field of 7 T at 1.8 K and the hysteresis loop persists up to 12 K. The effective energy barriers for 94 and 95 were found to be 198.8(1.4) cm−1 with t0 ¼ 3.7  10−9 s and 7.65(0.7) cm−1 with t0 ¼ 2.2  10−5 s respectively. Ab-initio calculations were useful in rationalizing this behavior as the ground state wave function for ErIII is almost pure 15/2 type with no mixing from other J projections. For the DyIII complex, however, the ground state is 9/2 type with major contributions from other J projections. 1,4-Bis(trimethylsilyl)cyclooctatetraenyl dianion (COT00 ) was used to synthesize [DyIII(COT00 )2Li(THF)(DME)] (96).69 Due to the presence of the trimethylsilyl group in the COT ring the compound attains a staggered conformation as confirmed from the crystal structure. (Fig. 38A). The complex shows SIM behavior at zero applied field (Fig. 38B) with a Ueff ¼ 12.5 cm−1 and t0 ¼ 6  10−6 s in the high temperature thermally activated region. In the lower temperature region the complex relaxes through QTM which can be suppressed with an optimum field of 600 Oe. A similar complex [Li(DME)3][ErIII(COT00 )2], (97) was synthesized where the terminal COT00 rings were tilted slightly from linearity with an angle of 3.6o.70 This complex showed a large coercive field of 6.25 T at 1.8 K which closes above 8 K. Analysis of the ac magnetic data provided an effective energy barrier of 130(0.7) cm−1 with t0 ¼ 4  10−8 s at zero applied dc field.

410

Organometallic Lanthanide Complexes as Single Molecule Magnets

Fig. 37 Molecular structures of [LnIII(COT)2] magnetic anions. (A) side and (B) top view for Ln ¼ Er; (C) side and (D) top view for Ln ¼ Dy. The [K(18-crown-6)] + cation, which is coordinated to one of the COT rings, is not shown for clarity. Color codes: Green (Er), gray (C), red (Dy) white (H). The transparent red surface shows the calculated electronic density in the ground state. Note the higher rotational symmetry of the electronic cloud close to the lanthanide ion than that expected from an octagonal group. Dashed lines show the calculated orientation of the main magnetic axis on the lanthanide ions in the ground (1) and first excited (2) Kramers doublet. For Er: g X,Y ¼ 3.5  106, gZ ¼ 17.96 for the ground doublet and g X,Y ¼ 5.4  104, gZ ¼ 15.53 for the first excited doublet; the angle between corresponding gZ axes is about 1.0 . For Dy: gX,Y ¼ 1.6  101; gZ ¼ 12.64 for the ground doublet (green dashed line) and gX,Y ¼ 5.8  102; gZ ¼ 13.84 for the first excited doublet (purple dashed line); the angle between corresponding gZ axes is about 21 . Adapted from Ref. Ungur, L.; Roy, J. J. L.; Korobkov, I.; Murugeshu, M.; Chibotaru, L. F. Angew. Chem. Int. Ed. 2014, 53, 1–6 with the permission of Wiley-VCH.

Fig. 38 Molecular structure (A) and w00 vs frequency plot (B) for 96. Color Codes: Yellow (Dy), gray (C), green (Si) with H atoms omitted for clarity. Adapted from Ref. Jelectic, M.; Lin, P. H.; Roy, J. J. L.; Korobkov, I.; Gorelsky, S. I.; Murugeshu, M. J. Am. Chem. Soc. 2011, 133, 19286–19289 with the permission of the American Chemical Society.

Using the same ligand, [DyIII(COT00 )2][Li(DME)3] (98) (Fig. 39A) and two double decker complexes, [DyIII2(COT00 )3] (99) and 71 00 were synthesized. COT00 coordinates to the DyIII anion in a Z8 fashion while the counter cation Li+ is [GdIII 2 (COT )3] (100) octahedrally coordinated by three 1,2-dimethoxyethane (DME). In 99 the terminal COT00 binds with DyIII in a Z8 mode and the central COT00 binds in a m-Z8Z8 bridging mode with a DydDy distance of 4.14 A˚ (Fig. 39B). The terminal COT00 rings are in near parallel position with a slight tilting angle of 1.86o. wMT vs T data for all the complexes revealed significant antiferromagnetic interactions. From ac magnetic susceptibility data, the effective energy barrier was found to be 17.4 cm−1 with t0 ¼ 6.1  10−6 s for 98 and 6.3 cm−1 with t0 ¼ 1.8  10−5 s for 99 at zero applied dc field.

Organometallic Lanthanide Complexes as Single Molecule Magnets

411

Fig. 39 (A) Molecular structure of 98 and Dy


COT00 inner centroid distance of 1.90 A˚ for 98. (B) Molecular structure of 99 and Dy


Dy distance of 4.14 A˚ , Dy


COT00 outer centroid distance of 1.79 A˚ , Dy


COT00 inner centroid distance of 2.07 A˚ for 99. Color Codes: Yellow (Dy), gray (C), green (Si) with H atoms omitted for clarity. Orientation of the main magnetic axes of the ground Kramers doublets on DyIII centers of (C) the initial complex 99; (D) the symmetrized 99; (E) the symmetrized 98. In (C, D, and E) DyIII is indicated by purple color. Adapted from Ref. Roy, J. J. L.; Jelectic, M. S.; Gorelsky, S. I.; Korobkov, I.; Ungur, L.; Chibotaru, L. F.; Murugeshu, M. J. Am. Chem. Soc. 2013, 135, 3502–3510 with the permission of the American Chemical Society.

Ab initio calculations revealed that the local magnetic axis does not coincide with the axis connecting two DyIII ions (Fig. 39C) and from the crystal field parameters calculations it was found that the equatorial ligand field strength is more than axial ligand field strength. Theoretical calculations on the model complex where the SiMe3 groups are replaced with hydrogen showed that the ground state J corresponds to J ¼ 9/2 and the magnetic axis is oriented along the symmetry axis of the molecule (Fig. 39D and E). Other double- and triple decker lanthanide complexes were synthesized using COT00 (1,4-trimethylsilyl cyclooctatetraene), 00 (101) (Fig. 40A) and K2(THF)4[LnIII (Ln ¼ Gd (102), Er (103); THF ¼ tetrahydrofuran, [ErIII 2 (COT )3] 2 (COT)4] COT ¼ cyclooctatetraenyl dianion).72 In the ErIII complex (101) the Er-COT00 -Er angle was found to be 175.7o. The triple decker complex was shown to have a tetralayer structure (Fig. 40B) where two sandwiched [Er(COT00 )2]− units bind to one K+ in a Z8 mode and another K(THF)4 binds with one terminal COT00 also in a Z8 fashion. The K1dEr1dK2 and Er1dK2dEr2 angles were found to be 179.4o and 169.6o respectively which indicates that in the latter the angle deviates significantly from linearity. The ErdEr distance is 8.82 A˚ in complex 103 which is much larger than the ErdEr distance in 101 (4.11 A˚ ). Both the complexes showed hysteresis loops which remain open until 12 K. Both the complexes show slow magnetic relaxation.

412

Organometallic Lanthanide Complexes as Single Molecule Magnets

Fig. 40 Molecular structures of 101 (A) and 103 (B). Color codes: gery, C; green, Si; light green, K; red, O; blue, Er. Adapted from Ref. Roy, J. J. L.; Ungur, L.; Korobkov, I.; Chibotaru, L. F.; Murugeshu, M. J. Am. Chem. Soc. 2014, 136, 8003–8010 with the permission of the American Chemical Society.

Fig. 41 (A) Solid-state structure of 104 with spheres representing Er (pink), Cl (green), O (red), and C (gray). Hydrogen atoms have been omitted for clarity. (B) outof-phase ac data (w00 ) versus frequency plot at various temperatures. Adapted from Ref. Hilgar, J. D.; Flores, B. S.; Rinehart, J. D. Chem. Commun. 2017, 53, 7322–7324 with the permission of the Royal Society of Chemistry.

A m-chloro bridged Er2 dimer [Er(m2-Cl)(COT)(THF)]2 (104) (Fig. 41A) was prepared where both the ErIII are crystallographically equivalent and were shown to have ferromagnetic coupling.73 The ac magnetic susceptibility studies revealed a frequency dependence of the out-of-phase ac data (Fig. 41B) at zero applied field [Ueff ¼ 113 cm−1 and relaxation time t0 ¼ 4.14  10−10 s]. A synthetic strategy where ErIII is coordinated with one COT ligand on the one side while the other side is open for coordination to other ligands was adopted and the complexes, Er(COT)I(THF)2 (THF ¼ tetrahydrofuran) (105) (Fig. 42A), Er(COT)I(Py)2 (Py ¼ pyridine) (106) (Fig. 42B), Er(COT)I(MeCN)2 (MeCN ¼ acetonitrile) (107) (Fig. 42C), and Er(COT)(Tp ) (108) (Tp ¼ tris(3,5-dimethyl-1-pyrazolyl)borate) (Fig. 42D) were prepared. All of these showed slow magnetic relaxation with multiple relaxation pathways.74 The effective energy barriers for 105–108 are Ueff ¼ 95.6(9), 102.9(3.1), 107.1(1.3), and 133.6 (2.2) cm−1 respectively. The higher energy barrier for 108 arises from the fact that this complex has a higher symmetry compared to the other complexes, which leads to better splitting between the ground and first excited KD of the J ¼  15/2 multiplet.

Organometallic Lanthanide Complexes as Single Molecule Magnets

413

Fig. 42 Solid-state structures of 105–108 with spheres representing Er (pink), I (purple), O (red), N (blue), C (gray), and B (salmon). Hydrogen atoms and outer-sphere solvent have been omitted for clarity. Black lines depict the direction of the main magnetic axis of the ground Kramers doublet. Adapted from Ref. Hilgar, J. D.; Bernbeck, M. G.; Flores, B. S.; Rinehart, J. D. Chem. Sci. 2018, 9, 7204–7209 with the permission of the Royal Society of Chemistry.

14.11.7.2 COT-ligated lanthanide complexes with heteroaromatic ligands Similarity between the boratabenzene and Cp ligands gave an opportunity to prepare complexes where the exocyclic substituent on boron can be changed and the effective energy barrier can be tuned. Keeping this in mind two series of complexes were synthesized, [(C5H5BR)Ln(COT)] where Ln ¼ DyIII, R ¼ H (109), Me (110), NMe2 (111) and for Ln ¼ ErIII, R ¼ H (112), Me (113), NMe2 (114).75 From the structural analysis it was found that the Ln-benzeneborate centroid distance (2.245–2.257 A˚ ) is larger than Ln-COT centroid distance (1.674–1.679 A˚ ). Ac magnetic susceptibility analysis showed that 109–111 show low effective energy barriers of 20–30 cm−1 under an applied field. On the other hand, 112–114 showed characteristic ac data for slow relaxation at zero applied field. Effective energy barriers for complex 112 (Fig. 43A) and 113 (Fig. 43B) are 259 cm−1 with t0 ¼ 5.3  10−12 s and 300 cm−1 with t0 ¼ 5.5  10−12 s respectively. These energy barriers are much higher than the energy barrier found in the analogs Er(COT)(Cp ) (89) (224.5 cm−1). These higher activation energy barriers are explained on the basis of poor electron donation ability of the boron atom which decreases the electronic repulsion between the f-orbitals and aromatic electron clouds along uniaxial direction and thus increasing the axial nature of the magnetic anisotropy. 114 revealed a comparatively low energy barrier of 174 cm−1 with no applied field. This decrease in Ueff is attributed to two facts, one is that NMe2 group is more electron donating than H or Me and second the boron atom in this complex is out of the boratabenzene plane as compared to the other two

Fig. 43 Molecular structures of 112 (A), 113 (B), and 114 (C). Color code: pink, Er, dark gray, C, yellow, B, blue, N. Hydrogen atoms are omitted for clarity. 109, 110 and 111 are isostructural with 112, 113 and 114 respectively. Adapted from Ref. Meng, Y. S.; Wang, C. H.; Zhang, Y. Q.; Leng, X. B.; Wang, B. W.; Chen, Y. F.; Gao, S. Inorg. Chem. Front. 2016, 3, 828–835 with the permission of the Royal Society of Chemistry.

414

Organometallic Lanthanide Complexes as Single Molecule Magnets

Fig. 44 Top view (A) and side view (B) of molecular structure of 115 ellipsoids (30% possibility). Color codes: pink, Er; orange, P; tan, Si; black, C. Adapted from Ref. Chen, S. M.; Xiong, J.; Zhang, Y. Q.; Yuan, Q.; Wang, B. W.; Gao, S. Chem. Sci. 2018, 9, 7540–7545 with the permission of the Royal Society of Chemistry.

complexes (Fig. 43C) and thus it may provide some transverse component to the ligand field leading to some loss of the magnetic anisotropy. 112 and 113 showed hysteresis loop up to 8 K and 6 K, respectively. Similar to the above, a phosphorous analog of the Cp , phosphacyclopentadienyl (phospholyl) (dsp−) ligand has been used to synthesize [Er(COT)(dsp)] (115) (Fig. 44).76 Replacement of a C by P makes dsp a softer base than Cp and thus coordinates less strongly than Cp and this affects the Er-COT distance. Thus the Er-COT distance in this complex (1.686 A˚ ) is less than in Er(COT) (Cp ) (89) (1.727 A˚ ). The tilting angle between the dsp− and COT2− is 10.5o. Ac magnetic susceptibility shows SIM behavior at zero applied magnetic field and an energy barrier of 248.8(2.1) cm−1 which is much higher than that of [Er(COT)2]− (93) (147(1) cm−1) and (Cp )Er(COT) (89) (136.9, 224.5 cm−1). Magnetic blocking temperature for this complex as found from the magnetization versus field hysteresis is 9 K which is higher than that found in 89. The magnetic relaxation parameters of the COT ligated lanthanide complexes behaving as SMMs/SIMs are summarized in Table 4.

14.11.8 Methanide and bismethane(diide) ligand based lanthanide complexes as SMMs In order to obtain linear complexes, bismethanide type of ligands were employed which resulted in the formation of [Dy(BIPMTMS) (BIPMTMSH)] (116); (BIPMTMS ¼ {C(PPh2NSiMe3)2}2−; BIPMTMSH ¼ {HC(PPh2NSiMe3)2}−] and [Dy(BIPMTMS)2] [K(18C6) (THF)2] (117) (Scheme 9).77

Scheme 9 Synthesis of 116 and 117. Adapted from Ref. Gregson, M.; Chilton, N.F.; Ariciu, A. M.; Tuna, F.; Crowe, I. F.; Lewis, W.; Blake, A. J.; Collison, D.; McInnes, E. J. L.; Winpenny, R. E. P.; Liddle, S. T. Chem. Sci. 2016, 7, 155–165 with the permission of the Royal Society of Chemistry

The DyIII centers in both 116 and 117 are hexacoordinated. While 116 is neutral as it contains one methanide and one methanediide ligands, 117 is anionic having two methanediides that are coordinated to the DyIII center. The C]DydC angle is 158.25(6)o for 116 (Fig. 45A) but for 117 (Fig. 45B), the C]Dy]C angle is almost linear at 176.6(2)o. Dy]C and DydC bond lengths are 2.3640(17) and 2.9001(18) A˚ for 116 and Dy]C bond distances are 2.434(6) and 2.433(6) A˚ for 117. As expected from the linearity of the CdDydC motif, 117 displayed effective energy barriers of 501 cm−1 with t0 ¼ 1.11(3)  10−12 s, and 565 cm−1 with t0 ¼ 5.65(20)  10−13 s (117 showed two relaxation pathways) which is much higher than the energy barrier of 177 cm−1 with t0 ¼ 3.55(9)  10−12 s obtained for 116. Also from the ab initio calculations it becomes clear that due to the substantial linearity of the C]Dy]C moiety and the higher charge accumulation due to the methanediide anions in 117, the ground state KD is stabilized and thus energy separation with the ground state to the excited state KDs increases. The blocking temperature for complex 117 has been estimated to be about 10–12 K.

Organometallic Lanthanide Complexes as Single Molecule Magnets

415

Fig. 45 Molecular structures 116 (A) and 117 (B). Color codes: gray, C; red, O; black, K; yellow, Si; pink, P; green, Dy. Thermal ellipsoids are at 30% probability level. Adapted from Ref. Gregson, M.; Chilton, N. F.; Ariciu, A. M.; Tuna, F.; Crowe, I. F.; Lewis, W.; Blake, A. J.; Collison, D.; McInnes, E. J. L.; Winpenny, R. E. P.; Liddle, S. T. Chem. Sci. 2016, 7, 155–165 with the permission of the Royal Society of Chemistry.

In a modification of the above system the equatorial imido donors (NCN) were replaced by softer (SCS) donor atoms so that the strength of the equatorial interactions can be reduced.78 Thus, the reaction of [Dy(SCS)(SCSH)(THF)] (118, SCS ¼ {C(PPh2S)2}2−) with alkali metal alkyls and auxiliary ethers afforded the bis-methanediide complexes [Dy(SCS)2] [Dy(SCS)2(K(DME)2)2] (119), [Dy(SCS)2][Na(DME)3] (120) and [Dy(SCS)2][K(2,2,2-cryptand)] (121) (Scheme 10).

Scheme 10 Synthetic routes to the bis-methanediide complexes, Ln ¼ Dy. Adapted from Ref. Thomas-Hargreaves, L. R.; Giansiracusa, M. J.; Gregson, M.; Zanda, E.; O’Donnel, F.; Wooles, A. J.; Chilton, N. F.; Liddle, S. T. Chem. Sci. 2021, 12, 3911–3920 with the permission of the Royal Society of Chemistry.

DyIII in 118 is presented in a C2S4O, seven coordinated environment while in 120 and 121 it is six coordinated in a C2S4 environment. Complex 119 exists as two discrete DyIII based ion pairs (referred here as 119 cation and 119 anion), both having C2S4 coordination environment. Complex 119 cation and 121 show the most linear type arrangement having C]Dy]C angles of 178.62 and 176.03(11)o respectively. Complex 119 anion and 120 deviate much from linearity having C]Dy]C angles of 166.1(3) and 164.01(11)o, respectively. Analysis of the ac data for 119, 120 and 121 reveals the effective energy barriers of 742.99 (89.66)/806.24(14.6), 705.46(22.2), and 770.79(48.7) cm−1, respectively. Open magnetic hysteresis up to 14, 12, and 15 K were observed for 119, 120, and 121 respectively. From the magneto structural correlation, it was found that with increase of the C] Dy]C angle (linearity of the complexes) the effective energy barrier increases (Fig. 46).

416

Organometallic Lanthanide Complexes as Single Molecule Magnets

Fig. 46 Ueff barrier as a function of C]Dy]C angle for 119–121. Adapted from Ref. Thomas-Hargreaves, L. R.; Giansiracusa, M. J.; Gregson, M.; Zanda, E.; O’Donnel, F.; Wooles, A. J.; Chilton, N. F.; Liddle, S. T. Chem. Sci. 2021, 12, 3911–3920 with the permission of the Royal Society of Chemistry.

14.11.9 Conclusions Various ligand classes, cyclobutadienyl (Cb), cyclopentadienyl (Cp), cyclooctatetraenyl (COT) etc. showed that each of these ligands has a very distinct mode of binding and provides large anisotropy when bound to lanthanide ions. For example Cp type of ligands provide mainly axial ligand field which is best to be used with oblate type of lanthanide ions (DyIII, TbIII, etc.) and from theoretical calculations it is found that COT type of ligands generally provides more equatorial ligand field than axial field and thus they are best suited for prolate type of lanthanide ions (ErIII, TmIII, etc.). Due to their specific binding capabilities these types of ligands provide an opportunity for a strategic design based synthetic approach allowing the assembly of complexes where the axial geometry can be tuned to near-linearity. Further, in all the cases the coordination capability and the ligand field strength can be modulated by stereo-electronic variations involving various substituents. This in turn allows a modulation of the magnetic properties. Within just a decade of the report of first organometallic lanthanide complex based SMM there has been a remarkable progress in this area. One of the measures of the progress is the effective energy barrier while the other is the blocking temperature. The remarkable progress starting from the effective energy barrier of 32.32  1.67 cm−1 for the first lanthanide organometallic SMM to the record breaking effective energy barrier of 1541 cm−1 for [(Z5-Cp )Dy(Z5-CpiPr5)][B(C6F5)4] which also has the blocking temperature for magnetization reversal of 80 K that is higher than the boiling point of liquid nitrogen. This indeed provides immense hope that in the not too distant future molecular magnetic bistability will be achieved around room temperature.

Acknowledgments BD is thankful to the Science and Engineering Research Board, New Delhi, India (file number PDF/2020/002670) for the award of a National Post-doctoral Fellowship. VC is thankful to the Department of Science and Technology, New Delhi, India for the National J. C. Bose Fellowship.

References 1. 2. 3. 4. 5. 6. 7.

Caneschi, A.; Gatteschi, D.; Sessoli, R.; Barra, A. L.; Brunel, L. C.; Guillot, M. J. Am. Chem. Soc. 1991, 113, 5874–5876. Gatteschi, D.; Sessoli, R. Angew. Chem. Int. Ed. 2003, 42, 268–297. Maniaki, D.; Pilichos, E.; Perlepes, S. P. Front. Chem. 2018, 6, 461–489. Neese, F.; Pantazis, D. A. Faraday Discuss. 2011, 148, 229–238. Ruiz, E.; Cirera, J.; Cano, J.; Alvarez, S.; Loose, C.; Kortus, J. Chem. Commun. 2008, 1, 52–54. Ako, A. M.; Hewitt, I. J.; Mereacre, V.; Clérac, R.; Wernsdorfer, W.; Anson, C. E.; Powell, A. K. Angew. Chem. Int. Ed. 2006, 45, 4926–4929. Liu, J. L.; Guo, F. S.; Meng, Z. S.; Zheng, Y. Z.; Leng, J. D.; Tong, M. L.; Ungur, L.; Chibotaru, L. F.; Heroux, K. J.; Hendrickson, D. N. Chem. Sci. 2011, 2, 1268–1272.

Organometallic Lanthanide Complexes as Single Molecule Magnets

8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78.

417

Ishikawa, N.; Sugita, M.; Ishikawa, T.; Koshihara, S. Y.; Kaizu, Y. J. Am. Chem. Soc. 2003, 125, 8694–8695. Bar, A. K.; Kalita, P.; Singh, M. K.; Rajaraman, G.; Chandrasekhar, V. Coord. Chem. Rev. 2018, 367, 163–216. Liu, J. L.; Chen, Y. C.; Tong, M. L. Chem. Soc. Rev. 2018, 47, 2431–2453. Carlin, R. L. Magnetochemistry; Springer, 1986. Acharya, J.; Kalita, P.; Chandrasekhar, V. Magnetochemistry 2021, 7, 1–29. Parmar, V. S.; Mills, D. P.; Winpenny, R. E. P. Chem. A Eur. J. 2021, 27, 7625–7645. Dey, A.; Acharya, J.; Chandrasekhar, V. Chem. Asian J. 2019, 14, 4433–4453. Chakraborty, A.; Acharya, J.; Chandrasekhar, V. ACS Omega 2020, 5, 9046–9054. Dey, A.; Tripathi, S.; Shanmugam, M.; Narayanan, R. S.; Chandrasekhar, V. Cobalt(II)/(III)–Lanthanide(III) Complexes as Molecular Magnets. In Organometallic Magnets, Springer, 2019. Dey, A.; Bag, P.; Kalita, P.; Chandrasekhar, V. Coord. Chem. Rev. 2021, 432, 213707. Dey, A.; Kalita, P.; Chandrasekhar, V. ACS Omega 2018, 3, 9462–9475. Lu, J.; Guo, M.; Tang, J. Chem. Asian J. 2017, 12, 2772–2779. Rinehart, J. D.; Long, J. R. Chem. Sci. 2011, 2, 2078–2085. Ungur, L.; Chibotaru, L. F. Inorg. Chem. 2016, 55, 10043–10056. Guo, F. S.; Day, B. M.; Chen, Y. C.; Tong, M. L.; Mansikkamäki, A.; Layfield, R. A. Angew. Chem. Int. Ed. 2017, 56, 11445–11449. Guo, F. S.; Day, B. M.; Chen, Y. C.; Tong, M. L.; Mansikkamäki, A.; Layfield, R. A. Science 2018, 362, 1400–1403. Dey, B. M.; Guo, F. S.; Layfield, R. A. Acc. Chem. Res. 2018, 51, 1880–1889. Liddle, S. T.; Slageren, J. V. Chem. Soc. Rev. 2015, 44, 6655–6669. McAdams, S. G.; Ariciu, A. M.; Kostopoulos, A. K.; Walsh, J. P. S.; Tuna, F. Coord. Chem. Rev. 2017, 346, 216–239. Ungur, L.; Chibotaru, L. F. Phys. Chem. Chem. Phys. 2011, 13, 20086–20090. Balanda, M. Acta. Phys. Pol. A 2013, 124, 964–976. Shao, D.; Wang, X. Y. Chin. J. Chem. 2019, 37, 1005–1018. Skomski, R. Simple Models of Magnetism; Oxford University Press: Oxford, 2008. Chilton, N. F. Inorg. Chem. 2015, 54, 2097–2099. Day, B. M.; Guo, F. S.; Giblin, S. R.; Sekiguchi, A.; Mansikkamaki, A.; Layfield, R. A. Chem. A Eur. J. 2018, 24, 16779–16782. Chakraborty, A.; Dey, B. M.; Durrant, J. P.; He, M.; Tang, J.; Layfield, R. A. Organometallics 2020, 39, 8–12. Durrant, J. P.; Tang, J.; Mansikkamaki, A.; Layfield, R. A. Chem. Commun. 2020, 56, 4708–4711. Ojea, M. J. H.; Maddock, L. C. H.; Layfield, R. A. Lanthanide Organometallics as Single-Molecule Magnets. Topics in Organometallic Chemistry Springer, 2019. Layfield, R. A.; McDouall, J. J. W.; Sulway, S. A.; Tuna, F.; Collison, D.; Winpenny, R. E. P. Chem. A Eur. J. 2010, 16, 4442–4446. Sulway, S. A.; Layfield, R. A.; Tuna, F.; Wernsdorfer, W.; Winpenny, R. E. P. Chem. Commun. 2012, 48, 1508–1510. Tuna, F.; Smith, C. A.; Bodensteiner, M.; Ungur, L.; Chibotaru, L. F.; McInnes, E. J. L.; Winpenny, R. E. P.; Collison, D.; Layfield, R. A. Angew. Chem. Int. Ed. 2012, 51, 1–6. Demir, S.; Zadrozny, J. M.; Long, J. R. Chem. A Eur. J. 2014, 20, 9524–9529. Errulat, D.; Gabidullin, B.; Mansikkamaki, A.; Murugesu, M. Chem. Commun. 2020, 56, 5937–5940. Meng, Y. S.; Zhang, Y. Q.; Wang, Z. M.; Wang, B. W.; Gao, S. Chem. A Eur. J. 2016, 22, 12724–12731. Pugh, T.; Chilton, N. F.; Layfield, R. A. Angew. Chem. Int. Ed. 2016, 128, 1–5. Collins, R.; Ojea, M. J. H.; Maniskkamaki, A.; Tang, J.; Layfield, R. A. Inorg. Chem. 2020, 59, 642–647. Pugh, T.; Tuna, F.; Ungur, L.; Collison, D.; McInnes, E. J. L.; Chibotaru, L. F.; Layfield, R. A. Nat. Commun. 2015, 6, 7492–7499. Pugh, T.; Vieru, V.; Chibotaru, L. F.; Layfield, R. A. Chem. Sci. 2016, 7, 2128–2137. Pugh, T.; Chilton, N. F.; Layfield, R. A. Chem. Sci. 2017, 8, 2073–2080. Burns, C. P.; Wilkins, B. O.; Dickie, C. M.; Latendresse, T. P.; Vernier, L.; Vignesh, K. R.; Bhuvanesh, N. S.; Nippe, M. Chem. Commun. 2017, 53, 8419–8422. Burns, C. P.; Yang, X.; Wofford, J. D.; Bhuvanesh, N. S.; Hall, M. B.; Nippe, M. Angew. Chem. Int. Ed. 2018, 57, 8144–8148. Trifonov, A. A.; Shestakov, B.; Long, J.; Lyssenko, K.; Guari, Y.; Larionova, J. Inorg. Chem. 2015, 16, 7667–7669. Goodwin, C. A. P.; Ortu, F.; Reta, D.; Chilton, N. F.; Mills, D. P. Nature 2017, 548, 439–442. McClain, K. R.; Gould, C. A.; Chakarawet, K.; Teat, S. J.; Groshens, T. J.; Long, J. R.; Harvey, B. G. Chem. Sci. 2018, 9, 8492–8503. He, M.; Guo, F. S.; Tang, J.; Mansikkamaki, A.; Layfield, R. A. Chem. Commun. 2021, 57, 6396–6399. Rinehart, J. D.; Fang, M.; Evans, W. J.; Long, J. R. Nat. Chem. 2011, 3, 538–542. Rinehart, J. D.; Fang, M.; Evans, W. J.; Long, J. R. J. Am. Chem. Soc. 2011, 133, 14236–14239. Demir, S.; Gonzalez, M. I.; Darago, L. E.; Evans, W. J.; Long, J. R. Nat. Commun. 2017, 8, 2144–2151. Demir, S.; Zadrozny, J. M.; Nippe, M.; Long, J. R. J. Am. Chem. Soc. 2012, 134, 18546–18549. Demir, S.; Nippe, M.; Gonzalez, M. I.; Long, J. R. Chem. Sci. 2014, 5, 4701–4711. Guo, F. S.; Layfield, R. A. Chem. Commun. 2017, 53, 3130–3133. Gould, C. A.; Darago, L. E.; Gonzalez, M. I.; Demir, S.; Long, J. R. Angew. Chem. Int. Ed. 2016, 129, 10237–10241. Jin, P. B.; Zhai, Y. Q.; Yu, K. X.; Winpenny, R. E. P.; Zheng, Y. Z. Angew. Chem. Int. Ed. 2020, 59, 9350–9354. Liu, S. S.; Ziller, J. W.; Zhang, Y. Q.; Wang, B. W.; Evans, W. J.; Gao, S. Chem. Commun. 2014, 50, 11418–11420. Liu, S. S.; Yan, B.; Meng, Z. S.; Gao, C.; Wang, B. W.; Gao, S. Inorg. Chem. Commun. 2017, 56, 312–314. Meng, Y. S.; Xu, L.; Xiong, J.; Yuan, Q.; Liu, T.; Wang, B. W.; Gao, S. Angew. Chem. Int. Ed. 2018, 57, 4673–4676. Harriman, K. L. M.; Le Roy, J. J.; Ungur, L.; Holmberg, R. J.; Korobkov, I.; Murugesu, M. Chem. Sci. 2016, 8, 231–240. Jiang, S. D.; Wang, B. W.; Sun, H. L.; Wang, Z. M.; Gao, S. J. Am. Chem. Soc. 2011, 133, 4730–4733. Jiang, S. D.; Liu, S. S.; Zhou, L. N.; Wang, B. W.; Wang, Z. M.; Gao, S. Inorg. Chem. 2012, 51, 3079–3087. Meihaus, K. R.; Long, J. R. J. Am. Chem. Soc. 2013, 135, 17952–17957. Ungur, L.; Roy, J. J. L.; Korobkov, I.; Murugeshu, M.; Chibotaru, L. F. Angew. Chem. Int. Ed. 2014, 53, 1–6. Jelectic, M.; Lin, P. H.; Roy, J. J. L.; Korobkov, I.; Gorelsky, S. I.; Murugeshu, M. J. Am. Chem. Soc. 2011, 133, 19286–19289. Roy, J. J. L.; Korobkov, I.; Murugeshu, M. Chem. Commun. 2014, 50, 1602–1604. Roy, J. J. L.; Jelectic, M. S.; Gorelsky, S. I.; Korobkov, I.; Ungur, L.; Chibotaru, L. F.; Murugeshu, M. J. Am. Chem. Soc. 2013, 135, 3502–3510. Roy, J. J. L.; Ungur, L.; Korobkov, I.; Chibotaru, L. F.; Murugeshu, M. J. Am. Chem. Soc. 2014, 136, 8003–8010. Hilgar, J. D.; Flores, B. S.; Rinehart, J. D. Chem. Commun. 2017, 53, 7322–7324. Hilgar, J. D.; Bernbeck, M. G.; Flores, B. S.; Rinehart, J. D. Chem. Sci. 2018, 9, 7204–7209. Meng, Y. S.; Wang, C. H.; Zhang, Y. Q.; Leng, X. B.; Wang, B. W.; Chen, Y. F.; Gao, S. Inorg. Chem. Front. 2016, 3, 828–835. Chen, S. M.; Xiong, J.; Zhang, Y. Q.; Yuan, Q.; Wang, B. W.; Gao, S. Chem. Sci. 2018, 9, 7540–7545. Gregson, M.; Chilton, N. F.; Ariciu, A. M.; Tuna, F.; Crowe, I. F.; Lewis, W.; Blake, A. J.; Collison, D.; McInnes, E. J. L.; Winpenny, R. E. P.; Liddle, S. T. Chem. Sci. 2016, 7, 155–165. Thomas-Hargreaves, L. R.; Giansiracusa, M. J.; Gregson, M.; Zanda, E.; O’Donnel, F.; Wooles, A. J.; Chilton, N. F.; Liddle, S. T. Chem. Sci. 2021, 12, 3911–3920.

14.12

Organometallic Receptors for Charged and Neutral Guest Species

Robert Hein and Paul D Beer, Department of Chemistry, University of Oxford, Oxford, United Kingdom © 2022 Elsevier Ltd. All rights reserved.

14.12.1 14.12.2 14.12.2.1 14.12.2.2 14.12.2.3 14.12.2.4 14.12.2.5 14.12.3 14.12.3.1 14.12.3.2 14.12.3.3 14.12.3.4 14.12.4 14.12.4.1 14.12.4.2 14.12.5 14.12.5.1 14.12.5.2 14.12.5.3 14.12.5.4 14.12.6 References

Introduction Organometallic cation receptors Cation receptors based on ferrocene Cation receptors based on cyclometalated iridium Cation receptors based on alkynyl gold motifs Cation receptors based on alkynyl platinum motifs Cation receptors based on other organometallic motifs Organometallic anion receptors Anion receptors based on ferrocene Anion receptors based on metal carbonyl complexes Anion receptors based on cyclometalated iridium Anion receptors based on other motifs Organometallic ion-pair receptors Ion-pair receptors based on ferrocene Other organometallic ion-pair receptors Organometallic receptors for neutral guests Receptors based on ferrocene Receptors based on half-sandwich complexes Receptors based on NHCs Receptors based on alkynyl platinum and gold motifs Conclusions and outlook

418 418 419 424 428 429 430 432 432 439 441 442 443 444 446 447 447 449 452 456 458 459

14.12.1 Introduction Molecular recognition of small charged and neutral guest species at artificial, supramolecular receptors has blossomed in recent decades. Supramolecular host-guest chemistry not only supplements fundamental investigations of intermolecular interactions but has also found widespread use in extraction, membrane transport, catalysis, materials design and, perhaps most importantly, sensing. As a consequence of their inherent redox- or photo-activity, the integration of reporter organometallic motifs into supramolecular host structural frameworks has received significant attention for sensing applications. Serving also as topological scaffolds and/or as binding sites for guest species, organometallic based host systems have proved to be potent receptors and sensors for ions and small molecules. Herein, we provide an overview of recent advances in this field, with a focus on work from the year 2007 onwards. For a comprehensive account of prior organometallic receptors the reader is referred to an earlier review.1 We first discuss cation and anion organometallic receptors separately (Sections 14.12.2 and 14.12.3), before highlighting systems in which charged species are bound as an ion-pair (Section 14.12.4). This is followed by a review of organometallic receptors for neutral guests (Section 14.12.5) and finally a conclusions and outlook discussion section (Section 14.12.6). All sections are subdivided according to the nature of the organometallic moiety. In particular, the important role the organometallic moiety plays in dictating the binding/sensing properties of the hosts is highlighted throughout. We restrict our discussion to organometallic receptors interacting with guests in a reversible, non-covalent manner. As such, chemodosimeters are not included in the review.

14.12.2 Organometallic cation receptors Metal cations are ubiquitous constituents of Nature and play vital roles in a plethora of biological, environmental and industrial processes. For example, the alkali cations Na+ and K+ are pivotal for many biological processes including neurotransmission, metabolism and the electrolyte balance, while transition metals fulfill important functions as the active sites of many enzymes. The latter are also increasingly important in a wide range of technological applications necessitating their extraction/recovery/remediation from ores or (waste) water streams. Similarly, the detection and removal of toxic metal cations, such as Hg2+, Cd2+ or Pb2+, remains highly important in a variety of environmental settings. As a result of these diverse functions the development of synthetic cation receptors capable of extracting or sensing these species has matured enormously over the last decades. Continuing explorations are mainly focused on the development of more selective and sensitive cation receptors and sensors and their application in real-life relevant settings and applications. Another recent focus

418

Comprehensive Organometallic Chemistry IV

https://doi.org/10.1016/B978-0-12-820206-7.00132-3

Organometallic Receptors for Charged and Neutral Guest Species

419

in the field of (cat)ion sensors is the development of dual channel sensors in which multiple output signals (e.g., electrochemical as well as optical) can be used to transduce an ion binding event.

14.12.2.1 Cation receptors based on ferrocene Ferrocene (Fc), the archetypical metallocene, has, since its discovery in the early 1950s, been incorporated in numerous receptors and sensors.2,3 This can be attributed to its straight-forward synthetic chemistry combined with favorable redox properties, characterized by a reversible, one-electron redox couple (ferrocene/ferrocenium, Fc/Fc+) at moderate potentials. This presents a facile means of generating redox-active receptors capable of electrochemically sensing guest molecules, first demonstrated in the 80s and 90s.4,5 Since then, Fc is undoubtedly the most commonly employed redox transducer in ion sensors and is continually explored in increasingly advanced systems. Mirroring early developments in the field,1 continuing attention has been paid to construct ferrocene crown-ether based receptors. For example in 2015, Bruña et al. reported a novel oxathiacrown macrocycle 1 appended with two Fc groups.6 As a result of the close proximity between the Fc moieties there exists significant electronic coupling such that two redox wave peaks are resolved voltammetrically, which represents the separate, step-wise oxidation of both Fcs. Electrochemical metal cation sensing studies were carried out with Li+, K+, Zn2+, Cd2+ and Hg2+ in acetonitrile (ACN), where notably only the latter induced a voltammetric response, which was characterized by a significant anodic perturbation and the merging of both Fc oxidation peaks into a single, but broad wave.

As a result of the significant toxicity of this transition metal cation, a large variety of other Hg2+ sensors have been developed, many also based on sulfur containing recognition motifs.7–11 For example, Zhang’s group developed thioether-appended ferrocenes 2 and 3 as highly selective optical and redox sensors for Hg2+ in ACN.7 Using UV-vis spectroscopy, both probes displayed limits of detection (LODs) of  0.5 mM and showed relatively modest anodic shifts of the Fc/Fc+ couple ( 40 mV) with an additional, new, irreversible peak at higher potentials. For the monosubstituted receptor 2 a 2:1 host-guest stoichiometric binding mode was suggested. The authors further demonstrated that these receptors can be easily chemically oxidized to the sulfones 4 and 5, a transformation that is associated with significant changes in the sensor’s selectivity. Specifically, these receptors are now selective towards Cu2+ which can be optically sensed with similar LODs and high selectivity over Hg2+. Electrochemically, no significant shift of the respective Fc/Fc+ redox couple was observed upon exposure to Cu2+, but a pronounced increase in the current density, was noted. These observations may arise due to oxidation of Fc by the cupric ion.

Long and co-workers reported the simple quinoline-appended ferrocene 6 as a selective Hg2+ sensor.12 Indeed, the quinoline fluorescence was significantly quenched in the presence of Hg2+ in ACN, albeit with smaller, but significant responses also towards Pb2+ and Cu2+. In the same solvent a peculiar voltammetric behavior towards Hg2+ was also reported: a −149 mV cathodic shift. The sensor was also immobilized onto screen-printed carbon electrodes by simple physisorption and used for electrochemical sensing of Hg2+ in water. In this case no shifts were observed but exposure to Hg2+ induced significant decreases in the peak currents. Similar Fc-quinolines, 7 and 8 were also studied by Ghosh and co-workers in 2015.13 These receptors were capable of Zn2+ recognition in

420

Organometallic Receptors for Charged and Neutral Guest Species

the predominantly aqueous solvent medium ACN/H2O 1:9, under formation of 1:1 host-guest stoichiometric complexes with K ¼ 25,500 and 28,900 M−1 for 7 and 8, respectively. Sensing was achieved by both a significant fluorescence turn-on as well as anodic voltammetric shifts of 70 and 90 mV, respectively. Of note are the recently reported triazole-linked Fc-quinoline derivatives 9 and 10 displaying a different sensory behavior, with a preference for Cu2+.14 Optically, both responded to Cu2+ in ACN/H2O 4:1 with K ¼ 680,000 and 59,700 M−1, respectively. Both receptors also displayed fluorescence quenching, however for 10 the emergence of a new emission peak at higher wavelength was also observed. Electrochemically, the response of 9 and 10 also differed slightly in that the former responded only to Cu2+ via decreased current densities while significant perturbations of the redox properties of 10 were also observed for Co2+, Ni2+, Hg2+ and Fe2+. As alluded to in earlier examples, it is likely that the redox properties of Cu2+ play a significant role in these sensing mechanisms. The authors also reported that analogous receptors containing an additional methylene spacer between the Fc and triazole also give similar sensory responses but display significantly lower selectivity. These observations illustrate the significant influence of even subtle changes in the design of the recognition sites on the binding and sensing performance of these receptors.

The 1,2,3-triazole building block has emerged as a versatile scaffold in supramolecular chemistry and is not only easily accessible via “click” chemistry, but can also bind to both cations as well as anions, and is now routinely employed in a large variety of (organometallic) cation, anion or ion-pair receptors (for examples of the latter two see Sections 14.12.3 and 14.12.4).15,16 In addition to the above-shown receptors 9 and 10, a large range of Fc-containing triazole-based cation receptors have been reported.17–23 For example, in 2011 Ghosh and co-workers investigated receptors 11–12, and demonstrated their utility as Hg2+ sensors.17 Both receptors displayed large anodic perturbations of 160 and 217 mV, for 11 and 12, respectively, towards this metal cation in ACN with good selectivity. In the same solvent a significant optical, naked-eye response was also observed for 12. Surprisingly, this sensor was also capable of sensitive fluorescence Hg2+ sensing in water with a LOD of 7.5 ppb and only small interference from Pb2+, Cd2+, Zn2+ and Mg2+. The same group also investigated the structurally related 13–14, and similarly demonstrated their potential as Hg2+ as well as 2+ Ni sensors.18 Interestingly, the mono-substituted receptor 13 displayed larger anodic voltammetric perturbations towards Hg2+ of 167 mV over the 1,1-di-substituted ferrocene receptor 14 (106 mV) in ACN. This trend was reversed towards Ni2+, with shifts of 136 and 187 mV, for 13 and 14, respectively. In this solvent a pronounced color change towards both ions was also observed for both probes. Notably, the electrochemical as well as optical responses were largely retained in the competitive aqueous solvent mixture ACN/H2O 1:4.

Organometallic Receptors for Charged and Neutral Guest Species

421

The differentiation of Fe2+ and Fe3+ by hydroxymethyl naphthyl appended Fc-triazole receptors 15 and 16 was reported by Thakur and coworkers.19 In ACN/H2O 1:4 both receptors responded selectively towards both iron cations, with small anodic shifts of 20 and 11 mV towards Fe3+, respectively. The dicationic ferrous cation induced significantly larger anodic perturbations of 90 and 59 mV, respectively. These observations arise from different sensing mechanisms, whereby the response towards Fe3+ arises from oxidation of the Fc+ moiety, while Fe2+ is bound via interactions with both the triazole and hydroxyl groups. Further studies also demonstrated naked eye and luminescent discrimination of both cations.

The combination of both redox active tetrathiafulvalene (TTF) and Fc-triazole scaffolds was investigated by Zhao et al. in receptors 17 and 18.20 In spite of only minor differences in their structures, both exhibited distinct redox and sensing properties. In DCM/ACN 1:1 17 displayed three reversible redox waves arising from two, one and one electron oxidation processes, indicative of an initial simultaneous oxidation of both Fc groups followed by subsequent two-step oxidation of TTF. In contrast, 18 displayed only two redox waves, whereby the first one encompasses both the first TTF oxidation as well as the oxidation of both Fc moieties. This in good agreement with diminished through space electronic communication between the Fc and TTF groups as a result of their larger separation. Furthermore, 17 did not display a significant redox response towards Ag+, Cd2+, Cu2+, Hg2+, Pb2+ or Zn2+, while 18 responded to the latter with unexpected cathodic shifts of the first and second waves of 53 and 59 mV, respectively. This was ascribed to conformational changes induced by metal cation binding thereby again altering the through space interaction between the different redox transducers. This highlights the potential of combining multiple redox transducers in attaining unique, highly tunable sensors.

Another example of a sensor integrating three redox groups is the tris-ferrocene receptor 19.21 In analogy to receptor 18 two distinct redox waves are observed in ACN, arising from the chemically inequivalent ferrocenes. Both waves displayed selective anodic perturbations of 145 and 92 mV towards Hg2+. The sensor also responded optically towards this cation (color change form yellow to purple) and furthermore displayed large and selective cathodic perturbations towards H2PO−4. This highlights the utility of the 1,2,3-triazole as a CdH hydrogen bond donor for anion as well as ion-pair recognition, as discussed in more detail in Sections 14.12.3 and 14.12.4. A structurally very similar receptor with different connectivity was similarly shown to be a good Hg2+ sensor.23

422

Organometallic Receptors for Charged and Neutral Guest Species

Gasser et al. demonstrated the utility of the related 1,2,4-triazole scaffold in receptors Fc-carbimine pyridyl-triazole 20 and Fc-carbamide pyridyl-triazole 21.24 Sensing studies were performed with Cu+, Ag+, Zn2+ and Cd2+ in DCM/ACN 1:1 both optically and electrochemically. In the latter format, moderate anodic shifts (32–57 mV) were observed in the presence of all these metal cations. Notably, the response towards the divalent cations was identical for both receptors and similar for Ag+. A significant difference in performance between the probes was only observed towards Cu+ which elicited a significantly larger shift (57 mV) for 21 over 20 (41 mV). Based on analysis of the binding stoichiometries the authors suggest formation of linear or grid-like 1:1 complexes.

The structurally similar, but S-BINOL appended receptor 22 was developed by Zhao and coworkers as a highly selective Al3+ probe.25 In DMSO/MeOH 99:1 a >200-fold fluorescence turn on was observed in the presence of this analyte, with good selectivity over many other di and trivalent cations. Similarly, small anodic shifts ( 30 mV) were observed in ACN/H2O/MeOH 7:3:1 only towards Al3+. Based on fluorescence and 1H NMR studies the authors suggested an interesting 1:2 host-guest binding stoichiometry involving both hydroxy, as well as the carbonyl and imine functional groups. Building on earlier work,1 the group of Tarraga and Molina have reported a variety of redox-active metal cation sensors based on nitrogenous recognition motifs.10,26,27 This includes imidazoquinoxaline receptors 23–24 as selective redox, chromogenic and fluorescent Pb2+ sensors.10 In ACN, 23 responded to this analyte with an anodic shift of 110 mV, the appearance of a new lower energy absorption band in UV-vis as well as a significant fluorescence increase (276-fold). The analogous pyridine containing 24 displayed even more impressive anodic shifts of 300 mV with similar optical responses. Of note is the role of the imidazole moiety, which enables voltammetric sensing of F− (via deprotonation; cathodic shifts of −300 and −270 mV, for 23 and 24, respectively) as well as HP2O3− 7 (via recognition with shifts of −80 and −90 mV).

Other Pb2+ sensors 25 and 26, based on pentakis(phenylthio)benzene imine motifs were also reported by the same group.26 The former displayed a selective optical and electrochemical sensor for Pb2+ with significant changes in both absorbance as well as an anodic shift of the Fc couple by 125 mV in ACN. The isomeric receptor 26 showed significant optical perturbations in the presence of both Pb2+ and Hg2+ as well as significantly larger cathodic perturbations of 340 mV towards both ions. Similar Fc-imine cyclophanes 27 and 28 were also developed as highly selective receptors for Zn2+ and Li+, respectively.27 The former responds selectively to Zn2+ in DCM/ACN 4:1 via 80 mV anodic shifts of its more cathodic Fc redox couple, while the more anodic redox couple of the other Fc remained unperturbed. This is indicative of complete and fast decomplexation of the Zn2+ cation upon mono-oxidation of the receptor. Neither Li+, Na+, K+, Mg2+, Ca2+, Cd2+ nor Ni2+ induced any voltammetric responses, while Cu2+ and Hg2+ cations induced receptor oxidation. As a result, these two cations, as well as Zn2+ also induced optical responses. A more detailed analysis of the UV-vis titration data suggested 2:1 host-guest stoichiometric binding of Zn2+ to the host with K ¼ 790,000 M−2 and an LOD of 5.7 mM. The iminophosphorane analog 28 displayed a notable preference for Li+ with a 1:1 binding stoichiometry as revealed by optical studies in THF (K  106 M−1). Electrochemically, the authors reported a similar behavior as for receptor 27, with significant perturbations of only the first Fc redox couple (in DCM). However, the reported anodic shift of 830 mV upon Li+ addition appears unreasonably large.

Organometallic Receptors for Charged and Neutral Guest Species

423

Mixed metallocene analogs of 27 have also been reported where either of the Fc groups is replaced with ruthenocene.28 These sensors also display a selective optical naked-eye response to Zn2+ in ACN/DCM 3:2. In addition to the ubiquitous crown ethers, there are a large variety of other macrocyclic host systems that have received enormous attention as synthetic receptors, including cyclodextrins, cucurbiturils, calixarenes, cyclophanes and many others. A variety of these have been further functionalized with organometallic groups as optical or electrochemical transducers. For example, Ruiz-Botella et al. reported a tetra-ferrocene benzimidazole resorcin[4]arene cavitand 29 as a redox-active quaternary ammonium cation binding and sensing receptor.29 1H NMR titration investigations in d6-benzene or d8-toluene revealed moderate binding of tetramethyl, tetraethyl and dodecyltrimethylammonium ions as well as the biologically relevant choline chloride (K ¼ 30–500 M−1), with negligible binding of tetrabutylammonium. In DCM, continuous anodic perturbation of the ferrocene redox signature of up to 182 mV for dodecyltrimethylammonium and 173 mV for choline chloride were observed. These large anodic perturbations are consistent with a complete release of the guest from the host cavity upon oxidation, highlighting that such systems can not only serve as sensors but potentially also as stimuli responsive assemblies or molecular machinery.

The polycyclic aromatic sumanene, a fullerene fragment, has gained attention as a concave binding motif for the recognition of Cs+ via cation–p interactions. In 2019, Kasprzak et al. investigated a range of Fc-functionalized sumanene receptors and demonstrated their potential for selective fluorescent Cs+ sensing.30 Preliminary results also indicated a redox-response towards this metal cation in 1:1 DCM/MeOH. Building on these results the same group recently reported the tris-Fc-appended sumanene 30 and demonstrated its potential for selective electrochemical Cs+ sensing in water.31 This was achieved by immobilization of the probe onto a glassy carbon electrode with the help of a Nafion membrane which enabled sensing in pure water. Upon exposure to Cs+ an unexpected, moderate cathodic perturbation of  35 mV and a concomitant increase in current was observed, which was proposed

424

Organometallic Receptors for Charged and Neutral Guest Species

to arise from a 2:1 host-guest stoichiometric binding mode. With a low LOD of 0.02 mM, no significant interference from Na+, K+ and Ba2+ and the potential for sensor re-use by simple washing, this work nicely illustrates some of the main advantages of surface-immobilization of receptors: more facile sensor reuse, device integration and surface effects,32 as discussed in more detail in Section 14.12.3.1.

For further examples of Fc-containing cation receptors within the last decade, (and due to space limitations) the reader is referred to a recent comprehensive review by Pal et al.33

14.12.2.2 Cation receptors based on cyclometalated iridium Over the last decades cyclometalated iridium(III) complexes have found wide-spread use in numerous applications, including light-emitting diodes, catalysis, imaging and sensing.34–38 This can be attributed to their high stability, tunability, accessible synthesis, high quantum yields and long-lived (luminescence/phosphorescence) lifetimes, which is particularly useful in biological (imaging) applications.39 The specific photophysical processes and sensing mechanisms that underpin the emission properties and changes therein upon analyte binding will not be discussed explicitly, but are well-documented in literature.40 As a result of their modular synthesis, a large variety of cyclometallated Ir(III) complexes have been developed, whereby most commonly 2-phenylpyridine (ppy) is employed as part of the ligand ensemble. A myriad of receptive functional groups have been appended directly to these cyclometalated ligands or other (typically nitrogenous) co-ligands such as 2,20 -bipyridyl (bipy) or phenanthroline (phen). For example, Schmittel developed receptor 31, comprising of a Ir(ppy)2 complex containing a bis(dithiaaza-15-crown-5)phen co-ligand as a luminescent probe for Ag+.41 In ACN/H2O 1:1, Ag+ binding at the dithia-aza crown ether recognition site resulted in a significant emission enhancement. By contrast, Hg2+ addition caused a quenching of the luminescence. The aza-18-crown-6 analog of this receptor was capable of sensing Ba2+ in ACN.42 A similar system containing aza-15-crown-5 appended ppy ligands 32, reported by Nabeshima and co-workers in 2010, displayed a very strong luminescence enhancement upon Mg2+ binding in ACN, with excellent selectivity over Na+, K+ and Ca2+.43 Interestingly, the analogous complex in which the crown-ether binding units are appended at the 4-position of the phenyl ligand (i.e., para with respect to the pyridine) does not bind, nor respond to, any of these cations.

Organometallic Receptors for Charged and Neutral Guest Species

425

The detection of Hg2+ in ACN/H2O 1:1 was achieved by Li et al. using receptor 33, containing a 4-(aza-18-crown-6)picolinate ligand.43 Luminescence quenching was observed towards the mercuric ion even in the presence of a large range of other metal cations, with the exception of Cu2+. In a recent study, Miller and co-workers conducted a detailed investigation of the tunability of organometallic aza-crown ether complexes by altering the charge, ligand environment or nature of the central metal center.44 In ACN, Ir(III)-based complex 34 displayed significantly diminished Li+ affinity (K ¼ 560 M−1) in comparison to its free ligand (130,000 M−1), undoubtedly as a result of N-lone pair coordination to the iridium metal center. However, metal coordination dramatically influences the Li+/Na+ selectivity, which is with KLi/KNa ¼ 29, significantly higher than that of the free ligand (KLi/KNa ¼ 1.6). A further modulation of the metal cation binding properties was achieved by ligand substitutions. Reductive elimination of HCl with the help of base afforded the analogous, neutral, Ir(I) complex, which displayed markedly reduced affinities and Li+ selectivity (KLi ¼ 180 M−1, KLi/KNa ¼ 7). Similarly, substitution of the Cl− ligand with an ACN ligand also diminished binding strength (KLi ¼ 45 M−1). Interestingly, even ligand isomerization had a significant impact on Li+ binding. Specifically, the analogous complex with a trans arrangement between the hydride and carbonyl ligands exhibited a slightly higher KLi ¼ 790 M−1, which was attributed to favorable interactions between the Cl− lone pairs and the bound lithium cation, an interaction that is favored when the Cl− ligand lies in the central plane of the complex (cis to hydride). The authors further demonstrated that substitution of the Ir(I) metal center (KLi ¼ 180 M−1) with isoelectronic d8 organometallic fragments containing Pd, Pt or Ni chloride complexes also predictably affected the strength of Li+ binding, which was strongest in the Pd complex (KLi ¼ 290 M−1) and weakest for the Ni complex (KLi ¼ 110 M−1).44,45 Very recently, Miller reported on the utility of the related complex 35 as a cation-switchable catalyst for olefin isomerization.46 Isomerization of 1-butene derivatives by the free receptor afforded 2-alkenes under kinetic control. In contrast, the Na+-complexed receptor gave the doubly isomerized thermodynamic 3-butene product. This difference was proposed to arise from a change in the pincer ligand binding mode when the Na+ cation binds to the aza-crown ether motif.

Numerous sulfur-based cyclometalated Ir(III) complexes have been investigated as luminescent sensors for Hg2+.47–50 For example, Huang’s group reported 36 as a highly selective luminescent and electrochemical sensor for Hg2+ in ACN, responding to binding of this metal cation via blue shifts of emission bands, the anodic perturbation of the Ir(III)/Ir(IV) couple by 390 mV as well as color changes.48 The same authors later reported the related receptors 3749 and 38.50 Both acted as highly selective probes for Hg2+ in DCM, displaying naked eye color changes as well as ratiometric emission modulations for the former and emission quenching for the latter. Surprisingly, the authors report formation of 2:1 stoichiometric host-guest complexes for 37, while 38 binds two Hg2+ cations (1:2 stoichiometry). These unexpected observations, as well as the significant difference in emission changes, may arise from removal of the ancillary acac ligand from complex 37 in the presence of Hg2+. In 2017, Song and co-workers developed an Ir(III) complex containing a cysteine ligand for Hg2+ sensing in water.51

426

Organometallic Receptors for Charged and Neutral Guest Species

A sulfur-free Hg2+ selective probe, 39, was described by Cao et al. in 2015.52 In the presence of the mercuric cation significant changes in the UV–vis absorption as well as emission quenching were observed in ACN. Additionally, the Ir(III)/Ir(IV) redox couple displayed reduced current densities. The heavy metal cation Pb2+ was selectively detected by pyridyl pyrazolate receptor 40 in ACN by quenching of the luminescence emission intensity.53 No interference was observed from various metal cations, including Na+, Ba2+, Ca2+, Mg2+, Zn2+, Hg2+ or Al3+. Similarly, selective 1:1 host-guest stoichiometric binding of Ag+ by receptor 41 was reported, resulting in absorbance shifts as well as emission quenching in DMF/H2O 10:1.54

By far the most common ligating motif to be appended to cyclometalated Ir(III) complexes is the dipycolylamine (DPA) unit for binding and sensing Cu2+ and Zn2+.55–57 Over the last decade significant efforts have been directed at employing these probes in biologically relevant, buffered aqueous media. For example, You, Nam, Lippard and coworkers employed receptor 42 for the luminescence sensing of Cu2+ in aqueous buffer (PIPES, pH 7.0)58 and was also used for ratiometric Cu2+ sensing in HeLa cells. The same authors developed 43 as the first phosphorescent probe for Zn2+ in biological samples.59 In analogy to 42, this receptor also displayed a dual modal emission (blue and yellow), wherein the lower energy emission band sensitively responded to Zn2+ binding by strong emission turn-on. While some interference from Fe2+, Co2+, Ni2+ and Co2+ was noted (inducing emission quenching), the probe was nevertheless successfully applied for imaging of free zinc ions in live A549 cells by confocal laser scanning microscopy. Further systematic studies varying the nature of the ligand and substitution patterns in related receptors were reported by Nam and You, highlighting the tunability of these zinc probes.60,61 Another Zn2+ ion sensor, 44, was reported by Ma et al. in 201462 which underwent a selective luminescence color change from blue to green in buffered neutral water, not observed most notably with Cu2+, Fe3+, Ni2+ and Co2+. The authors also demonstrated 44 was capable of Zn2+ visualization in live zebrafish.

Further examples of Ir(III)-DPA based probes for cation imaging in biological samples include the recently developed receptors 45 and 46.63 Both displayed a highly selective emission quenching response towards Cu2+ with low LODs ( 20 nM, in 1:1 MeOH/ H2O) and were further employed for Cu2+ visualization in HeLa cells. While 45 displayed significant cytotoxicity even at low concentrations, 46 remained largely nontoxic, even at concentrations as high as 100 mM.

Organometallic Receptors for Charged and Neutral Guest Species

427

Other recent examples of DPA-containing Cu2+ sensors include probes 47–48, developed by Sun and co-workers.62 Both sensors responded to this analyte in ACN/H2O 1:99 by emission quenching, with a larger response observed for receptor 48, appended with two DPA binding units (forming 1:2 host-guest stoichiometric complexes). While Zn2+ and Cd2+ also induced luminescence changes (red shifted emission), they did not interfere with Cu2+ determination.

Even the structurally simple receptor 49 was shown to be a potent Cu2+ sensor, displaying, in MeOH/H2O 1:99, significant emission quenching only in the presence of this metal cation, with minimal interference from various other ions.64 Interestingly, Zn2+ induced a moderate increase in emission intensity.

428

Organometallic Receptors for Charged and Neutral Guest Species

The detection of Cu2+ ions using Ir(III) receptors was also achieved by functionalization with other receptive motifs, as reported by Ma et al. in 2014.65 The salicylidene imino appended 50 displayed a naked eye color change (from colorless to yellow) as well as emission quenching upon Cu2+ binding, with excellent selectivity over many other metal cations, most notably Zn2+. The same group also developed related rhodamine appended complexes for naked eye detection of Cu2+ in tap or natural river water samples66 as well as the structurally similar 51 as a selective turn-on emission sensor for Al3+ in ACN.67

14.12.2.3 Cation receptors based on alkynyl gold motifs Gold(I) complexes, have as a result of their rich luminescent and spectroscopic properties, been exploited as potential reporter motifs in sensor design.68 Of particular note is monitoring aurophilic Au-Au interactions to effect the luminescent properties of Au(I) based receptors. For example, Yam and co-workers developed the diethylene glycol-appended tripodal alkynylgold(I) receptor 52 as a selective optical probe for Mg2+.69 Upon binding of this metal cation in DCM/MeOH 1:1, intramolecular Au-Au interactions are switched on such that significant changes in absorbance, as well as the emergence of a new lower energy emission band are observed. This sensing principle was also exploited by the same group in crown-ether modified calixarenes 53 and 54, which, in DCM/ACN 1:1, displayed fluorescence emission enhancements upon formation of sandwich complexes with K+ and Cs+, respectively.70 In a further example, benzo-15-crown-5 was incorporated into polynuclear gold(I)–copper(I) mixed-metal complexes and employed for the luminescent sensing of alkali metal cations.71

Yam’s group also demonstrated the sensing capabilities of the DPA-containing mononuclear alkynyl gold complex 55.72 In DCM/ MeOH 1:1 significant changes in absorbance, as well as fluorescence quenching were observed towards Cu2+ over a wide variety of other mono- and divalent cations. Zn2+ was also reported to bind to the receptor, but, interestingly, elicited no spectral changes.

Organometallic Receptors for Charged and Neutral Guest Species

429

In a recent example, Peris and co-workers developed Au(I)-containing metallo-tweezer 56 as a receptor for Ag+, Tl+ and Cu+ in DCM.73 Upon addition of these metal cations, significant fluorescence perturbations were resolved, which reveal a 2:1 host-guest stoichiometric binding mode with large binding constants of 2.7  109, 4.2  108, and 7.9  105 M−2 for Ag+, Tl+ and Cu+, respectively. Further investigations, including single-crystal X-ray diffraction studies, revealed the formation of an orthogonally assembled dimer which was formed as a result of p-p stacking as well as aurophilic interactions between the monovalent metal cations and the gold centers.

14.12.2.4 Cation receptors based on alkynyl platinum motifs In analogy to the above-described alkynyl gold motifs, corresponding alkynyl platinum-based hosts have gained attention as stable, photo-active sensors. The ubiquitous DPA binding motif has also been incorporated in alkynyl Pt(II) complexes.74,75 For example, in 2013, Zhang et al. developed receptor 57 as a luminescent Zn2+ sensor.74 In ACN/H2O 7:3 a significant emission turn on, with good selectivity over various other metal cations was reported.

Cyclometalated alkynyl Pt(II) Zn2+ sensors, 5876 and 5977 have also been developed by Qiu’s group. The former, carbazole-containing receptor, displayed strong luminescence emission enhancement towards Zn2+ and Cd2+ in DCM.76 In contrast, various other divalent metal cations, in particular Fe2+ and Ni2+ induced significant emission quenching. Of note is the formation of heterotrinuclear 1:2 host-guest stoichiometric complexes. This complex stoichiometry was also reported for binding of Zn2+ to 59, which again induced significant emission modulations in DCM.78 Specifically, the initial green emission of the free receptor was quenched with a concomitant appearance of a new, lower energy blue emission band. In contrast, complexation of 59 with Eu(III)(HFA)3 (where HFA ¼ hexafluoroacetylacetone), afforded a 1:1 host-guest complex displaying a significant red Eu(III) based emission, wherein the Pt(II) core acts as a photo-absorbing antenna. Related bis(arylamine) containing alkynyl Pt(II) receptors have also been developed for H+-tuneable Zn2+ sensing.79

430

Organometallic Receptors for Charged and Neutral Guest Species

Alkynyl Pt(II) motifs have also found wide-spread application in the assembly of supramolecular architectures, including host systems. For example, Mukherjee and co-workers assembled receptor 60 from an anthracene-based bis(ethynyl platinum) molecular “clip” and bis(pyridyl-imine) building blocks.80 The resulting rectangular shaped receptor was shown to recognize and sense transition metal cations Cu2+, Ni2+, Mn2+ and Fe3+ via emission quenching in MeOH. In contrast, Zn2+ and Cd2+ did not induce an optical response.

Such alkynyl platinum motifs have also found wide-spread use in the construction of more advanced metallacycles, such as crown-ether appended hexagons.77,81

14.12.2.5 Cation receptors based on other organometallic motifs A variety of other organometallic moieties have been incorporated into cation receptors, most notably those based on rhenium or ruthenium. For example, Godoy et al. appended benzo-15-crown-5 with a cyrhetrenyl ((Z5-C5H5)Re(CO)3) motif to afford receptor 61.82 In ACN, a blue shift of the MLCT absorption band was observed in the presence of various metal cations, with small perturbations induced by Li+, Na+ and K+, while Ba2+, Pb2+, Ca2+and Mg2+ elicited larger absorbance shifts. Similarly, all these metal cations induced significant emission quenching, whose magnitude was again larger for the divalent ions. Further analysis revealed a 1:1 host-guest stoichiometry for binding of Na+ and Ca2+ while 2:1 host-guest complexes were formed with Ba2+ and Pb2+. Notably, the Fc-analog of this receptor displays, as a result of its less electron-withdrawing nature, slightly larger binding constants towards all tested cations, but, as expected, no fluorescence. Similar aza-15-crown-5 appended Fc receptors were also reported for voltammetric sensing of Ba2+ and Ca2+.83

Re(I) complexes were also incorporated into multiple cation receptors by Yam’s group. In 2009, they reported a range of rhenium(I) bipyridyl coumarins 62–6584 and 66–69.85 Notably these systems display fluorescence resonance energy transfer (FRET) behavior wherein the coumarin acts as donor and the Re(I)(bipy) as acceptor. In ACN 62–65 displayed only small UV-vis changes towards the alkali metal ions Li+ and Na+, while the divalent metal cations Mg2+, Ca2+, and Ba2+, induced more significant perturbations, which where largest for Mg2+.84 No significant responses towards Zn2+, Cd2+ or Hg2+ were observed. In all cases, 1:1 host-guest stoichiometric binding was ascertained, whereby the shorter chain receptors 62 and 63 displayed strongest binding towards Ca2+ while receptors 64 and 65 preferentially bound the larger Ba2+. Emission spectrophotometric studies revealed interesting response patterns towards the different metal cations, whereby Li+ and Na+ generally induced a decreasing emission intensity of only the longer wavelength Re(I) complex, while Mg2+ and Ca2+ additionally induced emission enhancement of the coumarin fluorophore, with a larger response towards the former metal cation. In contrast, Ba2+ induced emission quenching of both fluorophores. The related receptors 66–69 generally displayed similar photophysical properties in ACN, with the same fluorescence emission pattern towards Mg2+, attributable to a decrease in FRET efficiency.85 These receptors also displayed absorbance changes in the presence of Li+, Na+, Mg2+, Ca2+, Ba2+ as well as Pb2+.

Organometallic Receptors for Charged and Neutral Guest Species

431

The related flavonol moiety, serving as both a fluorophore and cation receptor group, was appended to an alkynyl ruthenium complex by Fillaut et al., who developed receptor 70 as a Hg2+ selective optical probe.86 In ACN a selective absorbance change towards the mercuric ion was observed, while luminescence quenching was also observed for Zn2+ and Ni2+, albeit with slightly different spectral behaviors in each case (different magnitude of quenching as well as wavelength shifts).

Extensively studied in the early 2000s trinuclear ruthenium [12]metallacrown-3 complexes are well established and easily accessible organometallic receptors for Li+ with good selectivity over Na+.87 Building on this, Severin and co-workers reported a piperazine appended derivative 71, for fluorescent Li+ sensing in water.88 Specifically, a complex of the negatively charged fluorescent dye 8-hydroxy-1,3,6-pyrenetrisulfonate (HPTS) and, at physiological pH, protonated 71 was formed via electrostatic interactions. In neutral water this assembled system responded to Li+ via an increase in fluorescence intensity. Interestingly, this response does not arise from a typical dye displacement mechanism, but rather an impeded fluorescence quenching of the HPTS by electron transfer from the metallacrown in the presence of bound Li+. This assay not only displayed good selectivity over Na+ and Mg2+, but was also shown to work in deproteinized serum. However, a disadvantage was the long response time arising from slow binding kinetics. A further application of the parent ruthenium [12]metallacrown-3 complex 72 as well as the related (pentamethylcyclopentadienyl)rhodium(III) 73 was reported by Katsuta et al., who investigated these receptors as extraction agents.89 Specifically, the extraction of Li+, Na+, and K+ picrates from water into DCM was studied, revealing that both receptors displayed the same relative extraction trends of Li+ > Na+ > K+. The extraction constants Kex for all cations were slightly larger for the Ru metallacrown 72 (with Kex ¼ 5.25  106), while receptor 73 displayed a better selectivity of Li+ over Na+. Of note is the superior selectivity as well as the extraction capability of the Rh(III) complex for Li+ in comparison to the commonly employed Li+-selective crown ether dibenzo-14-crown-4.

432

Organometallic Receptors for Charged and Neutral Guest Species

In 2016, the Gilbertson group reported the 15-crown-5 appended pyridinediimine iron complex 74 as a redox active cation host.90 This receptor can, in THF, both be reversibly oxidized (ligand based redox) as well as reduced (metal based redox). In DCM only the oxidative couple can be resolved, which responds to Na+ and Li+ with moderate anodic shifts ( 50 mV). In THF the same perturbation was observed, but only for Na+. The same group later reported the related benzo-15-crown-5 containing 75, which, in DCM, displayed a smaller anodic perturbation of 31 mV towards Na+.87 These relatively small perturbations indicate only weak coupling between the crown ether binding site and the central iron complex. The authors exploited this in using the decoupled cation binding site within the secondary coordination sphere to catalyze the reduction of nitrite, which was accelerated  3.5-fold in the presence of Na+. This can be attributed to an ion-pair interaction between the receptor-bound Na+ and NO−2, thereby increasing the local concentration of the latter in proximity to the active catalytic site.

Similarly, cation-mediated regulation of catalytic activity has also been investigated for a range of rhodium-based hydroformylation catalysts91–93 and in related molybdenum model compounds.94,95

14.12.3 Organometallic anion receptors In analogy to metal cations, anions are important constituents of a wide variety of biological, environmental and industrial settings. For example, chloride, (bi)carbonate and phosphates play critical roles in various biological processes, while the latter, together with nitrate are linked to eutrophication of water systems. In view of these important functions, the recognition and sensing of anions has gained increasing attention. Due to their generally lower charge density, strong hydration and pH dependence, anion recognition remains, in comparison to cation recognition, a significant challenge, in particular in aqueous media. Nevertheless, significant advances in supramolecular anion chemistry have been made over the last three decades, whereby organometallic anion receptors, in particular those based on ferrocene, have often been at the forefront of the development of anion receptors and sensors. The significant anion binding enhancement afforded by oxidation of a Fc-containing receptor has early been recognized as a powerful strategy to address some of the aforementioned challenges (in particular with regards to binding affinity) and concomitantly affords a useful voltammetric sensing signal (see also Section 14.12.2.1). Unsurprisingly, Ferrocene remains the by far most common scaffold in organometallic anion receptors as highlighted in the following.

14.12.3.1 Anion receptors based on ferrocene Traditionally, Fc-containing anion receptors are based on NdH hydrogen bond donor receptive sites, a trend that persists in recent reports. This includes, for example Fc-appended amide96–98, imidazole11,99–102, (thio)urea103–106, hydrazone,107 amidine108 or guanidine109,110 conjugates. In these anion receptors, Fc mainly serves as a redox transducer, which typically undergoes well-defined cathodic voltammetric shifts upon anion binding, but has also been used as a purely structural, ball-bearing scaffold.111,112 Due to space limitations, the afore-mentioned sensors will not be discussed in detail, instead we will highlight more recent developments in Fc-containing anion receptors commencing with NdH, followed by CdH hydrogen bonding and halogen bonding systems. Mechanically interlocked molecules such as rotaxanes or catenanes, have received increasing attention as highly selective receptors for anions.113,114 In 2011, Evans and Beer reported the first example of a ferrocene-containing interlocked receptor, amide-based rotaxane 76, as a voltammetric anion sensor.115 In ACN, significant cathodic shifts towards Cl−, H2PO−4, BzO−, and HSO−4 were observed, which were largest for H2PO−4 (−100 mV), while Cl− induced only a modest response (−20 mV), but with strong quantitative binding. The same group also developed a Fc-appended [2]catenane, containing the same binding motifs, as a voltammetric chloride sensor exhibiting similar response patterns.116 The same redox behavior, i.e., a small cathodic perturbation of the Fc/Fc+ redox couple, but quickly plateauing response, indicative of strong guest binding, was also observed for Cl− binding to rotaxane 77 (in ACN/DCM 1:1), in which pentaphenyl ferrocene reporter groups were incorporated as stopper components in the axle.117 Importantly, this sensory response was not observed for the free axle.

Organometallic Receptors for Charged and Neutral Guest Species

433

Larger voltammetric perturbations upon Cl− binding were achieved with [3]rotaxane 78 in which the ferrocene transducer was incorporated into the center of the axle.118 As a result of two interlocked binding sites, 1:2 stoichiometric host-guest complexation of Cl− was observed in ACN, inducing cathodic shifts of −55 mV. The di-anionic SO2− 4 anion induced large magnitude cathodic perturbations of up to −270 mV (at 5 equiv.). A [3]rotaxane based on a C60-naphtalenediimide-functionalized axle and Fc-isophthalamide-containing macrocycles was also recently developed by Barendt et al. as an anion responsive molecular shuttle, wherein chloride binding-induced translocation of the Fc-macrocycle components suppresses electron transfer-quenching of the NDI fluorophore, restoring luminescence.119

434

Organometallic Receptors for Charged and Neutral Guest Species

Furthermore, Beer and co-workers developed a series of neutral isophthalamide Fc-containing [2]rotaxanes based on a pyridineN-oxide axle, which in ACN/DCM 1:1 displayed voltammetric perturbations towards Cl−, Br−, OAc− and H2PO−4 that were larger than that of the macrocycle alone.120 In 2007, Beer and co-workers also reported the first example of a mechanically interlocked surface-bound voltammetric anion sensor, the receptive self-assembled monolayer (SAM) 79SAM.121 This surface-rotaxane responded to Cl− in ACN (−40 mV) with a high selectivity over the oxoanions H2PO−4 and HSO−4. Notably, in solution, the macrocycle displayed a response to both these two oxoanions with only minimal perturbations in the presence of Cl−, while the axle alone did not significantly respond to any of the anions. These observations nicely illustrate the potency of an interlocked binding cavity in attaining a high degree of selectivity for a specific anionic guest. As alluded to in Section 14.12.2.1, such a receptor-confinement onto (conductive) surfaces is associated with a variety of benefits, including the potential for sensor re-use, flow sensing, more facile device integration and, importantly, surface-enhancement effects.32 For example, urea- and amide-containing calix[4]arene interface 80SAM, displayed, in DCM/ACN 1:1, large cathodic voltammetric shifts towards H2PO−4 (−250 mV), BzO− (−180 mV) and Cl− (−120 mV).122 Importantly, the magnitude of this response was significantly enhanced with respect to that observed in solution. This sensor also displayed a selective cathodic shift towards ReO−4 in pure water.

As a result of these benefits, a myriad of other interfacial voltammetric anion sensors have been developed over the last two decades. This includes electropolymerized thin films123 and surface-immobilized metallo-dendrimers, which have found wide-spread use in catalysis and as anion-sensors.124–126 Voltammetric, solution-phase anion sensing based on metallocenecontaining NdH hydrogen bonding dendrimers was extensively studied in the early 2000s (other more recent advances in anion-receptive organometallic dendrimers are detailed below), and has been increasingly explored at interfaces. This is facilitated by their strong tendency to adsorb onto electrode surfaces. For example, Astruc and co-workers modified electrodes with giant alkylferrocenyl dendrimers (generation six and seven dendrimers with 6561 and 19,683 Fc-termini, respectively).127,128 Even in the absence of a dedicated anion binding site these interfaces displayed cathodic perturbations towards H2PO−4 and ATP2− in DCM, as a result of electrostatic anion binding. Containing

Organometallic Receptors for Charged and Neutral Guest Species

435

a specific amide binding site, Villoslada et al. developed dendrimer film 81film (with n ¼ 4, 8, 16 or 32 side chains) as a voltammetric sensor for H2PO−4 in DCM.129 A positive dendrimeric sensing effects, that is a larger voltammetric perturbation for the higher generation dendrimer, was observed. The authors also reported sensing of H2PO−4 and ATP2− in pure water, demonstrating another crucial advantage of interfacial sensor systems in alleviating solubility constraints, which is often a particular issue for these architectures.

In another recent development, CdH hydrogen bond donors have gained attention as potent anion receptive motifs, most notably those based on imidazolium or 1,2,3-triazole heterocycles incorporated into a plethora of Fc-appended receptors. The former binding motif has been systematically studied by Jin and coworkers in receptors 82–87 with varying length of linker between the Fc and imidazolium groups.130,131 Expectedly, the magnitude of cathodic perturbation towards F−, Cl− and HSO−4 (in ACN) diminished with increasing linker length, whereby the magnitude of decrease in response was anion specific. For example, the response of 82–87 towards F− dropped from −230 mV for 82 to −12 mV for 85, while shifts of −65 mV and −10 mV towards Cl− were observed for the same receptors. Similar acyclic Fc-imidazolium sensors were developed by Zhuo et al.132–134 The C-H HB imidazole scaffold has also been incorporated into macrocyclic anion receptors.135,136 For example, Yu and co-workers demonstrated the utility of 88 as a voltammetric sensor for HSO−4.136 In ACN, this anion induced a cathodic shift of −52 mV, while much smaller perturbations were observed towards F−, H2PO−4 and OAc−.

The even more potent and versatile 5-proto-1,2,3-triazole has not only found wide-spread use in cation receptors (see Section 14.12.2), but is now also an established C-H donor motif in anion receptors and sensors.15 For example, in 2009, Tárraga and Molina developed Fc-triazole-pyrene receptor 89 for sensing HP2O3− 7 in DCM undergoing a cathodic response of −100 mV, while a large range of monovalent anions elicited no perturbations.137 By virtue of the appended pyrene fluorophore this probe also detected HP2O3− 7 via a significant fluorescence response. Kim and co-workers reported receptor 90 to selectively sense H2PO−4 and HP2O3− 7 in DCM, via voltammetric cathodic shifts of −200 and −180 mV, respectively.137 Additional amide binding sites were later incorporated into this scaffold between the triazole and Fc groups, which actually proved detrimental to the response magnitude towards H2PO−4 (−130 mV in DCM).138

436

Organometallic Receptors for Charged and Neutral Guest Species

A systematic study into a wide range of ferrocene triazole receptors 91–98 was conducted by the groups of Tárraga and Molina, − − − demonstrating these systems display in general selective voltammetric responses towards basic anions HP2O3− 7 , H2PO4, F and OAc , − − − − 16 while HSO4, NO3, Cl and Br typically elicited no response (the majority of studies were carried out in ACN). Expectedly, 91, 92 and 95 showed two redox waves, which, interestingly, in the case of the former two receptors merged into a single, cathodically shifted wave in the presence of anions (apart from the 91-F− complex). Notably, the triazolium receptors 92 and 98 were observed to exhibit a relatively enhanced cathodic response towards all anions in comparison to their neutral precursors 91 and 97. The triazolium motif was also integrated into an anthracene-amide-triazolium-ferrocene conjugate for fluorescent and voltammetric H2PO−4 sensing in ACN.139

Ferrocenyl-triazole motifs have also been incorporated into macrocyclic receptors.140,141 For example, Dai and co-workers investigated macrocyclic hosts 99 and 101 and their acyclic analogs 100 and 102.141 In DCM, all probes displayed a large response towards H2PO−4, while Cl−, F− and OAc− induced only minimal perturbations. Interestingly, the largest shift magnitude of −240 mV was observed for 100, with similarly large shifts also observed with 99 (−220 mV) and 102 (−200 mV), while macrocycle 101 exhibited a smaller perturbation (−79 mV). The generally enhanced response magnitude for the acyclic receptors in direct comparison with their macrocyclic analogs is most likely a reflection of the presence of two ferrocene groups in the former, affording more significant anion binding switch-on upon oxidation.

Organometallic Receptors for Charged and Neutral Guest Species

437

Click chemistry and the 1,2,3-triazole motif have also found use in the assembly of dendrimers, acting not only as a convenient linker motif but also as anion and/or cation binding site.142 Astruc reported the first examples of redox-active triazole-based dendrimers 103 (n ¼ 9,27,81) as voltametric ion sensors in 2007.143 In DCM, significant cathodic perturbations of up to −200 mV towards H2PO−4 as well as ATP2− were observed, while anodic shifts were reported in the presence of Pd2+, Pt2+, Cu2+ and Cu+. For all ions, apart from Cu+, a small negative dendritic effect was observed, that is the response magnitude of the higher generation dendrimers was diminished. Notably, the analogous cobaltocenium-based dendrimers displayed markedly altered response patterns and positive dendritic effects in DMF.144 Similarly, Fc-dendrimers 104 (n ¼ 27,81,243), containing an additional methylene spacer and opposite triazole connectivity showed a positive dendritic effect towards H2PO−4 in DCM with perturbations of up to −260 mV for 104 (n ¼ 243).145 Interestingly, all dendrimer generations responded to ATP2− with constant shifts of −180 mV. Upon physisorption of these receptors onto Pt electrodes an enhanced sensory response towards this anion was noted, with small positive dendritic effects. Anion as well as cation sensing was achieved with a range of biferrocene-triazole dendrimers 105 (n ¼ 9–729).146 As a result of electronic communication between the Fc moieties stepwise oxidation was observed, whereby the outer, more electron-rich Fc oxidizes at a  300 mV lower potential. Upon addition of ATP2− to 105 in DCM this redox couple displayed a cathodic perturbation, while addition of Pd2+ induced only changes of the higher potential couple, indicative of metal cation coordination at the adjacent triazole. The biferrocene moiety was also incorporated into amide receptors by Villena et al. and employed as H2PO−4 sensors in DMSO.147

A Fc-triazole dendrimeric interfacial sensor was also developed by a series of surface click reactions with alkyne-modified electrodes, which was subsequently employed for H2PO−4 sensing in DMF via scanning electrochemical microscopy.148 The 1,2,3-triazole scaffold has not only found significant use in C-H hydrogen bonding organometallic receptors (as highlighted above), but has more recently been expanded to the halogen bonding (XB) 5-iodo-1,2,3-triazole moiety, which quickly emerged as

438

Organometallic Receptors for Charged and Neutral Guest Species

a particularly potent anion binding motif.149 This has also been increasingly exploited in organometallic receptors. For example, in 2015 Beer and co-workers developed the first XB-based ferrocenyl voltammetric anion sensors 106 and 108.150 In ACN, both systems displayed moderate cathodic shifts towards Cl− (−32 and −30 mV, respectively) and Br− (−19 and −18 mV, respectively), a significantly enhanced response in comparison to their HB analogs 107 and 109. The related receptor 110 was later developed as an azide-selective voltammetric sensor, displaying a moderate, but enhanced response towards this anion (−40 mV), in comparison to Cl−, Br− and OAc− (max −22 mV for Br−; in ACN/H2O 99:1).151 The more stringent directionality requirements of the XB-anion interaction were directly exploited in the chiral (S)-BINOL-based redox active receptor 111, which in ACN showed different cathodic shifts for the enantiomers of N-Boc-Alanine (−57 mV (S), −63 mV (R)), N-Boc-Leucine (−55 mV (S), −77 mV (R)) and BINOL-phosphate (−55 mV (S), −38 mV (R)).152 Tucker and co-workers also reported enantioselective voltammetric anion sensors based on HB Fc-urea receptors, wherein the sensory chiral discrimination was typically less pronounced.153,154

In 2019, Lim et al. reported the first example of an all-halogen bonding redox active rotaxane 112 for voltammetric bromide sensing in ACN/acetone/H2O 45:45:10, with a notable selectivity over Cl− and SCN−.155. Hein and co-workers also recently conducted a systematic study of XB and HB Fc-isophthalamide-(iodo)triazole receptors as solution-phase and interfacial voltammetric anion sensors. Under diffusive conditions, the XB receptor displayed a significantly larger response than its HB analog towards a range of oxoanions and halides in both ACN and ACN/H2O 99:1, in good agreement with earlier studies detailed above. Upon surface-immobilization, affording 113SAM, a notable surface-enhancement effect was observed for both receptors, which was ascribed to surface-dielectric effects.156

Organometallic Receptors for Charged and Neutral Guest Species

439

14.12.3.2 Anion receptors based on metal carbonyl complexes Metal carbonyl complexes, in particular Re(I)(CO)3 motifs, continue to receive attention in anion receptor design. For instance, even Re(I)(CO)3 complexes with simple pyrazole, imidazole, or amino-pyridine ligands were shown to be potent anion receptors.157 Fletcher and co-workers developed the dinuclear amide-containing Re(I) receptor 114 as an optical and electrochemical phosphate sensor.158 In ACN, fluorescence emission enhancement towards this anion was observed, while voltammetric experiments reveal complex redox-properties and sensory behavior, characterized by two different Re+/Re2+ redox couples, which undergo first anodic and then cathodic perturbations upon H2PO−4 binding. This was attributed to intramolecular hydrogen bonding between a CO ligand and an amide binding site, thereby de-symmetrizing the receptor. This was further supported by 1H NMR studies which also revealed a 1:2 host-guest stoichiometric binding mode towards phosphate (b ¼ 3.5  105 M−2), notably stronger than Cl− binding (b ¼ 1.5  104 M−2). Other Re(CO)3-based optical probes include a receptor containing an imidazophenanthroline ligand for sensing of F− in DMSO.159 Tetra-cationic mixed metal Ru(bipy)3–Re(CO)3 metallomacrocycles were also developed for anion sensing in water. In the absence of any dedicated anion recognition binding sites, electrostatic interactions enabled binding and luminescent sensing of ATP2−.160. More recently, Re(I) carbonyl motifs have been incorporated into halogen-bonding receptors. For example, in 2015 Mole et al. reported the bimetallic pyrimidine-triazole receptors 115–116.161 In 1:1 CDCl3/MeOD binding of a large range of oxoanions and halides was reported, whereby the XB analog 115 displayed notably enhanced binding towards almost all anions, with a strong preference for halides over monovalent oxoanions. Iodide was bound most strongly with K > 10,000 M−1. In contrast, 116 displayed both reduced overall binding affinities and no notable selectivity trends. The integration of the Re(CO)3Cl motif into these receptors not only preorganized and polarized the halogen bond donor triazole binding sites but also enabled anion sensing via luminescence enhancements as observed for the halides, phosphate and sulfate in ACN.

Beer and co-workers also reported an all halogen bonding rotaxane 117 containing a Re(I)(bipy) reporter group for the highly selective sensing of halides.162 In ACN containing 10% or 20% H2O significant emission quenching upon binding of Cl−, Br− and I− were observed, while a large range of oxoanions elicited only very weak or negligible responses. In both solvent mixtures strong halide binding of 1:2 host-guest stoichiometry was observed, with K1 of up to 138,000 M−1 for I−. In all cases this anion was bound most strongly (I− > Br− > Cl−) with a lack of binding of F−. Following the same trends, strong binding and a fluorescence response were still observed in ACN/H2O 1:1, with K1 of up to 24,000 M−1 for I−. Langton et al. also reported a related all halogen bonding Re(I)(CO)3Cl-bistriazole rotaxane.163

440

Organometallic Receptors for Charged and Neutral Guest Species

Related molybdenum carbonyl motifs have also been investigated in anion receptors. For example, Ion et al. studied a range of amide and imidazole containing Mo(CO)x complexes 118–123.164 The neutral tetracarbonyl complexes 118 and 121 as well as the neutral methallyl chloride complexes 119 and 122 displayed very poor solubility such that preliminary anion binding studies could only be carried out in d6-DMSO, where moderate Cl−, H2PO−4 and OAc− binding, and no binding towards Br−, I−, HSO−4 and NO−3 with 118 was observed. The same trend, with even weaker binding was noted for 119. Receptor 122 displayed only modest Cl− and Br− affinities, while 121 could not be studied due to low solubility. The cationic methallyl tert-butylisocyanide-substituted complexes 120 and 123 were not only significantly more soluble, but also expectedly displayed stronger binding towards various anions. In ACN, 120 showed a preference for Cl− (K ¼ 17,200 M−1) with weaker binding of Br− > NO−3 > HSO−4 > OAc− > I− > ReO−4. 123 bound Cl− more weakly (K ¼ 11,500 M−1) and also displayed no significant discrimination between Br−, NO−3 and HSO−4 (K  10,500 M−1).

Very recently, Wright and co-workers developed a range of potent phosphazane anion receptors 124–125, preorganized by Rh(I) or Mo(0) carbonyl motifs.165 In DMSO, 2:1 host-guest stoichiometric chloride binding with K ¼ 1410 and 552 M−1, respectively, was observed, wherein the higher affinity of the former was attributed to enhanced electron-withdrawing ability of the higher oxidation state Rh(I) fragment. The related CF3-appended 126 displayed even stronger binding of K ¼ 4900 M−1, notably larger than that of structurally analogous thiourea or squaramide receptors. The authors further demonstrated that 126 was a potent transmembrane ionophore for Cl−.

Organometallic Receptors for Charged and Neutral Guest Species

441

Another anion receptive metal carbonyl complex [Fe4(m3-CO)4(Z5-C5H5)3(Z5-C5H4CONH(CH2)11SH)] was immobilized onto gold nanoparticles (127AuNP) by Astruc and co-workers and employed as redox sensor for phosphate anions.166 In DCM a single oxidative redox wave arising from the Fe4/Fe+4 couple was observed, which responded to the presence of ATP2− and H2PO−4 with cathodic shifts of −150 mV and −60 mV DCM, respectively. Such a high degree of ATP2+ selectivity is rarely achieved in related ferrocene-based voltammetric phosphate sensors (see Section 14.12.3.1).

14.12.3.3 Anion receptors based on cyclometalated iridium In spite of its cationic nature, cyclometalated iridium anion receptors remain, in comparison to cation receptors (Section 14.12.2.2), relatively rare. Based on an earlier study into (thio)urea appended Ir(ppy)(bipy) complexes,167 in 2017, Hyun and co-workers reported the chiral amino acid sensor 128.168 In ACN, the presence of various Boc-amino acids induced luminescence enhancements, which were notably different for the two enantiomers. While only moderate enantiodiscrimination of Boc-valine, Boc-threonine and Boc-phenylalanine was observed (with KD/KL  1.5), binding of D-Boc-phenylglycine (K ¼ 500,000 M−1) was fivefold larger than its L-enantiomer.

In 2007, a phenanthroline-benzimidazole appended Ir(ppy)2 receptor was reported as a luminescent anion sensor.169 Later, Rau’s group reported the related di-benzimidazole probe 129 and demonstrated emission intensity increases in the presence of Cl−, Br−, I− and HSO−4 in CHCl3, which were largest for the former two halides.170 The basic anions F−, OAc−, and H2PO−4 induced more complex response patterns, which were attributed to receptor deprotonation. The same group subsequently developed a more sensitive displacement assay based on the same receptor.171,172 Using the same XB 4,4-bis-iodotriazole bipy ligand of Beer’s all halogen bonding Re(I)(bipy) rotaxane 117 described above in Section 14.12.3.2, the group of Schubert in 2020 reported the first halogen bonding cyclometalated iridium receptor 130.173 This acyclic receptor’s chloride anion binding affinity (60,000 M−1) in ACN was found to be one order of magnitude larger than that of the HB analog 131 (5000 M−1), with weaker binding for Br− and OAc−. Emission studies also revealed a large, highly sensitive response of 130 towards Cl− (LOD ¼ 11 nM), with notably smaller, and similar, perturbations in the presence of the other two anions.

442

Organometallic Receptors for Charged and Neutral Guest Species

14.12.3.4 Anion receptors based on other motifs Organometallic platinum complexes have not only been applied in numerous cation sensors but have also been incorporated into anion receptors. It was demonstrated that direct coordination of the halides Cl− and Br− at the Pt center of Pt(II) NCN pincer presents a simple means of sensing these anions via emission enhancement,174 however this approach remains rare. More commonly, ligands functionalized with anion-receptive motifs are employed, typically amides.175–177 For example, Yam and co-workers developed 132 as an optical anion sensor.177 In acetone, significant fluorescence quenching towards various anions was observed, with the following binding trend: AcO− > F− > Cl− > Br− > I− with K ¼ 25,100 and 2570 M−1 for OAc− and I−, respectively.

The same group also carried out systematic studies into related mono and di-nuclear amide, urea and sulfonamide alkynylplatinum appended calixarenes and demonstrated the highly tuneable nature of the anion binding and sensing properties of these systems by judicious choice of the number and nature of the receptive Pt(II) complex.178 In 2018, the tricationic square planar gold(III) NHC complex 133 was reported as a surprisingly potent halide receptor.179 In d6-DMSO, formation of 1:2 host-guest stoichiometric complexes was observed in the presence of Cl−, Br− and I−, with strongest binding towards the former (K1 ¼ 11,500 M−1 and b ¼ 2.45  106 M−2). In water, 1:1 host-guest stoichiometric complexes were formed, with iodide bound the strongest (K ¼ 143 M−1). This tricationic receptor binds anions mainly by electrostatic interactions as well as additional CdH hydrogen bonds from the methyl and methylene groups. A recent report on similar Au(III) receptors with differing NHC ligand-linker group length demonstrated that these complexes can also respond to anions via changes in their absorbance.180 Altmann and Pöthig developed a similar, dinuclear Ni(II)-NHC receptor 134, which upon halide binding in CD3CN formed a dimeric tennis ball-like capsule.181 Of the halides, only in the presence of Cl− and Br−, was this 2:1 host-guest stoichiometric assembly formed, held together by electrostatic interactions between the Ni(II) centers and the halides as well as hydrogen bonds from the methylene-bridge protons which point into the capsule.

Organometallic Receptors for Charged and Neutral Guest Species

443

Anion recognition mediated by CdH hydrogen bonding interactions was also reported by Bedford et al. in palladacyclic receptor 135.182 Specifically, in CDCl3 binding of the halides Cl−, Br− and I− occurred via multiple CdH hydrogen bonds from the pre-organized and polarized Pd-ligated thiacrown ether. This binding was surprisingly strong with K ¼ 2200 M−1 for I− and even stronger association of Br− and Cl−, which could however not be reliably quantified as a result of some degree of ligand displacement. In a recent study, tri-iridium or rhodium p-metalated pillar[5]arenes showed binding of a triflate counteranion within the host’s cavity, ascribed to anion − p interactions.183 Building on this, Huang and co-workers investigated a series of p-metalated paracyclophanes as anion receptors.184 The tri-Ru(II)(cymene) receptor 136, displayed binding of the large, charge-diffuse anions ReO−4, ClO−4 and NO−3, with K ¼ 535, 186 and 143 M−1 in CD3CN, respectively. Notably, in water binding of the former two anions was even stronger, with K ¼ 1880 and 840 M−1, while NO−3 binding was attenuated (K ¼ 78 M−1). In this solvent moderate iodide − − − binding (K ¼ 182 M−1) was also observed, while SO2− 4 , H2PO4, Cl and Br displayed only very weak or negligible binding. It was − further shown that the receptor was capable of ReO4 extraction from water into nitromethane with no significant interference from various oxoanions, apart from ClO−4 if present in excess.

Other receptors based on the Ru(II)cymene fragment, appended with pyridyl-amine ligands, were also developed by Steed and co-workers as fluorescent anion sensors.185

14.12.4 Organometallic ion-pair receptors The concomitant recognition of both a cation as well as an anion within a heteroditopic host structural framework containing dedicated binding sites for both species, has gained increasing attention as a potent means of achieving a higher degree of binding strength and selectivity in comparison with monotopic host systems.186 This is afforded when binding of either ion cooperatively enhances binding of the other ion, an approach that is all the more appealing in real-life relevant applications, where ions are generally encountered as salts containing “coordinating” counter-ions.187,188 For instance, when anion binding is sought, the presence of its counter cation, typically an alkali metal cation, can be exploited to enhance anion binding by simultaneous cation recognition.

444

Organometallic Receptors for Charged and Neutral Guest Species

A receptor-bound cation can also directly serve as the anion binding site, as demonstrated in zinc(II) receptors, which, following cation recognition, can serve as phosphate receptors.189,190 For example, Yam and co-workers recently employed a DPA-appended Pt(II) alkynyl complex as a luminescent Zn2+ sensor (see Section 14.12.2.4 for related examples), which upon subsequent addition of pyrophosphate undergoes further optical changes and forms supramolecular oligomeric assemblies as a result of the crosslinking of the DPA-bound Zn2+ via the phosphate anions.190 Such examples will not be further discussed, instead, in the following section we focus our attention on organometallic receptors containing both discrete anion and cation binding sites.

14.12.4.1 Ion-pair receptors based on ferrocene As described in Sections 14.12.2.1 and 14.12.3.1, ferrocene-appended imidazoles can serve as both cation or anion receptors and have been employed as redox as well as optical ion sensors.100 Building on this, the groups of Tarraga and Molina developed a range of Fc-imidazole receptors that can act as potent ion-pair sensors. For example, in 2011, they reported Fc-imidazopyrene 137 as a voltammetric and fluorescent sensor for ion-pairs.191 This receptor underwent changes in its absorbance spectrum upon exposure to both cations (Pb2+, Hg2+ and Zn2+) or anions (OAc− and H2PO−4) in ACN, whereby the cations formed 2:1 host-guest stoichiometric complexes with b up to 8.71  1010 M−2 for Hg2+, while the anions formed 1:1 stoichiometric complexes, with strongest binding towards OAc− of K ¼ 8.22  105 M−1. All ions also induced notable fluorescence increases, which were largest for Pb2+ and generally larger for the cations than the anions. When the divalent metal cations were added to the 137-H2PO−4 complex, a further enhanced, red-shifted emission was observed, arising from formation of ion-pair complexes, with a suggested stoichiometry of two hosts, one metal and two anions. Again, the largest optical response was observed in the presence of Pb2+, while all ion-pair complexes with OAc− generally displayed smaller perturbations. Ion-pair binding was also ascertained from voltammetric experiments. As expected, the separate addition of cations and anions induced anodic or cathodic shifts, which where largest for Hg2+ (279 mV) and AcO− (−212 mV), respectively. In the presence of both Zn2+ and H2PO−4 intermediate responses, slightly more cathodically perturbed than the free receptor, were observed. The same group also developed a range of furan, thiophene and pyrrole-appended imidazole-quinoxaline ion-pair receptors 138–140.11,102,192 In ACN, receptors 138 and 139 responded to the divalent metal cations, Cd2+, Zn2+, Hg2+, and Pb2+ with anodic shifts, which for 138 were largest in the presence of Hg2+ (195 mV), while 139 responded most sensitively to Pb2+ (240 mV).11 In analogy to the related receptor 137, the probes also responded to OAc− and H2PO−4, while Cl−, Br−, NO−3 and HSO−4 induced no perturbations. Both receptors displayed a preference for OAc− with cathodic shifts of −205 and −170 mV, for 138 and 139, respectively. Even larger perturbations were observed with F− and H2PO3− 7 , albeit due to deprotonation. Interestingly, and in contrast to most voltammetric ion-pair sensors, the addition of metal cations to the H2PO−4-complexed receptors induced anodic shifts that were larger than those induced by the cations alone (between 225 and 265 mV with respect to the free receptor). Impressively, this behavior was also observed for Ni2+ and Mg2+, which alone do not induce perturbations of the free receptor. Similarly, all metal cations, including Ni2+ and Mg2+ induced a larger response in the presence of HSO−4 (between 220 and 275 mV with respect to the free receptor). It is noteworthy that HSO−4 alone does not induce a response, thus, for Mg2+/Ni2+ and HSO−4 a response is only obtained if both are present simultaneously. Additionally, both receptors also acted as fluorescent ion-pair sensors. While 138 only responded to Pb2+ and Zn2+ with emission enhancements, 139 additionally responded to Cd2+. In the presence of anions, neither receptor displayed significant fluorescence perturbations. In analogy to voltammetric experiments, the presence of both cations and anions induced an emergent behavior that was not observed for either ion alone. Specifically, ion-pair binding led to fluorescence enhancements that were much larger than those attained with the free metals, with emission turn-on of up to 965-fold in the presence of Pb2+ and H2PO−4. Ni2+ and HSO−4 together, again, induced a notable response (95-fold increase), in spite of no optical response towards either ion alone. Pyrrole-based receptor 140 displayed surprisingly different binding and especially sensory behavior towards ions and ion-pairs in ACN.102 UV-vis studies revealed perturbations only in the presence of cations with a 1:1 host-guest binding stoichiometry for Pb2 + and Hg2+, while Cd2+ and Zn2+ were bound in a 2:1 host-guest fashion. In more polar MeOH only Hg2+ induced perturbations. Addition of either Cd(OAc)2 or Pb(OAc)2 caused absorbance changes that were identical to those of the metals alone. In contrast, in the presence of Ni(OAc)2 and Hg(OAc)2 red shifted absorption bands were observed, indicative of ion-pair binding (Ni2+ alone did not bind). In MeOH only Hg(OAc)2 elicited such a response. Interestingly, the fluorescence emission of 140 was enhanced in the presence of Pb2+, Hg2+, Zn2+ as well as OAc− and H2PO−4, with up to 120-fold turn on (for Pb2+; in ACN), while upon ion-pair binding quenching was observed in all cases. Voltammetrically, 140 responded with the expected anodic shifts towards Zn2+ (50 mV), Cd2+ (80 mV), Hg2+ (225 mV), and Pb2+ (225 mV) as well as cathodic perturbations towards OAc− and H2PO−4 (both −120 mV). However, no response was observed towards any of the metal acetate salts. In 2015, Wan et al. developed another imidazole-based ion pair receptor 141, incorporating a 2,2-diferrocenylpropane motif.193 Upon binding of Pb2+ in ACN a naked eye color change from yellow to red as well as 42-fold fluorescence emission enhancement was noted. Zn2+ also induced increased emission (11-fold), both of which were further enhanced in the presence of HSO−4. Various other anions induced fluorescence quenching, ascribed to cation abstraction. In DCM this probe also sensed ions voltammetrically, with shifts of 151 mV for Pb2+ and 78 mV for Zn2+ of only the higher-potential couple arising from the imidazole appended Fc. In contrast, both Fc moieties responded with −70 mV cathodic shifts towards HSO−4. Addition of this anion to the 141-Hg2+ complex induced shifts that are  30 mV anodic with respect to the free receptor, indicative of ion-pair binding.

Organometallic Receptors for Charged and Neutral Guest Species

445

Tarraga, Molina and co-workers also developed ferrocene-triazole derivatives as ion pair receptors, including 142 and 143.194–196 The former displayed selective cathodic voltammetric shifts of −110 mV towards HP2O3− 7 in ACN/DCM/H2O 90:8:2, with no significant perturbations towards halides or monovalent oxoanions.194 Upon formation of 2:2 stoichiometric host-guest complexes, the pyrophosphate anion also induced changes in the absorbance as well as excimer fluorescence emission. In ACN/DCM 3:2, Pb2+ and Hg2+ induced significant anodic shifts of 75 and 155 mV, respectively, while Ni2+ and Zn2+ only induced moderate perturbations in high excess. Absorbance changes were also reported for the former two cations, with significant emission quenching observed only for Hg2+. Optical ion pair studies revealed addition of Ni2+, Zn2+ or Hg2+ to the 142-HP2O3− 7 complex induced dissociation. In contrast, upon exposure to Pb2+ a color change from yellow to green was observed. Additionally, the fluorescence excimer band of 142-HP2O3− 7 disappeared and the original emission was restored, albeit with lower intensity than that of the free receptor. Ion-pair binding of Pb2+ and HP2O3− 7 was also observed electrochemically, as indicated by intermediate voltammetric shifts, some 30 mV anodic with respect to the free receptor. The related receptor 143 also acted as an electrooptical sensor, but for a larger range of ion-pairs.195 In DCM/ACN 1:4 both H2PO−4 and OAc− induced cathodic shifts of −130 mV, while mono as well as divalent metal ions did not cause any perturbations. However, when Zn2+, Cd2+, Mg2+ or Ca2+ were added to the anion-receptor complex, moderate shifts back towards the original, free receptor potential were observed (between 29 and 48 mV). UV-vis experiments corroborated these results, with minimal binding towards the metal cations and binding of the anions, observed. The latter also resulted in a ratiometric fluorescence response arising from pyrene excimer formation (under 2:2 host-guest complexation). Subsequent addition of cations to 143-H2PO−4 caused significant emission enhancements (up to ninefold with Zn2+), while the addition of all cations to 143-OAc− resulted in emission quenching.

Recently, Roma nski and co-workers prepared benzo-15-crown-5 and benzo-18-crown-6 ether appended squaramide receptors 144 and 145 as redox active ion-pair sensors.197 Expectedly, the anion binding strength of both receptors was largely identical, following the order Cl− > NO−2 > Br− > NO−3 in ACN, as revealed by UV-vis studies. In the presence of Na+ both receptors displayed enhanced binding towards all anions, while in the presence of K+ 145 displayed even larger binding enhancements for all anions with an of up to 3.4-fold larger binding towards Cl−. Voltammetric studies revealed only small anodic perturbations for both receptors of 10 mV towards Na+ and K+, with somewhat larger cathodic perturbations towards anions. Furthermore, in most cases the response towards anions was not significantly larger in the presence of a co-bound cation. Only in the presence of one equivalent of both Na+/K+ and Cl−/Br− was a −20 mV more cathodic response observed, while at higher anion concentrations this difference was less pronounced.

446

Organometallic Receptors for Charged and Neutral Guest Species

14.12.4.2 Other organometallic ion-pair receptors As a result of their strong Lewis acidity, organotin compounds have been employed in various anion and ion-pair receptors, a field largely developed by the group of Jurkschat.198 For example, in 2007, 16-crown-5-appended receptor 146 was reported.199 Of note is an intramolecular O!Sn interaction (not shown), which, in CD2Cl2 is broken upon either Na+ crown ether binding or via Cl− coordination at the Sn center, highlighting the ditopic nature of the receptor. However, for this receptor no simultaneous recognition of Na+ and Cl− was achieved, as binding could not overcome the high lattice enthalpy of the salt. Only with NaSCN was ditopic binding observed in both C2D2Cl4 and CDCl3. Of note are slow kinetics (with equilibration over several days) and complex SCN− binding equilibria with both a single coordination to Sn as well as Cl−-ligand displacement. An improved receptor design, 147, containing an additional Sn anion recognition site, was shown to be able to overcome the high lattice energy of NaF and to bind this salt ditopically via convergent F− interactions with both Sn centers.200 This was achieved both in solution (in CDCl3 and CD3CN) and was later also demonstrated in the solid state.201 The Jurkschat group also developed a range of related 13-crown-4 as well as 19-crown-6 analogs of both receptors and demonstrated ditopic recognition of LiCl, LiI and CsF with the former as well as KF and KSCN binding with the latter.202,203 Ph2FSnCH2SnFPh-CH2-19-crown-6, the difluoro 19-crown-6 analog of receptor 147, was also employed as a transporter of KF through a liquid organic membrane.204 The transport rates of the ditopic receptor were up to 5.5 times larger than those of an equimolar mixture of the separate ionophores Ph2FSnCH2SnFPh2 and 18-crown-6.

In 2013, Jurkschat and co-workers further reported ferrocenophane organotin 13-crown-4 receptor 148.205 In agreement with other studies on similar monometallic receptors (see above), this receptor was not able to bind LiF, but instead formed ditopic complexes with LiCl in CD3CN. By virtue of the appended Fc moiety this host also responded to LiCl binding by cathodic shifts of −66 mV. In the same year bis(16-crown-5) receptors 149 and 150 were developed.206 In analogy to the aforementioned organotin compounds, the free receptors display intramolecular O ! Sn coordination, in this case from both crown ether appendages, affording octahedrally coordinated Sn centers. Interestingly, upon addition of NaI to 149 in CDCl3, the coordination of only one of the crown-ether oxygen atoms to the metal is disrupted by Na+ binding, while no significant I− anion binding to the Sn center was ascertained. In contrast, 150 ditopically binds NaBr, again by Na+ recognition at only one of the crown-ether moieties in polar CD3CN and MeOD solvent media.

Organometallic Receptors for Charged and Neutral Guest Species

447

14.12.5 Organometallic receptors for neutral guests In contrast to ion recognition, the development of receptors for neutral guest species has only gained significant attention more recently. This has largely been driven by a desire to recognize various organic molecules, in particular polycyclic aromatic hydrocarbons (PAHs), such as pyrene, perylene or coronene, due to their applications in organic electronics as well as significant concerns over their toxicity. As a result, various acyclic and macrocyclic receptors, typically containing large p-conjugated surfaces, have been developed, many based on organometallic fragments. In these receptors, organometallic motifs typically serve as structural scaffolds, enabling coordination-driven self-assembly, but have also been employed as reporter groups, as highlighted in the previous sections.207,208

14.12.5.1 Receptors based on ferrocene In 2013, Bivaud et al. reported the self-assembled ferrocene-cornered cages 151–152 for the recognition of perylene.209 As a result of two extended TTF panels and four (diphenylphosphino)ferrocene (dppf ) vertices, these receptors displayed two successive oxidation waves, with an exchange of four electrons each. In spite of the high charge this endows onto the cages, these assemblies remained intact, as indicated by only small changes in the voltammograms upon repeat cycling. Of note is that the exTTF panels are not flat, but adopt a bent conformation which imbues the receptors with an ovoid cavity. Binding studies in CD3NO2 indicated no interaction of pyrene or coronene, but strong 1:1 host-guest stoichiometric binding of perylene, with K ¼ 3900 and 3200 M−1 for 151 and 152, respectively.

The same Pd(II)/Pt(II)(dppf ) motif was also incorporated into carbazole-containing metallamacrocycles 153 and 154 by the groups of Chi, Stang and Mukherjee.210 These receptors adopt a bowl-shape structure with a large internal concave aromatic surface. This enables strong 1:1 host-guest stoichiometric binding of buckminsterfullerene (C60) in ACN (K ¼ 100,000 M−1), whereupon fluorescence quenching was observed. Related receptors for C60, C70 and their derivatives were also developed based on subphthalocyanine and dppf building blocks.211

448

Organometallic Receptors for Charged and Neutral Guest Species

Recently, Kasprzak et al. developed a ferrocene-imine trigonal prismatic cage as a molecular reactor for Suzuki-Miyaura reactions.212 In THF, the free cage 155 responded to binding of p-terphenyl, chrysene, pyrene and 1-formylpyrene by changes in absorbance and weak fluorescence quenching. For these large guests 1:1 stoichiometric binding was observed, with binding constants of up to 650 M−1 for 1-formylpyrene. In contrast, the smaller chlorobenzene and phenylboronic acid were bound with a 1:3 host-guest stoichiometry (K ¼  600 M−3 for both). Upon coordination of three Pd(II)Cl2 motifs between the imine groups, the cage was transformed into a Suzuki-Miyaura catalyst and employed for the synthesis of a range of 1,10 -biphenyls.

Ferrocene has also been appended to more established macrocyclic receptors for neutral molecules, in particular cyclodextrins (CDs). For example, Casas-Solvas et al. reported the ferrocene-triazole conjugate 156 in 2009.213 In water, this receptor forms a dimeric self-inclusion complex, wherein the Fc moiety binds to the cavity of b-CD. Upon addition of the bile salts cholate, deoxycholate or chenodeoxycholate, 1:1 host-guest stoichiometric inclusion complexes with the b-CD moiety are formed, with

Organometallic Receptors for Charged and Neutral Guest Species

449

K ¼ 2300 M−1 for deoxycholate, thereby displacing the Fc appendage. This was exploited in a voltammetric displacement assay, wherein guest binding induced significant cathodic shifts of the Fc couple upon its displacement into the more polar solvent environment. Binding of chenodeoxycholate induced the largest perturbations of −90 mV. A structurally related Fc-b-CD receptor was demonstrated to self-assemble into supramolecular aggregates via Fc binding within the CD cavity.214 Similarly, Smith and co-workers reported that a ureidopyrimidone-ferrocene dimer, formed by self-recognition in DCM, can be reversibly disassembled upon Fc oxidation.215

14.12.5.2 Receptors based on half-sandwich complexes Due to their well-defined coordination geometries, ruthenium, and to a lesser extent iridium and rhodium based half-sandwich complexes, have received considerable attention as facile building blocks in organometallic architectures and receptors.207 They are commonly employed for the assembly of macrocyclic metalla rectangles and prisms. For instance, the group of Therrien assembled the macrocyclic receptors 157–159 by reaction of a 1,4-naphthoquinonato (dhnq)-bridged Ru(II)(cymene)Cl molecular “clip” with the nitrogenous linkers pyrazine (pyz), 4,4-bipyridine (bipy) or 1,2-bis(4-pyridyl)ethylene (bipy(ethylene)).216 In CD3CN, 157 displayed no binding of pyrene, as elucidated by 1H NMR studies. In contrast, the bipy-containing 158 interacted with pyrene, albeit by formation of “exclusion” complexes, i.e., an interaction of the guest with the outside of the cage. Only for the larger 159, possessing a spacing of the Ru(II) clips of 13.6 A˚ , was the formation of an inclusion complex with pyrene observed. The same authors further conducted more detailed binding studies of the same bipy and (bipy(ethylene) cages, containing the larger anthracene (dhaq) and tetracene (dhtq) quinonato bridges (160–161 and 162–163).217 In CD3CN, DOSY NMR experiments indicated out-of-cavity interactions of the smaller receptors 158, 160 and 161 with anthracene, pyrene, perylene, and coronene. Even at −40  C only anthracene formed 1:1 host-guest stoichiometric inclusion complexes. In contrast, receptors 159, 162 and 163 displayed both “inside” and “outside” interactions only with anthracene, while all other guests formed inclusion complexes with strong binding, but poor selectivity. Specifically, in all cases the binding constants were between 52,000 and 69,000 M−1 with no noticeable trends with respect to either bridge or guest size. The authors further showed that in DCM perylene fluorescence is quenched by binding to host 159. Jin and co-workers showed that iridium or rhodium derivatives of these cages can recognize halocarbons in the solid state218 and can aid in the separation of disubstituted benzene derivatives.219

450

Organometallic Receptors for Charged and Neutral Guest Species

In 2011, Shanmugaraju et al. studied a range of metallorectangles assembled from Ru(II)(oxalate) clips, bridged by various bis(pyridines).220 Among these, 164 displayed fluorescence in MeOH, which, upon binding of the nitroaromatics nitrotoluene, nitrobenzene and 2,4,6-trinitrotoluene (TNT), underwent emission quenching. Other electron deficient aromatics like benzoic acid or 4-methoxybenzoic acid also induced such a response, albeit weaker. Although electron-deficient, benzoquinone induced fluorescence enhancements. Mukherjee, Chi and Stang also developed fluorescent sensors for nitroaromatics based on prismatic cage analogs of receptor 164.221

Organometallic Receptors for Charged and Neutral Guest Species

451

An even longer bispyridyl linker 165 was combined with the bimetallic clip [Ru2(p-cymene)2(dhtq)]2+ by Chi and co-workers.222 The so-generated [Ru4(p-cymene)4(dhtq)2(165)2]4+ displayed notable fluorescence quenching upon addition of C60 in 1,1,2,2-tetrachloroethane, forming 1:1 host-guest stoichiometric complexes.

In 2008, Therrien et al. developed a prismatic cage 166 consisting of two trigonal 2,4,6-tris(pyridin-4-yl)-1,3,5-triazine (tpt) panels linked by a bimetallic clip containing 2,5-dihydroxy-1,4-benzoquinonato (dhbq) bridges.223 The formation of this cage from its subcomponents can also be carried out in the presence of [Pd(acac)2] or [Pt(acac)2] to afford “complex-in-a-complex” assemblies. In water, the release of these hydrophobic drug guests is very slow, even at elevated temperature. In contrast, the more hydrophilic cisplatin quickly leaches out of the cage. As a result of their low water solubility, both [Pd(acac)2] and [Pt(acac)2] display no toxicity towards human A2780 ovarian cancer cells, while the empty cage displayed moderate toxicity with an IC50 of 23 mM. Encapsulation of the drugs within the receptor notably improved their efficacy, with IC50’s of 12 and 1 mM for the 166[Pt(acac)2] and 166-[Pd(acac)2] assemblies, respectively.

Recognition and delivery of drugs and fluorescent dyes based on related ruthenium224–228 and osmium-based229 prismatic cages has also been studied in detail. Similarly, Therrien’s group developed larger Ru(p-cymeme)-based cages for the recognition and cellular delivery of the photosensitizers porphin, phthalocyanine, and Zn-phthalocyanine.230,231 The same group also studied two prismatic cages [Ru6(p-cymene)6(dhnq)2(3-tpt)2]6+ and [Ru6(p-cymene)6(dhtq)2(3-tpt)2]6+ based on the isomeric tris(3-pyridyl)triazine (3-tpt) capping ligand.232 These cages were assembled in the presence of 1,3,5-tribromobenzene, phenanthrene, pyrene, or triphenylene as template, whereafter the large triphenylene is permanently encapsulated, while the smaller templates can be removed in refluxing toluene as indicated by 1H NMR and UV-vis studies. Of note is that cage [Ru6(p-cymene)6(dhnq)2(3-tpt)2]6+, containing the smaller dhnq bridge, was obtained as two isomers with different orientation of the pyridine units of the 3-tpt capping ligand. In CD3CN, triphenylene is bound more strongly than pyrene, as observed by qualitative displacement studies, and preferentially binds to the symmetrical cage isomer, in spite of the large portal size of the asymmetric isomer. As a result of its smaller portal size, receptor [Ru6(p-cymene)6(dhtq)2(3-tpt)2]6+ displayed slower guest exchange towards both pyrene and phenanthrene, the latter being expectedly faster.

452

Organometallic Receptors for Charged and Neutral Guest Species

In a separate study, binding of phenanthrene and pyrene to the analogous 4-tpt based [Ru6(p-cymene)6(4-tpt)2(dhtq)3]6+ receptor in CD3CN was quantified as 29,000 and 22,300 M−1, respectively, while the cage with the slightly smaller anthracene-based dhaq bridge displayed slightly weaker encapsulation of both guests of  20,000 M−1.233 Jin and co-workers developed analogous metalloprisms [M6(Cp∗)6(4-tpt)2(dhtq)3]6+ (M ¼ Ir/Rh; Cp ¼ pentamethylcyclopentadienyl) containing iridium(III) or rhodium(III)-based corners, and demonstrated the encapsulation of pyrene, coronene, [Pt(acac)2] and hexamethoxytriphenylene in CD3OD.234 In 2014, Therrien’s group incorporated the zwitterionic quinonoid bridging ligand 167 into 4-tpt-based cage [Ru6(p-cymene)6(4-tpt)2(167)3]6+.235 As a result of the redox-activity of this ligand multiple reductive couples are observed in THF, arising from reduction of the metal-quinonoid bridges as well as further reduction of the 4-tpt panels at lower potentials. Upon binding of coronene no significant potential shifts, but improved current stability upon repeat cycling was observed, thus imparting higher redox stability.

The Rh2(Cp )2(quinonato) metalloclip has very recently been exploited to form larger prismatic cages with the extended 1,3,5-tris(pyridin-4-ylethynyl)benzene (tpeb) capping panel.236 When the wider dhtq component was incorporated as the quinonato bridge, a monomeric metallacage [Rh6(Cp∗)6(tpeb)2(dhtq)3]6+ was observed. In contrast, the smaller dhnq and dhaq ligands induced formation of triply interlocked [2]catenane prisms. Upon binding of anthracene, pyrene, triphenylene or perylene in MeOH, these [2]catenanes dynamically reassembled to afford the monomeric, non-interlocked host-guest complexes.

14.12.5.3 Receptors based on NHCs Over the last decade, N-heterocyclic carbenes (NHCs) have been extensively studied as versatile building blocks in organometallic receptors, owing to their compatibility with a wide range of transition metals, strong MdC NHC bonds, and facile incorporation into extended p-conjugated systems.237 An early example of such a system, 168, was reported by Zeng, Xu, Zhang and coworkers in 2007.238 This calix[4]arene-type receptor displayed, by virtue of appended anthracene motifs, fluorescence, which, in ACN, was quenched upon binding of C60. This proceeded with a host-guest binding stoichiometry of 1:1 and K ¼ 348,000 M−1. The same authors later studied an extended receptor analog, containing an additional tert-butyl methoxybenzene unit, demonstrating luminescent sensing of both C60 and C70.239

Organometallic Receptors for Charged and Neutral Guest Species

453

Liu et al. also developed Ag(I)NHC cyclophanes 169 and 170 as optical sensors for p-phenylenediamine.240 In ACN both receptors displayed absorbance and fluorescence changes upon addition of the guest, with K ¼ 45,000 M−1 for 169, and slightly stronger binding of 106,000 M−1 for 170, however the specific binding mode was not elucidated.

Recently, Peris and co-workers constructed two nickel(II)-based metallorectangles 171–172, containing pyrenebisimidazolylidene NHC ligands, as PAH receptors.241 The smaller, pyrazine linked cage 171 formed 1:1 host-guest stoichiometric inclusion complexes with various aromatic guests in d6-acetone, with the following binding order: perylene < naphthalene < phenanthrene < anthracene < pyrene < triphenylene, clearly a reflection of complementary guest size, with a best “fit” and binding afforded for triphenylene (K ¼ 4400 M−1). The slightly larger perylene was bound weakest (130 M−1), while the even bigger coronene was not bound at all. The larger bipy bridged cage 172 was able to incorporate two pyrene and triphenylene guests, whereby the latter was again bound stronger with K1 and K2 of 181 and 40 M−1, respectively. This interesting 1:2 host-guest stoichiometric complex formation was also confirmed for the 172-(pyrene)2 adduct by single crystal XRD. As a proof-of-principle, the authors further showed that a suspension of cage 171 can be employed to sequester pyrene from heptane, whose fluorescence is quenched by binding to the receptor. The host-guest complex was then simply removed by filtration. The nickel(II) pyrene-bisimidazolylidene clip was also assembled into trigonal prisms in the presence of 1,3,5-tripyridyl-triazine capping ligands.242 This large cage bound both C60 and C70, with surprisingly large selectivity for the latter. Specifically, in acetone/ dichlorobenzene 1:4, C70 binding with K ¼ 35,000 M−1 was almost one order of magnitude larger than recognition of C60 (4700 M−1). This was rationalized by an entropic driving force for encapsulation, which is enhanced for the larger C70 as a larger number of solvent molecules are released from the C70 guest upon binding. The significant selectivity for C70 was further demonstrated by competitive binding experiments. When the receptor was exposed to an equimolar mixture of C60 and C70 in d6-acetone, only the uptake of C70 was observed. Similarly, one equivalent of C70 was capable of completely displacing bound C60 from the cage. This can potentially be exploited in the purification of these fullerenes.

Recognition of C60 and C70 was also achieved with the symmetric Pd(II)-allyl cornered square 173.243 In acetonitrile/dichlorobenzene 1:4 this cage displayed not only stronger binding than the afore-mentioned Ni(II)-based metalloprism, with K ¼ 5400 and 71,000 M−1 for C60 and C70, respectively, but also a larger selectivity for C70 (13-fold). This recognition process was, again,

454

Organometallic Receptors for Charged and Neutral Guest Species

entropically driven. Single crystal XRD studies of the free receptor as well as both fullerene adducts revealed an interesting guest-dependent adaptation of the receptor. The free cage 173 possessed a highly symmetrical structure with an average distance of 13.25 A˚ between the palladium atoms and only a small torsion angle between the two Pd-Pd axes of 10.6 . In the 173-C60 complex the receptor is significantly twisted, with a torsion angle of 21.7 , thereby shrinking its size. This is reflected in slightly smaller distances between opposite pyrene panels of 13.20 A˚ (in comparison to 13.39 A˚ in the free cage). In addition, the pyrene motifs are significantly curved so as to maximize their interactions with the fullerene guest. In contrast, the 173-C70 complex is somewhat expanded with larger pyrene-pyrene distances of 13.67 and 14.28 A˚ and significant bending of the pyrene motifs. This size-adaptation was also confirmed by consideration of the free volume of the cage calculated upon guest “removal” of 710, 650 and 696 A˚ 3 for 173, 173-C60 and 173-C70, respectively, corresponding to a shrinkage of 8.5% upon C60 binding. In a separate report, the authors demonstrated that the 173-fullerene complexes can serve as stable and efficient singlet oxygen photosensitizers.244 In ACN, both 173-C60 and 173-C70, but not the free receptor, produced 1O2 after irradiation with 532 nm laser light with quantum yields of 23% and 41%, respectively, somewhat lower than free C60 or C70 in benzene. The so-generated singlet oxygen was then utilized for a range of oxidation reactions of acyclic and cyclic alkenes. The versatility of 173 was further showcased in a binding study with PAHs.245 In CD3CN both pyrene and triphenylene formed 1:2 host-guest stoichiometric complexes with the cage, with K1 and K2 of 220 and 46, and 280 and 52 M−1, respectively. UV-vis studies also confirmed binding of coronene with the same stoichiometry and K1 and K2 of 50,000 and 7400 M−1, respectively. In the presence of the electron-deficient N,N0 -dimethyl-naphthalenetetracarboxydiimide (NTCDI), quintuple D-A-D-A-D stacks were formed whereby the NTCDI acceptors intercalated between the electron rich pyrene cage panels and the PAH guest, affording an overall 173/PAH/NTCDI ratio of 1:1:2.

In 2014, Hardie and co-workers reported the palladium(II) bis-(NHC)-containing cyclotriveratrylene cage 174 as a receptor for molecular iodine and dichlorobenzene.246 Importantly, 174 is, in comparison to analogous Pd3(en3) (en ¼ ethylenediamine) cages, which undergo rearrangements and transformations upon heating in DMSO, significantly more stable, highlighting the potency of the bis-(NHC) ligand as a protective “cap.” Crystals of this compound do not possess inherent porosity in the form of channels or large scale voids. Nevertheless, the large interior cavity of the cage of 697 A˚ 3, filled in the solid state by nitromethane molecules, can be utilized to adsorb I2 and 1,2-dichlorobenzene. The latter was incorporated into the cage by suspension of the crystals in the solvent, revealing an uptake of approximately three 1,2-dichlorobenzene molecules. This uptake occurs in a single crystal-to-single crystal fashion by displacement of the nitromethane molecules. Similarly, three molecules of I2 can be adsorbed into the crystal from solution. This can also be achieved by exposure to iodine vapors, however in this case only one molecule is taken up. The cage-bound iodine is significantly sequestered and remains within the crystal after washing with methanol, diethyl ether or if heated to 80  C under vacuum for 30 min.

Organometallic Receptors for Charged and Neutral Guest Species

455

More recently, Altmann and Pöthig developed a new class of macrocyclic hosts, which due their structural similarity with pilar[n] arenes were termed “pillarplexes”.247 Consisting of octanuclear Au(I) or Ag(I) NHC complexes, 175 and 176 form well-defined tubular cavities for linear alkane guests, exemplified by binding of 1,8-diaminooctane. The Au(I) analog 176 was easily obtained from 175 via transmetallation with (SMe2)AuCl. While emissive in the solid state, luminescence of 176 is quenched in ACN. In this solvent, 1,8-diaminooctane is bound in a 1:1 host-guest stoichiometric fashion with K ¼ 41,700 M−1. Crucially, anion exchange from the PF−6 salt to the OAc− salt endowed water solubility, in which guest binding was even stronger (K ¼ 84,000 M−1). Isothermal calorimetry studies revealed that this binding was enthalpically driven. Interestingly, in this solvent the receptor displayed emission, which upon binding of 1,8-diaminooctane was quenched. Of further note is that the pillarplexes display an unusual selectivity for this linear guest in comparison to 1,10 -dioctadecyl viologen bromide or p-phenylenediamine (no binding), both of which strongly bind to the similarly sized pillar[5]arene or cucurbit[6]uril. This can be attributed to a comparably longer, but narrower binding cavity. The same authors demonstrated that the strong binding of the longer 1,12-diaminododecane to 175 enables facile access to a [2] rotaxane via amide-bond formation stoppering in near quantitative yield.248 Not only can this interlocked structure be transmetallated to the corresponding Au(I) pillarplex rotaxane but can also be reversibly converted to a [3]rotaxane by acid-promoted demetallation.

456

Organometallic Receptors for Charged and Neutral Guest Species

The formation of metal-NHC complexes can also be employed for the templation of large molecules as demonstrated for the assembly of a large [16]-imidazolium cage via templating Ag(I)(NHC)2 formation.249

14.12.5.4 Receptors based on alkynyl platinum and gold motifs The incorporation of alkynyl metal motifs into receptors for neutral guests has been studied by Mukherjee and co-workers. The trigonal prism 177, assembled from a tri-topic alkynyl-Pt(II) panel and bis-pyridine clips, displayed fluorescence emission quenching upon binding of picric acid and TNT in DCM/DMF 4:1, with, K ¼ 48,100 and 19,600 M−1, respectively.250 Interestingly, the former induced quenching of the higher energy emission band at 413 nm, while TNT quenched the fluorescence at 546 nm.

In a subsequent study, a range of closely related metallo prisms were shown to also sense TNT in DMF via fluorescence modulation.251 Fluorescence sensing of nitroaromatics was also achieved with simple alkynyl palladium(II) molecular rectangles, a rare example of the use of this motif in organometallic receptors.252 The group of Yam have developed a series of alkynylplatinum(II) terpyridine receptors as luminescent sensors,253–256 such as the acyclic tweezer 178, and conducted detailed binding and sensing studies with various Pt(II) and Au(III) complexes, including 179–181.253 The addition of 179 to the receptor in DCM induced a naked-eye color change from yellow to red, concomitant with a decrease of the original luminescence emission band and an appearance of a new, lower energy emission. The 1:1 host-guest stoichiometric guest binding with K ¼ 77,600 M−1 proceeded via a “sandwich” binding mode driven by p-p and Pt-Pt interactions. Unsurprisingly, the anionic analog of this guest, 180, was bound with a higher affinity of K ¼ 1.47  108 M−1 and induced similar photophysical perturbations, while monocationic Pt(II) guest complexes elicited minor changes only in high excess. Related Pd(II), Au(I) and Au(III) complexes were also shown to bind to this host. For example, Au(III) complex 181 elicited fluorescence quenching, with K ¼ 18,600 M−1. This indicated weaker Pt(II)-Au(III) interactions compared with Pt(II)-Pt(II) interactions between 178 and 179. This was also observed in a comparison between related Pt(II) and Pd(II) complexes.

Organometallic Receptors for Charged and Neutral Guest Species

457

The same receptive scaffold also recognizes a range of PAHs254 and has later been employed by Wang and co-workers as a luminescent sensor for naphthol derivatives.257 In 2013, Yam’s group developed double-decker analogs of this receptor containing two discrete, stacked binding sites.255 They further reported a series of macrocyclic derivatives, including the rigid rectangle 182.256 In ACN, this receptor bound the guests 179 and 181 with K ¼ 209,000 and 35,500 M−1, respectively, significantly stronger than the acyclic analog 178. The binding strength of 179 towards a phenyl-extended, longer rigid box was enhanced by a factor of 10, while the flexible cage analog containing triethylene glycol linkers instead of the rigid pyridine framework displayed an order of magnitude weaker binding. As a result of the pyridine backbone, cage 182 displayed acid-switchable recognition properties. Upon addition of HCl the 179 guest was completely expelled from the binding cavity, a process that was switchable over multiple cycles by addition of HCl/triethylamine.

Platinum(II) based receptors for PAHs were also developed by Peris and coworkers, who combined polyaromatic NHC ligands with alkynyl platinum and alkynyl gold motifs. For instance, the acyclic Pt(II) tweezer 183, containing cis-oriented NHC ligands, was used for the recognition of the electron-deficient aromatics 2,4,7-trinitro-9-fluorenone (TNFLU), and 1,4,5,8-naphtalenetetracarboxylic dianhydride (NTCDA).258 In DMSO, no binding of pyrene was observed, while both NTCDA and TNFLU were bound weakly with K ¼ 10 and 22 M−1, respectively. In CHCl3 binding of TNFLU was somewhat stronger with K ¼ 47 M−1. Albeit weak, binding of these guests to 183 was notably enhanced in comparison to a receptor containing shorter pyrene-imidazole ligands.

458

Organometallic Receptors for Charged and Neutral Guest Species

The tetragold(I) organometallic rectangle 184 was shown to be a much more potent receptor for PAHs.259 In CD2Cl2 various guests were bound via formation of 1:1 host-guest stoichiometric complexes, in the following: naphthalene < anthracene < pyrene < triphenylene < perylene. While binding of naphthalene was weak (25 M−1), perylene was bound 6400-times more strongly (K ¼ 160,000 M−1). The binding constant for coronene was even larger, but could not be accurately quantified. As a result of inefficient p-stacking, the bowl-shaped corannulene was bound much more weakly (K ¼ 213 M−1). In general these binding strengths are significantly larger than that of an acyclic receptor analog developed earlier.260 The presence of additional OH-groups in these PAH guests was speculated to further enhance binding via hydrogen bonding interactions with the carbazole. However, neither 2-naphthol nor 1-pyrenemethanol displayed enhanced binding in comparison to their parent analogs, attributed to their relatively small size, incapable of spanning both the central p-binding and carbazole binding sites.261 Only the larger 3-perylenemethanol displayed significantly enhanced binding as confirmed by qualitative displacement studies. Interestingly, even 3-methylperylene displayed enhanced binding, albeit not as strongly as that of the methanol derivative. Peris and co-workers also developed the trigonal-prismatic analog of this receptor and demonstrated an one-order of magnitude stronger binding of coronene over perylene in CD2Cl2.262

A variety of other organometallic receptors have been developed and employed for the recognition of solvents263,264 or the binding and sensing of gasses.265,266

14.12.6 Conclusions and outlook As highlighted herein, organometallic receptors are ubiquitously employed for the recognition and sensing of a vast range of molecular species. The recognition of individual ions by organometallic receptors is now well-established and has matured to the extent that the use of these receptors in various applications, in particular sensors, is feasible. To this end, we believe future efforts will focus on further improving guest binding selectivity, especially in complex aqueous media, as well as the integration with specific device formats, for example via surface-confinement.

Organometallic Receptors for Charged and Neutral Guest Species

459

In contrast, ion-pair recognition at organometallic receptors remains comparably underdeveloped, undoubtedly as a result of the complexity of these systems. Nevertheless, significant advances have been made in gaining a more fundamental understanding of the processes that underpin ion-pair recognition as well as signal transduction in electrochemical voltammetric organometallic ion-pair sensors. The development of organometallic receptors for neutral guests has only gained significant attention over the past two decades and is recently evolving at a rapid pace. The recognition of polycyclic aromatic hydrocarbons and fullerenes has hereby been the focal point and is typically achieved by the integration of large p-extended surfaces into macrocyclic receptors, wherein organometallic motifs often play crucial roles as structural elements. Specifically, coordination-driven self-assembly has emerged as a versatile and highly adaptable approach in the construction of large, but well-defined receptive scaffolds. This enables a surprisingly high degree of control in designing highly selective hosts for specific hydrocarbon guests, a development that will undoubtedly lead to applications seeking to separate or sequester such species.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55.

Beer, P. D.; Bayly, S. R. In Comprehensive Organometallic Chemistry III; Mingos, D. M. P., Crabtree, R. H., Eds.; Elsevier: Oxford, 2007; pp 465–498. Astruc, D. Eur. J. Inorg. Chem. 2017, 2017, 6–29. Sun, R.; Wang, L.; Yu, H.; Abdin, Z.-U.; Chen, Y.; Huang, J.; Tong, R. Organometallics 2014, 33, 4560–4573. Boulas, P. L.; Gómez-Kaifer, M.; Echegoyen, L. Angew. Chem. Int. Ed. 1998, 37, 216–247. Hein, R.; Beer, P. D.; Davis, J. J. Chem. Rev. 2020, 120, 1888–1935. Bruña, S.; Martínez-Montero, I.; González-Vadillo, A. M.; Martín-Fernández, C.; Montero-Campillo, M. M.; Mó, O.; Cuadrado, I. Macromolecules 2015, 48, 6955–6969. Liu, Y.; Hu, J.; Teng, Q.; Zhang, H. Sens. Actuators B 2017, 238, 166–174. Jeon, H.; Suh, W.; Noh, D.-Y. Inorg. Chem. Commun. 2012, 24, 181–185. Huang, Y.; Lin, J.; Zhang, W.; Zhang, B.; Sun, Q. Inorg. Chem. Commun. 2019, 105, 129–134. Alfonso, M.; Tárraga, A.; Molina, P. Dalton Trans. 2010, 39, 8637–8645. Alfonso, M.; Espinosa Ferao, A.; Tárraga, A.; Molina, P. Inorg. Chem. 2015, 54, 7461–7473. Shi, L.; Song, W.; Li, Y.; Li, D.-W.; Swanick, K. N.; Ding, Z.; Long, Y.-T. Talanta 2011, 84, 900–904. Barik, S. K.; Borthakur, R.; Thakur, A.; Garai, B.; Jana, S.; Ghosh, S. RSC Adv. 2015, 5, 15690–15694. Xu, J.; Yang, Y.; Baigude, H.; Zhao, H. Spectrochim. Acta A 2020, 229, 117880. Hua, Y.; Flood, A. H. Chem. Soc. Rev. 2010, 39, 1262–1271. Romero, T.; Orenes, R. A.; Tárraga, A.; Molina, P. Organometallics 2013, 32, 5740–5753. Thakur, A.; Sardar, S.; Ghosh, S. Inorg. Chem. 2011, 50, 7066–7073. Thakur, A.; Ghosh, S. Organometallics 2012, 31, 819–826. Bhatta, S. R.; Bheemireddy, V.; Vijaykumar, G.; Debnath, S.; Thakur, A. Organometallics 2017, 36, 2141–2152. Zhao, B.-T.; Liu, L.-W.; Li, X.-C.; Qu, G.-R.; Belhadj, E.; Le Derf, F.; Sallé, M. Tetrahedron Lett. 2013, 54, 23–26. Thakur, A.; Mandal, D.; Ghosh, S. J. Organomet. Chem. 2013, 726, 71–78. Mandal, D.; Deb, P.; Mondal, B.; Thakur, A.; Ponniah, S. J.; Ghosh, S. RSC Adv. 2013, 3, 18614–18625. Mandal, D.; Thakur, A.; Ghosh, S. Polyhedron 2013, 52, 1109–1117. Gasser, G.; Carr, J. D.; Coles, S. J.; Green, S. J.; Hursthouse, M. B.; Cafferkey, S. M.; Stoeckli-Evans, H.; Tucker, J. H. R. J. Organomet. Chem. 2010, 695, 249–255. Zhou, Y.; Cao, Y.; Gong, G.; Zhang, Y.; Zhao, H.; Gao, X.; Zhao, G. Inorg. Chem. Commun. 2018, 96, 170–174. Caballero, A.; Espinosa, A.; Tárraga, A.; Molina, P. J. Org. Chem. 2008, 73, 5489–5497. Otón, F.; Ratera, I.; Espinosa, A.; Wurtz, K.; Parella, T.; Tárraga, A.; Veciana, J.; Molina, P. Chem. A Eur. J. 2010, 16, 1532–1542. Otón, F.; Espinosa, A.; Tárraga, A.; Molina, P. Organometallics 2007, 26, 6234–6242. Ruiz-Botella, S.; Vidossich, P.; Ujaque, G.; Vicent, C.; Peris, E. Chem. A Eur. J. 2015, 21, 10558–10565. Kasprzak, A.; Sakurai, H. Dalton Trans. 2019, 48, 17147–17152. Kasprzak, A.; Kowalczyk, A.; Jagielska, A.; Wagner, B.; Nowicka, A. M.; Sakurai, H. Dalton Trans. 2020, 49, 9965–9971. Evans, N. H.; Rahman, H.; Davis, J. J.; Beer, P. D. Anal. Bioanal. Chem. 2012, 402, 1739–1748. Pal, A.; Ranjan Bhatta, S.; Thakur, A. Coord. Chem. Rev. 2020, 213685. Omae, I. J. Organomet. Chem. 2016, 823, 50–75. Lo, K. K.-W.; Li, S. P.-Y.; Zhang, K. Y. New J. Chem. 2011, 35, 265–287. Ma, D.-L.; Wong, S.-Y.; Kang, T.-S.; Ng, H.-P.; Han, Q.-B.; Leung, C.-H. Methods 2019, 168, 3–17. Ma, D.-L.; Lin, S.; Wang, W.; Yang, C.; Leung, C.-H. Chem. Sci. 2017, 8, 878–889. Rota Martir, D.; Zysman-Colman, E. Coord. Chem. Rev. 2018, 364, 86–117. Caporale, C.; Massi, M. Coord. Chem. Rev. 2018, 363, 71–91. Zhao, N.; Wu, Y.-H.; Wen, H.-M.; Zhang, X.; Chen, Z.-N. Organometallics 2009, 28, 5603–5611. Schmittel, M.; Lin, H. Inorg. Chem. 2007, 46, 9139–9145. Lin, H.; Cinar, M. E.; Schmittel, M. Dalton Trans. 2010, 39, 5130–5138. Li, Y.; Yoon, U.-C.; Hyun, M.-H. Bull. Kor. Chem. Soc. 2011, 32, 122–126. Smith, J. B.; Camp, A. M.; Farquhar, A. H.; Kerr, S. H.; Chen, C.-H.; Miller, A. J. M. Organometallics 2019, 38, 4392–4398. Smith, J. B.; Kerr, S. H.; White, P. S.; Miller, A. J. M. Organometallics 2017, 36, 3094–3103. Camp, A. M.; Kita, M. R.; Blackburn, P. T.; Dodge, H. M.; Chen, C.-H.; Miller, A. J. M. J. Am. Chem. Soc. 2021, 143, 2792–2800. Zhao, Q.; Liu, S.; Li, F.; Yi, T.; Huang, C. Dalton Trans. 2008, 3836–3840. Zhao, Q.; Cao, T.; Li, F.; Li, X.; Jing, H.; Yi, T.; Huang, C. Organometallics 2007, 26, 2077–2081. Yan, F.; Mei, Q.; Wang, L.; Tong, B.; Xu, Z.; Weng, J.; Wang, L.; Huang, W. Inorg. Chem. Commun. 2012, 22, 178–181. Mei, Q.; Hua, Q.; Tong, B.; Shi, Y.; Chen, C.; Huang, W. Tetrahedron 2015, 71, 9366–9370. Yang, Z.; Huang, C.; Wang, C.; Zhao, Y.; Song, Q. ACS Sustain. Chem. Eng. 2017, 5, 4443–4448. Cao, H.-T.; Ding, L.; Shan, G.-G.; Sun, H.-Z.; Wu, Y.; Su, Z.-M. Dalton Trans. 2015, 44, 19997–20003. Ho, M.-L.; Cheng, Y.-M.; Wu, L.-C.; Chou, P.-T.; Lee, G.-H.; Hsu, F.-C.; Chi, Y. Polyhedron 2007, 26, 4886–4892. Tong, B.; Qiang, J.-Y.; Xu, Y.-Q.; Mei, Q.; Chen, Q.; Leung, W.-H.; Zhang, Q.-F. J. Solution Chem. 2012, 41, 1600–1609. Kim, H.-B.; Li, Y.; Hyun, M. H. Bull. Kor. Chem. Soc. 2013, 34, 653–656.

460 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126.

Organometallic Receptors for Charged and Neutral Guest Species Araya, J. C.; Gajardo, J.; Moya, S. A.; Aguirre, P.; Toupet, L.; Williams, J. A. G.; Escadeillas, M.; Le Bozec, H.; Guerchais, V. New J. Chem. 2010, 34, 21–24. Lee, P.-K.; Law, W. H.-T.; Liu, H.-W.; Lo, K. K.-W. Inorg. Chem. 2011, 50, 8570–8579. You, Y.; Han, Y.; Lee, Y.-M.; Park, S. Y.; Nam, W.; Lippard, S. J. J. Am. Chem. Soc. 2011, 133, 11488–11491. You, Y.; Lee, S.; Kim, T.; Ohkubo, K.; Chae, W.-S.; Fukuzumi, S.; Jhon, G.-J.; Nam, W.; Lippard, S. J. J. Am. Chem. Soc. 2011, 133, 18328–18342. Woo, H.; Cho, S.; Han, Y.; Chae, W.-S.; Ahn, D.-R.; You, Y.; Nam, W. J. Am. Chem. Soc. 2013, 135, 4771–4787. You, Y.; Cho, S.; Nam, W. Inorg. Chem. 2014, 53, 1804–1815. Shen, W.; Qi, Z.; Yan, L.; Tian, W.; Cui, X.; Yao, H.; Sun, Y. RSC Adv. 2016, 6, 16482–16488. Yu, T.; Wang, Y.; Zhu, Z.; Li, Y.; Zhao, Y.; Liu, X.; Zhang, H. Dyes Pigments 2019, 161, 252–260. Yang, H.; Zhu, Y.; Li, L.; Zhou, Z.; Yang, S. Inorg. Chem. Commun. 2012, 16, 1–3. Ma, D.-L.; He, H.-Z.; Chan, D. S.-H.; Wong, C.-Y.; Leung, C.-H. PLoS One 2014, 9, e99930. Wang, M.; Leung, K.-H.; Lin, S.; Chan, D. S.-H.; Kwong, D. W.; Leung, C.-H.; Ma, D.-L. Sci. Rep. 2014, 4, 1–7. Wang, W.; Mao, Z.; Wang, M.; Liu, L.-J.; Kwong, D. W.; Leung, C.-H.; Ma, D.-L. Chem. Commun. 2016, 52, 3611–3614. He, X.; Yam, V. W.-W. Coord. Chem. Rev. 2011, 255, 2111–2123. He, X.; Cheng, E. C.-C.; Zhu, N.; Yam, V. W.-W. Chem. Commun. 2009, 4016–4018. He, X.; Lam, W. H.; Zhu, N.; Yam, V. W.-W. Chem. A Eur. J. 2009, 15, 8842–8851. He, X.; Zhu, N.; Yam, V. W.-W. Dalton Trans. 2011, 40, 9703–9710. He, X.; Zhu, N.; Yam, V. W.-W. Organometallics 2009, 28, 3621–3624. Ibáñez, S.; Poyatos, M.; Peris, E. Angew. Chem. Int. Ed. 2017, 56, 9786–9790. Zhang, L.-K.; Xing, L.-B.; Chen, B.; Yang, Q.-Z.; Tong, Q.-X.; Wu, L.-Z.; Tung, C.-H. Dalton Trans. 2013, 42, 4244–4247. Lanoë, P.-H.; Fillaut, J.-L.; Guerchais, V.; Le Bozec, H.; Williams, J. A. G. Eur. J. Inorg. Chem. 2011, 2011, 1255–1259. Bi, D.; Feng, Y.; Zhao, Q.; Wang, H.; Zhu, Y.; Bao, X.; Fan, H.; Yu, L.; Yang, Q.; Qiu, D. RSC Adv. 2017, 7, 46980–46988. Xu, X.-D.; Yang, H.-B.; Zheng, Y.-R.; Ghosh, K.; Lyndon, M. M.; Muddiman, D. C.; Stang, P. J. J. Org. Chem. 2010, 75, 7373–7380. Qiu, Y.; Feng, Y.; Zhao, Q.; Wang, H.; Guo, Y.; Qiu, D. Dalton Trans. 2020, 49, 11163–11169. Qiu, D.; Li, M.; Zhao, Q.; Wang, H.; Yang, C. Inorg. Chem. 2015, 54, 7774–7782. Ghosh, S.; Chakrabarty, R.; Mukherjee, P. S. Inorg. Chem. 2009, 48, 549–556. Ghosh, K.; Hu, J.; Yang, H.-B.; Northrop, B. H.; White, H. S.; Stang, P. J. J. Org. Chem. 2009, 74, 4828–4833. Godoy, F.; Maldonado, T.; Flores, E.; Agurto, N.; González, R.; Ferraudi, G.; Lappin, G. Appl. Organomet. Chem. 2018, 32, e4115. Delavaux-Nicot, B.; Maynadié, J.; Lavabre, D.; Fery-Forgues, S. J. Organomet. Chem. 2007, 692, 3351–3362. Yam, V. W.-W.; Song, H.-O.; Chan, S. T.-W.; Zhu, N.; Tao, C.-H.; Wong, K. M.-C.; Wu, L.-X. J. Phys. Chem. C 2009, 113, 11674–11682. Li, M.-J.; Kwok, W.-M.; Lam, W. H.; Tao, C.-H.; Yam, V. W.-W.; Phillips, D. L. Organometallics 2009, 28, 1620–1630. Fillaut, J.-L.; Andries, J.; Marwaha, R. D.; Lanoë, P.-H.; Lohio, O.; Toupet, L.; Williams, J. G. J. Organomet. Chem. 2008, 693, 228–234. Burns, K. T.; Marks, W. R.; Cheung, P. M.; Seda, T.; Zakharov, L. N.; Gilbertson, J. D. Inorg. Chem. 2018, 57, 9601–9610. Gao, J.; Rochat, S.; Qian, X.; Severin, K. Chem. A Eur. J. 2010, 16, 5013–5017. Katsuta, S.; Iwabe, Y.; Kato, Y.; Kudo, Y.; Takeda, Y. Inorg. Chim. Acta 2008, 361, 103–108. Delgado, M.; Ziegler, J. M.; Seda, T.; Zakharov, L. N.; Gilbertson, J. D. Inorg. Chem. 2016, 55, 555–557. Mon, I.; Jose, D. A.; Vidal-Ferran, A. Chem. A Eur. J. 2013, 19, 2720–2725. Vidal-Ferran, A.; Mon, I.; Bauzá, A.; Frontera, A.; Rovira, L. Chem. A Eur. J. 2015, 21, 11417–11426. Rovira, L.; Vaquero, M.; Vidal-Ferran, A. J. Org. Chem. 2015, 80, 10397–10403. Owens, S. B., Jr.; Gray, G. M. Organometallics 2008, 27, 4282–4287. Martin, J. R.; Cagle, E. C.; Lucius, A. L.; Gray, G. M. Organometallics 2016, 35, 2609–2620. Huang, X.; Wu, B.; Jia, C.; Hay, B. P.; Li, M.; Yang, X. J. Chem. A Eur. J. 2013, 19, 9034–9041. Li, C.; Wang, L.; Deng, L.; Yu, H.; Huo, J.; Ma, L.; Wang, J. J. Phys. Chem. B 2009, 113, 15141–15144. Guo, D.-S.; Liu, Z.-P.; Ma, J.-P.; Huang, R.-Q. Tetrahedron Lett. 2007, 48, 1221–1224. Zapata, F.; Caballero, A.; Espinosa, A.; Tárraga, A.; Molina, P. J. Org. Chem. 2008, 73, 4034–4044. Alfonso, M.; Tárraga, A.; Molina, P. Inorg. Chem. 2013, 52, 7487–7496. Zapata, F.; Caballero, A.; Tárraga, A.; Molina, P. J. Org. Chem. 2010, 75, 162–169. Alfonso, M.; Tárraga, A.; Molina, P. Dalton Trans. 2016, 45, 19269–19276. Métay, E.; Duclos, M. C.; Pellet-Rostaing, S.; Lemaire, M.; Schulz, J.; Kannappan, R.; Bucher, C.; Saint-Aman, E.; Chaix, C. Eur. J. Org. Chem. 2008, 2008, 4304–4312. Otón, F.; Espinosa, A.; Tárraga, A.; Ratera, I.; Wurst, K.; Veciana, J.; Molina, P. Inorg. Chem. 2009, 48, 1566–1576. Métay, E.; Duclos, M. C.; Pellet-Rostaing, S.; Lemaire, M.; Schulz, J.; Kannappan, R.; Bucher, C.; Saint-Aman, E.; Chaix, C. Supramol. Chem. 2009, 21, 68–80. Huang, H.; Xin, Z.; Yuan, L.; Wang, B.-Y.; Cao, Q.-Y. Inorg. Chim. Acta 2018, 483, 425–430. Beeren, S. R.; Sanders, J. K. M. J. Am. Chem. Soc. 2011, 133, 3804–3807. Nieto, D.; González-Vadillo, A. M.; Bruña, S.; Pastor, C. J.; Kaifer, A. E.; Cuadrado, I. Chem. Commun. 2011, 47, 10398–10400. Otón, F.; Espinosa, A.; Tárraga, A.; de Arellano, C. R.; Molina, P. Chem. A Eur. J. 2007, 13, 5742–5752. Sola, A.; Orenes, R. A.; García, M.Á.; Claramunt, R. M.; Alkorta, I.; Elguero, J.; Tárraga, A.; Molina, P. Inorg. Chem. 2011, 50, 4212–4220. Beck, C. L.; Berg, S. A.; Winter, A. H. Org. Biomol. Chem. 2013, 11, 5827–5835. Beck, C. L.; Winter, A. H. J. Org. Chem. 2014, 79, 3152–3158. Langton, M. J.; Beer, P. D. Acc. Chem. Res. 2014, 47, 1935–1949. Ba˛ k, K. M.; Porfyrakis, K.; Davis, J. J.; Beer, P. D. Mater. Chem. Front. 2020, 4, 1052–1073. Evans, N. H.; Beer, P. D. Org. Biomol. Chem. 2011, 9, 92–100. Evans, N. H.; Rahman, H.; Leontiev, A. V.; Greenham, N. D.; Orlowski, G. A.; Zeng, Q.; Jacobs, R. M. J.; Serpell, C. J.; Kilah, N. L.; Davis, J. J.; Beer, P. D. Chem. Sci. 2012, 3, 1080–1089. Evans, N. H.; Serpell, C. J.; White, N. G.; Beer, P. D. Chem. A Eur. J. 2011, 17, 12347–12354. Evans, N. H.; Serpell, C. J.; Beer, P. D. Chem. Commun. 2011, 47, 8775–8777. Barendt, T. A.; Rašovic, I.; Lebedeva, M. A.; Farrow, G. A.; Auty, A.; Chekulaev, D.; Sazanovich, I. V.; Weinstein, J. A.; Porfyrakis, K.; Beer, P. D. J. Am. Chem. Soc. 2018, 140, 1924–1936. Lim, J. Y. C.; Cunningham, M. J.; Davis, J. J.; Beer, P. D. Dalton Trans. 2014, 43, 17274–17282. Bayly, S. R.; Gray, T. M.; Chmielewski, M. J.; Davis, J. J.; Beer, P. D. Chem. Commun. 2007, 2234–2236. Cormode, D. P.; Evans, A. J.; Davis, J. J.; Beer, P. D. Dalton Trans. 2010, 39, 6532–6541. Martínez, F. J.; González, B.; Alonso, B.; Losada, J.; García-Armada, M. P.; Casado, C. M. J. Inorg. Organomet. Polym. Mater. 2008, 18, 51–58. Astruc, D.; Boisselier, E.; Ornelas, C. Chem. Rev. 2010, 110, 1857–1959. Astruc, D.; Ruiz, J. J. Inorg. Organomet. Polym. Mater. 2015, 25, 2–11. Astruc, D.; Deraedt, C.; Djeda, R.; Ornelas, C.; Liu, X.; Rapakousiou, A.; Ruiz, J.; Wang, Y.; Wang, Q. Molecules 2018, 23, 966.

Organometallic Receptors for Charged and Neutral Guest Species 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. 180. 181. 182. 183. 184. 185. 186. 187. 188. 189. 190. 191. 192. 193. 194. 195. 196. 197. 198. 199.

461

Ornelas, C.; Ruiz, J.; Belin, C.; Astruc, D. J. Am. Chem. Soc. 2008, 131, 590–601. Ornelas, C.; Ruiz, J.; Astruc, D. Organometallics 2009, 28, 4431–4437. Villoslada, R.; Alonso, B.; Casado, C. M.; García-Armada, P.; Losada, J. Organometallics 2009, 28, 727–733. Weng, T.-Q.; Huang, Y.-F.; Xue, L.-S.; Cheng, J.; Jin, S.; Liu, S.-H.; Wu, D.-Y.; Chen, G. Z. Molecules 2019, 24, 238. Kong, D.; Weng, T.; He, W.; Liu, B.; Jin, S.; Hao, X.; Liu, S. J. Organomet. Chem. 2013, 727, 19–27. Zhuo, J.-B.; Lin, C.-X.; Wan, Q.; Xie, L.-L.; Yuan, Y.-F. J. Organomet. Chem. 2015, 791, 289–297. Zhuo, J.-B.; Zhu, X.-X.; Lin, C.-X.; Bai, S.; Xie, L.-L.; Yuan, Y.-F. J. Organomet. Chem. 2014, 770, 85–93. Zhuo, J.-B.; Zhang, C.-Y.; Lin, C.-X.; Bai, S.; Xie, L.-L.; Yuan, Y.-F. J. Organomet. Chem. 2014, 763, 34–43. Caballero, A.; White, N. G.; Beer, P. D. CrstEngComm 2014, 16, 3694–3698. Yao, L.-Y.; Qin, L.; Xie, T.-Z.; Li, Y.-Z.; Yu, S.-Y. Inorg. Chem. 2011, 50, 6055–6062. Cao, Q.-Y.; Pradhan, T.; Kim, S.; Kim, J. S. Org. Lett. 2011, 13, 4386–4389. Cao, Q.-Y.; Pradhan, T.; Lee, M. H.; Choi, D. H.; Kim, J. S. Tetrahedron Lett. 2012, 53, 4917–4920. Cao, Q.-Y.; Han, Y.-M.; Yao, P.-S.; Fu, W.-F.; Xie, Y.; Liu, J.-H. Tetrahedron Lett. 2014, 55, 248–251. White, N. G.; Beer, P. D. Beilstein J. Org. Chem. 2012, 8, 246. Li, C.-T.; Cao, Q.-Y.; Li, J.-J.; Wang, Z.-W.; Dai, B.-N. Inorg. Chim. Acta 2016, 449, 31–37. Astruc, D.; Liang, L.; Rapakousiou, A.; Ruiz, J. Acc. Chem. Res. 2012, 45, 630–640. Ornelas, C.; Ruiz Aranzaes, J.; Cloutet, E.; Alves, S.; Astruc, D. Angew. Chem. Int. Ed. 2007, 46, 872–877. Rapakousiou, A.; Wang, Y.; Belin, C.; Pinaud, N.; Ruiz, J.; Astruc, D. Inorg. Chem. 2013, 52, 6685–6693. Camponovo, J.; Ruiz, J.; Cloutet, E.; Astruc, D. Chem. A Eur. J. 2009, 15, 2990–3002. Djeda, R.; Rapakousiou, A.; Liang, L.; Guidolin, N.; Ruiz, J.; Astruc, D. Angew. Chem. Int. Ed. 2010, 49, 8152–8156. Villena, C.; Losada, J.; García-Armada, P.; Casado, C. M.; Alonso, B. Organometallics 2012, 31, 3284–3291. Leroux, Y. R.; Hui, F.; Ruiz, J.; Astruc, D.; Hapiot, P. Chem. A Eur. J. 2012, 18, 7041–7044. Brown, A.; Beer, P. D. Chem. Commun. 2016, 52, 8645–8658. Lim, J. Y. C.; Cunningham, M. J.; Davis, J. J.; Beer, P. D. Chem. Commun. 2015, 51, 14640–14643. Lim, J. Y. C.; Beer, P. D. Eur. J. Inorg. Chem. 2017, 2017, 220–224. Lim, J. Y. C.; Marques, I.; Ferreira, L.; Félix, V.; Beer, P. D. Chem. Commun. 2016, 52, 5527–5530. Willener, Y.; Joly, K. M.; Moody, C. J.; Tucker, J. H. R. J. Org. Chem. 2008, 73, 1225–1233. Mulas, A.; Willener, Y.; Carr-Smith, J.; Joly, K. M.; Male, L.; Moody, C. J.; Horswell, S. L.; Nguyen, H. V.; Tucker, J. H. R. Dalton Trans. 2015, 44, 7268–7275. Lim, J. Y. C.; Beer, P. D. Eur. J. Org. Chem. 2019, 2019, 3433–3441. Hein, R.; Li, X.; Beer, P. D.; Davis, J. J. Chem. Sci. 2021, 12, 2433–2440. Ion, L.; Nieto, S.; Pérez, J.; Riera, L.; Riera, V.; Díaz, J.; López, R.; Anderson, K. M.; Steed, J. W. Inorg. Chem. 2011, 50, 8524–8531. Pelleteret, D.; Fletcher, N. C.; Doherty, A. P. Inorg. Chem. 2007, 46, 4386–4388. Li, S.; Zhang, C.; Huang, S.; Hu, F.; Yin, J.; Liu, S. H. RSC Adv. 2012, 2, 4215–4219. Ahmad, H.; Hazel, B. W.; Meijer, A. J.; Thomas, J. A.; Wilkinson, K. A. Chem. A Eur. J. 2013, 19, 5081–5087. Mole, T. K.; Arter, W. E.; Marques, I.; Félix, V.; Beer, P. D. J. Organomet. Chem. 2015, 792, 206–210. Mullaney, B. R.; Thompson, A. L.; Beer, P. D. Angew. Chem. Int. Ed. 2014, 53, 11458–11462. Langton, M. J.; Xiong, Y.; Beer, P. D. Chem. A Eur. J. 2015, 21, 18910–18914. Ion, L.; Morales, D.; Nieto, S.; Pérez, J.; Riera, L.; Riera, V.; Miguel, D.; Kowenicki, R. A.; McPartlin, M. Inorg. Chem. 2007, 46, 2846–2853. Plajer, A. J.; Zhu, J.; Pröhm, P.; Rizzuto, F. J.; Keyser, U. F.; Wright, D. S. J. Am. Chem. Soc. 2020, 142, 1029–1037. Ruiz Aranzaes, J.; Belin, C.; Astruc, D. Chem. Commun. 2007, 3456–3458. Li, Y.; Liu, Y.; Nam, D.; Park, S.; Yoon, J.; Hyun, M. H. Dyes Pigments 2014, 100, 241–246. Hu, Y.; Li, Y.; Joung, J. F.; Yin, J.; Park, S.; Yoon, J.; Hyun, M. H. Sens. Actuators B 2017, 241, 224–229. Zhao, Q.; Liu, S.; Shi, M.; Li, F.; Jing, H.; Yi, T.; Huang, C. Organometallics 2007, 26, 5922–5930. Rommel, S. A.; Sorsche, D.; Dixit, A.; Rau, S. Eur. J. Inorg. Chem. 2016, 2016, 40–48. Rommel, S. A.; Sorsche, D.; Rau, S. Dalton Trans. 2016, 45, 74–77. Freys, J. C.; Bernardinelli, G.; Wenger, O. S. Chem. Commun. 2008, 4267–4269. Kampes, R.; Tepper, R.; Görls, H.; Bellstedt, P.; Jäger, M.; Schubert, U. S. Chem. A Eur. J. 2020, 26, 14679–14687. Dorazco-Gonzalez, A. Organometallics 2014, 33, 868–875. Gong, Z.-L.; Zhong, Y.-W. Inorg. Chem. 2016, 55, 10143–10151. Shanmugaraju, S.; Bar, A. K.; Chi, K.-W.; Mukherjee, P. S. Organometallics 2010, 29, 2971–2980. Yeung, M. C.-L.; Chu, B. W.-K.; Yam, V. W.-W. ChemistryOpen 2014, 3, 172–176. Hau, F. K.-W.; Lo, H.-S.; Yam, V. W.-W. Chem. A Eur. J. 2016, 22, 3738–3749. Baron, M.; Dall’Anese, A.; Tubaro, C.; Orian, L.; Di Marco, V.; Bogialli, S.; Graiff, C.; Basato, M. Dalton Trans. 2018, 47, 935–945. Baron, M.; Dall’Anese, A.; Miolato, A.; Cairoli, M. L. C.; Di Marco, V.; Graiff, C.; Pöthig, A.; Tubaro, C. New J. Chem. 2020, 44, 5343–5353. Altmann, P. J.; Pöthig, A. Chem. Commun. 2016, 52, 9089–9092. Bedford, R. B.; Betham, M.; Butts, C. P.; Coles, S. J.; Hursthouse, M. B.; Noelle Scully, P.; Tucker, J. H. R.; Wilkie, J.; Willener, Y. Chem. Commun. 2008, 2429–2431. Jiang, B.; Wang, W.; Zhang, Y.; Lu, Y.; Zhang, C.-W.; Yin, G.-Q.; Zhao, X.-L.; Xu, L.; Tan, H.; Li, X.; Jin, G.-X.; Yang, H.-B. Angew. Chem. Int. Ed. 2017, 56, 14438–14442. Zeng, H.; Liu, P.; Feng, G.; Huang, F. J. Am. Chem. Soc. 2019, 141, 16501–16511. Dickson, S. J.; Paterson, M. J.; Willans, C. E.; Anderson, K. M.; Steed, J. W. Chem. A Eur. J. 2008, 14, 7296–7305. Kim, S. K.; Sessler, J. L. Chem. Soc. Rev. 2010, 39, 3784–3809. McConnell, A. J.; Beer, P. D. Angew. Chem. Int. Ed. 2012, 51, 5052–5061. McConnell, A. J.; Docker, A.; Beer, P. D. ChemPlusChem 2020, 85, 1824–1841. Zeng, Z.; Torriero, A. A.; Bond, A. M.; Spiccia, L. Chem. A Eur. J. 2010, 16, 9154–9163. Wong, Y.-S.; Ng, M.; Yeung, M. C.-L.; Yam, V. W.-W. J. Am. Chem. Soc. 2021, 143, 973–982. Alfonso, M.; Espinosa, A.; Tárraga, A.; Molina, P. Org. Lett. 2011, 13, 2078–2081. Alfonso, M.; Espinosa, A.; Tárraga, A.; Molina, P. Chem. Commun. 2012, 48, 6848–6850. Wan, Q.; Zhuo, J.-B.; Wang, X.-X.; Lin, C.-X.; Yuan, Y.-F. Dalton Trans. 2015, 44, 5790–5796. Otón, F.; González, M. D. C.; Espinosa, A.; Ramírez de Arellano, C.; Tárraga, A.; Molina, P. J. Org. Chem. 2012, 77, 10083–10092. González, M. D. C.; Otón, F.; Espinosa, A.; Tárraga, A.; Molina, P. Chem. Commun. 2013, 49, 9633–9635. González, M. A. D. C.; Otón, F.; Orenes, R. A.; Espinosa, A.; Tárraga, A.; Molina, P. Organometallics 2014, 33, 2837–2852. Zaleskaya, M.; Jagleniec, D.; Karbarz, M.; Dobrzycki, Ł.; Romanski, J. Inorg. Chem. Front. 2020, 7, 972–983. Naseer, M. M.; Jurkschat, K. Chem. Commun. 2017, 53, 8122–8135. Reeske, G.; Schürmann, M.; Costisella, B.; Jurkschat, K. Organometallics 2007, 26, 4170–4179.

462 200. 201. 202. 203. 204. 205. 206. 207. 208. 209. 210. 211. 212. 213. 214. 215. 216. 217. 218. 219. 220. 221. 222. 223. 224. 225. 226. 227. 228. 229. 230. 231. 232. 233. 234. 235. 236. 237. 238. 239. 240. 241. 242. 243. 244. 245. 246. 247. 248. 249. 250. 251. 252. 253. 254. 255. 256. 257. 258. 259. 260. 261. 262. 263. 264. 265. 266.

Organometallic Receptors for Charged and Neutral Guest Species Reeske, G.; Bradtmöller, G.; Schürmann, M.; Jurkschat, K. Chem. A Eur. J. 2007, 13, 10239–10245. Reeske, G.; Schürmann, M.; Jurkschat, K. Dalton Trans. 2008, 3398–3400. Tagne Kuate, A. C.; Iovkova, L.; Hiller, W.; Schürmann, M.; Jurkschat, K. Organometallics 2010, 29, 5456–5471. Kuate, A. C. T.; Reeske, G.; Schürmann, M.; Costisella, B.; Jurkschat, K. Organometallics 2008, 21, 5577–5587. Kuate, A. C. T.; Naseer, M. M.; Jurkschat, K. Chem. Commun. 2017, 53, 2013–2015. Wendji, A. S.; Lutter, M.; Dietz, C.; Jouikov, V.; Jurkschat, K. Organometallics 2013, 32, 5720–5730. Arens, V.; Dietz, C.; Schollmeyer, D.; Jurkschat, K. Organometallics 2013, 32, 2775–2786. Thomas, J. A. Dalton Trans. 2011, 40, 12005–12016. Chan, A. K.-W.; Yam, V. W.-W. Acc. Chem. Res. 2018, 51, 3041–3051. Bivaud, S.; Goeb, S.; Croué, V.; Dron, P. I.; Allain, M.; Sallé, M. J. Am. Chem. Soc. 2013, 135, 10018–10021. Shanmugaraju, S.; Vajpayee, V.; Lee, S.; Chi, K.-W.; Stang, P. J.; Mukherjee, P. S. Inorg. Chem. 2012, 51, 4817–4823. Sánchez-Molina, I.; Grimm, B.; Krick Calderon, R. M.; Claessens, C. G.; Guldi, D. M.; Torres, T. J. Am. Chem. Soc. 2013, 135, 10503–10511. Kasprzak, A.; Gunka, P. A. Dalton Trans. 2020, 49, 6974–6979. Casas-Solvas, J. M.; Ortiz-Salmerón, E.; Fernández, I.; García-Fuentes, L.; Santoyo-González, F.; Vargas-Berenguel, A. Chem. A Eur. J. 2009, 15, 8146–8162. Munteanu, M.; Kolb, U.; Ritter, H. Macromol. Rapid Commun. 2010, 31, 616–618. Cedano, M. R.; Smith, D. K. J. Org. Chem. 2018, 83, 11595–11603. Barry, N. P. E.; Furrer, J.; Freudenreich, J.; Süss-Fink, G.; Therrien, B. Eur. J. Inorg. Chem. 2010, 2010, 725–728. Barry, N. P.; Furrer, J.; Therrien, B. Helv. Chim. Acta 2010, 93, 1313–1328. Han, Y.-F.; Fei, Y.; Jin, G.-X. Dalton Trans. 2010, 39, 3976–3984. Zhang, W.-Y.; Lin, Y.-J.; Han, Y.-F.; Jin, G.-X. J. Am. Chem. Soc. 2016, 138, 10700–10707. Shanmugaraju, S.; Bar, A. K.; Joshi, S. A.; Patil, Y. P.; Mukherjee, P. S. Organometallics 2011, 30, 1951–1960. Wang, M.; Vajpayee, V.; Shanmugaraju, S.; Zheng, Y.-R.; Zhao, Z.; Kim, H.; Mukherjee, P. S.; Chi, K.-W.; Stang, P. J. Inorg. Chem. 2011, 50, 1506–1512. Mishra, A.; Jung, H.; Lee, M. H.; Lah, M. S.; Chi, K.-W. Inorg. Chem. 2013, 52, 8573–8578. Therrien, B.; Süss-Fink, G.; Govindaswamy, P.; Renfrew, A. K.; Dyson, P. J. Angew. Chem. Int. Ed. 2008, 47, 3773–3776. Furrer, M. A.; Schmitt, F.; Wiederkehr, M.; Juillerat-Jeanneret, L.; Therrien, B. Dalton Trans. 2012, 41, 7201–7211. Yi, J. W.; Barry, N. P. E.; Furrer, M. A.; Zava, O.; Dyson, P. J.; Therrien, B.; Kim, B. H. Bioconjug. Chem. 2012, 23, 461–471. Suntharalingam, K.; Łe˛ czkowska, A.; Furrer, M. A.; Wu, Y.; Kuimova, M. K.; Therrien, B.; White, A. J. P.; Vilar, R. Chem. A Eur. J. 2012, 18, 16277–16282. Zava, O.; Mattsson, J.; Therrien, B.; Dyson, P. J. Chem. A Eur. J. 2010, 16, 1428–1431. Barry, N. P. E.; Zava, O.; Dyson, P. J.; Therrien, B. Chem. A Eur. J. 2011, 17, 9669–9677. Barry, N. P. E.; Zava, O.; Dyson, P. J.; Therrien, B. J. Organomet. Chem. 2012, 705, 1–6. Schmitt, F.; Freudenreich, J.; Barry, N. P. E.; Juillerat-Jeanneret, L.; Süss-Fink, G.; Therrien, B. J. Am. Chem. Soc. 2012, 134, 754–757. Freudenreich, J.; Dalvit, C.; Süss-Fink, G.; Therrien, B. Organometallics 2013, 32, 3018–3033. Freudenreich, J.; Furrer, J.; Süss-Fink, G.; Therrien, B. Organometallics 2011, 30, 942–951. Freudenreich, J.; Barry, N. P. E.; Süss-Fink, G.; Therrien, B. Eur. J. Inorg. Chem. 2010, 2010, 2400–2405. Han, Y.-F.; Li, H.; Zheng, Z.-F.; Jin, G.-X. Chem. Asian J. 2012, 7, 1243–1250. Yuan, M.; Weisser, F.; Sarkar, B.; Garci, A.; Braunstein, P.; Routaboul, L.; Therrien, B. Organometallics 2014, 33, 5043–5045. Zhang, Y.-Y.; Qiu, F.-Y.; Shi, H.-T.; Yu, W. Chem. Commun. 2021, 57, 3010–3013. Ibáñez, S.; Poyatos, M.; Peris, E. Acc. Chem. Res. 2020, 53, 1401–1413. Qin, D.; Zeng, X.; Li, Q.; Xu, F.; Song, H.; Zhang, Z.-Z. Chem. Commun. 2007, 147–149. Lin, C.-X.; Kong, X.-F.; Li, Q.-S.; Zhang, Z.-Z.; Yuan, Y.-F.; Xu, F.-B. CrstEngComm 2013, 15, 6948–6962. Liu, Q.-X.; Yao, Z.-Q.; Zhao, X.-J.; Chen, A.-H.; Yang, X.-Q.; Liu, S.-W.; Wang, X.-G. Organometallics 2011, 30, 3732–3739. Martínez-Agramunt, V.; Ruiz-Botella, S.; Peris, E. Chem. A Eur. J. 2017, 23, 6675–6681. Martínez-Agramunt, V.; Gusev, D. G.; Peris, E. Chem. A Eur. J. 2018, 24, 14802–14807. Martínez-Agramunt, V.; Eder, T.; Darmandeh, H.; Guisado-Barrios, G.; Peris, E. Angew. Chem. Int. Ed. 2019, 58, 5682–5686. Martínez-Agramunt, V.; Peris, E. Inorg. Chem. 2019, 58, 11836–11842. Martínez-Agramunt, V.; Peris, E. Chem. Commun. 2019, 55, 14972–14975. Henkelis, J. J.; Carruthers, C. J.; Chambers, S. E.; Clowes, R.; Cooper, A. I.; Fisher, J.; Hardie, M. J. J. Am. Chem. Soc. 2014, 136, 14393–14396. Altmann, P. J.; Pöthig, A. J. Am. Chem. Soc. 2016, 138, 13171–13174. Altmann, P. J.; Pöthig, A. Angew. Chem. Int. Ed. 2017, 56, 15733–15736. Hua, K.; An, Y.-Y.; Wang, Y.-Y.; Han, Y.-F. Chem. A Eur. J. 2020, 26, 7190–7193. Ghosh, S.; Mukherjee, P. S. Organometallics 2008, 27, 316–319. Ghosh, S.; Gole, B.; Bar, A. K.; Mukherjee, P. S. Organometallics 2009, 28, 4288–4296. Bar, A. K.; Gole, B.; Ghosh, S.; Mukherjee, P. S. Dalton Trans. 2009, 6701–6704. Tanaka, Y.; Wong, K. M.-C.; Yam, V. W.-W. Chem. Sci. 2012, 3, 1185–1191. Tanaka, Y.; Wong, K. M.-C.; Yam, V. W.-W. Chem. A Eur. J. 2013, 19, 390–399. Tanaka, Y.; Wong, K. M.-C.; Yam, V. W.-W. Angew. Chem. Int. Ed. 2013, 52, 14117–14120. Chan, A. K.-W.; Lam, W. H.; Tanaka, Y.; Wong, K. M.-C.; Yam, V. W.-W. Proc. Natl. Acad. Sci. 2015, 112, 690–695. Fu, T.; Han, Y.; Ao, L.; Wang, F. Organometallics 2016, 35, 2850–2853. Nuevo, D.; Gonell, S.; Poyatos, M.; Peris, E. Chem. A Eur. J. 2017, 23, 7272–7277. Ibáñez, S.; Peris, E. Angew. Chem. Int. Ed. 2020, 59, 6860–6865. Biz, C.; Ibáñez, S.; Poyatos, M.; Gusev, D.; Peris, E. Chem. A Eur. J. 2017, 23, 14439–14444. Ibáñez, S.; Gusev, D. G.; Peris, E. Organometallics 2020, 39, 4078–4084. Ibáñez, S.; Peris, E. Angew. Chem. Int. Ed. 2019, 58, 6693–6697. Han, Y.-F.; Jia, W.-G.; Lin, Y.-J.; Jin, G.-X. Angew. Chem. Int. Ed. 2009, 48, 6234–6238. Han, Y.-F.; Lin, Y.-J.; Jin, G.-X. Dalton Trans. 2011, 40, 10370–10375. Cid, C. C.; Jimenez-Cadena, G.; Riu, J.; Maroto, A.; Xavier Rius, F.; Batema, G. D.; van Koten, G. Sens. Actuators B Chem. 2009, 141, 97–103. Sun, L.-Y.; An, Y.-Y.; Ma, L.-L.; Han, Y.-F. Chin. J. Chem. 2019, 37, 763–768.

14.13

Surface Organometallic Chemistry and Catalysis

Walid Al Maksouda, Sandeep Mishraa,b, Aya Saidia, Manoja K Samantaraya, and Jean Marie Basseta, aKing Abdullah University of Science and Technology (KAUST), Kaust Catalysis Center (KCC), Thuwal, Saudi Arabia; bDepartment of Physics & NMR Research Center, Indian Institute of Science Education and Research, Pune, India © 2022 Elsevier Ltd. All rights reserved.

14.13.1 14.13.2 14.13.2.1 14.13.2.2 14.13.2.3 14.13.2.4 14.13.2.5 14.13.2.6 14.13.2.7 14.13.2.8 14.13.2.9 14.13.2.10 14.13.2.11 14.13.2.12 14.13.2.13 14.13.2.14 14.13.2.15 14.13.2.16 14.13.2.17 14.13.2.18 14.13.2.19 14.13.2.20 14.13.2.21 14.13.2.22 14.13.2.23 14.13.2.24 14.13.3 14.13.3.1 14.13.3.2 14.13.3.3 14.13.3.4 14.13.4 14.13.5 14.13.6 14.13.7 14.13.8 14.13.9 14.13.10 14.13.10.1 14.13.10.2 14.13.11 14.13.12 14.13.13 14.13.14 14.13.15 14.13.16 References

Introduction Solid-state NMR spectroscopy: An unique tool in SOMC characterization Nuclear magnetic resonance (NMR) spectroscopy Solution-state vs. solid-state NMR (ssNMR) The ssNMR and nuclei with quadrupolar magnetic moment Reintroduction of DD interaction in ssNMR and proximity information NMR sensitivity Hyperpolarization and DNP Concept of dynamic nuclear polarization (DNP) Experimental approaches of DNP and DNP-MAS Effective surface enhancements and DNP SENS DNP SENS ssNMR in SOMC: Challenges and solutions The ssNMR with and without DNP SENS in the characterization of SOMC 15 N ssNMR spectroscopy 17 O MAS NMR spectroscopy 31 P ssNMR spectroscopy 1D MAS and 2D HETCOR ssNMR experiments 1D 29Si CP-MAS and 2D 1Hd29Si HETCOR ssNMR in characterization of SOMC 2D DNP SENS 13Cd111Cd HETCOR spectra of CdSe QDs 27 Al ssNMR spectroscopy HETCOR between 1H and Quadrupolar nuclei HETCOR between a low abundant spin ½ and quadrupolar nuclei 2D INADEQUATE ssNMR spectroscopy The multiple quantum (MQ) NMR spectroscopy 2D 1Hd1H SQ-MQ correlation ssNMR experiments 195 Pt DNP SENS ssNMR Alkane metathesis Linear alkane metathesis with [W]dH and [Ta]dH (1st generation catalyst) Liquid phase alkane metathesis reaction Cycloalkane metathesis Understanding alkane metathesis with a bi-metallic catalyst Low temperature hydrogenolysis of alkanes Cross-metathesis of alkanes Imine metathesis Hydroamination reaction Hydrometathesis of olefins Catalytic reduction of N2 toward NH3 Light alkanes aromatization Introduction Aromatization by SOMC methodology (catalysis by design) Catalytic oxidation reaction by SOMC with O2: A new route to acetaldehyde Hydro peroxide decomposition Olefin epoxidation The Baeyer-Villiger reaction Catalytic CO2 conversion by SOMC Conclusions

Comprehensive Organometallic Chemistry IV

https://doi.org/10.1016/B978-0-12-820206-7.00133-5

463 464 464 464 465 466 466 467 467 468 468 468 469 470 470 470 470 473 475 475 475 476 476 478 479 481 481 482 487 488 488 490 492 495 496 497 499 506 506 508 512 515 515 518 519 527 527

463

464

Surface Organometallic Chemistry and Catalysis

14.13.1 Introduction Surface Organometallic Chemistry (SOMC) is a branch of chemistry that bridges the gap between homogeneous and heterogeneous catalysis.1 Catalysis is a molecular phenomenon, it must involve a well-defined catalyst/intermediate along with a reactant during a catalytic reaction. Homogeneous catalysis has progressed a lot in the last 50 years. However, heterogeneous catalysis, which is extremely important in the industry, has not progressed at the same pace. The primary reason for this discrepancy is the lack of understanding of the nature of the active sites.1 In the field of catalysis, it is only possible to target a reaction or to develop a new catalyst for boosting the productivity of a particular reaction if the catalyst is well-defined in nature. If a catalyst is well-defined, one can pre-determine catalysis and develop the catalyst accordingly by changing substituents around the metal center, including support. In the case of the classical heterogeneous catalyst, the so-called “active sites” are not well-defined, which makes it very difficult to pre-determine or target a particular catalytic reaction. Another difficulty in obtaining the structure-activity relationship in the heterogeneous catalyst is the small number of active sites. Apart from that, due to the unknown number or structure of the active sites, it is very difficult to develop new catalysts in this field though chemical industries are mainly dependent on heterogeneous catalysts. In other words, we need a catalyst that can behave like a heterogeneous catalyst (stable at comparable high temperature, recyclable, easily separable from reaction mixture, etc.) but also be well-defined like a homogeneous catalyst.2 In order to bridge the gap between the well-defined homogeneous catalyst and not so well-defined heterogeneous catalyst, new chemistry has emerged called Surface Organometallic Chemistry (SOMC). This SOMC strategy is aiming to bring the concepts of molecular chemistry especially organometallic chemistry to surface science and the heterogeneous catalyst. Surface organometallic technique is a beautiful technique where one can predict catalysis based on the surface organometallic fragment (SOMF) or surface coordination fragment (SCF) which is linked to the surface. It is called “predictive approach” or “catalysis by design”.3,4 Not only one can predict the catalysis, but it allows one to tune the SOMF or SCF based on the need to enhance the efficiency of a catalyst, just like in the case of homogeneous catalysis. Additionally, surface organometallic technique allows one to outline a catalytic cycle by isolating the reaction intermediate, which is very difficult in the case of homogeneous/ classical heterogeneous catalysis conditions. With all these advances of surface organometallic catalysis (SOMCcat), it projects itself as a good contender for industrial use in the near future. In this chapter, we are going to discuss the recent advances of surface organometallic catalyst (SOMCcat) in various catalytic reactions like alkane metathesis, imine metathesis, ammonia production, oxidation, hydrogenation, CO2 chemistry, to name a few. Alongside catalysis, we also provide here insight into the solid-state NMR technique, which is a major tool used for characterizing the SOMCcat precisely.

14.13.2 Solid-state NMR spectroscopy: An unique tool in SOMC characterization Solid-state nuclear magnetic resonance (ssNMR) spectroscopy5,6 is proven as a robust analytical technique in the characterization and investigation of heterogeneous catalysts, synthesized using surface organometallic chemistry (SOMC).7–12 NMR spectroscopy is a powerful tool in the characterization of surface organometallic fragments (SOMFs) grafted on silica nanoparticles (SiNPs).13,14 Still, the scope of ssNMR spectroscopy in the study of such kinds of systems is enormous. The upcoming sections of this chapter explore the utility and drawbacks of NMR spectroscopy, along with the possible alternative solutions for the existing problems. The discussion further extended to various ssNMR experiments in the characterization of SOMC catalysts where the variety of one dimensional (1D) and two dimensional (2D) experiments on different NMR active nuclei are being demonstrated.

14.13.2.1 Nuclear magnetic resonance (NMR) spectroscopy NMR spectroscopy assesses the resonance frequencies of magnetically sensitive nuclei in the presence of a huge external magnetic field (B0). NMR spectroscopy has proven to be the most valuable among existing spectroscopic techniques due to its versatility, non-destructive nature, and application in extracting both the structure and quantitative information with high reproducibility. Therefore, the applications of the NMR spectroscopic technique incessantly expanding in physics, chemistry, biology, geology, material science, and medicine, for spectroscopic studies and or imaging purposes. Small molecules to the large systems can be investigated by NMR in the liquid state, solid state, gel and as well as in vivo. NMR spectroscopy is a well-established quantitative tool which covers highly diverse area such as structure interpretation, pharmaceuticals, natural-products, mixture analysis, material sciences, and structural biology etc. It permits the characterization of chemical compounds at the atomic-level and provides information on structure and dynamics.

14.13.2.2 Solution-state vs. solid-state NMR (ssNMR) NMR spectroscopy is firmly concerned with the interactions between an isolated network of spins. The Hamiltonian operator in the presence of B0 for an isolated spin network is given by ĤNMR ¼ ĤZ + ĤCS + ĤD + ĤJ + ĤQ, where ĤZ denotes the Zeeman interaction, HCS for chemical shift (shielding (d)), ĤD, ĤJ, and ĤQ stands for dipolar (D), scalar coupling (J) and quadrupolar interactions (Q), respectively. The chemical shift (d) and spin-spin coupling (J-splitting) are the interactions mainly reflecting in the solution NMR

Surface Organometallic Chemistry and Catalysis

465

spectra, and the orientation-dependent interactions such as the dipole-dipole couplings (DD) and chemical shift anisotropy (CSA), get averaged due to the fast isotropic tumbling and molecular motions in solution. CSA arises due to the non-spherical distribution of electrons in atoms or molecules; more clearly, the non-spherical molecules orient differently concerning the magnetic field B0, which results in CSA. However, the molecular motion in solids is generally arrested. Hence the CSA and the DD interactions have significant effects on the solid-state NMR spectrum, which results in much broader and featureless signals compared to the solution-state NMR. Furthermore, the arrested molecular motion in solids increases the transverse relaxation rate (T2 and T2)15 as a consequence of extensive line broadening. The excited magnetic moment relaxes back to the equilibrium, involving the relaxation mechanisms, more specifically the spin-lattice relaxation (called longitudinal or T1) and spin-spin relaxation (called transverse or T2). The T2 defines the decay (dephasing) of excited magnetization in the transverse plane perpendicular to the B0. The line-width of a spectrum is related to the T2 and also influenced by magnetic field inhomogeneity. Combined, the T2 and magnetic field inhomogeneity is the actual transverse relaxation rate or T2 .15 To acquire a high-resolution ssNMR spectra, one needs to use magic-angle spinning (MAS) where the rotor containing sample needs to spin rapidly at an angle of y equals to 54.74 with respect to the external magnetic field (B0), which averages the anisotropies such as CSA and DD close to their isotropic values. In present days the maximum achieved commercially available spinning at a magic angle for ssNMR spectroscopy is 130 kHz. The CSA and DD interactions are orientation dependent on the relation, 1/2(3 cos2y −1). The cancellation of DD and reduction of CSA on MAS can be understood from the given relations.

D1,2 ¼ g1 g2 ħ=r1,2 3 1=2 3 cos 2 y −1 ð3 I1,2 I1,2 −ðI1 :I2 ÞÞ 

 szz ¼ siso + ð1=2Þ z 3cos2 y −1 + ZCS sin 2 y cos 2 j Where D1,2 is the dipolar interaction between two nuclei, g’s are the gyromagnetic ratios of spins, ħ is the reduced plank constant, r is the distance in space between two nuclei, y and j are the angles describing the location of nuclear spins concerning the external magnetic field, I stands for the nuclear spin number, szz is the chemical shift with anisotropy, siso is the isotropic chemical shift tensors and Z is asymmetry parameter which defines the shape of the electric field gradient (EFG). The rate of MAS should be equal or greater to the anisotropic interaction to average to its isotropic values (Fig. 1). Additionally, the molecular symmetry determines the number of resonances in the solution state NMR spectrum. On the other hand, in the solids, the molecules behave as part of a crystal lattice hence the crystallographic symmetry determines the number of peaks in the ssNMR spectrum. That is the reason for getting a greater number of peaks in solid-state NMR comparing to the solution state on many occasions.

14.13.2.3 The ssNMR and nuclei with quadrupolar magnetic moment Another parameter that causes broadening in ssNMR spectrum is quadrupolar interaction (Q), which appears for the nuclei having spin angular moment I > 1/2. Generally, nuclei with nuclear spin I ¼ 1/2, such as 1H, 13C, 19F, 29Si, 15N, 31P etc. are studied by ssNMR spectroscopy due to slow spin-spin relaxation they provide additional space for spin gymnastic by use of pulses and delays in various 1D and 2D experiments, addition to this the resolution for spin ½ nuclei is fairly good, and the result interpretation is straightforward. Nearly 100 out of 130 NMR active isotopes have nuclear spin I > ½, and the electric quadrupole interaction of these nuclei reflects as highly broad solid-state NMR spectrum. However, the high desire for characterization of inorganic materials led to developing the effective techniques to obtain the high-resolution ssNMR spectra of quadrupolar nuclei. The effect of electric

Fig. 1 Pictorial presentation of angle y of 54.74 , named “the magic angle” can potentially bring the dipolar term (3cos2y−1) to zero, which in co-ordination with fast-spinning resulting the highly resolved MAS ssNMR spectrum.

466

Surface Organometallic Chemistry and Catalysis

quadrupole interaction gets reduced at the higher magnetic fields, which are now commercially available higher than 21.1 T for solid-state NMR spectrometers. Also, the approaches of ssNMR traditionally utilized in the study of spin ½ nuclei have been adopted for quadrupolar nuclei, e.g., the echo methods, sample spinning at the magic angle (MAS), multi-dimensional spectroscopy, the adiabatic transfer, and double resonance. In addition to this, the special techniques, e.g., double rotation (DOR)16 and multiple-quantum transition in combination with fast-spinning (MQMAS),17 were developed for half-integer quadrupolar nuclei. The NMR study of quadrupolar nuclei with I > 1/2 such as 11B (I ¼ 3/2), 27Al (I ¼ 5/2), 17O (I ¼ 5/2), 93Nb (I ¼ 9/2), Rb, Cs, etc. by ssNMR is quite difficult due to the fast quadrupolar relaxation which causes extensive line broadening in the observed signal and limits the spin gymnastics in multi-pulse, multi-nuclear, and multi-dimensional experiments to extract the desired information.9,10 Due to these circumstances, the study of quadrupolar nuclei on surface organometallic chemistry (SOMC) is yet rarely reported in the literature.18–22 Nevertheless, many SOMC systems involve the quadrupolar nuclei. Hence, the NMR study of those nuclei is crucial to reveal the structural information.

14.13.2.4 Reintroduction of DD interaction in ssNMR and proximity information The presence of DD interactions in the ssNMR spectrum is not always the drawback. These parameters can be used to extract the structural information, dynamics and make NMR sensitive to the neighboring environment of the observed nuclei. The DD interactions are distinguished as the traceless second-rank tensors and reduced to zero under MAS condition, resulting in narrow spectral lines and making the analysis easy. In contrast, the valuable structural information of dipolar coupling constant is being lost, which is directly leading to the internuclear distances (r). Therefore, many approaches for the MAS NMR are introduced which can revive the homonuclear or heteronuclear DD interaction. DD recoupling can be achieved in the MAS NMR by applying the radio frequency pulses in synchronous with magic angle spinning to interfere with the averaging. Dipolar homonuclear recoupling in MAS NMR can be done by several recoupling sequences such as, radio-frequency driven recoupling (RFDR)23,24, finite pulse (fp)-RFDR applicable under fast MAS25 back-to-back (BABA),26,27 dipolar recoupling with a windowless multi-pulse irradiation (DRAWS),28 dipolar recovery at the magnetic angle (DRAMA), windowless version of DRAMA (DRAWS),29 RIL,30 C7 (sequence involves sevenfold symmetric phase shift of rotor-synchronized rf pulse cycles),31 POST-C7,32 CMR7,33 homonuclear rotary resonance (HORROR),34,35 and R-sequences36 etc. The coherence transfer pathway of 1H-1H BABA MAS NMR and pulse sequence is given in Fig. 2. The 2D NMR spectrum with rotor-synchronization can be acquired when the t1 increment becomes equal to a rotor period. This results in reducing t1 and therefore increasing the MQ spectral width, leads to a selective-MQ MAS spinning sideband pattern, which can be analyzed quantitatively in forms of coupling and internuclear distances.

14.13.2.5 NMR sensitivity The major drawback associated with NMR spectroscopy is its relatively low sensitivity, mainly due to the poor polarization of nuclear spin in the presence of external magnetic field results in lengthy measurement time for diluted samples. The polarization of nuclear spin can be determined by the Boltzmann equation at thermal equilibrium,

Fig. 2 The pulse sequence of 1H DQ BABA MAS NMR experiments (upper trace) and diagram of coherence transfer pathways (lower trace).37

Surface Organometallic Chemistry and Catalysis

467

P ¼ tanh ½g ℏ B0 =2Kb TŠ Where g is the gyromagnetic ratio of the spin, ħ is the reduced Planck constant, B0 is the external magnetic field, Kb is the Boltzmann constant, and T is the temperature. According to the Boltzmann equation of polarization, at 14.1 T the polarization of 1H nuclear spin is 0.000008 at 300 K, and 0.000002 for low-g 13C nuclei. The NMR experiments are generally repeated n times (referred to as number of scans, ns) to obtain the desired signal-to-noise (S/N), and this S/N ratio grows only with √ns. The minimal required concentration for a solution is in the micromolar range to obtain a good quality 1H NMR spectrum. The sensitivity of NMR instruments has been reasonably improved by using a high B0 and cryogenically cooled probes for solutions to suppress the thermal noise added by the electronics. However, expanding the B0 is not so simple and elevates the equipment cost drastically. The key challenge in applying the ssNMR on SOMC is the minute population difference and the existence of additional interactions discussed before. The implementation of MAS has improved the resolution of solid-state NMR,38 up to a great extent. However, the sensitivity of NMR spectrometers remains limited for several systems and materials of interest.

14.13.2.6 Hyperpolarization and DNP The “hyperpolarization” techniques are capable of achieving the nuclear polarization far higher than the thermal equilibrium and thus resulting in drastic signal enhancement. The most popular hyperpolarization approaches are spin-exchange optical pumping (SEOP)39 of noble gases, use of para-hydrogen induced polarization (PHIP),40 chemically induced dynamic nuclear polarization (CIDNP),41,42, and dynamic nuclear polarization (DNP).10,43–47 Among all the polarization techniques, the DNP found more efficient and promising in NMR applications, which can be utilized to extract the selective information for surface sites (vide infra). The DNP includes the transfer of electron polarization to nuclei through microwave irradiation in presence of a strong magnetic field. DNP has shown great potential in enhancing the sensitivity of solid-state as well as liquid-state NMR where the much higher thermal polarization of electrons, which arises from very high gyromagnetic ratio (g ¼ 28,024.952 MHz/T,) almost 660 times when compared with the 1H is being utilized. Hence the theoretically possible maximum improvement by DNP can be 660 (ge/g1H 660).

14.13.2.7 Concept of dynamic nuclear polarization (DNP) The concept of DNP is the transfer of huge polarization of the electron spins (compare to the nuclear spin) to the surrounding nuclear spins on microwave irradiation at, or near, the electron para-magnetic resonance (EPR) transitions. Based on the experimental conditions, the DNP phenomenon in the solids and in the solution can occur via several mechanisms namely, the Overhauser effect (OE)48,49 the solid effect (SE)50,51 the cross effect (CE),52 and the thermal mixing (TM)53 The DNP mechanisms demand a radicals species with very distinct electron spin resonance (ESR) lines. The most popular radicals being used in DNP are, 2,2,6,6-tetramethylpiperidin-1-yloxy (TEMPOL)54 TEKPol55,56 AMUPol57, TOTAPOL,58 BDPA59–61 trityl62 etc. The nitroxide based biradical TEKPol exhibits the high DNP enhancement (e 200) in the organic medium at 9.4 T and 100 K, furthermore, a urea-based biradical AMUPol, displayed the large polarization enhancement (e 235) in aqueous medium at 9.4 T and 100 K. Polarization transfer through cross effect needs strong dipolar interaction among two electrons and one nucleus which is generally 1H. The condition for polarization transfer is matching the Larmor frequency of 1H with the frequency difference of ESR lines for two electrons according to the given equation [e1 − e2] ¼ 1H. In the polarization transfer process, the microwaves of high-energy and high-frequency are produced by gyrotrons and being used to saturate one of the ESR transitions, which ensures the continuous transfer of polarization from electrons to 1H via three-spin “flip-flop-flip” process. The temperature 100 K, or lower, is required to slow down the electron spin relaxation which ensures the efficient polarization transfer. This DNP mechanism usually depends on nuclear spin diffusion, indeed the polarization transfer from electron to the neighbor core nuclear spins takes place in the first step, and the propagation of this polarization toward the bulk nuclear spin takes place by nuclear spin diffusion in the second step. To obtain the maximal DNP enhancement in the cross effect process, we need an unpaired electron source (fee radical source) as polarizing agents, the glass-forming solvent, low-temperature MAS unit, and a high-energy microwaves source. The whole concept of DNP/DNP SENS through cross effect is depicted pictorially in Fig. 3.63 Finally, the cross-polarization (CP)64,65 or other schemes of polarization-transfer, e.g., PRESTO66,67 can be utilized to transfer the polarization from 1H to the comparatively much less sensitive hetero nuclei of interest. CP from an abundant nuclear spin to a dilute spin and high gamma spin to low gamma spin is a double-resonance phenomenon68 which solves the two major problems of ssNMR. The first is the minute sensitivity of isotopically low abundant nuclei, particularly for low gyromagnetic ratio nuclei. The second problem associated with dilute nuclei of nuclear spin ½ is the long spin-lattice relaxation times (in the order of seconds and minutes) in solids, which increases the experiment time drastically. The CP technique facilitates the transfer of polarization from the nuclei of the high gyromagnetic moment (mostly 1H) to the nuclei of the low gyromagnetic moment, utilizes T1 of high gamma nucleus, and results in shorter experimental time. The CP in solids stunts the fact that the dilute and abundant nuclei are coupled via DD, which are located in close proximity. The transfer of polarization is only possible when the magnetizations of spins I and S fulfill the Hartmann-Hahn condition in the rotating frame.65 The combined application of CP with magic angle spinning (CP-MAS)69 yields the ssNMR spectra of a heteronucleus with high sensitivity and better resolution.

468

Surface Organometallic Chemistry and Catalysis

Fig. 3 Schematic presentation of cross effect DNP(SENS). A radical species is introduced into the pores by infusion. Magnified polarization is devoluted to the protons of the solvent and organic functionalities. CP is then used to devolute the elevated polarization to dilute spin. Reproduced with permission from ref. Liao, W.-C.; Ghaffari, B.; Gordon, C. P.; Xu, J.; Copéret, C., Curr. Opin. Colloid Interface Sci. 2018, 33, 63–71. Copyright Elsevier.

14.13.2.8 Experimental approaches of DNP and DNP-MAS The execution of DNP under MAS condition in the early 1990s turned as a game-changer in huge elevation of nuclear spin polarization, resulted in high S/N in the NMR spectrum, which is being improved day to day. The DNP-MAS can be executed using a MAS probe equipped with waveguide and beam launcher, which facilitates the radio frequency (rf ) irradiation of the sample with microwave under MAS conditions at 100 K temperature. The continuous microwave irradiation is being ensured by an external gyrotron with the frequency approximately to the electron spin resonance frequency of the NMR spectrometer. The in situ experimental setup makes the process more compatible for the majority of the ssNMR experiments, including 1D and 2D NMR techniques. The elevated sensitivity in NMR signal using DNP has been demonstrated in several practical applications, such as polymorphs,70,71 zeolites72–77 colloidal quantum dots78,79, silica materials80–85 including SOMC catalysts,7-10,86 etc., acquire various multinuclear,1D, and multi-dimensional NMR experiments that are hard-to-achieve even with enriched NMR active isotopes.

14.13.2.9 Effective surface enhancements and DNP SENS The close proximity between radicals, and the polarization-transfer medium (1H of solvents), and the surface of an analyte is quintessential for fast construction of the nuclear hyperpolarization, which results in maximum signal enhancement by DNP.10,43–47 The above-mentioned demand encouraged the NMR community to develop the “DNP surface-enhanced NMR spectroscopy” (DNP SENS).46,80 The DNP SENS technique46,80,87 is practically able to enhance the signal to noise (S/N) in order of two- to threefold and has been applied to a variety of surface systems,81,88–98 for characterization and establishing the three-dimensional (3D) structures. In a typical DNP-CENS, ssNMR experiment, the incipient wetness impregnation (IWI)99 of the analyte substrate is used before analysis, where the analytes are impregnated with the sufficient radical solution to uniformly wet the surface of the substrate or fill the pores of porous materials such as SiNPs. The IWI approach does not dilute the solid analyte substrates and is very simple. Typically, the materials with a higher surface area such as, nanoparticles and micro/mesoporous materials are found ideal for DNP SENS for which the radical concentration can be optimized to achieve the optimal enhancement.

14.13.2.10

DNP SENS ssNMR in SOMC: Challenges and solutions

The more sensitive DNP SENS ssNMR technique can be utilized for the structural determination of SOMC and heterogeneous catalysts45 with the condition, they should not be sensitive to the free radical,89,92,100 else the active site(s) should be protected.97,98,101–105 Unfortunately, most of the SOMC complexes are extremely sensitive and prone to react with the free radicals.97 Many strategies are reported to protect SOMFs to conquer the challenge; herein, we will discuss the protection strategy, named confinement. The confinement strategy enables the use of DNP SENS in the characterization of highly sensitive SOMFs grafted on surfaces by ssNMR spectroscopy. For example, the DNP SENS ssNMR was applied to [(^SidOd)W(^CtBu) (dCH2tBu)2] supported by mesoporous silica of various pore sizes (dpore ¼ 6.0, 3.0, and 2.5 nm).106 This mesoporous support

Surface Organometallic Chemistry and Catalysis

469

Fig. 4 1D 1H ! 13C CP-MAS DNP SENS spectrum (100 K, 9.4 T/263GHz) of [(^SidOd)W(^CtBu)(dCHt2Bu)2] in a 16 mM TEKPol solution in TCE. Reproduced with permission from ref. Pump, E.; Viger-Gravel, J.; Abou-Hamad, E.; Samantaray, M. K.; Hamzaoui, B.; Gurinov, A.; Anjum, D. H.; Gajan, D.; Lesage, A.; Bendjeriou-Sedjerari, A.; Emsley, L.; Basset, J.-M., Chem. Sci. 2017, 8 (1), 284–290.

prevents direct contact between large nitroxide biradical TEKPol55,56 (dradical  2 nm) and the complex on the surface. The samples of the SOMC complex were prepared by IWI99 with a solution of TEKPol in dichlorobenzene (DCB) or tetrachloroethane (TCE). On SBA-15 support (with large pore diameter), no surface signals were observed in 1H ! 13C CP-MAS DNP SENS ssNMR spectrum, most likely due to the reaction of TEKPol with the surface complex. On the other hand, the 1H ! 13C CP-MAS DNP SENS ssNMR spectrum of [(^SidOd)W(^CtBu)(dCHt2Bu)2] contains all expected peaks of the surface complex as, (CH3) 33 ppm, (dC(CH3)3) 52 ppm and [[W](dCH2dC^)] 95 ppm, when MCM-41 (dpore ¼ 3.0 nm) used as support. Even the good carbyne signal at 317 ppm ([W]^Cd) was also detected only in 8000 scans, while the conventional ssNMR needs at least 70,000 scans to observe only the alkyl fragments (32, 52, and 95 ppm) (Fig. 4) and the carbyne signal at 317 ppm ([W]^Cd) being undetected.106 This example illustrates the DNP SENS as a powerful characterization technique in the limited time frame when compared to conventional NMR. It turned out that the DNP SENS ssNMR technique is even applicable when the SOMF is grafted on the surface of spherical silica if the SOMF is small enough and the concentration is moderate.106 In such a situation, the silica particles are aggregated and create crowding around the surface complex; this crowding creates hindrance, which might be unfavorable for the expected bulky intermediate complex formation between SOMF complex and biradical. The consequence to this the SOMF complex grafted on the surface being protected from the reaction with biradical.

14.13.2.11

The ssNMR with and without DNP SENS in the characterization of SOMC

ssNMR is proven as a powerful spectroscopic technique for the investigation and characterization of heterogeneous catalysts synthesized using SOMC7–10 by grafting SOMFs on the surface of the porous materials such as SiNPs where catalysts connected to the tether of the surface.107 The traditional NMR spectroscopy has extremely low sensitivity in the characterization of SOMFs grafted to silanols of SiNPs, because the SOMFs are mainly organometallic complexes of early transition metals, which can be grafted 1–5 wt% only, to design a well-defined heterogeneous catalyst108 using the “Catalysis by Design” strategy.4 Besides, the 2D methods are often utilized to determine the structures of SOMC catalysts at a molecular level demand much extended experimental time. More specifically, the 2D 13Cd13C Incredible Natural Abundance DoublE QUAntum Transfer Experiment (INADEQUATE)109–111 correlates the 13C normal frequencies with the 13C double quantum (DQ) frequencies on its natural abundance or partially enriched 13C which establishes the connectivity of carbon skeleton, the proton SQ resonance with DQ112–114 and TQ113,115,116 correlates in 2D 1Hd1H double DQ-SQ and 2D 1Hd1H TQ-SQ ssNMR experiments respectively or the resonances of protons with other heteroatoms in heteronuclear correlation spectroscopy (HETCOR) experiments117,118 the combined rotation and multiple sequences (CRAMPS)119,120 to achieve the narrowest proton linewidths and many more advanced experiments demand a huge amount of spectrometer time in normal conditions. In such a situation, the use of CP-MAS and DNP-SENS in ssNMR proved as a wonderful tool for fast data acquisition and characterization of SOMFs and other heterogeneous catalysts.45,46 There are many reports where DNP SENS is successfully used to derive the structure of complex molecules46,80,87 grafted on silica’s surface,121 alumina,122–124 zeolites72,73 and functionalized heterogeneous catalysts.73,79,90,97,98,101–106 Here, we will illustrate how ssNMR spectroscopy using CP-MAS and DNP SENS help obtain and understand the structure of catalyst designed by SOMC strategy and supports with the help of some derived examples the development of more efficient catalysts.

470

Surface Organometallic Chemistry and Catalysis

14.13.2.12

15

N ssNMR spectroscopy

The 15N is a spin ½ nucleus, which is the best for NMR studies; however, the natural abundance of this nucleus is very low (0.37%), which makes it unfavorable for NMR investigations when the sensitivity is concerned. Particularly, it is extremely challenging to obtain a reasonably good ssNMR spectrum on this nucleus, the situation becomes worst when the complex is grafted on a surface in a well-defined system where loading of SOMF is only 1–5% concerning silica or other supports. The application of DNP in the ssNMR spectroscopy facilitates the characterization of 15N containing materials up to a great extent. On using DNP enhancement, the several heterogeneous systems (hybrid materials, SOMC catalysts, etc.) bearing nitrogen-based moieties were successfully investigated. Several examples, including peptides125 aniline-functionalized silica95 and nitridated silica surfaces126 are also characterized and reported. The 15N DNP-SENS ssNMR spectra given in Fig. 5 were utilized in the characterization of [(^SidOd)Hf(Z2,p-MeNCH2)(Z1-NMe2)(Z1-HNMe2)] and [(^SidOd)Hf(]NMe)(Z1-NMe2)] SOMC catalysts of imine metathesis. The 15N spectrum of [(^SidOd)Hf(Z2,p-MeNCH2)(Z1-NMe2)(Z1-HNMe2)]127 exhibits two resonances at 7 ppm and 32 ppm where peak at 32 ppm was assigned to the tri-coordinated 15N of (Z2,p-MeNCH2) and dNMe2 functionalities. The signal at 7 ppm assigned to the 15N of NH(CH3)2 moiety. The spectrum of [(^SidOd)Hf(]NMe)(Z1-NMe2)] contains the 15N signals at 113 ppm, 34 ppm and 7 ppm where the 15N peak at 113 ppm is assigned to the hafnium imido moiety (]NMe).

14.13.2.13

17

O MAS NMR spectroscopy

In the characterization of supported metal oxide species, the 17O ssNMR spectroscopy is found as an interesting tool. Additionally, the hydrogen bonding (HB) and dynamics on surfaces of catalytically important materials can also be analyzed. The major challenge associated with 17O NMR spectroscopy is extremely low natural abundance (0.037%) and its quadrupolar nature (I ¼ 5/2).128 To obtain a reasonably good ssNMR spectra of 17O on silica surfaces, one needs to either treat the support material with 17O enriched water129,130 or with the 17O2 gas.131 From 17O MAS ssNMR spectra given in Fig. 6, one can gather the key information on SOMC complexes about distinct environments of oxygen such as, silanols which appears around 0–10 ppm, siloxanes nearly 20–50 ppm, and metal-bound siloxides mostly resonates in the range of 150–250 ppm. The 17O MAS ssNMR spectra on its natural abundance for mesoporous SBA-15 was also obtained using DNP enhanced ssNMR and reported.132,133

14.13.2.14

31

P ssNMR spectroscopy

Key structural features were revealed from 1D solid-state 31P MAS ssNMR spectroscopy,83 the 1D 31P CPMAS ssNMR spectra of Mat-RuF and Mat-RuR are given in Fig. 7. Here for Mat-RuF only one 31P signal at 47 ppm is characteristic of P(V) products that result from the reaction of free PCy3 with the silica surface.134,135 The absence of 31P NMR signal associated to the coordination of PCy3 to Ru, concluded as two PCy3 ligands, which were previously present in [(Cl)2Ru(]CHPh)(PCy3)2], liberated while grafting and the NHCdRu center associated to the surface having a flexible tether without phosphine bound and interacts with oxygen at the surface of the material, probably an OSiMe3 moiety or a siloxane bridge, to stabilize the otherwise.

14.13.2.15 1

1D MAS and 2D HETCOR ssNMR experiments 13

The H and C are the most explored nuclei in NMR spectroscopy among all magnetically sensitive nuclei. The 1D 1H, and 13C MAS experiments provide the resonance frequencies of mentioned nuclei which are directly related to the chemical environment surrounding them. A major strength of contemporary NMR spectroscopy is the versatile and rich structurally informative multinuclear, multi-dimensional correlation spectra. From the 2D NMR technique catalog, the most widely explored 2D ssNMR

Fig. 5 15N MAS DNP SENS spectra of (A) [(^SidOd)Hf(Z2,p-MeNCH2)(Z1-NMe2)(Z1-HNMe2)] and (B) [(^SidOd)Hf(]NMe)(Z1-NMe2)]. Reproduced with permission from ref. Aljuhani, M. A.; Barman, S.; Abou-Hamad, E.; Gurinov, A.; Ould-Chikh, S.; Guan, E.; Jedidi, A.; Cavallo, L.; Gates, B. C.; Pelletier, J. D. A.; Basset, J. M., Acs Catal. 2018, 8 (10), 9440–9446. Copyright American Chemical Society.

Surface Organometallic Chemistry and Catalysis

471

Fig. 6 1D 17O MAS ssNMR spectra (18.8 T, 19 kHz MAS for (A–C) and 21 kHz MAS for (D)) of (A) ZrdSiO2-200, (B) TadSiO2-200, (C) WdSiO2-200, and (D) WdSiO2-200. The spinning side bands are identified by asterisks. Reproduced with permission from ref. Merle, N.; Trébosc, J.; Baudouin, A.; Rosal, I. D.; Maron, L.; Szeto, K.; Genelot, M.; Mortreux, A.; Taoufik, M.; Delevoye, L.; Gauvin, R. M., J. Am. Chem. Soc. 2012, 134 (22), 9263–9275. Copyright 2012 American Chemical Society.

Fig. 7 1D 31P CPMAS ssNMR spectra of (A) Mat-RuF and of (B) Mat-RuR. All peaks marked with asterisks are spinning sidebands separation equal to 10 kHz MAS frequency. Reproduced with permission from ref. Samantaray, M. K.; Alauzun, J.; Gajan, D.; Kavitake, S.; Mehdi, A.; Veyre, L.; Lelli, M.; Lesage, A.; Emsley, L.; Copéret, C.; Thieuleux, C., J. Am. Chem. Soc. 2013, 135 (8), 3193–3199. Copyright 2013 American Chemical Society.

472

Surface Organometallic Chemistry and Catalysis

Fig. 8 (A) 1D 1H MAS and (B) 1H ! 13C CP-MAS NMR spectrum and (C) 2D 1Hd13C HETCOR NMR spectrum of silica-supported rhenium alkylidene species [(^SiO)Re(^CtBu)(]CHtBu)(dCHt2Bu)2], the contact time of 0.5 ms for cross polarization in HETCOR experiment was used. Panels a, b and c are reproduced from ref. Chabanas, M.; Baudouin, A.; Coperet, C.; Basset, J.-M.; Lukens, W.; Lesage, A.; Hediger, S.; Emsley, L., J. Am. Chem. Soc. 2003, 125 (2), 492–504, with permission and modified from the American Chemical Society 2003.

experiment in materials chemistry is HETCOR, particularly involving 1H correlating with another nucleus. The correlation can be mediated either through J-coupling or through dipolar coupling. The correlation of 1H to 13C, 29Si, 31P, and other nuclei are very popular. Similarly, the correlation between two non-proton spins can be made. As an example, the 2D 1Hd13C HETCOR experiments in ssNMR providing the correlation between 1H and 13C resonances. These experiments can yield short as well as long-range 1Hd13C correlations on demand. Combined the above discussed 1D and 2D experiments have proved their importance as a characterization technique in nanotechnology, surface chemistry, and heterogeneous catalysis. ssNMR facilitates the characterization of well-defined solids, grafted linkers, hybrid materials, and heterogeneous catalysts. We are discussing here some important examples of 1D 1H, 13C, and 1H-X HETCOR ssNMR experiments (spectra are given in Fig. 8) explored to understand the structure of SOMC catalysts. The 2D 1Hd13C HETCOR MAS NMR spectrum of [(^SiOd)Re(^CtBu)(]CHtBu)(dCH2tBu)2] shows a correlation (a) between protons around 1.4 ppm and carbon at 30 ppm. Moreover, the 13C resonance of the methylene group of CHt2Bu at 44 ppm showing two correlations b and c with proton resonances at 2.6 and 3.0 ppm, corresponding to the two diastereotopic CH2 protons. Another correlation peak (d) between proton at 11.0 ppm and the carbon at 246 ppm, confirming the assignment as a carbynic group. Using longer contact time, generally greater 1 ms, it is possible to observe correlation peaks, arise through long-range dipolar interactions. The 1Hd13C HETCOR spectrum of [(^SiOd)Re(^CtBu)(]CHtBu)(dCH2tBu)2] with a contact

Surface Organometallic Chemistry and Catalysis

473

Fig. 9 The long-range 1Hd13C HETCOR ssNMR spectra (A) Full spectrum (B) of an expanded region, of the surface species [(^SiO)Re(^CtBu)(]CHtBu) (CHt2Bu)2] acquired with a 5 ms contact time for cross-polarization. Reproduced with permission from ref. Chabanas, M.; Baudouin, A.; Coperet, C.; Basset, J.-M.; Lukens, W.; Lesage, A.; Hediger, S.; Emsley, L., J. Am. Chem. Soc. 2003, 125 (2), 492–504. Copyright 2003 American Chemical Society.

time of 5 ms reported in Fig. 9. Using the long-range correlations in this spectrum, the much complex surface species [(^SiOd)Re(^CtBu)(]CHtBu)(dCHt2Bu)2] is fully characterized. In this article on utilizing the 1D and 2D ssNMR experiments, the authors also proved the presence of two syn and anti-species formed on heating the surface complex [(^SiOd)Re(^CtBu) (]CHtBu)(dCHt2Bu)2] at 120  C.136

14.13.2.16

1D 29Si CP-MAS and 2D 1Hd29Si HETCOR ssNMR in characterization of SOMC

In the SOMC approach, the organometallic complexes are grafted on solids supports (silica, silica-alumina, etc.) contains silica in their molecular framework. Besides the structural determination, the ssNMR also facilitates the complete characterization of unfunctionalized support materials such as silica, alumina, silica-alumina, etc. being used for SOMC. Among the available supports for SOMC, partially dehydroxylated silica is most common. 29Si NMR spectroscopy7,137,138 can distinguish the environment of the silicon atoms using the 29Si chemical shift, which is dependent on silicon −oxygen bonds (Q-sites). The peaks of 29Si in (i) Q4-sites, where a Si atom is connected to other Si atoms via siloxane bridges only (bulk silica (SiO2)) appear in the range of −115 to −105 ppm (ii) Q3-sites, related to single surface silanols (^SidOH), found around −100 ppm and (iii) Q2-sites describe adjacent bis-silanols [(^Si(dOH)2)], appear in the range of −85 and −95 ppm in the NMR spectrum.138 On partial dehydroxylation of SBA-15 at 700  C, the 29Si CP/MAS ssNMR spectrum showed three peaks at −101, −92, and −110 ppm, where the most prominent peak at −101 ppm belongs to Q3-sites, indicating the presence of surface silanols in excess, the peaks at −92 ppm belongs to Q2-sites (adjacent silanols), and the signal at −110 ppm refers to Q4-sites.137,138 Additionally, the 29Si NMR spectroscopy can also distinguish the T-sites, which describes the environment of silicon atom concerning SidC bonds (following the same systematic

474

Surface Organometallic Chemistry and Catalysis

Fig. 10 2D 1Hd29Si HETCOR DNP SENS ssNMR spectra at 9.4 T for (A) Mat-ImR and (B) Mat-ImF. The blue vertical bands show the correlations between the surface Tn sites and the ortho 1H (panel A) and the CH2dSi protons (panel B) of the ligands. In panel b, the red bands show the correlations of surface Qn silicon with the methyl and aromatic protons of the ligand. The DNP enhancements were eH ¼ 30 and 26 for Mat-ImR (A) and Mat-ImF (B), respectively. A minute amount of sodium 2,2-dimethyl-2-silapentane-5-sulfonate (DSS), was used as an internal standard for 1H and 29Si chemical shifts. A green box (panel a) reports the Tn sites with a twofold increased intensity. Reproduced with permission from ref. Samantaray, M. K.; Alauzun, J.; Gajan, D.; Kavitake, S.; Mehdi, A.; Veyre, L.; Lelli, M.; Lesage, A.; Emsley, L.; Copéret, C.; Thieuleux, C., J. Am. Chem. Soc. 2013, 135 (8), 3193–3199. Copyright 2013 American Chemical Society.

as Q-sites). The T-sites typically appear by the migration of alkyl group(s) from the metal complex to Si of the surface (Fig. 10).139,140 Similar to 13C, the 29Si ssNMR spectroscopy is also suffering from low sensitivity due to its low natural abundance (4.7%) and low gyromagnetic ratio. The DNP enhanced ssNMR spectroscopy helps to toggle the sensitivity issue, and spectra can be acquired in a reasonable time, either with CP or only DNP.121 The 2D 1Hd29Si HETCOR ssNMR experiments could be found much informative in the characterization of SOMC catalysts81 as silica is the common nucleus in mostly all kinds of supports. The 2D 1 Hd29Si HETCOR ssNMR spectra of Mat-ImR and Mat-ImF, acquired using DNP with a short mixing time of 400 ms for cross polarization, to ensure the appearance of only the short rage correlations. On assigning the resulted 1Hd29Si HETCOR spectra, the authors observed that the flexible tether allows the phenyl ring and the attached methyl groups to fold back onto the surface. The large separation between ligand and surface in Mat-ImR was concluded by the absent correlation between them.83

Surface Organometallic Chemistry and Catalysis

14.13.2.17

475

2D DNP SENS 13Cd111Cd HETCOR spectra of CdSe QDs

The HETCOR spectra between two hetero nuclei rather than 1H could be acquired to gather the proximity information among them. In order to this DNP enhanced 13Cd111Cd 2D dipolar heteronuclear multiple-quantum coherence (D-HMQC) spectrum (Fig. 11) with partially 13C-labeled carboxylate carbon of a surface ligand oleate-capped ZB-CdSe QDs was found convenient.79 The correlations between 13C and 111Cd at −317 ppm (Cd dimension) in the resulting spectrum proves the close proximity between these heteronuclei which is interpreted as an evidence for the CdSecore(CdX2)shell model.

14.13.2.18

27

Al ssNMR spectroscopy

The ssNMR spectroscopy for a quadrupolar nucleus, such as 27Al is commonly less popular compare to spin half nuclei9 due to their intricate spin dynamics in the presence of the B0, which leads to extensive spectral broadening resulting from the complex Zeeman interactions. The quadrupolar effect (Q) accelerates the T2 relaxation, which leads to rapid dephasing of spin magnetization in transverse plane and consequence to this extensive line broadening and S/N-ratios close to the detection limit. The quadrupolar effect can be minimized by MAS up to first order but cannot be completely removed. The application of 27Al ssNMR spectroscopy is very crucial in heterogeneous catalysis,9 especially for zeolites, single-atom catalysts, or with well-defined SOMC-catalysts. Mostly three types of aluminium atoms can be easily distinguished by NMR spectroscopy, where the tetra coordinated AlT or Al(IV) appearing around 70 ppm, Penta-coordinated Al(V) (or AlP)nearly 35 ppm and Hexa coordinated Al(VI) (or AlH) aluminium nuclei resonation close to 8 ppm in 27Al NMR spectrum. A unique category of mesoporous silica with well-defined single-site and terminal selective AldH was designed in 2016,141 where the trimeric-isoBu2AlH was selectively reacted with the surface silanols without affecting the siloxane bridges. The starting material that is aluminium-isobutyl complex and the final hydride of aluminium were completely characterized by the application of advanced ssNMR techniques (1H, 13C, 2D 1Hd1H MQ-SQ correlation, and 27Al spectroscopy). An intense signal in the 27Al ssNMR spectrum at 68 ppm corresponding to Al(IV) was observed on the other hand the characteristic resonances of Al(V) and Al(VI) were not detected (Fig. 12).

14.13.2.19

HETCOR between 1H and Quadrupolar nuclei

Gauvin, Taoufik, Scott et al.142,143 successfully characterized the structure of SOMC catalysts (methyl-trioxorhenium (MTO) on g-Al2O3) by 27Al ssNMR spectroscopy. The 2D 1Hd27Al dipolar heteronuclear multiple quantum coherence (D-HMQC) MAS

Fig. 11 DNP enhanced 2D 13Cd111Cd D-HMQC spectrum correlating 13C of 1-oleate ligands and 111Cd of ZB-CdSe. Projections of 2D spectrum in dark blue, 1D 111 Cd CPMAS DNP NMR spectrum in light gray are shown. The isotropic shifts of the surface and core Cd-species are highlighted by solid and dotted black curves. The spinning sidebands are identified by asterisks. Reproduced with permission from ref. Piveteau, L.; Ong, T.-C.; Rossini, A. J.; Emsley, L.; Copéret, C.; Kovalenko, M. V., J. Am. Chem. Soc. 2015, 137 (43), 13964–13971. Copyright 2015 American Chemical Society.

476

Surface Organometallic Chemistry and Catalysis

Fig. 12 1D 27Al ssNMR spectrum after thermal treatment of tetracoordinated AldH. Reproduced with permission from ref. Werghi, B.; Bendjeriou-Sedjerari, A.; Jedidi, A.; Abou-Hamad, E.; Cavallo, L.; Basset, J.-M., Organometallics 2016, 35 (19), 3288–3294. Copyright 2016 American Chemical Society.

ssNMR spectrum of MTO/CldAl2O3 is given in Fig. 13, where all three types of aluminium sites were detected. In this spectrum, the 1 H resonances in between 1.5 and 3.5 ppm (under green dotted rectangle) are mostly attributed to m2-OH and the resonances range from 3.5–6 ppm are assigned to m3-OH. The m3-OH signals of the 1H frequency domain are strongly correlated with AlH and, to a lesser extent with AlT. The signal of CldAl2O3 region is mostly unaffected by grafting, kept under dashed rectangles in Fig. 13C. The area marked by solid rectangle includes the 1H chemical shift of RedCH3, which is showing the correlations with all types of aluminium atoms with the relative proportions of AlH  AlT > AlP where the lower chemical shift region correlates with AlT/AlH, the high chemical shift region with AlP/AlH. The above discussed pairwise correlations indicate the presence of strong interactions between Re and bridging oxygen in the AlnOAlH, (n ¼ T or P) fashion as suggested in Fig. 13C.

14.13.2.20

HETCOR between a low abundant spin ½ and quadrupolar nuclei

The correlation experiments between two less abundant, low gamma and quadrupolar nuclei on normal NMR spectrometer are very challenging because the polarization transfer to or from quadrupolar nuclei is a very challenging addition to the poor sensitivity. In such a situation, the use of DNP SENSE ssNMR spectroscopy was found as a vigorous tool. In a similar condition, DNP SENS 2D 29 Sid27Al HETCOR ssNMR spectra (with the INEPT and dipolar transfer schemes) of Al/SiO2 was acquired to get the direct insight into the interaction between Alumina and SiO2, on changing the alumina loading (Fig. 14 spectra A–C).82 In the obtained spectra the 29Sid27Al INEPT HETCOR correlates the only scalar coupled aluminium with the silicon atoms connected by bridging oxygen (black traces); on the other hand, the 29Sid27Al D-HETCOR spectra are providing the proximity information between both bonded and non-bonded aluminium and silicon atoms (red contours/traces). The correlations in 29Sid27Al INEPT HETCOR reveals that the silica tetrahedra are directly bound to aluminium tetrahedra in Al/SiO2, and the correlations in 29Sid27Al D-HETCOR spectra indicates the proximity between 29Si Q4 sites with 27Al(IV) and octahedral.82

14.13.2.21 13

2D INADEQUATE ssNMR spectroscopy

The 2D Cd13C INADEQUATE109–111 is a powerful NMR tool to identify the neighboring carbons connected through covalent bonds. However, it is extremely inconsiderate as 13C on its natural abundance is only 0.01%. The resulting spectrum having 13C chemical shifts on f2 axis and the 13C DQ frequency on f1 axis. The 2D 13Cd13C INADEQUATE ssNMR spectrum in the Fig. 15B evidenced the metallacycle (TBP-1) formation in the intermediate during the olefin metathesis reaction of ethylene with the SOMC catalyst [(^SiOd)W(]NAr)(]CHtBu)(dOtBuF9)]144. Before acquiring the INADEQUATE spectrum the SOMC catalyst [(^SiOd)W(NAr)(]CHtBu)(dOtBuF9)] was reacted with 10 equivalents of ethylene (13C, 2H-labeled) which was expected to

Surface Organometallic Chemistry and Catalysis

477

Fig. 13 (A) 27Al chemical shift projection, (B) 1H MAS projection, and (C) 2D 1Hd27Al D-HMQC MAS (18.8 T, 20 kHz MAS recoupling time 500 ms) ssNMR spectra of methyltrioxorhenium (MTO)/CldAl2O3. Reproduced with permission from ref. Gallo, A.; Fong, A.; Szeto, K. C.; Rieb, J.; Delevoye, L.; Gauvin, R. M.; Taoufik, M.; Peters, B.; Scott, S. L., J. Am. Chem. Soc. 2016, 138 (39), 12935–12947. Copyright 2016 American Chemical Society.

Fig. 14 The 29Sid27Al INEPT HETCOR DNP SENS ssNMR spectra of Al/SiO2 interface demonstrating the direct bonding between 29Si Q4 and tetrahedral aluminium and the proximity with octahedral aluminium sites. The scalar refocused INEPT (black traces/contours) and dipolar refocused R3-INEPT (red traces/ contours) (A) 5.7 wt% Al/SiO2, (B) 15.0 wt% Al/SiO2, (C) 24.8 wt% Al/SiO2. Reproduced with permission from ref. Valla, M.; Rossini, A. J.; Caillot, M.; Chizallet, C.; Raybaud, P.; Digne, M.; Chaumonnot, A.; Lesage, A.; Emsley, L.; van Bokhoven, J. A.; Copéret, C., J. Am. Chem. Soc. 2015, 137 (33), 10710–10719. Copyright 2015 American Chemical Society. t 13 13 yield the reaction intermediate [(^SiOd)W(]NAr)(13CD13 Cd13C INADEQUATE 2 CD2 CD2)(dO BuF9)]. In the resulting 2D 13 t 96 13 13 ssNMR spectrum of the surface catalyst [(^SiOd)(W]NAr)( CD2 CD2 CD2)(dO BuF9)], as expected, the authors observed a one-bond CadCb (marked in the Fig. 15A) correlation, along with a weaker correlation between CadCa, which are two-bond apart, which evidenced the formation of tungsten TBP metallacycle (TBP-1) as a reaction intermediate.

478

Surface Organometallic Chemistry and Catalysis

t 13 Fig. 15 400 MHz (A) 1D DNP SENS 13C CP MAS (B) 2D 13Cd13C refocused INADEQUATE DNP SENS ssNMR spectra of [(^SiO)W(NAr)(13CD13 2 CD2 CD2)(O BuF9)], Ar ¼ 2,6-iPr2C6H3. Asterisks indicate spinning sidebands. Reproduced with permission from ref. Ong, T.-C.; Liao, W.-C.; Mougel, V.; Gajan, D.; Lesage, A.; Emsley, L.; Coperet, C., Angew. Chem. Int. Ed. 2016, 55 (15), 4743–4747.

14.13.2.22

The multiple quantum (MQ) NMR spectroscopy

The first experimental evidence of MQ phenomenon in NMR spectroscopy was made at the end of the 1950s, and the group of A. Pines demonstrated it for the solid-state145–147 in the early 1980s. The NMR signals are the result of nuclear spin transitions from one spin state to others. In solids, these transitions are more prominent through dipolar interactions. Depending on the effective dipolar coupling strength, the n number of MQ coherences can be excited. In MQ experiments the observed frequency appears as the addition of SQ frequencies if those spins interacting through dipolar interaction or through scalar coupling, more specific the resonance frequency of chemically equivalent nuclei in a DQ experiment appearing as double, and three times in TQ experiment

Surface Organometallic Chemistry and Catalysis

479

compared to SQ resonance frequency. The resonance frequencies of chemically distinguished nuclei just appear at the sum of both the frequencies in DQ experiment. The application of selective MQ excitation could extract the internuclear distances, residual dipolar couplings (RDCs), and quadrupolar coupling (Q) parameters of a system. Additionally, the dependency of dipolar interaction on internuclear distances and molecular orientations can be used to determine the structural and dynamics information.

14.13.2.23

2D 1Hd1H SQ-MQ correlation ssNMR experiments

The breakthrough in SOMC was grafting a very sensitive molecule WMe6 on partially dehydroxylated silica. The homogeneous structure of the WMe6 complex was reported in 1996 by Seppelt et al.147 Since then, it took nearly 18 years, until 2014 when this complex was first grafted on partially dehydroxylated silica SiO2-700, which turned as a powerful implication in the area of catalysis.139 The application of the CP/MAS technique in NMR helps to precisely measure the resonance frequency of the organometallic complex grafted on oxide supports.139 The low-temperature ssNMR experiment findings proved the pseudotrigonal-prismatic arrangement of [(^SidOd)WMe5] where two types of distinct methyl groups showing a 2:3 stoichiometry were observed. On the other hand, at room temperature, just one signal for all the five methyl 1H/13C was observed because of coalescence.139 In the 2D 1Hd1H DQ-SQ spectrum of [(^SidOd)WMe5] (Fig. 16B, top trace) the 1H SQ frequency (1.93 ppm) shows correlation (at 3.86 ppm) with DQ frequency confirms the chemical equivalence between two protons. Additionally, this SQ frequency showing the correlation in TQ dimension of 2D 1Hd1H TQdSQ spectrum at 5.79 ppm (Fig. 16B, bottom trace), which

Fig. 16 (A) 1D 1H MAS solid-state NMR spectrum (B) 2D 1Hd1H DQ-SQ (upper trace) and 2D 1Hd1H TQdSQ (lower trace) correlation NMR spectra. (C) 1H ! 13C CP-MAS NMR (d) 2D 1H ! 13C CP-MAS 2D 1Hd13C dipolar HETCOR spectrum, of [(]SidOd)W(CH3)5]. Reproduced with permission from ref. Samantaray, M. K.; Callens, E.; Abou-Hamad, E.; Rossini, A. J.; Widdifield, C. M.; Dey, R.; Emsley, L.; Basset, J.-M., J. Am. Chem. Soc. 2014, 136 (3), 1054–1061. Copyright 2014 American Chemical Society.

480

Surface Organometallic Chemistry and Catalysis

is exactly thrice the SQ frequency confirming the presence of three chemically equivalent protons (CH3 group). The above-discussed type of correlations are known as self-correlations. The structure and the well-defined nature of the SOMF in the catalyst was further confirmed by 1Hd13C dipolar HETCOR experiment (Fig. 16D). The DQ and TQ correlation experiments are also useful in finding the spatial distance between protons of different functional groups if they are present close in proximity. For example, in the 1H ssNMR spectrum of a bimetallic surface complex [(^SidOd)W(CH3)5/(^SidOd)Zr(Np)3] the CH3 of (W(CH3)5) resonates at 2 ppm, and CH2 and CH3 of ZrNp at 1.2 and 0.9 ppm respectively (Fig. 17A). The 2D 1Hd1H DQ-SQ spectrum of this complex contains the self-correlated peaks in DQ dimension for CH3 of (W(CH3)5), CH2, and CH3 of ZrNp at 4.0, 2.4, and 1.8 ppm respectively, as expected. Additionally, the CH3 of (W(CH3)5), and the CH3 of ZrNp showing peaks at 2.9 ppm in DQ frequency dimension, are assigned as DQ (SQ(W(CH3)5) + SQZrNp) correlation between CH3 groups of both surface complexes (Fig. 17B). This correlation proves the close proximity among grafted WMe5 and ZrNp3.148 The 2D 1Hd1H TQ-SQ spectrum of the same molecule showing the self-correlation in TQ dimension for CH3 of (W(CH3)5), and CH3 of ZrNp protons at 6.0 and 2.7 ppm respectively. Also, the CH3 of (W(CH3)5), and the CH3 of ZrNp show peaks at 4.9 ppm in TQ frequency dimension, which are assigned as TQ (DQ(W(CH3)5) + SQZrNp) correlation between both CH3 groups (Fig. 17C). These correlations prove the close proximity between two protons of CH3 of (W(CH3)5), and one proton of CH3 of ZrNp in the complex. The above-obtained information from Fig. 17C can be used to derive the 3D structure of the complex. The Multiple quantum (MQ) correlation techniques are also useful to distinguish different metal hydrides on solid surfaces. For example, on treating the (^SidOd)Zr(Np)3 with H2 at 150  C, the formation of zirconium hydride was determined by solid-state 1 H MAS NMR spectrum showing two distinct hydride peaks at 10.1 and 12.1 ppm (Fig. 18A).112 Nevertheless, the number of protons present in the [ZrdHx] metal complex, were not properly quantified. However, very different longitudinal relaxation (T1) was observed for two different hydride species. A very short T1 (2 s) for a signal at 12.1 ppm was noticed, whereas it was much longer for the signal at 10.1 ppm (about 30 s), which indicates the presence of two electronically distinguished protons on Zr as [ZrHx]. The 2D 1Hd1H MQ correlation MAS ssNMR experiment was carried out to assign the structure of the complex. The 1H peak at 12.1 ppm showed a strong autocorrelation at 24.2 ppm in the o1 (DQ) dimension in the 2D 1Hd1H DQ-SQ correlation spectrum (Fig. 18B), which indicates the presence of zirconium-bis-hydride [ZrH2]. A weak correlation at 14.5 ppm in the o1 (DQ) dimension (Fig. 18B) for the protons resonating at 10.1 ppm and 4.4 ppm indicates the correlation between ZrH (10.1 ppm) with SidH (4.4 ppm) of the surface. The above findings prove that these species are very close to each other in space.

Fig. 17 (A) 1D 1H MAS ssNMR spectrum of [(^SidOd)W(CH3)5/(^SidOd)Zr(Np)3] and (B) 2D 1Hd1H DQ-SQ (C) TQ-SQ correlation ssNMR spectra of the surface complex [(^SidOd)W(dCH3)5/(^SidOd)Zr(Np)3]. Reproduced with permission from ref. Samantaray, M. K.; Dey, R.; Kavitake, S.; Abou-Hamad, E.; Bendjeriou-Sedjerari, A.; Hamieh, A.; Basset, J.-M., J. Am. Chem. Soc. 2016, 138 (27), 8595–8602. Copyright 2016 American Chemical Society.

Surface Organometallic Chemistry and Catalysis

481

Fig. 18 (A) 1H MAS ssNMR spectrum of [ZrHx] species obtained during the hydrolysis of (^SidOd)Zr(Np)3. (B) 2D 1Hd1H DQ-SQ correlation MAS spectrum of [ZrHx] species. Reproduced with permission from ref. Rataboul, F.; Baudouin, A.; Thieuleux, C.; Veyre, L.; Copéret, C.; Thivolle-Cazat, J.; Basset, J.-M.; Lesage, A.; Emsley, L., J. Am. Chem. Soc. 2004, 126 (39), 12541–12550. Copyright 2004 American Chemical Society.

14.13.2.24

195

Pt DNP SENS ssNMR

195

Recently the Pt DNP SENS ssNMR spectra for [(COD)Pt(dOSi(dOtBu)3)2] and [Pt(dOSi(OtBu)3)2(COD)]/SiO2–700149 (COD ¼ 1,5-cyclooctadiene)150 complexes grafted on the surface of partially dehydroxylated silica at 700  C acquired for the characterization. The applications of ssNMR spectroscopy in the SOMC catalysts are vast, and still, its huge potential need to be explored.

14.13.3 Alkane metathesis In 1968 Heckelsberg and Banks reported an unprecedented dual catalyst based on tungsten oxide on silica and chromia-alumina that could transform propane to heavier olefins and paraffins at 600  C.151 They observed that along with the main product, propylene, a lot of cracking products were also observed. This work was extended by Burnett and Hughes, they converted propane in the presence of a dual catalytic system where Pt/Al2O3 used as dihydrogen catalyst and WO3/Silica used as metathesis catalyst.152 They converted propane at low temperature (400  C) and named the reaction as “disproportionation of alkane”. Almost 25 years later, in 1997, Basset et al. for the first time, observed that propane could be converted to a mixture of ethane, butane, and pentane at a much lower temperature ( 150  C) in the presence of single-site silica supported [Ta]dH catalyst.153 This new reaction is named as “alkane metathesis” as it resembles olefin metathesis reaction (Eq. 1). 2Cn H2ðn + 1Þ ! Cðn-iÞ H2ðn-iÞ + 2 + Cðn + iÞ H2ðn + iÞ + 2

(1)

Where i ¼ 1, 2,. n− 1; with i ¼ 1, 2 favored for n < 4. This discovery opened up doors to synthesize a plethora of new catalysts for this type of reaction. In this section, we will be discussing on metathesis of propane, n-decane, and cyclooctane using various well-defined oxide supported catalysts. We divided our catalysts into generations based on their development. We consider the 1st generation catalysts are metal hydrides of Ta and W, the 2nd generation is Ta-methyl, and W-methyl supported over various oxides, and the 3rd generation is the combination of group IV and group VI metal alkyls (bi-metallic system).

482

Surface Organometallic Chemistry and Catalysis

14.13.3.1 Linear alkane metathesis with [W]dH and [Ta]dH (1st generation catalyst) In 1997, during a reaction of propane in the presence of [(^SidOd)2TaH], lower and higher homologs of propane were observed.153 The reaction resembles olefin metathesis, and it is called alkane metathesis. During the catalytic reaction, it was noticed that to propagate the reaction, the presence of a carbene/hydride fragment in the coordination sphere of metal is necessary. To understand the reaction mechanism of the alkane metathesis reaction, kinetic experiments of propane metathesis were carried out using a continuous flow reactor. The surface complex [(^SidOd)2TaH] was prepared from its parent complex [(^SidOd)xTa(]CHCtBu)3-x] (x ¼ 1 or 2) with hydrogen at 150  C for 15 h followed by 200  C for 2 h.154 When propane is contacted with [Ta]dH, [Ta]-n-propyl, and [Ta]-isopropyl are formed by s-bond metathesis with the evolution of molecular hydrogen. Once metal alkyl is formed by s-bond metathesis, there are two possibilities for metal to abstract H to formed metal hydride: 1. a-H abstraction leads to metal carbene-hydride 2. b- H abstraction leads to release of an olefin with the formation metal hydride. Once olefin is formed, it again reacts back with the metal-carbene-hydride fragment to propagate the reaction. At the end, the new olefins, which are generated during the catalytic cycle, get reduced to form new alkanes. The concentration of the product depends upon the approach of the olefin toward to metal-carbene-hydride fragment (Scheme 1).

Scheme 1 Metal hydrides of group IV, V, and VI are presented here. Reproduce with permission from ref. Samantaray, M. K.; Pump, E.; Bendjeriou-Sedjerari, A.; D’Elia, V.; Pelletier, J. D. A.; Guidotti, M.; Psaro, R.; Basset, J.-M., Chem. Soc. Rev. 2018, 47 (22), 8403–8437. Copyright 2018 Royal Society of Chemistry.

Surface Organometallic Chemistry and Catalysis

483

It was observed that the formation of n + 1 alkanes is higher than that of n + 2 alkanes. Similarly, the formation of n-1 alkanes is higher than that of n −2 alkanes. Additionally, it was also noticed that linear products are favored over branched ones. To understand the products trend, the flow rate of the propane was varied from 1 mL min−1 to 100 mL min−1.155 At low contact time, the formation of hydrogen and olefin increases while that of alkane decreases. Similarly, at high contact time, mainly alkanes were observed. This experiment confirms that the initial products are olefin and hydrogen in the alkane metathesis reaction (Fig. 19). With the above experimental evidence, a mechanism was drawn for alkane metathesis (Scheme 2). Reproduced with permission from ref.155 Copyright 2005 American Chemical Society.

Fig. 19 (A) Conversion and (B) selectivities obtained during propane metathesis catalyzed by [(^SidOd)2TadH], (5.33 wt% Ta) in a continuous flow reactor (150  C, 1 atm, 1 mL min−1, VHSV ¼ 38 h−1). Selectivity vs. inverse space velocity expressed in [(min)(volume of catalyst)/(volume of propane)] for (C) alkanes, (D) olefins, and (E) H2. (F) alkanes/olefins ratio vs. inverse space velocity. Reproduced with permission from ref. Basset, J. M.; Copéret, C.; Lefort, L.; Maunders, B. M.; Maury, O.; Le Roux, E.; Saggio, G.; Soignier, S.; Soulivong, D.; Sunley, G. J.; Taoufik, M.; Thivolle-Cazat, J., J. Am. Chem. Soc. 2005, 127 (24), 8604–8605. Copyright 2005 American Chemical Society.

484

Surface Organometallic Chemistry and Catalysis

Scheme 2 Mechanistic Pathway derived from kinetic and DFT calculations for the formations of linear and branched alkane. Reproduced with permission from ref. Samantaray, M. K.; D’Elia, V.; Pump, E.; Falivene, L.; Harb, M.; Ould Chikh, S.; Cavallo, L.; Basset, J.-M., Chem. Rev. 2020, 120 (2), 734–813. Copyright 2020 American Chemical Society.

From the kinetic studies and the observed products, it is suggested that olefin metathesis is part of alkane metathesis. To enhance the productivity in alkane metathesis reaction, it was shifted from group V [Ta] to group VI [W] as it is reported that [(^SidOd)W(^CtBu)(dCHt2Bu)2] is a good catalyst in olefin metathesis reaction as compared to [(^SidOd)Ta(]CHtBu) (dCH2tBu)2].156 During the reaction it was noticed that the silica-supported [(^SidOd)W(^CtBu)(dCHt2Bu)2] is not active in alkane metathesis reaction whereas a TON of 30 could be achieved with silica-alumina [(^SidOd)SiO2dAl2O3W(^CtBu) (dCHt2Bu)2] supported and alumina supported [(^AlsdOd)W(^CtBu)(dCHt2Bu)2] catalysts.157 It was known that hydrides are prone to react faster with alkanes as compared to metal-neopentyl, the corresponding [W]-hydrides of the silica-alumina supported catalyst was prepared and used for alkane metathesis reaction. As expected, a TON of 129 was observed in the propane metathesis reaction.157 As a carbene hydride is an essential part of the coordination sphere for the reaction to proceed effectively, to synthesize a metalcarbene-hydride, the authors moved from metal neopentyl to metal methyl. Two new silica-supported catalyst [(^SidOd)W(CH3)5] and [(^SidOd)Ta(CH3)4] were synthesized by the reaction of silica dehydroxilated at 700  C (SiO2–700) with the corresponding homogeneous complexes W(CH3)6 and Ta(CH3)5 (Scheme 3).139,158 The silica-supported complexes are fully characterized using advanced solid-state NMR, IR, elemental analysis and gas quantification methods. After characterization, the pre-catalyst [(^SidOd)W(CH3)5] was used for propane metathesis with a TON of 160.139 Please note that the corresponding W-neopentyl complex was inactive in propane metathesis.

Surface Organometallic Chemistry and Catalysis

485

Scheme 3 Synthesis of [(^SidOd)W(CH3)5] and [(^SidOd)Ta(CH3)4] on silica surface and their corresponding thermal treatment products. Reproduced with permission from ref. Samantaray, M. K.; Pump, E.; Bendjeriou-Sedjerari, A.; D’Elia, V.; Pelletier, J. D. A.; Guidotti, M.; Psaro, R.; Basset, J.-M., Chem. Soc. Rev. 2018, 47 (22), 8403–8437. Copyright 2018 Royal Society of Chemistry.

This activity of [(^SidOd)W(CH3)5] in propane metathesis encouraged the authors to obtain the expected metal-carbenehydride catalyst. We believed the W-carbene-hydride could be isolated in two ways: (i) by carrying out controlled hydrogenolysis in the presence of molecular hydrogen and (ii) by thermal treatment of the parent [(^SidOd)W(CH3)5] species. To obtain the required metal carbene/hydride or metal carbene/methyl to enhance the productivity of the alkane metathesis reaction, the SOMC complex [(^SidOd)W(CH3)5] reacted with hydrogen at various temperatures ranging from −78  C to 150  C. At −78  C, [(^SidOd)W(CH3)5] reacts with hydrogen to produce [(^SidOd)W(]CH2)H3] (Scheme 4).159 The formation of the carbene hydride species [(^SidOd)W(]CH2)H3] was confirmed by solid-state NMR apart from elemental analysis and IR studies. 1H NMR showed a peak at 15 ppm, which is assigned to W-hydride whereas a peak at 231 ppm in 13C NMR, which is the characteristic peak for the metal-carbene species, confirms the formation of a [W]]CH2 on the surface. Additionally, DFT studies on hydrogenolysis of [(^SidOd)W(CH3)5] confirmed that the most stable species could be [(^SidOd)W(]CH2) H3] on the surface.159

486

Surface Organometallic Chemistry and Catalysis

Scheme 4 Synthetic procedure of [(^SidOdW)(]CH2)H3] by the reaction of [(^SidOd)W(CH3)5] with molecular hydrogen at low temperature. Reproduced with permission from ref. Maity, N.; Barman, S.; Callens, E.; Samantaray, M. K.; Abou-Hamad, E.; Minenkov, Y.; D’Elia, V.; Hoffman, A. S.; Widdifield, C. M.; Cavallo, L.; Gates, B. C.; Basset, J.-M., Chem. Sci. 2016, 7 (2), 1558–1568. Copyright 2018 Royal Society of Chemistry.

We also carried out the second approach to isolate [W](]CH2)Hx on the surface. We treated [(^SidOd)W(CH3)5] from 80  C to 120  C to obtain [W](]CH2)Hx species. During our characterization in 1H NMR, we observed several peaks from 7.6 to 4.2 ppm, and in 13C, we observed a peak at 298 ppm along with other peaks. The peak at 298 ppm confirms the formation of a [W^CH] species on the surface instead of [W]CH2] (Scheme 3).139 W-carbyne species [(^SidOd)W(^CH)(CH3)2] was reacted with propane, the expectation was the reaction should not proceed. At the end of the reaction, propane metathesis products (methane, ethane, butane, pentanes, etc.) were observed. It was much unexpected to understand how a carbyne species could be active in alkane metathesis reaction as it does not possess a metal-carbene fragment which is necessary for the reaction to proceed. To understand the reaction and to isolate the reactive intermediate, the carbyne species [(^SidOd)W(^CH)(CH3)2] was reacted with either phosphine or cyclooctene (Scheme 5). After the necessary workup, the product was characterized by solid-state NMR. 13C CP/MAS solid-state NMR of the resulting species shows the disappearance of the signal at 298 ppm and the appearance of two signals at 356 ppm and 252 ppm.160 Additionally, HETCOR experiment was carried out to understand the peak belongs to which fragments. The peak at 252 ppm shows a correlation with the proton at 4.2 ppm, and the peak at 356 ppm correlates with the proton chemical shift at 7 ppm. The peak at 356 ppm is assigned to W-carbyne fragment, and the peak at 252 is assigned to W-carbene species (Scheme 5).160 This experiment confirms that during the reaction due to s-bond metathesis between propane and W-methyl, propene results and as soon as propene results it coordinates with the W-carbyne fragment and splits it into a bis carbene [(^SidOd)W(]CH2)2CH3] species which is carrying out the catalytic reaction.160

Scheme 5 Well-defined silica-supported [(^SidOdW(CH3)5)] and the expected bis-carbene species after treatment with phosphine/olefin.

Similarly, Ta(CH3)5 was grafted on silica dehydroxylated at 700  C (SiO2–700) to obtained [(^SidOd)Ta(CH3)4]. The surface complex is thoroughly characterized by solid-state NMR, IR, elemental analysis, and gas quantification methods. To generate tantalum methylidene species on the surface, the surface complex [(^SidOd)Ta(CH3)4] was thermally treated at 150  C for 4 h,158 the resulting product is analyzed by solid-state NMR. NMR studies confirm that a mixture of mono [(^SidOd)Ta(]CH2) (CH3)2], and bi-podal [(^SidOd)2Ta(]CH2)CH3] species are formed on the surface (Scheme 5).158 The surface mixture complex was tested for propane metathesis at 150  C for 5 days. At the end of the reaction, only a TON of 49 could be achieved.

Surface Organometallic Chemistry and Catalysis

487

14.13.3.2 Liquid phase alkane metathesis reaction [(^SidOd)W(CH3)5] was not only explored in the propane metathesis reaction, but it was also explored in liquid phase n-decane metathesis reaction. During the n-decane metathesis reaction, a range of products starting from C3 to C34 were observed.161 The broad range of products could be explained only by “Iso-metathesis”. Iso-metathesis is derived from alkane metathesis, normally alkane metathesis is mainly explored in low molecular weight alkane, more specifically propane, when a longer chain alkane like n-decane is used for alkane metathesis, once olefin is generated from the alkane due to b-H elimination, the presence of [W]dH on surface isomerizes the double bond which is called as chain walking of the double bond. Because of this chain walking, various olefins were formed at various intervals of time (Scheme 6). As soon as the olefin is formed, they cross-metathesize with each other in the presence of a catalyst to generate a broad range of new olefins, and at the end of the reaction, the new olefins are reduced to new alkanes (Scheme 6). n-decane metathesis using [(^SidOd)W(CH3)5] as a catalyst precursor could achieve only a TON of 150.161 To improve the TON in alkane metathesis, W(CH3)6 was grafted on silica-alumina dehydroxylated at 500  C. A mixture of mono and bi-podal supported complex was found. The silica-alumina species was thermally treated at 120  C to achieve a mixture of carbyne species. The silica-alumina species were used for n-decane metathesis with an improved TON of 350.161

Scheme 6 A schematic representation of how isomerization occurs during alkane metathesis of higher alkane and the reaction mechanism is present here. Reproduced with permission from ref. Samantaray, M. K.; Pump, E.; Bendjeriou-Sedjerari, A.; D’Elia, V.; Pelletier, J. D. A.; Guidotti, M.; Psaro, R.; Basset, J.-M., Chem. Soc. Rev. 2018, 47 (22), 8403–8437. Copyright 2018 Royal Society of Chemistry.

488

Surface Organometallic Chemistry and Catalysis

14.13.3.3 Cycloalkane metathesis So far only linear alkanes were used for metathesis reaction, and it was observed that using propane as a reactant, we could achieve a maximum TON of 264 while using [(^Si-O-)W(]CH2)H3] as a catalyst precursor, and in case of n-decane metathesis a TON of 350 with W(CH3)6 supported silica-alumina catalyst could be achieved. Macrocycles are the building blocks for pharmaceutical intermediates, fragrance, and as a biomarker for B. Braunii. To obtained macrocycles, cyclooctane was used as a starting material. It was observed that cyclooctane undergoes self-metathesis in the presence of [(^SidOd)W(CH3)5] to form a broad range of cycloalkanes without any polymeric products.162 The overall selectivity is limited to macrocyclic carbon ranging from cC12dcC40 and ring contracted products cC5dcC7, a TON of 450 is achieved after 190 h of reaction (Scheme 7). The broad range of products is because of the competition of olefin metathesis vs. double bond migration, which is called “isomet” (isomerization metathesis). It was also observed that double bond migration is much faster, leading to various cyclic olefins followed by olefin metathesis and hydrogenation of the newly formed cyclic olefins to new macrocycles and ring contracted cyclic alkanes.162

Scheme 7 Formation of ring contracted (left) and ring expanded (right) product during the cyclooctane metathesis reaction using [(^SidOd)W(CH3)5].

To improve the selectivity in cycloalkane metathesis reaction, instead of conventional silica, mesoporous silica with various pore diameters (dpore from 2.5 to 6 nm) were used as support for grafting the catalyst.163 During the catalytic test, it was observed that the selectivity of cC16 is higher while using a lower pore diameter, whereas, with a higher pore diameter, the selectivity of ring contracted products was high.163 DFT calculation was carried out to understand the selectivity of the cyclic products. The computational analysis found that double bond migration is reduced using small pore diameter silica because the intermediates and transition states are bulkier. The TONs are significantly improved in large pore diameter silica because of improved site accessibility.163

14.13.3.4 Understanding alkane metathesis with a bi-metallic catalyst Over the period of time, the TON in alkane metathesis was substantially improved; [Ta]dH ¼ 60 to [W] ¼ CH2 ¼ 260 in propane metathesis, 150 in n-decane metathesis, and 450 in cyclooctane metathesis with [(^SidOd)W(CH3)5].159,161,164 During the catalytic reaction it was understood that the critical step is the dehydrogenation of alkane to olefin which plays the important role to determine the efficiency of the alkane metathesis reaction. If the dehydrogenation step can be improved, then the reactivity can be enhanced, resulting in higher TON. To increase the efficiency of the catalyst in alkane metathesis reaction, for the first time, a bi-metallic catalyst was designed where one part of the catalyst could help for improved dehydrogenation of alkanes to olefin, and at the same time, it should not affect the olefin metathesis catalyst to carry out metathesis reaction. In the bi-metallic catalyst, group IV metals like Zr and Ti were used for the dehydrogenation, and W was used as a metathesis catalyst. At first, Zr(Np)4 and W(CH3)6 were grafted on the surface of silica sequentially to form [(^SidOd)W(CH3)5/(^SidOd)ZrNp3] (Scheme 8).148 The supported complex was characterized by solid-state NMR (Fig. 17). The NMR double-quantum (DQ) and triple quantum (TQ) study showed strong autocorrelation peaks for WdCH3 groups and [Zr] neopentyl groups, which confirms that the grafted W(CH3)6 and Zr(Np4) are in close proximity to each other on the surface. Along with the expected autocorrelation peaks, another cross-correlation peak is observed corresponds to the methyl peak of [Zr]d(CH2dC(CH3)3) and [W]dCH3 thus confirming that of [W]dCH3 and [Zr]d(CH2dC(CH3)3) are in close vicinity to each other (Fig. 17). The corresponding hydrides of [(^SidOd)W(CH3)5/(^SidOd)ZrNp3] were prepared by reacting molecular hydrogen at room temperature or by heating at 100  C for 12 to 15 h.

Surface Organometallic Chemistry and Catalysis

489

Scheme 8 Synthetic scheme of a bi-metallic [W]/[Zr] and [W/Ti] precursor catalyst. Reproduced with permission from ref. Samantaray, M. K.; Pump, E.; Bendjeriou-Sedjerari, A.; D’Elia, V.; Pelletier, J. D. A.; Guidotti, M.; Psaro, R.; Basset, J.-M., Chem. Soc. Rev. 2018, 47 (22), 8403–8437.

The bi-metallic pre-catalyst was used for n-decane metathesis with an improved TON of 1436. A broad range of products were observed as expected. To further improve the reactivity of bi-metallic catalyst, in place of Zr, Ti metal was introduced, and a new variant of the bi-metallic catalyst [(^SidOd)W(CH3)5/(^SidOdTiNp3)] was prepared.165 The bi-metallic catalyst was fully characterized by advanced solid-state NMR, IR, elemental analysis, and gas quantification method. The [W]/[Ti] bi-metallic pre-catalyst was used for propane metathesis reaction in batch reactor conditions and continuous flow reactor conditions. In batch reactor conditions, a TON of 1800 was achieved, whereas in flow reactor conditions, a TON of 10,000 was achieved. This is the highest obtained TON till date by any alkane metathesis catalyst known so far (Fig. 20).165 NMR characterization of both the [W]/[Zr] and [W]/[Ti] bi-metallic catalyst gives an impression that both the surface organometallic fragments (SOMFs) like W and Zr in the case of [W]/[Zr] and W and Ti in [W]/[Ti] bi-metallic catalysts are very close to each other. To understand the effect of togetherness of the metals known as synergistic effect of the metal fragments, three catalyst beds were prepared: (1) the [W]/[Ti] catalyst bed, (2) a bed of mechanically mixed [(^SidOd)W(CH3)5 + (^SidOd)TiNp3] and (3) a layer of [(^SidOd)W(CH3)5 and (^SidOd)TiNp3 separated by a layer of glass wool.165 In all the cases, the metal loading was kept constant. At the end of the reaction with all above mentioned catalysts, it was found that the grafted catalyst [(^SidOd)W(CH3)5/(^SidOd)TiNp3] produced a TON approximately 10,000 better than both of the mixture of

Fig. 20 Turnover number vs. time for pre-catalyst [(^SidOd)W(CH3)5]/[(^SidOd)TiNp3] (0.34 wt% W and 1.02 wt% Ti)(violet color), the physical mixture of [(^SidOd)W(CH3)5] and [(^SidOd)TiNp3] (0.27 wt% W, 1.05 wt% Ti)(green color), and a layer of [(^SidOd)W(CH3)5] (1.1 wt% Ti) followed by a layer of [(^SidOd)W(CH3)5] (0.34 wt% W) (red color). TON is expressed in moles of propane transformed per mole of W. Reproduced with permission from ref. Samantaray, M. K.; Kavitake, S.; Morlanes, N.; Abou-Hamad, E.; Hamieh, A.; Dey, R.; Basset, J.-M., J. Am. Chem. Soc. 2017, 139 (9), 3522–3527. Copyright 2017 American Chemical Society.

490

Surface Organometallic Chemistry and Catalysis

[(^SidOd)W(CH3)5 + (^SidOd)TiNp3] and the layer of [(^SidOd)W(CH3)5] and (^SidOd)TiNp3] (TONs are 5204 and 639) (Fig. 20).165 This study proved that when both the catalysts are grafted on the surface, they have very unique way of functioning, which enhances their activity in the alkane metathesis reaction.165

14.13.4 Low temperature hydrogenolysis of alkanes Hydrogenolysis of wax or polyolefins can lead to more valuable transportation range alkanes. This idea was first observed when [(^SidOd)ZrNp3] was made to react with hydrogen at 150  C, and at the end of the reaction, instead of neopentane, methane and ethane were observed.166,167 The single-site [(^SidOd)ZrNp3] was prepared by grafting ZrNp4 with silica dehydroxylated at 500  C (SiO2–500), and it was fully characterized by IR, gas quantification method, solid-state NMR, and elemental analysis.166,167 The resulting complex [(^SidOd)ZrNp3] was reacted with hydrogen at 150  C to give corresponding metal-hydride [(^SidOd)(4-x)Zr(H)x].112 The hydride complex was fully characterized by solid-state NMR apart from elemental analysis. Spectroscopic data reveals that the existence of zirconium mono hydride [(^SidOd)3ZrH] and bis-hydride [(^SidOd)2ZrH2]. Similarly, other group IV metal hydrides were prepared and used for hydrogenolysis reaction. It was observed that [(^SidOd)(4-x)Zr(H)x] and [(^SidOd)(4-x)Hf(H)x] (x ¼ 1 or 2) produce a 3:1 ratio of methane to ethane whereas [(^SidOd)(4-x)Ti(H)x] (x ¼ 1 or 2) gives 1:1 ratio of methane to ethane at end of the reaction.155,168,169 It was also observed that neither [Zr]dH/[Hf]dH nor [Ti]dH further react with ethane. From this hydrogenolysis experiment with silica-supported group IV metal hydride complexes, it was fully understood that skeletal rearrangement takes place in the case of [Ti]dH as catalyst precursor leads to 1:1 ratio of methane to ethane whereas all the above three metal hydride catalyst precursors do not break down ethane to methane. To further understand the reaction, silica-alumina supported group IV metal hydrides ([Zr]SALdHx and [Hf]SALdHx), and group V [Ta]SALdHx and group VI [W]SALdHx metal hydrides were prepared by reaction of hydrogen with their parent silica-alumina supported neopentyl complexes (Scheme 9). These catalysts were tested for hydrogenolysis of ethane to verify the

Scheme 9 Synthesis of silica-alumina supported Group IV, V, and VI metal hydride. Reproduced with permission from ref. Samantaray, M. K.; D’Elia, V.; Pump, E.; Falivene, L.; Harb, M.; Ould Chikh, S.; Cavallo, L.; Basset, J.-M., Chem. Rev. 2020, 120 (2), 734–813. Copyright 2020 American Chemical Society.

Surface Organometallic Chemistry and Catalysis

491

Fig. 21 Comparison of the reactivity of the different metal hydrides (Zr, Hf, Ta, W) supported on silica-alumina in the hydrogenolysis of ethane. Left and right parts show the conversion with respect to time on stream and the cumulated TON after 1350 min. Reproduced with permission from ref. Norsic, S.; Larabi, C.; Delgado, M.; Garron, A.; de Mallmann, A.; Santini, C.; Szeto, K. C.; Basset, J.-M.; Taoufik, M., Cat. Sci. Technol. 2012, 2 (1), 215–219. Copyright 2012 Royal Society of Chemistry.

Fig. 22 Comparison of activities of different metal hydrides (Zr, Hf, Ta, and W) supported on silica-alumina in the hydrogenolysis of n-butane. Left and right parts show the conversion with respect to time on stream and the cumulated turnover number after 2300 min, respectively. Reproduced with permission from ref. Norsic, S.; Larabi, C.; Delgado, M.; Garron, A.; de Mallmann, A.; Santini, C.; Szeto, K. C.; Basset, J.-M.; Taoufik, M., Cat. Sci. Technol. 2012, 2 (1), 215–219. Copyright 2012 Royal Society of Chemistry.

reactivity of these hydrides toward ethane. As observed earlier, [Zr]SALdHx and [Hf]SALdHx do not react with ethane, whereas [Ta]SALdHx and [W]SALdHx react with ethane to give methane.170 In the case of [Ta]SALdHx, the conversion stabilizes at 27%. However, in the case of [W]SALdHx, it stabilizes at 22% (Fig. 21), but in the case of [Zr]SALdHx and [Hf]SALdHx, the observed conversions were 0.8% and 0.07%, respectively.170 These hydride catalysts were used for the hydrogenolysis of butane. It was observed that [Zr]SALdHx is very active and converts around 98% of butane to methane, ethane, and propane, followed by [Hf]SALdHx with 76% conversion, [Ta]SALdHx 50% conversion, and [W]SALdHx 44% conversion. (Fig. 22).170 The above results told that group IV metal hydrides react differently with alkanes as comparative group V and VI hydrides during the hydrogenolysis reaction. Group IV hydrides prefer b-alkyl transfer, whereas group V and VI hydride prefer a-alkyl transfer mechanism. Based on the product obtained from hydrogenolysis of ethane and butane, a mechanism was drawn (Scheme 10).170

492

Surface Organometallic Chemistry and Catalysis

Scheme 10 Proposed mechanism for hydrogenolysis of alkanes, waxes, and polyolefins by d0 [Zr]-Hx (b-alkyl transfer) and d0 [Ta]-Hx (a-alkyl transfer). Reproduced with permission from ref. Samantaray, M. K.; Pump, E.; Bendjeriou-Sedjerari, A.; D’Elia, V.; Pelletier, J. D. A.; Guidotti, M.; Psaro, R.; Basset, J.-M., Chem. Soc. Rev. 2018, 47 (22), 8403–8437.

Recently, a Ti-methyl complex [(^SidOd)Ti(CH3)3] was synthesized and was tested for hydrogenolysis reaction. The surface complex [(^SidOd)Ti(CH3)3] was synthesized by the reaction of silica de-hydroxylated at 700  C and Ti(CH3)4 at −60  C (Scheme 11). The surface complex was fully characterized using advanced solid-state NMR, IR, gas quantification methods.171

Scheme 11 Grafting of Ti(CH3)4 on silica dehydroxylated at 700  C. Reproduced with permission from ref. Saidi, A.; Al Maksoud, W.; Samantaray, M. K.; Abou-Hamad, E.; Basset, J.-M., Chem. Commun., 2020, 56 (87): 13401–13404. Copyright 2020 Royal Society of Chemistry.

After characterization of the surface complex [(^SidOd)Ti(CH3)3], it was tested for hydrogenolysis of propane and n-butane. During hydrogenolysis reaction, it was observed that when propane is used as a reactant, products selectivity’s are almost similar (50: 50, methane to ethane), whereas, in the case of n-butane methane, ethane and propane were observed with almost similar selectivities (Fig. 23).171

14.13.5 Cross-metathesis of alkanes Cross metathesis of olefins is well-known in the literature, but cross-metathesis of alkanes is very rare because of the low reactivity as well as the difficulty to distinguish the cross-metathesis product from the alkane metathesis product. For the first time, a cross-metathesis reaction was carried out using propane and n-decane as reactants. To understand the cross-metathesis product, it was necessary to label the alkanes taking part in the cross-metathesis reaction. During the cross-metathesis reaction of propane and n-decane, propane was chosen to be labeled.172 The idea to select 13C labeled propane was to distinguish cross-metathesis products from homo metathesis products. A previous study revealed that during the self-metathesis reaction of propane, the product’s length should not go beyond C5 (pentane). If during cross-metathesis reaction, 13C inducted alkane is longer than C5 could observe, it would be confirmed that it is coming from cross-metathesis of 13C labeled propane and n-decane.172 A mixture of 13C n-propane and n-decane (non-labeled) with a molar ratio of C3 to C10 ¼ 2.25 was allowed to react in the presence of [(^SidOd)W(CH3)2(H)3] at 150  C for 5 days. At the end of the reaction, the products were analyzed by GC/GC–MS. The mass spectra of the products were taken and compared with the control (propane non labeled reacted with n-decane). A clear enrichment of 13C is observed for all the hydrocarbons in the range C4dC19, indicating that a cross-metathesis reaction takes place between 13C n-propane and n-decane (Scheme 12).172

Surface Organometallic Chemistry and Catalysis

493

Fig. 23 Evolution of conversion (blue curve) and corresponding TON over time (red curve) for hydrogenolysis reaction of (A) propane (B) n-butane using 29. The product selectivities over time for hydrogenolysis reaction of (C) propane (D) n-butane using [(^SidOd)Ti(CH3)3]. A TON of 419 was observed in the case of propane and 578 in the case of n-butane as reactants. Reproduced with permission from ref. Saidi, A.; Al Maksoud, W.; Samantaray, M. K.; Abou-Hamad, E.; Basset, J.-M., Chem. Commun. 2020, 56 (87): 13401–13404. Copyright 2020 Royal Society of Chemistry.

Scheme 12 Three stages of cross-metathesis represent here. Reproduced with permission from ref. Morlanés, N.; Kavitake, S. G.; Rosenfeld, D. C.; Basset, J.-M., ACS Catal. 2019, 9 (2), 1274–1282. Copyright 2019 American Chemical Society.

A mechanism was proposed for this reaction. At first, CdH bond activation takes place (s-bond metathesis), followed by b-H elimination with the liberation of olefin. This olefin could react back with the W-carbene present in the complex (formed by a-H abstraction) to form a metallacycle. Once a metallacycle is formed, it undergoes metathetical cleavage to generate a new olefin; this process goes on and on; at the end of the reaction, the new olefins get reduced, and new alkanes are formed. This cross-metathesis reaction between labeled n-propane and n-decane is believed to occur in three catalytic cycles. Self-metathesis of n-decane (blue cycle), self-metathesis of propane (red cycle), and cross-metathesis between propane and n-decane (green cycle).172 For the later, several possible cross-metathesis steps between the alkylidene species from the other alkane with the olefin intermediate from the other alkane may occur during the olefin metathesis step (Scheme 13).

494

Surface Organometallic Chemistry and Catalysis

Scheme 13 Proposed alkane cross-metathesis between propane and n-decane. The cross-metathesis of propane and n-decane is also dependent on the ratio of propane to n-decane. Reproduced with permission from ref. Morlanés, N.; Kavitake, S. G.; Rosenfeld, D. C.; Basset, J.-M., ACS Catal., 2019, 9 (2), 1274–1282. Copyright 2019 American Chemical Society.

Surface Organometallic Chemistry and Catalysis

495

It was reported that the percentage of cross-metathesis increases with C3/C10 ratio from 1 to 10. A maximum cross-metathesis product of around 50% was obtained when C3/C10 ¼ 10. With further increase in the ratio, self-metathesis increases instead of cross-metathesis.

14.13.6 Imine metathesis Imine metathesis reaction is a reaction where two different imines react in the presence of a catalyst to give a statistical mixture of all possible ¼ NR exchange products (Scheme 14).

Scheme 14 A general method for imine metathesis in the presence of a catalyst.

Imine metathesis is well-known in literature with homogeneous catalysts, the mechanism of this reaction proceeds by two separate pathways. Bergman and co-workers established a Chauvin type mechanism based on isolated diazametallacyclic intermediate in the reaction of CpCp’Zr(]NR)THF(Cp’]Cp, Cp ) with imines, whereas kinetic studies by Mountford and co-workers revealed that an amine-catalyzed process to explain the reaction of (py)3Cl2Ti(]NR) with imines. Though imine metathesis was a well-established reaction in the homogeneous condition, it was not reported in heterogeneous catalysis. For the first time, a silica-supported SOMC based Zr complex was synthesized and used for imine metathesis reaction. The surface complex [(^SidOd)Zr(dNEt2)3] was prepared by the reaction of silica dehydroxylated at 700  C with Zr(dNEt2)4 (Scheme 15).173 During thermal evacuation at 200  C, a new imido complex is generated [(^SidOd)Zr(]NEt) (dNEt2)]. These two complexes were fully characterized by solid-state NMR, IR, and elemental analysis. The silica-supported complex [(^SidOd)Zr(]NEt)(dNEt2)] was thermally reacted with benzyl amine in toluene at 80  C to produce [(((^SidOd) Zr)dNEt2)(]NPh)].173 The new surface complex [(^SidOd)Zr(NEt2)(]NPh)] was fully characterized by solid-state NMR, IR, and elemental analysis. Similarly, a hafnium imido catalyst [(^SidOd)Hf(]NMe)(NMe2)] was prepared and used for imine metathesis reaction (Scheme 15).127 During the catalytic reaction, it was observed that Hf surface species are more active than [Zr] surface species. Based on the isolation of the intermediate, a detailed mechanism was drawn (Scheme 16) and also calculated by DFT.127

Scheme 15 Synthesis of surface [Zr] and [Hf] imido complex on SiO2–700. Reproduced with permission from ref. Samantaray, M. K.; D’Elia, V.; Pump, E.; Falivene, L.; Harb, M.; Ould Chikh, S.; Cavallo, L.; Basset, J.-M., Chem. Rev. 2020, 120 (2), 734–813. Copyright 2018 American Chemical Society and Copyright 2016 Royal Society of Chemistry.

496

Surface Organometallic Chemistry and Catalysis

Scheme 16 Imine metathesis mechanism carried out by surface [M]NR2]. Reproduced with permission from ref. Samantaray, M. K.; D’Elia, V.; Pump, E.; Falivene, L.; Harb, M.; Ould Chikh, S.; Cavallo, L.; Basset, J.-M., Chem. Rev. 2020, 120 (2), 734–813. Copyright 2020 American Chemical Society.

14.13.7 Hydroamination reaction Hydroamination reaction is when an NdH bond is added over CdC multiple bonds to give nitrogen-containing cyclic or acyclic compounds, which can be very useful as bulk or specialty chemicals (Scheme 17).

Scheme 17 A schematic representation of hydroamination reaction.

Though this reaction is well documented in homogeneous catalysis, a few supported catalysts have been explored in this reaction. For this reaction, a supported tantalum complex [(^SidOd)Ta(1s-NEtMe)2(]NtBu)] was chosen. The surface tantalum complex was synthesized by the reaction of silica partially dehydroxylated at 700  C (SiO2–700) with [Ta(1sNEtMe)3(]NtBu)] in pentane at room temperature.174 The surface complex [(^SidOd)Ta(1sNEtMe)2(]NtBu)] was fully characterized by solid-state NMR, IR, and elemental analysis and was further confirmed by EXAFS. After the full characterization of the surface complex [(^SidOd)Ta(1sNEtMe)2(]NtBu)], it was tested for hydroamination reaction with a large range of substrate. With 1-octyne as an unsaturated carbon source and aniline as an amine source. The reaction was carried out at 80  C for 16 h in the presence of tantalum surface complex.174 At the end of the reaction, the product was analyzed by GC/GC–MS. Both Markovnikov and anti-Markovnikov products were observed. Usually, aliphatic alkynes lead to an anti-Markovnikov product. Though a range of alkyne and amine sources were used for this reaction, the best conversion comes from the reaction of 1-octyne and aniline (63% conversion).174 It was important to propose a mechanism for this reaction and, most importantly, to draw a mechanism by isolating the reactive intermediates. To isolate intermediates, the catalyst [(^SidOd)Ta(1s-NEtMe)2(]NtBu)] was reacted with aniline after washing off the unreacted product and followed by vacuum treatment, the imine adduct tantalum intermediate [(^SidOd)Ta(1s-NEtMe)2(NHtBu)(NHC6H5)] was isolated and was fully characterized by solid-state NMR. Similarly, 1-octyne was reacted with the surface tantalum complex [(^SidOd)Ta(1s-NEtMe)2(]NtBu)] at 80  C in a batch reactor for 16 h.174 The resulting complex was washed thoroughly and was characterized by solid-state NMR to confirm the structure to be [(^SidOd)Ta(1s-NEtMe)2(2NtBuCH]C7H13)]. Based on the above experiment a mechanism was drawn (Scheme 18).

Surface Organometallic Chemistry and Catalysis

497

Scheme 18 Catalytic cycle for propyne hydroamination with aniline in the presence of the supported tantalum complex. Reproduced with permission from ref. Aljuhani, M. A.; Zhang, Z.; Barman, S.; El Eter, M.; Failvene, L.; Ould-Chikh, S.; Guan, E.; Abou-Hamad, E.; Emwas, A.-H.; Pelletier, J. D. A.; Gates, B. C.; Cavallo, L.; Basset, J.-M., ACS Catal., 2019, 9 (9), 8719–8725. Copyright 2019 American Chemical Society.

14.13.8 Hydrometathesis of olefins During the alkane metathesis reaction, it was observed that with changing reaction conditions, using new catalysts (single metal catalysts) does not affect the productivity of the reaction in terms of TON appreciably. It was also observed that catalysts are not able to dehydrogenate alkane efficiently, which affects the olefin metathesis step; eventually, the productivity of the reaction is hard to understand (please note that alkane metathesis comprises three steps dehydrogenation, olefin metathesis, and hydrogenation). To increase the productivity of the reaction, Basset et al. for the first time used a mixture of olefin and hydrogen in the presence of [Ta]Hx and named the reaction as hydro-metathesis (Eq. 2).175 2ðCn H2n + H2 Þ Ð Cn − i H2ðn − iÞ + 2 + Cn + 1 H2ðn + iÞ + 2 + Cn H2n + 2 n  2 i ¼ 1, 2:: . . . ðn − 1Þ

(2)

The first catalyst for this reaction was a surface [Ta]dH supported on KCC-1. To understand the effectiveness of the hydrometathesis reaction, a 1:1 mixture of propane and hydrogen was allowed to react in the presence of [Ta]dH in a continuous flow reactor at 150  C.175 Propene was chosen as reactant because one could directly compare the results with propane metathesis reaction. The reaction proceeds efficiently with a conversion of 40%. The major products were butane and ethane, apart from minor products like isobutane, pentanes, and methane. The maximum TON was 768, which is far more as compared to alkane metathesis of propane (TON 60).175 Along with hydro-metathesis products, an appreciable amount of propane was also observed. Propane comes from the reduction of propene in the presence of hydrogen. Hydro-metathesis reaction was explored using surface W catalyst. Silica and silica-alumina supported W surface complexes were used for hydro-metathesis reaction to explore other metal complexes on hydro-metathesis as well as to understand the effect of support on this catalysis.

498

Surface Organometallic Chemistry and Catalysis

30

30

24C

Propene:H2=1:1

Propene:H2=1:2

25

12A

12 C 20

0.20

0.10

15

Alkanes distribution

C9H20

C8H18

C8H16

C9H20

C8H16

C7H14

C9H20

C9H20

C8H18

C7H16

C6H14

C5H12

n-C4H10

iso-C4H10

C2H6

C10H22

C9H20

C8H18

C7H16

C6H14

C5H12

n-C4H10

0

iso-C4H10

0

CH4

C10H22

C9H20

C8H18

C7H16

iso-C4H10

10

5

C2H6

C7H16

0.05

5

CH4

0.15

0.00

CH4

10

0.20

0.10

0.05 0.00

C8H18

C6H12 0.25

mmol

mmol

mmol

mmol

15

0.15

C7H14

0.30

0.25 0.20

C7H16

C6H12

C5H10

12D

20

0.30

C6H14

C4H8

n-C4H10

Propene:H2=1:2

C8H18

12B

Product distribution

25

Propene:H2=1:1

20

24D

iso-C4H10

C10H22

C9H20

C8H16

C7H14

C7H16

C6H12

C6H14

C5H10

C4H8

C5H12

C8H18

30

Product distribution

C7H16

25

n-C4H10

C3H8

iso-C4H10

C2H4

CH4

0

C2H6

0

C2H4

10

5

24B

0.10 0.05

5

30

0.15

0.00

C9H20

C8H18

C8H16

C7H14

10

C7H16

0.00

iso-C4H10

0.05

15

C3H8

0.10

CH4

0.15

mmol

0.25

0.20

C2H6

15

0.30

0.25

mmol

mmol

0.30

CH4

mmol

20

C5H12

24A 25

Alkanes distribution

Fig. 24 24A and 24C represent the total distribution of products olefins ( ), alkanes ( ), and propane ( ) when the ratio of propene to hydrogen 1:1.for 24A and 1:2 for 24C, 24B and 24D represents only alkanes ( ) except propene ( ) when propene to hydrogen 1:1 for 24B and 1:2 for 24D in a continuous flow reactor for hydro-metathesis reaction catalyzed by W/silica. Reproduced with permission from ref. Tretiakov, M.; Samantaray, M. K.; Saidi, A.; Basset, J. M., J. Organomet. Chem. 2018, 863, 102–108. Copyright 2018 Elsevier.

The catalytic reaction with W/silica with propene to hydrogen ratio of 1:1 at 150  C for 24 h yielded a 51% conversion with a cumulative TON of 1635.176 Propane was also observed coming from the reduction of propene in the presence of hydrogen and W catalyst, along with other desired products. It was also observed that along with the hydro-metathesis product, some olefins were also observed (C2, C4, and C5 olefins). To obtain a good TON, the propane to hydrogen ratio was increased from 1:1 to 1:2. As expected, the TON increased from 1635 to 2104. With a further increase in the ratio of propene to hydrogen, the TONs did not increase appreciably (Fig. 24).176 It is reported that W grafted on silica-alumina is very stable as compared to W on silica. During alkane metathesis reaction, it was observed that W grafted on silica-alumina is more active than W grafted on silica. To understand the effect of support (silica-alumina versus silica) on hydro-metathesis reaction, W grafted on silica alumina was prepared and used in hydro-metathesis reaction. In a continuous flow reactor, a 1:1 ratio and 1:2 ratio of propene to hydrogen were taken and heated at 150  C in the presence of W/silica-alumina catalyst. As expected, a TON of 3279 was obtained for propene to hydrogen ratio 1:1, and a TON of 4577 was obtained for a 1:2 ratio.176 It was also observed that in all the cases, whether it is W/silica or W/silica-alumina, the TONs are much better as compared to Ta/silica.

Surface Organometallic Chemistry and Catalysis

499

Furthermore, the hydro-metathesis reaction expanded from the gas to the liquid phase. This time n-decene was used as a reactant with three different catalysts supported on silica [(^SidOd)W(CH3)5], [(^SidOd)Ta(]CHtBu)(CHt2Bu)2], and [(^SidOd)Mo(^CtBu)(CHt2Bu)2].177 In a typical reaction in a batch reactor, 40 mg of catalyst precursor was taken, and 1 mL of dried 1-decene was added into it inside a glovebox. The containers were cooled in liquid nitrogen, degassed under vacuum, and 0.8 mbar of dried hydrogen was introduced into it. The reactors were placed in an oil bath at 150  C for 3 days. At the end of the reaction, the reactors were cooled under liquid nitrogen and were quenched with a fixed amount of dichloromethane. The reaction mixture was filtered and analyzed by GC/GC–MS. A wide range of products was obtained (C3–C34 alkanes) along with a few unreacted olefins. With [(^SidOd)W(CH3)5] a TON of 818 was observed, whereas with [(^SidOd)Ta(]CHtBu)(CHt2Bu)2], and [(^SidOd)Mo(^CtBu)(CHt2Bu)2] TON of 334 and 808 were obtained respectively (Fig. 24).177 To understand the reaction mechanism of 1-decene hydro-metathesis reaction, olefin metathesis was carried out. As expected, all the catalysts give a good conversion of olefins with a range of products from C3 to C22. This reaction ruled out that hydro-metathesis reaction proceeds with olefin metathesis followed by reduction of new olefins to alkanes as products were observed greater that C22 alkanes. The long-range alkanes can only be produced if alkane metathesis is also happening during the hydro-metathesis reaction. At the end, it was concluded that hydro-metathesis reaction is a combination of olefin metathesis, alkane metathesis, and reduction of the newly formed olefins to new alkanes. Based on the study, a mechanism was proposed (Scheme 19).

Scheme 19 Proposed catalytic mechanism of 1-decene hydro-metathesis using [M-catalyst](M]W, Mo, and Ta). Reproduced with permission from ref. Saidi, A.; Samantaray, M. K.; Tretiakov, M.; Kavitake, S.; Basset, J. M., ChemCatChem 2018, 10 (8), 1912–1917. Copyright 2018 Wiley.

14.13.9 Catalytic reduction of N2 toward NH3 The ammonia synthesis is one of the largest-volume industrial chemicals synthesis in the world, owing to the use of NH3 as the source of most fertilizers for agriculture, various nitrogen-containing chemicals and materials, nitric acid, clean-burning fuels, refrigerant fluid, and H2 storage and distribution as a significant energy carrier of the future.178–183 There is a strong correlation between the increase of the human population in the early 20th century to nowadays and the demand for worldwide ammonia production (Fig. 25).184,187 In this context, the global output in NH3 reached 0.175 Gt in 2016 and is expected to grow by over 5% between 2020 and 2025. However, it is significant to note that the amount of CO2 equivalents released into the atmosphere from the production of ammonia is around 400 Mt./year with 12 Gt/year of CO2 equivalents produced from the entire ammonia food chain. This carbon footprint represents approximately 1.5% of all greenhouse global gas emissions.180,185,186,188 Today, the Haber-Bosh process is considered the main way to produce ammonia industrially, departing from natural gas or other fossil fuel. Earlier in the 20th century, Osmium or Uranium were the first industrial catalysts to achieve a high conversion of nitrogen at high pressure and high temperature.189 However, due to the expensive prices of these metals, a cheaper and abundant alternative needed to be implemented. In this context, a highly effective ammonia catalyst was finally developed early in 1910. A

500

Surface Organometallic Chemistry and Catalysis

Fig. 25 (A) Trends in the human population and nitrogen use throughout the twentieth century. (B) Annual global CO2 emissions from fossil fuel, gas flaring, and cement production since the industrial revolution from 1870 to 2014. Reproduced with permission from ref. Erisman, J. W.; Sutton, M. A.; Galloway, J.; Klimont, Z.; Winiwarter, W., Nat. Geosci. 2008, 1, 636. Mebrahtu, C.; Krebs, F.; Abate, S.; Perathoner, S.; Centi, G.; Palkovits, R., CO2 Stud. Surf. Sci. Catal, Albonetti, S.; Perathoner, S.; Quadrelli, E. A., Eds. Elsevier: 2019; Vol. 178, 85–103. Copyright 2019 Elsevier.

pure magnetite iron (Fe3O4) promoted by fusing it with K2O and Al2O3 was prepared.189,190 This particular synthesis leads to the resulting a-Fe phase’s specific surface properties, which is commonly termed “ammonia iron” and is still widely used in industry today. Two other catalytic processes were also actively used for N2 fixation, as it is shown in Scheme 20: - The biological system using the nitrogenase enzyme as a catalyst for ammonia production.191 - The well-defined homogeneous catalyst based on a molecular monometallic molybdenum complex was also used to produce ammonia.192 Additionally, these systems used different sources of hydrogen (gas phase, NADH, and inorganic acids in solution) and different cofactors (inorganic promoters, coenzymes, and organometallic redox system) to catalyze the stepwise hydrogenation of N2 to ammonia, which is unexpectedly different from a catalytic system to others.

Surface Organometallic Chemistry and Catalysis

501

Scheme 20 Simplified proposed reaction mechanisms for catalytic ammonia synthesis in (A) heterogeneous catalysis by surface science studies on Haber-Bosh model systems; (B) biocatalysis in studies on FeMoco nitrogenase enzymes; (C) homogeneous catalysis by the organometallic molecule Mo[N(CH2CH2N(HIPT))3]. Reproduced with permission from ref. Avenier, P.; Taoufik, M.; Lesage, A.; Solans-Monfort, X.; Baudouin, A.; de Mallmann, A.; Veyre, L.; Basset, J.-M.; Eisenstein, O.; Emsley, L.; Quadrelli, E. A., Science 2007, 317 (5841), 1056–1060.

As described previously, the ammonia synthesis reaction by the Haber-Bosch process, using the iron-based catalyst, requires very harsh conditions: high temperatures and pressures varying from 300 to 525  C and 10 to 80 MPa, respectively due to thermodynamic limitation. Taking the whole process into consideration (catalyst based on iron, which is widely abundant and cheap material, significant lifetime, the energy needed, and CO2 production), the search for new catalytic material and/or processes for ammonia synthesis is an area of special interest. In the past few years, there was so strong driver on developing an alternative process with enhanced sustainability, such as low-temperature and localized production of NH3 synthesis. Following this strategy, a different system for ammonia synthesis has been investigated using: photocatalysts method, plasma, electrochemical process, and organometallic complexes.192,194–196 To date, while some of these systems showed a successful performance in ammonia synthesis even at low temperatures nevertheless, the efficiency of such strategies is still limited for industrial application. Given the fact that thermodynamically ammonia synthesis is favored by the lower reaction temperature, relatively good performances prepared catalysts based on Ruthenium, Nickel, Cobalt, and/or Molybdenum have been reported in the literature.197–206 Furthermore, in order to boost the performance of the traditional Ru or Fe-based catalysts, the addition of promoters such as alkali and alkaline earth metal oxides (Cs2+xO, K2O, BaOx, etc.) is primordial.207–212 The alkali promotion is commonly assumed to proceed via “electron transfer” from the promoters to the antibonding bonds of N2 through transition metals (Fe, Ru, Co, and Co3Mo3N) to lower the dissociation energy of the N^N bond, which is known as the rate-limiting step in ammonia synthesis. 179,193,210,213–218 Recently, new results obtained with catalysts derived from calcination of Fe phthalocyanine doped with K2O indicated that K2O is just selectively poisoning the surface of Fe without any strong modification of the electronic state of Fe. Even though dissociated forms of N2 are likely formed on the surface during ammonia synthesis, their role in the mechanism of ammonia synthesis remains under investigation. More and more DFT methods have proposed a non-dissociative mechanism 193,205,219–222 for ammonia synthesis relatively with the dissociative mechanism. Furthermore, based on this hypothesis, the catalysts prepared by surface organometallic chemistry methodology (SOMC) confirm for the first time that the partial hydrogenation of N2 to amido ligand could occur on an isolated surface tantalum atom supported on silica under low pressure of dinitrogen and dihydrogen and at moderate temperature (250  C) (Scheme 21 A). Additionally, a theoretical study was inspired by this discovery to elucidate the mechanism of N2 partial hydrogenation on the single Ta atom (Scheme 21 B).115,193,221,223,224

502

Surface Organometallic Chemistry and Catalysis

Scheme 21 (A) Reaction routes for the synthesis of the silica-supported tantalum hydrides and the final silica-supported [(^SiO)2Ta(]NH)(NH2)]. (B) Computed (DFT) Gibbs energy profile for the reduction of N2 by H2 at the single surface Ta(III) and Ta(V) centers with [silica_cluster_model] ¼ Si2O7H4 (by Eisenstein et al., black line) and Si14O20H18 (by Li, red line). Novel routes proposed: (1) direct protonation routes from coordinated H2 for first hydride transfer and second hydride transfer to cis-diazenido (blue); (2) alternative H2 catalyzed third hydride transfer (green). Reproduced with permission from ref. Jia, H.-P.; Quadrelli, E. A., Chem. Soc. Rev. 2014, 43 (2), 547–564.

The silica-grafted tantalum hydrides [(^SiO)2TaH] 3a and [(^SiO)2TaH3] 3b are formed by hydrogenolysis of grafted [(^SiOd)Ta(dCH2CMe3)2(]CHCMe3)] 5 with dihydrogen at 150  C, where hydrides 3a and 3b can partly interconvert depending on H2 pressure. The main product of the reaction with N2 and H2, at 250  C and 0.5 bar, is the silica-supported Ta(v) amido imido surface complex [(^SiOd)2Ta(]NH)(dNH2)] 4, as characterized by IR, solid-state NMR, elemental analysis and EXAFS.193 The silica-supported Ta(v) amido imido surface complex 4 was also observed by direct room temperature NdH bond activation of ammonia with the same starting tantalum hydrides.115 But no catalytic ammonia production can be observed from this specific system since the imido amido is thermodynamically too stable to release ammonia. As for the Ta-hydride complex, Shima et al. recently demonstrated the reactivity for H2 and N2 with the polynuclear Ti complex.225 The initial

Surface Organometallic Chemistry and Catalysis

503

(CpMe4SiMe3)Ti(dCH2SiMe3)3 complex reacts with H2 gas to form a capped trinuclear titanium cluster, and further reaction with N2 gas yields a nitride/imido/hydrido complex. While this is an extraordinary example of imido group formation from N2 and H2 gas, the authors claim in this study, similar to the Ta-hydride complex, that no NH3 is formed. 181,226,227 Furthermore, this discovery was later supported by DFT calculations which were performed on the cluster models [m-O[(HO)2SiO]2]TaH1 and {m-O [(HO)2SiO]2}TaH3 (Scheme 21) (calculated DG for complex 35 formations form N2 ¼ − 524.7 and −338.5 kJ mol−1 for the models of tantalum monohydride (7) and tris-hydride (8), respectively).221,224 The whole calculations confirmed that the direct hydride transfer to nitrogen-based ligands in [(^SiO)2TaHd(Z2-N2)] III and [(^SiO)2TaH(Z2-HNNH)] VI is extremely energetically demanding (Scheme 21 B). This very large activation energy is consistent with the lack of energetically accessible empty orbitals in the negatively charged coordinated dinitrogen and diazenido ligands for the incoming hydride. The proposed alternative steps involve tantalum coordinated dihydrogen adducts and avoid direct hydride transfers. The novel steps consist of proton transfer from coordinated dihydrogen, dihydride reductive coupling to yield coordinated H2 and H2 heterolytic splitting across the TadN bond. Such alternative elementary steps systematically provide more energetically accessible pathways than the mechanism involving direct hydride transfer to N2Hx moieties and better match the available experimental evidence. Furthermore, a very recent study appeared on the catalytic hydrogenation of N2 to ammonia-based on molybdenum hydride grafted on silica and prepared by SOMC methodology (Scheme 22). 228 The authors demonstrated, contrarily to the example with the single atom TadHx grafted on silica described before,193 that the ammonia was produced at a good rate, especially with the temperature increase from 300  C to 550  C and pressure from 0.1 to −1   3 MPa. The rate of ammonia produced increased from 1276 m mol h−1 g −1 g −1 Mo at 300 C to reach 10.000 m mol h Mo at 550 C. However, one has to take these results with caution regarding the structure of the Mo hydride. It has to be checked that Mo remains as a single atom during treatment and or catalysis.228

Scheme 22 Schematic experimental procedure for the synthesis of [(^SidOd)Mo(^CdtBu)(Np)2] and proposed mechanism for N2 into-NH3 conversion by chemical hydrogenation under 400  C and 1 atm. Reproduced with permission from ref. Azofra, L. M.; Morlanes, N.; Poater, A.; Samantaray, M. K.; Vidjayacoumar, B.; Albahily, K.; Cavallo, L.; Basset, J.-M., Angew. Chem. Int. Ed. 2019, 58 (20), 6476.

In parallel, DFT calculations were carried out on the monopodal and bipodal Mo-hydride complexes inspired by the mechanism of hydrogenation of N2 on Ta-hydride supported on silica.193 Additionally, the rate-determinate step (RDS) in the case of Mo-hydride is the transfer of the fourth H from [Mo]@ NHNH2 to give [Mo]]NH and release of one NH3 molecule (Fig. 26).228,229 Those works were essential to understand the chemistry of N2 activation and ammonia production by using well-defined grafted tantalum and molybdenum hydride complexes (single site) and further opened a loophole to build a reasonable pathway of N2 activation and hydrogenation toward ammonia production.

504

Surface Organometallic Chemistry and Catalysis

Fig. 26 Gibbs free energy profile (room temperature, in kcal mol−1), for N2-into-NH3 reaction mechanism catalyzed by monopodal [(^SidOd)Mo(IV)H3]. Si6O10H5 cluster has been used as the silica model. The Mo atom was assumed to switch between MoIV and MoVI oxidation states during the catalytic steps. Energy results are shown at the M06/TZVP//PBE/SVP(Si, O, and H)/TZVP(Mo) computational level; O red, Si beige, Mo light blue, N dark blue, H white. Ref.228

Furthermore, recently Tanaka and al. have published two examples of the catalytic formation using a dinitrogen-bridged dimolybdenum complex bearing a pincer ligand using ambient conditions (temperature ¼ 25  C and atmospheric pressure).230,231 An important theoretical study has explored this work and showed a new reaction pathway where the binuclear structure of the dinitrogen bridged dimolybdenum-dinitrogen complex plays decisive roles in exhibiting catalytic ability for the transformation of molecular dinitrogen into ammonia. The synergy between the two molybdenum moieties connected with a bridging dinitrogen ligand has been observed at the protonation of the coordinated dinitrogen ligand (Scheme 23).

Scheme 23 A possible reaction pathway: (A) Protonation of a terminal dinitrogen ligand in I followed by an exchange of the dinitrogen ligand trans to the NNH group for OTf group. Protons and electrons are supplied by lutidinium and cobaltocene, respectively. Energy changes and activation energies (in parenthesis) for individual reaction steps were calculated at the B3LYP /BS2 level of theory (units in kcal mol−1). NB means that the corresponding reaction has no activation barrier. (B) A sequential protonation/reduction of IV and separation of bimetallic complexes leading to the formation of ammonia and the monometallic nitride complex XI. Reproduced with permission from ref. Tanaka, H.; Arashiba, K.; Kuriyama, S.; Sasada, A.; Nakajima, K.; Yoshizawa, K.; Nishibayashi, Y., Nat. Commun. 2014, 5 (1), 3737. Copyright 2014 Spring Nature.

Surface Organometallic Chemistry and Catalysis

505

Similar to the SOMC approach for the N2 activation, an alternative approach was also explored in the literature using single-atom catalysis (SAC). This approach is considered as a new discipline of heterogeneous catalysis. In this approach, a single atom at the surface of a solid is permitted to activate various substrates to produce desirable products. Various studies are based on different transition metals such as Fe, Ru, or Co.86 SAC is considered as a particular case of SOMC and SOMF. SOMC and SOMF enter into the detail of the bonding between the metal atom and the surface, which is behaving as an X or L ligand in the MLH formalism. In this context, a recent paper reported by Wand et al.205 in which the authors claim that the development of nitrogen-anchored Co single-atom catalyst (“CodNdC”) in which the active metal is exclusively dispersed as single atoms. This approach could provide a good entry to make N2 dissociation impossible, while N2 stepwise hydrogenation is more feasible. It was also reported that the coordination of transition metals with N could result in dramatic modification of surface electronic properties by electron transfer to the antibonding p-orbitals of N2. Furthermore, the presence of surface N defects on cobalt molybdenum catalysts could enhance NH3 synthesis activity in mild conditions.201,205,232 The authors show a nitrogen-anchored Co single-atom catalyst (named CodNdC) with dual active sites for efficient and stable ammonia synthesis catalyst at mild condition (1 MPa, and −1 350  C) with a high rate of ammonia, which is around 116.35 mmolNH3.g−1 Co.h . However, to predict the structure and the nature of the active sites of the CodNdC single-atom catalyst, the authors use very developed analytical tools such as high-angle annular dark-field (HAADF) imaging of aberration-corrected scanning transmission electron microscopy (AC-STEM), operando X-ray absorption spectroscopy (XAS), and in situ X-ray photoemission spectroscopy (XPS), together with 15N2 isotopic-labeling experiments. Furthermore, the DFT calculations were performed to understand the pathway of NH3 synthesis over CodNdC catalyst. The authors show the good activity is related to the single Co sites in the form of steady-state Co1dN3.5 and single Co sites in the form of dynamic CodNx (0 < x  1.5). That enables N2 adsorption and hydrogenation as well as the subsequent formation of NH3 via the breaking of  NH2dNH4 bond following the Eley-Rideal mechanism,205,232,233 (Scheme 24) while the latter afford pyridinic surface N to anchor single Co atoms for NH3 production via chemical looping. The results demonstrate that the dual active sites release NH3 synthesis from the bottleneck of N^N dissociation, leading to superior NH3 production under mild conditions.

Scheme 24 Theoretical investigation and reaction pathway. (A) Changes of free energy for the formation of NH3 on CodNdC. (B) NH3 synthesis pathway on single Co sites in the form of steady-state Co1dN3.5 (A represents the active sites). (C) NH3 production on dynamic cyclic sites via chemical-looping pathway (x is in the range of 0 < x  1.5 and V∗ N stands for an anionic nitrogen vacancy; herein CoN6/C only represents the molar ratio of each element in the “CodNdC” catalyst). Reproduced with permission from ref. Wang, X.; Peng, X.; Chen, W.; Liu, G.; Zheng, A.; Zheng, L.; Ni, J.; Au, C.-T.; Jiang, L., Nat. Commun. 2020, 11 (1), 653. Copyright 2020 Spring Nature.

506

Surface Organometallic Chemistry and Catalysis

Fig. 27 Homogeneous and heterogeneous Fe3 cluster with N2 adsorption. (A) Schematic representation of N2 coordinated with three Fe(I)-ion homogeneous complexes in the side-on/end-on/end-on (m3 − Z2:Z1Z1) configuration; (B): schematic representation of N2 coordinated with heterogeneous Fe3/y−Al2O3(010) in the same configuration; (C): optimized Fe3 cluster on y-Al2O3(010); (D), and (E): optimized configurations of N2 adsorption on Fe3/y− Al2O3(010); (F): N2 adsorption configuration on the C7 site of Fe(211) surface. Reproduced with permission from ref. Azofra, L. M.; Morlanes, N.; Poater, A.; Samantaray, M. K.; Vidjayacoumar, B.; Albahily, K.; Cavallo, L.; Basset, J.-M., Angew. Chem., Int. Ed. 2019, 58 (20), 6476.; Liu, J.-C.; Ma, X.-L.; Li, Y.; Wang, Y.-G.; Xiao, H.; Li, J., Nat. Commun. 2018, 9 (1), 1610. Copyright 2020 American Chemical Society.

Furthermore, another important theoretical work appeared recently, in which the authors studied by DFT the complete catalytic mechanism for the conversion of N2 toward NH3 on Fe3/ y-Al2O3 (010) that is distinctly different from the industrially used catalyst based on Fe or Ru metal surface. They claim that the 3 Fe are linked together in a triangular form, which is the most stable clusters on Al2O3 and kinetically constant against aggregation. They found that the ammonia synthesis reaction follows the “associative mechanism” at the Fe3 single cluster. The hydrogenation of N2 to  NNH is much faster than the “dissociative mechanism” of N2.219 This “associative mechanism” is attributed to the multi-step redox capacity, large spin polarization, and the low oxidation state of the metal. This enables efficient N2 activation due to the spin-polarized charge transfer from Fe’s 3d orbital to N2 p orbitals (Fig. 27). The partial occupations of N2’ s b-spin p orbitals both lowers the NdN bond order. This phenomenon leads to an associative mechanism for N2 activation to ammonia production, which mimics the initiation process in the nitrogenase as well as artificial metal complexes for homogeneous catalysis involving nitrogen fixation.195,234,235 Previous theoretical studies showed the possibility of N2 reduction toward ammonia at low temperature and pressure; nevertheless, a very small TOF for ammonia synthesis was obtained due to the Brønsted-Evans-Polanyi (BEP) relationship on the metal surfaces.236 Thus, the BEP relationship demands the balance between N2 dissociation and NHx species desorption. Remarkably, the associative mechanism bypasses the BEP relation; hence the limitation was underlying one side of the volcano curve.198,237,238 In the future, the research on ammonia synthesis reaction on the metal surface should be motivated to fund an optimum between TOF and reaction conditions (lower temperature and pressure). A study based on the preparation of Fe NPs embedded in a kind of “carbon-nitrogen” support from a molecular Fe phthalocyanine (FePc) precursor is currently undergoing in our laboratory.222,239 The role of alkali promoters on the catalyst structure was elucidated experimentally and theoretically (DFT calculations). First, it was found that alkali is present as alkali oxide (e.g., K2O). The main effect of alkali is related to the geometrical repartition of this alkali oxide on the Fe surface. The alkali leaves at medium coverage of the surface some exposed Fe(0) for non-dissociative chemisorption of N2 (by an end-on type).

14.13.10

Light alkanes aromatization

14.13.10.1

Introduction

Aromatics, especially Benzene, Toluene, and Xylenes (BTX), are among the most important basic chemicals in the chemical industry, which is extensively used in the production of styrene, phenol, polymers, plastics, pharmaceuticals, rubbers, paints, adhesives, and others. Furthermore, aromatics represent about one-third of the market for commodity petrochemicals 240–242. Benzene is a

Surface Organometallic Chemistry and Catalysis

507

multipurpose chemical building block with a varied array of end uses. The most important application that utilizes benzene includes styrene, phenol, nylon, and aniline production.243 Due to its low toxicity and high octane number, mostly toluene is typically mixed with unleaded gasoline. The other fraction of toluene is used for conversion into benzene and xylenes or as a solvent for different applications.244 Concerning the xylenes, the para-xylene is the most valuable isomer. It is converted into terephthalic acid and dimethyl terephthalate, which produce polyethylene terephthalate fiber, resins, and films. In contrast, the ortho-xylene and meta-xylene are converted to phthalic anhydride or isophthalic acid, respectively.245 A large quantity of aromatics comes from catalytic cracking/reforming of naphtha.241,246–253 However, the new natural gas sources recently discovered, and considered as the cleanest fossil fuel, increased the interest to valorized natural gas to more valuable products like aromatics.254 Another route to produce aromatics via methanol after it production from natural gas.241,248,249,255–257 During the last decades, the catalytic approaches to producing aromatics from light alkanes such as ethane, propane, n-butane, or isobutene were also well investigated and have gained high importance for both academic and industrial communities.242,258–264 Furthermore, due to the complexity of the aromatization process, which required a multi-step reaction, the necessity of a multifunctional catalyst is highly desired. Zeolites are well known to be used in diverse industrial applications that range from heterogeneous catalysis for various applications: methanol to olefins (MTO), fluid catalytic cracking (FCC), and hydrocracking technology, a vital process in the oil-refining industry-based, and ion and molecular separations to drug delivery and sensing. The choice of zeolite is due to its porosity system leading to shape selectivity, chemical composition, hydrophobicity/hydrophilicity, acidic properties, and (hydro) thermal stability.250,265–267 Furthermore, the acidity of zeolite is considered as the most important property. In other terms, the conversion in the catalytic reaction, catalyst lifetime, product yield/selectivity, and potential regeneration is directly related to the nature and distribution of acid sites on the zeolite crystals.268–270 Several metals have been used as promoters with zeolites; for example, Ga, Mo, Pt, Fe, Ni, and Zn have been frequently used with zeolites to convert light alkanes to aromatics. 112,256,271–278 The most studied catalytic system for the aromatization reaction is Ga modified zeolites, which showed a higher activity over the other metals.268,271,276,279,280 Furthermore, coke formation is the main obstacle for the aromatization process catalysts by blocking the access to the active sites and reduce the catalytic activity with time on stream. This formation of coke is due to the production of poly-aromatics at high reaction temperatures.281–283 Various technologies were applied in the industry for converting light alkanes and LPG., The “Cyclar Process” developed by UOP and BP in 1991, transforms liquefied petroleum gas (LPG) into aromatics in a single operation.247,272,282 This technology operates with a multifunctional catalyst ZSM-5 zeolite impregnated with a Gallium (Ga) solution salt. This impregnation method is used to avoid the location of the large hydrated Ga3+ cations in the zeolite pores.284 There is a consensus that Ga+ is involved in the dehydrogenation of alkanes to olefins by an oxidative addition step followed by oligomerization, cyclization, and aromatization steps on the acid sites of zeolites.285 Table 1 summarizes the various technologies developed by different companies worldwide to convert light alkanes and other feedstocks to valuable products such as aromatics. Zeolite is the typical support used in those various technologies described in Table 1 owing to its acidity properties and ability to support a transition or a non-transition metal. Furthermore, the wetness impregnation is the principal method to modify ZSM-5 by Ga, Mo, Pt, or Zn. The mechanical mixing of oxides and chemical vapor deposition of precursors like GaMe3 are reported.297–300 Another tool for preparing the bifunctional catalysts is to prepare them by Surface Organometallic Chemistry (SOMC) methodology. This method usually leads to a well-defined single-site catalyst described previously, with a uniform distribution of active sites, an essential parameter in the aromatization process.4,86,159,242,301,302 Based on the assumption that alkane aromatization occurs by dehydrogenation of the paraffin into an olefin followed by the aromatization (oligomerization, cyclization, dehydrogenation) of the obtained olefin by an acid-catalyzed mechanism, the choice and the localization of the metal for the first step is a crucial parameter to consider.

Table 1

Different technologies used worldwide to convert light alkanes or LPG to aromatics (BTX).

Process

Z-Forming

Cyclar

Aroforming

Alpha

Licensor Feedstock Reactor

Mitsubishi Oil/Chiyoda C3, C4, light naphtha Fixed, radial flow, adiabatic Switching reactors Zeolite, cylindrical, formed from Al-silicate and specialized binder 200 bbl/day demo-plant Nov. 1990–Dec. 1991

IFP/Salutec C3, C4, light naphtha Fixed tubular, isothermal Swing reactors Zeolite, cylindrical

Sanyo/Asahi C3dC7 olefins Fixed beds, adiabatic

Catalysts regeneration Catalysts

BP/UOP C3, C4, light naphtha Radial flow, four arranged in a vertical stack, adiabatic CCR Zeolite, spherical with non-noble metal promoter 1000 bbl/day demo-plant Dec. 1989–Dec. 1991

286–289

247,272,290–295

Status

Ref.

500 bbl/day demo-plant under design 296

Swing reactors Zeolite, cylindrical 3500 bbl/day commercial since Jul. 1993 281

508

Surface Organometallic Chemistry and Catalysis

14.13.10.2

Aromatization by SOMC methodology (catalysis by design)

The aromatization of light alkanes is also an interesting area from an academic viewpoint as it represents an example of a complex and multistage reaction. Most of the studies have focused on using modified ZSM-5 and have reported that the yield of aromatics can be increased essentially by using a bifunctional system. The zeolite provides acidity, and one or more metals are added to provide dehydrogenation activity.284 Several metals have been considered, mainly Ga and Zn, and the results indicate that Ga modified zeolites are the most efficient catalysts for light alkanes aromatization.282,296 There is a consensus that Ga species are involved in dehydrogenation reaction and that the Brønsted acid sites of zeolites catalyze the oligomerization, cyclization, and dehydrogenation steps. As described previously, various preparation methods have been reported, in which extra framework gallium can be incorporated into the micropores of HZSM-5: -ion ex-change in aqueous solution, -impregnation, -chemical vapor deposition (CVD) of GaCl3, and—a physical mixture of Ga2O3 or GaCl3.262,296 Nevertheless, the catalytic properties are independent of the preparation method since the active site is mostly generated after a severe pre-treatment step before the reaction.279 It is also possible to obtain isomorphous substitution of aluminium by gallium in MFI-type zeolites by steam activation.303 With aqueous ion exchange, the hydrated Ga cations tend to stay on the external zeolite surface. The subsequent reduction or oxidation of the Ga species is necessary to disperse them into the micropores. Sublimation of GaCl3 is a more efficient method for depositing cationic Ga species directly into the micropores replacing Brønsted acid sites.300 However, the presence of water during the pre-treatment at elevated temperature enhances Ga mobility and generates multiple Ga species that are difficult to identify. As an alternative route and to avoid any addition of water, Van-Santen et al. developed an anhydrous procedure consisting of chemical vapor deposition (CVD) of trimethylgallium Ga(CH3)3 with subsequent removal of the methyl groups by treatment with H2 or O2, H2 is preferred. This is a kind of SOMC approach developed by S. Scott on silica.303 With H2 methane is produced, which causes less modification on the zeolite framework, resulting in better Ga dispersion. During the chemical vapor deposition, the reaction of Ga(CH3)3 occurred only with the [^SidOH] terminal group and left the Brønsted acid sites intact.304,305 However, further heating of Ga(CH3)3/ZSM-5 resulted in the total consumption of Brønsted acid sites as described in (Scheme 25). This observation was supported since no aromatics were observed during the propane conversion, which gives a high selectivity for de propylene, methane, and ethylene as hydrocarbon products.297,306

Scheme 25 Reaction of Ga(CH3)3 with Brønsted sites of ZSM-5 followed by (i) hydrogen treatment at 773 K, (ii) He at 823 K and (iii) N2O at 473 K. These reactions lead to [Ga+]ZSM-5, [Ga]O+]ZSM-5, and [GaH+2 ]ZSM-5 species as reported by Van Santen and collaborators. Reproduced with permission from ref. Hensen, E. J. M.; Pidko, E. A.; Rane, N.; van Santen, R. A, Studies in Surface Science and Catalysis, Xu, R.; Gao, Z.; Chen, J.; Yan, W., Eds. Elsevier: 2007; Vol. 170, 1182–1189.

Furthermore, A DRIFT study was performed to investigate the nature of gallium species in the Ga/ZSM-5 catalyst prepared by chemical vapor deposition of Ga(CH3)3 at different times and during the reduction under H2 at 550  C (Fig. 28). The authors claim that the elimination of acidic protons via reaction of the bridging hydroxyl groups with Ga(CH3)3 starts slightly above room temperature. This results in a gradual decrease of the band at 3610 cm−1 with CdH stretching bands appearing between 2900 and 2600 cm−1 from the grafted Ga(CH3)2 species. To eliminate the bulky methyl groups and to increase the accessibility of protons for Ga(CH3)3, the catalyst was treated under hydrogen at 823 K. Repetition of this procedure several times resulted in a complete replacement of protons by the univalent gallium ions according to the Eq. (3):  +  + ZO − ⋯H + + GaðCH3 Þ3 ! ZO − ⋯ GaðCH3 Þ2 + CH4 , ZO − ⋯ GaðCH3 Þ2 + H2 ! 2CH4 + ZO − ⋯Ga + (3)

Surface Organometallic Chemistry and Catalysis

509

2800

2964 cm-1

2911 cm-1

2834 cm-1

Kubelka-Munk

3610 cm-1

3000

3200

3400

3600

3800

Wavenumbers (cm-1) Fig. 28 Room-temperature DRIFT spectra of the hydroxyl region and of anchored dimethyl gallium species: (a) dehydrated parent H-ZSM-5, (b) H-ZSM-5 contacted with Ga(CH3)3 for 1 h at 323 K, (c) subsequent reduction in H2 for 1.5 h at 823 K, (d) subsequent contacting with Ga(CH3)3 for 1 h at 323 K, and (e) subsequent reduction in H2 at 823 K for 3 h. Reproduced with permission from ref. Hensen, E. J. M.; Pidko, E. A.; Rane, N.; van Santen, R. A, Studies in Surface Science and Catalysis, Xu, R.; Gao, Z.; Chen, J.; Yan, W., Eds. Elsevier: 2007; Vol. 170, 1182-1189. Copyright 2007 Elsevier.

Other studies such as Ga K-edge XANES spectroscopy and DFT calculations confirmed the hypothesis that the Ga species was reacted with the brønsted acidity and the production of aromatics is significantly minimized after ethane propane and n-butane conversion.285,306 However, only the dehydrogenation reaction of those alkanes was performed, and the selectivity for the analog olefins was very high (Fig. 29). Later, the same group investigated by DFT cluster modeling approach, various gallium species (Ga+, GaH+2, and GaH+2) in reduced Ga-ZSM-5. Indeed, extra-framework gallium incorporated by traditional impregnation methods results in Ga deposition on the external surface of the zeolite crystals. Hydrated Ga3+ are too bulky to enter the elliptical channels of ZSM-5.285 The DFT study for ethane dehydrogenation revealed that univalent Ga+ ions at cation sites of high-silica zeolites are the most probable active species. The catalytic reaction can also occur with GaH+2 and GaH2 +, but these proposed paths were not favored. Nevertheless, all of the gallium species acted as Lewis acids promoting heterolytic CdH cleavage involving the basic oxygen atoms of the zeolite lattice.285 Moreover, it was recently demonstrated by synthesizing a series of molecular model compounds and grafted surface organometallic Ga species. The XANES edge energy cannot be used to distinguish Ga3+ alkyls or hydrides from Ga+.307 Recently, Scott, Taoufik et al. studied the conversion of propane to aromatics (BTX) with a catalyst resulting from the grafting of Ga(iBu)3 on mesoporous zeolite [H-ZSM-5]. The catalyst was prepared by the SOMC technique, a powerful method to prepare by design a well-defined bi-functional catalyst (dehydrogenation of alkanes, aromatization of the resulting olefins) with an optimal ratio of Lewis and brønsted acid sites.86,262 They claim that the choice of the organometallic precursor is crucial to avoid any reaction with the Brønsted acid sites located on the micropores and minimize dimerization to digallium species. For this reason, they chose a bulky organogallium such as Ga(iBu)3 (size of 10 A˚ ). Furthermore, the grafted reaction between Ga(iBu)3 and [H-ZSM-5250] after dehydroxylation at 250  C under vacuum was conducted at room temperature to ensure selective reaction Ga(iBu)3 with the silanol group. Different powerful spectroscopic tools (solid-state NMR, IR, EXAFS spectroscopy, elemental analysis) were used to confirm that the grafting of the organogallium precursor occurred selectively on the silanol groups and resulting in uniform monopodal gallium species [(^SiOd)Ga(iBu)2]. The IR spectra of the pristine [Meso-H-ZSM-5250] and the [(^SiOd)Ga(iBu)2] catalysts revealed significant consumption of isolated silanols as indicated by the attenuation of the intensity of the 3743 cm−1 band. Simultaneously, two series of bands appeared at 2800–3000 cm−1 and 1300–1500 cm−1, attributed to the vibration of alkyl moieties. However, as expected, the band at 3601 cm−1 attributed to brønsted acidity remained intact (Fig. 30 A). The 1H NMR spectrum of [(^SiOd)Ga(iBu)2] consists of signals due to the iBu protons, with a peak at 1.0 ppm accompanied by a shoulder at 0.9 ppm (CH3 and CH2 moieties, respectively) and another at 2.0 ppm (CH groups) (Fig. 30 B). The peaks that appeared at 2.7 and 4.2 ppm are associated with the extra frame work aluminium OH and Brønsted acidic sites. No signal for residual silanols was detected. Further insight was gained from the 1Hd1H DQ-SQ correlation spectrum (Fig. 30 B). The main correlation is due to the isobutyl groups, for which cross-peaks

510

(A)

Surface Organometallic Chemistry and Catalysis

(B)

Fig. 29 Reaction paths for the “alkyl activation” mechanism of ethane dehydrogenation over (A) Ga+ Zd; (B) GaH2Zd. Reproduced with permission from ref. Pidko, E. A.; Kazansky, V. B.; Hensen, E. J. M.; Van Santen, R. A., J. Catal. 2006, 240 (1), 73-84. Copyright 2006 Elsevier.

Fig. 30 (A): DRIFT spectra of [Meso-H-ZSM-5250] (black) and [(^SiOd)Ga(iBu)2] (red). (B): [(^SiOd)Ga(iBu)2]: (A) 1H MAS and (B) 1Hd1H DQ MAS NMR spectra (18.8 T, 20 kHz).  Methylenic protons of pentane. Reproduced with permission from ref. Szeto, K. C.; Gallo, A.; Hernandez-Morejudo, S.; Olsbye, U.; De Mallmann, A.; Lefebvre, F.; Gauvin, R. M.; Delevoye, L.; Scott, S. L.; Taoufik, M., J. Phys. Chem. C 2015, 119 (47), 26611–26619.

Surface Organometallic Chemistry and Catalysis

511

at (2.0, 3.0) and (1.0, 3.0) appear due to dipolar interactions between CH groups (peak 2 in Fig. 30 B) and CH3 groups (peak 3 in Fig. 30 B). The CH groups also give rise to a weak autocorrelation signal, as expected from the presence of two isobutyl groups on each Ga center. The EFA protons give rise to a self-correlation (3.6–2.3 ppm in the direct dimension), indicating the presence of several similar hydroxyl groups in close proximity, as expected from the range of structural types for Al(OH)n postulated in the literature.140 Their wider CS range indicates that grafting of the gallium species significantly perturbs their environment. No self-correlation was detected for Brønsted acidic protons using the same experimental conditions (Fig. 30 B). Furthermore, a stable propane conversion was found for [(^SiOd)Ga(iBu)2] catalyst at 500  C with a high selectivity toward aromatics (around 65% for benzene and toluene) with time on stream (2500 min). They claim that the high selectivity toward aromatics is attributed to the organogallium precursor, which is mainly responsible for the dehydrogenation of propane. It is selectively grafted on the silanol groups and not on the brønsted sites. These Brønsted acid sites are thus preserved to perform an aromatization reaction. As compared to the catalyst prepared by Van-Santen and collaborators described previously,300 this catalyst showed very high selectivity for aromatics. In contrast, the catalysts prepared by vapor deposition of trimethylgallium on [ZSM-5] showed high selectivity toward propene, and no aromatics were produced. Inspired by this study, our team developed recently an efficient catalyst for propane to aromatics based on titanium grafted on [H-ZSM-5].242 The selection of Ti was based on various proprieties: it is a very active metal for catalytic transformations in particular for Ziegler Natta polymerization as it is a versatile, abundant, inexpensive, with low-toxicity and established biocompatibility308,309 and among the oxide-supported group IV metal hydrides, titanium hydrides are the best catalysts for low-temperature hydrogenolysis of paraffin.168 This hydrogenolysis is assumed to come from a sigma bond metathesis step between the paraffin’s CdH bond and the d0 metal hydride.165,166,170,310 A bulky, organometallic complex of titanium (TiNp4) (size 9–10 A˚ ) with the expectation, based on modeling, that it could not enter the pores of [H-ZSM-5] (ca. 5–6 A˚ ). To follow our strategy, the [H-ZSM-5] was first dehydrated at 300  C under vacuum, then the TiNp4 was reacted with [H-ZSM-5300] in pentane for 8 h at room temperature to obtain [Ti/ZSM-5] catalyst. Prior to propane conversion, the freshly prepared catalyst [Ti/ZSM-5] was treated under H2 at 550  C, to give the [Ti-H/ZSM-5] catalyst. The catalysts [Ti/ZSM-5] and [Ti-H/ZSM-5] were further thoroughly characterized by ICP, CHNS, DRIFT, XRD, N2-physisorption, multinuclear solid-state NMR, XPS, and HR-TEM analysis, including STEM imaging. Similar to the case of Ga(iBu)3 described previously,262 all the elemental analysis, H and C solid-state NMR and DRIFT confirmed that the TiNp4 was grafted on the external hydroxyl groups (^SidOH), with uniform distribution of Ti on the surface [H-ZSM-5]. Furthermore, the high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) analysis of the [Ti-H/ZSM-5] catalyst confirmed that titanium was uniformly distributed mostly on the external surface of the [H-ZSM-5300] crystals and left the Brønsted acidity untouched (Fig. 31). The catalytic tests were carried out in a continuous flow reactor at 550  C with a space velocity ¼ 16.08 h−1, and a total flow rate of pure propane of 20 mL min−1. High conversion of propane (ca. 55%) was recorded over [Ti-H/ZSM-5] especially at the beginning of the reaction as compared to the pristine [H-ZSM5300]. As time on stream goes on, this conversion decreased slightly over time, and the conversion of propane remained significant

Fig. 31 (1) HAADF-STEM image for [Ti-H/ZSM-5] after treatment at 550  C for 2 h under H2. EDX elemental mapping images of [Ti-H/ZSM-5]; (2) Si (red), (3) Ti (yellow), and (4) Ti is displayed on the surface of Si. Reproduced with permission from ref. Al Maksoud, W.; Gevers, L. E.; Vittenet, J.; Ould-Chikh, S.; Telalovic, S.; Bhatte, K.; Abou-Hamad, E.; Anjum, D. H.; Hedhili, M. N.; Vishwanath, V.; Alhazmi, A.; Almusaiteer, K.; Basset, J. M., Dalton Trans. 2019, 48 (19), 6611–6620.

512

Surface Organometallic Chemistry and Catalysis

even after 60 h (25%) compared to [H-ZSM-5300] (ca. 13%). It should be noted that the deactivation is very similar for the catalysts, in agreement with the fact that titanium is outside the pores and does not modify the internal Brønsted acidity responsible for coke formation.277 Consequently, the [Ti-H/ZSM-5] catalyst reported the highest yield of aromatics at the initial stage (30%). Furthermore, this yield dropped slightly to 10% after 60 h of time on stream without any regeneration. In the case of pristine [H-ZSM-5300], the yield for aromatics dropped drastically from 15% at the beginning of the reaction to less than 3% after 60 h of time on stream. Inspired by the work of Guisenet et al., a pathway for the propane conversion and aromatics productions over [Ti-H/ZSM-5] catalyst was proposed, and titanium was assumed to play a significant role in the dehydrogenation of propane to propylene. This dehydrogenation reaction occurred following the well-known mechanism of sigma bond metathesis and b-H elimination, followed by the oligomerization cyclization and aromatization by the carbocationic mechanism on the brønsted acid sites (Scheme 26).170,273,282

Scheme 26 A proposed pathway for the aromatization reaction of propane over [Ti-H/ZSM-5300] catalyst. Reproduced with permission from ref. Al Maksoud, W.; Gevers, L. E.; Vittenet, J.; Ould-Chikh, S.; Telalovic, S.; Bhatte, K.; Abou-Hamad, E.; Anjum, D. H.; Hedhili, M. N.; Vishwanath, V.; Alhazmi, A.; Almusaiteer, K.; Basset, J. M., Dalton Trans. 2019, 48 (19), 6611–6620.

Obviously, the aromatization reaction of light alkanes, methanol, and natural gas remains a very large and challenging domain to deal with in both academic and industrial, especially with the concerns about sustainability such as climate change, waste pollution, energy security, and resource depletion.311 The SOMC seems a promising strategy to prepare a multifunctional catalyst with a uniform distribution of metal around the ZSM-5. More research needs to be deployed in this research axis to improve the catalytic activity, and the active species’ behavior to better make the structure-activity relationship. Moreover, as we know, this reaction is one of the most complicated chemical processes. The principal investigators seem to agree that improvement of aromatics yield can be increased essentially by using a bifunctional catalyst, mostly metal modified ZSM-5. The rapid deactivation of the catalyst by coke formation is the major challenge for the aromatization reaction, which is inexorable in the ZSM-5 based catalysts.312,313

14.13.11

Catalytic oxidation reaction by SOMC with O2: A new route to acetaldehyde

Acetaldehyde is one of the most important aldehydes as a building block for various important products, such as pyridine derivatives, vinyl acetate, penta-erythritol, crotonaldehyde, and resins. The dominating industrial process responsible for the annual production of 800,000 tons of acetaldehyde is the Wacker process, involving oxidation of ethylene in an aqueous solution of CuCl2/ PdCl2 as catalysts (Fig. 32).86,314–316 Expensive Pd is used to oxidize ethylene to acetaldehyde by reducing Pd(II) to Pd(0), followed by the re-oxidation of Pd(0) to Pd(II) by CuCl2 in a cocatalytic cycle. Industrially, the two catalytic cycles can operate in a single reactor or in two separate reactors depending on the purity of the reactant gas feed. The higher cost of the two-reactors method is balanced by using less expensive

Surface Organometallic Chemistry and Catalysis

513

Fig. 32 The Wacker process. Reproduced with permission from ref. Samantaray, M. K.; D’Elia, V.; Pump, E.; Falivene, L.; Harb, M.; Ould Chikh, S.; Cavallo, L.; Basset, J.-M., Chem. Rev. 2020, 120 (2), 734–813. Copyright 2020 American Chemical Society.

Fig. 33 Catalysis by Design: Possible Analogy between metathetic oxidation and olefin metathesis. Reproduced with permission from ref. Le Quéméner, F.; Barman, S.; Merle, N.; Aljuhani, M. A.; Samantaray, M. K.; Saih, Y.; Szeto, K. C.; De Mallmann, A.; Minenkov, Y.; Huang, K.-W.; Cavallo, L.; Taoufik, M.; Basset, J.-M., ACS Catal. 2018, 8 (8), 7549–7555. Copyright 2018 American Chemical Society.

diluted gases.317 This homogeneous process generates highly corrosive HCl upon the catalytic reaction, which imposes the use of devoted materials (normally ceramic coated titanium) to construct the reactor and the plant. Furthermore, the homogeneous nature of the reaction also complicates the regeneration of the catalysts. Nevertheless, this process is quite reliable and has been used for decades. On the other hand, the oxidative transformation of alkenes into carbonyl compounds is considered a widely used synthetic strategy to cleave the C]C bonds and to build C]O functionalities. While the heme and nonheme oxygenase oxidize olefins with molecular oxygen under mild conditions, 318–320 the use of ozone and a stoichiometric amount of reductive workup reagents, such as Zn and dimethyl sulfide, or a combination of other stoichiometric oxidants, such as KMnO4 and OsO4/HIO4, are still the common practice in synthetic chemistry. Using O2 as the sole oxidant is economically and environmentally attractive, but literature examples are scarce, and the reactant scopes are limited. To the best of our knowledge, no direct catalytic oxidation of internal or a-olefins to aldehydes via single-step catalysis using molecular oxygen by metathesis has been reported. To find a sustainable alternative to the Wacker process, the analogy between olefin metathesis and its mechanism321 and the unreported “metathetic oxidation” of olefin by molecular oxygen was one of the first strategic hypothesis based on catalysis by design. In olefin metathesis, the double bond of 2-butene can be cleaved by ethylene to yield propylene via the Chauvin mechanism (Fig. 33). Therefore, the double bond of 2-butene could react with the double bond of oxygen to afford acetaldehyde during metathetic oxidation (Fig. 33). Because olefin metathesis occurs with molybdenum (Mo) complexes, either in solution or after grafting to a support,322,323 we chose by design to test this metal for “metathetic” oxidation. Indeed, supported Mo-oxo species can be activated by olefins during the initiation step, yielding Mo-carbene species that are active in olefin metathesis,324 and these metal carbenes undergo pseudo-Wittig reactions with molecular oxygen to afford a metal oxo and an aldehyde. To achieve single-site catalysis by design which could lead to a high selectivity in the product’s distribution, it was assumed that the active site of the catalyst would consist of one or preferably two Mo-oxo (Mo]O) functionality supported on silica. We chose (O])Mo(dOtBu)4 as the precursor and successfully transformed it into a silica-supported catalyst that was able to promote the conversion of propylene or 2-butene to acetaldehyde via molecular oxygen with high selectivity.323 Spectroscopic characterization of the developed catalyst, including EXAFS, XANES, IR, DRIFT, Microanalysis, unambiguously showed that it corresponded to the well-defined (^SiOd)2Mo(VI)(]O)2322 (Scheme 27).

Scheme 27 Grafting of Mo(]O)(dOtBu)4 on silica(200) followed by its thermal rearrangement to (^SidOd)2Mo(]O)2. Reproduced with permission from ref. Le Quéméner, F.; Barman, S.; Merle, N.; Aljuhani, M. A.; Samantaray, M. K.; Saih, Y.; Szeto, K. C.; De Mallmann, A.; Minenkov, Y.; Huang, K.-W.; Cavallo, L.; Taoufik, M.; Basset, J.-M., ACS Catal., 2018, 8 (8), 7549–7555. Copyright 2018 American Chemical Society.

514

Surface Organometallic Chemistry and Catalysis

The catalyst gave a steady-state conversion of about 20%, which was constant with time on stream, with a selectivity of 68%. (Fig. 34). The other by-products were CO2, (98 80 20 81

46 20 3 14

Conditions: Dry CH3CN solvent; 100 mg cat. 1.0 mmol Limonene; 2.0 mmol aq. H2O2 or 1.1 mmol dry TBHP; solvent reflux temperature; glass batch reactor. Limonene conversion. b Selectivity to limonene epoxide. c Endocyclic/exocyclic epoxide ratio. d Specific activity ([mol converted limonene] [mol Nb h]−1) after 1 h. a

Reproduced with permission from ref. Gallo, A.; Tiozzo, C.; Psaro, R.; Carniato, F.; Guidotti, M., J. Catal. 2013, 298, 77–83. Copyright 2013, Elsevier.

Surface Organometallic Chemistry and Catalysis

519

Nb and Ti on silica have been used with H2O2 as the oxidant.338 Sn-BEA was successfully tested in the oxidation of delfone into d-decalactone (an important fruity aroma for the flavors and fragrances industry). The desired product was obtained in high yields (up to 86%), the zeolite remained active for a long time, and turn-over numbers of ca. 10,000 were attained.337 Moreover, carrying out the reaction of the R-isomer of delfone (the most useful for flavoring purposes), the enantiomeric configuration of the migrating carbon atom was retained. This is an important feature when enantiopure ketone oxidization is necessary to be carried out (Scheme 33).

Scheme 33 Active site and mechanism of the Baeyer-Villiger oxidation of cyclohexanone over Sn-BEA.337 Reproduced with permission from ref. Boronat, M.; Concepción, P.; Corma, A.; Renz, M., Catal. Today 2007, 121 (1), 39–44. Copyright 2007, Elsevier.

14.13.15

Catalytic CO2 conversion by SOMC

After the Industrial Revolution, the continuous increase in anthropogenic CO2 emission due to the combustion of huge amounts of fossil fuels (coal, petroleum, natural gas, and their derivatives) is outpacing the natural carbon cycle. Unavoidably related environmental interest has recently taken the international spotlight from different levels, politically, scientifically, and legislation.338,339 The increase in CO2 emission (but also Methane) has increased the temperature in the atmosphere, which leads to climate change, affecting everyday life on the planet (desertification, flooding, ocean acidification, etc.).339–341 Deniably, in the period 2000–2014, the anthropogenic CO2 emissions increased with an average rate of 2.6% per year compared to 1.72% between 1970 and 2000.342 It became commonplace to discuss future emission trajectories in terms of scenarios. The IEA (International Energy Agency) project that a world commensurate with no more than 2  C of warming above pre-industrial levels is when total anthropogenic CO2 emissions are reduced to something less than 20 GtCO2 per year by 2050, with further reductions to near-zero or even net-negative emissions by the end of the century. This is typically referred to as the two-degree scenario (2DS). At the other end of the spectrum, allowing anthropogenic emissions to increase to 60 GtCO2 per year by 2050 is commensurate with a warming of approximately 6  C above pre-industrial levels six-degree scenario, 6DS as described in (Fig. 37).345

520

Surface Organometallic Chemistry and Catalysis

Fig. 37 Illustration of the calculated mitigation challenge rated to climate change. Reproduced with permission from ref. Mac Dowell, N.; Fennell, P. S.; Shah, N.; Maitland, G. C., Nat. Clim. Change 2017, 7 (4), 243–249. Copyright 2017 Spring Nature.

Fig. 38 Illustration of the utilization of CO2 as a feedstock to synthesize synthetic products and fuels through the formation of various chemicals.343,344

According to historical data from BP data,343 the low-mitigation scenario chosen here is the IEA’s 6DS, and the objective is to meet the IEA 2DS in 2050. In this example, the mitigation challenge equates to approximately 800 GtCO2 (total) from 2020 to 2050. It is becoming more apparent to reduce the net amount of anthropogenic CO2 released in the atmosphere by using renewable energy with high-energy efficiency and carbon capture and utilization (CCU). Owning from the large scale and rate of CO2 production compared to that utilization allows long-term sequestration by chemical sorbents. Geological storage of captured CO2 (CCS) on a large scale also gained high importance in the last decade. However, it encounters non-negligible limitations in energy, cost, CO2 transportation, and process durability.344–348 The transformation of CO2 into high-value chemicals is gaining more attention in the last decade and may result in a paradigm shift in which CO2 turns from waste to commodity (Fig. 38).343,344,349 Several technologies able to convert CO2 and water to economically attractive chemicals have been proposed in this context. These technologies can be divided into the following categories: biochemical, photochemical, electrochemical, and thermochemical.343,350,351 Despite high efficiencies to transform CO2 to certain products like CO, ethylene, methane, methanol, and aromatics, the high conversion and economically attractive production rates have proven challenging. The difficulty of applying them in chemical synthesis is due to the low reactivity originating from its inherent thermodynamic stability and kinetic inertness 352 These factors may easily offset the economic and environmental advantages of using CO2, especially using very efficient heterogeneous catalyzed transformation, becoming a desirable technology.353,354 Furthermore, a highly integrated chemical process able to direct capture and convert CO2 from the point source to high-value chemicals could be able to maximize the benefit of CO2 recycling without incurring most of the indirect penalties.355 To this extent, some studies have shown good potential to convert CO2 coming from flue gas. They include a non-catalytic process from CO2 to carbonates. They also include CO2 tri-reforming which is a combination of three reactions: steam methane reforming (SRM), dry methane reforming (DRM), and partial oxidation of methane (POM).356–358 Both methods have specific limitations and have not yet been implemented on a large scale. The use of multifunctional catalysts (a combination of a metal/metal oxide) for the direct valorization of CO2 to useful chemicals such as cyclic carbonate, oxazolidinones, acrylates sungas, formic acid methanol, methane, and CO have gained great importance over the past few years.12,349,351,359–364 More specifically, the catalytic conversion of CO2 over well-defined single site catalysts was studied “by design” using SOMC concepts and methods.18,365 Thus various studies recently reported converting CO2 into cyclic carbonates or polycarbonates by transition metals such as Zr, Nb, and Y after immobilizing their organometallic precursors in SiO2.12,86,365

Surface Organometallic Chemistry and Catalysis

521

In parallel, well-distributed nanoparticles of Cu, Ni, and Au were prepared after grafting of their organometallic precursors in inorganic supports (SiO2, Al2O3) by surface organometallic chemistry followed by reduction under H2 (Fig. 39). These metallic nanoparticles issued from the SOMC strategy served as catalysts to convert CO2 and methane to H2 and CO (dry reforming) to achieve CO2 methanation or conversion of CO2 to methanol.366–377

Fig. 39 (A) General SOMC strategy to generate isolated sites or supported nanoparticles. (B) and (C) Utilizing isolated metal site supports generated from SOMC for metal and metal-alloy nanoparticle formation. Reproduced with permission from ref. Lam, E.; Noh, G.; Chan, K. W.; Larmier, K.; Lebedev, D.; Searles, K.; Wolf, P.; Safonova, O. V.; Copéret, C., Chem. Sci. 2020, 11 (29), 7593–7598. Copyright 2020 Royal Society of Chemistry.

The first example for activating CO2 experimentally using organometallic surface chemistry with zirconium hydride supported on silica was reported in 2004. The CO2 molecule was inserted on the ZrdH bond to produce the corresponding formate complex (Scheme 34).12,112 This strategy was later investigated by DFT calculations to study the mechanism of CO2 insertion on the ZrdH complex by using a tripodal zirconium mono-hydride and zirconium bis-hydrides [(^SidOd)3ZrH] and [(^SidOd)2ZrH2], respectively model complexes.378

Scheme 34 Various reactions of CO2 with silica-supported zirconium mono and bis hydrides. Reproduced with permission from ref. Rataboul, F.; Baudouin, A.; Thieuleux, C.; Veyre, L.; Copéret, C.; Thivolle-Cazat, J.; Basset, J.-M.; Lesage, A.; Emsley, L., J. Am. Chem. Soc. 2004, 126 (39), 12541–12550. Copyright 2004, American Chemical Society.

522

Surface Organometallic Chemistry and Catalysis

This computational study shows that the CO2: (i) reacted first with the zirconium on the surface by formation of a weak adduct leading (ii) then to a formate species, denoted as (1-Z1-O2CH) and (1-Z2-O2CH), respectively, with high reaction energies of −135 and −183 kJ mol−1. Additionally, no isolated formic acid was reported with the zirconium complexes; even with homogeneous catalysts based on Ru, Rh, and Ir complexes, formic acid isolation was produced after a hydrogen treatment step.12,379 Inspired by the feasibility of CO2 activation in the supported organometallic complexes, the catalytic process of converting CO2 to carbonates under mild conditions was later reported without requiring high energy and potentially hazardous reactants such as hydrogen.380 The first example of the supported organometallic complex for carbonate synthesis at mild conditions was reported with niobium.18 This heterogeneous process was inspired by the homogeneous system of the NbCl5 in the presence of nucleophiles such as imidazolium bromides or others, which showed an efficient catalyst for industrially preparing carbonates after reacting CO2 with propylene epoxide at ambient temperature.381,382 Furthermore, an in-depth study experimentally and theoretically was reported recently by the grafting of NbCl4 on two silica supports (SiO2–200 and SiO2–700 dehydroxylated at 200  C and 700  C, which correspond to 0.78 mmol.g−1 and 0.3 mmol.g−1 respectively). Divers niobium species grafted on silica were generated. The monopodal niobium complex [(^SidO)NbCl4OEt2] (36) was obtained after reacting NbCl5 with SiO2–700 support, whereas with SiO2–200, a mixture of two grafted Nb complexes such us vicinal monopodal (37) and bipodal complexes [(^SidOd)2NbCl3OEt2] (38) (Fig. 40). 31P and 1H ssNMR on their PMe3 derivatives (360 ), (370 ), and (380 ) after reacting (36), (37), and (38) with PMe3 were investigated. It is assumed that the phosphine would displace the ether molecule and remains bound to the surface niobium without being physisorbed or chemisorbed on the surface of the silica. The obtained result led to the unambiguous assignment of (36) as a single-site monopodal Nb species.18 At the same time, (36) and (37) were found to present two distinct surface-supported components, with 2a being mostly monopodal [(^SidOd)NbCl4OEt2] and (37) being mostly bipodal [(^SidOd)2NbCl3OEt2] (Scheme 35). A double-quantum/ single-quantum 31P NMR correlation experiment carried out on (370 ) supported the existence of vicinal Nb centers on the silica surface for this species. A DFT study carried out on the NbCl5 model supported on SiO2–200. The obtained results display that the Nb complex was grafted in position A. The grafted of the second one in the vicinal position is also possible, suggesting a viable pathway of preparation. The difference in catalytic efficiency on the reaction of CO2 with propylene oxide in the presence of TBAB under mild conditions (60  C and 10 bar of CO2) was correlated with an unprecedented cooperative effect between two neighboring Nb centers on the surface of 2a.18

(A) 31P

(B) 1H

SS NMR

~30

SS NMR

(B)

~70

(38’)

~80

~20

(37’)

(36’)

Fig. 40 Comparison of 31P SS NMR and 1H ssNMR of (380 ) (top trace), (370 ) (middle trace) and (360 ) (bottom trace) showing the different proportions of monopodal and bipodal species in each compound. Reproduced with permission from ref. D’Elia, V.; Dong, H.; Rossini, A. J.; Widdifield, C. M.; Vummaleti, S. V. C.; Minenkov, Y.; Poater, A.; Abou-Hamad, E.; Pelletier, J. D. A.; Cavallo, L.; Emsley, L.; Basset, J.-M., J. Am. Chem. Soc. 2015, 137 (24), 7728–7739. Copyright 2015, American Chemical Society.

Surface Organometallic Chemistry and Catalysis

523

Scheme 35 Illustration of the proposed structures for niobium species supported on silica 1a–3a. Reproduced with permission from ref. D’Elia, V.; Dong, H.; Rossini, A. J.; Widdifield, C. M.; Vummaleti, S. V. C.; Minenkov, Y.; Poater, A.; Abou-Hamad, E.; Pelletier, J. D. A.; Cavallo, L.; Emsley, L.; Basset, J.-M., J. Am. Chem. Soc. 2015, 137 (24), 7728–7739. Copyright 2015, American Chemical Society.

The catalytic activity of the surface-supported Nb species was tested in the conversion of CO2 to carbonate after reacting with propylene oxide under mild conditions (60  C, 1 bar of CO2) in the presence of TBAB (Tetrabutylammonium bromide) as a nucleophile. The results showed a poor conversion of propylene oxide obtained with the monopodal Nb complexes (36). Contrary, excellent conversion of propylene oxide was obtained with the vicinal neighboring monopodal and bipodal niobium complexes (Fig. 41). The recyclability of the grafted niobium complexes was studied, and the obtained results showed a loss of catalytic activity of the catalysts in the second run as compared to the first run.18

Fig. 41 Catalytic cycloaddition of CO2 to propylene oxide (7 mL, 100 mmol) under mild conditions (60  C, 10 bar, 18 h) using [(^SidOd)NbCl4OEt2] (36) (1.1 g; 0.33 mmol of Nb), [(^SidOd)NbCl3OEt2] (37) (0.5 g; 0.33 mmol of Nb) or [(^SidOd)2NbCl3OEt2] (38) (0.7 g; 0.33 mmol Nb) in the presence of NBu4Br (1 mmol). For run 2, the recovered silica materials (ca. 95% of the initial amounts) were reused in the presence of 1 mmol NBu4Br. Reproduced with permission from ref. D’Elia, V.; Dong, H.; Rossini, A. J.; Widdifield, C. M.; Vummaleti, S. V. C.; Minenkov, Y.; Poater, A.; Abou-Hamad, E.; Pelletier, J. D. A.; Cavallo, L.; Emsley, L.; Basset, J.-M., J. Am. Chem. Soc. 2015, 137 (24), 7728–7739. Copyright 2015, American Chemical Society.

524

Surface Organometallic Chemistry and Catalysis

Recently, based on the early transition metal such as Zr and Y, various studies on a homogeneous and supported system, as Lewis acids for cyclic carbonate synthesis using an untreated flue CO2, were reported.347,383 This process of using untreated CO2 flue as feedstock to bypass the limitation of using clean CO2 and reduce the intensive energy for the purification, compressed, and CO2 separation.384 Furthermore, in 2017, Kelly et al. reported a study with ZrCl4 complex grafted on silica. As observed in the case of the supported NbCl5 complex, the grafting of ZrCl4(OEt)2 on SiO2–200 in various amounts led to a mixture of monopodal and bipodal species in different portions [(^SidOd)ZrCl3OEt2(Os(dSi^)2)] and [(^SidOd)ZrCl2OEt2(Os(dSi^)2)] (400 ), and (4000 ). The surface complexes’ identity was thoroughly investigated by FT-IR, elemental microanalysis, and solid-state NMR and applied as a recoverable and reusable heterogeneous catalyst for carbonate production using pure CO2 and flue gas samples from a cement factory in the presence of TBAB at the mild conditions. The activity of the catalysts suggests that there is not a considerable influence of the podality of the surface complexes on the catalytic performance. In contrast, cooperating neighboring zirconium centers seem to play a crucial role, which is the main difference from what has been observed for supported NbCl5 (Scheme 36).

Scheme 36 [(^SidOd)(ZrCl3
OEt2)2] and [(^SidOd)ZrCl3
OEt2
(Os(-Si^)2)] complexes in (39) and [(^SidOd)ZrCl3
OEt2
(Os(dSi^)2)] and [(^SidOd)2ZrCl2
OEt2
(Os(dSi^)2)] complexes present in different proportions in (40). Reproduced with permission from ref. Kelly, M. J.; Barthel, A.; Maheu, C.; Sodpiban, O.; Dega, F.-B.; Vummaleti, S. V. C.; Abou-Hamad, E.; Pelletier, J. D. A.; Cavallo, L.; D’Elia, V.; Basset, J.-M., J. CO2 Util. 2017, 20, 243–252. Copyright 2017, Elsevier.

The recyclability of the catalysts (39), (400 ), and (4000 ) was also performed; the catalytic activity was found to be decreased after consecutive cycles, due to the partial leaching of zirconium and probably due also a possible change of the coordination environment of the zirconium center.383 DFT calculations were also performed to understand the behavior of the activity of the different species obtained with ZrCl4. The obtained result suggests that the presence of a bimetallic pathway involving two zirconium atoms is effective in lowering the barrier for CO2 insertion. Nevertheless, it does not have an effect on the rate-determining step of ring closure (Fig. 42). As a consequence, both pathways can occur, depending on the concentration of zirconium complexes on the silica surface. Based on the well-established catalytic process of the coupling of CO2 and epoxides by ZrCl4 grafted complex described before, very recently, a new catalytic way was reported to convert CO2 on cyclic carbonate based on zirconium methyl complex grafted on silica.365 In this study, the molecular ZrMe4 complex was for the first time isolated and fully characterized as well as its silica-supported counterpart. The solid-state NMR and the elemental analysis showed the formation of a mixture of monopodal and bipodal ([(^SidOd)Zr(CH3)3(THF)2] and [(^SidOd)2Zr(CH3)2(THF)2]) complexes. Additionally, the 1H magic-angle spinning (MAS) solid-state NMR spectrum of [(^SidOd)Zr(CH3)3(THF)2] and [(^SidOd)2Zr(CH3)2(THF)2] complexes includes two broad signals at 3.5 and 1.4 ppm, which were found to be auto-correlated in double quantum (DQ) and triple quantum (TQ) experiments at 20 kHz MAS, as shown in (Fig. 43 A), respectively. The 13C CP/MAS NMR spectrum includes four resonances, at 67, 58, 23 ppm, and −5 ppm. Those at 67, 58, and 23 ppm were correlated with proton resonances at 3.5, 1.5, and 1.4 ppm, as indicated in the 2D 1Hd13C HETCOR NMR spectrum recorded with a contact time of 0.2 ms (Fig. 43 E).

Surface Organometallic Chemistry and Catalysis

525

Fig. 42 Computed free energy surface for the cycloaddition of propylene oxide (PO) and CO2 catalyzed by ZrCl4 ∙(OEt2)2-cis/TBAB (black dashed lines). Free energies in solution (PO was considered as the solvent) are given in kcal/mol relative to the starting species ZrCl4(OEt2)2-cis. The energy values in red represent the energies of the transition states. Free energy values in the brackets (connected through green bars and dashed lines) correspond to the bimetallic pathway. Reproduced with permission from ref. Kelly, M. J.; Barthel, A.; Maheu, C.; Sodpiban, O.; Dega, F.-B.; Vummaleti, S. V. C.; Abou-Hamad, E.; Pelletier, J. D. A.; Cavallo, L.; D’Elia, V.; Basset, J.-M., J. CO2 Util. 2017, 20, 243–252. Copyright 2017, Elsevier.

The grafted zirconium complexes ([(^SidOd)Zr(CH3)3(THF)2] and [(^SidOd)2Zr(CH3)2(THF)2]) served as a precursor to catalyze the conversion of CO2 to carbonates after reaction with PO, and the obtained catalytic activity was more than twofold higher as compared to the previously reported catalyst [(^SidOd)ZrCl3(THF)2] with a TON of 4221 versus 1640, respectively.365 Furthermore, Sodpiban et al. reported in 2019, for the first time, a sustainable approach to prepare a well-defined YCl3/TBAB complex grafted on silica [(^SidOd)YCl(dOCH(CH3)CH2Cl)] as powerful Lewis acids prepared by surface organometallic chemistry (SOMC) for the addition of CO2 on several epoxides.385 The obtained material was then characterized by several analytical and spectroscopic techniques demonstrating the successful grafting of the precursor and allowing a careful assignment of the surface structure with the assistance of DFT calculations (Fig. 44). Taken together, elemental analysis, XPS investigation, and DFT calculations indicate B3 as the most likely structure for the surface complexes of 1 with B2 as a minor species. Therefore, the materials’ activity was tested as heterogeneous Lewis acids for the cycloaddition of CO2 to several epoxides to afford cyclic carbonates at mild conditions (60  C and atmospheric pressure). The yttrium complexes led to high conversion under atmospheric and even ambient conditions and could be recovered and recycled.385,386 The transformation of CO2 to valuable chemicals is a promising route for reducing CO2 emission in the atmosphere and tackling contemporary energy crisis and environmental issues. In the literature, diverse pathways, including thermochemical, photochemical, electrochemical, biochemical, and plasma chemical reactions, have been reported recently. Especially, many interesting chemical transformations of CO2, including cyclization, carboxylation, condensation, formylation, methylation, and reduction of CO2, have been accomplished with the careful selection of suitable approaches. In particular, some transformations of CO2, for example, the cyclization of epoxides with CO2 to cyclic carbonates and hydrogenation of CO2 to methanol, may come into

526

Surface Organometallic Chemistry and Catalysis

Fig. 43 (A) One-dimensional (1D) 1H MAS solid-state NMR spectrum of [(^SidOd)Zr(CH3)3(THF)2] and [(^SidOd)2Zr(CH3)2(THF)2] complexes acquired at 600 MHz (14.1 T) with a 22 kHz MAS frequency, a repetition delay of 5 s, and 8 scans. (B) Two-dimensional (2D) 1Hd1H double-quantum (DQ)/single-quantum (SQ) spectra of ([(^SidOd)Zr(CH3)3(THF)2] and [(^SidOd)2Zr(CH3)2(THF)2]) complexes (acquired with 32 scans per t1 increment, 5 s repetition delay, 32 individual t1 increments). (C) Two-dimensional (2D) 1Hd1H Triple quantum (TQ)/single-quantum (SQ) spectra of [(^SidOd)Zr(CH3)3(THF)2] and [(^SidOd)2Zr(CH3)2(THF)2] complexes (acquired with 32 scans per t1 increment, 5 s repetition delay, 32 individual t1 increments) (D) 13C CP/MAS NMR spectrum of ([(^SidOd)Zr(CH3)3(THF)2] and [(^SidOd)2Zr(CH3)2(THF)2]) complexes acquired at 9.4 T with a 10 kHz MAS frequency, 5000 scans, a 4 s repetition delay, and a 2 ms contact time. Exponential line broadening of 80 Hz was applied prior to Fourier transformation. (E) 2D 1Hd13C CP/MAS dipolar HETCOR spectrum of [(^SidOd)Zr(CH3)3(THF)2] and [(^SidOd)2Zr(CH3)2(THF)2] complexes (acquired at 9.4 T with a 10 kHz MAS frequency, 4000 scans per t1 increment, a 4 s repetition delay, 64 individual t1 increments, and a 0.2 ms contact time). Reproduced with permission from ref. Al Maksoud, W.; Saidi, A.; Samantaray, M. K.; Abou-Hamad, E.; Poater, A.; Ould-Chikh, S.; Guo, X.; Guan, E.; Ma, T.; Gates, B. C.; Basset, J.-M., Chem. Commun. 2020, 56 (24), 3528–3531. Copyright 2020 Royal Society of Chemistry.

Fig. 44 Proposed surface structures of 1 based on characterization studies and DFT calculations. Reproduced with permission from ref. Sodpiban, O.; Del Gobbo, S.; Barman, S.; Aomchad, V.; Kidkhunthod, P.; Ould-Chikh, S.; Poater, A.; D’Elia, V.; Basset, J.-M., Cat. Sci. Technol. 2019, 9 (21), 6152–6165. Copyright 2019 Royal Society of Chemistry.

Surface Organometallic Chemistry and Catalysis

527

commercialization shortly. However, for most transformation routes of CO2, the challenges to achieving the industrial applications due to the relatively low efficiency or unfavorable operating conditions such as CO2 separation, purification, compression, and transportation in a safe way are still remaining. But with CO2 it appeared that the grafting of early transition metals (e.g., silica supported ZrCl4 or NbCl5) lead to a richer reactivity in catalysis.

14.13.16

Conclusions

In this chapter book, we describes the recent development of Surface Organometallic catalyst (SOMCcat) in various catalytic reactions. We also observed that SOMCcat plays a crucial role in isolating intermediates in some reactions where it is not possible in the case of the homogeneous or classical heterogeneous catalyst. Hence it facilitates the understanding of the reaction mechanism, and based on that, one can predict a reaction mechanism of a particular reaction. The single-site nature of SOMCcat is very different from that of the classical heterogeneous catalyst, where multiple sites are observed very frequently. Because of site isolation, the surface organometallic fragments (SOMF) stay intact with the surface. Hence bi-molecular decomposition is almost not observed, which is very frequent in the homogeneous catalyst. The single-site approach also allows to target a particular reaction, and based on the need to design, one can add or remove or change a spectator ligand from the coordination sphere of the metal center. Once an organometallic catalyst is grafted on the surface, it remains intact hence the leaching of the metal is almost not observed in the case of SOMCcat. The well-defined nature of the SOMCcat allows easy identification of the structure by combining various tools of molecular chemistry and surface science like ssNMR, IR, XPS, and microanalysis and gas quantification methods. The beauty of SOMCcat is that one can graft more than one well-defined organometallic complex on the surface to achieve cascade. This can be achieved very easily with the available modern techniques of surface science (ssNMR, IR, Gas quantification methods, etc.), which is not possible in the case of homogeneous or heterogeneous catalysis conditions. In this regard, it was observed how the development occurred in alkane metathesis reaction from a mono-metallic catalyst with a TON of 60 to a bi-metallic catalyst with a TON of 10,000 in propane metathesis reaction, by designing two metals with two functions. This type of catalyst tuning can only be possible if the catalyst is well-defined in nature so that further development can be achieved. Surface organometallic catalyst (SOMCcat) brings the molecular concept of homogeneous catalysis to heterogeneous catalysis. One can observe the essence of the well-defined nature of the catalyst as well as the stability of the catalyst in various conditions. With these advantages, SOMCcat projects itself as a good contender in the field of catalysis.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.

Copéret, C.; Chabanas, M.; Petroff Saint-Arroman, R.; Basset, J.-M. Angew. Chem. Int. Ed. 2003, 42 (2), 156–181. Tangeysh, B.; Palmer, C.; Metiu, H.; Gordon, M. J.; McFarland, E. W. Cat. Sci. Technol. 2020, 10 (3), 625–629. Basset, J.-M.; Pelletier, J. D. A. Natl. Sci. Rev. 2018, 5 (5), 633–635. Pelletier, J. D. A.; Basset, J.-M. Acc. Chem. Res. 2016, 49 (4), 664–677. Duer, M. J. Solid State NMR Spectroscopy: Principles and Applications; Wiley, 2002. Bakhmutov, V. I. Solid-State NMR in Materials Science; Taylor & Francis Group, LLC, 2012. Blanc, F.; Copéret, C.; Lesage, A.; Emsley, L. Chem. Soc. Rev. 2008, 37 (3), 518–526. Pourpoint, F.; Lafon, O.; Gauvin, R.; Amoureux, J. P.; Delevoye, L., Modern Methods in Solid-state NMR: A Practitioner’s Guide. The Royal Society of Chemistry: 2018; 97-133. Grekov, D.; Vancompernolle, T.; Taoufik, M.; Delevoye, L.; Gauvin, R. M. Chem. Soc. Rev. 2018, 47 (8), 2572–2590. Copéret, C.; Liao, W.-C.; Gordon, C. P.; Ong, T.-C. J. Am. Chem. Soc. 2017, 139 (31), 10588–10596. Saidi, A.; Samantaray, M. K.; Poater, A.; Tretiakov, M.; Cavallo, L.; Basset, J.-M. ChemCatChem 2020, 12 (23), 6067–6075. Samantaray, M. K.; Pump, E.; Bendjeriou-Sedjerari, A.; D’Elia, V.; Pelletier, J. D. A.; Guidotti, M.; Psaro, R.; Basset, J.-M. Chem. Soc. Rev. 2018, 47 (22), 8403–8437. Anwander, R. Chem. Mater. 2001, 13 (12), 4419–4438. Coperet, C.; Comas-Vives, A.; Conley, M. P.; Estes, D. P.; Fedorov, A.; Mougel, V.; Nagae, H.; Nunez-Zarur, F.; Zhizhko, P. A. Chem. Rev. 2016, 116 (2), 323–421. Chavhan, G. B.; Babyn, P. S.; Thomas, B.; Shroff, M. M.; Haacke, E. M. Radiographics 2009, 29 (5), 1433–1449. Samoson, A.; Lippmaa, E.; Pines, A. Mol. Phys. 1988, 65 (4), 1013–1018. Frydman, L. Encyclopedia of Nuclear Magnetic Resonance; Wiley, 2002; vol. 9; pp 262–274. D’Elia, V.; Dong, H.; Rossini, A. J.; Widdifield, C. M.; Vummaleti, S. V. C.; Minenkov, Y.; Poater, A.; Abou-Hamad, E.; Pelletier, J. D. A.; Cavallo, L.; Emsley, L.; Basset, J.-M. J. Am. Chem. Soc. 2015, 137 (24), 7728–7739. Tu, B.; Liu, X.; Wang, H.; Wang, W.; Zhai, P.; Fu, Z. Inorg. Chem. 2016, 55 (24), 12930–12937. Mazoyer, E.; Trébosc, J.; Baudouin, A.; Boyron, O.; Pelletier, J.; Basset, J.-M.; Vitorino, M. J.; Nicholas, C. P.; Gauvin, R. M.; Taoufik, M.; Delevoye, L. Angew. Chem. Int. Ed. 2010, 49 (51), 9854–9858. Li, Y.; Wu, X.-P.; Jiang, N.; Lin, M.; Shen, L.; Sun, H.; Wang, Y.; Wang, M.; Ke, X.; Yu, Z.; Gao, F.; Dong, L.; Guo, X.; Hou, W.; Ding, W.; Gong, X.-Q.; Grey, C. P.; Peng, L. Nat. Commun. 2017, 8 (1), 581. Bryce, D. L. Dalton Trans. 2019, 48 (23), 8014–8020. Bennett, A. E.; Griffin, R. G.; Ok, J. H.; Vega, S. J. Chem. Phys. 1992, 96 (11), 8624–8627. Bennett, A. E.; Rienstra, C. M.; Griffiths, J. M.; Zhen, W.; Griffin, R. G. J. Chem. Phys. 1998, 108 (22), 9463–9479. Angeli, C.; Cimiraglia, R.; Evangelisti, S.; Leininger, T.; Malrieu, J.-P. J. Chem. Phys. 2001, 114 (23), 10252–10264. Yan, Y. J.; Mukamel, S. J. Chem. Phys. 1986, 85 (10), 5908–5923. Feike, M.; Demco, D. E.; Graf, R.; Gottwald, J.; Hafner, S.; Spiess, H. W. J. Magn. Reson, Series A 1996, 122 (2), 214–221. Gregory, D. M.; Mitchell, D. J.; Stringer, J. A.; Kiihne, S.; Shiels, J. C.; Callahan, J.; Mehta, M. A.; Drobny, G. P. Chem. Phys. Lett. 1995, 246 (6), 654–663. Baldus, M.; Tomaselli, M.; Meier, B. H.; Ernst, R. R. Chem. Phys. Lett. 1994, 230 (4), 329–336. Brinkmann, A.; Edén, M.; Levitt, M. H. J. Chem. Phys. 2000, 112 (19), 8539–8554.

528 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89.

Surface Organometallic Chemistry and Catalysis Lee, Y. K.; Kurur, N. D.; Helmle, M.; Johannessen, O. G.; Nielsen, N. C.; Levitt, M. H. Chem. Phys. Lett. 1995, 242 (3), 304–309. Fan, T. W. M. Prog. Nucl. Magn. Reson. Spectrosc. 1996, 28 (2), 161–219. van Rossum, B. J.; Boender, G. J.; de Groot, H. J. M. J. Magn. Reson., Series A 1996, 120 (2), 274–277. Nielsen, N. C.; Bildsoe, H.; Jakobsen, H. J.; Levitt, M. H. J. Chem. Phys. 1994, 101 (3), 1805–1812. Sun, B. Q.; Costa, P. R.; Kocisko, D.; Griffin, R. G. J. Chem. Phys. 1995, 102 (2), 702–707. Griffin, R. G. Nat. Struct. Biol. 1998, 5 (supplement), 508–512. Gullion, T.; Vega, A. J. Prog. Nucl. Magn. Reson. Spectrosc. 2005, 47 (3), 123–136. Andrew, E. R.; Bradbury, A.; Eades, R. G. Nature 1958, 182 (4650), 1659. Walker, T. G.; Happer, W. Rev. Mod. Phys. 1997, 69 (2), 629–642. Natterer, J.; Bargon, J. Prog. Nucl. Magn. Reson. Spectrosc. 1997, 31 (4), 293–315. Lawler, R. G. Prog. Nucl. Magn. Reson. Spectrosc. 1973, 9 (3), 147–210. Hore, J.; Broadhurst, R. W. Prog. Nucl. Magn. Reson. Spectrosc. 1993, 25 (4), 345–402. Maly, T.; Debelouchina, G. T.; Bajaj, V. S.; Hu, K. N.; Joo, C. G.; Mak-Jurkauskas, M. L.; Sirigiri, J. R.; van der Wel, P. C.; Herzfeld, J.; Temkin, R. J.; Griffin, R. G. J. Chem. Phys. 2008, 128 (5), 052211. Barnes, A. B.; Paëpe, G. D.; van der Wel, P. C.; Hu, K. N.; Joo, C. G.; Bajaj, V. S.; Mak-Jurkauskas, M. L.; Sirigiri, J. R.; Herzfeld, J.; Temkin, R. J.; Griffin, R. G. Appl. Magn. Reson. 2008, 34 (3-4), 237–263. Kobayashi, T.; Perras, F. A.; Slowing, I. I.; Sadow, A. D.; Pruski, M. ACS Catal. 2015, 5 (12), 7055–7062. Rossini, A. J.; Zagdoun, A.; Lelli, M.; Lesage, A.; Copéret, C.; Emsley, L. Acc. Chem. Res. 2013, 46 (9), 1942–1951. Lilly Thankamony, A. S.; Wittmann, J. J.; Kaushik, M.; Corzilius, B. Prog. Nucl. Magn. Reson. Spectrosc. 2017, 102-103, 120–195. Can, T. V.; Caporini, M. A.; Mentink-Vigier, F.; Corzilius, B.; Walish, J. J.; Rosay, M.; Maas, W. E.; Baldus, M.; Vega, S.; Swager, T. M.; Griffin, R. G. J. Chem. Phys. 2014, 141 (6), 064202. Overhauser, A. W. Phys. Ther. Rev. 1953, 92 (2), 411–415. Jeffries, C. D. Phys. Ther. Rev. 1957, 106 (1), 164–165. Jeffries, C. D. Phys. Ther. Rev. 1960, 117 (4), 1056–1069. Corzilius, B. Phys. Chem. Chem. Phys. 2016, 18 (39), 27190–27204. Karabanov, A.; Kwiatkowski, G.; Perotto, C. U.; Wisniewski, D.; McMaster, J.; Lesanovsky, I.; Köckenberger, W. Phys. Chem. Chem. Phys. 2016, 18 (43), 30093–30104. Kryukov, E. V.; Newton, M. E.; Pike, K. J.; Bolton, D. R.; Kowalczyk, R. M.; Howes, A. P.; Smith, M. E.; Dupree, R. Phys. Chem. Chem. Phys. 2010, 12 (22), 5757–5765. Zagdoun, A.; Casano, G.; Ouari, O.; Schwarzwälder, M.; Rossini, A. J.; Aussenac, F.; Yulikov, M.; Jeschke, G.; Copéret, C.; Lesage, A.; Tordo, P.; Emsley, L. J. Am. Chem. Soc. 2013, 135 (34), 12790–12797. Kubicki, D. J.; Casano, G.; Schwarzwälder, M.; Abel, S.; Sauvée, C.; Ganesan, K.; Yulikov, M.; Rossini, A. J.; Jeschke, G.; Copéret, C.; Lesage, A.; Tordo, P.; Ouari, O.; Emsley, L. Chem. Sci. 2016, 7 (1), 550–558. Sauvée, C.; Rosay, M.; Casano, G.; Aussenac, F.; Weber, R. T.; Ouari, O.; Tordo, P. Angew. Chem. Int. Ed. 2013, 52 (41), 10858–10861. Song, C.; Hu, K.-N.; Joo, C.-G.; Swager, T. M.; Griffin, R. G. J. Am. Chem. Soc. 2006, 128 (35), 11385–11390. Koelsch, C. F. J. Am. Chem. Soc. 1957, 79 (16), 4439–4441. Haze, O.; Corzilius, B.; Smith, A. A.; Griffin, R. G.; Swager, T. M. J. Am. Chem. Soc. 2012, 134 (35), 14287–14290. Lelli, M.; Chaudhari, S. R.; Gajan, D.; Casano, G.; Rossini, A. J.; Ouari, O.; Tordo, P.; Lesage, A.; Emsley, L. J. Am. Chem. Soc. 2015, 137 (46), 14558–14561. Ardenkjær-Larsen, J. H.; Laursen, I.; Leunbach, I.; Ehnholm, G.; Wistrand, L. G.; Petersson, J. S.; Golman, K. J. Magn. Reson. 1998, 133 (1), 1–12. Liao, W.-C.; Ghaffari, B.; Gordon, C. P.; Xu, J.; Copéret, C. Curr. Opin. Colloid Interface Sci. 2018, 33, 63–71. Pines, A.; Gibby, M. G.; Waugh, J. S. J. Chem. Phys. 1972, 56 (4), 1776–1777. Hartmann, S. R.; Hahn, E. L. Phys. Ther. Rev. 1962, 128 (5), 2042–2053. Perras, F. A.; Kobayashi, T.; Pruski, M. Phys. Chem. Chem. Phys. 2015, 17 (35), 22616–22622. Zhao, X.; Hoffbauer, W.; Schmedt auf der Günne, J.; Levitt, M. H. Solid State Nucl. Magn. Reson. 2004, 26 (2), 57–64. Pines, A.; Gibby, M. G.; Waugh, J. S. J. Chem. Phys. 1973, 59 (2), 569–590. Cole, H. B. R.; Torchia, D. A. Chem. Phys. 1991, 158 (2), 271–281. Rossini, A. J.; Widdifield, C. M.; Zagdoun, A.; Lelli, M.; Schwarzwälder, M.; Copéret, C.; Lesage, A.; Emsley, L. J. Am. Chem. Soc. 2014, 136 (6), 2324–2334. Pinon, A. C.; Rossini, A. J.; Widdifield, C. M.; Gajan, D.; Emsley, L. Mol. Pharm. 2015, 12 (11), 4146–4153. Wolf, P.; Valla, M.; Rossini, A. J.; Comas-Vives, A.; Núñez-Zarur, F.; Malaman, B.; Lesage, A.; Emsley, L.; Copéret, C.; Hermans, I. Angew. Chem. Int. Ed. 2014, 53 (38), 10179–10183. Gunther, W. R.; Michaelis, V. K.; Caporini, M. A.; Griffin, R. G.; Román-Leshkov, Y. J. Am. Chem. Soc. 2014, 136 (17), 6219–6222. Wolf, P.; Valla, M.; Núñez-Zarur, F.; Comas-Vives, A.; Rossini, A. J.; Firth, C.; Kallas, H.; Lesage, A.; Emsley, L.; Copéret, C.; Hermans, I. ACS Catal. 2016, 6 (7), 4047–4063. Wolf, P.; Liao, W.-C.; Ong, T.-C.; Valla, M.; Harris, J. W.; Gounder, R.; van der Graaff, W. N. P.; Pidko, E. A.; Hensen, E. J. M.; Ferrini, P.; Dijkmans, J.; Sels, B.; Hermans, I.; Copéret, C. Helv. Chim. Acta 2016, 99 (12), 916–927. Berkson, Z. J.; Messinger, R. J.; Na, K.; Seo, Y.; Ryoo, R.; Chmelka, B. F. Angew. Chem. Int. Ed. 2017, 56 (19), 5164–5169. Harris, J. W.; Liao, W.-C.; Di Iorio, J. R.; Henry, A. M.; Ong, T.-C.; Comas-Vives, A.; Copéret, C.; Gounder, R. Chem. Mater. 2017, 29 (20), 8824–8837. Protesescu, L.; Rossini, A. J.; Kriegner, D.; Valla, M.; de Kergommeaux, A.; Walter, M.; Kravchyk, K. V.; Nachtegaal, M.; Stangl, J.; Malaman, B.; Reiss, P.; Lesage, A.; Emsley, L.; Copéret, C.; Kovalenko, M. V. ACS Nano 2014, 8 (3), 2639–2648. Piveteau, L.; Ong, T.-C.; Rossini, A. J.; Emsley, L.; Copéret, C.; Kovalenko, M. V. J. Am. Chem. Soc. 2015, 137 (43), 13964–13971. Lesage, A.; Lelli, M.; Gajan, D.; Caporini, M. A.; Vitzthum, V.; Miéville, P.; Alauzun, J.; Roussey, A.; Thieuleux, C.; Mehdi, A.; Bodenhausen, G.; Coperet, C.; Emsley, L. J. Am. Chem. Soc. 2010, 132 (44), 15459–15461. Grüning, W. R.; Rossini, A. J.; Zagdoun, A.; Gajan, D.; Lesage, A.; Emsley, L.; Copéret, C. Phys. Chem. Chem. Phys. 2013, 15 (32), 13270–13274. Valla, M.; Rossini, A. J.; Caillot, M.; Chizallet, C.; Raybaud, P.; Digne, M.; Chaumonnot, A.; Lesage, A.; Emsley, L.; van Bokhoven, J. A.; Copéret, C. J. Am. Chem. Soc. 2015, 137 (33), 10710–10719. Samantaray, M. K.; Alauzun, J.; Gajan, D.; Kavitake, S.; Mehdi, A.; Veyre, L.; Lelli, M.; Lesage, A.; Emsley, L.; Copéret, C.; Thieuleux, C. J. Am. Chem. Soc. 2013, 135 (8), 3193–3199. Conley, M. P.; Mougel, V.; Peryshkov, D. V.; Forrest, W. P.; Gajan, D.; Lesage, A.; Emsley, L.; Coperet, C.; Schrock, R. R. J. Am. Chem. Soc. 2013, 135 (51), 19068–19070. Geiger, Y.; Gottlieb, H. E.; Akbey, Ü.; Oschkinat, H.; Goobes, G. J. Am. Chem. Soc. 2016, 138 (17), 5561–5567. Samantaray, M. K.; D’Elia, V.; Pump, E.; Falivene, L.; Harb, M.; Ould Chikh, S.; Cavallo, L.; Basset, J.-M. Chem. Rev. 2020, 120 (2), 734–813. Lelli, M.; Gajan, D.; Lesage, A.; Caporini, M. A.; Vitzthum, V.; Miéville, P.; Héroguel, F.; Rascón, F.; Roussey, A.; Thieuleux, C.; Boualleg, M.; Veyre, L.; Bodenhausen, G.; Coperet, C.; Emsley, L. J. Am. Chem. Soc. 2011, 133 (7), 2104–2107. Conley, M. P.; Drost, R. M.; Baffert, M.; Gajan, D.; Elsevier, C.; Franks, W. T.; Oschkinat, H.; Veyre, L.; Zagdoun, A.; Rossini, A.; Lelli, M.; Lesage, A.; Casano, G.; Ouari, O.; Tordo, P.; Emsley, L.; Copéret, C.; Thieuleux, C. Chem. A Eur. J. 2013, 19 (37), 12234–12238. Berruyer, P.; Lelli, M.; Conley, M. P.; Silverio, D. L.; Widdifield, C. M.; Siddiqi, G.; Gajan, D.; Lesage, A.; Copéret, C.; Emsley, L. J. Am. Chem. Soc. 2017, 139 (2), 849–855.

Surface Organometallic Chemistry and Catalysis

529

90. Lafon, O.; Thankamony, A. S. L.; Kobayashi, T.; Carnevale, D.; Vitzthum, V.; Slowing, I. I.; Kandel, K.; Vezin, H.; Amoureux, J.-P.; Bodenhausen, G.; Pruski, M. J. Phys. Chem. C 2013, 117 (3), 1375–1382. 91. Eedugurala, N.; Wang, Z.; Chaudhary, U.; Nelson, N.; Kandel, K.; Kobayashi, T.; Slowing, I. I.; Pruski, M.; Sadow, A. D. ACS Catal. 2015, 5 (12), 7399–7414. 92. Gutmann, T.; Liu, J.; Rothermel, N.; Xu, Y.; Jaumann, E.; Werner, M.; Breitzke, H.; Sigurdsson, S. T.; Buntkowsky, G. Chem. A Eur. J. 2015, 21 (9), 3798–3805. 93. Romanenko, I.; Gajan, D.; Sayah, R.; Crozet, D.; Jeanneau, E.; Lucas, C.; Leroux, L.; Veyre, L.; Lesage, A.; Emsley, L.; Lacôte, E.; Thieuleux, C. Angew. Chem. Int. Ed. 2015, 54 (44), 12937–12941. 94. Werner, M.; Heil, A.; Rothermel, N.; Breitzke, H.; Groszewicz, P. B.; Thankamony, A. S.; Gutmann, T.; Buntkowsky, G. Solid State Nucl. Magn. Reson. 2015, 72, 73–78. 95. Hamzaoui, B.; Bendjeriou-Sedjerari, A.; Pump, E.; Abou-Hamad, E.; Sougrat, R.; Gurinov, A.; Huang, K.-W.; Gajan, D.; Lesage, A.; Emsley, L.; Basset, J.-M. Chem. Sci. 2016, 7 (9), 6099–6105. 96. Ong, T.-C.; Liao, W.-C.; Mougel, V.; Gajan, D.; Lesage, A.; Emsley, L.; Coperet, C. Angew. Chem. Int. Ed. 2016, 55 (15), 4743–4747. 97. Pump, E.; Viger-Gravel, J.; Abou-Hamad, E.; Samantaray, M. K.; Hamzaoui, B.; Gurinov, A.; Anjum, D. H.; Gajan, D.; Lesage, A.; Bendjeriou-Sedjerari, A.; Emsley, L.; Basset, J.M. Chem. Sci. 2017, 8 (1), 284–290. 98. Liao, W.-C.; Ong, T.-C.; Gajan, D.; Bernada, F.; Sauvée, C.; Yulikov, M.; Pucino, M.; Schowner, R.; Schwarzwälder, M.; Buchmeiser, M. R.; Jeschke, G.; Tordo, P.; Ouari, O.; Lesage, A.; Emsley, L.; Copéret, C. Chem. Sci. 2017, 8 (1), 416–422. 99. De Jong, K. P. Synthesis of Solid Catalysts; John Wiley & Sons, 2009. 100. Werghi, B.; Pump, E.; Tretiakov, M.; Abou-Hamad, E.; Gurinov, A.; Doggali, P.; Anjum, D. H.; Cavallo, L.; Bendjeriou-Sedjerari, A.; Basset, J.-M. Chem. Sci. 2018, 9 (14), 3531–3537. 101. Valentine, K. G.; Mathies, G.; Bédard, S.; Nucci, N. V.; Dodevski, I.; Stetz, M. A.; Can, T. V.; Griffin, R. G.; Wand, A. J. J. Am. Chem. Soc. 2014, 136 (7), 2800–2807. 102. Fernández-de-Alba, C.; Takahashi, H.; Richard, A.; Chenavier, Y.; Dubois, L.; Maurel, V.; Lee, D.; Hediger, S.; De Paëpe, G. Chem. A Eur. J. 2015, 21 (12), 4512–4517. 103. Takahashi, H.; Hediger, S.; De Paëpe, G. Chem. Commun. 2013, 49 (82), 9479–9481. 104. Viger-Gravel, J.; Berruyer, P.; Gajan, D.; Basset, J. M.; Lesage, A.; Tordo, P.; Ouari, O.; Emsley, L. Angew. Chem. Int. Ed. 2017, 56 (30), 8726–8730. 105. Silverio, D. L.; van Kalkeren, H. A.; Ong, T.-C.; Baudin, M.; Yulikov, M.; Veyre, L.; Berruyer, P.; Chaudhari, S.; Gajan, D.; Baudouin, D.; Cavaillès, M.; Vuichoud, B.; Bornet, A.; Jeschke, G.; Bodenhausen, G.; Lesage, A.; Emsley, L.; Jannin, S.; Thieuleux, C.; Copéret, C. Helv. Chim. Acta 2017, 100 (6), e1700101. 106. Pump, E.; Bendjeriou-Sedjerari, A.; Viger-Gravel, J.; Gajan, D.; Scotto, B.; Samantaray, M. K.; Abou-Hamad, E.; Gurinov, A.; Almaksoud, W.; Cao, Z.; Lesage, A.; Cavallo, L.; Emsley, L.; Basset, J.-M. Chem. Sci. 2018, 9 (21), 4866–4872. 107. Yadav, N.; Seidi, F.; Del Gobbo, S.; D’Elia, V.; Crespy, D. Polym. Chem. 2019, 10 (26), 3571–3584. 108. Zhuravlev, L. T. Eng. Aspects 2000, 173 (1), 1–38. 109. Bax, A.; Freeman, R.; Kempsell, S. P. J. Am. Chem. Soc. 1980, 102 (14), 4849–4851. 110. Lesage, A.; Auger, C.; Caldarelli, S.; Emsley, L. J. Am. Chem. Soc. 1997, 119 (33), 7867–7868. 111. Lesage, A.; Bardet, M.; Emsley, L. J. Am. Chem. Soc. 1999, 121 (47), 10987–10993. 112. Rataboul, F.; Baudouin, A.; Thieuleux, C.; Veyre, L.; Copéret, C.; Thivolle-Cazat, J.; Basset, J.-M.; Lesage, A.; Emsley, L. J. Am. Chem. Soc. 2004, 126 (39), 12541–12550. 113. Schnell, I.; Lupulescu, A.; Hafner, S.; Demco, D. E.; Spiess, H. W. J. Magn. Reson. 1998, 133 (1), 61–69. 114. Graf, R.; Demco, D. E.; Gottwald, J.; Hafner, S.; Spiess, H. W. J. Chem. Phys. 1997, 106 (3), 885–895. 115. Avenier, P.; Lesage, A.; Taoufik, M.; Baudouin, A.; De Mallmann, A.; Fiddy, S.; Vautier, M.; Veyre, L.; Basset, J.-M.; Emsley, L.; Quadrelli, E. A. J. Am. Chem. Soc. 2007, 129 (1), 176–186. 116. Friedrich, U.; Schnell, I.; Demco, D. E.; Spiess, H. W. Chem. Phys. Lett. 1998, 285 (1), 49–58. 117. Chabanas, M.; Quadrelli, E. A.; Fenet, B.; Coperet, C.; Thivolle-Cazat, J.; Basset, J.-M.; Lesage, A.; Emsley, L. Angew. Chem. Int. Ed. Engl. 2001, 40 (23), 4493–4496. 118. Ishii, Y.; Tycko, R. J. Magn. Reson. 2000, 142 (1), 199–204. 119. Gerstein, B. C.; Pembleton, R. G.; Wilson, R. C.; Ryan, L. M. J. Chem. Phys. 1977, 66 (1), 361–362. 120. Blanc, F.; Copéret, C.; Thivolle-Cazat, J.; Basset, J.-M.; Lesage, A.; Emsley, L.; Sinha, A.; Schrock, R. R. Angew. Chem. Int. Ed. 2006, 45 (8), 1216–1220. 121. Lafon, O.; Rosay, M.; Aussenac, F.; Lu, X.; Trébosc, J.; Cristini, O.; Kinowski, C.; Touati, N.; Vezin, H.; Amoureux, J.-P. Angew. Chem. Int. Ed. 2011, 50 (36), 8367–8370. 122. Vitzthum, V.; Miéville, P.; Carnevale, D.; Caporini, M. A.; Gajan, D.; Copéret, C.; Lelli, M.; Zagdoun, A.; Rossini, A. J.; Lesage, A.; Emsley, L.; Bodenhausen, G. Chem. Commun. 2012, 48 (14), 1988–1990. 123. Lee, D.; Duong, N. T.; Lafon, O.; De Paëpe, G. J. Phys. Chem. C 2014, 118 (43), 25065–25076. 124. Lee, D.; Takahashi, H.; Thankamony, A. S. L.; Dacquin, J.-P.; Bardet, M.; Lafon, O.; De Paëpe, G. J. Am. Chem. Soc. 2012, 134 (45), 18491–18494. 125. Brodrecht, M.; Kumari, B.; Thankamony, A. S. S. L.; Breitzke, H.; Gutmann, T.; Buntkowsky, G. Chem. A Eur. J. 2019, 25 (20), 5214–5221. 126. Lilly Thankamony, A. S.; Lion, C.; Pourpoint, F.; Singh, B.; Perez Linde, A. J.; Carnevale, D.; Bodenhausen, G.; Vezin, H.; Lafon, O.; Polshettiwar, V. Angew. Chem. Int. Ed. 2015, 54 (7), 2190–2193. 127. Aljuhani, M. A.; Barman, S.; Abou-Hamad, E.; Gurinov, A.; Ould-Chikh, S.; Guan, E.; Jedidi, A.; Cavallo, L.; Gates, B. C.; Pelletier, J. D. A.; Basset, J. M. Acs Catal. 2018, 8 (10), 9440–9446. 128. Michaelis, V. K.; Corzilius, B.; Smith, A. A.; Griffin, R. G. J. Phys. Chem. B 2013, 117 (48), 14894–14906. 129. Merle, N.; Trébosc, J.; Baudouin, A.; Rosal, I. D.; Maron, L.; Szeto, K.; Genelot, M.; Mortreux, A.; Taoufik, M.; Delevoye, L.; Gauvin, R. M. J. Am. Chem. Soc. 2012, 134 (22), 9263–9275. 130. Grekov, D.; Bouhoute, Y.; Szeto, K. C.; Merle, N.; De Mallmann, A.; Lefebvre, F.; Lucas, C.; Del Rosal, I.; Maron, L.; Gauvin, R. M.; Delevoye, L.; Taoufik, M. Organometallics 2016, 35 (13), 2188–2196. 131. Merle, N.; Girard, G.; Popoff, N.; De Mallmann, A.; Bouhoute, Y.; Trébosc, J.; Berrier, E.; Paul, J.-F.; Nicholas, C. P.; Del Rosal, I.; Maron, L.; Gauvin, R. M.; Delevoye, L.; Taoufik, M. Inorg. Chem. 2013, 52 (17), 10119–10130. 132. Perras, F. A.; Chaudhary, U.; Slowing, I. I.; Pruski, M. J. Phys. Chem. C 2016, 120 (21), 11535–11544. 133. Perras, F. A.; Kobayashi, T.; Pruski, M. J. Am. Chem. Soc. 2015, 137 (26), 8336–8339. 134. Sommer, J.; Yang, Y.; Rambow, D.; Blümel, J. Inorg. Chem. 2004, 43 (24), 7561–7563. 135. Hilliard, C. R.; Bhuvanesh, N.; Gladysz, J. A.; Blümel, J. Dalton Trans. 2012, 41 (6), 1742–1754. 136. Chabanas, M.; Baudouin, A.; Coperet, C.; Basset, J.-M.; Lukens, W.; Lesage, A.; Hediger, S.; Emsley, L. J. Am. Chem. Soc. 2003, 125 (2), 492–504. 137. Bendjeriou-Sedjerari, A.; Pelletier, J. D. A.; Abou-hamad, E.; Emsley, L.; Basset, J.-M. Chem. Commun. 2012, 48 (25), 3067–3069. 138. Lei, S.; Yan, Z.; Heyong, H. Chem. Lett. 2001, 30 (11), 1164–1165. 139. Samantaray, M. K.; Callens, E.; Abou-Hamad, E.; Rossini, A. J.; Widdifield, C. M.; Dey, R.; Emsley, L.; Basset, J.-M. J. Am. Chem. Soc. 2014, 136 (3), 1054–1061. 140. Pelletier, J.; Espinas, J.; Vu, N.; Norsic, S.; Baudouin, A.; Delevoye, L.; Trébosc, J.; Le Roux, E.; Santini, C.; Basset, J.-M.; Gauvin, R. M.; Taoufik, M. Chem. Commun. 2011, 47 (10), 2979–2981. 141. Werghi, B.; Bendjeriou-Sedjerari, A.; Jedidi, A.; Abou-Hamad, E.; Cavallo, L.; Basset, J.-M. Organometallics 2016, 35 (19), 3288–3294. 142. Zhang, F.; Szeto, K. C.; Taoufik, M.; Delevoye, L.; Gauvin, R. M.; Scott, S. L. J. Am. Chem. Soc. 2018, 140 (42), 13854–13868. 143. Gallo, A.; Fong, A.; Szeto, K. C.; Rieb, J.; Delevoye, L.; Gauvin, R. M.; Taoufik, M.; Peters, B.; Scott, S. L. J. Am. Chem. Soc. 2016, 138 (39), 12935–12947. 144. Mougel, V.; Copéret, C. Chem. Sci. 2014, 5 (6), 2475–2481. 145. Emid, S.; Bax, A.; Konijnendijk, J.; Smidt, J.; Pines, A. Phys. Ther. 1979, 96 (3), 333–336. 146. Munowitz, M.; Pines, A. Science 1986, 233 (4763), 525–531.

530

Surface Organometallic Chemistry and Catalysis

147. Pfennig, V.; Seppelt, K. Science 1996, 271 (5249), 626–628. 148. Samantaray, M. K.; Dey, R.; Kavitake, S.; Abou-Hamad, E.; Bendjeriou-Sedjerari, A.; Hamieh, A.; Basset, J.-M. J. Am. Chem. Soc. 2016, 138 (27), 8595–8602. 149. Venkatesh, A.; Lund, A.; Rochlitz, L.; Jabbour, R.; Gordon, C. P.; Menzildjian, G.; Viger-Gravel, J.; Berruyer, P.; Gajan, D.; Copéret, C.; Lesage, A.; Rossini, A. J. J. Am. Chem. Soc. 2020, 142 (44), 18936–18945. 150. Ruddy, D. A.; Jarupatrakorn, J.; Rioux, R. M.; Miller, J. T.; McMurdo, M. J.; McBee, J. L.; Tupper, K. A.; Tilley, T. D. Chem. Mater. 2008, 20 (20), 6517–6527. 151. Heckelsberg, L. F.; Banks, R. L.; Bailey, G. C. Ind. Eng. Chem. Prod. Res. Dev. 1968, 7 (1), 29–31. 152. Burnett, R. L.; Hughes, T. R. J. Catal. 1973, 31 (1), 55–64. 153. Vidal, V.; Théolier, A.; Thivolle-Cazat, J.; Basset, J.-M. Science 1997, 276 (5309), 99–102. 154. Vidal, V.; Théolier, A.; Thivolle-Cazat, J.; Basset, J.-M.; Corker, J. J. Am. Chem. Soc. 1996, 118 (19), 4595–4602. 155. Basset, J. M.; Copéret, C.; Lefort, L.; Maunders, B. M.; Maury, O.; Le Roux, E.; Saggio, G.; Soignier, S.; Soulivong, D.; Sunley, G. J.; Taoufik, M.; Thivolle-Cazat, J. J. Am. Chem. Soc. 2005, 127 (24), 8604–8605. 156. Le Roux, E.; Taoufik, M.; Chabanas, M.; Alcor, D.; Baudouin, A.; Coperet, C.; Thivolle-Cazat, J.; Basset, J.-M.; Lesage, A.; Hediger, S.; Emsley, L. Organometallics 2005, 24 (17), 4274–4279. 157. Le Roux, E.; Taoufik, M.; Copéret, C.; de Mallmann, A.; Thivolle-Cazat, J.; Basset, J.-M.; Maunders, B. M.; Sunley, G. J. Angew. Chem. Int. Ed. 2005, 44 (41), 6755–6758. 158. Chen, Y.; Abou-hamad, E.; Hamieh, A.; Hamzaoui, B.; Emsley, L.; Basset, J.-M. J. Am. Chem. Soc. 2015, 137 (2), 588–591. 159. Maity, N.; Barman, S.; Callens, E.; Samantaray, M. K.; Abou-Hamad, E.; Minenkov, Y.; D’Elia, V.; Hoffman, A. S.; Widdifield, C. M.; Cavallo, L.; Gates, B. C.; Basset, J.-M. Chem. Sci. 2016, 7 (2), 1558–1568. 160. Callens, E.; Abou-Hamad, E.; Riache, N.; Basset, J. M. Chem. Commun. 2014, 50 (30), 3982–3985. 161. Samantaray, M. K.; Dey, R.; Abou-Hamad, E.; Hamieh, A.; Basset, J.-M. Chem. A Eur. J. 2015, 21 (16), 6100–6106. 162. Riache, N.; Callens, E.; Samantaray, M. K.; Kharbatia, N. M.; Atiqullah, M.; Basset, J.-M. Chem. A Eur. J. 2014, 20 (46), 15089–15094. 163. Pump, E.; Cao, Z.; Samantaray, M. K.; Bendjeriou-Sedjerari, A.; Cavallo, L.; Basset, J.-M. ACS Catal. 2017, 7 (10), 6581–6586. 164. Riache, N.; Callens, E.; Samantaray, M. K.; Kharbatia, N. M.; Atiqullah, M.; Basset, J.-M. Chem. A Eur. J. 2015, 21 (24), 8656. 165. Samantaray, M. K.; Kavitake, S.; Morlanes, N.; Abou-Hamad, E.; Hamieh, A.; Dey, R.; Basset, J.-M. J. Am. Chem. Soc. 2017, 139 (9), 3522–3527. 166. Corker, J.; Lefebvre, F.; Lécuyer, C.; Dufaud, V.; Quignard, F.; Choplin, A.; Evans, J.; Basset, J.-M. Science 1996, 271 (5251), 966–969. 167. Lecuyer, C.; Quignard, F.; Choplin, A.; Olivier, D.; Basset, J.-M. Angew. Chem. Int. Ed. 1991, 30 (12), 1660–1661. 168. Rosier, C.; Niccolai, G. P.; Basset, J.-M. J. Am. Chem. Soc. 1997, 119 (50), 12408–12409. 169. Tosin, G.; Santini, C. C.; Baudouin, A.; De Mallman, A.; Fiddy, S.; Dablemont, C.; Basset, J.-M. Organometallics 2007, 26 (17), 4118–4127. 170. Norsic, S.; Larabi, C.; Delgado, M.; Garron, A.; de Mallmann, A.; Santini, C.; Szeto, K. C.; Basset, J.-M.; Taoufik, M. Cat. Sci. Technol. 2012, 2 (1), 215–219. 171. Saidi, A.; Al Maksoud, W.; Samantaray, M. K.; Abou-Hamad, E.; Basset, J.-M. Chem. Commun. 2020, 56 (87), 13401–13404. 172. Morlanés, N.; Kavitake, S. G.; Rosenfeld, D. C.; Basset, J.-M. ACS Catal. 2019, 9 (2), 1274–1282. 173. Hamzaoui, B.; Pelletier, J. D. A.; Abou-Hamad, E.; Basset, J.-M. Chem. Commun. 2016, 52 (25), 4617–4620. 174. Aljuhani, M. A.; Zhang, Z.; Barman, S.; El Eter, M.; Failvene, L.; Ould-Chikh, S.; Guan, E.; Abou-Hamad, E.; Emwas, A.-H.; Pelletier, J. D. A.; Gates, B. C.; Cavallo, L.; Basset, J.M. ACS Catal. 2019, 9 (9), 8719–8725. 175. Polshettiwar, V.; Thivolle-Cazat, J.; Taoufik, M.; Stoffelbach, F.; Norsic, S.; Basset, J.-M. Angew. Chem. Int. Ed. 2011, 50 (12), 2747–2751. 176. Tretiakov, M.; Samantaray, M. K.; Saidi, A.; Basset, J. M. J. Organomet. Chem. 2018, 863, 102–108. 177. Saidi, A.; Samantaray, M. K.; Tretiakov, M.; Kavitake, S.; Basset, J. M. ChemCatChem 2018, 10 (8), 1912–1917. 178. Klerke, A.; Christensen, C. H.; Nørskov, J. K.; Vegge, T. J. Mater. Chem. 2008, 18 (20), 2304–2310. 179. Kammert, J.; Moon, J.; Cheng, Y.; Daemen, L.; Irle, S.; Fung, V.; Liu, J.; Page, K.; Ma, X.; Phaneuf, V.; Tong, J.; Ramirez-Cuesta, A. J.; Wu, Z. J. Am. Chem. Soc. 2020, 142 (16), 7655–7667. 180. Wang, L.; Xia, M.; Wang, H.; Huang, K.; Qian, C.; Maravelias, C. T.; Ozin, G. A. Joule 2018, 2 (6), 1055–1074. 181. Kobayashi, Y.; Tang, Y.; Kageyama, T.; Yamashita, H.; Masuda, N.; Hosokawa, S.; Kageyama, H. J. Am. Chem. Soc. 2017, 139 (50), 18240–18246. 182. Abedin, A.; Radenahmad, N.; Cheok, Q.; Shams, S.; Kim, J.; Azad, A. Renew. Sustain. Energy Rev. 2016, 60, 822–835. 183. Morlanés, N.; Katikaneni, S. P.; Paglieri, S. N.; Harale, A.; Solami, B.; Sarathy, S. M.; Gascon, J. Chem. Eng. J. 2021, 408, 127310. 184. Smith, C.; Hill, A. K.; Torrente-Murciano, L. Energ. Environ. Sci. 2020, 13 (2), 331–344. 185. Jackson, R. B.; Canadell, J. G.; Le Quéré, C.; Andrew, R. M.; Korsbakken, J. I.; Peters, G. P.; Nakicenovic, N. Nat. Clim. Change 2016, 6 (1), 7–10. 186. Wang, Q.; Guo, J.; Chen, P. J. Energy Chem. 2019, 36, 25–36. 187. Erisman, J. W.; Sutton, M. A.; Galloway, J.; Klimont, Z.; Winiwarter, W. Nat. Geosci. 2008, 1, 636. 188. Mebrahtu, C.; Krebs, F.; Abate, S.; Perathoner, S.; Centi, G.; Palkovits, R. In CO2 Stud. Surf. Sci. Catal; Albonetti, S., Perathoner, S., Quadrelli, E. A., Eds.; Elsevier, 2019; vol. 178; pp 85–103. 189. Kandemir, T.; Schuster, M. E.; Senyshyn, A.; Behrens, M.; Schlögl, R. Angew. Chem. Int. Ed. 2013, 52 (48), 12723–12726. 190. Mittasch, A.; Frankenburg, W. In Advances in Catalysis; Frankenburg, W. G., Komarewsky, V. I., Rideal, E. K., Eds.; Academic Press, 1950; vol. 2; pp 81–104. 191. Howard, J. B.; Rees, D. C. Proc. Natl. Acad. Sci. U. S. A. 2006, 103 (46), 17088–17093. 192. Yandulov, D. V.; Schrock, R. R. Science 2003, 301 (5629), 76–78. 193. Avenier, P.; Taoufik, M.; Lesage, A.; Solans-Monfort, X.; Baudouin, A.; de Mallmann, A.; Veyre, L.; Basset, J.-M.; Eisenstein, O.; Emsley, L.; Quadrelli, E. A. Science 2007, 317 (5841), 1056–1060. 194. Liu, H. Ammonia Synthesis Catalysts Innovation and Practice; World Scientific, 2013. 195. van der Ham, C. J. M.; Koper, M. T. M.; Hetterscheid, D. G. H. Chem. Soc. Rev. 2014, 43 (15), 5183–5191. 196. Qiu, W.; Xie, X.-Y.; Qiu, J.; Fang, W.-H.; Liang, R.; Ren, X.; Ji, X.; Cui, G.; Asiri, A. M.; Cui, G.; Tang, B.; Sun, X. Nat. Commun. 2018, 9 (1), 3485. 197. Jacobsen, C. J. H. Chem. Commun. 2000, 12, 1057–1058. 198. Jacobsen, C. J. H.; Dahl, S.; Clausen, B. S.; Bahn, S.; Logadottir, A.; Nørskov, J. K. J. Am. Chem. Soc. 2001, 123 (34), 8404–8405. 199. Kitano, M.; Inoue, Y.; Sasase, M.; Kishida, K.; Kobayashi, Y.; Nishiyama, K.; Tada, T.; Kawamura, S.; Yokoyama, T.; Hara, M.; Hosono, H. Angew. Chem. Int. Ed. 2018, 57 (10), 2648–2652. 200. Ogura, Y.; Sato, K.; Miyahara, S.-I.; Kawano, Y.; Toriyama, T.; Yamamoto, T.; Matsumura, S.; Hosokawa, S.; Nagaoka, K. Chem. Sci. 2018, 9 (8), 2230–2237. 201. Zeinalipour-Yazdi, C. D.; Hargreaves, J. S. J.; Catlow, C. R. A. J. Phys. Chem. C 2018, 122 (11), 6078–6082. 202. Inoue, Y.; Kitano, M.; Tokunari, M.; Taniguchi, T.; Ooya, K.; Abe, H.; Niwa, Y.; Sasase, M.; Hara, M.; Hosono, H. ACS Catal. 2019, 9 (3), 1670–1679. 203. Song, X.; Yang, P.; Wang, J.; Zhao, X.; Zhou, Y.; Li, Y.; Yang, L. Inorg. Chem. Frontiers 2019. 204. Sato, K.; Miyahara, S.-I.; Ogura, Y.; Tsujimaru, K.; Wada, Y.; Toriyama, T.; Yamamoto, T.; Matsumura, S.; Nagaoka, K. ACS Sustainable Chem. Eng. 2020, 8 (7), 2726–2734. 205. Wang, X.; Peng, X.; Chen, W.; Liu, G.; Zheng, A.; Zheng, L.; Ni, J.; Au, C.-T.; Jiang, L. Nat. Commun. 2020, 11 (1), 653. 206. Ye, T.-N.; Park, S.-W.; Lu, Y.; Li, J.; Sasase, M.; Kitano, M.; Tada, T.; Hosono, H. Nature 2020, 583 (7816), 391–395. 207. van Ommen, J. G.; Bolink, W. J.; Prasad, J.; Mars, P. J. Catal. 1975, 38 (1), 120–127. 208. Aika, K.; Kumasaka, M.; Oma, T.; Kato, O.; Matsuda, H.; Watanabe, N.; Yamazaki, K.; Ozaki, A.; Onishi, T. Appl. Catal. 1986, 28, 57–68. 209. Kitano, M.; Kanbara, S.; Inoue, Y.; Kuganathan, N.; Sushko, P. V.; Yokoyama, T.; Hara, M.; Hosono, H. Nat. Commun. 2015, 6, 6731. 210. Bielawa, H.; Hinrichsen, O.; Birkner, A.; Muhler, M. Angew. Chem. Int. Ed. 2001, 40 (6), 1061–1063.

Surface Organometallic Chemistry and Catalysis 211. 212. 213. 214. 215. 216. 217. 218. 219. 220. 221. 222. 223. 224. 225. 226. 227. 228. 229. 230. 231. 232. 233. 234. 235. 236. 237. 238. 239. 240. 241. 242. 243. 244. 245. 246. 247. 248. 249. 250. 251. 252. 253. 254. 255. 256. 257. 258. 259. 260. 261. 262. 263. 264. 265. 266. 267. 268. 269. 270. 271. 272. 273. 274. 275. 276. 277. 278. 279.

531

Aika, K.-I. Catal. Today 2017, 286, 14–20. The Chemical Physics of Solid Surfaces. Bowker, M., King, D. A., Woodruff, D. P., Eds.; In Chapter 7 - Promotion in Ammonia Synthesis; Elsevier, 1993; vol. 6; pp 225–268. Egeberg, R.; Dahl, S.; Logadottir, A.; Larsen, J. H.; Nørskov, J. K.; Chorkendorff, I. Surf. Sci. 2001, 491 (1-2), 183–194. Gong, Y.; Wu, J.; Kitano, M.; Wang, J.; Ye, T.-N.; Li, J.; Kobayashi, Y.; Kishida, K.; Abe, H.; Niwa, Y.; Yang, H.; Tada, T.; Hosono, H. Nat. Catal. 2018, 1 (3), 178–185. Wang, S.; Ichihara, F.; Pang, H.; Chen, H.; Ye, J. Adv. Funct. Mater. 2018, 28 (50), 1803309. Ertl, G.; Lee, S. B.; Weiss, M. Surf. Sci. 1982, 114 (2), 527–545. Ertl, G.; Prigge, D.; Schloegl, R.; Weiss, M. J. Catal. 1983, 79 (2), 359–377. Li, J.; Cheng, X.; Zhang, C.; Wang, J.; Dong, W.; Yang, Y.; Li, Y. J. Chem. Technol. Biotechnol. 2017, 92 (6), 1472–1480. Liu, J.-C.; Ma, X.-L.; Li, Y.; Wang, Y.-G.; Xiao, H.; Li, J. Nat. Commun. 2018, 9 (1), 1610. Reiners, M.; Baabe, D.; Münster, K.; Zaretzke, M.-K.; Freytag, M.; Jones, P. G.; Coppel, Y.; Bontemps, S.; Rosal, I. D.; Maron, L.; Walter, M. D. Nat. Chem. 2020. Solans-Monfort, X.; Chow, C.; Goure, E.; Kaya, Y.; Basset, J.-M.; Taoufik, M.; Quadrelli, E. A.; Eisenstein, O. Inorg. Chem. 2012, 51 (13), 7237–7249. Al Maksoud, W.; Rai, R. K.; Morlanés, N.; Harb, M.; Ahmad, R.; Ould-Chikh, S.; Anjum, D.; Hedhili, M. N.; Al-Sabban, B. E.; Albahily, K.; Cavallo, L.; Basset, J.-M. J. Catal. 2020. https://doi.org/10.1016/j.jcat.2020.10.031. Jia, H.-P.; Quadrelli, E. A. Chem. Soc. Rev. 2014, 43 (2), 547–564. Li, J.; Li, S. Angew. Chem. Int. Ed. 2008, 47 (42), 8040–8043. Shima, T.; Hu, S.; Luo, G.; Kang, X.; Luo, Y.; Hou, Z. Science 2013, 340 (6140), 1549–1552. Ballmann, J.; Munhá, R. F.; Fryzuk, M. D. Chem. Commun. 2010, 46 (7), 1013–1025. Tanabe, Y.; Nishibayashi, Y. Coord. Chem. Rev. 2013, 257 (17), 2551–2564. Azofra, L. M.; Morlanes, N.; Poater, A.; Samantaray, M. K.; Vidjayacoumar, B.; Albahily, K.; Cavallo, L.; Basset, J.-M. Angew. Chem. Int. Ed. 2019, 58 (20), 6476. Azofra, L. M.; Morlanes, N.; Poater, A.; Samantaray, M. K.; Vidjayacoumar, B.; Albahily, K.; Cavallo, L.; Basset, J.-M. Angew. Chem. Int. Ed. 2018, 57 (48), 15812–15816. Tanaka, H.; Arashiba, K.; Kuriyama, S.; Sasada, A.; Nakajima, K.; Yoshizawa, K.; Nishibayashi, Y. Nat. Commun. 2014, 5 (1), 3737. Eizawa, A.; Arashiba, K.; Tanaka, H.; Kuriyama, S.; Matsuo, Y.; Nakajima, K.; Yoshizawa, K.; Nishibayashi, Y. Nat. Commun. 2017, 8 (1), 14874. Li, X.; Ao, Z.; Liu, J.; Sun, H.; Rykov, A. I.; Wang, J. ACS Nano 2016, 10 (12), 11532–11540. Wang, M.; Liu, S.; Qian, T.; Liu, J.; Zhou, J.; Ji, H.; Xiong, J.; Zhong, J.; Yan, C. Nat. Commun. 2019, 10 (1), 341. Sellmann, D.; Sutter, J. Acc. Chem. Res. 1997, 30 (11), 460–469. Rodriguez, M. M.; Bill, E.; Brennessel, W. W.; Holland, P. L. Science 2011, 334 (6057), 780–783. Logadottir, A.; Rod, T.; Nørskov, J. K.; Hammer, B.; Dahl, S.; Jacobsen, C. J. H. J. Catal. 2001, 197, 229–231. Bligaard, T.; Nørskov, J. K.; Dahl, S.; Matthiesen, J.; Christensen, C. H.; Sehested, J. J. Catal. 2004, 224 (1), 206–217. Dahl, S.; Logadottir, A.; Jacobsen, C. J. H.; Nørskov, J. K. Appl. Catal. Gen. 2001, 222 (1), 19–29. Morlanes, N.; Almaksoud, W.; Rai, R. K.; Ould-Chikh, S.; Ali, M. M.; Al-Sabban, B. E.; Vidjayacoumar, B.; Al-Bahily, K.; Basset, J.-M. Cat. Sci. Technol. 2020, 10, 844–852. Miao, D.; Ding, Y.; Yu, T.; Li, J.; Pan, X.; Bao, X. ACS Catal. 2020, 10 (13), 7389–7397. Niziolek, A. M.; Onel, O.; Floudas, C. A. AIChE J. 2016, 62 (5), 1531–1556. Al Maksoud, W.; Gevers, L. E.; Vittenet, J.; Ould-Chikh, S.; Telalovic, S.; Bhatte, K.; Abou-Hamad, E.; Anjum, D. H.; Hedhili, M. N.; Vishwanath, V.; Alhazmi, A.; Almusaiteer, K.; Basset, J. M. Dalton Trans. 2019, 48 (19), 6611–6620. Plotkin, H. A. W. B. G. R. J. S. Industrial Organic Chemicals; Encyclopedia.com, 2012; pp 323–373. Plotkin, H. A. W. B. G. R. J. S. Industrial Organic Chemicals; Encyclopedia.com, 2012; pp 375–382. Plotkin, H. A. W. B. G. R. J. S. Industrial Organic Chemicals; Encyclopedia.com, 2012; pp 383–405. Xiao, H.; Zhang, J.; Wang, P.; Zhang, Z.; Zhang, Q.; Xie, H.; Yang, G.; Han, Y.; Tan, Y. RSC Adv. 2015, 5 (112), 92222–92233. Corbetta, M.; Manenti, F.; Pirola, C.; Tsodikov, M. V.; Chistyakov, A. V. Comput. Chem. Eng. 2014, 71, 457–466. Zaidi, H. A.; Pant, K. K. Catal. Today 2004, 96 (3), 155–160. Rownaghi, A. A.; Hedlund, J. Ind. Eng. Chem. Res. 2011, 50 (21), 11872–11878. Vollmer, I.; Yarulina, I.; Kapteijn, F.; Gascon, J. ChemCatChem 2018. [Ahead of Print]. De Klerck, A. Fischer-Tropsch Refining; Wiley-VCH Verlag GmbH & Co. KGaA, 2011; pp 301–334. De Klerck, A. Fischer-Tropsch Refining; Wiley-VCH Verlag GmbH & Co. KGaA, 2011; pp 393–406. De Klerck, A. Fischer-Tropsch Refining; Wiley-VCH Verlag GmbH & Co. KGaA, 2011; pp 441–463. Robert, A.; Meyers, P. D. Handbook of Petroleum Refining Processes, 4th ed.; McGraw-Hill Education: New York, 2016. Yarulina, I.; Bailleul, S.; Pustovarenko, A.; Martinez, J. R.; De Wispelaere, K.; Hajek, J.; Weckhuysen, B. M.; Houben, K.; Baldus, M.; Van Speybroeck, V.; Kapteijn, F.; Gascon, J. ChemCatChem 2016, 8 (19), 3057–3063. Haag, W. O.; Lago, R. M.; Rodewald, P. G. J. Mol. Catal. 1982, 17 (2), 161–169. Mokrani, T.; Scurrell, M. Catal. Rev. 2009, 51 (1), 1–145. Wannapakdee, W.; Wattanakit, C.; Paluka, V.; Yutthalekha, T.; Limtrakul, J. RSC Adv. 2016, 6 (4), 2875–2881. Dehertog, W. J. H.; Fromen, G. F. Appl. Catal. A 1999, 189 (1), 63–75. Liu, R.-L.; Zhu, H.-Q.; Wu, Z.-W.; Qin, Z.-F.; Fan, W.-B.; Wang, J.-G. J. Fuel Chem. Technol. 2015, 43 (8), 961–969. Xiao, H.; Zhang, J.; Wang, X.; Zhang, Q.; Xie, H.; Han, Y.; Tan, Y. Cat. Sci. Technol. 2015, 5 (8), 4081–4090. Szeto, K. C.; Gallo, A.; Hernandez-Morejudo, S.; Olsbye, U.; De Mallmann, A.; Lefebvre, F.; Gauvin, R. M.; Delevoye, L.; Scott, S. L.; Taoufik, M. J. Phys. Chem. C 2015, 119 (47), 26611–26619. Vosmerikova, L. N.; Danilova, I. G.; Vosmerikov, A. A.; Barbashin, Y. E.; Vosmerikov, A. V. Pet. Chem. 2018, 58 (3), 237–244. Fan, H.; Nie, X.; Wang, H.; Janik, M. J.; Song, C.; Guo, X. Cat. Sci. Technol. 2020, 10, 8359–8373. Palcic, A.; Valtchev, V. Appl. Catal. Gen. 2020, 606, 117795. Tian, P.; Wei, Y.; Ye, M.; Liu, Z. ACS Catal. 2015, 5 (3), 1922–1938. Zaarour, M.; Cazemier, J.; Ruiz-Martínez, J. Cat. Sci. Technol. 2020, 10, 8140–8172. Hoff, T. C.; Thilakaratne, R.; Gardner, D. W.; Brown, R. C.; Tessonnier, J.-P. J. Phys. Chem. C 2016, 120 (36), 20103–20113. Petrovic, I.; Navrotsky, A.; Davis, M. E.; Zones, S. I. Chem. Mater. 1993, 5 (12), 1805–1813. Yeung, K. L.; Han, W. Zeolites and Molecular Sieves in Fuel Cell Applications; Wiley-VCH Verlag GmbH & Co. KGaA, 2010; pp 827–861. Schulz, P.; Baerns, M. Appl. Catal. 1991, 78 (1), 15–29. Giannetto, G.; Monque, R.; Galiasso, R. Catal. Rev. Sci. Eng. 1994, 36 (2), 271–304. Choudhary, V. R.; Devadas, P.; Kinage, A. K.; Sivadinarayana, C.; Guisnet, M. J. Catal. 1996, 158 (2), 537–550. Choudhary, V. R.; Kinage, A. K.; Choudhary, T. V. Chem. Commun. 1996, 22, 2545–2546. El-Malki, E.-M.; van Santen, R. A.; Sachtler, W. M. H. J. Phys. Chem. B 1999, 103 (22), 4611–4622. Yuan, S.; Derouane-Abd Hamid, S. B.; Li, Y.; Ying, P.; Xin, Q.; Derouane, E. G.; Li, C. J. Mol. Catal. A: Chem. 2002, 184 (1), 257–266. Vu, B. K.; Song, M. B.; Ahn, I. Y.; Suh, Y.-W.; Suh, D. J.; Kim, J. S.; Shin, E. W. J. Ind. Eng. Chem. 2011, 17 (1), 71–76. Wang, S.; Huang, Z.; Luo, Y.; Wang, J.; Fang, Y.; Hua, W.; Yue, Y.; Xu, H.; Shen, W. Cat. Sci. Technol. 2020, 10 (19), 6562–6572. Bhan, A.; Nicholas Delgass, W. Catal. Rev. 2008, 50 (1), 19–151.

532 280. 281. 282. 283. 284. 285. 286. 287. 288. 289. 290. 291. 292. 293. 294. 295. 296. 297. 298. 299. 300. 301. 302. 303. 304. 305. 306. 307. 308. 309. 310. 311. 312. 313. 314. 315. 316. 317. 318. 319. 320. 321. 322. 323. 324. 325. 326. 327. 328. 329. 330. 331. 332. 333. 334. 335. 336. 337. 338. 339. 340. 341.

342. 343. 344.

Surface Organometallic Chemistry and Catalysis Almutairi, S. M. T.; Mezari, B.; Magusin, P. C. M. M.; Pidko, E. A.; Hensen, E. J. M. ACS Catal. 2012, 2 (1), 71–83. Nagamori, Y.; Kawase, M. Micropor. Mesopor. Mat. 1998, 21 (4), 439–445. Guisnet, M.; Gnep, N. S. Catal. Today 1996, 31 (3), 275–292. Choudhary, V. R.; Sivadinarayana, C.; Devadas, P.; Sansare, S. D.; Magnoux, P.; Guisnet, M. Micropor. Mesopor. Mat. 1998, 21 (1-3), 91–101. Fricke, R.; Kosslick, H.; Lischke, G.; Richter, M. Chem. Rev. 2000, 100 (6), 2303–2405. Pidko, E. A.; Kazansky, V. B.; Hensen, E. J. M.; Van Santen, R. A. J. Catal. 2006, 240 (1), 73–84. Al-Zahrani, S. Dev. Chem. Eng. Mineral Process 2008, 6, 101–120. Adachi, K. Kagaku Kogaku 1994, 58 (2), 140–142. Wakai, T.; Hirabayashi, K.; Kondoh, T.; Inoue, S. Stud. Surf. Sci. Catal. 1995, 92, 449–452. Hirabayashi, K.; Kondo, T. Aromatik Kusu 1998, 50 (7/8), 265–273. Mowry, J. R.; Martindale, D. C.; Hall, A. H. P. Arabian J. Sci. Eng. 1985, 10 (4), 367–375. Mowry, J. R.; Weiszmann, J. A.; Johnson, J. A. Proc., Gas Process. Assoc. 1986, 65–70. Doolan, P. C.; Pujado, P. R. Hydrocarbon Process., Int. Ed. 1989, 68 (9), 72–74. 76. Gosling, C. D.; Wilcher, F. P.; Sullivan, L.; Mountford, R. A. Hydrocarbon Process., Int. Ed. 1991, 70 (12), 69–72. Imai, T.; Kocal, J. A.; Gosling, C. D.; Hall, A. H. P. The UOP-BP Cyclar Process. 1994, 21, 144–151. Marui, M. UOP/BP Cyclar Process. Kagaku Kogaku 1994, 58 (2), 143–145. Meriaudeau, P.; Naccache, C. Catal. Rev. Sci. Eng. 1997, 39 (1 & 2), 5–48. Kazansky, V. B.; Subbotina, I. R.; van Santen, R. A.; Hensen, E. J. M. J. Catal. 2005, 233 (2), 351–358. Choudhary, T. V.; Kinage, A.; Banerjee, S.; Choudhary, V. R. Energy Fuel 2006, 20 (3), 919–922. Rane, N.; Overweg, A. R.; Kazansky, V. B.; Van Santen, R. A.; Hensen, E. J. M. J. Catal. 2006, 239 (2), 478–485. Garcia-Sanchez, M.; Magusin, P. C. M. M.; Hensen, E. J. M.; Thune, P. C.; Rozanska, X.; Van Santen, R. A. J. Catal. 2003, 219 (2), 352–361. Mohandas, J. C.; Abou-Hamad, E.; Callens, E.; Samantaray, M. K.; Gajan, D.; Gurinov, A.; Ma, T.; Ould-Chikh, S.; Hoffman, A. S.; Gates, B. C.; Basset, J.-M. Chem. Sci. 2017, 8 (8), 5650–5661. Szeto, K. C.; Jones, Z. R.; Merle, N.; Rios, C.; Gallo, A.; Le Quemener, F.; Delevoye, L.; Gauvin, R. M.; Scott, S. L.; Taoufik, M. ACS Catal. 2018, 8 (8), 7566–7577. Bayense, C. R.; van Hooff, J. H. C.; de Haan, J. W.; van de Ven, L. J. M.; Kentgens, A. P. M. Catal. Lett. 1993, 17 (3), 349–361. Fleischman, S. D.; Scott, S. L. J. Am. Chem. Soc. 2011, 133 (13), 4847–4855. Kazansky, V. B.; Subbotina, I. R.; van Santen, R. A.; Hensen, E. J. M. J. Catal. 2004, 227 (2), 263–269. Hensen, E. J. M.; Pidko, E. A.; Rane, N.; van Santen, R. A. In Studies in Surface Science and Catalysis; Xu, R., Gao, Z., Chen, J., Yan, W., Eds.; Elsevier, 2007; vol. 170; pp 1182–1189. Getsoian, A. B.; Das, U.; Camacho-Bunquin, J.; Zhang, G.; Gallagher, J. R.; Hu, B.; Cheah, S.; Schaidle, J. A.; Ruddy, D. A.; Hensley, J. E.; Krause, T. R.; Curtiss, L. A.; Miller, J. T.; Hock, A. S. Cat. Sci. Technol. 2016, 6 (16), 6339–6353. Hanna, T. E.; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc. 2004, 126 (45), 14688–14689. Collins, R. A.; Russell, A. F.; Mountford, P. Appl. Petrochem. Res. 2015, 5 (3), 153–171. Watson, P. L.; Roe, D. C. J. Am. Chem. Soc. 1982, 104 (23), 6471–6473. Yue, D.; Kim, M. A.; You, F. ACS Sustain. Chem. Eng. 2013, 1 (8), 1003–1014. Kosinov, N.; Uslamin, E. A.; Meng, L.; Parastaev, A.; Liu, Y.; Hensen, E. J. M. Angew. Chem. Int. Ed. 2019, 58 (21), 7068–7072. Zhang, Y.; Wu, S.; Xu, X.; Jiang, H. Cat. Sci. Technol. 2020, 10 (3), 835–843. Keith, J. A.; Henry, P. M. Angew. Chem. Int. Ed. 2009, 48 (48), 9038–9049. Imbao, J.; van Bokhoven, J. A.; Nachtegaal, M. Chem. Commun. 2020, 56 (9), 1377–1380. Auldridge, M. E.; McCarty, D. R.; Klee, H. J. Curr. Opin. Plant Biol. 2006, 9 (3), 315–321. Dhumal, R. S.; Biradar, S. V.; Paradkar, A. R.; York, P. Pharm. Res. 2008, 25 (12), 2835–2844. Braaz, R.; Armbruster, W.; Jendrossek, D. Appl. Environ. Microbiol. 2005, 71 (5), 2473–2478. Lara, M.; Mutti, F. G.; Glueck, S. M.; Kroutil, W. J. Am. Chem. Soc. 2009, 131 (15), 5368–5369. Salameh, A.; Copéret, C.; Basset, J.-M.; Böhm, V. P. W.; Röper, M. Adv. Synt. Catal. 2007, 349 (1–2), 238–242. Jean-Louis Hérisson, P.; Chauvin, Y. Die Makromolekulare Chemie 1971, 141 (1), 161–176. Le Quéméner, F.; Barman, S.; Merle, N.; Aljuhani, M. A.; Samantaray, M. K.; Saih, Y.; Szeto, K. C.; De Mallmann, A.; Minenkov, Y.; Huang, K.-W.; Cavallo, L.; Taoufik, M.; Basset, J.-M. ACS Catal. 2018, 8 (8), 7549–7555. Merle, N.; Le Quéméner, F.; Bouhoute, Y.; Szeto, K. C.; De Mallmann, A.; Barman, S.; Samantaray, M. K.; Delevoye, L.; Gauvin, R. M.; Taoufik, M.; Basset, J.-M. J. Am. Chem. Soc. 2017, 139 (6), 2144–2147. Chisholm, M.; Hoffman, D.; Huffman, J. J. Am. Chem. Soc. 1984, 106. Saint-Arroman, R. P.; Didillon, B.; de Mallmann, A.; Basset, J. M.; Lefebvre, F. Appl. Catal. Gen. 2008, 337 (1), 78–85. Saint-Arroman, R. P.; Basset, J.-M.; Lefebvre, F.; Didillon, B. Appl. Catal. Gen. 2005, 290 (1), 181–190. Yates, F. Org. Process Res. Dev. 2001, 5 (4), 456–457. Cavani, F. Angew. Chem. Int. Ed. 2014, 53 (30), 7707. Ugo, R.; Psaro, R.; Zanderighi, G. M.; Basset, J. M.; Theolier, A.; Smith, A. K. In Fundamental Research in Homogeneous Catalysis; Tsutsui, M., Ed.; Plenum Press: New York and London, 1979; vol. 3; pp 579–601. Ruddy, D. A.; Tilley, T. D. Chem. Commun. 2007, 32, 3350–3352. Ruddy, D. A.; Tilley, T. D. J. Am. Chem. Soc. 2008, 130 (33), 11088–11096. Guillo, P.; Lipschutz, M. I.; Fasulo, M. E.; Tilley, T. D. ACS Catal. 2017, 7 (4), 2303–2312. Morlanés, N.; Notestein, J. M. J. Catal. 2010, 275 (2), 191–201. Morlanes, N.; Notestein, J. M. Appl. Catal. Gen. 2010, 387 (1), 45–54. Zwaschka, G.; Rondelli, M.; Krause, M.; Rötzer, M. D.; Hedhili, M. N.; Heiz, U.; Basset, J. M.; Schweinberger, F. F.; D’Elia, V. New J. Chem. 2018, 42 (4), 3035–3041. Thornburg, N. E.; Nauert, S. L.; Thompson, A. B.; Notestein, J. M. ACS Catal. 2016, 6 (9), 6124–6134. Boronat, M.; Concepción, P.; Corma, A.; Renz, M. Catal. Today 2007, 121 (1), 39–44. Gallo, A.; Tiozzo, C.; Psaro, R.; Carniato, F.; Guidotti, M. J. Catal. 2013, 298, 77–83. Sanford, T.; Frumhoff, P.; Luers, A.; Gulledge, J. Nat. Clim. Change 2014, 4, 164–166. Dokania, A.; Ramirez, A.; Bavykina, A.; Gascon, J. ACS Energy Lett. 2019, 4 (1), 167–176. IPCC, Qin, D.; Plattner, G.-K.; Tignor, M.; Allen, S. K.; Boschung, J.; Nauels, A.; Xia, Y.; Bex, V.; Midgley, P. M. Climate Change 2013: The Physical Science Basis; In Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, United Kingdom/New York, NY, USA, Vol. 1535, 2013. Mac Dowell, N.; Fennell, P. S.; Shah, N.; Maitland, G. C. Nat. Clim. Change 2017, 7 (4), 243–249. He, M.; Sun, Y.; Han, B. Angew. Chem. Int. Ed. 2013, 52 (37), 9620–9633. Peters, M.; Köhler, B.; Kuckshinrichs, W.; Leitner, W.; Markewitz, P.; Müller, T. E. ChemSusChem 2011, 4 (9), 1216–1240.

Surface Organometallic Chemistry and Catalysis

533

345. Krishnamurthy, S.; Rao, V. R.; Guntuka, S.; Sharratt, P.; Haghpanah, R.; Rajendran, A.; Amanullah, M.; Karimi, I. A.; Farooq, S. AIChE J. 2014, 60 (5), 1830–1842. 346. Boot-Handford, M. E.; Abanades, J. C.; Anthony, E. J.; Blunt, M. J.; Brandani, S.; Mac Dowell, N.; Fernández, J. R.; Ferrari, M.-C.; Gross, R.; Hallett, J. P.; Haszeldine, R. S.; Heptonstall, P.; Lyngfelt, A.; Makuch, Z.; Mangano, E.; Porter, R. T. J.; Pourkashanian, M.; Rochelle, G. T.; Shah, N.; Yao, J. G.; Fennell, P. S. Energ. Environ. Sci. 2014, 7 (1), 130–189. 347. Barthel, A.; Saih, Y.; Gimenez, M.; Pelletier, J. D. A.; Kuhn, F. E.; D’Elia, V.; Basset, J.-M. Green Chem. 2016, 18 (10), 3116–3123. 348. Yang, Z.-Z.; He, L.-N.; Gao, J.; Liu, A.-H.; Yu, B. Energ. Environ. Sci. 2012, 5 (5), 6602–6639. 349. Ding, P.; Zhao, H.; Li, T.; Luo, Y.; Fan, G.; Chen, G.; Gao, S.; Shi, X.; Lu, S.; Sun, X. J. Mat. Chem. A 2020, 8 (42), 21947–21960. 350. Chen, C.; Khosrowabadi Kotyk, J. F.; Sheehan, S. W. Chem 2018, 4 (11), 2571–2586. 351. Álvarez, A.; Bansode, A.; Urakawa, A.; Bavykina, A. V.; Wezendonk, T. A.; Makkee, M.; Gascon, J.; Kapteijn, F. Chem. Rev. 2017, 117 (14), 9804–9838. 352. Song, C. Catal. Today 2006, 115 (1), 2–32. 353. Prieto, G. ChemSusChem 2017, 10 (6), 1056–1070. 354. von der Assen, N.; Jung, J.; Bardow, A. Energ. Environ. Sci. 2013, 6 (9), 2721–2734. 355. Pekdemir, T. In Carbon Dioxide Utilisation; Styring, P., Quadrelli, E. A., Armstrong, K., Eds.; Elsevier: Amsterdam, 2015; pp 253–272. 356. Sun, Y., Park, S.-E., Chang, J.-S., Lee, K.-W., Eds.; In Studies in Surface Science and Catalysis; Elsevier, 2004; vol. 153; pp 9–16. 357. Wei, T., Wang, M., Wei, W., Sun, Y., Zhong, B., Park, S.-E., Chang, J.-S., Lee, K.-W., Eds.; In Studies in Surface Science and Catalysis; Elsevier, 2004; vol. 153; pp 41–46. 358. Song, C.; Pan, W.; Srimat, S. T. In Environmental Challenges and Greenhouse Gas Control for Fossil Fuel Utilization in the 21st Century; Maroto-Valer, M. M., Song, C., Soong, Y., Eds.; Springer US: Boston, MA, 2002; pp 247–267. 359. Song, Q.-W.; Zhou, Z.-H.; He, L.-N. Green Chem. 2017, 19 (16), 3707–3728. 360. Centi, G.; Quadrelli, E. A.; Perathoner, S. Energ. Environ. Sci. 2013, 6 (6), 1711–1731. 361. Kulkarni, A. P.; Hos, T.; Landau, M. V.; Fini, D.; Giddey, S.; Herskowitz, M. Sustain. Energy Fuels 2021. 362. Artz, J.; Müller, T. E.; Thenert, K.; Kleinekorte, J.; Meys, R.; Sternberg, A.; Bardow, A.; Leitner, W. Chem. Rev. 2018, 118 (2), 434–504. 363. Bonin, J.; Maurin, A.; Robert, M. Coord. Chem. Rev. 2017, 334, 184–198. 364. Hu, J.; Buelens, L.; Theofanidis, S.-A.; Galvita, V. V.; Poelman, H.; Marin, G. B. J. CO2 Util. 2016, 16, 8–16. 365. Al Maksoud, W.; Saidi, A.; Samantaray, M. K.; Abou-Hamad, E.; Poater, A.; Ould-Chikh, S.; Guo, X.; Guan, E.; Ma, T.; Gates, B. C.; Basset, J.-M. Chem. Commun. 2020, 56 (24), 3528–3531. 366. Gajan, D.; Guillois, K.; Delichere, P.; Basset, J.-M.; Candy, J.-P.; Caps, V.; Coperet, C.; Lesage, A.; Emsley, L. J. Am. Chem. Soc. 2009, 131 (41), 14667–14669. 367. Zhou, L.; Li, L.; Wei, N.; Li, J.; Takanabe, K.; Basset, J.-M. ChemCatChem 2015, 7 (16), 2508–2516. 368. Al Sabban, B.; Falivene, L.; Kozlov, S. M.; Aguilar-Tapia, A.; Ould-Chikh, S.; Hazemann, J.-L.; Cavallo, L.; Basset, J.-M.; Takanabe, K. Appl Catal B 2017, 213, 177–189. 369. Zhou, L.; Enakonda, L. R.; Harb, M.; Saih, Y.; Aguilar-Tapia, A.; Ould-Chikh, S.; Hazemann, J.-L.; Li, J.; Wei, N.; Gary, D.; Del-Gallo, P.; Basset, J.-M. Appl Catal B 2017, 208, 44–59. 370. Aguilar-Tapia, A.; Ould-Chikh, S.; Lahera, E.; Prat, A.; Delnet, W.; Proux, O.; Kieffer, I.; Basset, J.-M.; Takanabe, K.; Hazemann, J.-L. Rev. Sci. Instrum. 2018, 89 (3), 035109/1–035109/8. 371. Hengne, A. M.; Samal, A. K.; Enakonda, L. R.; Harb, M.; Gevers, L. E.; Anjum, D. H.; Hedhili, M. N.; Saih, Y.; Huang, K.-W.; Basset, J.-M. ACS Omega 2018, 3 (4), 3688–3701. 372. Al Abdulghani, A. J.; Park, J.-H.; Kozlov, S. M.; Kang, D.-C.; Al Sabban, B.; Pedireddy, S.; Aguilar-Tapia, A.; Ould-Chikh, S.; Hazemann, J.-L.; Basset, J.-M.; Cavallo, L.; Takanabe, K. J. Catal. 2020, 392, 126–134. 373. Chen, H.; Ma, N.; Cheng, C.; Zhang, H.; Yuan, W.; Liu, P.; Feng, X.; Liu, J.; Yang, Q.; Zhou, S. Appl. Surf. Sci. 2020, 515, 146038. 374. Tang, Y.; Kobayashi, Y.; Tassel, C.; Yamamoto, T.; Kageyama, H. Adv. Energy Mater. 2018, 8 (23), 1800800. 375. Qureshi, Z. S.; Sarawade, P. B.; Hussain, I.; Zhu, H.; Al-Johani, H.; Anjum, D. H.; Hedhili, M. N.; Maity, N.; D’Elia, V.; Basset, J.-M. ChemCatChem 2016, 8 (9), 1671–1678. 376. Larmier, K.; Liao, W.-C.; Tada, S.; Lam, E.; Verel, R.; Bansode, A.; Urakawa, A.; Comas-Vives, A.; Copéret, C. Angew. Chem. Int. Ed. 2017, 56 (9), 2318–2323. 377. Lam, E.; Noh, G.; Chan, K. W.; Larmier, K.; Lebedev, D.; Searles, K.; Wolf, P.; Safonova, O. V.; Copéret, C. Chem. Sci. 2020, 11 (29), 7593–7598. 378. Kalhor, M. P.; Wischert, R.; Copéret, C.; Chermette, H. New J. Chem. 2014, 38 (8), 3717–3721. 379. Hutschka, F.; Dedieu, A.; Eichberger, M.; Fornika, R.; Leitner, W. J. Am. Chem. Soc. 1997, 119 (19), 4432–4443. 380. Comerford, J. W.; Ingram, I. D. V.; North, M.; Wu, X. Green Chem. 2015, 17 (4), 1966–1987. 381. Monassier, A.; D’Elia, V.; Cokoja, M.; Dong, H.; Pelletier, J. D. A.; Basset, J.-M.; Kühn, F. E. ChemCatChem 2013, 5 (6), 1321–1324. 382. Wilhelm, M. E.; Anthofer, M. H.; Reich, R. M.; D’Elia, V.; Basset, J.-M.; Herrmann, W. A.; Cokoja, M.; Kühn, F. E. Cat. Sci. Technol. 2014, 4 (6), 1638–1643. 383. Kelly, M. J.; Barthel, A.; Maheu, C.; Sodpiban, O.; Dega, F.-B.; Vummaleti, S. V. C.; Abou-Hamad, E.; Pelletier, J. D. A.; Cavallo, L.; D’Elia, V.; Basset, J.-M. J. CO2 Util. 2017, 20, 243–252. 384. Otto, A.; Grube, T.; Schiebahn, S.; Stolten, D. Energ. Environ. Sci. 2015, 8 (11), 3283–3297. 385. Sodpiban, O.; Del Gobbo, S.; Barman, S.; Aomchad, V.; Kidkhunthod, P.; Ould-Chikh, S.; Poater, A.; D’Elia, V.; Basset, J.-M. Cat. Sci. Technol. 2019, 9 (21), 6152–6165. 386. Sopeña, S.; Martin, E.; Escudero-Adán, E. C.; Kleij, A. W. ACS Catal. 2017, 7 (5), 3532–3539.

14.14 Common Precursors and Surface Mechanisms for Atomic Layer Deposition Seán Thomas Barrya, Peter George Gordona, and Vincent Vandalonb, aDepartment of Chemistry, Carleton University, Ottawa, ON, Canada; bDepartment of Applied Physics, Eindhoven University of Technology, Eindhoven, The Netherlands © 2022 Elsevier Ltd. All rights reserved.

14.14.1 Introduction 14.14.2 Alkyl compounds 14.14.3 Cyclopentadienyl compounds 14.14.4 Amide compounds 14.14.5 Chelate nitrogen compounds 14.14.6 Alkoxide compounds 14.14.7 Beta-diketonate compounds 14.14.8 Carbonyl compounds 14.14.9 Summary Acknowledgment References

534 535 538 540 542 544 545 547 548 548 548

14.14.1 Introduction Atomic layer deposition (ALD) has become a well-known thin film deposition technique that brings together the foundational work of atomic layer epitaxy and molecular layering into a technique that bridges engineering, material science, and metal-organic chemistry.1 Simply, ALD deposits films in a layer-by-layer fashion by introducing two or more precursor compounds sequentially in the gas phase (separated by “purge” steps to prevent them from interacting in the gas phase), and designing a deposition process such that each precursor encounters a monolayer of reactive sites at the surface to promote film growth. A typical cartoon of ALD (Fig. 1) shows surface bonded precursor moieties reacting with vapor phase compounds to advance film growth. The precursor is entrained to a reaction volume (typically a furnace) in the gas phase (Fig. 1A) where it undergoes chemical reaction at the surface and chemisorbs into a monolayer (Fig. 1B). After the gas phase is cleared of excess precursor and volatile side products, a second precursor is entrained (Fig. 1C) where it reacts with the previous monolayer to form a layer of the target film (Fig. 1D). This “ALD cycle” can be repeated to achieve a desired film thickness. Naturally, this is an idealized view of reactivity which in practice can be quite complicated, depending on the chemical nature of the precursors. Central to ALD is the chemical precursor, typically an inorganic or organometallic compound that has thermal characteristics favorable for use in a vapor phase deposition technique. Halide compounds of metals are often used due to their excellent thermal stability, low onset of volatility, and ease of synthesis and handling.2 However, halide ligands can produce problematic side products (particularly the conjugate acids of the halides, HX, which can be damaging to deposited films and parts of the reactor) and so organometallic species of many varieties have been studied and used in ALD processes (Fig. 2). In 2019, the Plasma and Materials Processing group at the Eindhoven University of Technology launched an online database of ALD processes, and this has become a valuable tool in surveying potential processes and precursors for ALD.3 This chapter will make use of this database to identify the main classes of organometallic compounds used in ALD and discuss the surface chemistry for chemisorption and film formation. To survey a wider variety of surface chemistries, we will use an expanded definition of “organometallic compound,” where organic ligands will be considered even if an MdC bond does not exist in the compound. This chapter is not meant to be comprehensive, and there are several ligand families that do not have representations here. Rather, we chose to focus on several conventional organometallic ligand families that would be familiar to all chemists, and tried to represent the variety of surface chemistries that are exhibit in ALD. This chapter will focus on alkyls, carbonyls, alkoxides, and amides. The precursors that are discussed can be homoleptic or heteroleptic, and so the heteroleptic compounds will be grouped with the ligand family that is most important to their surface chemistry. There is a myriad of ligands that do not fall under these strict headings, and there are three ligand families that are very important both to the context of precursor development, as well as the specific surface chemistries in ALD. These additional ligands are either haptic (the cyclopentadienyl ligand) or chelating (amidinates and beta-diketonates) in nature, and each highlight a different variety of surface chemistry. Finally, there are many ligands or precursors not represented here, and the authors encourage the reader to further explore the ALD database for a fuller picture of the scope of organometallic species in ALD.3

534

Comprehensive Organometallic Chemistry IV

https://doi.org/10.1016/B978-0-12-820206-7.00117-7

Common Precursors and Surface Mechanisms for Atomic Layer Deposition

535

Fig. 1 A simple cartoon of a typical ALD process.

Fig. 2 The introduction of common organometallic ligands into atomic layer deposition precursors. Data from Atomic Limits: ALD Database (2021) https://www. atomiclimits.com/alddatabase/. https://doi.org/10.6100/alddatabase.

14.14.2 Alkyl compounds Precursors containing alkyl ligands are the most studied class of precursors. Metal alkyl compounds are straightforward to make and modify, giving a wide control over thermochemical characteristics, and the prototypical ALD reaction is the deposition of alumina from trimethylaluminum(III) (TMA) and water:4 2Me3 Al + 3H2 O ! Al2 O3 + H2 O The model of a hydroxylated surface is commonly considered in ALD to be the starting point for deposition and makes sense in the case of an oxide: precursor nucleation occurs at surface defects in the oxide extended structure. These surface defects would be similar for nitrides (–NH2 and –NH) and other binary extended structures. Additionally, TMA can act as a prototype reactant for many metal alkyls. TMA has a tendency to react in a straightforward manner at the surface due to its high electropositivity and lack of redox activity (it would be unusual for aluminum to adopt a + 1 oxidation state, for instance). There are many metal alkyl precursors that can be found in ALD processes. Table 1 shows a broad representation of these alkyls, and many more can be found in the ALD database in the Atomic Limits website.3 The reaction of TMA and water is an excellent starting point for discussing the chemistry of precursors, because even the commonly used trivial acronym of the aluminum precursor already contains misleading structural information. TMA exists in the gas phase in equilibrium with hexamethyldialuminum(III), the dimer of TMA.21 However, a straightforward model of TMA reacting with a hydroxylated surface shows good reproducibility of the growth measured by ALD.22 Here, TMA was found to “nucleate” at the surface by reacting with one lone hydroxyl, or two proximate hydroxyls to give two different chemisorbed surface moieties (Fig. 3).

536

Common Precursors and Surface Mechanisms for Atomic Layer Deposition

Table 1

A representative selection of metal alkyls used in ALD.

Precursor

Co-precursors

Target film

First reported

GaMe3 AlMe3 GeH2Et2 InMe3 CdMe2, HgMe2 SnMe4 ZnEt2 PbEt4 Pt(CpMe)Me3 ZrCp2Me2 HfCp2Me2 RuCp(CO)2Et BeMe2 BiMe3 Au(PMe3)Me3 Re(O)3Me

AsH3 AsH3 Thermal AsH3 HgMe2, TeMe(allyl) N2O4 AlMe3, H2O H2S O2 H2O H2O O2 H2O Fe(C6H8)(CO)3, H2O O2 plasma, H2O AlMe3

GaAs AlAs Ge InAs HgTe, CdTe, HgCdTe SnO2 Al:ZnO PbS Pt ZrO2 HfO2 Ru BeO BiFexOy Au ReAlxOy

19855 19856 19897 19898 19929 199410 199411 200112 200313 200314 200515 200916 201117 201318 201619 201720

Data from Atomic Limits: ALD Database (2021) https://www.atomiclimits.com/alddatabase/. https://doi.org/ 10.6100/alddatabase.

Fig. 3 Two potential surface nucleation reactions between TMA and a hydroxylated surface. Data from Puurunen, R.L. Growth per Cycle in Atomic Layer Deposition: Real Application Examples of a Theoretical Model. Chem. Vap. Depos. 2003, 9 (6), 327–332. https://doi.org/10.1002/cvde.200306266.

The elimination of methane as a volatile, non-reactive side product is well established,23 and has been demonstrated in several examples of alumina ALD.22 This surface chemistry likely proceeds through the donation of a lone pair from the surface hydroxyl to the acid p orbital of TMA and subsequent transfer of the surface proton to a methyl ligand to produce CH4. It is straightforward to imagine the reaction of the chemisorbed surface with water, where the water protonates the methyl group to produce a surface hydroxyl (Fig. 4A). With ozone and oxygen plasma, a surface hydroxyl is still produced, but with combustion of the alkyl group (Fig. 4B).24 Reactivity with oxygen plasma can be difficult to balance, since the plasma will naturally contain cations, anions, radicals, electrons, and light. In many oxygen plasma deposition systems, the conditions are such mostly radicals contribute to the surface reactions. In this way, the reaction of TMA with ozone can be like the reaction of oxygen plasma in radical enhanced ALD. In the reaction of TMA with ozone, it was found that the “radical” oxygen would insert itself into the existing AldC ligand bonds, as well as the aluminum bond to the surface (below, “╟” represents the Si surface):25

(A)

(B)

Fig. 4 The contrast of the reactivity of water and oxygen plasma with a chemisorbed methylaluminum moiety from Ref. 24. Data from Goldstein, D. N.; McCormick, J. A.; George, S. M. Al2O3 Atomic Layer Deposition with Trimethylaluminum and Ozone Studied by In Situ Transmission FTIR Spectroscopy and Quadrupole Mass Spectrometry. J. Phys. Chem. C 2008, 112(49), 19530–19539. https://doi.org/10.1021/jp804296a.

Common Precursors and Surface Mechanisms for Atomic Layer Deposition

537

╟AlðCH3 Þ2 + 3O3 ! ╟OAlðOCH3 Þ2 + 3O2 This reactivity is inherently difficult to characterize, but a potential path is through coordination of the ozone to the empty p orbital on the aluminum, followed by a (formal) 1,1-migratory insertion of the oxygen into the AldC bond, as well as the Al-surface bond. Further combustion of the surface methoxy group will then produce a bicarbonate group, which can eliminate CO2 to form a surface hydroxyl: ╟OAlðOCH3 Þ2 + 3O2 ! ╟OAlðHCO3 Þ2 + 2H2 O ╟OAlðHCO3 Þ2 ! ╟OAlðOHÞ2 + 2CO2 The production of water in this reaction can lead to the formation of surface hydroxyls and methane formation.26 It is important to note that reaction with ozone and oxygen plasma has been found to also eliminate ethene,27 as well as producing formate groups at the surface.28 These examples further demonstrate the manifold reactions that can contribute to a simple ALD process like the deposition of alumina using ozone. Naturally, reaction at a metal surface does not need to follow this general reactivity of nucleation at a surface hydroxyl, although many metals, even noble metals, can form hydroxyl nucleation defects.29 An example of a different reaction pathway has been reported for the case of trimethyl(trimethylphosphine)gold(III), where the surface reactivity is initiated by loss of the coordinated phosphine and coordination of both moieties to the gold surface.30 Metal surfaces can also act as a site for catalytic decomposition of a precursor’s ligand system. The compound RuCp(CO)2Et is known to produce Ru0 on hydrogen terminated silicon.16 Being a noble metal, ruthenium is known to react with dioxygen to make surface bound oxygen atoms, rather than an oxide; this is used to combust the ligand system, leaving Ru0 behind. This has naturally lead to the speculation that Ru0 is deposited by combustion of the ligand system to CO2 and H2O.31 The surface reactivity has been found to be more complicated, however, with H2 being found as the predominant decomposition product for hydrogen.32 Oxygen forms a monolayer, which reacts with the incoming precursor to release CO2, H2O, and H2, depositing ruthenium and leaving a partial carbon monolayer (Fig. 5A). This monolayer reacts with the subsequent oxygen pulse to release CO and CO2 and re-establish the oxygen monolayer (Fig. 5B). This combustion reactivity is not a straightforward reaction, and so it is difficult to fully characterize what ligand is lost upon chemisorption. However, it seems likely that a ruthenium bound carbonyl reacts with surface oxygen to release CO2 and allows a RudRu interaction with the surface. Another interesting case of reactivity, although less common, is represented by the deposition of elemental germanium from diethyldihydridogermanium(IV).7 This reaction has the hallmarks of a high temperature CVD process, and the ALD occurs by modulating the temperature after the precursor has chemisorbed to the surface. The authors show this proceeding with loss of both hydrides from the precursor, leaving an ethyl-terminated surface (Fig. 6A). A possible fate of the hydride is to migrate to an open surface site on the Ge0 surface as a formal hydride, and then undergo a reductive elimination with a second hydride to form dihydrogen. The second step, a rapid heating, is said to eliminate “Et,” supposedly as a radical. The evidence cited is the increase in a mass 27 amu peak in a quadrupole mass spectrometer (QMS) that was used to monitor the gas phase as temperature was increased. Since an ethyl radical has a mass of 29 amu, this might not be the case. Another plausible scenario is the beta-hydrogen elimination from the ethyl group to form ethene (28 amu), and formation of a Ge-H, which then undergoes reductive elimination to again form dihydrogen (Fig. 6B). The QMS might then be detecting the cation of ethene as a 27 amu species.

(A)

(B)

Fig. 5 Surface reactivity of oxygen at a growing ruthenium metal surface. Data from Aaltonen, T.; Rahtu, A.; Ritala, M.; Leskelä, M. Reaction Mechanism Studies on Atomic Layer Deposition of Ruthenium and Platinum. Electrochem. Solid-State Lett. 2003, 6(9), C130. https://doi.org/10.1149/1.1595312.

538

Common Precursors and Surface Mechanisms for Atomic Layer Deposition

(A)

(B)

Fig. 6 Possible surface reaction for the deposition of elemental germanium. Data from Takahashi, Y.; Ishii, H.; Fujinaga, K. Germanium Atomic Layer Epitaxy Controlled by Surface Chemical Reactions. J. Electrochem. Soc. 1989, 136(6), 1826–1827. https://doi.org/10.1149/1.2097029.

14.14.3 Cyclopentadienyl compounds Cyclopentadienyl (Cp) precursors for ALD were introduced in 1991 and saw steady development through the first decade of the millennium. Much of the development of Cp-based ALD precursors was carried out at the Technical University of Helsinki (now Aalto University) in the research group of Lauri Niinistö (Table 2). The earliest example of a Cp precursor, cyclopentadienyl(triethylphosphine)copper(I), was used to match the process temperature required for the application, and was chosen from “among a very limited number of candidates.”33 The first to receive and detailed discussion of the mechanism of deposition was the deposition of Ru0 from RuCp2, where the ligand system was considered to be combusted completely to CO2 and H2O:31 RuCp2 + 25O ∗ ! Ru0 + 10CO2 + 5 H2 O This was later to be found to also release H2 as a reaction by-product (as discussed in the previous section).16 A later paper on cobalt metal deposition found that metallocenes in general have difficulty nucleating on a hydrogen terminated silicon surface,44 and suggested that a common decomposition pathway for Cp is the formation of the dihydrofulvalene. Decomposition of CoCp2 in a hydrogen atmosphere was found to produce the dihydrofulvalene, cyclopentadiene, cyclopentene, and cyclopentane, suggesting sequential reactions with dihydrogen (Fig. 7).56 Table 2

A representative selection of metal cyclopentadienyls used in ALD.

Precursor

Co-precursors

Target film

First reported

CuCp(PEt3) MgCp2 Ba(CpMe5)2, Sr(CpiPr3)2 ZrCp2Cl2, ZrCp2Me2 ScCp3 NiCp2 Pt(CpMe)Me3 RuCp2 YCp3, Y(CpMe)3 Lu[Cp(SiMe3)]2Cl Er(CpMe)3 Gd(CpMe)3 CoCp2, CoCp(CO)2 MoCp(CO)2(NO) Ca(CpiPr3)2 InCp Mn(CpMe)2 Ir(CpEt)(CHD) W(CpEt)(CO)2(NO) LaCp3 FeCp2 TaCp(NtBu)(NEt2)2 OsCp2 Mg(EtCp)2

GaEt3, H2S H2O Ti(OiPr)4, H2O O3 H2O H2O, NH3 plasma O2 O2 H2O H2O H2O H2O NH3 plasma H2/N2 plasma HfCl4, H2O O3 As(NMe2)3 O2/H2 plasma H2 plasma H2O O2 NH3 O2 HF

CuGaxSy MgO BaTixOy ZrO2 Sc2O3 NiO, Ni Pt Ru Y2O3 Lu2O3 Er2O3 Gd2O3 Co MoNx CaO, CaCO3, CaHfxOy In2O3 MnAs IrO2 WNxCy La2O3 FeOx TaNx Os MgF2

199133 199334 199935 200136 200137 200238 200313 200339 200440 200441 200542 200543 200644 200645 200646 200647 200748 200749 200750 200751 200852 201053 201154 201655

Data from Atomic Limits: ALD Database (2021) https://www.atomiclimits.com/alddatabase/. https://doi.org/10. 6100/alddatabase.

Common Precursors and Surface Mechanisms for Atomic Layer Deposition

539

Fig. 7 By-product of Cp reactions with dihydrogen or hydrogen plasma. Data from Lee, H.-B.-R.; Kim, H. High-Quality Cobalt Thin Films by Plasma-Enhanced Atomic Layer Deposition. Electrochem. Solid-State Lett. 2006, 9(11), G323. https://doi.org/10.1149/1.2338777.

This reactivity with H2 was explored for Co0 metallic films, and it was speculated that this reactivity will minimize carbon impurities in metallocene deposited films. In the case of using an ammonia plasma to reduce the cobaltocene to Co0, carbon impurities could not be detected by X-ray photoelectron spectroscopy. An interesting ALD process using nickelocene deposited carbon rich NiO from NiCp2 and water, and the oxide film was subsequently reduced with hydrogen radicals produced from an ammonia plasma.38 Presumably, the NiCp2 chemisorbed at the growing nickel surface through loss of a Cp ligand. The high carbon content of the oxide film demonstrates that dihydrofulvalene formation does not occur at all surfaces, and that some Cp ligand can remain following chemisorption. The ammonia plasma in this process was then used to clean the surface, removing the Cp ligands entirely and reducing the nickel center to Ni0. Many of the Cp precursors are heteroleptic, and this begs the question of which ligand will be lost first during chemisorption. From a first-principles standpoint, one might expect the ligand whose conjugate acid has the highest pKa to be lost first. Cyclopentadiene has a pKa of 16, and so it might be expected to be retained compared to a methyl group (pKa  50 for methane). However, surface reactivity can be less straightforward than solution chemistry: in a solution, the reactants and products are commonly in the same thermodynamic system for a timeframe that would allow equilibrium to be established, and so specific emergent reactions can become dominant. However, chemisorption at a surface, particularly under ALD conditions, can mean loss of the ligand to a gas flow and subsequent removal from the system, with no opportunity for equilibrium to be established. This can allow less-stable reaction pathways to compete more readily with more thermodynamically spontaneous pathways. For example, ZrCp2Me2 would be expected to react through loss of methyl when chemisorbing at a surface hydroxyl. However, both Cp and methyl are lost during this chemisorption (Fig. 8).57 By careful QMS analysis, it was found that the balanced chemisorption reaction showed two contributions from the reaction in Fig. 8A, and one each from the reactions in Fig. 8B and C. Although methyl groups are majoritively lost, CpH is also a by-product of the nucleation reaction. This contrasts with when an ansa-zirconocene is used: chemisorption occurs through the loss of two methyl moieties only. This is due to the chelate effect stabilizing the ansa-bridged Cp ligands.58 The deposition mechanism of manganese fluoride by ALD contributes another contrasting chemisorption mechanism with respect to Cp chemistry.55 The fluoride-containing precursor for this process is the pyridine adduct of hydrofluoric acid (pyridine
HF), and it is used with Mn(CpEt)2 to deposit MgF2 at 150  C. Interestingly, quartz crystal microbalance analysis of the process shows that the manganocene does not lose a ligand upon coordinating to the surface. It appears to coordinate intact, and release two equivalents of cyclopentadiene during the pulse of pyridine
HF. The surface interactions are likely due to fluoride in the substrate bridging to the manganese center (Fig. 9). Manganocene is a 17-electron species and accommodating further electron density might require a Cp to ring-slip into an Z3 coordination geometry. Further modeling of this interaction would clarify metallocene-surface interactions.

(A)

(C)

(B)

Fig. 8 Chemisorption reactions of ZrCp2Me2 at a hydroxylated surface. Data from Niinistö, J.; Rahtu, A.; Putkonen, M.; Ritala, M.; Leskelä, M.; Niinistö, L. In Situ Quadrupole Mass Spectrometry Study of Atomic-Layer Deposition of ZrO2 Using Cp2 Zr(CH3)2 and Water. Langmuir 2005, 21(16), 7321–7325. https://doi.org/10. 1021/la0500732.

540

Common Precursors and Surface Mechanisms for Atomic Layer Deposition

Fig. 9 Potential chemisorption geometry of manganocene at a MnF2 surface. Data from Lee, Y.; Sun, H.; Young, M. J.; George, S. M. Atomic Layer Deposition of Metal Fluorides Using HF–Pyridine as the Fluorine Precursor. Chem. Mater. 2016, 28(7), 2022–2032. https://doi.org/10.1021/acs.chemmater.5b04360.

(A)

(B)

Fig. 10 Cyclopentadienylindium(I) undergoing (A) volatilization, and (B) reaction at an oxide surface. Data from Mizutani, F.; Higashi, S.; Inoue, M.; Nabatame, T. Atomic Layer Deposition of Stoichiometric In 2O3 Films Using Liquid Ethylcyclopentadienyl Indium and Combinations of H2O and O2 Plasma. AIP Adv. 2019, 9(4), 045019. https://doi.org/10.1063/1.5081727.

InCp47 and its more volatile cousin In(CpEt)59 are noteworthy Cp-bearing precursors since they are metallocene “halfsandwiches” that exist as coordination polymers in the solid state,60 but volatilize to be monomeric (Fig. 10A).21 These compounds volatilize with pressures of 0.2 Torr (InCp) and 0.3 Torr (In(CpEt)) at room temperature and produce independent InCp molecules.59 The surface chemistry suggests that the indium reacts with a hydroxyl defect eliminating cyclopentadiene (i.e., the typical Bronsted acid-base surface reaction), or alternately coordinates to an open oxygen valence on the growing In2O3 film while retaining the Cp ligand (Fig. 10B).47,59 Although subsequent reaction with O2 plasma produces both terminal oxo and hydroxyl groups (presumably through reaction with water formed in the combustion) to allow regeneration of the surface in Fig. 9B,47 it was found that more efficient growth occurred when water was pulsed prior to the oxygen plasma to hydrolyse off any retained ligand; the oxygen plasma then oxidizes In+ to In3+. Interestingly, the oxygen plasma is not needed for oxidation. Using water and oxygen either in separate pulses or mixed together produces uniform In2O3 with InCp.61 This reactivity demonstrates the susceptibility of the In+ surface to oxidation, and was proved to be a welcome innovation over using ozone.

14.14.4 Amide compounds The amide precursors got an early start in atomic layer deposition, with bis(hexamethylsilazido)zinc(II) being used in 1992 with H2Se to make ZnSe.62 The reach of amides is now across the periodic table, but their uptake on many metal centers is sparse, and later than expected (Table 3). Although the first amide paper discusses the effect of nitrogen incorporation in the resulting film, it does not discuss Si incorporation. However, if an oxygen source is used, the oxygen will scavenge silicon from the precursor to form a silicate.74,78 Presumably, this is the oxophilic nature of silicon and the combustion of the ligand system by the oxygen source acting together to ensure Si incorporation. In general, amide precursors follow the same Bronsted acid-base surface reaction chemistry seen for alkyls: the amine (pKa  11) is released upon reaction with a surface hydroxyl or amido group, chemisorbing the remaining precursor moiety. An interesting case arose early in amide precursor use when Ti(NR2)4 compounds were found to react with ammonia to produce TiN, with a reduction of the Ti4+ to a Ti3+ during the ALD process.86 The initial chemisorption (at a amine-terminated surface) is considered to undergo typical transamination at Ti4+ (Fig. 11A). However, the redox step is not fully understood. It is commonly quoted in the CVD and ALD literature that dinitrogen is evolved during reaction with ammonia:

Common Precursors and Surface Mechanisms for Atomic Layer Deposition

Table 3

541

A representative selection of metal amides used in ALD.

Precursor

Co-precursors

Target film

First reported

Zn[N(SiMe3)2]2 Ni(dimethylglyoximate)2 Ti(NEt2)4 La[N(SiMe3)2]3 Ta(NtBu)(NEt2)3 W(NtBu)2(NMe2)2 Bi[N(SiMe3)2]3 Hf[N(SiMe3)2]2Cl2 Pr[N(SiMe3)2]3 Zr[N(SiMe3)2]2Cl2 GaMe2NH2 Al(NEt2)3 Lu[N(SiMe3)2]3 Mo(NtBu)2(NMe2)2 Sn(NMe2)4 Nb(NtBu)(NEtMe)3 Li(N(SiMe3)2) Cu(iPr2-NHC)(N(SiMe3)2) In(N(SiMe3)2)Et2 Ge(Z2-((NtBu)CHMeCHMe(NtBu))) Fe(N(SiMe3)2)2 V(NEtMe)4 Co[N(SiMe3)2]2(THF) Pb[N(SiMe3)2])2

H2Se O2 NH3 H2O, (tBuO)3SiOH H2 plasma H2/N2 plasma Ta(OEt)5, H2O H2O H2O H2O O2 plasma H2O H2O, O3 NH3 H2O H2 plasma H2O, O3 H2 plasma H2O Te(SiMe3)2 H2O2 In(iPrAMD)3, H2S H2O SnI4

ZnSe NiO TiN La2O3, LaSixOy TaNx WCx BiOx, BiTaxOy HfO2 PrOx ZrO2 Ga2O3 Al2O3 LuSixOy MoNx SnO2 NbN LiSixOy Cu In2O3 GeTe FeOx VInxS3y CoO/Co(OH)2 PbI2

199262 199863 199864 200165 200166 200367 200468 200469 200470 200471 200572 200673 200674 200775 200876 201077 201278 201379 201480 201681 201682 201683 201984 201985

Data from Atomic Limits: ALD Database (2021) https://www.atomiclimits.com/alddatabase/. https://doi.org/10. 6100/alddatabase.

Fig. 11 Surface reaction and redox chemistry of Ti(NR2)4. Data from Kan, B.-C.; Boo, J.-H.; Lee, I.; Zaera, F. Thermal Chemistry of Tetrakis(Ethylmethylamido) Titanium on Si(100) Surfaces. J. Phys. Chem. A 2009, 113(16), 3946–3954. https://doi.org/10.1021/jp8102172.


2╟NH + TiðNMe2 Þ4 ! ╟N − 2 TiðNMe2 Þ2 + 2HNMe2

╟N − 2 TiðNMe2 Þ2 + 4=3NH3 ! ╟N − 2 TiNH2 + 2HNMe2 + 1=6N2 Although there is support for this mechanism in Ref. 86, it has two potential flaws. The first is that the reaction starts with two nucleation sites and ends with one nucleation site, which is not sustainable during multiple film growth cycles. The second is the formation of N2, which has no proposed mechanism and is difficult to assess. A hint at the redox chemistry was subsequently found for the similar compound Ti(NEtMe)487, where the nucleation step was found to be the same, but was followed by a beta-hydrogen elimination and formation of a chemisorbed imine (Fig. 11B). This leads to a formal Ti2+ at the surface, which might then undergo comproportionation with a neighboring Ti4+ site to produce the appropriate +3 oxidation state. Notably, reaction of Ti(NR2)4 with water gives TiO2, with titanium retained in the +4 oxidation state.

542

Common Precursors and Surface Mechanisms for Atomic Layer Deposition

An interesting class of amides are found with the mid-transition refractory metals (Nb5+, Mo6+, Ta5+, W6+) where the large metal size and high valency can support imides as well as amides. These compounds have the general structure M(NR)x(NR0 2)4− x, where x is determined by the valency of the metal. Due to the strong imide bond, these compounds generally react to deposit the nitride, except in the case of W(NtBu)2(NMe2)2, which deposits a carbide with N2/H2 plasma, and a nitride when reacted thermally with ammonia.88

14.14.5 Chelate nitrogen compounds There are several nitrogen-containing chelate ligands in atomic layer deposition, and here we will focus on the N,N-chelates (Fig. 12). The most important classes of N,N-chelates, and those whose chemistry will be discussed in detail are the amidinates (including guanidinates and iminopyrrolidinates). It should be noted that several ligand frameworks have also started to be used, including the diazabutadienes,89 the betadiketiminates,90 and the triazenides.91 Unlike many examples of ligands in ALD, the introduction of the amidinates can be traced back to a key paper from 2003, when Roy Gordon (from Harvard University) published the ALD processes of seven different metals and oxides using a variety of amidinates (Table 4).92 This foundational paper triggered research into amidinates and their surface chemistry that has lasted almost 20 years.

Fig. 12 N,N-chelate ligands used in ALD. Based on years of research.

Table 4

A representative selection of metal amidinates used in ALD.

Precursor

Co-precursors

Target film

First reported

H2, H2O H2 H2, H2O H2O H2 O3 H2O H2O H2O NH3 H2O H2O H2O H2S H2O H2O H2S O3 H2O In(iPrAMD)3, H2S H2O Sn(iPrAMD)2, H2S

Co, CoOx Cu Fe, FeOx La2O3 Ni Er2O3 Y2O3 Sc2O3 PrOx Ru Al2O3 ZrO2 Ta2O5 SnS Dy2O3, Gd2O3 TiO2 In2S3 MoON WO3 VInxS3y CeO2 CaSnS

200392 200392 200392 200392 200392 200593 200594 200695 200696 200797 200898 200999 2010100 2011101 2012102 2012103 2014104 2016105 2016105 201683 2018106 2019107

i

Co( PrAMD)2 [Cu(iPrAMD)]2 Fe(tBuAMD)2 La(iPrAMD)3 Ni(iPrAMD)2 Er(tBuAMD)3 Y(iPrAMD)3 Sc(iPrAMD)3 Pr(iPrAMD)3 Ru(tBuAMD)2(CO)2 Al(iPrAMD)Et2 Zr(MeAMD)4 Ta(NtBu)(iPrAMD)2(NMe2) Sn(iPrAMD)2 Dy(DPDMG)3, Gd(DPDMG)3 Ti(OiPr)3(iPrAMD) In(iPrAMD)3 MoO2(tBuAMD)2 WO2(tBuAMD)2 V(iPrAMD)3 Ce(iPrCp)2(iPrAMD) Ca(iPrFMD)2

Data from Atomic Limits: ALD Database (2021) https://www.atomiclimits.com/alddatabase/. https://doi.org/10. 6100/alddatabase.

Common Precursors and Surface Mechanisms for Atomic Layer Deposition

543

The nucleation of the homolytic amidinates occurs through formation of the conjugate acid (pKa  12) at a surface hydroxyl, as is typical of ALD. In the case of MO2(tBuAMD)2 (M ¼ Mo, W) that reaction at the surface hydroxyl does protonate the amidinate ligand, but it remained hydrogen-bonded to the surface.105 In the case of molybdenum, this led to a controllable amount of nitrogen in the film, forming MoON. The heteroleptic Al(iPrAMD)Et2 was used with water to deposit Al2O3 as low as 125  C. Here, the nucleation was expected to occur through loss of the ethyl groups with the amidinate remaining at the surface. This reaction mechanism was demonstrated through solid-state NMR characterization using the analogous gallium precursor Ga(iPrAMD)Et2, with the precursor losing both ethyl ligands (Fig. 13).108 Deposition of Ru0 from Ru(tBuAMD)2(CO)2 deviates from the normal thermal behavior of amidinate compounds.97 This precursor was used with ammonia to deposit Ru0, and addition of carbon monoxide as the carrier gas suppressed deposition, suggesting that loss of carbonyl was the chemisorption step for this process (Fig. 14A). It is unclear if the CO leaves immediately or physisorbs to the surface, but introduction of ammonia promotes the reductive elimination of a coupled amidinate species (Fig. 14B). The mechanistic details of this process are still unexplored. Amidinates are known to decompose readily at a growing surface at elevated temperatures, limiting their use (in most cases) to low-temperature ALD.109 The copper amidinate dimer undergoes dissociation to surface-adsorbed monomers on Ni0 at temperatures below about −125  C (150 K) (Fig. 15).

Fig. 13 Nucleation of the heteroleptic species Ga(iPrAMD)Et2. Data from Pallister, P. J.; Buttera, S. C.; Barry, S. T. Quantitative Surface Coverage Calculations via SolidState NMR for Thin Film Depositions: A Case Study for Silica and a Gallium Amidinate. J. Phys. Chem. C 2014, 118(3), 1618–1627. https://doi.org/10.1021/jp4102674. (A)

(B)

Fig. 14 The nucleation and unusual reductive elimination for deposition of ruthenium metal from Ru(tBuAMD)2(CO)2. Data from Li, H.; Farmer, D. B.; Gordon, R. G.; Lin, Y.; Vlassak, J. Vapor Deposition of Ruthenium from an Amidinate Precursor. J. Electrochem. Soc. 2007, 154(12), D642. https://doi.org/10.1149/1.2789294.

Fig. 15 Thermal surface decomposition of [Cu(sBuAMD)]2. Data from Ma, Q.; Guo, H.; Gordon, R. G.; Zaera, F. Surface Chemistry of Copper(I) Acetamidinates in Connection With Atomic Layer Deposition (ALD) Processes. Chem. Mater. 2011, 23 (14), 3325–3334. https://doi.org/10.1021/cm200432t.

544

Common Precursors and Surface Mechanisms for Atomic Layer Deposition

This adsorption is, of course, beneficial to ALD. However, the onset of surface, thermal decomposition is seen as low as −73  C. It is worth noting that the onset of such a low onset of thermal decomposition does not limit the use of this precursor, which shows self-limiting ALD behavior at temperatures as high as 280  C.92 Noticeably, these two experiments seems to contradict each other. However, the thermal decomposition at −73  C was established under high vacuum on an ultra-clean Ni0 surface, while the deposition study was carried out under roughing-pump vacuum on a growing Cu0 surface; these are quite different thermodynamic systems. However, the main difference is the kinetics: decomposition on the Ni0 surface was observed over more than an hour, while the ALD of copper happened with cycle times measured in seconds. The copper amidinate surface species is decomposing slowly enough during ALD that it does not noticeably affect the self-limiting behavior of the process. This highlights the how important the kinetics of surface decomposition are to establishing an effective ALD process. The surface decomposition proceeds by betahydrogen abstraction from the alkyl group by the ligand’s nitrogen center. This truncated ligand fragment can be released by the copper adatom at temperatures as low as room temperature. This can be remedied through ligand design, specifically by replacing thermally susceptible hydrogens with methyls. In the case of an isopropyl side-chain, this has been shown to improve thermal stability by greater than 200  C.110

14.14.6 Alkoxide compounds Alkoxides have been used for 30 years in ALD but fill the specific niche of forming oxide films (Table 5). In general, metal-oxygen bonds are stable enough to cause significant oxygen inclusion in films and are generally detrimental. Alkoxides are extremely good Bronsted bases (pKa  17) and are typically believed to simply undergo formation of the conjugate acid alcohol in reaction with a surface nucleation defect. Most of the early development of alkoxide precursors were done by Markku Leskelä and Mikko Ritala at the University of Helsinki. Alkoxide precursors are particularly useful for binary or higher order oxides. The earliest example comes from the Ukraine,112 and these precursors have found to be excellent for the deposition of perovskites.124,127 The case of deposition of lithium tantalum oxide is noteworthy:121 The lithium tert-butoxide compound is considered to react twice at the same nucleation site (Fig. 16). This first equivalent of the precursor exchanges a lithium cation for a proton at the surface, while the second equivalent of lithium tert-butoxide chemisorbs without ligand loss at the same nucleation site. Although the paper does not have direct mechanistic proof of this mechanism, the stoichiometry of lithium centers in the resulting film requires a mechanism where it is deposited at a greater than one equivalent in the film. There is an interesting surface mechanism in the deposition of elemental tellurium films, which exploits the high oxophilicity of silicon.125 The authors show that the by-product of reaction between Te(SiMe3)2 and TeOEt4 is Me3SiOEt, and that these two precursors have been designed to leave Te0 in both half-cycles of the reaction: 2TeðSiMe3 Þ2 + TeOEt4 ! 3Te0 + 4Me3 SiOEt In this deposition, the Te(SiMe3)2 must nucleate with a direct Te–Te bond to the growing surface (Fig. 17B). Table 5

A representative selection of metal alkoxides used in ALD.

Precursor

Co-precursor

Target film

First reported

Al(OEt)3 B(OMe)3 Si(OEt)4 Ti(OEt)4 Ta(OEt)5 Nb(OEt)5 Hf(OtBu)4 Zr(OtBu)4 [LaAl(OiPr)6(iPrOH)]2 Fe2(tBuO)6 Sn(OtBu)4 Ge(OMe)4 Sb(OEt)3 LiOtBu NaOtBu, KOtBu [NdAl(OiPr)6(PrOH)]2 RbOtBu Te(OEt)4 CsOtBu

H2O, O2 POCl3 TiCl4 H2O H2O H2O O2 plasma O2 plasma H2O H2O CH3COOH Sb(OEt)3, Te(SiMe3)2 Te(SiMe3)2 Ta(OEt)5, H2O H2O H2, H2O, O2 Nb(OEt)5, H2O Te(SiMe3)2 Nb(OEt)5, H2O

Al2O3 BPxOy SiTixOy TiO2 Ta2O5 Nb2O5 HfO2 ZrO2 LaAlxOy FeOx SnO2 GeSbxTey Sb2Te3 LiTaxOy Na2O, K2O NdAlxOy RbNbxOy Te CsNbxOy

1991111 1993112 1993112 1994113 1995114 1997115 2003116 2003116 2006117 2007118 2011119 2012120 2012120 2013121 2014122 2015123 2017124 2019125 2020126

Data from Atomic Limits: ALD Database (2021) https://www.atomiclimits.com/alddatabase/. https://doi.org/10.6100/ alddatabase.

Common Precursors and Surface Mechanisms for Atomic Layer Deposition

545

Fig. 16 Chemisorption of more than one equivalent of lithium species (either a cation or tert-butoxolihium(I)) at a nucleation site. Data from Liu, J.; Banis, M. N.; Li, X.; Lushington, A.; Cai, M.; Li, R.; Sham, T.-K.; Sun, X. Atomic Layer Deposition of Lithium Tantalate Solid-State Electrolytes. J. Phys. Chem. C 2013, 117(39), 20260–20267. https://doi.org/10.1021/jp4063302.

(A)

(B)

(C)

Fig. 17 Surface mechanism of Te0 deposition. Data from Cheng, L.; Adinolfi, V.; Weeks, S. L.; Barabash, S. V.; Littau, K. A. Conformal Deposition of GeTe Films With Tunable Te Composition by Atomic Layer Deposition. J. Vac. Sci. Technol. A 2019, 37(2), 020907. https://doi.org/10.1116/1.5079661.

A pulse of TeOEt4 then reacts to form two equivalents of Me3SiOEt and leaving a -OEt terminated surface (Fig. 17C). This surface then reacts with one equivalent of Te(SiMe3)2 to leave a Te-terminated surface (Fig. 17A). It is notable that two equivalents of Te(SiMe3)2 has to react in the Te(SiMe3)2 pulse, and that the whole cycle formally deposits three equivalents of Te.

14.14.7 Beta-diketonate compounds Beta-diketonates are a special class of alkoxide ligand that have seen wide and varied use in ALD (Table 6). These have a framework with adjacent ketones, and both historical and “systematic” (non-IUPAC) naming conventions have given rise to a variety of acronyms (Fig. 18). Like many ALD precursors, metal beta-diketonates (pKa  9) generally undergo an acid-base reaction with protonated surface nucleation defects to lose a protonated ligand. Interestingly, the beta-diketone “hfac” is one of the few fluorine-bearing ligands that is tolerated in ALD; generally hfac does not leave appreciable residual fluorine in growing films. It is possible to grow metallic films from beta-diketones without significant oxygen impurities in the metal films. In the case of both Pt(acac)2 and Cu(acac)2, films could be deposited using H2 to provide a reducing environment for the metal film.134 The acac ligand was shown to chemisorb to the surface rather than eliminate in the gas phase, and undergo partial decomposition on the surface to produce CO and CO2.147 This surface-mediated combustion is further highlighted in the deposition of Ir0,138 and Rh0.139 This naturally will not work with electropositive metals, and oxide films will preferentially form. Iron oxide has been deposited on a dehydroxylated yttrium/zirconium surface, showing interesting nucleation.136 In this case the Fe(acac)3 sheds its ligands to the surface (as happens in Ref. 147 as well) prior to undergoing combustion (Fig. 19). Notably, beta-diketonates tend to chemisorb to the surface in some key mechanisms, rather than to eliminate as their conjugate acid. Although this needs further research, it is notable that the pKa of beta-diketonates are much lower than other ligands explored in this chapter. Because hfac is an easily available fluorine-containing ligand, it has been used as a fluorine source for ALD.143 The hfac ligand had previously been shown to leave surface-bonded fluorine during thermal ALD.148 Exploiting hfac as a precursor for fluorine-containing films in elegant idea, since the fundamental fluorinating agent (HF) is toxic and difficult to handle safely. It should be noted that other fluorinating schemes have been employed in ALD, including using a pyridine
HF adduct,55 as well as using TiF4.149

546

Common Precursors and Surface Mechanisms for Atomic Layer Deposition

Table 6

A representative selection of metal beta-diketonates used in ALD.

Precursor

Co-precursor

Target film

First reported

Cr(acac)3 Ni(acac)2 Ga(acac)3 Cu(acac)2 Co(acac)3 In(acac)3 Pt(acac)2 La(thd)3 VO(acac)2 Fe(acac)3 Pd(hfac)2 Ir(acac)3 Rh(acac)3 Er(thd)3 Ag(hfac)(cod) Sn(acac)2 Ca(hfac)2 Mg(thd)2 Sr(thd)2 Ru(CO)2(mhdk)2 Na(thd)

O2 O2 O2, H2O O2 O2 H2S H2 Al(acac)3, O3 O2 O2 H2, glyoxylic acid O2 O2 Ga(acac)3, O3 propanol H2S Hhfac, O3 Hhfac, O3 Co(acac)3, O3 NH3 plasma Co(acac)3, O3

CrOx NiO Ga2O3 CuOx CoOx In2S3 Pt LaAlxOy VOx FeOx Pd Ir Rh ErGaxOy Ag SnS CaF2 MgF2 SrCoxOy Ru NaxCoO2

1994128 1994129 1996130 1996131 1998132 2000133 2000134 200137 2002135 2002136 2003137 2004138 2005139 2007140 2010141 2010142 2011143 2011143 2014144 2015145 2020146

Data from Atomic Limits: ALD Database (2021) https://www.atomiclimits.com/alddatabase/. https://doi.org/10.6100/ alddatabase.

Fig. 18 Key beta-diketonates used in ALD. Based on years of research.

Fig. 19 Surface adsorption of acetylacetonate ligands. Data from de Ridder, M.; van de Ven, P. C.; van Welzenis, R. G.; Brongersma, H. H.; Helfensteyn, S.; Creemers, C.; Van Der Voort, P.; Baltes, M.; Mathieu, M.; Vansant, E. F. Growth of Iron Oxide on Yttria-Stabilized Zirconia by Atomic Layer Deposition. J. Phys. Chem. B 2002, 106 (51), 13146–13153. https://doi.org/10.1021/jp0211640.

Reference 143 has a mechanism included, but it suffers from some flaws in stoichiometry and charge balance. However, the authors have discovered that CaF2 is formed by combustion of chemisorbed Ca(thd)2 with ozone. The combustion of Ca(thd)2 with ozone at the surface (shown to be calcium carbonate in Ref. 143) is then followed by reaction with H-hfac, and again combusted with ozone. This leaves a film of microcrystalline CaF2 with undetectable oxygen and carbon impurities.

Common Precursors and Surface Mechanisms for Atomic Layer Deposition

547

14.14.8 Carbonyl compounds Carbonyl ALD precursors are quite rare and are mainly represented by heteroleptic systems (Table 7). A trait that limits carbonyl-containing precursors is their low thermal stability. For example, although the vapor pressure of CoCp(CO)2 is almost two orders of magnitude higher than CoCp2, it was shown to undergo gas phase nucleation through loss of CO.152 Although these gas phase reactions can be beneficial for CVD, the are deleterious for ALD. Carbonyl compounds can be expected to nucleate through loss of carbonyl. In the deposition of WNx from W(CO)6 and ammonia, loss of carbonyl was found to occur when nucleating on the growing film (Fig. 20).153 The resulting films had