Nanofibers: Fabrication, Performance, and Applications 1607419475, 9781607419471

Citation preview

NANOTECHNOLOGY SCIENCE AND TECHNOLOGY SERIES Safe Nanotechnology Arthur J. Cornwelle 2009. ISBN: 978-1-60692-662-8 National Nanotechnology Initiative: Assessment and Recommendations Jerrod W. Kleike (Editor) 2009. ISBN 978-1-60692-727-4 Nanotechnology Research Collection 2009/2010. DVD edition James N. Ling (Editor) 2009. ISBN 978-1-60741-293-9 Nanotechnology Research Collection 2009/2010. PDF edition James N. Ling (Editor) 2009. ISBN 978-1-60741-292-2 Safe Nanotechnology in the Workplace Nathan I. Bialor (Editor) 2009. ISBN 978-1-60692-679-6 Strategic Plan for NIOSH Nanotechnology Research and Guidance Martin W. Lang (Author) 2009. ISBN: 978-1-60692-678-9 Nanotechnology in the USA: Developments, Policies and Issues Carl H. Jennings (Editor) 2009. ISBN: 978-1-60692-800-4 New Nanotechnology Developments Armando Barrañón (Editor) 2009. ISBN: 978-1-60741-028-7

Electrospun Nanofibers and Nanotubes Research Advances A. K. Haghi (Editor) 2009. ISBN: 978-1-60741-220-5

Nanostructured Materials for Electrochemical Biosensors Umasankar Yogeswaran, S. Ashok Kuma and Shen-Ming Chen 2009. ISBN: 978-1-60741-706-4 Magnetic Properties and Applications of Ferromagnetic Microwires with Amorpheous and Nanocrystalline Structure Arcady Zhukov and Valentina Zhukova 2009. ISBN 978-1-60741-770-5 Electrospun Nanofibers Research: Recent Developments A.K. Haghi (Editor) 2009. ISBN 978-1-60741-834-4 Nanotechnology: Environmental Health and Safety Aspects Phillip S. Terrazas (Editor) 2009. ISBN: 978-1-60692-808-0 Nanofibers: Fabrication, Performance, and Applications W. N. Chang (Editor) 2009. ISBN: 978-1-60741-947-1

Nanotechnology Science and Technology Series

NANOFIBERS: FABRICATION, PERFORMANCE, AND APPLICATIONS

No part of this digital document may be reproduced, stored in a retrieval system or transmitted in any form or by any means. The publisher has taken reasonable care in the preparation of this digital document, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained herein. This digital document is sold with the clear understanding that the publisher is not engaged in rendering legal, medical or any other professional services.

Nanotechnology Science and Technology Series

NANOFIBERS: FABRICATION, PERFORMANCE, AND APPLICATIONS

W. N. CHANG EDITOR

Nova Science Publishers, Inc. New York

Copyright © 2009 by Nova Science Publishers, Inc. All rights reserved. No part of this book may be reproduced, stored in a retrieval system or transmitted in any form or by any means: electronic, electrostatic, magnetic, tape, mechanical photocopying, recording or otherwise without the written permission of the Publisher. For permission to use material from this book please contact us: Telephone 631-231-7269; Fax 631-231-8175 Web Site: http://www.novapublishers.com NOTICE TO THE READER The Publisher has taken reasonable care in the preparation of this book, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained in this book. The Publisher shall not be liable for any special, consequential, or exemplary damages resulting, in whole or in part, from the readers’ use of, or reliance upon, this material. Any parts of this book based on government reports are so indicated and copyright is claimed for those parts to the extent applicable to compilations of such works. Independent verification should be sought for any data, advice or recommendations contained in this book. In addition, no responsibility is assumed by the publisher for any injury and/or damage to persons or property arising from any methods, products, instructions, ideas or otherwise contained in this publication. This publication is designed to provide accurate and authoritative information with regard to the subject matter covered herein. It is sold with the clear understanding that the Publisher is not engaged in rendering legal or any other professional services. If legal or any other expert assistance is required, the services of a competent person should be sought. FROM A DECLARATION OF PARTICIPANTS JOINTLY ADOPTED BY A COMMITTEE OF THE AMERICAN BAR ASSOCIATION AND A COMMITTEE OF PUBLISHERS. LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA Nanofibers : fabrication, performance, and applications / W.N. Chang, editor. p. cm. Includes index. ISBN 978-1-61668-288-0 (E-Book) 1. Nanofibers. I. Chang, W. N. TA418.9.F5N36 2009 620'.5--dc22 2009021232

Published by Nova Science Publishers, Inc. Ô New York

CONTENTS Preface Chapter 1

Chapter 2

Chapter 3

Chapter 4

Chapter 5

vii Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers Sk. F. Ahmed and K. K. Chattopadhyay

1

Permeability Studies of Electrospun Chitin and Chitosan Nanofibrous Membranes Jessica D. Schiffman and Caroline L. Schauer

61

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning: Preparation and Biomedical Application D. Paneva, М. Ignatova, N. Manolova and I. Rashkov

73

A Novel Approach for Analysis of Processing Parameters in Electrospinning of Nanofibers M. Ziabari, V. Mottaghitalab and A. K. Haghi

153

Carbon Nano-Fibers and their Applications: Derived from Electrospinning and Vapor Grown Processes S. K. Nataraj, B. H. Kim and K. S. Yang

183

Chapter 6

Carbon Nanofibers as Sensors Sharlene A. Lewis and Charles M. Lukehart

225

Chapter 7

Processing-Structure Relationships of Electrospun Nanofibers Xiangwu Zhang

239

Chapter 8

Glycosylated Nanofibers for Protein Adsorption and Recognition Ai-Fu Che, Ling-Shu Wan and Zhi-Kang Xu

271

Chapter 9

Porphyrinated Polymer Nanofibers by Electrospinning Yuan-Yuan Lv, Jian Wu, Zhen-Mei Liu and Zhi-Kang Xu

303

Chapter 10

A Nanofibrillar Prosthetic Modified with Fibroblast Growth Factor2 for Spinal Cord Repair Sally Meiners, Suzan L. Harris, Roberto Delgado-Rivera, Ijaz Ahmed, Ashwin N. Babu, Ripal P. Patel and David P. Crockett

327

vi Chapter 11

Chapter 12

Chapter 13

Chapter 14

Index

Contents Fabrication, Performance, and Biomedical Application of Collagen-, Gelatin- or Keratin-Containing PHBV Nanofibers Inn-Kyu Kang, Zhi-Cai Xing, Jiang Yuan, Oh Hyeong Kwon, Jung Chul Kim and Yoshihiro Ito

345

Fabrication and Characterization of Polypropylene Fiber Reinforced by Carbon Nanofiber Yuanxin Zhou and Shaik Jeelani

375

Thermal Analysis of Carbon Nanotubes Incorporated Polyurethanes Nanocomposites Shahrul Azam Abdullah and Lars Frormann

395

Carbon Nanotubes Incorporated Polyurethanes Nanocomposites for Thermal and Electrical Conductive Applications Shahrul Azam Abdullah and Lars Frormann

411 425

PREFACE Nanofibers are defined as fibers with diameters on the order of 100 nanometers. They can be produced by interfacial polymerization and electrospinning. Contrastly, carbon nanofibers are graphitized fibers produced by catalystic synthesis. Nanofibers are included in garments, insulation and in energy storage. They are also used in medical applications, which include drug and gene delivery, artificial blood vessels, artificial organs and medical facemasks. This book presents new research in this dynamic field. Chapter 1 - From the authors experimental observation, it is proposed that the nanothermometer can be constructed using MWCNT more easily. As the emission current vary linearly with temperature for a particular applied electric field, so temperature can be directly measured. The sensitivity of the nanothermometer can be adjusted by choosing the area of the MWCNT film or appropriate applied electric field. The study cited shows that the temperature dependent field emission property of CNFs and MWCNTs has potential for development of direct thermal-to-electrical power conversion applications. Continued improvements in the PECVD of CNFs/CNTs and related nanostructures are indeed required to explore the potential utility of these structures in advanced applications and future largescale integration. Chapter 2 - Electrospinning has been utilized to fabricate fibrous membranes composed of polymer nanofibers, which have large surface area-to-volume ratios and small pores. Electrospun nanofibrous membranes have potential uses in a variety of industries such as energy, environment, medicine, packaging, and automotive, with specific applications including air filtration, protective clothing, fuel cells, and nanocomposites. Nanofibrous membranes composed of biopolymers have potential uses that harness their inherent biocompatibility. Chitin, the second most abundant, naturally occurring polysaccharide after cellulose, is found in shells of crabs and shrimp. Chitosan, the acid soluble form of chitin, is a non-toxic, biodegradable, biopolymer consisting primarily of ί(1 4) linked 2-amino-2deoxy-ί-D-glucopyranose units, and is currently used in tissue engineering, antifouling coatings, separation membranes, stent coatings, enzyme immobilization matrices, and the removal of heavy metals from ground and wastewater. Chitosan is a commercially interesting compound because of its high nitrogen content (6.89%), making it a useful chelating agent for metal ions. Before these chitin or chitosan nanofibrous membranes can be used in the myriad of industries their physical properties, such as permeability, must be known. This chapter focuses on the fabrication and flow cell testing of chitin and chitosan nanofibrous membranes.

viii

W. N. Chang

Additionally, it explores the potential applications of biopolymer and synthetic polymer electrospun membranes. Chapter 3 - At present increased attention is paid on fibrous materials from the natural polymer chitosan because of its numerous beneficial properties (biocompatibility, biodegradability, inherent antibacterial and haemostatic activity). The presence of both hydroxyl and amino groups enables the tailored modification of chitosan into derivatives having targeted properties. The materials containing chitosan or its derivatives are considered as very promising candidates for versatile applications in medicine, pharmacy, food industry, and agriculture. Nowadays the preparation of nanosized fibrous materials is of special interest because of their unique properties, in particular their high surface area-to-volume and aspect ratios. Electrospinning is a cutting edge technique for fabrication of continuous polymer micro- and nanofibers. The basic principles and the effect of the process parameters on the morphology of the electrospun fibers and fibrous materials are briefly discussed in the present Chapter. The first successful attempt to prepare chitosan-containing electrospun materials dates from 2004. This has been achieved by the addition of a non-ionogenic, water-soluble polymer into the spinning solution. The application of this approach for preparation of chitosan-containing fibers is thoroughly discussed in the Chapter. The preparation of neat chitosan nanofibers by electrospinning is outlined as well. The application of suitable chitosan derivatives soluble in water or low toxic organic solvents enables the design of novel non-woven textiles in absence/presence of a non-ionogenic polymer. The preparation of such non-toxic, environmentally friendly materials is detailed. The applied two-step procedures (heat or UV treatment, use of appropriate crosslinking agents) for imparting water-insolubility to the obtained micro- and nanofibrous materials are described. The main approaches that have been used for preparation of electrospun materials combining the beneficial properties of chitosan and aliphatic polyesters based on poly(L-lactide): simultaneous electrospinning or electrospinning of the polyester, followed by coating of the non-woven textile with a thin chitosan layer, are summarized. Moreover, the recently developed routes for preparation of chitosan-containing micro- and nanofibers, such as reactive electrospinning, combination of electrospinning and polyelectrolyte complex formation as well as yarns formation, are discussed. The advantages of the one-step imparting of water-insolubility of chitosan fibers by reactive electrospinning and polyelectrolyte complex formation as compared to the twostep procedures are emphasized. Last but not least the potential biomedical application of the obtained micro- and nanofibers are outlined. Chapter 4 - The precise control of fiber diameter during electrospinning is very crucial for many applications. A systematic and quantitative study on the effects of processing variables enables us to control the properties of electrospun nanofibers. In this contribution, response surface methodology (RSM) was employed to quantitatively investigate the simultaneous effects of four of the most important parameters, namely solution concentration (C), spinning distance (d), applied voltage (V) and volume flow rate (Q) on mean fiber diameter (MFD) as well as standard deviation of fiber diameter (StdFD) in electrospinning of polyvinyl alcohol (PVA) nanofibers. Chapter 5 – Electrospun (ES) and vapor grown carbon nanofibers (VGCFs) are attractive building blocks for functional nanoscale devices. They are promising candidates for various applications, including filtration, protective clothing, polymer batteries and sensors. The continuous progress of nanotechnology in material science has led to the development of nanostructure materials with unique chemical, physical, and thermal properties. Nanofibers

Preface

ix

possesses significant characteristic that exhibit enormous availability of surface area per unit mass. Furthermore, their high surface-to-volume ratio renders them attractive as catalyst supports, energy storage devices, as well as in drug delivery and tissue engineering. Since the discovery of carbon nanotubes in 1991 and based on the results obtained from the characterization of these nanostructures, many other carbon-based nanomaterials have been developed. Of these, carbon nanofibers (CNFs), fullerenes, carbon nanohorns, and nanoporous structures are the subject of extensive experimental and theoretical studies for specific applications. Carbon is a truly remarkable element existing as four allotropes, viz. diamond, graphite, carbynes and fullerenes, each having significant scientific and technological importance. Its most abundant allotrope, graphite, can take many forms with respect to microstructure, amorphous to highly crystalline structure, highly dense with density 2.2 g/cm3 to highly porous with density 0.5 g/cm3 and different shapes. These types of graphites are called synthetic carbons and in technical terms, engineered carbons. Carbon nanofibers are unique in the fact that their whole surface area can be activated. Since carbon nanofibers have a much larger functionalized surface area compared to that of nanotubes, the surface-active groups-to-volume ratio of these materials is much larger than that of the glassy-like surface of the carbon nanotubes. This characteristic, combined with the fact that the number and type of functional groups on the outer surface of the carbon fibers can be well controlled, is expected to allow for the selective immobilization and stabilization of functional biomolecules such as proteins, enzymes, and DNA. Also, the high conductivity of carbon nanofibers seems to be ideal for the electrochemical signal transduction. The oxygen-containing activated sites are ideal for the immobilization and stabilization of biomaterials is an important feature. Industrial applications of these new materials include: polymer and elastomer fillers, commercial hydrogen storage systems, radiowave-absorbing composites, lithium battery electrodes, construction composites, oil additives, gas-distribution layers for fuel cells, absorbents and filters as well as in capacitive deionization (CDI) processes for water treatment. Many applications are being developed for field emission display, electrodes of secondary battery and reinforcement of materials. Among the many future possibilities includes; soft protective vests stronger than Kevlar, bandages that can contract to put pressure on, artificial muscles powered by electricity those expected much lighter than current hydraulics, would make it easier to incorporate electronic sensors and actuators into clothing. All of these possible applications derive from the remarkable properties of carbon nanofiber, the ability to conduct both heat and electricity along with the extreme toughness of the fiber. This chapter presents the various features of carbon nanofibers with elaborated properties description in connection with their different application, produced from electrospinning and vapor grown techniques. Carbon fibers are fibrous carbon materials with carbon content more than 90%. They are transformed from organic matter by 1000-1500oC heat treatment, which are substance with imperfect graphite crystalline structure arranged along the fiber axis. Chapter 6 - Carbon nanofibers (CNFs) are platelet or conical (herringbone) carbon nanostructures consisting of nested cup-shaped or platelet graphene sheets stacked along the long fiber axis. CNFs typically have diameters on the nanometer scale and lengths on the micrometer scale and possess attractive properties, such as large surface area, high electrical conductivity, and good mechanical strength and thermal stability. As-prepared CNFs have surfaces along the long axis that terminate in C(sp2)-H edge sites that are suitable for

x

W. N. Chang

chemical functionalization. The surface charge, wettability, dispersibility, and chemical reactivity of CNFs can be altered through chemical and physical modifications of these CNF surface sites. CNFs or surface-functionalized CNFs have been used as sensor media either as pristine nanofibers or as CNF-based composites. Large changes in electrical properties, such as electrical resistance, are observed depending on the presence or absence of gaseous analytes that adsorb, bind, or electrochemically react with the CNF component. CNFs functionalized with biomolecules, such as enzymes or DNA oligomers, have been used as biosensors to detect complexation of specific proteins or complementary DNA oligomers. CNFs also act as mechanical sensors. Deflection along the vertical axis of CNFs by acoustic fields results in the generation of electrical fields that can be detected. This chapter summarizes diverse applications in which CNFs are used as sensor media. Chapter 7 - Electrospinning is a simple and versatile method for producing nanofibers from various materials including polymers, composites, carbons, ceramics, and metals. One unique and important aspect of electrospinning is its ability to manipulate the structures of nanofibers through careful control of processing parameters, including: i) intrinsic properties of the spinning solution such as rheological behavior, conductivity, surface tension, polymer molecular weight, and solution concentration; and ii) operational conditions such as voltage, solution flow rate, nozzle diameter, spinneret-collector distance, spinneret configuration, and motion of the collector. This chapter addresses the fundamental relationships between processing and structures of electrospun nanofibers and the utilization of such first-principle knowledge to achieve nanofibers with desirable structures. Nanofiber structures that are covered include fiber diameter, primary pore structure, secondary pore structure, and other secondary pore structures. The focus of this chapter is on polymer nanofibers, but electrospun fibers of other materials, such as composites, carbons, ceramics, and metals, are also discussed. Chapter 8 - As a kind of biomacromolecules, carbohydrates are found on the external surface of cell membranes in the forms of glycoproteins, glycolipids and polysaccharides. They play essential roles in biological processes such as cell adhesion, blood coagulation, viral infection, immune response and apoptosis. Numerous biological phenomena are based on the carbohydrate-protein interaction. In nature, carbohydrates always interact with specific proteins through multivalent interaction, namely “cluster glycoside effect”. A great number of glycopolymers have been designed and synthesized to mimic the multivalent functions of natural glycoconjugates. It is expected that nanostructured materials with morphology similar to the native extracellular matrix will be more interesting for the mimicking of “cluster glycoside effect”. Therefore, a series of polyacrylonitrile-based nanofibers with glycosylated surfaces were studied in our laboratory. Two protocols were used to fabricate these glycosylated nanofibers. One is the synthesis of glycopolymers followed by electrospinning and the other is the surface modification of polyacrylonitrile nanofibers having reactive groups. The authors found that the morphology of the glycosylated nanofibers could be modulated by the characteristics of glycopolymers and the parameters of electrospinning and surface modification. These glycosylated nanofibers were studied for protein adsorption and recognition. Concanavalin A (Con A), peanut agglutinin (PNA) and bovine serum albumin (BSA) were used for comparison. Water contact angle measurement confirms that the glycosylated nanofiber surface is hydrophilic which facilitates the resistance to the nonspecific adsorption of proteins. Because of the specific interaction of Con A and glucose

Preface

xi

residues, nanofibers with glucose groups have strong affinity with Con A, but present no binding with PNA or BSA. By contrast, those with galactose groups can selectively recognize PNA instead of Con A or BSA. The results suggest that the glycosylated nanofibers possess the capability to recognize the corresponding protein, which is strongly dependent on the specific carbohydrate-protein interaction. Furthermore, the adsorbed surface can be regenerated by incubation with high concentration of sugar solutions and then be reused. As a consequence, it is believed that the glycosylated nanofibers have potential applications in separation and purification of proteins. Chapter 9 - Electrospinning has been suggested as a useful method to prepare non-woven fabrics of sub-micron or nano-scale fibers, which have high porosity and large surface areato-volume ratio, small pore size between the depositing fibers of the electrospun mats, and vast possibility for surface functionalization. These characteristics make the non-woven fabrics attractive for many applications, such as functional membranes, photocatalysts, biosensors, and nanoelectronics. On the other hand, porphyrins play important roles in biological processes and much attention has been paid to design and synthesize porphyrinfunctionalized polymers for potential applications including molecular recognition or molecular imprinting, sensors, interactions with biological systems, and enzyme mimics for catalysis. Combining the merits of electrospinning with the bioinspired applications of porphyrinated polymers may generate functionalized nanofibers for more multiple purposes. Following this idea, various porphyrinated polymers, which include polyacrylonitrile, polyimide and polypeptide, were synthesized by either physical blending or chemical copolymerization. They were fabricated into nanofibrous membranes by electrospinning process. The authors found these porphyrinated polymer nanofibers not only preserved their nature characteristics but also endowed new spectroscopy properties of porphyrins. On the other hand, it is well known that porphyrins have been used as red emitting materials that have reasonable fluorescence efficiency and good thermal stability. Based on this, fluorescent microspheres or nanofibers with different diameters were prepared from the porphyrinated polymers by changing the parameters for electrospinning process, such as solution concentration and molecular weight of the polymers. Confocal laser scanning microscopy (CLSM) showed that red light emitted uniformly through out the nanofibers in spite of fiber morphologies. Besides being used as emitting materials, the authors expect that these luminescent nanofibers may be latent materials applied in many areas such as catalysis, molecular imprinting, biosensors, and light/energy conversion. Chapter 10 - Thousands of new cases of spinal cord injury occur each year in the USA alone. However, despite recent advances, there is at present no cure for the resulting paraplegia or quadriplegia. This chapter evaluates a spinal cord prosthetic (SCP) developed in our laboratoy that is comprised of longitudinally bundled strips of nanofibers whose surfaces have been modifed with fibroblast growth factor-2 (FGF-2). The SCP is designed to be a prefabricated implant that can be grafted into the lesion site not only to provide structural but also to provide chemical cues that permit regenerating axons to cross the lesion site. For a comparative study, two separate SCPs were produced with one containing unmodified nanofibers and the other containing FGF-2-modified nanofibers. Both SCPs correctly guided regenerating axons across the injury gap created by an over-hemisection to the adult rat thoracic spinal cord and encouraged revascularization of the injury site. Neither SCP initiated glial scarring when implanted into the injured rat spinal cord. However, devices that incorporated nanofibers modified with FGF-2 encouraged more axonal regrowth and

xii

W. N. Chang

significantly better functional recovery than did devices that incorporated unmodified nanofibers as assessed using the Basso, Beattie, Bresnahan (BBB) locomotor rating scale. As such, the FGF-2-modified SCP provides a multi-faceted approach to spinal cord repair. Chapter 11 - Electrospinning has recently emerged as a leading technique for the formation of nanofibrous structures made of synthetic and natural extracellular matrix components. In this chapter, nanofibrous scaffolds were obtained by electrospinning a combination of poly(3-hydroxybutyrate-co-3-hydroxyvalerate)(PHBV) and cell attachment factor such as type-I collagen, gelatin and keratin in 1,1,1,3,3,3-hexafluoro-2-isopropanol (HIFP). The resulting fibers ranged from 300 to 800 nm in diameter. Their surfaces were characterized by attenuated total reflection Fourier transform infrared spectroscopy (ATRFTIR), electron spectroscopy for chemical analysis (ESCA) and atomic force microscopy (AFM). The PHBV and protein components such as collagen and keratin were biodegraded by PHB depolymerase, type-I collagenase and trypsin solution, respectively. The results of cell adhesion experiment showed that NIH 3T3 cells more adhered to the PHBV/protein nanofibrous mats than to the PHBV nanofibrous one. It was also found, from a BrdU assay, that the PHBV/protein nanofibrous mats could accelerate the proliferation of fibroblast cells more effectively than the PHBV nanofibrous mats, suggesting the good scaffolds for tissue engineering. Chapter 12 - In this study, vapor grown carbon nanofiber (CNF) has been used to improve thermal and mechanical properties of polypropylene. The CNFs were first dispersed over the polypropylene particles using sonication coating method, and then extruded into filaments with a single screw extruder. The thermal properties of neat and nanophased polypropylene were characterized by TGA and DSC. TGA thermograms showed that the nanoparticle-infused systems are more thermally stable, and DSC results indicated that CNFs have no effect on melting temperature. Tensile tests were performed on the single filament at a strain rate range from 0.02/min to 2/min. Results indicate that both neat and nanophased polypropylene were strain rate-strengthening material. The tensile modulus and yield strength both increased with increasing strain rate. Experiment results also show that infusing polypropylene with nanofibers increases tensile modulus and yield strength, but decreases ductility. Finally, based on the tensile test results, a nonlinear constitutive equation was developed to describe strain rate-sensitive behavior of neat and nanophased polypropylene. Chapter 13 - Carbon nanotubes (CNTs) have a number of outstanding mechanical and physical properties which make them attractive as reinforcement in polymer matrix. CNTs reinforced polyurethane nanocomposites provide the possibility to tailor the material strength, stiffness and thermal behavior of polyurethane. Multi-walled carbon nanotubes (MWNTs) filled polyurethane composites were prepared by mixing and injection molding and its thermal characteristics were investigated. The analysis of the influences of MWNTs particles and composites mixing methods were done using dynamic mechanical analysis (DMA), differential scanning calorimetry (DSC) and thermogravimetry analysis (TGA). The storage moduli of the composites increased with increasing MWNTs loading which indicate a good matrix/filler adhesion and increased the stiffness of the composites. However, the increase in processing speed has decrease the storage modulus. Addition of MWNTs filler also broadened and lowered the peak of tan δ denotes that the polyurethane composite became more elastic. Thermal stability of the polyurethane was improved with MWNTs loading which is associated to high thermal stability of CNTs. A very high processing speed reduced the composites thermal stability making it easier to degrade. DSC analysis indicated that the

Preface

xiii

inclusion of MWNTs increased the melting temperature and act as restriction sites for the polyurethanes soft segments. The present chapter revealed the potential of MWNTs as agent for better thermal properties of polyurethane nanocomposites and their properties depend strongly on the dispersion and distribution of nanotubes in polyurethane matrix. Chapter 14 - Polyurethane composites filled with multi-walled carbon nanotubes (MWNTs) were prepared by mixing and injection molding and its thermal as well as electrical conductivity characteristics were investigated. The influences of MWNTs addition and mixing methods on thermal and electrical conductivity of MWNT/polyurethane nanocomposites were investigated using a high resistance meter and thermal conductivity analyzer. The electrical resistivity of MWNTs incorporated polyurethanes were decreased in relation to filler concentration which is attributed by the formation of a conductive path made up from MWNTs particle. Increasing the processing speed will further decrease the resistivity because the dispersion of CNTs in polymer is improved. Higher processing speed samples shows resistivity values closer to the theoretical value because of better dispersion of CNTs in polyurethane and more conductive pathway were formed. The result shows that the addition of MWNTs fillers improved the thermal conductivity of the polyurethane composites. Higher filler concentration and higher shear rate results in better thermal conductivity because better formation of thermally conductive networks along polymer matrix to ensure the thermal was conducted through the matrix and the network along the polymer composites. The theoretical thermal conductivity comparisons approximately agree with the experimental measurements for the composites studied. The present study revealed the potential of MWNTs as agent for better thermal and electrical conductivities of polyurethane nanocomposites and their properties depend strongly on the dispersion and distribution of nanotubes in polyurethane matrix.

In: Nanofibers: Fabrication, Performance, and Applications ISBN 978-1-60741-947-1 Editors: W. N. Chang © 2009 Nova Science Publishers, Inc.

Chapter 1

SYNTHESIS AND ELECTRON FIELD EMISSION FROM DIFFERENT MORPHOLOGY CARBON NANOFIBERS Sk. F. Ahmed1,a and K. K. Chattopadhyay2,b 1

Future Fusion Technology Laboratory, Korea Institute of Science and Technology, P.O. Box 131, Cheongryang, Seoul 130-650, Republic of Korea 2 Thin Film and Nanoscience Laboratory, Department of Physics, Jadavpur University, Kolkata, India

1. INTRODUCTION Carbon Nanofiber and Carbon Nanotube Carbon, which belongs to group IV of the periodic table, is the lightest element in this group, and it possesses countless interesting physical and chemical properties. Among the different types of supports used in heterogeneous catalysis carbon materials attract a growing interest due to their specific characteristics which are mainly: (i) resistance to acid / basic media, (ii) possibility to control, up to certain limits, the porosity and surface chemistry and (iii) easy recovery of precious metals by support burning resulting in a low environmental impact. In contrast to Si, Ge and Sn, which have the same number of electrons in the outermost shell as carbon and can only exist in cubic sp3 hybridization, carbon not only exhibits sp3 hybridization (diamond), but also planar sp2 hybridization as in the graphite structure and sp1 hybridization as in carbynes. Each carbon atom has six electrons which occupy 1s2, 2s2 and 2p2 atomic orbitals. The 1s2 orbital contains two strongly bound core electrons. Four more weakly bound electrons occupy the 2s22p2 valence orbitals. In the crystalline phase, the valence electrons give rise to 2s, 2px, 2py and 2pz orbitals which are important in forming covalent bonds in carbon materials. Since the energy difference between the upper 2p energy levels and the lower 2s level in carbon is small compared with the binding energy of the chemical bonds, the electronic wave functions for these four electrons aE-mail: [email protected] (Sk. F. Ahmed). bE-mail: [email protected] (K. K. Chattopadhyay, Corresponding author).

2

Sk. F. Ahmed and K. K. Chattopadhyay

can readily mix with each other, thereby changing the occupation of the 2s and three 2p atomic orbitals so as to enhance the binding energy of the carbon atom with its neighboring atoms. The general mixing of 2s and 2p atomic orbitals is called hybridization, whereas the mixing of a single 2s electron with one, two, or three 2p electrons is called spn hybridization with n = 1,2,3 [1,2]. The bonding structures of diamond, graphite and nanotubes, or fullerenes are shown in Figure 1.1. When a graphite sheet is rolled over to form a nanotube, the sp2 hybrid orbital is deformed for rehybridization of sp2 toward sp3 orbital or σ−π bond mixing. This rehybridization structural feature, together with electron confinement, gives nanofibers/nanotubes unique, extraordinary electronic, mechanical, chemical, thermal, magnetic, and optical properties [3-5]. The physical reason why these nanostructures form is that a graphene layer (defined as a single 2D layer of 3D graphite) of finite size has many edge atoms with dangling bonds, index dangling bonds and these dangling bonds correspond to higher energy states. Therefore the total energy of a small number of carbon atoms (30 -100) is reduced by eliminating dangling bonds, even at the expense of increasing the strain energy, thereby promoting the formation of closed cage clusters such as fullerenes and carbon nanotubes. For example, diamond and layered graphite forms of carbon are well known, but the same carbon also exists also in planar sheet, rolled up tubular, helical spring, rectangular hollow box, and nanoconical forms. Elemental carbon in the sp2 hybridization can form a variety of amazing structures. Apart from the well-known graphite, carbon can build closed and open cages with honeycomb atomic arrangement. First such structure to be discovered was the C60 molecule by Kroto et al. in 1985 [1]. Although various carbon cages were studied, it was only in 1991, when Iijima [6] observed for the first time tubular carbon structures. Two years later, Iijima and Ichihashi [7] and Bethune et al. [8] synthesized single-walled carbon nanotubes (SWNTs). Actually carbon nanotubes (CNTs) are allotropes of carbon. Nanotubes are members of the fullerene structural family, which also includes buckyballs. Whereas buckyballs are spherical in shape, a nanotube is cylindrical, with at least one end typically capped with a hemisphere of the buckyball structure. The nature of the bonding of a nanotube is described by applied quantum chemistry, specifically, orbital hybridization. Nanotubes are composed entirely of sp2 bonds, similar to those of graphite. This bonding structure, which is stronger than the sp3 bonds found in diamond, provides the molecules with their unique strength. Nanotubes naturally align themselves into "ropes" held together by Van der Waals forces. Under high pressure, nanotubes can merge together, trading some sp2 bonds for sp3 bonds, giving great possibility for producing strong, unlimited-length wires through highpressure nanotube linking [9]. A single-wall carbon nanotube (SWCNT) is best described as a rolled-up tubular shell of graphene sheet (Figure 1.2(a)) which is made of benzene-type hexagonal rings of carbon atoms [10-12]. There are many possible orientations of the hexagons on the nanotubes, even though the basic shape of the carbon nanotube wall is a cylinder. A single walled carbon nanotube is a graphene sheet appropriately rolled into a cylinder of nanometer size diameter [13,14]. The planar sp2 bonding, which is characteristic of graphite, plays a significant role in carbon nanotubes. The body of the tubular shell is thus mainly made of hexagonal rings (in a sheet) of carbon atoms, whereas the ends are capped by half-dome shaped half-fullerene molecules. The internal diameter of these structures can vary between 0.4 to 2.5 nm and the length ranges from few microns to several millimeters.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers

3

Figure 1.1. Bonding structures of diamond, graphite, nanotubes and fullerenes[From ref. 3].

Since the single wall carbon nanotube is only one atom thick and has a small number of atoms around its circumference, only a few wave vectors are needed to describe the periodicity of the nanotubes. These constraints lead to quantum confinement of the wave functions in the radial and circumferential directions, with plane wave motion occurring only along the nanotube axis corresponding to a large number or closely spaced allowed wave

4

Sk. F. Ahmed and K. K. Chattopadhyay

vectors. Thus, although carbon nanotubes are closely related to a 2D graphene sheet, the tube curvature and the quantum confinement in the circumferential direction lead to a host of properties that are different from those of a graphene sheet. A multi-wall carbon nanotube (MWCNT) is a rolled-up stack of graphene sheets of coaxial SWCNTs, with the ends again either capped by half-fullerenes or kept open. Both SWCNTs and MWCNTs have physical characteristics of solids and are nanocrystals with high aspect ratios of 1000 or more, although their diameter is close to molecular dimensions. The number of walls present can vary from two (double wall nanotubes) to several tens, so that the external diameter can reach upto 100 nm. The concentric walls are regularly spaced by 0.34 nm similar to the inter graphene distance evidenced in turbostatic graphite materials. The main difference between nanotubes and nanofibers consists in the lack of a hollow cavity for the latter. The diameters of carbon nanofiber (CNF) are generally higher than the ones presented by nanotubes and can easily reach 500 nm. A nomenclature (n,m), used to identify each single-wall nanotube, refers to integer indices of two graphene unit lattice vectors corresponding to the chiral vector of a nanotube. Chiral vectors determine the directions along which the graphene sheets are rolled to form tubular shell structures and perpendicular to the tube axis vectors. Figure 1.2 shows the schematic representation of the construction of a nanotube by rolling-up an infinite strip of graphite sheet (so called graphene). In Figure 1.2(a) the chiral vector Ch = na1 + ma2 connects two lattice points O and A on the graphene sheet, where n and m are integers, a1 and a2 the unit cell vectors of the two-dimensional lattice formed by the graphene sheets. The direction of the nanotube axis is perpendicular to this chiral vector. An infinite strip is cut from the sheet through these two points, perpendicular to the chiral vector. The strip is then rolled-up into a seamless cylinder. T = t1a1 + t2a2 is the primitive translation vector of the tube [15]. The nanotube is uniquely specified by the pair of integer numbers n, m or by its radius R = Ch / 2π and chiral angle θ which is the angle between Ch and the nearest zigzag of C–C bonds. All different tubes have angles θ between zero and 30o. Special tube types are the achiral tubes (tubes with mirror symmetry): when n = m, the nanotube is called “armchair” type (γ = 0o) [Figure 1.2(b)]; when m = 0, then it is of the “zigzag” type (γ = 30o) [Figure 1.2(c)]. Otherwise, when n = m, it is a “chiral” tube and γ takes a value between 0o and 30o [Figure 1.2(d)]. The value of (n,m) determines the chirality of the nanotube and affects the optical, mechanical and electronic properties. Nanotubes with |n - m| = 3q are metallic and those with |n - m| = 3q ± 1 are semiconducting (q is an integer). Exhaustive studies concerning electronic properties of both SWCNT [16] and MWCNT [17] are available in the literature, whereas carbon nanofiber (CNF) are often considered as conductive substrates that can exert electronic perturbations similar to those of graphite [18]. In the case of SWCNT, studies have demonstrated that they behave like pure quantum wires (1D-system) where the electrons are confined along the tube axis. Electronic properties are mainly governed by two factors: the tube diameter and the helicity, which is defined by the way in which the graphene layer is rolled up (armchair, zigzag or chiral) [8]. In particular, armchair SWCNTs are metallic and zigzag ones display a semi-conductor behavior. Studies on MWCNTs electronic properties have revealed that they behave like an ultimate carbon fiber at high temperature their electrical conductivity may be described by semi-classical models already used for graphite, whereas at low temperature they reveal 2D-quantum transport features [17]. Nanotubes can be as small as 1 nm in diameter and as long as 100,000 nm.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers

5

Figure 1.2. Schematic representation of the construction of a nanotube by rolling-up an infinite strip of graphite sheet.

These tubes are extremely strong, approaching the strength of diamond, and also dissipate heat better than any other known material. Carbon nanofibers/nanotubes are one of the strongest and stiffest materials known, in terms of tensile strength and elastic modulus respectively.

6

Sk. F. Ahmed and K. K. Chattopadhyay

This strength results from the covalent sp2 bonds formed between the individual carbon atoms. Depending on how they are configured, CNFs/CNTs are good conductors of electricity and can also act as semi-conductors for molecular electronics. CNFs/CNTs are three dimensional as opposed to the current silicon based electronics that are two-dimensional. They appear to be able to extend the miniaturization process by several additional orders of magnitude over current methods. As they conduct electricity better than copper and can also act as semi-conductors. CNFs/CNTs have very good elasto-mechanical properties because the two dimensional arrangement of carbon atoms in a graphene sheet allows large out-of-plane distortions, while the strength of carbon-carbon in-plane bonds keeps the graphene sheet exceptionally strong against any in-plane distortion or fracture. Some of the properties of carbon nanotube, which are forming a driving force for their wide range of applications, are shown in Table -1. Table 1. Some fundamental properties of carbon nanofiber/nanotube Properties Average Diameter: SWNT's MWCNTs CNFs Young's Modulus Maximum Tensile Strength Band Gap: For Metallic For Semi-Conducting Thermal Conductivity (Room Temp.) Carrier mobility Semi-Conducting NT Maximum Current Density Turn-on field Threshold field for CNTs/CNFs

Value 1.2 – 1.4 nm 2 – 100 nm 10 nm – 100 μm ~ 1 TPa ~ 63 Gpa 0 eV 0.18 – 1.8 eV 3000 W/mK 105 cm2/Vsec 109 A/m2 1.5 – 7.5 V/μm 1.5 – 9.5 V/μm

References [19] [20] [21] [22,23] [24] [25] [25] [26] [27] [28] [29,30] [31,32]

History of Carbon Nanofiber and Nanotube We provide here a brief review of the history of carbon fibers, the macroscopic analog of carbon nanotubes, as carbon nanotubes have become the focus of recent developments in carbon fibers. Since last decade, new carbon forms like carbon nanofibers (CNF) or nano filaments and carbon nanotubes (CNT) have generated an interest in the scientific community. However, it has got to be remembered that carbon nano filaments have been synthesized for very long as products from the action of a catalyst over the gaseous species originating from the thermal decomposition of hydrocarbons. One of the first evidence that the nano filaments thus produced could have been nanotubes, exhibiting an inner cavity, can be found in the transmission electron microscope micrographs published by Hillert et al. in the year of 1958 [33]. Radushkevich et el. published clear images of 50 nanometer diameter tubes made of carbon in the year 1952 [34]. This discovery was largely unnoticed, the article was published in the Russian language. The production of graphite nanofibers is even older and the first reports date of more than a century [35,36]. Their efforts were mostly directed toward the study of vapor grown carbon filaments, showing filament growth from the thermal

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers

7

decomposition of hydrocarbons. The second applications-driven stimulus to carbon fiber research came in the 1950’s from the needs of the space and aircraft industry for strong, stiff light-weight fibers that could be used for building lightweight composite materials with superior mechanical properties. This stimulation led to great advances in the preparation of continuous carbon fibers based on polymer precursors, including rayon, polyacrylonitrile (PAN) and later mesophase pitch. The late 1950’s and 1960’s was a period of intense activity at the Union Carbide Corporation, the Aerospace Corporation and many other laboratories worldwide. This stimulation also led to the growth of a carbon whisker [37], which has become a benchmark for the discussion of the mechanical and elastic properties of carbon fibers. The growth of carbon whiskers was also inspired by the successful growth of single crystal whisker filaments at that time for many metals such as iron, non-metals such as Si, and oxides such as Al2O3, and by theoretical studies [38], showing superior mechanical properties for whisker structures [39]. Parallel efforts to develop new bulk synthetic carbon materials with properties approaching single crystal graphite led to the development of highly oriented pyrolytic graphite (HOPG) in 1962 by Ubbelohde and co-workers [40,41], and HOPG has since been used as one of the benchmarks for the characterization of carbon fibers. While intense effort continued toward perfecting synthetic filamentary carbon materials, and great progress was indeed made in the early 1960’s, it was soon realized that long term effort would be needed to reduce fiber defects and to enhance structures resistive to crack propagation. New research directions were introduced because of the difficulty in improving the structure and microstructure of polymer-based carbon fibers for high strength and high modulus applications, and in developing graphitizable carbons for ultra-high modulus fibers. Because of the desire to synthesize more crystalline filamentous carbons under controlled conditions, synthesis of carbon fibers by a catalytic Chemical Vapor Deposition (CVD) process was developed, laying the scientific basis for the mechanism and thermodynamics for the vapor phase growth of carbon fibers in the 1960’s and early 1970’s. In parallel to these scientific studies, other research studies focused on control of the process for the synthesis of vapor grown carbon fiber [42-45], leading to the more recent commercialization of vapor grown carbon fibers in the 1990’s for various applications. Concurrently, polymer-based carbon fiber research has continued worldwide, mostly in industry, with emphasis on greater control of processing steps to achieve carbon fibers with ever-increasing modulus and strength, and on fibers with special characteristics, such as very high thermal conductivity, while decreasing costs of the commercial products. As research on vapor grown carbon fibers on the micrometer scale proceeded, the growth of very small diameter filaments less than 10 nm, was occasionally observed and reported [46-47], but no detailed systematic studies of such thin filaments were carried out. Oberlin et al. clearly showed hollow carbon fibres with nanometer-scale diameters using a vapour-growth technique [48]. The interest in fibrous carbon has since then been recurrent and a significant boost in the research in carbon nanostructure field coincides with the discovery of multiwall carbon nanotubes (MWNT) by Iijima in 1991 [6]. It is likely that carbon nanotubes were produced before this date, but the invention of the transmission electron microscope allowed the direct visualization of these structures. Carbon nanotubes have been produced and observed under a variety of conditions prior to 1991. The arc discharge technique was well known to produce the famed Buckminster fullerene on a preparative scale [49] and these results appeared to extend the run of accidental discoveries relating to fullerenes. The original observation of fullerenes in mass

8

Sk. F. Ahmed and K. K. Chattopadhyay

spectrometry was not anticipated [5], and the first mass-production technique by Kratschmer et al. was used for several years before realising that it produced fullerenes [49].

2. SYNTHESIS AND GROWTH MECHANISM OF CARBON NANOTUBE AND CARBON NANOFIBER BY DIFFERENT PROCESS There have been major developments in the synthesis processes and characterization techniques of CNT and CNF. It is possible to produce CNTs by a wide range of deposition methods. The most popular deposition techniques are arc discharge [6,50], laser ablation [51], chemical vapour deposition (CVD) [52,58-62] plasma enhanced chemical vapour deposition [53] and solvothermal process [54-55]. The earliest approach to produce nanotubes was an arc process [56] as pioneered by Iijima in 1991. This was shortly followed by a laser ablation technique developed at Rice University [57]. Chemical vapor deposition (CVD) has become a common technique to grow nanotubes in the last ten years [58-62]. The figure-of-merit for an ideal growth process depends on the application. For development of composites and other structural applications, the expected metric is the ability to achieve controlled growth of specified thickness on patterns is important for applications in nano electronics, field emission, displays, and sensors. The arc process involves striking a dc arc discharge in an inert gas (such as argon or helium) between a set of graphite electrodes [6,56]. The electric arc vaporizes a hollow graphite anode packed with a mixture of a transition metal (such as Fe, Co or Ni) and graphite powder. The inert gas flow is maintained at 50-600 Torr. Nominal conditions involve 2000 - 3000° C, 100 amps and 20 volts. This produces SWCNTs in mixture of MWCNTs and soot. In the arc-discharge synthesis of nanotubes, Bethune et al. in 1993 used as anodes thin electrodes with bored holes which were filled with a mixture of pure powdered metals (Fe, Ni or Co) and graphite [8]. The electrodes were vaporized with a current of 95-105 A in 100-500 Torr of He. The gas pressure, flow rate, and metal concentration can be varied to change the yield of nanotubes, but these parameters do not seem to change the diameter distribution. Typical diameter distribution of SWCNTs by this process appears to be 0.7 - 2 nm. On the other hand in laser ablation method, a target consisting of graphite mixed with a small amount of transition metal particles such as such as Co, Ni, Fe etc., catalyst is placed at the end of a quartz tube enclosed in a furnace [57]. A neodymium-yttrium-aluminum-garnet laser was employed to vaporize the target and helium or argon carrier gas was flowed through the tube. Argon gas flowing through the reactor, heated about 1200 °C by a furnace, carries the vapor and nucleates the nanotubes which continue to grow. The nanotubes are deposited on the cooler walls of the quartz tube downstream from the furnace. Both the above-mentioned methods were used to synthesize SWNTs in relatively large quantities. They are based on the condensation of hot carbon gases through vaporizing solid carbon, where temperatures of > 3000 K are initialized by either arc or laser. Due to such a high temperature, carbon nanotubes obtained exhibit high straightness and high crystallinity. However, depositions by both approaches are not directly on the substrates and are in a form of either powder or mat. Thus applications require additional manipulation or processes to deposit the CNTs on substrates. Lastly, the products obtained normally contain a large quantity of catalyst and carbon particles so that a purification step becomes necessary.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers

9

Chemical vapor deposition (CVD) methods have been successful in making carbon fiber and filament since more than 10-30 years ago [46,63,64], but in recent years, chemical vapor deposition (CVD) has progressed rapidly towards the growth of carbon nanotubes and has been adopted worldwide [65]. In this method, transition catalysts such as Fe, Co, Ni are deposited on the substrates (silicon, quartz, molybdenum etc.). A mixture of precursors such as hydrogen, methane, acetylene, and ammonia are flowed into the chamber. Assisted by either direct heating or by an external source such as plasma and hot filament, carbon gases are decomposed at the catalyst surface and carbon nanotubes are grown directly on the substrates at temperatures between ~ 700-1400 K [52]. The advantage of this method is that the nanotubes can be deposited directly on the substrate, which facilitates nanotube applications and mass production. Nevertheless, MWNTs are often obtained by CVD and only until recently SWNT growth by CVD was possible [66]. The drawback of the catalytic CVD-based nanotube production is the inferior quality of the structures that contain gross defects (twists, tilt boundaries etc.), particularly because the structures are created at much lower temperatures (600-1000 oC) compared to the arc or laser processes (~2000 oC). Several plasma based growth techniques have been reported [67-69] and in general, the plasma-grown nanotubes appear to be more vertically oriented than that is possible by thermal CVD. Since the plasma is very efficient in tearing apart the precursors and creating radicals, it is also hard to control and keep the supply of carbon low to the catalyst particles and hence, plasma based growth always results in MWCNTs and filaments. Different CVD systems which employ different approaches to dissociate the precursor gases have been used including thermal CVD [70], hot filament chemical vapour deposition (HFCVD) [53], and plasma enhanced chemical vapor deposition (PECVD) [71]. Several types of plasma systems which have been used including dc-plasma [72], radio frequency (RF) plasma [73], microwave plasma [74], and electron cyclotron resonance (ECR). In CVD technique, CNTs and CNFs are grown using the catalytic decomposition of hydrocarbons over transition metal catalysts [52]. The function of these metals is to facilitate the decomposition of the hydrocarbon gases and the formation of the tubular graphene structures. Chemical composition and particle size of the catalyst is expected to crucially affect the diameter and the number of walls of the carbon nanotubes [75]. The metal catalysts have been prepared by several methods including wet chemical solution [76], thin metal films [71], thick metal films/substrates [77], colloids [75], and solgel techniques [61]. In some cases surface treatments such as wet chemical etching [67], plasma etching [77], ion beam sputtering [62] and annealing [71] have been used to enable the formation of nano-particles before the growth. Effect of catalyst on growth of carbon nanofibers have been studied by Kamada et al. [78]. They have been successfully grown carbon nanofiber (CNF) films on Pd-Se, Fe-Ni, and Ni-Cu alloy catalysts at low temperatures by a thermal chemical vapor deposition method. Among these alloy catalysts, Ni-Cu alloy catalyst was found to be most suitable for low temperature growth of CNF. The CNFs grown using Pd-Se catalyst were found to have more defective structure than that obtained with the other catalysts, and exhibited excellent field emission property with threshold field ~ 1.1 V/μm. Merkulov et al. [79] evaporated Ni on n-type Si by E-gun. Shyu et al. deposited Fe-Ni with various components by e-beam evaporation [80]. Kin et al. [81] prepared copper-nickel powder by coprecipitation of the metal carbonates from mixed nitrate solutions using ammonium bicarbonate and a sequence complex treatment process including drying, calcining and reducing, etc. Carbon nanofibers (CNFs) were grown on a Ni–P alloy catalyst deposited on a silicon substrate in a microwave heating chemical vapor deposition system

10

Sk. F. Ahmed and K. K. Chattopadhyay

with methane gas at 650 oC [82]. The nanosized clusters on the clustered surface of the Ni–P alloy catalyst film directly provided the nucleation sites for CNFs without any pretreatment before the growth of the CNFs. The CNFs grown on the Ni–P alloy catalyst showed random orientation and it composed of parallel graphite planes. Figure 2.1 shows scanning electron microscopy images of CNFs grown at substrate temperature approximately 650 oC for 7 min. It reveals that the growth rate of CNFs is related to the thickness of Ni–P alloy catalyst film. The results indicate that the growth rate of the CNFs decreases as the thickness of the catalyst film increases. The above phenomenon explained by the diffusion of carbon atom into the catalyst particle. The growth of carbon nanostructures, including CNTs and CNFs, occurs by diffusion driven precipitation of carbon atoms from the supersaturated catalyst particles [83]. The size of catalyst particle increases and that causes the diffusion length to increase and the gradient of supersaturation to decrease. These factors will decrease the growth rate of CNFs. So, the thin catalyst film has a larger growth rate than the thick catalyst film. The result also proves that diffusion of carbon through the catalyst particle is the rate-determining step in the growth of carbon nanostructures using a Ni-P alloy catalyst.

Figure 2.1. SEM images of CNFs grown at 650 oC substrate temperature for 7 min. The corresponding thickness of Ni–P alloy catalyst film is 20 and 40 nm in images (a) and (b) respectively [From ref. 82].

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 11 Figure 2.2 is the SEM micrograph showing the surface morphology of CNFs grown at approximately 650 oC for 10 min. All of these SEM images show that these CNFs grown on the catalyst film with various thicknesses have similar morphology and are not vertically aligned but randomly tangled. The diameter of the CNFs in Figure 2.2(a)-(c) is approximately 30-70, 50-120 and 70-150 nm, respectively. The diameter of the CNFs increases as the size of the catalyst clusters increases with catalyst film thickness. The results show that the diameter of the CNFs is dependent on the initial thickness of the pre-deposited catalyst film. Wei et al. [69], Yudasaka et al. [84] and Bower et al. [85] also reported same type results using Fe, Co and Ni catalyst.

Figure 2.2. SEM images of CNFs grown at 650 oC substrate temperature for 10 min. The corresponding thickness of Ni–P alloy catalyst film is 20, 30 and 40 nm in images (a), (b) and (c), respectively [from ref. 82].

Vertically aligned carbon nanofiber and naotubes were synthesized by different technique and there are different growth mechanism proposed by different groups [86-92]. The physical and chemical characteristics of vertically aligned carbon nanofiber (VACNF) structures, in comparison to ideal multiwalled carbon nanotubes, offer inherent processing advantages imparted by their vertical architecture. The ability to control the VACNF growth rate is an important practical aspect of the synthesis process because some applications require high growth rates, whereas others would benefit from a lower growth rate but a high degree of uniformity and control over the final VACNF length. Thus understanding the factors that

12

Sk. F. Ahmed and K. K. Chattopadhyay

determine the growth rate is essential not only from the fundamental science point of view, but also from the point of view of practical applications. In conventional thermal CVD the growth of carbon nanofibers/nanotubes occurs in three main steps [63]; (i) decomposition of the carbonaceous gas molecules at the surface of the catalyst nanoparticle, (ii) diffusion of the resultant carbon atoms through the catalyst nanoparticle from the nanoparticle/gas interface towards the nanoparticle/nanofiber interface due to the concentration gradient, and (iii) precipitation of carbon atoms at the nanoparticle/nanofiber interface. Each of these steps can be a complex process by itself [93,94] and the whole picture is not completely understood, mainly due to the difficulty of conducting imaging and surface analysis in situ during the growth. In thermal CVD, the carbon feedstock is the molecules of the carbonaceous gas used in the growth process. The growth rate is determined by all of the three steps, but under typical growth conditions appears to be diffusion limited as suggested by the equality of the activation energies for the nanofiber growth rate and for the diffusion of carbon atoms through the catalyst [95]. Chuang et al. [86] prepared carbon nanofibers and carbon nanotubes using CH4 as a precursor material of carbon on Ni/Si and Ni/Ti/Si substrates at 640 oC and 700 oC by thermal chemical vapor deposition method. They have explained the growth of carbon nanofibers (CNF) and carbon nanotubes (CNT) on Ni/Si substrate through tip growth mechanism, and the growth mechanism of carbon nanotubes on Ni/Ti/Si substrate through root growth mechanism [87]. Figure 2.3(a) schematically shows the solid amorphous carbon nanofiber grown on Si substrate at 640 oC under the catalysis of Ni particle. During the heating of Ni/Si substrate from room temperature to 640 oC, the Ni film on Si surface would agglomerate into Ni nanoparticles, which still are in solid state. In the growth chamber the CH4 molecules are adsorbed on the upper surface of Ni particles and decomposed into C and H atoms. These C atoms are absorbed into Ni particles and diffuse to lower part through inside of Ni particles. Due to the difference of both temperature and concentration of carbons between upper and lower parts of Ni particles and the weak adhesion force between Ni particle and Si surface, once the concentration of C atoms exceeds the saturation solubility of Ni, the C atoms would precipitate on the lower surface of Ni particles and form carbon nanofiber. Because of the roughness of Ni particle surface, the precipitated carbons could not form graphene planes on the lower surface and hence only amorphous solid carbon nanofibers are obtained. Figure 2.3(b) schematically displays the hollow carbon nanofiber, i.e. carbon nanotube, is grown on Si substrate at 700 oC under the catalysis of Ni particle. When the growth temperature is raised from 640 - 700 oC, the Ni nanoparticles become fluid-like due to the lower melting temperature of Ni–C alloy and the effects of size and interfacial stress between Ni particle and carbon tube [96]. The surface of fluid-like Ni nanoparticle is much smoother than solid Ni nanoparticle, and the diffusion of C atoms on fluid-like Ni nanoparticle surface is much higher than on the surface of solid Ni nanoparticle. The C atoms are much easy to precipitate on the surface of middle part, not the lower part, of Ni nanoparticle, and form the parallel grapheme planes. Therefore, the carbon nanotubes are grown through the tip growth mechanism. Figure 2.3(c) schematically shows the hollow carbon nanofiber, i.e. carbon nanotube, is grown on Ti/Si substrate at 700 oC under the catalysis of Ni particle. Due to the strong adhesion force between Ni nanoparticle and Ti film surface, the growth force cannot push Ni nanoparticle up. Therefore, the carbon nanotubes are grown through the root growth mechanism. When the temperature of Ni/Ti/Si substrate increased, a part of Ni diffuse through Ti layer and into Si substrate to form NiSi and the rest of Ni would agglomerate to

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 13 form Ni particles [97]. The C atoms from the decomposition of CH4 under the catalysis of Ni particles diffuse into Ni particles and Ti layer. When ternary alloy of Ni-Ti-C was formed at the lower part of Ni particle due to the interdiffusion of Ni, Ti and C, the ternary alloy layer at the bottom of Ni particle would melt due to the lower melting point of the Ni-Ti-C ternary alloy [98]. Although the eutectic temperature of Ni-C is 1327 °C, there are many reports [99,100] indicating that the Ni catalyst particle had melted or was behaved fluid like during CNTs growth in the temperature of 600-900 °C due to the size effect of catalyst at nanometer level and the interfacial effect between catalyst and carbon. The eutectic temperature of NiTi-C is 1265-1295 °C, lower than 1327 °C, hence the Ni-Ti-C alloy layer at the bottom of Ni particle can melt due to the same reasons for Ni-C. Therefore Ni particles could sink into and firmly adhere to the Ti layer. When C atoms in Ni particles exceed the saturation solubility, they would form CNTs and grow up under the catalysis of Ni particles by root growth mechanism. Once Ni particles were surrounded by CNTs, the possibility for CH4 to touch Ni particles and decompose into C and H would decrease largely. For making CNT growth able to continue, the C atoms supplied from Ti layer. Figure 2.4 schematically shows the layer structure of substrates before and after CNTs growth. The three-layered Ni/Ti/Si substrate becomes four layers structure after CNTs growth. From the AES depth profile measurement of each element and the layer structure of sample they confirmed that the Ti interlayer couldn't prevent the outward diffusion of Si and the inward diffusion of Ni, C and Ti; however it can supply C atoms to continue the CNTs growth at later stage in CNT growth process.

Figure 2.3. Schematic diagrams of one dimensional carbon growth mechanism for solid amorphous carbon nanofiber grown through tip mechanism (a), carbon nanotube grown through tip mechanism (b), and carbon nanotube grown through root mechanism (c) [ From Ref. 86].

14

Sk. F. Ahmed and K. K. Chattopadhyay

Figure 2.4. Schematic diagrams for substrate (a) before CNTs growth and (b) mechanism for CNT growth [From Ref. 87].

Plasma enhanced chemical vapor deposition technique has achieved lower growth temperatures compared to other methods and vertical alignment of the nanotubes, which facilitates the CNT based device fabrication with current silicon technologies. It has been suggested [88] that in PECVD only VACNFs grown from the tip are aligned specifically due to the presence of the plasma electric field in the growth process, whereas VACNFs grown from the base are aligned mainly due to the crowding effect. Consequently, in the case of the base-type growth, deterministic synthesis of isolated VACNFs is expected to be rather difficult. Vertically aligned carbon nanofibers (VACNFs) were synthesized by direct-current plasma enhanced chemical vapor deposition using acetylene and ammonia as the gas source as shown in Figure 2.5 [101]. Generally in the case of PECVD growth the activation energies were found to be different [71] and the growth rate was suggested to be limited by the supply of carbon from the gas phase. Huang et al. showed that the VACNF growth rate depends quite strongly upon the gas mixture and plasma power used in the PECVD process [102]. This was attributed to changes in the chemical composition of the excited gas species. These species include (i) simple radicals created in the plasma as a result of direct dissociation of C2H2 and NH3 molecules and (ii) larger radicals that form due to collision and consequent attachment of radicals to each other as they move towards the substrate. It was suggested that changing the gas mixture or plasma power changes the chemical distribution of the excited species. Since different species are expected to have different decomposition rates at the catalyst surface, the nanofiber growth rate also changes. The fact that reduction of the C2H2 content resulted in a several-fold increase of the growth rate as well as a dramatic increase of the nitrogen content within the nanofibers strongly suggests that the growth occurs mainly due to the species created in the plasma, not due to unexcited C2H2 molecules.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 15

Figure 2.5. SEM image of an array of VACNFs grown by PECVD for 5 min and of randomly oriented CNFs/CNTs produced after the plasma was turned off and the growth run continued via purely thermal CVD [From Ref.101].

Figure 2.5 shows the SEM image of a sample for which VACNFs were first grown using a plasma, then the plasma was turned off, the NH3 flow was stopped, and C2H2 was flowed over the sample that was kept at the same high temperature as during the PECVD part of the growth (~700 oC). During the purely thermal phase, the VACNFs did not seem to increase in length, but very long thin non-aligned CNFs/CNTs that were not present during the PECVD part of the growth were produced. The non-aligned CNFs seemed to originate from the bases of VACNF, where perhaps some of the catalyst was still left. This indicates that the catalytic activities are quite different for the base- and tip-type growth modes. While the base-type catalytic growth can produce CNFs in the purely thermal process, the tip-type growth, requires the presence of radicals. The inability to re-grow VACNFs using only thermal CVD confirms the idea of the feedstock for the VACNF growth consisting mainly of radicals and not C2H2 molecules. There is another reason that a carbon shell may form around the catalyst nanoparticles sitting at the tips of VACNFs after the plasma is turned off, which prevents decomposition of carbonaceous species at the catalyst surface and consequently the VACNF growth. Since some applications may require very long VACNFs, it is highly desirable to develop controllable ways to further increase the growth rate. One way to relax the limit for the VACNF growth rate, imposed by the process of radical diffusion towards the substrate, is to change the radical transport mechanism from diffusive to forced flow by applying a pressure gradient perpendicular to the substrate surface to force more radicals to impinge on the surface [103]. The radical flux in this case will be given by F = Cv, where C is the radical concentration and v is an average velocity towards the substrate. Thus, by increasing the gas flow and therefore the radical velocity one can expect to achieve substantial increase in the VACNF growth rate. The high gas flow during the growth can be produced highly conical structures even for dense forests of VACNFs. Isolated VACNFs tend to assume conical shape during the growth due to reactive species emerging for the discharge and attaching to the

16

Sk. F. Ahmed and K. K. Chattopadhyay

sidewalls of the nanofibers [104]. In contrast, dense VACNF forests nominally consist of essentially cylindrical nanofibers due to the shielding of the sidewalls by neighboring VACNFs [79]. However, if the velocity of the incoming radicals is high enough, they will be able to penetrate the inter-fiber space for even densely spaced VACNFs and precipitate at the sidewalls thus forming conical structures [101]. Also the degree of conicity can be modified by changing the gas mixture. Merkulov et al. [88] proposed the following model to explain the vertical alignment of PECVD grown CNFs. The growth of CNFs occurs via decomposition of the carbonaceous gas molecules at the catalyst particle surface or in the glow discharge, diffusion of the carbon atoms through the particle, and subsequent precipitation at the particle/fiber interface [63]. The axis of a CNF growing perpendicular to the substrate coincides with the direction of the applied electrostatic force, resulting in a uniform tensile stress across the entire nanofiber/catalyst particle interface, as shown in Figure 2.6(a) and (d) Consequently, carbon uniformly precipitates across the interface and the fiber continues to grow vertically (perpendicular to the substrate). However, if there were a spatial fluctuation in the C precipitation at the interface, CNF growth would deviate from vertical alignment, as shown in Figure 2.6(c) and (b). In the case of nanofibers growing from the tip (catalyst particle at the tip), the electrostatic force produces a compressive stress at the particle/nanofiber interface where the greater rate of growth is seen [Figure 2.6(c)]. Likewise, a tensile stress is applied to the particle/nanofiber interface where the lesser rate of growth is seen. These opposing stresses favor subsequent C precipitation at the interface experiencing tensile stress and the lesser rate of growth. The net result is stable, negative feedback that acts to equalize the growth rate around the entire periphery of the particle/nanofiber interface, and vertically aligned CNFs are grown. The presence of the preferred direction of C precipitation can be caused by stress-induced diffusion [105] due to the stress gradient in the catalyst particle and possibly by the variation in the stress-dependent sticking of diffusing C atoms to the C side of the Ni-C interface. Since the nanofiber base is attached to the substrate, the stress created at the particle/ nanofiber interface with the greater growth rate is tensile [Figure 2.6(d)] and acts to continue the increased growth rate, thus causing the CNF to bend even further. The inherent instability of positive feedback control systems leads to the wildly varying CNF orientation. Ngo et al. synthesized vertically aligned carbon nanofibers (VACNFs) using palladium as a catalyst by plasma enhanced chemical vapor deposition (PECVD) [106]. Figure 2.7 shows the TEM micrographs of vertically aligned carbon nanofiber grown on thick Pd catalyst. They observed that the thick Pd films lead to a variety of growth morphologies including a hybrid tip growth phenomenon, as well as small cone angles that are imparted by the elongation/wetting of the inner cavity of the CNFs by the Pd catalyst. Huang et al. reported growth of core/shell carbon nanofibers and nanotubes using metal sulfide (FeS, CoS and NiS) as a catalyst by arc discharge technique [107]. Figure 2.8 shows the schematic growth model of the core–shell carbon nanofibers and large-cavity carbon nanotubes. The catalyst seed of metal sulfide come from melted metal sulfides or from the combination reaction of evaporated metal and sulfur ions dissociated from metal sulfides. The linkage of sulfur with metal and carbon, as well as the low-temperature environment prevents the core materials from extruding out of the carbon nanofibers during growth. If the core/shell carbon nanofibers are not exposed to high temperature during and after their growth, the stuffed sulfides will remain in the carbon shell and contract and separate upon cooling.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 17

Fig. 2.6 Alignment mechanism of carbon nanofibers. If a CNF grows vertically (along the electric-field lines), electrostatic force F creates a uniform tensile stress across the entire catalyst particle/nanofiber interface, regardless of whether the particle is located at the tip (a) or at the base (b). If during the growth the CNF starts to bend due to spatial fluctuations in carbon precipitation at the particle/nanofiber interface, nonuniform stresses are created at the particle/nanofiber interface. For the nanoparticles at the tip (c) and at the base (d) the stresses are distributed in the opposite way, which leads to the nanofiber alignment in the first (c) but not in the second (d) case. White ellipses indicate the interface regions where the stresses occur [From Ref. 88].

When the temperature is high (> 1600 oC), such as in areas where the arc plasma reaches, the metal sulfide core materials released from the carbon shells and subsequently the carbon nanofibers will be annealed to become carbon nanotubes with large cavities, the main component of the deposit on the bottom of the bowl-like cathode. The intensive gas flow inside the cathode bowl and the gravity of the filled carbon nanofibers could be the force to bring the nanofibers into the plasma environment for the annealing process. On the other hand, in the plasma region, the catalyst seeds of metal sulfides could also lead directly to the growth of large cavity carbon nanotubes. The growth model is similar to that of metal catalyzed growth of carbon nanotubes and core/shell carbon nanofibers are no longer the intermediates of the carbon nanotubes.

18

Sk. F. Ahmed and K. K. Chattopadhyay

Figure 2.7. TEM micrographs of vertically aligned carbon nanofiber [From ref. 106].

Figure 2.8. Schematic growth mechanism proposed for the formation of the core/shell carbon nanofibers and carbon nanotubes [From ref 107].

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 19 Carbon nanofibers have been synthesized by the thermal decomposition of acetylene with a copper nanocatalyst derived from cupric nitrate trihydrate at a low temperature of 260 oC [108]. Figure 2.9 shows the typical TEM image of as-prepared regularly helical nanofibers with a symmetric growth mode. The copper nanoparticles changed from initial irregular shapes to regular shapes during the growth of nanofibers.

Figure 2.9. TEM image of typical helical carbon nanofibers [From ref. 108].

Figure 2.10 shows the schematic diagram of growth mechanism of helical nanofibers. The mechanism of morphological changes of copper catalysts and the growth of carbon nanofibers are proposed in five steps: (1) the dehydration of cupric nitrate trihydrate, (2) the decomposition of cupric nitrate into Cu oxide, (3) the reduction of Cu oxide, (4) the formation of Cu nanoparticle before the growth of carbon nanofibers, and (5) the reaction with acetylene for the growth of carbon nanofibers.

Figure 2.10. The schematic diagram of growth mechanism of helical carbon nanofibers [From ref. 108].

20

Sk. F. Ahmed and K. K. Chattopadhyay

Hansen et al. reported that copper nanocrystals would undergo dynamic reversible shape changes in response to changes in the gaseous environment, and the shape changes were caused both by adsorbate-induced changes in surface energies and by changes in the interfacial energy [109]. The gas adsorption on the surface of Cu nanoparticles and the surface energy of different crystallographic planes of a single crystal were the main driving force for the gas-induced surface reconstruction and reshaping of the Cu nanoparticles. In a small metal particle, surface energies associated with different crystallographic planes are usually different. The catalyst particles undergo surface reconstruction to form geometrical shapes, which were able to promote the formation of carbon nanofibers with certain growth conditions of catalysts, gas, and temperature [110,111]. The shape changes of copper nanoparticles were induced by the adsorption of gases on the surfaces of particles. During the reaction, the active sites of the copper nanoparticles were changed from one place to another by following the surface reconstruction. The copper nanoparticle size has a considerable effect on the morphology of carbon nanofibers. The helical carbon nanofibers with a symmetric growth mode were grown on copper catalyst nanoparticles with a grain size less than 50 nm. When the catalyst particle size was around 50-200 nm, straight carbon nanofibers were obtained dominantly. It is reasonable to assume that it is possible to control the diameter of carbon nanofibers by controlling the size of the catalyst particle. A novel method for the direct growth of a single carbon nanofiber (CNF) onto the tip of a commercially available scanning probe microscope (SPM) using Ar+ ion irradiation was reported by Tanemura et al. [112]. This method was proposed on the basis of the experimental fact that the Ar+ ion bombardment of carbon coated substrates induced the formation of conical protrusions that possessed a single CNF at their tip. Commercially available Si SPM tips were coated with carbon and then were Ar+ ion bombarded at room temperature and at 200 oC. Figure 2.11(a) and (b) shows SEM image of a commercially available Si SPM chip of the tip region before and after sputtering respectively. The CNF thus grown was ~30 nm in diameter and 1.5 μm in length. The length was controlled between 0.5 and 1.5 μm by varying the sputtering duration.

Figure 2.11. SEM image of Si SPM chip at tip region (a) before and (b) after sputtering [From ref. 112].

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 21 The formation of ion-induced CNFs explained in terms of erosion and/or growth processes during sputtering. In a case where CNFs are formed by the erosion process alone, so-called ‘‘seed’’ materials, which differ from the surface-constituent materials, are necessary for the CNF formation. These are known as “left-standing” protrusions [113]. The seed materials act as a protection against sputtering during the continuous erosion of the surrounding surface, thus yielding the protrusions tipped with the seed materials [Figure 2.12(b)]. The protrusions thus formed must be linear in shape and possess no conical base. TEM observation confirmed that there was no seed material on the CNF top and possessing CNF-tipped cone structure, which disagree with those of projections formed by the erosion process alone. This clearly suggests that the diffusion process plays a dominant role in the formation of CNFs. Based on the TEM observations, they proposed the following growth mechanism of ion-induced CNFs: (i) Formation of conical protrusions triggered by surface defects such as grain boundaries and small amounts of impurities, (ii) deposition of carbon atoms sputter-ejected from the surface onto the sidewall of the conical protrusions, and (iii) surface diffusion of the deposited carbon atoms toward the tips during sputtering. Since the diffusion is the thermal process, sputtering at elevated temperatures must enhance the diffusion of deposited carbon atoms, thus yielding the longer CNF on the tip of the conical protrusions. Van Vechten et al. demonstrated that carbon atoms readily migrated as far as ~20 mm on the surface during sputtering [114]. In addition to the thermal diffusion, the ionbombardment-enhanced diffusion, which is widely known to occur during sputtering, [115] is likely responsible for the CNF growth. One of the most successful approaches to obtain oriented arrays of nanotubes uses a nano channel alumina template for catalyst patterning [116]. First, aluminum is anodized on a substrate such as Si or quartz, which provides ordered vertical pores. Anodizing conditions are varied to tailor the pore diameter, height and spacing between pores. This is followed by electrochemical deposition of a cobalt catalyst at the bottom of the pores. The use of a template not only provides uniformity but also vertically oriented nanotubes.

Figure 2.12. Schematic representation of the possible formation mechanism of surface projections. (a) left-standing model based on ion etching process alone and (b) a growth model based on the diffusion processes. [From ref. 112].

22

Sk. F. Ahmed and K. K. Chattopadhyay

3. BASIC THEORY OF ELECTRON FIELD EMISSION Field emission (FE) is based on the physical phenomenon of quantum tunneling, in which electrons are injected from the surface of materials into vacuum under the influence of a strong external electric field [117]. The potential barrier is rectangular when no electric field is present, and becomes triangular when a negative potential is applied to the solid. The slope of the latter depends on the amplitude of the local electric field E just above the surface. This local electric field is drastically enhanced if the structure of the emitter is very sharp and protruding (high aspect ratio) as in the case of a CNT. Compared to thermionic electron emission and photo electron emission where electrons have sufficient energy to overcome the potential barrier (work function ϕ ), field emitted electrons tunnel into the vacuum because of a strongly deformed barrier under an electrical field (shown in Figure 3.1).

Figure 3.1. Potential-energy diagram for electrons at a metal surface under an applied electric field, where a strongly deformed potential results from the combination of the applied field and the image charges induced by the emitted electrons.

In the presence of high electric field, flat thin films of some materials emit electrons at macroscopic fields of about 1 to 10 V/μm, although cold field emission of electrons normally occurs only at fields of about 2 V/μm or above. I

This occurs because the thin film is an electrically nanostructured heterogeneous (ENH) material. II Internal nanostructure creates geometrical field enhancement at or near the film or vacuum interface, so local fields are much greater than macroscopic fields. III Electron emission at low macroscopic field strengths is a property of all ENH materials under appropriate conditions. IV Field enhancement is the primary effect, but there also be the secondary effects that contribute to facilitating emission at low fields.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 23 A simple mathematical expression for field emission can be obtained from the Heisenberg uncertainty principle: ∆p. ∆x ≅ ħ/2

(3.1)

where ∆p is the uncertainty of the electron momentum and ∆x is the corresponding uncertainty in position. Considering electrons near the Fermi level, the barrier height is the work function φ, so the uncertainty of the electron momentum ∆p = (2m φ) 1/2 and ∆x can be obtained from equation. (3.1) ∆x = h / 2 (2m φ) 1/2

(3.2)

Thus when ∆x is of the order of the barrier width, the probability that electrons would penetrate the barrier is high. The barrier width is x=φ/E

(3.3)

where E is the electrical field. Combining equations 3.2 and 3.3, we can obtain an estimate of the electrical field required for electron field emission:

E = 2( 2 m / h 2 )

1 2

ϕ2 3

(3.4)

e

A more detailed mathematical description has been done with the Wentzel- KramersBrillouin (WKB) method. The emission current density (J) can be obtained as a function of the electrical field, E, and work function, φ, which is described by Fowler-Nordheim (F-N) equation [118]: 3 ⎡ 2 ϕ AE 2 ⎢− Bα ( y ) exp J = ⎢ E ϕ .t 2 ( y ) ⎢⎣

⎤ ⎥ A / cm 2 ⎥ ⎥⎦

(3.5)

where A and B are constants with values of 1.54 x 10-6 and 6.87 x 107, respectively, and 1/ 2 α ( y ) = 0.95 − y 2 , where y = 3 .8 x10 − 4 E , Both t 2 ( y ) and α ( y ) are contributions

φ arising from the image potential, which is due to positive image charges induced by the emitted electrons at the surface. Image charges can cause a further lowering of the barrier 2 2 height by a factor of − e (Figure 3.1). Under typical conditions, t ( y ) and

4x

α ( y ) are close

to unity and normally omitted in practice. The standard physical assumptions of F-N theory are that the metal: (i) has a free-electron band structure; (ii) has electrons that are in thermodynamic equilibrium and obey Fermi-Dirac statistics; (iii) is at zero temperature; (iv) has a smooth flat surface; and (v) has a local work function that is uniform across the emitting

24

Sk. F. Ahmed and K. K. Chattopadhyay

surface and is independent of external field. It is also assumed that: (vi) there is a uniform electric field above the emitting surface; (vii) the exchange-and-correlation interaction between the emitted electron and the surface can be represented by a classical image potential; and (viii) barrier penetration coefficients may be evaluated using the JWKB approximation. In addition, the F-N equation is based on the assumption of a flat surface and only valid at 0 K, therefore, a modified F-N equation has to be used for an irregular surface and at different temperatures. In order to initiate field emission, an extremely large electrical field has to be employed, which is difficult to achieve from flat surface. However, with a tip structure, a high local electrical field can be obtained around high curvature regions. For instance, an electrical field E at a surface of a sphere, with a radius r and a potential V, is E = V/r. When r becomes smaller, E will become larger. Therefore, to account for the geometrical effects on the local electric field, a field enhancement factor β is introduced in the F-N equation as follows:

J = 1.54 x10 −6

3 ⎤ ⎡ 2 (βE ) 7 φ ⎥ ⎢ exp − 6.87 x10 A / cm 2 ⎥ ⎢ ϕ βE ⎥⎦ ⎢⎣ 2

(3.6)

An experimental F-N plot is modeled by the tangent, which can be written in the form [119-121]

ln(

I S )=a− 2 V V

(3.7)

where a is a constant and

S =−

6 . 83 x10 7 ϕ 2 d 3

β

(3.8)

A linear relationship can be obtained when plotting ln(I/V2) vs I / V, and if we know the work function (φ) of the material, specially for carbon nanotubes, we will assume a work function of 5 eV, [122,123] the inter-electrode distance (d), and the slope (S) of the F-N plot, the field enhancement factor β can be obtained. In geometrical configurations resembling a parallel plate capacitor, the macroscopic field EM is defined by: EM = V/d,

(3.9)

where V is the voltage applied across a gap of thickness d. The local field E is the field, close to the emitting surface (within 1-2 nm of the surface atoms), that determines the barrier through which field emitted electrons tunnel. The field E is some times called the barrier field. E is typically a few V/nm, and is often significantly higher than EM. Their ratio defines a field enhancement factor β

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 25

β = E / EM

(3.10)

Considering the simple physical models of a ‘floating sphere at the emitter plane potential’ and a ‘hemisphere on a post’ the corresponding mathematical expressions have the form:

β=m+h/r

(3.11)

where r is the radius of the sphere or the hemisphere, h is its height above the emitter plane, and m is a constant generally taken as 0, 2 or 3 [124].

4. FIELD EMISSION FROM CARBON BASED MATERIALS Physically the length of CNT is several mm and diameters down to 10 Å, so the nanotubes exhibit a very high geometric aspect ratio (h/r), and has been shown that the applied electric field is concentrated precisely at the nanotube tips [125,126], which results in a large field enhancement factor β, typically ~10,000. In addition to the geometric field enhancement factor shown in equation (3.6), the effect of adsorbates i.e., forming a tunneling state at the surface [127] the tube cap and tip structures such as open or close [128,129] and the surface morphologies and contaminations i.e. catalyst particles, amorphous carbon [130] have been proposed as affecting their extraordinary field emission behavior. Carbon nanotubes have the right combination of properties such as nanometer size diameter, structural integrity, high electrical conductivity, and chemical stability that make good electron emitters. Electron field emission from carbon nanotubes was first demonstrated by Rinzler et al. in 1995 [131], and has since been studied intensively on various carbon nanotube materials. Field emission properties of different types of carbon nanotubes have been reported, including individual nanotubes [131-133], MWCNTs embedded in epoxy matrices [134], MWCNT films [135], SWCNTs [136], aligned MWCNT films [65,137] and hollow carbon nanotubes [132]. Random and aligned MWCNTs were found to have threshold fields slightly larger than that of the SWCNT films and are typically in the range of 3-5 V/µm for a 10 mA/cm2 current density [65]. These values for the threshold field are all significantly better than those from conventional field emitters such as the Mo and Si tips which have a threshold electric field of 50-100 V/µm. It is interesting to note that the aligned MWCNT films do not perform better than the random films. This is due to the electrical screening effect arising from closely packed nanotubes [138]. The case of emission from the large variety of carbon-based materials suggests that the NEA is not a prerequisite, and more general emission models are desired for these systems. Many models have been proposed to discuss the origin of electron emission from ta-C films and other amorphous carbon (a-C) films. Robertson suggested that field emission at a low electric field is due to the low electron affinity of ta-C films [139]. Other models such as space-charge-induced band bending [140], surface dipole-controlled emission [139], field enhancement due to the film microstructure [141] and conductive paths caused by localized sp2 sites were also proposed [142]. Thermal annealing is an effective method to alter structure of DLC films. Ilie et al. [143] and Carey et al. [144] proposed that the presence of sp2 clusters

26

Sk. F. Ahmed and K. K. Chattopadhyay

within the insulating sp3 matrix could give rise to field enhancement in amorphous carbon (aC) films containing large defect densities (>1019 cm−3). It was proposed that the presence of such dielectric inhomogeneity is responsible for field enhancement in these films. Since sp2 clusters will have different dielectric constants, the application of the external field will result in local field enhancements around the clusters and will aid in the emission of electrons. Groning et al. [142] explained the emission mechanism from DLC films in a way that, like a freestanding conductive tip in the vacuum, sp2 bonded carbon clusters are assumed to form a conductive channel in an insulating matrix, which leads to local field enhancement and hence to an enhanced electron emission. Since the sp2 clusters are located at or near the Fermi level and high concentration sp2 carbon clusters in the films play a more important role in determining the electron emission property of the films. The emission can depend on various parameters such as negative electron affinity [145], band gap [146], surface termination [145,147], depletion layers [148] and film thickness [149]. In these cases, the emission can be interpreted in terms of homogeneous films and a well-defined band structure [150]. In amorphous carbon, the sp3 content controls the band gap and electron affinity. Nanostructured carbon, nanocrystalline diamond, and carbon nanotubes are the types of carbon that emit at lowest applied field. In microcrystalline diamond, emission is found to occur from grain boundaries [151,152], that is, nm-scale sp2-bonded regions of positive electron affinity. Similarly, emission from carbon nanotubes [153] occurs from 1 nm curved regions. CNFs films were synthesized by plasma enhanced chemical vapor deposition about 100 nm in diameter and about 10 μm in length using P doped n-type Si (100) wafers and indium tin oxide (ITO) coated glasses as substrates [154]. Figure 4.1(a) shows the SEM image of the CNF film on Si substrate and (b) shows the TEM image of a tip of a CNF where as (c) shows a schematic illustration of the CNF structure. The CNFs ranged 50-100 nm in diameter and over 10 μm in length, which were randomly oriented to the substrate showed electron field emission characteristic (as shown in Figure 4.2(d)). Since the nanofibers were grown to random orientation, electrons can be emitted with any directions from the protrusions on the fiber. The threshold electric field (Eth) was estimated 2.4 V/μm, which is comparable with the threshold field of several carbon nanostructures such as, Eth of single wall carbon nanotube (SWCNT) by arc discharge about 2.2 V/μm, multi wall carbon nanotube (MWCNT) by microwave plasma CVD about 1.8 V/μm and nanostructured carbon film about 3.0 V/μm respectively [155-157]. The reason of the excellent field emission characteristics is due to the CNF film has many protrusions which are 10 nm in width and 30 nm in length, as shown in Figure 4.1(b). Since the nanofibers were grown to random orientation, electrons can be emitted with any directions from the protrusions on the fiber. Aligned carbon nanofibers and hollow carbon nanofibers were grown by micro wave ECR-CVD method using methane and argon mixture gas at a temperature of 550 oC showed good electron field emission (Figure 4.2) [158]. The aligned carbon nanofibers give a high current density 7.25 mA/cm2 at 12.5 V/μm in comparison with the value 0.69 mA/cm2 at 12.5 V/μm of the hollow carbon nanofibers.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 27

Figure 4.1.(a) SEM image of the CNF film on Si substrate, (b) TEM image of a tip of a CNF, (c) schematic illustration of the CNF structure and (d) J-E curve and in inset the corresponding F-N plot [From ref. 154].

28

Sk. F. Ahmed and K. K. Chattopadhyay

Figure 4.2. (a) I-V graph of the carbon nanofiber (b) and the hollow carbon nanofiber [From ref. 158].

Effect of catalysts on the field emission property was studied by Kamada et al. [78]. Carbon nanofiber films on Pd-Se, Fe-Ni, and Ni-Cu alloy catalysts have been synthesized at low temperatures by a thermal CVD technique. Figure 4.3(a) shows the field electron emission characteristics of the CNF films grown at 600 oC. The threshold electric fields of the CNFPd-Se, CNFNi-Cu, and CNFFe-Ni films are estimated to be 1.1, 2.8, and 3.8 V/μm, respectively. The CNFs grown using Pd-Se catalyst were found to have more defective structure than that obtained with the other catalysts, and exhibited best field emission property. It is likely that defects play a role as electron emission sites. Figure 4.3(b) shows the threshold electric field obtained for the CNFPd-Se, CNFFe-Ni, and CNFNi-Cu films as a function of growth temperature. The threshold electric field was not strongly dependent on the catalyst type and decreased with growth temperature. Ilie et al. [159] reported that the surface electronic properties introduced by defects could provide a local field enhancement to facilitate the field emission. From this, it is suggested that the excellent field emission property obtained for CNFPd-Se originates from numerous defects in the body of the CNFs. In this sense, good crystallinity of the CNFs is not required to obtain good field emission characteristics. Carbon nanofibers (CNFs) were grown on a Ni-P alloy catalyst deposited on a silicon substrate via MWCVD technique with methane gas [82]. The CNFs grown on the Ni-P alloy catalyst showed random orientation and it composed of parallel graphite planes with defects tilted from their axis. Figure 4.4 shows the electron emission current density versus electric field (J-E) curves of CNFs with different thickness of the catalyst.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 29

Figure 4.3. (a) Emission current density as a function of applied electric field and (b) threshold electric field for the CNFPd-Se, CNFFe-Ni, and CNFNi-Cu films as a function of growth temperature [From ref. 78].

For CNF grown on catalyst with thickness 20 nm, the turn-on field was approximately 0.11 V/μm with an emission current density of 10 mA/cm2 and the threshold field was 3.1 V/μm with an emission current density of 10 mA/cm2. CNF grown on catalyst thickness 30 and 40 nm, have almost the same turn-on field approximately 0.22 V/μm, but the threshold field is 3.4 and 4.1 V/μm, respectively.

30

Sk. F. Ahmed and K. K. Chattopadhyay

Figure 4.4. The current density vs. electric field curves for CNFs deposited with different thickness of catalyst and in Inset corresponding F-N plot [From ref 82].

The excellent field emission properties of Ni-P alloy catalyzed-CNFs may be attributed to the random and defects of CNFs. Davydov et al. [160] have also pointed out that perfectly aligned CNTs were less efficient field emitters and had lower field enhancement than chaotic CNTs. Some reports relating defect densities to field emission properties have also been proposed [161,162]. The enhanced emission may originate from the defect-induced energy bands that are formed within the band gap of graphite. The energy barrier that the electrons must tunnel through to be emitted is reduced, so the electrons residing at these defect levels can be emitted directly into vacuum from these bands or be transported to the surface states for emission [163]. Obraztsov et al. [164] have also found that the field emission properties were improved by increasing the density of structural defects. Figure 4.4 also indicates that the field emission properties of CNFs with small diameter are better than those of the CNFs with large diameter. A clear fluctuation of the I-V curve at higher voltages for CNFs is seen in Figure 4.4, indicating some emission sites are damaged or destroyed. It is reasonable to suggest that the CNFs with small diameter are more easily damaged by ion bombardment than the CNFs with large diameter [165]. The CNF is synthesized on the iron-evaporated Si substrate by microwave plasma chemical vapor deposition and nitrogen (N2) plasma treatment is carried out to modify the CNF surface [166]. A reduction in the turn-on electric field is achieved by N2 plasma treatment for the CNF and the stability of the electron emission current is also improved by nitridation of the CNF surface. In Figure 4.5 the field emission characteristics are compared between as grown and N2 plasma-treated samples. Potential barrier height is reduced by nitridation of the CNF surface. The excellent field emission of carbon nanofibers and nanotubes have stimulated their applications as electron sources in x-ray applications [137], and field emitters in electron microscopes [132], with their environmental sensitivities of electrical conduction.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 31

Figure 4.5. Field emission characteristics (a) untreated and (b) N2 plasma treated CNFs [From ref. 166].

5. SYNTHESIS AND FIELD EMISSION PROPERTY OF DIFFERENT CARBON NANOSTRUCTURE 5.1. Synthesis and Field Emission Property of Carbon Fibrous Films Among the different techniques for the production of carbon nanotubes, plasma enhanced chemical vapor deposition is a high yield and controllable method for the production carbon nanotubes/nanofibers [167] with mass production. Thin film catalyst layers have been successfully employed in the carbon nanofiber growth. The use of metal catalysts such as Ni, Fe, Co, Pt and some of their alloys has been explored in an effort to control the size and morphology of carbon nanostructures formed through the decomposition of hydrocarbons. We have used Ni as a catalyst for the formation of carbon fibers, because carbon nanostructure formed on Ni are more crystalline than those formed with other catalysts [168,169]. The target used for sputtering was a Ni plate of thickness ~1 mm with a diameter 2.5 cm (with purity 99.99 %, Aldrich). The Ni target has been sputtered on Si substrates via dc sputtering technique to produce thin film of Ni catalyst. The substrates were 10×10 mm2 cleaned Si (400) wafer. The Si substrates were etched in HF (~20%) for 5 minutes to remove the surface oxide layer and finally cleaned in an ultrasonic cleaner. The sputtering was done at a pressure 0.2 mbar sending argon as a sputtering gas with an inter-electrode distance 1.6 cm at room temperature for deposition time 5 minutes, which yielded a Ni film with a thickness ~10 nm, as measured by quartz crystal thickness monitor. For sputtering, we

32

Sk. F. Ahmed and K. K. Chattopadhyay

maintained high voltage 2.5 KV and corresponding current density was 19.5 mA cm-2. Table 5.1 shows the deposition parameters for Ni catalyst thin films. After deposition of the catalyzed film, the sample was immediately transferred in to the CVD chamber where nanofibers growth has been performed. Deposition chamber used for the synthesis of carbon fibers was made of stainless steel (SS). The plasma was produced between two parallel plates SS electrodes as usual. The lower plate was grounded and the upper plate was used as the cathode. The schematic diagram of the DC-PECVD unit is shown in Figure 5.1. The deposition chamber was initially evacuated by a standard rotary and a diffusion pump arrangement up to a base pressure of 10−6 mbar. The substrate (Si) was placed on a molybdenum substrate holder, which could be directly heated. When the chamber pressure attained 10−6 mbar, the Mo substrate holder was started to heat by sending current through it. The substrate temperature could be varied by varying the current through the Mo substrate holder, which was connected to the secondary of a step down transformer. The temperature of the substrate was measured by a disappearing filament type pyrometer (PYROPTO, IT65). Acetylene (C2H2) gas was used in PECVD process as a precursor of carbon. Acetylene (C2H2) gas was allowed to flow maintaining the CVD chamber pressure 50 mbar. Deposition was performed at 2.0 kV DC supply with corresponding current density 25 mA cm-2 for 30 min. Substrate temperature was varied from 700 to 850 oC for different set of experiment. The deposition parameters for the synthesis of carbon fibrous thin films via PECVD have been shown in Table - 5.2.

Figure 5.1. Schematic diagram of the DC -PECVD unit.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 33 Table 5.1. Deposition parameters used for dc-sputtered deposited Ni catalyst Deposition parameters Deposition time dc voltage Electrode distance Sputtering gas Gas pressure Substrates used Substrate temperature

Corresponding values 5 min. 2.5 kV 1.6 cm Argon (Ar) 0.2 mbar Si (400) 300 oK

Table - 5.2. Deposition parameters used for the synthesis of carbon fibrous thin films Deposition parameters 1. Deposition time 2. dc voltage 3. Electrode distance 4. Precursor material Gas pressure Substrates used Substrate temperature

Corresponding values 30 min. 2.0 kV 1.4 cm Acetylene (C2H2) 50 mbar Ni catalyzed Si (400) 700 - 850 oC

Figure 5.2 showed the SEM micrographs of the deposited films, which showed the existence of carbon fibers in the films. The morphologies of the films have been changed with the change of substrate temperature. At 700 oC substrate temperature, only particles have been found but at 750 oC, some carbon nanofibers have also been grown. At 800 oC substrate temperature, carbon fibers have been grown with length ~ 1000 nm and the corresponding diameter ~ 400 nm. Finally at 850 oC substrate temperature, the best quality carbon fibers have been grown with length ~ 2000 nm with corresponding diameter ~ 400 nm. It is clear from these studies of substrate temperature variations that at lower substrate temperature only particles are grown and at higher substrate temperature the morphology changes from particles to nanotubes or fibers like structure i.e. quasi one dimensional growth takes place at higher substrate temperature. The electron field emission properties of the CNFs deposited on Si substrates have been studied by our high vacuum (~10-7 mbar) field emission setup as shown in Figure 5.3. Field emission measurements were carried out by using a diode configuration consisting of a cathode (the film under test) and a stainless steel tip anode mounted in a liquid nitrogen trapped rotary-diffusion vacuum chamber with appropriate chamber baking arrangement. The measurements were performed at a base pressure of ~5 x 10-7 mbar and at different temperature, which was controlled with a controller and measured with a thermocouple. The tip-sample distance was continuously adjustable to a few hundred μm by spherometric arrangement with screw-pitch of 10 μm. The anode-sample spacing was set at a particular value by rotating the micrometer screw which served as an anode electrode. Field emission current-voltage measurements were done with the help of an Agilent multimeter (model 3440-1A). Emission characteristics were registered and analyzed with the help of a personal computer.

34

Sk. F. Ahmed and K. K. Chattopadhyay

Figure 5.2. SEM image for different substrate temperature (a) 750 oC, oval shaped particles, (b) 800 oC, fiber like and (c) 850 oC; fibers [From ref. 168].

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 35

Figure 5.3. Schematic diagram of the field-emission apparatus [From ref. 170].

We have used the simplified F-N equation for field emission analysis. Figure 5.4(a) shows the emission current (I) vs. macroscopic field (E) curves of carbon nanofibers thin film deposited on Si substrate for anode-sample separation (d) of 60 μm. The macroscopic field is calculated from the external voltage applied (V), divided by the anode-sample spacing (d). Theoretically, the emission current I is related to the macroscopic electric field E by

− b vF φ I = A a t φ ( β E ) exp{ βE −2 F

−1

2

3

2

}

(5.1)

where, φ is the local work-function, β is the field enhancement factor, A is the effective emission area, a is the first Fowler-Nordheim Constant (1.541434 x 10-6 A eV V-2), b is the second F-N Constant (6.830890 x 109 eV-3/2 V m-1), and vF and tF are the values of the special field emission elliptic functions [119] v and t, evaluated for a barrier height φ. In so-called Fowler-Nordheim coordinates, this equation takes the form:

(v bφ 2 β −1 ) I ln{ 2 } = ln{t F− 2 A a φ −1 β 2 } − F E E 3

(5.2)

36

Sk. F. Ahmed and K. K. Chattopadhyay

An experimental F-N plot is modeled by the tangent, which can be written in the form [119-121]:

I ( s bφ 2 β −1 ) ln{ 2 } = ln{rA a φ −1 β 2 } − E E 3

(5.3)

where r and s are appropriate values of the intercept and slope correction factors, respectively. Typically, s is of the order of unity, but r may be of order 100 or greater. Both r and s are relatively slowly varying functions of 1/E, so a F-N plot (plotted as a function of 1/E) is expected to be a good straight line. The F-N plot of our sample is shown in Figure 5.4(b). It has been observed that the I-E curve in the present work is closely fitted with straight line. This suggests that the electrons are emitted by cold field emission process. The turn-on field, which we define as the macroscopic field needed to get an emission current I = 0.09 μA, (which corresponds to an estimated macroscopic current density, Jest = 14.5 μA/cm2, where Jest = I/A, A = anode-tip area) were lying in the range 2.57 to 9.71 V/μm for variation substrate temperature films. This value is quite lower than that of nanocrystalline carbon 6.4 V/μm [171] and carbon nanofiber arrays (~3 V/μm) reported by Cao et al. [172]. According to the F-N plot (Figure 5.4(b)), the slope m (given by equation (5.4)) would represent the combined effect of work function and enhancement of local electric field and is given by,

m= −

bφ

β

3

2

(5.4)

The effective work function relation [173]

φE =φ / β

2

3

φ E is related with the true work function φ through the

. Using φ = 5 eV is the work function of CNF [174]; the field

enhancement factor was calculated from the slope of the F-N plot, lies in the range 8090 to 1945 and the corresponding effective work function φC lies in the range 0.0124 to 0.0321 eV for films deposited with different substrate temperature, which is comparable with CNT films [173]. The plots of I-E graph for different electrode distance (d) are shown in Figure 5.5(a) and (b) the corresponding F-N plot. The turn-on field was found to vary in the range 6.87 to 2.87 V/μm for a variation of anode sample spacing 80 - 120 μm for the carbon fibrous film deposited at 850 oC. In the I-E graph (Figure 5.5(a)), we observed a parallel shift of curves with respect to anode-sample separation (d) i.e., for a particular electric field the current density increases with increasing the anode-sample separation. Zhou et al. [175] reported similar type of observation for their β-SiC nanorods. We suppose that this type of shift observed in our sample may be due to the change in the effective emission area of the sample for different anode-sample separation. The change of effective emission area with respect to d may be related to the geometry of the anode.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 37

Figure 5.4. Emission current vs. macroscopic field curves of carbon fibrous film deposited for different substrate temperature (a) and corresponding. F-N plot (b) [From ref. 168].

In our experiment we have used a conical shape anode with tip diameter 1 mm, therefore the lines of force immerging from the edge of the anode tip and terminating to the sample surface are diverging in nature, whereas the lines of force immerging from the flat surface of the tip are parallel in nature. Hence, the effective emission area of the sample increases with increasing d as shown in Figure 5.6. Au et al. [176] performed field emission of silicon nanowires using a spherical-shaped stainless steel probe with a tip diameter of 1 mm as an anode. They also found a parallel shift in their I-V curve.

38

Sk. F. Ahmed and K. K. Chattopadhyay

Figure 5.5. I-E curves carbon fibrous film for different anode-sample separation (a) and (b) corresponding F-N plot (b) [From ref. 168].

Okano et al. [177] reported that their macroscopic current density for diamond films was independent of the anode-sample separation. Their field emission apparatus consisted of a parallel plate arrangement of the anode and sample, separated by spacers. So the electric lines of force between the anode and the sample were parallel in nature, hence effective emission area remained independent of the anode-sample spacing.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 39

Figure 5.6. Schematic diagram for the dependence of effective emission area as an increasing function of anode-sample separation.

5.2. Synthesis and Field Emission Property of Vertically Aligned CNFS Vertically aligned carbon nanofibers (VACNFs) have been synthesized by direct current PECVD technique and for the synthesis of VACNFs, Ni catalyst was deposited in thin film form on Si substrates via RF magnetron sputtering technique. For sputtering, we maintained RF power 200 watt and corresponding chamber pressure 0.1 mbar. We have synthesized Ni thin film having different thickness varying from 10-20 nm, as measured by quartz crystal thickness monitor (HindHivac, Digital thickness monitor, Model: DTM-101). The deposition parameters for Ni catalyst have been shown in Table- 5.3. After deposition of the Ni film, the substrates were immediately transferred into the CVD chamber where nanotube growth has been performed. Acetylene (C2H2) gas was used in PECVD process as a precursor of carbon and during deposition the chamber pressure was maintained at 30 mbar. Deposition was performed at 2.2 kV DC with corresponding current density 21.5 mA cm-2 for 25 min. Table – 5.3. Deposition parameters used for rf-magnetron sputtered deposited Ni catalyst Deposition parameters RF-power Electrode distance Sputtering gas Gas pressure Substrates used Substrate temperature Deposition time

Corresponding values 200 Watt 3 cm Argon (Ar) 0.2 mbar Si (400) 300 oK 4 -10 min.

40

Sk. F. Ahmed and K. K. Chattopadhyay

An atomic force microscope (AFM-NT-MDT, Solver Pro) was used to analyze the surface topography of the grown CNFs films. Figure 5.7 shows the AFM images of the carbon nanofibrous films deposited on Ni catalyst having different thicknesses. From the figure it is clear that the diameter of the CNFs increase and length decreases with the increase of Ni catalyst film thickness. The diameter and lengths of the CNFs deposited on thinner catalyst (Ni film thickness 10 nm) are ~ 150 nm and 2.5 μm whereas the diameter and length of the CNFs deposited on thicker catalyst (Ni film thickness 20 nm) are ~ 250 nm and 1.0 μm. The morphology of the catalyst film is known to play a critical role in CNF growth. So, the thickness of the Ni catalyst film will affect the growth and the properties of CNFs. The diameter of the CNFs decreased as the thickness of the catalyst film decreased.

Figure 5.7. AFM 3D pictures of VACNFs deposited on different thickness of Ni catalyst (a) for 10 nm, (b) for 15 nm and (c) for 20 nm [From ref. 178].

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 41 Figure 5.8(a) shows the emission current density (J) vs. macroscopic field (E) curves for the VACNF films deposited at a fixed anode-sample separation (d) of 120 μm. Field emission characteristics of the films were analyzed using the help of simplified Fowler-Nordheim (FN) theory [119-121]. The F-N plots of our sample are shown in Figure 5.8(b).

Figure 5.8. J-E graph of VACNFs for different aspect ratios (a) and corresponding F-N plot (b) [From ref. 178].

42

Sk. F. Ahmed and K. K. Chattopadhyay

It has been observed that all the J-E curves in the present work are satisfactorily fitted with a straight line, which suggests that the electrons are emitted by field emission process. The threshold field, which we define as the macroscopic field needed to get an emission current density J = 10 μA/cm2, were lying in the range 4.3 to 5.4 V/μm for VACNFs deposited on Ni catalyst having different thicknesses. As our deposited CNFs have different aspect ratios so the local field enhancement occurs and changes in the threshold field was observed for different type of VACNFs. The field enhancement factor (β) as well as emission current density is strongly dependent on the aspect ratio of canon nanofibers (shown in Figure 5.9). Tsai et al. [82] showed that the CNFs with small diameter and many defects exhibited excellent field emission properties than the CNFs with larger diameter.

Figure 5.9. The variation of threshold field and emission current density with aspect ratio carbon nanofiber [From ref. 178].

5.3. Synthesis and Field Emission Property of Multiwalled Carbon Nanotubes For the deposition of Ni catalyst in thin film form by dc sputtering process we have used a Ni plate of thickness ~1 mm with a diameter 2.5 cm (with purity 99.99 %, Aldrich). The substrates were 10×10 mm2 cleaned Si (400) wafer. The Si substrates were etched in HF (~20%) for 5 minutes to remove the surface oxide layer and finally cleaned in an ultrasonic cleaner. The sputtering was done at a pressure 0.1 mbar sending argon as a sputtering gas with an inter-electrode distance 1.6 cm at room temperature for deposition time 5 minutes, which yielded a Ni film with a thickness ~ 12 nm, as measured by quartz crystal thickness monitor. For sputtering, we maintained high voltage 3.0 KV and corresponding current

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 43 density was 27.5 mA cm-2. After deposition of the catalyzed film, the sample was immediately transferred to the CVD chamber where the synthesis of nanotubes was performed. The deposition procedure was described in section 5.1. Table 5.4 shows the deposition parameters for the synthesis of MWCNTs thin films via DC-PECVD technique. Table 5.4. Deposition parameters used for the synthesis of MWCNTs thin films Deposition parameters 1. Deposition time 2. dc voltage 3. Electrode distance 4. Precursor material 5. Gas pressure 6. Substrates used 7. Substrate temperature

Corresponding values 30 min. 2.0 kV 1.4 cm Acetylene (C2H2) 30 mbar Ni catalyzed Si (400) 900 0K

Figure 5.10(a) shows the FESEM micrographs of the deposited films, which showed the existence of carbon nanotubes in the films. The diameter of the carbon nanotubes are ~ 12 nm and few micrometer in length. The transmission electron micrographs of multiwalled carbon nanotubes have been shown in Figure 5.10(b). It could be observed that the carbon nanotubes are multiwalled with diameter ~12 nm.

Figure 5.10. FESEM micrograph (a) and HRTEM lattice image of the MWCNT (b).

44

Sk. F. Ahmed and K. K. Chattopadhyay

Figure 5.11(a) shows the emission current density (J) vs. macroscopic field (E) curves for MWCNT films for different anode-sample separation (d). The F-N plot of our sample is shown in Figure 5.11(b). The threshold fields were found to vary in the range 4.7 - 3.6 V/μm for a variation of anode-sample separation in the range 80 - 150 μm.

Figure 5.11. (a) J-E curves for the MWCNTs for different anode-sample separation (d) and (b) corresponding F-N plots.

In the J-E graph (Figure 5.11(a)), we observed a parallel shift of curves with respect to anode-sample separation (d) i.e., for a particular electric field the current density increases with increasing the anode-sample separation. For example, at a field of 5 V/μm, the J values were found to be 0.2 mA/cm2 (for d = 80 μm), 1.2 mA/cm2 (for d = 110 μm) and 3.7 mA/cm2 (for d = 150 μm). This type of shift observed in our sample is due to the change in the effective emission area of the sample for different anode-sample separation, which is described in section 5.1.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 45

6. EFFECT OF TEMPERATURE ON THE ELECTRON FIELD EMISSION FROM VERTICALLY ALIGNED CARBON NANOFIBERS AND MWCNTS The synthesis of vertically aligned carbon nanofiber (VACNF) thin films has been described in previous section 5.2 and for synthesis of MWCNTs, Ni catalyst deposited in thin film form on Si substrates via RF sputtering technique. The target used for sputtering was a Ni plate of thickness ~1 mm with a diameter 2.5 cm (with purity 99.99 %, Aldrich). For sputtering, we maintained RF power 180 watt and corresponding chamber pressure 0.2 mbar. After deposition of the Ni film, the substrates were immediately transferred into the CVD chamber where nanotube growth has been performed. Acetylene (C2H2) gas was used in PECVD process as a precursor of carbon. C2H2 gas was allowed to flow maintaining the CVD chamber pressure 40 mbar. Deposition was performed at 2.5 kV DC supply with corresponding current density 27.5 mA cm-2 for 30 min at 700 oC substrate temperature. AFM image (Figure 6.1(a)) of the carbon nanofibrous films shows that the vertically aligned CNFs are having an average diameter ~ 250 nm and length 2.0 μm. Figure 6.1(b) shows the FESEM micrographs of the carbon nanotubes, which showed that the diameter of the carbon nanotubes are ~ 20 nm and few micrometer in length. The transmission electron micrographs of multiwalled carbon nanotubes have been shown in inset of Figure 6.1(b). It could be observed that the carbon nanotubes are multiwalled with diameter ~ 20 nm.

Figure 6.1.(a) AFM 3D pictures of vertically aligned carbon nanofiber thin films, (b) FESEM micrograph and in inset HRTEM micrograph of the MWCNT [From ref. 179].

46

Sk. F. Ahmed and K. K. Chattopadhyay

For temperature dependent field emission total current density (J = JE + JT, where JE and JT are the field current and thermionic current density respectively) given by simplified F–N equation and Richardson equation as [180,181]:

J = J E+ J T J =a φ

−1

(6.1)

− sb φ ( β E ) exp( β E 2

3

2

φ ⎡ θ ⎤ 2 − KT + ADT e )⎢ ⎥ ⎣ sin( θ ) ⎦

(6.2)

where A is a constant about 120 A/(cmK)2, D is the average transmission coefficient of emitter surface, T is the temperature in Kelvin, φ is the work function of CNF/CNT, k is the

Boltzmann constant and θ is the temperature correction factor, and is given by

2.2π (kT / q)φ θ≈ 1.959 E

1

2

(6.3)

For CNF/CNT with a work function 5 eV [174,182] and temperature below 1000 K, the value of [θ/(sin(θ))] in eqn. (6.2) is always 1.0 and in our studied temperature range, (< 400 o C), the highest contribution of thermionic emission is much smaller than the field emission current density i.e., the measured emission property is predominated by field emission current because below 1000 K, the thermionic emission effect is less significant than the field emission effect [183]. Hence eqn. (6.2) reduced as

( s bφ 2 β −1 ) J ln{ 2 } = ln{r a φ −1 β 2 } − E E 3

(6.4)

Figure 6.2(a) and 6.3(a) shows the emission current density (J) vs. macroscopic field (E) curves of carbon nanofibers and MWCNTs thin films respectively, for different temperature and corresponding F-N plot is shown in Figure 6.2(b) and 6.3(b). It has been observed that the J-E curve in the present work is closely fitted with straight line. This suggests that the electrons are emitted by cold field emission process. The threshold field, which we define as the macroscopic field needed to get an emission current density J = 10 μA/cm2, were lying in the range 5.1 to 2.6 V/μm for CNFs and lying in the range 4.0 to 1.4 V/μm for MWCNTs for the variation of temperature from 300 K to 650 K.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 47

Figure 6.2. J-E graph of CNFs for different temperature (a) and corresponding F-N plot of CNFs (b) [From ref. 179].

48

Sk. F. Ahmed and K. K. Chattopadhyay

Figure 6.3. J-E graph of MWCNTs for different temperature (a) and corresponding F-N plot of MWCNTs (b) [From ref. 179].

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 49 It is clear from the Figure 6.4(a) and (b) that with the increase of temperature, threshold field decreases and current density increases for CNFs as well as MWCNTs. The emission current density is strongly dependent on the work function and as well as on the aspect ratio. For an example to get emission current density J = 5 mA/cm2, 9.5 V/μm field needed for CNFs but for MWCNTs 4.5 V/μm field needed for the same current density at 650 K ambient temperature. The work function of materials is temperature dependent. Therefore, the decrease of threshold field with the increase of temperature may be due to the decrease of work function of CNF and MWCNT films.

Figure 6.4. The variation of threshold field and emission current density with temperature (a) for CNFs (b) for MWCNTs [From ref. 179].

50

Sk. F. Ahmed and K. K. Chattopadhyay

Figure 6.5 shows the variation of the effective work function with temperature for CNFs and MWCNTs. It shows that the effective work function decreases with increase of temperature. The variation of the effective work function with temperature is consistent with the variation of the emission current density as observed. From the quantum mechanical tunneling phenomena, we know that the Fermi energy determines the field emission current. The work function (φ) is given by φ = EV - EF, where EV is the fixed vacuum level and EF is the Fermi level [184]. For low temperature emission, Fermi level is lower and the electrons have to transmit through a much boarder barrier as shown in Figure 6.6(a) and for high temperature field emission, Fermi level increase so barrier width decrease as shown in Figure 6.6(b). Hence the emission current increases under same field for high temperature field emission (as shown in Figure 6.4).

Figure 6.5. Variation of the effective work function with temperature for CNFs and MWCNTs [From ref. 179].

Figure 6.6. Schematic diagram of the transmission of electrons from the MWCNTs at low (a) and high (b) temperatures under applied field [From ref. 170].

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 51 As the field emission characteristic has a complicated dependence on electric field and temperature when these two factors coexist, and cannot be explained simply by F–N or thermionic emission model, the information of β can still provide fruitful information. According to the F-N plot (Figure 6.2(b) and 6.3(b)), the slope m (given by equation (6.5)) would represent the combined effect of work function and enhancement of local electric field and is given by,

m= −

bφ

β

3

2

(6.5)

The effective work function relation [183]

φC = φ / β

2

3

φC is related with the true work function φ through the

. Using φ = 5 eV is the work function of CNF/CNT, the field

enhancement factor was calculated from the slope of the F-N plot, lies in the range 4589 to 9917 and the corresponding effective work function φC lies in the range 0.018 to 0.011 eV for the CNFs with different ambient temperature. For MWCNTs field enhancement factor lies in the range 5838 to 11448 and the corresponding effective work function φC lies in the range

0.015 to 0.010 eV. The field enhancement factor β increases monotonously with the temperature, which explains very well the increase in emission current density with measuring temperature (shown in Figure 6.4). But physically the field enhancement factor should depend on the geometric shape of the emitter rather than temperature. Hence there may be other factors responsible for such temperature dependence of emission current. Although the exact explanation of the observed temperature dependence of the emission current needs more research, the following effects may have strong influence. The presence of defects or surface states is predominant in nanomaterials like CNT or CNF. Wang et al. and Xu et al. also proposed that the field emission property is related with the defect densities [161,162]. These states might have small activation energies and when the temperature in increased, the carriers trapped in these states are activated into the conduction band and more emission currents results. Chen et al. [185] observed the field emission of different oriented CNTs and they discovered that the CNTs oriented parallel to the substrate have a lower onset applied field than those oriented perpendicular to the substrate. They also suggested that the defect emission mechanism is a reason for the low onset electrical field. Obraztsov et al. [164] have also found that the field emission properties were improved by increasing the density of structural defects. The field enhancement factor also depends on aspect ratio of carbon nanostructure. The aspect ratio (h/r, where r is the average radius and h is the length of the tubes, respectively) of our MWCNTs is greater than that of CNFs. From our experimental result we also see that the field emission properties of CNTs are better than those of the CNFs with large diameter. Another possible reason is that the screening effect, which diminishes the electric field near the CNFs. Nilson et al. proposed from their experimental observation that to overcome screening effect the distance between nanotubes will be 1-2 times the tube height [186]. As our deposited vertically aligned CNFs are compact so screening effects is much remarkable. Since it is well known that MWCNTs are mostly

52

Sk. F. Ahmed and K. K. Chattopadhyay

conductive, their conductivity should decrease with higher temperature and thus causes the screen effect less remarkable.

7. APPLICATION OF CARBON NANOFIBER AND CARBON NANOTUBE Along with the improvement of the production and characterization techniques for nanotubes, progress is being made in their applications. MWCNTs have also exhibited ballistic transport [187]. The reasons for high electron conductance in carbon nanotubes are as follows; i) physical perfection: a smooth surface with no chemical dangling bonds and no edges reducing surface states, which affects backscattering; ii) strong covalent bonds in CNT and iii) no low-energy dislocations or defects, which also reduces backscattering and provide stability for high current transport. Carbon nanofibers/nanotubes have many properties from their unique dimensions to an unusual current conduction mechanism that make them ideal components of electrical circuits. Due to their semi conducting properties, nanofibers/nanotubes may be the building blocks for smaller, faster computers. Other potential applications in electronics and computers include, storage devices, It was proposed to use nanotubes as central elements of electronic devices including field-effect transistors, single-electron transistors [188] and rectifying diodes [189] and for logic circuits [134]. The geometric properties of nanotubes such as the high aspect ratio and small tip radius of curvature, coupled with the extraordinary mechanical strength and chemical stability, make them an ideal candidate for electron field emitters [129,186]. CNT field emitters have several industrial and research applications; flat panel displays [131], outdoor displays, traffic signals and electron microscopy. De Heer et al. [135] demonstrated the earliest high intensity electron gun based on field emission from a film of nanotubes. The properties of carbon nanotubes (CNTs) and the less crystalline carbon nanofibers (CNFs) have attracted considerable interest for both scientific and technological issues [17,190]. Their impressive mechanical properties [191], high current carrying ability [192], and field emission performance [135] have opened the way to a number of applications such as field emission devices [193], interconnects [194], sensors [195], super-capacitors [196], fuel cells [197] and battery electrodes [198]. The vertical geometry of carbon nanofibers (CNFs) is particularly useful in technologies such as nanoelectronics [89], electrodes for biosensing/stimulation [90], nanomechanical [91], and thermal interface materials [92]. Achieving near-ohmic contact at the nanotube-metal interface as well as investigating the affect of nanotube crystallinity is critical for evaluating and modeling the electrical performance of on-chip interconnects. Sim et al. reported that the carbon-nanofiber-based (CNF) ionization gas sensing devices on plastic substrates [199]. The device is configured as diode structure with a Cu plate and a CNF film as anode and cathode respectively. For a fixed applied voltage of 600 V, the ionization current of that device exhibits two regions of linearity with respect to gas pressure below and above 5 Pa, suggesting that the device can be employed as vacuum ion gauge. Ngo et al. reported that due to thermal conductance properties CNF-Cu composite material could be use as a thermal interface material in both IC packaging and equipment cooling applications [200]. The concept of nanothermometer using CNT was first proposed by Gao et al. [201,202]. They used gallium filled CNT as nanothermometer and transmission electron microscope is necessary for observation of CNT during temperature measurement.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 53 From our experimental observation, it is proposed that the nanothermometer can be constructed using MWCNT more easily. As the emission current vary linearly with temperature for a particular applied electric field, so temperature can be directly measured. The sensitivity of the nanothermometer can be adjusted by choosing the area of the MWCNT film or appropriate applied electric field. The above study shows that the temperature dependent field emission property of CNFs and MWCNTs have potential for development of direct thermal-to-electrical power conversion applications. Continued improvements in the PECVD of CNFs/CNTs and related nanostructures are indeed required to explore the potential utility of these structures in advanced applications and future large-scale integration.

REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]

[14] [15] [16] [17] [18] [19] [20]

H.W. Kroto, J.R. Heath, S.C. Obrien, S.C. Curl and R.E. Smalley, Nature, 318, (1985) 162. B.T. Kelly, Physics of Graphite, Applied Science, London, (1981). M. Meyyappan, (Eds.) Carbon nanotubes Science and Applications, CRC Press, (2005). M.L. Weeks, T. Rahmanb, P.D. Frymier, S.K. Islam and T.E. McKnight, Sensors and Actuators B, 133, (2008) 53. M. Endo, C. Kim, T. Karaki, Y. Nishimura, M.J. Matthews, S.D.M. Brown and M.S. Dresselhaus, Carbon 37, (1999) 561. S. Iijima, Nature, 354, (1991) 56. S. Iijima and T. Ichihashi, Nature, 363, (1993) 603. D.S. Bethune, C.H. Kiang, M.S.D. Vries, G. Gorman, R. Savoy, J. Vazquez and R. Beyers, Nature, 363, (1993) 605. T. Yildirim, O. Gulseren, C. Kılıc and S. Ciraci, Phys. Rev. B, 62, (2000) 12648. M.S. Dresselhaus, G. Dresselhaus and P.C. Eklund, Science of Fullerenes and Carbon Nanotubes, Academic Press, London, (1996). R. Saito, G. Dresselhaus and M.S. Dresselhaus, Physical Properties of Carbon Nanotubes, Imperial College Press, London (1998). S.C. Tsang, Y.K. Chen, P.J.F. Harris and M.L.H. Green, Nature, 372, (1994) 159. J. Liu, A. Rinzler, H. Dai, J. Hafner, R. Bradley, P. Boul, A. Lu, T. Iverson, K. Shelimov, C. Huffman, F.R. Macias, Y. Shon, R. Lee, D. Colbert and R.E. Smalley, Science, 280, (1998) 1253. P.M. Ajayan and S. Iijima, Nature, 361, (1993) 333. M.S. Dresselhaus, G. Dresselhaus and R. Saito, Phys. Rev. B, 45, (1992) 6234. J.E. Fischer and A.T. Johnson, Solid State Mater. Sci., 4, (1999) 28. J.P. Issi, J.C. Charlier, K. Tanaka, and T. Yamabe and K. Fukui, (Eds.) The Science and Technology of Carbon Nanotubes, Elsevier, Amsterdam, (1999). C.A. Bessel, K. Laubernds, N.M Rodriguez and R.T.K. Baker, J. Phys. Chem., 105, (2001) 1115. J.A. Elliott, J.K.W. Sandler, A.H. Windle, R.J. Young and M.S.P. Shaffer, Phys. Rev. Lett., 92, (2004) 95501. C.S. Kuo, A. Bai, C.M. Huang, Y.Y. Li, C.C. Hu and C.C. Chen, Carbon, 43, (2005) 2760.

54

Sk. F. Ahmed and K. K. Chattopadhyay

[21] M.S. Dresselhaus, G. Dresselhaus and P. Avouris, (Eds.) Carbon nanotubes: Synthesis, properties and applications, Spinger-verlog, (2000). [22] E.W. Wong and P.E. Sheehan, C.M. Lieber, Science, 277, (1997) 1971. [23] P. Poncharal, Z.L. Wang, D. Ugarte and W.A.D. Heer, Science, 283, (1999) 1513. [24] W.D. Callister, Materials science and engineering, an introduction, New York, Wiley, (2003). [25] J.N. Coleman, U. Khan, W.J. Blau and Y.K. Gunko, Carbon, 44, (2006) 1624. [26] M.A. Osman and D. Srivastava, Nanotechnology, 12, (2001) 21. [27] B.M. Kim and A.M.S. Fuhrer, J. Phys. Cond. Matter., 16, (2004) 553. [28] P.L. Mceuen and J.Y. Park, M.R.S. Bulletin, 29, (2004) 272. [29] P.G. Collins and A. Zettl, Appl. Phys. Lett., 69, (1996)1969. [30] L. Song, S. Liu, G.M. Zhang, L.F. Liu, W.J. Ma, D.F. Liu, X.W. Zhao, S.D. Luo, Z.X. Zhang, Y.J. Xiang, J. Shen, J.J. Zhou, G. Wang and W.Y. Zhou, Chinese Phys. 15, (2006) 422. [31] C. Bower, O. Zhou, W. Zhu, A.G. Ramirez, G.P. Kochanski and S. Jin, Mater. Res. Soc. Symp., 593, (1999). [32] Y. Wang, H. Chen, R.W. Hoffman and J.C. Angus, J. Mater. Res. 5 (1990) 2378. [33] ] M. Hillert and N.Z. Lange, Kristallogr, 111, (1958) 24. [34] L.V. Radushkevich and V.M. Lukyanovich, Russian J. of Phys. Chem., 26, (1952) 88. [35] P. Schutzenberger and L.C.R. Schutzenberger, Acad. Sci. Paris, 111, (1890) 774. [36] C. Pelabon and H.C.R. Pelabon, Acad. Sci. Paris, 137, (1903) 706. [37] R. Bacon, J. Appl. Phys., 31, (1960) 283. [38] C. Herring and J.K. Galt, Phys. Rev., 85, (1952) 1060. [39] A.P. Levitt, Whisker Technology, Wiley-Interscience, New York, 1970. [40] A.W. Moore, A.R. Ubbelohde and D.A. Young, J. Appl. Phys., 13, (1962) 393. [41] L.C.F. Blackman and A.R. Ubbelohde, Proc. Roy. Soc.A, London 266, (1962) 20. [42] T. Koyama, Carbon 10, (1972) 757. [43] M. Endo, T. Koyama and Y. Hishiyama, Jap. J. Appl. Phys., 15, (1976) 2073. [44] G.G. Tibbetts, Appl. Phys. Lett., 42, (1983) 666. [45] G.G. Tibbetts, J. Cryst. Growth, 66, (1984) 632. [46] M. Endo, The growth mechanism of vapor-grown carbon fibers, Orleans, France, (1975). [47] M. Endo, PhD thesis, Nagoya University, Japan, (1978). [48] A. Oberlin, M. Endo and T. Koyama, J. Cryst. Growth, 32, (1976) 335. [49] W. Kratschmer, L.D. Lamb, K. Fostiropoulos and D.R. Huffman, Nature, 347, (1990) 354. [50] T.W. Ebbesen, J. Mater. Res., 13, (1998) 2432. [51] A. Thess, R. Lee, P. Nikolaev, H. Dai and J. Robert, Science, 273, (1996) 483. [52] H. Dai, M.S. Dresselhouse, G. Dresselhouse and P. Avouris, (Eds.) Carbon nanotubes: synthesis, structure, properties and application. New York, Springer, (2001). [53] Z.F. Ren, Z.P. Huang, J.W. Xu, J.H. Wang, P. Bush and M.P. Siegal, Science, 282, (1998) 1105. [54] Y. Gogotsi, J.A. Libra and M. Yoshimura, J. Mater. Res., 15, (2000) 2591. [55] J.M.C. Moreno and M. Yoshimura, J. Am. Chem. Soc., 123, (2001) 741. [56] C.H. Kiang, W.A. Goddard, R. Beyers and D.S. Bethune, Carbon, 33, (1995) 903.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 55 [57] T. Guo, P. Nikolaev, A. Thess, D.T. Colbert and R.E. Smalley, Chem. Phys. Lett., 243, (1995) 49. [58] H. Dai, A.G. Rinzler, P. Nikolaev, A. Thess, D.T Colbert. and R.E. Smalley, Chem. Phys. Lett., 260, (1996) 471. [59] J. Kong, H.T. Soh, A.M. Cassel, C.F. Quate and H. Dai, Nature, 395, (1998) 878. [60] H. Kind, J.M. Bonard, L. Forro, K. Kern, K. Hernadi, L. Nilsson and L. Schlapbach, Langmuir, 16, (2000) 6877. [61] M. Su, B. Zheng and J. Liu, Chem. Phys. Lett., 322, (2000) 321. [62] L. Delzeit, B. Chen, A. Cassell, R. Stevens, C. Nguyen and M. Meyyappan, Chem. Phys. Lett., 348, (2001) 368. [63] R.T.K. Baker, Carbon, 27, (1989) 315. [64] R.T.K. Baker, L.L. Murrell, (Eds.), Novel Materials in Heterogeneous Catalysis,Washington, DC (1990) 32. [65] C. Bower, W. Zhu, S. Jin and O. Zhou, Appl. Phys. Lett., 77, (2000) 830. [66] W. Kim, A. Javey, O. Vermesh, O. Wang, Y.M. Li and H.J. Dai, Nano Lett., 3, (2003) 193. [67] Y.C. Choi, D.W. Kim, T.J. Lee and Y.H. Lee, Synthetic metals, 117, (2001) 81. [68] Q. Wei, J. Sankos and J. Narayan, Surf. Coat. Technol., 146-147, (2001) 250. [69] Y.Y. Wei, G. Eres, V.I. Merkulov and D.H. Lowndes, Appl. Phys. Lett., 78, (2001) 1394. [70] M. Joseyacaman, M. Mikiyoshida, L. Rendon and J.G. Santiesteba, Appl. Phys. Lett., 62, (1993) 657. [71] M. Chhowalla, K.B.K. Teo, C. Ducati, N.L. Rupesinghe, G.A.J. Amaratunga, A.C. Ferrari, D. Roy, J. Robertson and W.I. Milne, J. Appl. Phys., 90, (2001) 5308. [72] V.I. Merkulov, D.H. Lowndes, Y.Y. Wei, G. Eres and E. Voelkl, Appl. Phys. Lett., 76, (2000) 3555. [73] G.W. Ho, A.T.S. Wee, J. Lin and W.C. Tjiu, Thin Solid Films, 388, (2001) 73. [74] L.C. Qin, D. Zhou, A.R. Krauss and D.M. Gruen, Appl. Phys. Lett., 72, (1998) 3437. [75] C.L. Cheung, H. Park and C.M. Lieber, J. Phys. Chem. B, 106, (2002) 2429. [76] H. Kind, J.M. Bonard, C. Emmenegger, L.O. Nilsson, K. Hernadi, E.M. Schaaler, L. Schlapbach, L. Forro and K. Kern, Adv. Mater., 11, (1999), 1285. [77] Y. Chen, Z.L. Wang, J.S. Yin, D.J. Johnson and R.H. Prince, Chem. Phys. Lett,. 272, (1997) 178. [78] K. Kamada, T. Ikuno, S. Takahashi, T. Oyama, T. Yamamoto, M. Kamizono, S. Ohkura, S. Honda, M. Katayama, T. Hirao and K. Oura, Appl. Surf. Sci., 212, (2003) 383. [79] V.I. Merkulov, A.V. Melechko, M.A. Guillorn, D.H. Lowndes and M.L. Simpson, Appl. Phys. Lett., 80, (2002) 476. [80] Y.M. Shyu and F.C.N. Hong, Diam. Relat. Mater., 10, (2001) 1241. [81] M.S. Kin, N.M. Rodriguez and R.T.K. Baker, J. Catal. Growth., 131, (1991) 60. [82] T.K. Tsai, C.C. Chuang, C.G. Chao and W.L. Liu, Diam. Relat. Mater., 12, (2003) 1453. [83] N.M. Rodriguez, J. Mater. Res., 8, (1993) 3233. [84] M. Yudasaka, R. Kikuchi, T. Matsui, Y. Ohki, S. Yoshimura and E. Ota, Appl. Phys. Lett., 67, (1995) 2477. [85] C. Bower, O. Zhou, W. Zhu, D.J. Werder and S. Jin, Appl. Phys.Lett., 77, (2000) 2767.

56

Sk. F. Ahmed and K. K. Chattopadhyay

[86] C.C. Chuang, W.L. Liu, W.J. Chen and J.H. Huang, Appl. Surf. Sci., 254, (2008) 4681. [87] C.C. Chuang, W. L. Liu, W.J. Chen and J.H. Huang, Surf. Coat. Technol., 202, (2008) 2121. [88] V.I. Merkulov, A.V. Melechko, M.A. Guillorn, D.H. Lowndes and M.L. Simpson, Appl. Phys. Lett., 79, (2001) 2970. [89] X. Yang, M.A.Guillorn, D. Austin, A.V. Melechko, H. Cui, H.M. Meyer III, V.I. Merkulov, J.B.O. Caughman, D.H. Lowndes and M.L. Simpson, Nano Lett, 3, (2003) 1751. [90] J. Li, H.T. Ng, A. M. Cassell, W. Fan, H. Chen, Q. Ye, J. Koehne, J. Han and M. Meyyappan, Nano Lett, 3, (2003) 597. [91] H.J. Qi, K.B.T. Teo, K.K.S. Lau, M.C. Boyce, W.I. Milne, J. Robertson, K.K. Gleason, J. Mech. Phys. Sol., 51, (2003) 2213. [92] H.F. Chuang, S.M. Cooper, M. Meyyappan and B.A. Cruden, J. Nanosci. Nanotechnol., 4, (2004) 964. [93] I. Alstrup, J. Catalysis, 109, (1988) 241 [94] D. Nolan, D.C. Lynch and A.H. Cutler, J. Phys. Chem. B, 102, (1998) 4165. [95] R.T.K Baker and J.J. Chludzinski Jr., J. Catalysis, 64, (1980) 464. [96] C.M. Hsu, C.H. Lin, H.J. Lai and C.T. Kuo, Thin Solid Films, 471, (2005) 140. [97] Q. Jiang, Y.W. Wang and J.C. Li, Appl. Surf. Sci., 152, (1999) 156. [98] C.C. Lin, C.C.I.C. Leu, J.H. Ten and M.H. Hon, Diam. Relat. Mater., 14, (2005)1901. [99] X. Chen, R. Wang, J. Xu and D. Yu, Micro, 35, (2004) 455. [100] K. Bartsch and A. Leonhard, Carbon, 42, (2004) 1731. [101] V.I. Merkulov, A.V. Melechko, M.A. Guillorn, D.H. Lowndes and M.L. Simpson, Chem. Phys. Lett., 361, (2002) 492. [102] Z.P. Huang, J.W. Xu, Z.F. Ren, J.H. Wang, M.P. Siegal and P.N. Provencio, Appl. Phys. Lett., 73, (1998) 3845. [103] F. Jensen, Plasma-enhanced Chemical Vapor Deposition, American Vacuum Society, New York, (1997). [104] V.I. Merkulov, M.A. Guillorn, D.H. Lowndes, M.L. Simpson and E. Voelkl, Appl. Phys. Lett., 79, (2001) 1178. [105] M. Laudon, N.N. Carlson, M.P. Masquelier, M.S. Daw and W. Windl, Appl. Phys. Lett., 78, (2001) 201. [106] Q. Ngo, A. M. Cassell, V. Radmilovic, J. Li, S. Krishnan, M. Meyyappan and C. Y. Yang, Carbon, 45, (2007) 424. [107] H. Huang, H. Kajiura, Y. Murakami and M. Ata, Carbon, 41, (2003) 615. [108] L. Yu, L. Sui, Y. Qin and Z. Cui, Chem. Eng. Journal, 144, (2008) 514. [109] P.L. Hansen, J.B.Wagner, S. Helveg, J.R.R. Nielsen, B.S. Clausen and H. Topsoe, Science, 295, (2002) 2053. [110] N.M. Rodriguez, A. Chmbers and R.T.K. Baker, Langmuir, 11, (1995) 3862. [111] A. Chambers and R.T.K. Baker, J. Phys. Chem. B, 101, (1997) 1621. [112] M. Tanemura, M. Kitazawa, J. Tanaka, T. Okita, R. Ohta1, L. Miao and S. Tanemura, Jpn. J. Appl. Phys., 45, (2006) 2004. [113] O. Auciello and R. Kelly, (Eds.), Ion Bombardment Modification of Surfaces, Elsevier, Amsterdam, (1984). [114] J.A.V. Vechten, W. Solberg, P.E. Batson, J.J. Cuomo and S.M. Rossnagel, J. Cryst. Growth, 82, (1987) 289.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 57 [115] P. Sigmund, (Eds.) Fundamental Processes in Sputtering of Atoms and Molecules, Royal Danish Academy of Sciences and Letters, Copenhagen, (1993). [116] J. Li, C. Papadopoulos, J.M. Xu and M. Moskovits, Appl. Phys. Lett., 75, (1999) 367. [117] R. Gomer, Field Emission and Field Ionization, American Institute of Phys., New York, (1993). [118] R.H. Fowler and L.W. Nordheim, Proc. R. Soc. London, series A, 119, (1928) 173. [119] E.L. Murphy and Jr. R.H. Good, Phys. Rev., 102, (1956) 1464. [120] R.G. Forbes, Ultramicroscopy, 79, (1999)11. [121] R.G. Forbes, Proceedings of 4th Moscow International ITEP School of Physics (Eds: A.L. Suvorov, Yu.G. Abov, V.G. Firsov) Moscow, (2001) p.62. [122] L. Nilsson, O. Groening, P. Groening, O. Kuettel and L. Schlapbach, J. Appl. Phys., 90, (2001) 768. [123] R. Gao, Z. Pan and Z.L. Wang, Appl Phys. Lett., 78, (2001) 1757. [124] R.G. Forbes, C.J. Edgcombe and U. Valdre, Ultramicroscopy, 95, (2003) 57. [125] J. Cumings, A. Zettl, M.R. McCartney and J.C.H. Spence, Phys. Rev. Lett., 88, (2002) 56804. [126] X. Zheng, G.H. Chen, Z. Li, S. Deng and N. Xu, Phys. Rev. Lett., 92, (2004) 106803. [127] R. Collazo, R. Schlesser and Z. Sitar, Appl. Phys. Lett., 78, (2001) 2058. [128] S.M. Yoon, J. Chae and J.S. Suh, Appl. Phys. Lett., 84, (2004) 825. [129] S.C. Kung, K.C. Hwang and I.N. Lin, Appl. Phys. Lett., 80, (2002) 4819. [130] S.H. Jo, Y. Tu, Z.P. Huang, D.L. Carnahan, J.Y. Huang, D.Z. Wang and Z.F. Ren, Appl. Phys. Lett., 84, (2004) 413. [131] A.G. Rinzler, J.H. Hafner, P. Lou Nikolaev, L. Kim, S.G. Tomanek, D. Colbert and R.E. Smalley, Science, 269, (1995) 1550. [132] Y. Saito, K. Hamaguchi, K Hata., K. Uchida, Y. Tasaka, F. Ikazaki, M. Yumura, A. Kasuya and Y. Nishina, Nature, 389, (1997) 554. [133] Y. Saito, K. Hamaguchi, T. Nishino, K. Hata, K. Tohji, A. Kasuya and Y. Nishina, Jpn. J. Appl. Phys., 36, (1997) 1340. [134] P.C. Collins, M.S. Arnold and P. Avouris, Science, 292, (2001) 706. [135] W.A. De Heer, A. Chatelain and D. Ugerate, Science, 270, (1995) 1179. [136] W. Zhu, C. Bower, O. Zhou, G. Kochanski and S. Jin, Appl. Phys. Lett., 75, (1999) 873. [137] H. Sugie, M. Tanemura, V. Filip, K. Iwata, K. Takahashi and F. Okuyama, Appl. Phys. Lett., 78, (2001) 2578. [138] J. Robertson, Mater. Res. Soc. Symp. Proc., 498, (1998) 197. [139] J. Robertson and M.J. Rutter, Diam. Relat. Mater., 7, (1998) 620. [140] G.A.J. Amaratunga and S.R.P. Silva, App Phys Lett., 68, (1996) 2529. [141] F.Y. Chuang, C.Y. Sun, H.F. Cheng, W.C. Wang and I.N. Lin, Appl. Surf. Sci., 113, (1997) 259. [142] O. Groning, O.M. Kuttel, P. Groning and L. Schlapbach, Appl. Surf. Sci., 111, (1997) 135. [143] A. Ilie, A.C. Ferrari, T. Yagi and J. Robertson, Appl. Phys. Lett., 7, (2000) 2627. [144] J.D. Carey, R. D. Forrest, R.U. A. Khan and S.R.P. Silva, Appl. Phys. Lett., 77, (2000) 2006. [145] M.W. Geis, J.C. Twichell, J. Macaulay and K. Okano, Appl. Phys. Lett., 67, (1995) 1328.

58

Sk. F. Ahmed and K. K. Chattopadhyay

[146] B.S. Satyanarayana, A. Hart, W.I. Milne and J. Robertson, Appl. Phys. Lett., 71, (1997) 1430. [147] A. Hart, B.S. Satyanarayana, J. Robertson and W.I. Milne, Appl. Phys. Lett., 74, (1999) 1594. [148] M.W. Geis, J.C. Twichell and T.M. Lyszczarz, J. Vac. Sci. Technol. B, 14, (1996) 2060. [149] R.D. Forrest, A.P. Burden, S.R.P. Silva, L.K. Cheah and X. Shi, Appl. Phys. Lett., 73, (1998) 3784. [150] J. Robertson, Mater. Res. Soc. Symp. Proc. 417, (1997) 217. [151] N.S. Xu, Y. Tzeng and R.V. Latham, J. Phys. D, 26, (1993) 1776. [152] A.V. Karabutov, V.D. Frolov, S.M. Pimenov and V.I. Konov, Diam. Relat. Mater., 8, (1999) 763. [153] O.M. Kuttel, O. Groning, C. Emmenegger, L. Nilsson, E. Maillard, L. Diederich and L. Schlapbach, Carbon, 37, (1999) 745. [154] T. Ikunoa, J.T. Ryu, T. Oyamaa, S. Ohkura, Y.G. Baek, S. Honda, M. Katayama, T. Hirao and K. Oura, Vacuum, 66, (2002) 341. [155] J.M.. Kim, W.B. Choi, N.S. Lee and J.E. Jung, Diam. Relat. Mater., 9, (2000) 1184. [156] H. Murakami, M. Hirakawa, C. Tanaka and H. Yamakawa, Appl. Phys. Lett., 76, (2000) 1776. [157] A.N. Obraztov, A.P. Volkov and I. Pavlovsky, Diam. Relat. Mater., 9, (2000) 1190. [158] F. Hoshi, K. Tsugawa, A. Goto, T. Ishikura, S.Yamashita, M. Yumura, T. Hirao, K. Oura and Y. Koga, Diam. Relat. Mater., 10, ( 2001) 254. [159] A. Ilie, A.C. Ferrari, T. Yagi, S.E. Rodil, J. Robertson, E. Barborini and P. Milani, J. Appl. Phys., 90, (2001) 2024. [160] D.N. Davydov, P.A. Sattari, D. AlMawlawi, A. Osika and T.L. Haslett, J. Appl. Phys., 86, (1999) 3983. [161] C. Wang, A. Garcia, D.C. Ingram, M. Lakea and M.E. Kordesch, Electron. Lett., 27, (1991) 1459. [162] N.S. Xu, R.V. Latham and Y. Tzeng, Electron. Lett., 29, (1993) 1596. [163] W. Zhu, C. Bower, G.P. Kochanski and S. Jin, Solid State Electron., 45, (2001) 921. [164] A.N. Obraztsov, I.Y. Pavlovsky, A.P. Volkov, A.S. Petrov, V.I. Petrov, E.V. Rakova, V.V. Roddatis, Diam. Relat. Mater., 8, (1999) 814. [165] J.M. Bonard, J.P. Salvetat, T. Stockli, W.A. de Heer, L. Forro and A. Chatelain, Appl. Phys. Lett., 73, (1998) 918. [166] C. Kimura, Y. Yamamuro, H. Aoki and Takashi Sugino, Diam. Relat. Mater., 16, (2007) 1383. [167] K.B.K. Teo, S.B. Lee, M. Chhowalla, V. Semet, V.T. Binh, O. Groening, M. Castignolles, A. Loiseau, G. Pirio, P. Legagneux, D. Pribat, D.G. Hasko, H. Ahmed, G.A.J. Amaratunga and W.I. Milne, Nanotechnology, 14, (2003) 204. [168] S.F. Ahmed, S. Das, M.K. Mitra and K.K. Chattopadhyay, Ind. Journal of Pure and Applied Physics, 44, (2006) 700. [169] V.L. Merkulov, D.H. Lowndes, Y. Wei and G. Eres, Appl. Phys. Lett., 76, (2000) 3555. [170] S.F. Ahmed, Ph.D. Thesis, Some aspects of nanocrystalline silicon incorporated diamond like carbon and carbon nanostructures, Jadavpur University, Kolkata, India, (2008). [171] S. Gupta, B.L. Weiss, B.R. Weiner and G. Morell, J. Appl. Phys., 89, (2001) 5671.

Synthesis and Electron Field Emission from Different Morphology Carbon Nanofibers 59 [172] A.Y. Cao, X.F. Zhang, X. Xiao, M.Q. Ding, D.M. Zhuang, C.L. Xu, B.Q. Wei, J., Liang and D.H. Wu, Mater. Lett., 51, (2001) 371. [173] L. Xinghui, Z. Changchun, L. Weihua, Z. Fanguang and T. Changhui, Mater. Chem. and Phys., 93, (2005) 473. [174] K.H. Park, J.H. Yim, S. Lee and K.H. Koh, J Vac Sci Technol B, 23, (2005) 776. [175] X.T. Zhou, H.L. Lai, H.Y. Peng, F.C.K. Au, L.S. Liao, N. Wang, I. Bello, C.S. Lee and S.T. Lee, Chem. Phys. Lett., 318, (2000) 58. [176] F.C.K. Au, K.W. Wong, Y.H. Tang, Y.F. Zhang, I. Bello and S.T. Lee, Appl. Phys. Lett., 75, (1999)1700. [177] K. Okano, S. Koizumi, S.R.P. Silva and G.A.J. Amartunga, Nature, 381, (1996) 140. [178] S. Das, S.F. Ahmed, M.K. Mitra and K.K. Chattopadhyay, Appl. Phys. A, 91, (2008) 429. [179] S.F. Ahmed, M.K. Mitra and K.K. Chattopadhyay, Appl. Surf. Sci., 254, (2007) 610. [180] W. Zhu, Vacuum Microelectronics, Wiley, New York, (2001) Chapter 3. [181] A. Modinos, Field, Thermionic, and Secondary Electron Emission Spectroscopy, Plenum, New York, (1984). [182] M. Shiraishi and M. Ata, Carbon, 39, (2001) 1913. [183] M.C. Kan, J.L. Huang, J.C. Sung, K.H. Chen and B.S.Yau, Carbon, 41, (2003) 2839. [184] M.J. Fransen, T.L.V.Rooy and P. Kruit, Appl. Surf. Sci., 146, (1999) 312. [185] Y. Chen, D.T. Shaw and L. Guo, Appl. Phys. Lett., 76, (2000) 2469. [186] L. Nilsson, O. Groening, C. Emmenegger, O. Kuettel, E. Schaller, L. Schlapbach H. Kind, J.M. Bonard and K. Kern, Appl. Phys. Lett., 76, (2000) 2071. [187] S. Frank, P. Poncharal, Z.L. Wang and W.A.D. Heer, Science, 280, (1998) 1744. [188] C. Dekker, W. Henk, C. Postma, T. Teepen, Z. Yao and M. Grifoni, Science, 293, (2001) 5527. [189] A. Bachtold, P. Hadley, T. Nakanishi and C. Dekker, Science, 294, (2001) 1317. [190] R.H. Baughman, A.A. Zakhidov and W.A. de Heer, Science, 297, (2002) 787. [191] M. Yu, O. Lourie, M.J. Drayer, K. Molini, T.F. Kelly and R.S. Ruoff, Science, 287, (2000) 637. [192] B.Q. Wei, R. Vajtal and P.M. Ajayan, Appl Phys Lett., 79, (2001) 1172. [193] W.I. Milne, K.B.K. Teo, G.A.J. Amaratunga, P. Legagneux, L. Gangloffa and J.P. Schnell, J. Mater. Chem., 14, (2004) 933. [194] G.S. Duesberg, A.P. Graham, M. Liebau, R. Seidel, E. Unger and F. Kreupl, Nano Lett., 3, (2003) 257. [195] J. Kong, N.R. Franklin, C. Zhou, M.G. Chaplene, S. Peng and K. Cho, Science, 287, (2000) 622. [196] C. Emmenegger, P. Mauron, P. Sudan, P. Wenger, V. Hermann and R. Gallay, J. Power Sources, 124, (2003) 321. [197] T. Yoshitake, Y. Shimakawa, S. Kuroshima, H. Kimura, T. Ichihashi and Y. Kubo, Physica B, 323, (2002) 124. [198] M. Endo, Y.A. Kim, T. Hayashi, K. Nishimura, T. Matusita and K. Miyashita, Carbon, 39, (2001) 1287. [199] H.S. Sim, S.P. Lau, L.K. Ang, M. Tanemura and K. Yamaguchi, Diam. Relat. Mater., 17, (2008) 1959. [200] Q. Ngo, B.A. Cruden, A.M. Cassell, G. Sims, M. Meyyappan, J. Li and C.Y. Yang, Nano Letters, 4, (2004) 2403.

60

Sk. F. Ahmed and K. K. Chattopadhyay

[201] Y.H. Gao and Y. Bando, Nature, 415, (2002)599. [202] Y.H. Gao, Y. Bando, Z.W. Liu and D. Golberg, Appl. Phys. Lett., 83, (2003) 2913.

In: Nanofibers: Fabrication, Performance, and Applications ISBN: 978-1-60741-947-1 Editor: W. N. Chang ©2009 Nova Science Publishers, Inc.

Chapter 2

PERMEABILITY STUDIES OF ELECTROSPUN CHITIN AND CHITOSAN NANOFIBROUS MEMBRANES Jessica D. Schiffman and Caroline L. Schauer* Drexel University, Philadelphia, PA, USA

1. ABSTRACT Electrospinning has been utilized to fabricate fibrous membranes composed of polymer nanofibers, which have large surface area-to-volume ratios and small pores. Electrospun nanofibrous membranes have potential uses in a variety of industries such as energy, environment, medicine, packaging, and automotive, with specific applications including air filtration, protective clothing, fuel cells, and nanocomposites. Nanofibrous membranes composed of biopolymers have potential uses that harness their inherent biocompatibility. Chitin, the second most abundant, naturally occurring polysaccharide after cellulose, is found in shells of crabs and shrimp. Chitosan, the acid soluble form of chitin, is a non-toxic, biodegradable, biopolymer consisting primarily of ί(1 4) linked 2-amino-2-deoxy-ί-D-glucopyranose units, and is currently used in tissue engineering, antifouling coatings, separation membranes, stent coatings, enzyme immobilization matrices, and the removal of heavy metals from ground and wastewater. Chitosan is a commercially interesting compound because of its high nitrogen content (6.89%), making it a useful chelating agent for metal ions. Before these chitin or chitosan nanofibrous membranes can be used in the myriad of industries their physical properties, such as permeability, must be known. This chapter focuses on the fabrication and flow cell testing of chitin and chitosan nanofibrous membranes. Additionally, it explores the potential applications of biopolymer and synthetic polymer electrospun membranes.

* Telephone: (215) 895-6797, Fax: (215) 895-6760, E-mail: [email protected]

62

Jessica D. Schiffman and Caroline L. Schauer

2. INTRODUCTION 2.1. Electrospinning Non-woven fibrous membranes, which have large surface area-to-volume ratios can be fabricated utilizing the electrospinning process. A simple schematic of a standard laboratory set-up is given in Figure 1. This is the only production method that produces fibers by utilizing electrical forces as opposed to a mechanical pulling force. As a result of this, submicron fibers are created and deposited in the form of a non-woven membrane.

Figure 1. Diagram displaying the basic components of an electrospinning apparatus including (a) syringe needle loaded with a polymeric solution, which is often advanced at a constant rate by a metering pump, (b) voltage supply, and (c) target onto which the electrospun fibers accumulate.

In a 2007 review article, which discussed the state of electrospun filtration membranes, Barhate and Ramakrishna [1] included a table of enterprises that work in the nanofiber production area. Their table included eight companies in the United States of America, three in Germany, three in Japan, two in South Korea, as well as one in Canada, the Czech Republic, the Republic of Estonia, and Finland. Therefore, there is a widespread invested interest in nanofibrous filtration.

2.2. Select Electrospinning Examples The specific investigations that are discussed in this section are meant to represent (1) the diversity of precursor materials that have been electrospun and (2) the wide range of filtration related applications from said materials. All of the articles discussed in this section have been published within the previous year and utilize synthetic polymers. Electrospun membranes could potentially act as pre-filters to rid solid particulates from a flow so that the down-stream filtration membranes will become less fouled and experience an increased lifecycle between required cleanings or replacements. The functionality of electrospun nylon-6 (dissolved in 75% formic acid) was tested as a pre-filter by attempting to

Permeability Studies of Electrospun Chitin and Chitosan Nanofibrous Membranes

63

pass polystyrene (PS) micro-particles (Ø= 0.5 - 10 µm) through the membrane.[2] It was determined that the membranes became irreversibly fouled. Kim et al. [3] have also electrospun pre-filters from nylon 6/formic acid solutions and investigated their efficiency utilizing polystyrene latex particles with known diameters ranging from 0.02 - 1.0 µm. Membranes with a mean diameter of 100 nm displayed a lower pressure drop in comparison to commercial high efficiency particulate air (HEPA) filters. The collection of aerosol particles by an electrospun polyacrylamide solution containing 2,2′-(bisacrylamino) diethyl disulfide (BAC) was tested by Vetcher et al.[4] Potentially, these filters could be utilized for indoor and outdoor testing of microbes, pollen allergens, and toxins. Liquid filtration applications have also been investigated since electrospun membranes could have higher flux and anti-fouling properties than other ultra-purification membranes. Veleirinho and Lopes-da-Silva [5] have electrospun a solution of poly(ethylene terephthalate) (PET) dissolved in trifluoracetic acid (TFA) : dichloromethane (DCM) (80:20 v/v) to investigated its applicability to apple juice clarification. The clarified juice exhibited physicochemical characteristics comparable when conventional techniques were employed. In addition to the typically randomly accumulated fibrous membranes, hollow fibers have been electrospun at the University of Washington [6, 7] utilizing a multilayer microlayer electrospinning device. Into one microchannel, Srivastava et al. [8] loaded titanium isopropoxide (Ti(OiPr)4) stock solution (ethanol/acetic acid) that was added to poly(vinylpyrrolidone)/ethanol solution. A heavy mineral oil was added via another microchannel and was the core material. From this, hollow core/sheath nanofibers were electrospun. Robust hollow fibers with diameters ranging from 85 to 350 nm were obtained from these networks, whose geometry could be customized according to the end application.

2.3. Chitin and Chitosan The recent articles discussed in section 2.2 utilized synthetic polymers. We have taken a slightly different approach by electrospinning and testing the biopolymers chitin and chitosan. We believe that these biopolymers hold a strong potential for filtration applications because they are renewable resources that have attractive intrinsic properties. Chitin is the second most abundant natural polysaccharide and consists of N-acetyl-ί-D-glucosamine chains. The functional polysaccharide chitin, is produced primarily by arthropods and crustaceans as ordered crystalline microfibrils; it also exists as the structural component in the cell walls of fungi and yeast.[9] The distribution of the chitin produced from arthropods annually is displayed in Figure 2.[10] Chitin exists in two main crystalline polymorphic forms: α and ί. Alternating sheets of parallel and antiparallel chains tightly pack into an orthorhombic cell, which are obtained from insect cuticles, shrimp and crab shells, fungal and yeast cells, lobster and crab tendons, it is known as α-chitin. [9, 11, 12] A second polymorph, which is far less common, ί-chitin, is found in squid pens, in the tubes synthesized by vestimetiferan and pogonophoran worms, Aphrodite chaetae, and in the monocrystalline spines excreted by the diatom. In this form, the chains are arranged parallel within a monoclinic unit cell.[13] A third rare form, ΰ-chitin, is thought to be a mixture of the α- and ί- chitin forms, containing both parallel and antiparallel arrangements.[14, 15]

64

Jessica D. Schiffman and Caroline L. Schauer

Figure 2. Breakdown of the total chitin produced by arthropods (tons per year) annually from three different environments. Data adapted from Cauchie.[10]

Since chitin is a highly insoluble material, chitosan, the N-deacetylated derivative of chitin is often studied. As a result of the increase in available free amine groups, chitosan can be dissolved in aqueous acidic solvents that chitin cannot dissolve in, such as formic acid, acetic acid (AA), and malic acid. Solubility is achieved by protonating the –NH2 function on the C-2 of the D-glucosamine repeat unit, creating a polyelectrolyte in acidic solutions.[9, 16] Both biopolymers have many of the same attributes including antibacterial activity, biocompatibility, chelating capabilities, biodegradability, and adsorption properties.[16] As a result of these properties, chitin and chitosan can be applied to an extensive array of applications in fields ranging from biomedical to environmental. In the remainder of this chapter, we will focus on our current electrospinning of these two biopolymers, as well as initial results concerning their permeability.

3. RESEARCH RESULTS Both chitin [17-20] and chitosan [21-28] fibrous membranes have previously been fabricated utilizing the electrospinning process. In the case of electrospinning chitin and chitosan, the given biopolymers were dissolved in 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP) for at least 72 h and TFA for 24 h, respectively. The polymeric solutions were loaded into (5 mL) Luer-lock syringes (Becton Dickinson & Co, Franklin Lakes, NJ), which were capped with Precision Glide 21-gauge needle (Becton Dickinson & Co, Franklin Lakes, NJ) and placed on a metering pump (Harvard Apparatus, Plymouth Meeting, PA) that was set at a constant advancing speed around 1.2 mL/h. When a voltage of approximately 25 kV was applied by the high voltage supply (Gamma High Voltage Research Inc., Ormond Beach, FL) to the metal needle and the target, a Taylor cone [29] was formed. This is a critical step for the initiation and propagation of fiber formation. Next, the electrostatic force needs to overcome the surface tension force of the Taylor cone so that a thin jet can form and thin out over three stages. They include: jet initiation and extension in a straight line, whipping instability, and jet solidification and fiber collection. In all of our experiments, a copper plate

Permeability Studies of Electrospun Chitin and Chitosan Nanofibrous Membranes

65

wrapped with store bought aluminum foil was utilized as the target. Dry chitin and chitosan fibers were deposited onto the target since most of the solvent utilized to dissolve the biopolymers evaporates over the course of the separation distance (gap between the tip of the needle and the target). The separation distance utilized for the electrospinning of chitin and chitosan was 6.4 cm and 6.0 cm, respectively. Utilizing the aforementioned parameters, chitin and chitosan fibrous membranes were successfully electrospun into nanofibrous membranes comprised of randomly oriented, continuous cylindrical fibers. The practical grade (PG) of chitin from crab shells (SigmaAlrdich, St Louis, MO) that was spun has a degree of deacetylation (DD) of 9% as determined by Fourier transform infrared spectroscopy (FTIR).[19] The DD and molecular weights (MW) of the chitosans that were successfully electrospun [23] are given in Table 1. Table 1. Various chitosan biopolymers, which were purchased from Sigma-Aldrich (St Louis, MO) and successfully electrospun. Their degree of deacetylation (DD) [23] and molecular weight (MW) are given. Chitosan Practial grade (PG) Low molecular weight Medium molecular weight High molecular weight *As determined by FTIR **As supplied by Sigma-Aldrich

DD (%)* 75 74 83 72

MW** 190-375,000 70,000 190-310,000 500-700,000

The average fiber diameter of all membranes electrospun was determined utilizing a Zeiss Supra 50/VP field emission scanning electron microscope (FESEM) by measuring the diameters of fifty random fibers. The term as-spun implies that no additional coatings or alterations were made to the fibrous membranes. As displayed on Table 2, the average fiber diameters of all as-spun chitin and chitosans are within standard deviation from each other. In order for as-spun fibrous membranes to be used in field devices, the chemical stability of the chitosan fibers needed improvement. This was determined as a result of the initial solubility testing conducted on as-spun chitosan membranes. (See Table 2, Solubility: Post 72 h.) The solubility of the chitin and chitosan fibrous membranes were tested [23] utilizing three 15-mm2 petri dishes (Becton Dickinson, Franklin Lakes, NJ), which contained 30 mL of three various solutions: basic (1 M sodium hydrocxide, NaOH), acidic (1 M AA), and aqueous (ultrapure H2O). Two samples of electrospun membrane, (2.54 cm x 1.27 cm) were placed into each solution. After 15 min, if possible, one of the membranes was removed, while the other remained in the solution for 72 h.

66

Jessica D. Schiffman and Caroline L. Schauer

Table 2. Properties of various electrospun membranes including: their average fiber diameter (n=50), the biopolymer that was electrospun and cross-linking method (when applicable), and the fiber membrane stability after fibrous membranes were submerged into 1M solutions of acetic acid (AA), ultrapure aqueous (H2O), and sodium hydroxide (NaOH) for 72 h. All data displayed is for electrospun (PG) practical grade or (MMW) medium molecular weight biopolymers. Electrospun Biopolymer

Stability: Post 72h

Average Fiber

AA

H2O

NaOH

Diameter (nm)

As-spun PG chitin

√

√

√

152

±

70

As-spun PG chitosan

X

X

√

58

±

20

As-spun MMW chitosan X One-step cross-linked MMW chitosan X Two-step cross-linked MMW chitosan √

X

√

77

±

29

√

√

128

±

40

√

√

172

±

75

3.1. Cross-Linking Studies As evident from Table 2, electrospun medium molecular weight (MMW) chitosan membranes did not survive when immersed in aqueous and acetic solutions. Hence, initial investigations regarding two different cross-linking methods were employed. A two-step cross-linking method was first identified, [23] utilizing vapor-phase glutaraldehyde (GA). The first step of this process consists of electrospinning the chitosan solution; step two is exposing the as-spun fibrous membranes overnight to vapor-GA in a vaporization chamber. Shortly after this work, we successfully demonstrated the electrospinning of cross-linked chitosan fibers in one-step, utilizing GA-liquid.[24] In this in-situ method, a small amount of GA liquid is added just prior to the electrospinning of the chitosan solution, and the resultant electrospun fibrous membranes are insoluble in aqueous, basic, and acidic solutions by the completion of the electrospinning session. Both cross-linking methods transform the electrospun chitosan fibrous membranes so that they are stable for at least 72 h in AA, aqueous, and NaOH solutions. Fibrous membranes composed of chitin are stable in the previously mentioned solutions without further processing.

3.2. Lead Ion Testing The PG chitin and MMW two-step cross-linked chitosan fibrous membranes are both chemically stable and have potential to act as filtration devices based on their intrinsic chelation capabilities. Therefore, their capability to bind with lead ions was tested. Solutions containing various amounts of lead acetate were created and PG chitin and two-step crosslinked fibrous membranes of MMW chitosan were subjected to the solutions for 1 h. After this hour, they were removed and rinsed repeatedly with DI water. These samples were next prepared for analysis utilizing a FESEM equipped with energy dispersive x-ray spectroscopy

Permeability Studies of Electrospun Chitin and Chitosan Nanofibrous Membranes

67

(EDS) system. EDS distinctly identified that when fibrous membranes were subjected to solutions with 50 ppm lead, lead was incorporated into the membrane. Chitin and two-step cross-linked MMW chitosan membranes were additionally tested in solutions containing 5 ppm, 500 ppb, and 50 ppb lead. Spectra acquired on these samples indicate that lead has been incorporated into the membranes. However, further work must be done to quantify these amounts, determine binding capacity of the fibrous membranes, and identify the lowest detectable limits of lead for the EDS system.

3.3. Permeability Studies The permeability (k) of two-step cross-linked PG chitosan and as-spun PG chitin membranes were evaluated according to ASTM standard D4491. A constant head test utilizing the flow cell apparatus displayed in Figure 3(c) was employed. A constant head of water was maintained on the electrospun membrane and the quantity of flow was measured over a specific amount of time. In order to calculate the membrane permeability (k), the permittivity (ψ) was first determined according to the following equation: ψ = k/t = q / (∆h * A)

Equation 1

Where: ψ is permittivity (s-1), k is permeability (m/s), t is thickness (m), q is quantity of flow or flow rate (m3/s), ∆h is head lost (m), and A is area of the test specimen (m2). A constant head test over a circular (Ø = 5.08 cm) area was employed. A constant head of water was maintained on the electrospun membrane and the quantity of flow was measured over a specific amount of time. As seen in Figure 3(b), fibrous membranes that were approximately (7.62 cm x 6.35 cm) were electrospun; however the water only flowed through a 5.08 cm diameter circular section of the membranes. To ensure that the membranes did not move during the remaining preparation of the flow cell apparatus or during the experimental run, the membranes were placed between two thin sheets of metal mesh. These background metal meshes were tested independently of the fibrous membranes and it was determined that their permeability is so high that they do not restrict or have a negligible influence on the flow during the testing. The fibrous membranes, sandwiched between the metal mesh was clamped between (5.08 cm diameter) polyvinyl chloride (PVC) pipes that were then clamped together. This area is labeled as “sample holder” in Figure 3(b). After a sample was carefully loaded, de-aired water was added into the discharge pipe until the system was backfilled. The bleed valve was utilized to get rid of any air bubbles that were trapped within the apparatus during this process. Once backfilled, water was added until it reached the overflow outlet, then, the rate of water being added was held constant at a reduced rate. Next, the rotating discharge pipe was moved so that (1) the gauge for measuring head (see Figure 3(c)) displayed a constant height difference (∆h) and (2) there was a constant water runoff from the discharge pipe. At this time, the quantity of flow (q) was recorded for a particular time (s), the readings were averaged, and the permittivity determined according to the equation previously given. This value can then easily be converted to permeability. Multiple samples were utilized and the permittivity of each sample was determined in repetition four times and converted to permeability by utilizing their average thickness

68

Jessica D. Schiffman and Caroline L. Schauer

measurements. It is important to note that this conversion can yield increased amount of standard deviation error. The average permeability of the chitin and two-step cross-linked chitosan membranes was determined to be 0.144 ± 0.013 µm/s and 0.344 ± 0.197 µm/s, respectively. According to ASTM standards, a coefficient of permeability (cm/s) 10-3 > k > 10-5 is the equivalent to sand, dirty sand, or silty sand and has a “low” degree of permeability. The membranes tested fall between this and the permeability category that has a “very low” rating (where 10-5 > k > 10-7); here the membranes fall into the soil rating of silt or silty clay.[30]

Figure 3. Set-up of flow cell apparatus displaying (a) water storage, (b) two-step cross-linked PG chitosan fibrous membrane being loaded into sample holder, and (c) the apparatus utilized to determine permeability of the membranes.

The (5.08 cm diameter) circular area that the water flowed through looks consistent with the rest of the membrane. This indicates that the membranes were both chemically resistant and mechanically strong enough to withstand the forces from the flow cell. Additionally, FESEM micrographs displayed in Figures 4(a) and (c) demonstrates that cylindrical fiber morphology was retained. Both the two-step cross-linked PG chitosan and the PG chitin fibrous membranes display fine cylindrical and continuous fibers. The micrographs were acquired after desiccating the membranes for at least 24 h. The average fiber diameters before and after the flow cell experiment were determined to be within standard deviation of each other before and after use in the flow cells. They were 152

Permeability Studies of Electrospun Chitin and Chitosan Nanofibrous Membranes

69

± 70 nm and 326 ± 171 nm for chitin membranes and 172 ± 75 nm and 253 ± 108 nm for the two-step cross-linked PG chitosan, pre- and post-cell respectively. It is important to note that the biopolymers chitin and chitosan are known to swell in aqueous solutions. Thus, during the flow cell experiments, it can be assumed that the water flowing through the cell is causing the fibers to increase in diameter. These membranes were desiccated prior to the acquisition of the fiber diameter distribution data.

Figure 4. Displays FESEM micrographs and digital images of the fibrous membranes after they were removed from the flow cell. Images (a) and (b) are of a PG chitin membrane, whereas (c) and (d) are a PG chitosan membrane.

The water that has been used throughout the flow cell investigation is unpurified Philadelphia, Pennsylvania tap water. A sample of this water was sent to Robertson Microlit Laboratories (Madison, NJ) to be analyzed. It was determined that the water utilized in this study contains 33 ppm fluorine and less than 1 ppm iron. EDS conducted on the fibrous membranes post-permeability studies were unable to confirm the presence of either of these elements at their aqueous levels in the membranes. Future investigations should include permeability studies that utilize heavy metal ion contaminated waters through the fibrous membranes to determine the detection limits of these fibrous membranes. Additionally, more analysis concerning the uniformity of the fibrous membranes after their use should be evaluated.

4. APPLICATIONS OUTLOOK As devices and their components become miniaturized and contain nano-sized features, higher proportions of atoms are on the surface. This results in new and often enhanced properties such as increased quantum efficiency, surface energy and reactivity, thermal and electrical conductivity, high strength-to-weight ratios, and superparamagnetism.[31, 32] Certainly, by fabricating membranes composed of nanofibers very high surface area-tovolume ratios exist.

70

Jessica D. Schiffman and Caroline L. Schauer

Nanofibrous membranes are appropriate for a wide variety of applications based upon their low weight, high permeability, and small pore sizes. Biopolymer nanofibrous membranes could be appropriate as antimicrobial filters, ultrafiltration membranes, affinity filters, [1] catalytic filters, coalescence filters, ion exchange media, affinity filters, particle filters in vivo, biomedical sutures, [33] filters for metal recovery, [34] as templates, [35-37] and for protective clothing that is both chemically and biologically protective.[38] The porosity of electrospun membranes can be altered [39-41] and thus the properties of these membranes, such as the number of anchoring points for cells, wetting-properties, and degradation rates can vary. Medical textiles, chemical filtration, fuel cell membranes, catalysis, electrochemical cells, and nano-reinforcements would benefit from having a porosity that could be engineered for the needs of the particular application.[42, 43]

5. CONCLUSION Electrospun membranes composed of chitin or chitosan offer many advantages to membranes composed of synthetic polymers. The biopolymers are sustainable and ecoefficient, while also being biocompatible, antibacterial, and biodegradable. These characteristics become even more exciting when the biopolymers are processed into nanofibrous membranes due to the nano-effects that occur. As we start to investigate and understand the permeability properties that electrospun chitin and chitosan membranes have, we foresee these membraneerials will behave as well as synthetic polymers, while having less of an impact on the environment.

REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14]

Barhate, R. S.; Ramakrishna, S. J. Membr. Sci. 2007, 296, 1-8. Aussawasathien, D.; Teerawattananon, C.; Vongachariya, A. J. Membr. Sci. 2008, 315, 11-19. Kim, G. T.; Ahn, Y. C.; Lee, J. K. Korean J. Chem. Eng. 2008, 25, 368-372. Vetcher, A. A.; Gearheart, R.; Morozov, V. N. Polymers for Advanced Technologies 2008, 19, 1276-1285. Veleirinho, B.; Lopes-da-Silva, J. A. Process Biochem. 2009, 44, 353-356. Li, D.; McCann, Jesse T.; Xia, Y. Small 2005, 1, 83-86. McCann, J. T.; Li, D.; Xia, Y. J. Mater. Chem. 2005, 15, 735-738. Srivastava, Y.; Loscertales, I.; Marquez, M.; Thorsen, T. Microfluidics and Nanofluidics 2008, 4, 245-250. Rinaudo, M. Prog. Polym. Sci 2006, 31, 603-632. Cauchie, H. M. Hydrobiologia 2002, 470, 63-96. Schiffman, J. D.; Schauer, C. L. Mater. Sci. Eng., C 2009, 29, 1370-1374. Minke, R.; Blackwell, J. J. Mol. Biol. 1978, 120, 167-181. Gardner, K. H.; Blackwell, J. Biopolymers 1975, 14, 1581-1595. Kurita, K. Prog. Polym. Sci 2001, 26, 1921-1971.

Permeability Studies of Electrospun Chitin and Chitosan Nanofibrous Membranes

71

[15] Lavall, R. L.; Assis, O. B. G.; Campana-Filho, S. P. Bioresour. Technol. 2007, 98, 2465-2472. [16] Kumar, M. N. V. R. React. Funct. Polym. 2000, 46, 1-27. [17] Min, B. M.; Lee, S. W.; Lim, J. N.; You, Y.; Lee, T. S.; Kang, P. H.; Park, W. H. Polymer 2004, 45, 7137-7142. [18] Noh, H. K.; Lee, S. W.; Kim, J. M.; Oh, J. E.; Kim, K. H.; Chung, C. P.; Choi, S. C.; Park, W. H.; Min, B. M. Biomaterials 2006, 27, 3934-3944. [19] Schiffman, J. D.; Stulga, L. A.; Schauer, C. L. Polym. Eng. Sci. 2009, 49, DOI:1002/pen.21434 [20] Schiffman, J. D.; Schauer, C. L. Polymer Reviews 2008, 48, 317-352. [21] Ohkawa, K.; Cha, D.; Kim, H.; Nishida, A.; Yamamoto, H. Macromol. Rapid Commun. 2004, 25, 1600-1605. [22] Ohkawa, K.; Ken-Ichi Minato; Kumagai, G.; Hayashi, S.; Yamamoto, H. Biomacromolecules 2006, 7, 3291-3294. [23] Schiffman, J. D.; Schauer, C. L. Biomacromolecules 2007, 8, 594-601. [24] Schiffman, J. D.; Schauer, C. L. Biomacromolecules 2007, 8, 2665-2667. [25] Sangsanoh, P.; Supaphol, P. Biomacromolecules 2006, 7, 2710-2714. [26] Matsuda, A.; Kagata, G.; Kino, R.; Tanaka, J. J. Nanosci. Nanotechnol. 2007, 7, 852855. [27] Geng, X.; Kwon, O.-H.; Jang, J. Biomaterials 2005, 26, 5427-5432. [28] De Vrieze, S.; Westbroek, P.; Van Camp, T.; Van Langenhove, L. J. Mater. Sci. 2007, 42, 8029-8034. [29] Taylor, G. Proc. R. Soc. London, Ser. A 1964, 280, 383-397. [30] Hoopes, R. J. In The Design and Application of Controlled Low-strength Materials (flowable Fill); Howard AK, Hitch JL, Conshohocken, 1998. ASTM International: 87102. [31] He, J.-H.; Wan, Y.-Q.; Xu, L. Chaos, Solitons and Fractals 2007, 33, 26-37. [32] Bean, C. P.; Livingston, J. D. J. Appl. Phys. 1959, 30, 120S-129S. [33] Zarkoob, S.; Reneker, R. H.; Eby, R. K.; Hudson, S. D.; Erley, D.; Adams, W. W. U.S. Patent 6,110,590. 2000. [34] Ki, C. S.; Gang, E. H.; Um, I. C.; Park, Y. H. J. Membr. Sci. 2007, 302, 20-26. [35] Bognitzki, M.; Hou, H.; Ishaque, M.; Frese, T.; Hellwig, M.; Schwarte, C.; Schaper, A.; Wendorff, J. H.; Greiner, A. Adv. Mater. 2000, 12, 637-640. [36] Caruso, R. A.; Schattka, J. H.; Greiner, A. Adv. Mater. 2001, 13, 1577-1579. [37] Muller, K.; Quinn, J. F.; Johnston, A. P. R.; Becker, M.; Greiner, A.; Caruso, F. Chem. Mater. 2006, 18, 2397-2403. [38] Gibson, P.; Schreuder-Gibson, H.; Rivin, D. Colloids Surf., A 2001, 187-188, 469–481. [39] McCann, J. T.; Marquez, M.; Xia, Y. J. Am. Chem. Soc. 2005, 128, 1436-1437. [40] Casper, C. L.; Stephens, J. S.; Tassi, N. G.; Chase, D. B.; Rabolt, J. F. Macromolecules 2004, 37, 573-578. [41] Bognitzki, M.; Czado, W.; Frese, T.; Schaper, A.; Hellwig, M.; Steinhart, M.; Greiner, A.; Wendorff, J. H. Adv. Mater. 2001, 13, 70-72. [42] Dayal, P.; Kyu, T. J. Appl. Phys. 2006, 100, 043512 043511-043516. [43] Greiner, A.; Wendorff, Joachim H. Angew. Chem. Int. Ed. 2007, 46, 5670-5703.

In: Nanofibers: Fabrication, Performance, and Applications ISBN: 978-1-60741-947-1 Editor: W. N. Chang ©2009 Nova Science Publishers, Inc.

Chapter 3

NOVEL CHITOSAN–CONTAINING MICRO- AND NANOFIBROUS MATERIALS BY ELECTROSPINNING: PREPARATION AND BIOMEDICAL APPLICATION D. Paneva, М. Ignatova, N. Manolova and I. Rashkova Laboratory of Bioactive Polymers, Institute of Polymers, Bulgarian Academy of Sciences, 1113 Sofia, Bulgaria

ABSTRACT At present increased attention is paid on fibrous materials from the natural polymer chitosan because of its numerous beneficial properties (biocompatibility, biodegradability, inherent antibacterial and haemostatic activity). The presence of both hydroxyl and amino groups enables the tailored modification of chitosan into derivatives having targeted properties. The materials containing chitosan or its derivatives are considered as very promising candidates for versatile applications in medicine, pharmacy, food industry, and agriculture. Nowadays the preparation of nanosized fibrous materials is of special interest because of their unique properties, in particular their high surface area-to-volume and aspect ratios. Electrospinning is a cutting edge technique for fabrication of continuous polymer micro- and nanofibers. The basic principles and the effect of the process parameters on the morphology of the electrospun fibers and fibrous materials are briefly discussed in the present Chapter. The first successful attempt to prepare chitosan-containing electrospun materials dates from 2004. This has been achieved by the addition of a non-ionogenic, water-soluble polymer into the spinning solution. The application of this approach for preparation of chitosan-containing fibers is thoroughly discussed in the Chapter. The preparation of neat chitosan nanofibers by electrospinning is outlined as well. The application of suitable chitosan derivatives soluble in water or low toxic organic solvents enables the design of novel non-woven textiles in absence/presence of a non-ionogenic polymer. The preparation of such nontoxic, environmentally friendly materials is detailed. The applied two-step procedures (heat or UV treatment, use of appropriate crosslinking agents) for imparting waterinsolubility to the obtained micro- and nanofibrous materials are described. The main a

e-mail: [email protected].

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

74

approaches that have been used for preparation of electrospun materials combining the beneficial properties of chitosan and aliphatic polyesters based on poly(L-lactide): simultaneous electrospinning or electrospinning of the polyester, followed by coating of the non-woven textile with a thin chitosan layer, are summarized. Moreover, the recently developed routes for preparation of chitosan-containing micro- and nanofibers, such as reactive electrospinning, combination of electrospinning and polyelectrolyte complex formation as well as yarns formation, are discussed. The advantages of the one-step imparting of water-insolubility of chitosan fibers by reactive electrospinning and polyelectrolyte complex formation as compared to the two-step procedures are emphasized. Last but not least the potential biomedical application of the obtained microand nanofibers are outlined.

LIST OF ABBREVIATIONS αeq (Me)soln (ne)soln

φp

c* ce AFS C CECh Ch Ch-g-oligo(D,L)LA Ch-g-PLLA CMCh DAS DCM DDA DMF DMPA DMSO DSC GА HFIP HMW LbL LMW Me PAA PAAm PAH PAMPS PBS PEC

equilibrium swelling degree entanglement molecular weight in solution entanglement number of the macromolecules in the solution polymer volume fraction in solution critical chain overlap concentration entanglement concentration applied field strength cardboard N-carboxyethylchitosan chitosan chitosan-graft-oligo(D,L-lactic acid) chitosan-graft-poly(L-lactide) carboxymethylchitosan 4,4’-diazidostilbene-2,2’-disulfonic acid disodium salt dichloromethane deacetylation degree dimethylformamide 2,2-dimethoxy-2-phenylacetophenone dimethylsulphoxide differential scanning calorimetry glutaraldehyde 1,1,1,3,3,3-hexafluoro-2-propanol high-molecular-weight layer-by-layer low-molecular-weight entanglement molecular weight in melt poly(acrylic acid) polyacrylamide poly(allylamine hydrochloride) poly(2-acrylamido-2-methylpropanesulphonic acid) phosphate buffer solution polyelectrolyte complex

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 75 PEG PEG-g-Ch PEG-N,O-Ch PEG-N-Ch PEO PHEMA PLA PLDLA PLGA PLLA PP PVA PVP PЕϦ QCh SQ SЕϠ TEGDA TFA XRD ϦHF ϦЕϠ

poly(ethylene glycol) poly(ethylene glycol)-graft-chitosan PEG-N,O-chitosan PEG-N-chitosan poly(ethylene oxide) poly(2-hydroxyethyl methacrylate) poly(lactic acid) poly(L-lactide-co-D,L-lactide) poly(lactide-co-glycolide) poly(L-lactide) polypropylene poly(vinyl alcohol) poly(vinyl pyrrolidone) poly(ethylene terephtalate) quaternized chitosan 7-iodo-8-hydroxyquinoline-5-sulphonic acid scanning electron microscopy triethylene glycol diacrylate trifluoroacetic acid X-ray diffraction analysis tetrahydrofuran transmission electron microscopy

1. INTRODUCTION In the recent years much interest is focused on nanofibrous materials. Because of their inherent large specific surface area and their small pore size nanofibrous materials may find a variety of applications, e.g. in military protective clothing and filter applications, in fuel cells, in drug delivery carriers, cosmetics, in nanosensors (thermal, piezoelectric, biochemical and fluorescence optical chemical sensors), or in electronics. Electrospinning is recognized as the most efficient for producing significant in length polymer fibers with diameters in the nanoscale range. Moreover, the transfer of an electrospinning technology from laboratory to industrial scale can be easily achieved. Special attention is given to the possibility of obtaining 3D-scaffolds for cell and tissue engineering, and wound healing dressings as well. Electrospinning of natural polymers, such as collagen, fibrinogen, chitosan and its derivatives, cellulose and its derivatives, is considered as a very promising method for the development of a new generation of fibrous materials for medical applications. It is expected that these materials should have surface structure and topology mimicking those of natural fibrous materials, e.g. of extracellular matrix. Such structure is propitious for the application of electrospun materials as cell and tissue engineering scaffolds. Amongst the natural polymers, chitosan is the most attracting for use in nanofibrous materials. The interest in preparation of chitosan-based polymer materials is due to the beneficial properties of chitosan in terms of its potential application in the biomedical field. Chitosan exhibits inherent antibacterial and haemostatic activity. Combining

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

76

these properties with the potentialities of the electrospinning can enable the preparation of large variety of materials, e.g. of new generation wound healing dressings of high efficiency. It is worth to be noted that studies on the potential applications of fibrous materials containing chitosan or its derivatives are still in an early stage of development. In addition, solutions have to be found for obtaining composite materials combining chitosan or its derivatives with the widely used in medical practice aliphatic polyesters such as poly(lactic acid) and poly(εcaprolactone). In the present Chapter, after a brief description of basics of electrospinning, the preparation of nanofibrous materials containing chitosan or chitosan derivatives by electrospinning is surveyed. Special attention is focused on the newest trends in the development of electrospinning such as reactive electrospinning as well as polyelectrolyte complex formation during electrospinning, and on the possibilities of biomedical applications of electrospun materials.

2. ELECTROSPINNING Nanofibers can be prepared by different processing techniques such as: 1) template synthesis [1-3], 2) self-assembly [4,5], 3) phase separation [6,7], 4) drawing [8], 5) meltblowing [9,10] and 6) electrospinning [11-14]. Among them electrospinning stands out as the most promising route for fabrication of fibers having diameters within the micro- and the nanoscale and of length reaching tens of meters and more. Although the electrostatic spinning process has been discovered long time ago, electrospinning has gained much interest only by the end of the 20th century.

900 Number of publications

800 700 600 500 400 300 200 100 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Years Figure 1. Number of scientific publications on electrospinning per year for the period 1999-2008 (source: Scopus®; Elsevier B.V).

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 77 The first patents for obtaining polymer fibers by means of electrostatic spinning belong to J. F. Cooley and W. J. Morton [15,16]. Electrospinning is an attractive technique because of its simplicity and the possibility of its facile transfer from laboratory to the industrial scale. The increase of the number of scientific publications on electrospinning in the period from 1999 to 2008 is an indication of the growing interest towards this method (Figure 1). There are USA and Singapore companies that manufacture electrospun non-woven textile [http://www. spinrati.com]. This Section deals in brief with the electrospinning principle and set-up, as well as with the main process parameters affecting the diameter and the morphology of the electrospun micro- and nanofibers.

2.1. Electrospinning Set-Up and Fundamentals The electrospinning process involves application of a high electric field to a polymer solution or melt. A scheme of an electrospinning set-up is shown in Figure 2. There are three basic components: a high voltage supply, a reservoir with a capillary tip for the spinning solution (or melt) and a metallic collector. The polymer solution is delivered through the capillary by means of an appropriate pump. One electrode lead of a high voltage power supply is immersed into the solution or connected to the capillary tip of the reservoir, and the other is connected to the collector. Applying high voltage (between 10 and 50 kV) on the solution induces electric charges. The mutual charge repulsion creates force acting oppositely to the surface tension. As the applied field strength (AFS) is increased, the hemispherical solution surface at the tip of the capillary deforms into a conical shape (Taylor cone) (Figure 2). When the AFS exceeds a threshold value, the repulsive electrostatic force overcomes the surface tension and the charged jet is ejected from the tip of the Taylor cone.

Syringe High voltage supply

Polymer solution

Jet

Collector

Figure 2. A schematic representation of an electrospinning set-up.

78

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

The small jet diameter permits rapid mass exchange and the solvent usually evaporates during its traveling from the capillary to the collector acting as a counter electrode. As a result, charged polymer fiber is deposited on the collector. The fiber charges are gradually neutralized in the environment. The end-product of the process usually consists of randomly deposited fibers (mat) with diameters ranging from nanometers to micrometers. Under certain conditions the charged jet violates its continuity and instead of fibers nano- and/or microparticles of various forms are formed on the collector. This process is called electrospraying and is particularly appropriate for obtaining nano- and/or microparticles. The distance to the counter electrode usually varies between 10 and 25 cm. The substrates used so far for fiber deposition are either the collector of the electrospinning set-up or specific substrates chosen in dependence on the targeted application of the electrospun material. The basic knowledge accumulated in recent years on the jet behavior during electrospinning has been presented in a comprehensive review by Reneker and Yarin [17]. It has been found that jet flight time from the capillary tip to the collector is ca. 0.2 s. Initially the jet only follows a direct path towards the counter electrode. Then it becomes unstable performing a series of bending coils and jet’s radius is continuously enlarged. Investigations with the help of a high-speed digital video camera show that the jet is only one and it moves and bends very quickly (Figure 3).

Figure 3. Stereographic, stroboscopic picture, recorded during electrospinning with a digital video camera, that illustrated the bending paths of the jet (6% solution of poly(ethylene oxide) (PEO) with molar mass 400 000 g/mol, in a mixture of 75% water and 25% ethanol). Reproduced from Reneker and Yarin [17] by permission of Elsevier.

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 79

2.2. Processing Parameters Fiber formation by electrospinning is a complex process affected by a great number of parameters. They are generally divided into: i) solution parameters, ii) process variables, iii) ambient parameters. Solution parameters include polymer type (synthetic or natural, ionogenic or nonionogenic, linear or branched), polymer molar mass, solution characteristics (volatility, dielectric constant, viscosity), and absence/presence of a low-molecular-weight organic or inorganic salt, spinning solution conductivity, surface tension of the solution. Process variables include electric potential at the capillary tip, the gap (distance between the tip and the collector), feeding rate. Ambient parameters include air temperature and humidity. Knowledge and control of these parameters is of great importance for the successful preparation of micro- and nanofibers of polymers with desired morphology. The nature and the molar mass of the polymer hold great significance for the selection of the correct approach to prepare micro- or nanofibers by electrospinning. The successful electrospinning of non-ionogenic synthetic polymers can be performed using various solvents or solvent systems. In the case of ionogenic synthetic and natural polymers the choice of a solvent system is very limited. This imposes the necessity to search for innovative systems for their electrospinning. The solution viscosity is one of the key parameters that affect the formation and morphology of the fibers. Viscosity increase favors the formation of cylindrically shaped fibers of larger and more uniform diameters [18,19]. Vice-versa, at low solution viscosity, fibers having defects along their length are formed [20,21]. With the increase of viscosity the morphology of the obtained fibers is gradually changed: starting from fibers with bead-like defects, through fibers with spindle-like defects to defect-free fibers. SЕϠ micrographs of fibers prepared by electrospinning of poly(L-lactide) (PLLA)/poly(ethylene glycol) (PEG) bicomponent solutions at three polymer concentrations are shown in Figure 4 [22]. There are a number of models describing the effect of the spinning solution concentration on the diameters and the morphology of the prepared fibers [18,19,23,24]. It has been propounded that the obtaining of continuous fibers is feasible when polymer chain entanglements occur in a solution at sufficiently high polymer concentration. The macromolecule entanglements allow the formation of an elastic network of long distance order, thus stabilizing the liquid jet and preventing its disintegration into individual drops.

A

B

C

Figure 4. SEM micrographs of PLLA/PEG fibers (weight ratio PLLA/PEG =70/30): (А) fibers with bead-like defects (concentration 5 wt. %); (B) fibers with spindle-like defects (concentration 7 wt. %) and (C) defect-free fibers (concentration 9 wt. %). Reproduced from Spasova et al. [22] by permission of SAGE.

80

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

In the case of solutions of low concentration and/or by using polymers of low molar masses, the number of macromolecule entanglements is not sufficient and the jet disperses into nano- or microbeads (electrospraying). Models are developed on the basis of the Huggins equation describing the concentration dependence of viscosity for homogenous solutions of a linear polymer: βsp(c)=[β]c + kH([β]c)2 +

(1)

where βsp(c) is the specific viscosity, [β] is the intrinsic viscosity, Ѕ is the polymer concentration and кн is the Huggins coefficient. The dimensionless value obtained by the product of the intrinsic viscosity and the concentration ([β]c), is referred to as the Berry number (Be) [25]. This number is a measure of chain overlap in solution. In case of a very dilute solution of a polymer in a good solvent, the polymer molecules are too remote from each other, they rarely come in contact and the Berry number in such case is less than unity. At higher polymer concentration the individual macromolecules interact and entangle; in such case the Berry number is greater than unity. Linear homopolymers of poly(methyl metacrylate) have been used for more detailed study on the effect of the solution viscosity (in a good solvent, DMF, 25°ϥ) on fiber formation during electrospinning [23]. It has been shown that during electrospinning of dilute poly(methyl metacrylate) solutions (at Ѕ/Ѕ*6 for all polymers with narrow molar mass distribution (Mw/Mn ∼ 1.03-1.35), however for polymers having relatively broader molar mass distribution (Mw/Mn ∼ 1.62 and ∼2.12), defect-free fibers are formed at much higher concentrations, viz. Ѕ/Ѕ*∼9.7 and 10.1, respectively. The authors have also determined the dependence of fiber diameter on concentration [d ∼(Ѕ/Ѕ*)3.1], as well as of fiber diameter on zero shear solution viscosity [d ∼(ηo)0.71]. These scaling relationships are in agreement with the results obtained by McKee et al. [18,26] in the case of electrospinning of poly(ethyleneterephthalate-co-ethylene isophthalate) solutions in chloroform/DMF (70/30 v/v). A semi-empirical method of predicting the transition from electrospraying to electrospinning in a good solvent at concentrations c>c* has been suggested. The entanglement number of the macromolecules in the solution (ne)soln is defined as the ratio between weight-average molar mass of the polymer ( M W ) and the entanglement molecular weight in solution (Me)soln:

(2) where Me is entanglement molecular weight in melt, φp is polymer volume fraction in solution [19]. This analysis shows that fiber formation is initiated at (ne)soln∼2, while the critical polymer concentration necessary for continuous fiber formation (defect-free fibers)

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 81 corresponds to (ne)soln ≥ 3.5. The validity of this method has been confirmed for polystyrene, poly(lactic acid) (PLA), PEO and poly(vinyl pyrrolidone) (PVP) solutions. It should be pointed out that this method is valid only for polymer solutions in good solvents where polymer-polymer interaction can be neglected. As for systems where it is possible that interaction between polymer chains takes place (based on hydrogen bonding, ion interactions, liquid-liquid or solid-liquid phase separation, hydrophobic interactions), intermolecular interactions stabilize physical entanglements. In such cases the concentration at which defectfree nanofibers are obtained is lower than the one calculated on the basis of the semiempirical method of chain entanglement [24,27-29]. The presence of a low-molecular-weight organic or inorganic salt in the spinning solution affects to a high extent the average diameters of the prepared fibers. On increasing the solution conductivity, the density of the jet charges increases. Stronger elongation forces are imposed to the jet due to the repulsion of the excess charges along the jet length thus resulting in formation of fibers with smaller diameters. For instance the addition of an inorganic salt such as KH2PO4, NaH2PO4, and NaCl to the spinning solutions leads to obtaining defect-free fibers from poly(D,L-lactic acid) with smaller diameters [30]. The addition of 0.8 wt. % of the ionogenic organic compound pyridinium formate leads to an increase in the electrical conductivity of dichloromethane (DCM) spinning solutions and to obtaining defect-free PLLA fibers with smaller average diameters [31]. A substantial decrease of fiber average diameter has been observed during electrospinning of PVP in the presence of 4,4’-diazidostilbene-2,2’-disulfonic acid disodium salt (DAS) [32]. The addition of a lowmolecular-weight salt has been applied for preparing nanofibers by electrospinning of poly[(2-dimethylamino)ethyl methacrylate] aqueous solution [33]. The addition of an organic or inorganic salt to the spinning solution [34] may lead to obtaining micro- and nanofibrous bundles (see also Section 6.2). The surface tension of spinning solutions also affects the morphology of the prepared fiber. It has been shown that defect fibers are obtained from solutions with higher surface tension. As mentioned before, in order to form a spinning jet, the surface tension of the solution must be overcome by the electric voltage. As with solution electrical conductivity, the use of additives, even in low concentrations, may affect the surface tension of polymer solutions. Surfactants may play a decisive role in obtaining defect-free fibers. Nanofibers of polystyrene electrospun from DMF/ϦHF mixtures (5-15% w/v) yield defect-free nanofibrous mats only when 0.03-30 mmol/l of the cationic surfactant dodecyl trimethyl ammonium bromide is added to the solution [35]. An important feature in electrospinning is the rapid evaporation of the solvent leading to thinning of the jet. That is why the volatility of the solvent has an impact on the morphology of the obtained fibers. The vapor pressure of the solvent affects the degree and rapidity of its evaporation. Although THF is a good solvent for a number of polymers, and moreover is of high volatility, working with it causes difficulties due to blockage of the capillary tip. In electrospinning of poly(vinyl chloride) from THF, a very broad distribution of fiber diameters is obtained, whereas poly(vinyl chloride) electrospinning from DMF results in a narrow fiber diameter distribution and in an average fiber diameter of about 200 nm. Electrospinning of poly(vinyl chloride) from THF/DMF mixtures leads to decrease of the average fiber diameter with the increase of the DMF amount [36]. Very rapid evaporation of the solvent may therefore impede the formation of fibers with smaller average diameters. Depending on the type of the system, fibers of smaller average diameter can be obtained on increasing the

82

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

solvent vapor pressure. For example, it has been reported that when polystyrene solutions in various solvents are electrospun, the average fiber diameter decreases with increasing boiling point of the solvent [37]. The solvent vapor pressure also influences the fiber surface features, such as porosity. When applying solvents with rapid evaporation such as DCM (vapor pressure 46.6 kPa) PLLA fibers with pores along the fiber length are obtained [38]. The process variables affecting fiber diameter and morphology are: (1) applied electric potential, (2) spinning solution feeding rate and (3) the gap (distance between the tip and the collector). Many authors use AFS, which represents the relation between electric potential and the distance between the capillary tip and the collector. AFS is measured in kV/cm. Applying voltage upon the spinning solution induces electric charges. The increase of the applied voltage causes an increase of the electrostatic forces of repulsion between the individual charges acting on the liquid jet [39-41]. This leads to obtaining fibers with smaller diameters. A decrease in the average fiber diameter with the increase of the applied voltage has been observed in electrospinning of numerous polymers: polyacrylonitrile in DMF [42], polystyrene in THF [43], poly(vinyl alcohol) (PVA) in water [44], chitosan/PEO in a dilute acetic acid solution [45], quaternized chitosan (QCh)/PVA in water [28], DNA in ethanol [46]. There are several reports suggesting that applied voltage has no significant effect on the average fiber diameter. This has been observed during electrospinning of hydroxypropyl cellulose in ethanol or propanol [47], of poly(D,L-lactic acid) in chloroform/acetone (2/1 v/v) [48], of PVA in water [49] and of N-carboxyethylchitosan (CECh)/polyacrylamide (PAAm) in water, of poly(2-acrylamido-2-methylpropanesulphonic acid) (PAMPS)/PVA in water and of P(AMPS-co-acrylic acid)/PVA in water [50]. It has also been reported that average polystyrene fiber diameter increases from 0.31 to 1.72 μm with the increase of applied voltage from 5 to 25 kV in electrospinning of a 12.5-22.5 % w/v polystyrene solution in chloroform [51]. For the electrospinning of 17 wt. % aqueous solutions of PVP or of PVPiodine complex, the average fiber diameters increases from 580 to 640 nm for the PVP system and from 150 to 225 nm for the PVP-iodine system, on increasing the AFS from 0.8 to 1.7 kV/cm [32]. Such discrepancy in experimental observations is probably due to differences in the feeding rate of spinning solution, the gap distances, or the concentrations used in the different studies. Another process variable affecting morphology and diameter of the obtained fibers is the gap distance. Decreasing the gap distance shortens jet flight time, thus decreasing the degree of jet elongation as well as the time available for evaporation of the solvent. Baker et al. [51] demonstrate that when electrospinning a polystyrene solution in chloroform (17.5 wt %) on increasing the gap distance between the capillary tip and the collector from 5 to 25 cm (voltage 15 kV) a decrease of the nanofiber average diameter from 1 to 0.66 μm is observed. Hong et al. [52] have studied the effect of the gap distance on the morphology of PVA nanofibers. It has been shown that as the gap distance decreases, drying of the nanofibers is not sufficiently efficient and they are fused when deposited on the collector. Feeding rate of the spinning solution is another parameter with impact on the fiber diameter and morphology. At low solution feeding rate the electrospinning can be intermittent with the Taylor’s cone being depleted. At increase of solution feeding rate either an increase in the fiber diameter [53] or bead-like defects formation is observed. The effect of solution feeding rate on nanofiber morphology has been discussed in a number of publications [5456].

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 83 Ambient parameters (air temperature and humidity) also affect the fiber morphology. The effect of temperature on the morphology of polyuretanecarbamide fibers obtained by electrospinning of their solutions in DMF has been studied. At higher temperature (70°ϥ) fibers of narrower diameter distribution are obtained, compared to diameter distribution of fibers obtained at room temperature [57]. Average diameter of PVP or cellulose acetate nanofibers [58] is the largest at 20°ϥ, and the fibers obtained at 10°ϥ and at 30°ϥ are with smaller average diameters. This particular temperature impact is explained by the authors by two parameters affected by temperature. The first one is evaporation rate of the solvent decreasing exponentially by the decrease of temperature; consequently, it takes a longer time for the jet to solidify. The second parameter is polymer chain rigidity. At higher temperatures the polymer chains have more freedom of movement, leading to a decrease of the solution viscosity, to a higher jet stretching rate and to thinner fibers. At temperature of 10°ϥ the effect of the first parameter probably predominates over the second one due to the exponential variation of the solvent evaporation rate with temperature. At higher temperature of 30°ϥ the effect of the second parameter is predominant due to the exponential viscosity decrease when temperature increases. The majority of reported data on electrospinning of polymer solution are collected from experiments performed in air. Kim et al. [59] have studied the impact of air relative humidity on the diameter of polystyrene electrospun fibers. It has been shown that with the increase of relative humidity from 10% to 70%, the average diameter of polystyrene electrospun fibers increased from 130 to 380 nm. It has been suggested that the formation of thicker fibers at higher relative humidity is probably due to the fact that the electrostatic charges on the surface of polymer solution are easily discharged at higher humidity level, causing reducing of repulsion forces.

2.3. Morphology and Alignment of Electrospun Fibers Nanofibers prepared by electrospinning are usually monolithic with cylindrical shapes (Figure 5 A) [11,60]. However, more often some deviations are observed, such as obtaining ribbon-like fibers (“nanoribbons”) [61-63], fibers with bead-like defects (Figure 5 B) and spindle-like defects (Figure 5 C) [20,21,63]. Beside the described defects, it is possible to obtain also porous fibers (Figure 6) [64-67]. Under a regime of electrospraying, instead of fibers, beads and structures with various shapes non-connected with the fibers can be formed. In low-viscosity spinning solutions polygonal spheres and other mushroom-like structures can be formed [68], as well as “buns” or “cups” [69,70]. Criteria for the complex evaluation of the morphology and alignment of electrospun fibers have been systematized [63]. The main characteristics needed to evaluate the morphology of defect-free fibers comprise: average fiber diameter, minimal and maximal fiber diameter, standard deviation of the diameter and fiber diameter distribution. The main parameters to be used to characterize the occurring defects (bead-like or spindle-like) along the fiber length have been pointed out. The most frequently used electrospinning set-up permits obtaining mats formed by randomly deposited fibers.

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

84

А

B

C

Figure 5. A sketch of: (А) a defect-free fiber; (B) fiber with bead-like defects and (C) fiber with spindle-like defects. Reproduced from Spasova et al. [63] by permission of SAGE.

Figure 6. SEM micrographs of porous fibers of PLLA, electrospun from DCM at concentration 7 wt. %. Spasova et al. [66], Electrospun chitosan-coated fibers of poly(L-lactide) and poly(Llactide)/poly(ethylene glycol): preparation and characterization, Macromol. Biosci., 2008, 8, 153-162, Copyright Wiley-VCH Verlag GmbH and Co. KGaA. Reproduced with permission.

When applying collectors of special geometry [12,71-73] aligned fibers may be obtained. A schematic representation of some collectors used so far, are shown in Figure 7. One of the collectors used is a rotating cylindrical collector (Figure 7 А and B) [71,74-76] or a rotating thin disc (Figure 7 C) [77-79]. High effectiveness in fiber deposition is achieved when such collectors are used. The low degree of fiber alignment is a disadvantage of this collector type. In order to achieve good fiber alignment static collectors of parallel electrode type are applied (Figure 7 D and E), as well as static collectors with blades placed in line (Figure 7 F) with conductive strips separated by a gap of different width (larger than several centimeters)

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 85 [63,72,80-82]. Often there are difficulties in getting highly aligned nanofibers over a large area of substantial thickness. Fiber alignment is disturbed on increasing the thickness of the material deposited on the collector. The productivity of electrospinning process can be increased by simultaneous use of multiple spinnerets (multiple nozzles) (Figure 7 G) [83].

A

B

C

D

E

F

G Figure 7. Schematic representation of various electrospinning set-ups to obtain aligned fibers (A-F) and for multiple spinnerets (G). Reprinted from IOP, Teo, W., Ramakrishna S. [12], A review on electrospinning design and nanofibre assemblies, Nanotechnology, 2006, 17, 89-106, with permission from IOP.

Multiple spinnerets have also been used to prepare bicomponent and multicomponent blend nanofibrous mats [84,85]. Reported in literature appliances to electrospinning set-ups for obtaining nanofibrous yarns have been discussed in detail in review articles [12,73].

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

86

2.4. Applications Electrospinning offers diverse possibilities for the design of fibrous materials with targeted composition and morphology. It also allows one-step incorporation of additives of different nature into micro- and nanofibers, such as drugs, metallic nanoparticles, carbon materials (carbon nanotubes, fullerenes). New materials that can find application in many different fields may be fabricated using electrospinning. So far electrospinning has been used to obtain new-generation filters for highly efficient gas and liquid filtration [86]; to prepare high-sensitivity nanosensors [87]; in the design of clothing of new generation - extremely light in weight and with desired degree of permeability [88]; to obtain new hybrid fibrous materials combining the useful properties of inorganic nanosized materials and polymer fibers [89]; they are very promising for the development of highly effective heterogenic catalytic systems, also for the needs of optoelectronics; to obtain new materials with application in pharmacy and medicine, for instance as carriers of low-molecular-weight bioactive substances, as cell- and tissue engineering scaffolds, as wound healing dressings [90]. A number of reviews and books have appeared in recent years [90-98], discussing the possibilities of applying the electrospinning process for preparation of new nanomaterials for biomedical applications. In the next Sections the main trends of obtaining nanofibrous materials containing the natural polysaccharide chitosan or its derivatives by electrospinning have been outlined. Results obtained so far on some potential biomedical applications of electrospun fibrous materials are summarized.

3. CHITOSAN – A VERSATILE POLYMER Chitin is a polysaccharide which is the second in abundance after cellulose. It is a basic structural component in the crustaceans’ exoskeleton, in insects’ shells, as well as in the cell walls of some fungi and bacteria [99,100]. Its annual biosynthesis is estimated to approximately 1010 tons [101]. Chitin is a linear polymer with a polymer chain built of poly[β(1→4)-2-acetamido-2-deoxy-D-glucopyranose] (Figure 8A). It is a structural analogue of cellulose however differing from it by the acetamido [-NH(C=O)CH3] group at the C2 carbon atom. Chitosan is obtained by chitin deacetylation and can be regarded as a copolymer built of β(1→4)-bound 2-acetamido-2-deoxy-D-glucopyranose and 2-amino-2-deoxy-Dglucopyranose units (Figure 8B). Chitosan is a basic structural element of some fungi cell walls (Zygomycetes, Aspergillus and Fusarium) [102]. In industrial production of chitosan, chitin deacetylation is usually performed in alkali medium [103]. In recent years biosynthesis methods and the enzymatic deacetylation of chitin are considered as alternative methods of obtaining chitosan. Depending on the chitin type and the hydrolysis conditions chitosan of different molar masses and deacetylation degrees (DDA) are obtained. The molar mass of chitosan may exceed 1000 kDa; the DDA of most of the commercially available products varies from 75% to 95% [103-105].

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 87 OH

NHCOCH3

HO

O

O

NHCOCH3

NHCOCH3

OH

NH2

HO

O

O

HO

OH (A)

OH

O

O O

HO

O

O O

HO

OH

O

NH2

OH

HO

O NHCOCH3

(B) Figure 8. Chain sequence of chitin (A) and of chitosan (B).

It should be noted that chitosan is a natural polymer having behavior of weak polybase (pKa=6.5 [106]) in aqueous solutions, and its solubility highly depends on the medium Єϡ [107]. It is soluble in aqueous medium with Єϡ 5 000 000 g/mol enables the occurrence of sufficient number of chain entanglements by adding a small amount of a non-ionogenic polymer, and continuous fibers can be obtained at chitosan/PEO = 90/10 (w/w) [216]. Solvent mixture of formic acid (HCOOH)/water for electrospinning of chitosan in the presence of PVA has been used [207]. As seen from Ϧable 1, the used up to date PVA is of

92

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

molar mass between 75 000 and 186 000 g/mol. In order to obtain nanofibers enriched in chitosan in this case it is necessary to modulate the main parameters of the spinning solution.

A

B

C

D Figure 10. SEM micrographs of nanofibers electrospun from mixed solutions of chitosan/PEO prepared at AFS 1 kV/cm at weight ratio chitosan/PEO: 0.18; ×1 000 (A), 0.33; ×1 000 (B) (total polymer concentration of 5 wt. %); 1.0; ×5 000 (C), 4.0; ×10 000 (D) (total polymer concentration of 3 wt. %). Reprinted from e-Polymers, Spasova M. et. al., 2004, no. 056 [45], with permission from e-Polymers Foundation.

Table 1. Optimal conditions found for preparation of chitosan-containing nanofibers by electrospinning in the presence of a nonionogenic polymer (the corresponding reference, macromolecular characteristics of chitosan and non-ionogenic polymer as well as the supplier of chitosan are listed in the Table; generally the voltage does not exceed 30 kV, and the gap between the nozzle and the collector is less than 20 cm) Chitosan (molar mass; DDA; supplier)

PEO 600 000 g/mol 80 % Sigma 654 000 g/mol 90 % Qingdao Hisound Biological Engineering (China) 276 000 g/mol 81.7 % Vanson HaloSource (Redmond, WA, USA) 190 000 g/mol 85 % Sigma

85 % Sigma

Non-ionogenic polymer (molar mass)

Solvent system

Chitosan/ Non-ionogenic polymer [w/w]; Total polymer concentration

Mean fiber diameter [nm]

Defects [nm]

Reference

PEO 800 000 g/mol

aq. CH3COOH (2 wt.%)

up to 1/1 3 wt.%

40

Beads-like defects 750/410

[45]

PEO 600 000 g/mol 1 500 000 g/mol 2 300 000 g/mol 4 000 000 g/mol PEO 5 000 000 g/mol

aq. CH3COOH (2 wt.%)

up to 2/1 2–8 wt.%

124±19

-

[206]

Solution formulation Kamterter II, LLC, Lincoln, NE, USA

No data available

-

[223]

PEO 900 000 g/mol

aq. CH3COOH (0.5 M); Triton X-100TM (0.3 wt.%) DMSO (10 %) aq. CH3COOH (3 wt.%) + DMSO CH3COOH/DMSO = 10/1 (w/w)

Nominal concentration of chitosan – 1 wt.% up to 90/10 2.05 wt.%

40

-

[218]

114±19

-

[216]

PEO UHMWPEO > 5 000 000 g/mol

95/5 3 wt. %

Table 1. (Continued) Chitosan (molar mass; DDA; supplier)

PEO 1 400 000 g/mol 87 %; 70 %; 67% Primex Inc. 100 000 g/mol 83 % Sigma 85 % Shen Chiu 90 % Pharmaceutical-grade chitosan was obtained from the Naval Research Laboratory (Washington, DC) 600 000 g/mol; 400 000 g/mol; 148 000 g/mol 75 - 85 % Fluka 1 000 000 g/mol 80 % Primex Inc

Non-ionogenic polymer (molar mass)

Solvent system

Chitosan/ Non-ionogenic polymer [w/w]; Total polymer concentration

Mean fiber diameter [nm]

Defects [nm]

Reference

PEO 900 000 g/mol 300 000 g/mol

aq. CH3COOH

95/5 1.33 wt.% (70 ºC)

80±35

-

[224]

PEO 3 000 000 ~ 5 000 000 g/mol

aq. CH3COOH (1 wt.%)

40/60

ca. 30

Spindle-like defects no data on the sizes

[225]

PEO 600 000 g/mol

aq. CH3COOH (2 wt.%) + CH3OH (4 wt. %)

1/1 3 wt.%

40±9.14

Spindle-like defects no data on the sizes

[226]

PEO 900 000 g/mol

aq. CH3COOH (32 wt.%)

8/9 3.4 wt.%

60±9

-

[211]

PEO 900 000 g/mol

aq. CH3COOH (90 wt.%) + surfactant

3/1 1.6 wt.%

140

-

[227]

Chitosan (molar mass; DDA; supplier)

PVA 210 000 g/mol 78 % 1 300 000 g/mol 77 % Wako Pure Chemical Industries, Ltd., Japan 1 600 000 g/mol 82.5 % Aldrich Low molecular weight 75 – 85 % Aldrich 120 000 g/mol 82.5 % Zhejiang GoldenShell Biochemical Co., Ltd. (Taizhou, China) 78 % Sichuan Biochem-ZX Research Co., Ltd., China 165 000 g/mol 90 % Zhejing Yuhuan Ocean Biochemistry, China

Non-ionogenic polymer (molar mass)

PVA 88 000 g/mol

PVA 124 000 – 186 000 g/mol 87-89% hydrolysed PVA 146 000 – 186 000 g/mol 98-99% hydrolysed PVA 154 000 g/mol 88 % hydrolysed PVA 94 000 g/mol 96 % degree of hydrolysis PVA 80 000 g/mol 98 % degree of hydrolysis

Solvent system

Chitosan/ Non-ionogenic polymer [w/w]; Total polymer concentration

Mean fiber diameter [nm]

Defects [nm]

Reference

HCOOH/H2O

1/1

120

-

[207]

aq. CH3COOH (2 wt.%)

up to 25 % chitosan 6 wt.%

20±5

Spindle like defects no data on the sizes

[219]

aq. CH3COOH (2 wt.%)

5-8 wt.% PVA solution containing 1 wt.% chitosan

160±38

-

[228]

aq. acrylic acid (90 %)

up to 95/5

290

-

[229]

aq. CH3COOH

up to 30/70 7.4 wt.%

125

Spindle-like defects no data on the sizes

[214]

aq. CH3COOH (2 wt.%)

40/60 7%

100±21

Beads-like defects no data on the sizes

[215]

Table 1. (Continued) Chitosan (molar mass; DDA; supplier)

PVA 100 000 g/mol 88 % Zhejiang Golden-Shell Biochemical Co. (Yuhuan, Zhejiang, China) 200 000 g/mol 88 % Zhejiang Golden-Shell Biochemical Co. (Yuhuan, Zhejiang, China) > 10 000 g/mol 100 % water-soluble Hittolife Co. (Kyongki-Do, Korea) Low-viscosity chitosan High-viscosity chitosan Samsung Chitopia Co., Ltd. (Siheung, Korea) PAAm 1 400 000 g/mol (HMW) 80 % Primex 100 000 g/mol (LMW) 70-80 % Sigma

Non-ionogenic polymer (molar mass)

Solvent system

Chitosan/ Non-ionogenic polymer [w/w]; Total polymer concentration

Mean fiber diameter [nm]

Defects [nm]

Reference

260

-

[221]

PVA 170 000 g/mol 88 % degree of hydrolysis

aq. CH3COOH (90 wt.%) + H2O

80/20 7 wt. % chitosan/ acetic acid (90 wt. %) and 10 wt. % PVA aqueous solution

PVA 75 000 g/mol 88 % degree of hydrolysis

aq. CH3COOH (88 wt.%) + hydroxyapatite nanoparticles

90/10 7.3 wt.%

100 - 700

-

[222]

PVA 74 000 g/mol 99.9 % degree of hydrolysis

double distilled water

up to 4/6 12.5 wt.%

~ 200

Beads-like defects

[217]

PVA 44 000 g/mol

HCOOH/H2O AgNO3 or TiO2 nanoparticles

up to 15/85 chitosan was dissolved in formic acid (5 wt %)

290-360

Spindle-like defects no data on the sizes

[220]

aq. CH3COOH (50 wt.%) 25, 40 and 70 ºC

up to 75/25 (25 ºC) up to 90/10 (70 ºC) 1.4 wt.%

50 – 350 depending on chitosan molar mass and spinning temperature

Increasing the temperature the formation of beads-like defects decreases

[230]

PAAm 5 000 000 g/mol

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 97 Until now the studies have aimed at searching for an appropriate solvent that enables the electrospinning of chitosan/PVA system at chitosan/PVA weight ratio higher than 1. The use of dilute acetic acid allows nanofibers from chitosan/PVA to be formed at weight ratio between the polymer partners up to 40/60 [214,215,219]. The preparation of chitosan/PVA fibers at weight ratio 85/15 in HCOOH/H2O solvent has been reported [220]. Using 90 % aqueous solution of acetic acid enables the successful preparation of chitosan/PVA fibers at weight ratio of 80/20 and 90/10 [221,222]. It has been found that after immersion in NaOH aq. solution bath only fibers composed of chitosan/PEO= 90/10 (w/w) preserve their integrity [218]. The electrospinning of chitosan oligomers in the presence of PVA has been reported very recently [217]. Continuous and defect-free fibers with average diameter of ca. 200 nm are obtained using aqueous solutions at total polymer concentration 12.5 wt. % and at weight ratio oligomers/PVA = 40/60. The increase of PVA content enhances the tensile strength of the fibers. Zhou et al. [229] apply acrylic acid for preparation of chitosan/PVA mixed solutions. In this system the electrical conductivity is higher in dilute acrylic acid and at higher chitosan concentration. While continuous defect-free fibers up to chitosan/PVA = 60/40 (w/w) are prepared from 4 % aq. solution of acrylic acid, fibers of higher chitosan content (up to 90/10) are obtained when 90 % acrylic acid is used. Membranes of poly(lactide-co-glycolide) (PLGA), chitosan and PVA nanofibers with average diameter of ca. 300 nm have been obtained by using an electrospinning set-up with two spinnerets, allowing simultaneous electrospinning of a solution containing PLGA and a chitosan/PVA solution (weight ratio 60/40 in 2 wt.% acetic acid) [231].

Figure 11. Fiber diameter (FD, left) and bead density (BD, right) of 1.4 wt. % high molecular weight chitosan/high molecular weight PAAm blend fibers at different air temperature. (Error bars represent standard deviation (n = 60 for FD, and n = 3 for BD), letters indicate significant difference at p < 0.05). Reproduced from Desai et al. [230] by permission of Elsevier.

PAAm is another water-soluble non-ionogenic polymer able to form hydrogen bonds with chitosan. PААm has been applied to facilitate the electrospinning of the chitosan derivative CECh [50]. These studies are discussed in Section 5. Later PAAm has been used for the preparation of chitosan-containing nanofibrous materials [231] from solutions enriched in chitosan [chitosan/PAAm = 95/5 (w/w)]. The effect of the spinning solution temperature (25 ºϥ, 41 ºϥ and 70 ºϥ) on the fibers morphology at different weight ratios

98

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

chitosan/PAAm has been evaluated (Figure 11). The temperature increase results in preparation of fibers with significantly smaller number of defects however, with higher average diameters. Phase separation occurs in fibers electrospun from chitosan/PEO = 1/1 (w/w) solutions [206]. While the fibers with larger diameters consist mainly of PEO, those of smaller diameters are enriched in chitosan. For chitosan/PVA pair at chitosan/PVA = 17/83 (w/w), Li et al. have also shown a phase separation between chitosan and the non-ionogenic polymer [219]. This fact has been used to obtain nanoporous fibers by removal of PVA by treatment with NaOH aq. solution. A ϦЕϠ micrograph of the porous chitosan fiber thus obtained is presented on Figure 12. The fibers consist of chitosan as determined gravimetrically as well as by DSC and IR-spectroscopy. At high chitosan content [chitosan/PEO = 90/10 and 95/5 (w/w)] no phase separation occurs and the fiber composition is almost the same as the feed one, i.e. the polymer partners are homogenously distributed in the fibers [216,230].

Figure 12. TEM of NaOH treated (1 M, 12 h) 17/83 chitosan/PVA bicomponent fibers. Reproduced from Li et al. [219] by permission of Elsevier.

The crystallization of the polymers during electrospinning is hampered since the high rate of transition of the liquid jet into dry fibers impedes crystallites formation [30,232] for example in the case of nanofibers electrospun from chitosan/PVA [214] or chitosan/PEO [45] mixed solutions. Recently, the approach of electrospinning of chitosan in the presence of a non-ionogenic polymer has been applied for preparation of hybrid nanofibers containing hydroxyapatite nanoparticles [221,222], titanium dioxide or silver nitrate [220], as well as silver nanoparticles [233]. The incorporation of bioactive inorganic substances gains significant attention because of the possibility additional beneficial properties to be imparted to the electrospun non-woven textile. For instance, nanosized hydroxyapatite aids the bone cells proliferation; thus becoming a beneficial component in the design of scaffolds, as well as of implants for bone tissue regeneration. The preparation of hybrid nanofibrous materials that

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 99 combine chitosan biodegradability and biocompatibility with hydroxyapatite activity to accelerate the bone cell proliferation is of outstanding importance for the development of a new generation scaffolds for cell and tissue engineering. Ag and TiO2 nanoparticles possess a broad spectrum antimicrobial activity. It is found that hybrid nanoparticles consisting of chitosan and Ag nanoparticles display synergistic effect [234]. Thus, the preparation of hybrid nanosized materials from chitosan and Ag nanoparticles is very promising for the design of novel wound healing dressings [234-237]. The electrospinning of a dispersion containing chitosan/PVA/hydroxyapatite nanoparticles in 88 wt. % acetic acid has been demonstrated [222]. However, the fibers obtained are with broad diameter distribution and their values vary from 100 to 700 nm. Besides, the loading with hydroxyapatite in an amount higher than 10 wt. % is hampered because of nanoparticles agglomeration. The growth of the hydroxyapatite mineral phase can take place onto the electrospun scaffolds. In this case the hydroxyapatite growth and its uniform distribution are supported by the presence of carboxyl groups in the polymer matrix and in the incubation solution. Thus, PAA has been added into the mineralizing bath [221] and better results in respect to the attachment and proliferation of mouse fibroblasts on the surface of CECh/PVA scaffolds as compared to chitosan/PVA scaffolds have been obtained. The preparation of hybrid chitosan/PVA nanofibers, containing AgNO3 or TiO2 nanoparticles has been reported [220]. They use formic acid as a solvent. It is difficult to assess the usefulness of these materials since no data have been provided in terms of crosslinking of at least one of the polymer partners. In addition, no data on weight loss or the fiber morphology after their stay in aqueous solution are presented. The concentrated HCOOH is able to reduce silver ions to Ag nanoparticles [238]. This property of formic acid has been used for preparation of hybrid chitosan/PEO or CECh/PEO nanofibers containing in situ synthesized Ag nanoparticles [233]. At polymer weight ratio of 1/1 cylindrical defect-free fibers with average diameter of 100±29 nm have been obtained with Ag nanoparticles content being 10 wt. % in respect to the total polymer weight. It is worth to be noted that the use of HCOOH as a solvent for electrospinning of chitosan and CECh has an additional beneficial feature. It allows the crosslinking of the polysaccharide under the action of a crosslinking agent to be slowed down, giving opportunity the crosslinking and electrospinning to be performed simultaneously for a long period of time. The obtained nanofibers are waterinsoluble. The one-step approaches for imparting water-insolubility of chitosan-containing nanofibers are discussed in Section 6 of the Chapter. Very recently, the preparation of nanofibers containing surfactant micelles by electrospinning of chitosan/PEO blend solutions has been reported [227]. The used spinning solution consists of chitosan and PEO at weight ratio equal to 3/1 (w/w) and the surfactant concentration is chosen in such a manner so as to exceed the critical micelle one. It is claimed that the micelles can serve as carriers of lipophilic substances, such as lipophilic drugs. Bicomponent fibers consisted of chitosan core and PEO sheath have been prepared [239]. For this purpose two spinnerets, one placed in the other, ending with coaxially positioned needles have been used. The inner spinneret contains chitosan solution, and the outer one – PEO solution. Both solutions form a bicomponent pendant drop at the capillary end, resulting in the formation of a bicomponent Taylor cone. It is claimed that the mixing of the two polymers is limited due to the rapid solvent evaporation.

100

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

4.2. Electrospinning of Chitosan Until now a number of studies have been performed on the electrospinning of chitosan alone. As already mentioned, the natural polymer cannot be electrospun alone from dilute or concentrated solutions of formic acid, lactic acid and propionic acid. [207,208,210]. There are data on electrospinning of chitosan from its solutions in concentrated (90 %) acetic acid [70,208]. Here the known up to date on the electrospinning of bare chitosan will be presented. Chitosan (molar mass of 210 000 g/mol) can be electrospun alone using TFA as a solvent [207,240]. The fibers prepared at chitosan concentration equal or less than 6 wt. % have defects along their axis. At a polymer concentration of 7 wt. % the fibers obtained are mainly with average diameter of 490 nm. However, beads-like defects are observed again along their axis. In order to obtain defect-free fibers a mixture of TFA/DCM has been used as a solvent. The optimal chitosan molar mass and chitosan concentrations for preparation of defect-free fibers from chitosan/TFA system depending on chitosan molar mass are summarized in Ϧable 2. Table 2. Concentration of chitosan above which continuous defect-free fibers are prepared using chitosan solutions in TFA depending on the molar mass of chitosan, according to Ohkawa et al. [240] Chitosan molar mass [g/mol] 210 000 1 310 000 1 580 000 1 800 000

Chitosan concentration [wt.%] 8.00 4.25 3.25 2.00

Avarage fiber diameter [nm] 200±24 103±16 83±11 60±22

TFA is the preferred solvent for electrospinning of bare chitosan in subsequently reported studies [210,241-243]. The use of solvents with lower boiling point and lower surface tension as compared to water facilitates the electrospinning of chitosan. According to Torres-Giner et al. [210] the mixed solvent TFA/CH2Cl2 (70:30 v/v) is the most suitable one for electrospinning of chitosan alone. The tendency of defect-free fibers formation is enhanced increasing chitosan molar mass. The combination of the good physico-mechanical properties and degradability of poly(εcaprolactone) and (co)copolymers of PLA with chitosan antibacterial activity is a promising strategy for obtaining new nanofibrous materials suitable for the design of novel wound healing dressings. The use of TFA as a solvent enables the preparation of composite fibers from chitosan or its derivatives and water-insoluble polyesters such as poly(ethylene terephtalate) (PЕϦ) [244], (co)polymers of PLA [245,246] and poly(ε-caprolactone) [247]. Chitosan/PET nanofibers with average diameters from 500 to 800 nm have been electrospun from their mixed solution in TFA [244]. The chitosan content in the materials imparts hydrophilicity and enhances their inhibition activity against pathogenic organisms S. aureus and Klebsiella pneumoniae. Successfully novel composite nanofibrous materials have been prepared by electrospinning of mixed solutions of chitosan and poly(L-lactide-co-D,L-lactide) (PLDLA) in the common solvent TFA/CH2Cl2 [70/30 (v/v)] [246].

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 101

A

B Figure 13. SEM micrographs of mats that have been incubated in S. aureus cell culture (107 cells/mL) for 24 h at 37 °C: PLDLA mat (A), and crosslinked chitosan/PLDLA mat [50/50 (w/w)] (B); magnification: × 2500. Ignatova et al. [246], Electrospun non-woven nanofibrous hybrid mats based on chitosan and PLA for wound-dressing applications, Macromol. Biosci., 2009, 9, 102-111, Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.

Continuous, defect-free and cylindrical chitosan/PLDLA nanofibers with average diameter value of 840 nm are obtained. In order to impart stability of the bicomponent electrospun chitosan/PLDLA nanofibers in aqueous solutions, the protonated by TFA amino groups of chitosan are neutralized under ammonia vapors, and after that the nanofibers are crosslinked with GА vapors. The microbiological screening reveals that the novel chitosan/PLDLA materials are effective to prevent adhesion and to suppress the growth of the Gram-positive bacteria S. aureus and the Gram-negative bacteria E. coli (Figure 13). The fluorinated solvent 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP) is also suitable in terms of electrospinning of chitosan. Composite fibers of diameters 300-500 nm have been prepared from chitosan and poly(ε-caprolactone) by electrospinning of their solution in TFA and HFIP at weight ratios poly(ε-caprolactone)/chitosan equal to 40/60 [247]. Chitosan fibers have been obtained by electrospinning of chitin precursor in HFIP solvent and subsequent deacetylation of the obtained defect-free fibers in a 40 % NaOH aq. solution [248]. No significant changes in the fiber morphology and their average diameters as a result of the deacetylation have been observed.

102

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

Chitosan fibers with average diameter of 130 nm have been prepared by electrospinning of chitosan with molar mass of 106 000 g/mol and DDA of 54 % in 90 % acetic acid [70]. The feasibility of electrospinning when using concentrated acetic acid is attributed to the fact that the surface tension of chitosan solution decreases from 54.6 dyn/cm when 10 % acetic acid solution is used, to 31.5 dyn/cm when 90 % acetic acid is used. Successful preparation of defect-free nanofibers from chitosan with DDA 75 – 85 % from its solutions in 90 % acetic acid has been reported [208]. Attempts to electrospin fibers from chitosan with molar mass of 148 000 g/mol and DDA of 80 % applying the approach of Geng et al. [70] and of De Vrieze et al. [208] have been made [211]. The inability to obtain fibers has been explained by the differences in the DDA values of the used chitosan. It is worth to be noted that the three research groups have performed their studies using chitosan purchased by different suppliers. This is an additional indication that the physico-chemical characteristics of chitosan have a remarkable impact on its behavior in terms of the electrospinning process. The chitosan nanofibrous materials electrospun in absence/presence of a non-ionogenic polymer are soluble in body fluids. Thus, the development of suitable and easily feasible approaches to render them water-insoluble is of essential importance. One of the routes for imparting water-insolubility of chitosan-containing nanofibrous materials is their stay in alkaline aqueous medium [218,219,243] or their treatment with ammonia vapors [246] where the protonated amino groups are converted into non-protonated ones and chitosan turns into its insoluble in neutral and alkaline medium form. The possibility of imparting waterinsolubility to chitosan fibers obtained by electrospinning from solutions in TFA by immersion of the mats in NaOH or Na2CO3 aq. solutions has been studied [243]. Neutralizing by 5 Ϡ NaOH aq. solution has led to partial preservation of the fiber structure of the nonwoven textile. The use of Na2CO3 instead of NaOH leads to materials that preserve their fibrous structure even after a long (12 weeks) stay in phosphate buffer solution (Єϡ = 7.4) or in distilled water. Crosslinking of chitosan nanofibers under GA vapors has been applied [241] and fibers that preserve their morphology after contact with aq. medium have been obtained. This route for imparting water-insolubility to chitosan fibrous materials is applied both for chitosan fibers [66,221,222,226,231,233,246], and for fibers from chitosan derivatives (particular examples are given in Section 5 of the Chapter). In this case chitosan crosslinking occurs at a second stage after the electrospinning. One-step preparation of chitosan fibers crosslinked with GA has been reported as well, and is discussed in details in Section 6 of the Chapter. Genipin is a very attractive crosslinking agent of chitosan (Figure 6) [249,250]. This is a natural product obtained from geniposide by means of enzyme hydrolysis under the action of ί-glucosidase. Geniposide is isolated from the fruits of Genipa americana (South America) and Gardenia jasminoides Ellis (Asia), where its content is in the range 4-6 %. Genipin has found its application as a crosslinking agent of natural polymers having primary amino groups, such as chitosan and proteins (serum albumin, gelatin, fibrinogen) [251]. Because of its natural origin genipin is a preferred agent for obtaining crosslinked natural polymers for the design of new polymer devices for biomedical applications. It has been reported that two reactions that proceed at a different rate are responsible for chitosan crosslinking. The more rapid reaction is a nucleophilic attack on genipin in position ϥ3 (Figure 14) by the primary chitosan amino groups leading to the formation of a genipin heterocyclic compound linked to the glucosamine residue of chitosan. The second, the slower reaction, is a nucleophilic

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 103 substitution of the ester group at ϥ11 (Figure 14), accompanied by release of methanol and formation of amide bond between genipin and chitosan. Dinan et al. [247] have used genipin for crosslinking of electrospun nanofibrous materials from chitosan and poly(ε-caprolactone). For this purpose the electrospun materials has been kept in 1 wt. % of genipin aq. solution for 24 h. 11

COOCH3

6 7

5 8

HOH2C

10

9

4 1

3

O

OH

Figure 14. Genipin formula.

Table 3. Reported data on the biocompatibility of electrospun chitosan-containing nanofibrous non-woven textiles in respect to tissue cells Chitosan-containing electrospun non-woven textile Chitosan/PEO Chitosan/PEO PLGA-chitosan/PVA membranes Chitosan/poly(ε-caprolactone) Chitosan/PVA/HA Chitosan Chitosan/PET Chitosan/PEO

Cell Type

Reference

canine chondrocytes chondrocytes (HTB-94) osteoblasts (MG-63) human embryo skin fibroblasts (hESFs) Schwann cells mouse fibroblasts (L929) Schwann cells NIH 313 fibroblasts osteosarcoma cells (MG-63)

[223] [218] [231] [247] [222] [243] [244] [225]

The combination of the biological activity of chitosan and the high surface-to-volume ratio of the electrospun materials is an excellent prerequisite for preparation of new generation materials that can find application as wound healing dressings, as well as in the cell and tissue engineering [210,220,244,246]. The ability of chitosan-containing nanofibers to serve as drug carriers has also been demonstrated [45]. Owing to the appropriate behavior of chitosan in respect to attachment, proliferation and viability of tissue cells, it is regarded as very suitable for obtaining scaffolds for cell and tissue engineering [252;253]. It is known that chitosan can stimulate proliferation of cells such as chondrocytes, osteoblasts, fibroblasts. This is the reason most of the studies on the biological activity of chitosan-containing nanofibrous materials to be performed using these type of cells (Ϧаble 3). The obtained up to date results are highly encouraging and reveal the great potential of these new materials as scaffolds for cell and tissue engineering. It is worth to be noted that the

104

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

obtained results are preliminary ones and thorough research on the possibility chitosan nanofibrous materials to be used for such kind of applications has to be performed.

4.3. Chitosan-Coated Nanofibers As evident from the data presented above, in recent years there is a growing interest towards combining the good physico-mechanical properties and biocompatibility of polyesters such as (co)polymers of lactides, poly(ε-caprolactone) and PET, with chitosan antibacterial activity [231,244-247]. This has been achieved by using an electrospinning setup with two spinnerets for simultaneous electrospinning of a polyester solution and a chitosan/PVA solution [231] or by electrospinning of a common solution of polyester and chitosan using TFA as a solvent or mixture of TFA and CH2Cl2 or HFIP [244-247]. Recently, an original and elegant two-step approach for combining the beneficial properties of polyesters and chitosan has been used [66]. The new strategy applied for preparation of the composite fibrous materials is based on the formation of a thin chitosan coating on electrospun PLLA and PLLA/PEG fibers. It is known that proteins and pathogenic microorganism cells adhere onto PLLA based materials due to the hydrophobic nature of the latter [254]. This fact somewhat limits the potentialities of these materials as implants and wound healing dressings. A widely used approach to diversify their application is their physical or chemical modification with PEG. The polymer products thus obtained are characterized by a higher hydrophilicity which is a prerequisite for decreasing the undesired cells and proteins adhesion. The incorporation of PEG in PLLA fibers during the electrospinning process allows hydrophilicity to the obtained materials to be imparted [22]. It is expected that significantly less blood cells and pathogenic microorganisms would adhere onto PLLA/PEG fibers. These materials are not able to inhibit pathogenic microorganisms’ growth and incorporation of appropriate drugs into them is needed. Based on the knowledge on chitosan inherent bactericidal activity, it has been suggested that the formation of a thin chitosan film on PLLA and PLLA/PEG fibers would lead to obtaining new composite materials [66]. Such materials should combine the physico-mechanical properties of the polyester fibrous materials and the antibacterial and haemostatic activity of chitosan. As mentioned in Section 3 chitosan degrades under the action of enzymes such as lysozyme which is found in the blood serum [255]. Under the action of these enzymes, after a certain period of contact with blood the integrity of the chitosan coating will be disrupted. It has been demonstrated that PLLA and PLLA/PEG electrospun materials are appropriate scaffolds for cell proliferation and tissue regeneration [22,256,257]. That is why it is expected that after the degradation of chitosan coating the polyester-containing mats would serve as a scaffold for regeneration of the injured tissue. Applying the described strategy studies have been performed on the possibility of obtaining hybrid fibers from PLLA, PEG and chitosan as well as on their behavior in contact with blood cells and with the pathogenic microorganism S. aureus [66]. A triple chitosan coating of ca. 20 ± 2 nm thickness has been obtained by immersion of PLLA or PLLA/PEG non-woven textile in 0.05 wt. % chitosan solution. The coating has been crosslinked by treatment with GA vapors. In order to evaluate the interaction of the obtained hybrid materials with blood cells the uncoated and the coated with crosslinked chitosan mats from PLLA and PLLA/PEG have been put in contact with whole human blood

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 105 for an hour. Figure 15 shows SEM micrographs of fibrous materials after contact with whole blood.

A

B

C

D

Figure 15. SEM micrographs of PLLA mat (A and B) and PLLA/PEG (C and D) after 1 h contact with whole blood: pristine (A and C) and triple-coated with chitosan (crosslinked) (B and D). Spasova et al. [66], Electrospun chitosan-coated fibers of poly(L-lactide) and poly(L-lactide)/poly(ethylene glycol): preparation and characterization, Macromol. Biosci., 2008, 8, 153-162, Copyright Wiley-VCH Verlag GmbH and Co. KGaA. Reproduced with permission.

PLLA mats show no haemostatic activity. Only single erythrocytes with preserved morphology are observed on the surface of PLLA mats (Figure 15А). Most probably their presence is due to mechanical attachment to the highly porous structure of the mat. High haemostatic activity is displayed by mats with triple chitosan coating. In this case agglutination, deformation and aggregation of the erythrocytes are observed (Figure 15B). Figures 15C and D show SEM micrographs of uncoated and triple-coated with crosslinked chitosan PLLA/PEG fibrous mats after contact with blood. The presence of 30 wt. % PEG in the mats affects the interaction of chitosan coated materials with the blood cells. On the surface of PLLA/PEG mats single erythrocytes can be detected (Figure 15C); the blood cells adhered onto the surface of the coated mats preserve their specific morphology and are not deformed (Figure 15D). The erythrocytes number is less than 5 cells per 1 000 μm2 on triplecoated mat is 55.

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

106

A

B

C

D

Figure 16. SEM micrographs of S. aureus cells adhered into PLLA: pristine (A) and triple-coated with chitosan (crosslinked) (B); and onto PLLA/PEG (70/30): pristine (C) and triple-coated with chitosan (crosslinked) (D). Spasova et al. [66], Electrospun chitosan-coated fibers of poly(L-lactide) and poly(Llactide)/poly(ethylene glycol): preparation and characterization, Macromol. Biosci., 2008, 8, 153-162, Copyright Wiley-VCH Verlag GmbH and Co. KGaA. Reproduced with permission.

Compared to a bare PLLA mat, the erythrocyte number on the triple-coated with chitosan mat is approx. twofold higher - 130. This is an indication that PEG reduces the blood cell adhesion. In contact with the pathogenic microorganism S. aureus, a significant number of cells (more than 200 cells/100 μm2) adhere onto the PLLA mat surface (Figure 16). The increased hydrophilicity of PEG-containing fibers leads to some decrease of the number of adhered cells, and chitosan coating on the fibers leads to a substantial decrease of the number of the adhered cells (ca. 5 cells/100 μm2). The inhibition of the pathogenic cells adhesion combined with the haemostatic activity of the chitosan coated PLLA and PLLA/PEG mats render them very promising materials for application as wound healing dressings.

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 107

5. CHITOSAN DERIVATIVES-CONTAINING NANOFIBERS As mentioned above, electrospinning of chitosan is feasible only from a concentrated acid solutions – TFA [207, 210, 241-243] and acetic acid [70,208] or from dilute acid solutions in the presence of a non-ionogenic water-soluble polymer with sufficiently high molar mass [45,206,218,219]. A number of chitosan derivatives are water-soluble at Єϡ ≤ 7. Such are those obtained by carboxymethylation [258] or carboxyethylation [162], sulfonation [259] and quaternization [144]. Solubility in organic solvents is easily attained by acylation of chitosan [260]. The chitosan derivatives obtained by PEGylation [261] are soluble both in water and in organic solvents. The synthesis of chitosan derivatives soluble in water or in easily volatile low-toxic organic solvents may allow the preparation of fibers from acid-free solutions. Thus, the production of nanofibrous materials from chitosan derivatives will be rendered friendly for the environment, and it will offer additional advantages for the biological and biomedical applications of the nanofibers. In this Section the existing up to now studies aimed at preparation of chitosan derivatives - containing micro- and nanofibers will be surveyed. The approaches applied to stabilize chitosan derivatives-containing fibers against dissolving in water and their potential biomedical applications will also be discussed.

5.1. Electrospinning of Chitosan Derivatives in the Presence of Synthetic Polymers QCh derivatives [144], carboxymethylchitosan (CMCh) [160, 258, 262] and CECh [162] (Figure 17) are easily synthesized and purified. Unlike chitosan which is only soluble in acidic medium with Єϡ6.5. Their biological properties are similar to those of chitosan – they are non-toxic, biocompatible and biodegradable polymers and they can find application in medicine and pharmacy [144,162,263-268]. QCh derivatives have shown higher antibacterial activity, broader spectrum of activity, and higher killing rate as compared to those of chitosan [144,265,266], and thus are potential candidates for design of biologically active wound dressings of new generation that actively take part in the wound healing process. Carboxyalkylated chitosan derivatives attract attention because of their antioxidant and anti-tumor effects. It has been reported that the biodegradability of these water-soluble chitosan derivatives is even higher than that of chitosan [117]. No fibers are obtained in the case of electrospinning of concentrated aqueous solutions of the polyelectrolyte QCh and of the polyampholyte-polyzwitterions CMCh and CECh [28,32,50,82,269,270]. This might be due to the fact that the repulsive forces between ionogenic groups within polymer backbone interfere with the formation of an elastic network of polymer chains of long distance order and impede the formation of continuous fibers. Continuous QCh-containing fibers are formed only when mixed aqueous solutions of QCh and a non-ionogenic water-soluble polymer – PVA or PVP are electrospun [28,32]. It has been shown that the found optimal conditions for electrospinning of QCh/PVA system are: total polymer concentration 8 wt. % and AFS values from 1.5 to 3.5 kV/cm, and for QCh/PVP system - total polymer concentration 20 wt. % and AFS values from 1.6 to 2.8 kV/cm.The average QCh/PVA fiber diameters are in the range of 60 - 200 nm [28]. The effect

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

108

of the composition and properties of spinning solutions and of AFS on the morphology and the average diameters of the prepared fibers has been studied. With the increase of the ionogenic polymer QCh content a decrease of the average diameter of the nanofibers, as well as an increase in the number of spindle-like defects is observed. This change in the nanofiber morphology might be explained with the increase in the charge density and with increase of the solution conductivity on increasing the polyelectrolyte content. It has been shown that cylindrically shaped and defect-free nanofibers are formed by electrospinning of mixed aqueous solutions at weight ratio of QCh/PVA = 1/4 (Figure 18А). Spindle-like defects along the fibers are observed at weight ratio of QCh/PVA (w/w) = 2/3, 1/1 and 3/2 (Figure 18B). In all these cases (ne)soln for PVA is significantly lower than 3.5, estimated by the semi-empirical method, proposed by Shenoy et al. [19,24]. Nanofiber feasibility under such conditions is explained with the ability of PVA solutions to undergo physical gelation due to the presence of inter- and intramolecular hydrogen bonding. Similar effect has also been observed by other authors in cases of electrospinning of aqueous PVA solutions [19,62]. At weight ratio of QCh/PVA = 4/1 and higher only non-connected with the fibers “tailed” beads are observed. R4 R R3 + 3_ N I

R1O O

OR2 O R1O

O

R1O O

O

NHR O

NHR

OR2

OR2

R = H(14%), C(O)CH3 (20%); R1 = H (87%), CH3 (13%); R2 = H(12%), CH3 (88%); R3 = CH3 and R4 = C4H9 (66%)

quaternized chitosan (QCh)

OCH2COOH O HO

OH

OH O

O

NHCH2COOH

HO

NH2

N,O-carboxymethylchitosan (CMCh)

HO

HO

O

NHCH2CH2COOH

N-carboxyethylchitosan (CECh)

OCH2COOH O

O

O

OH O

O

NH2

O-carboxymethylchitosan (CMCh)

HO

O

NHCH2COOH

N-carboxymethylchitosan (CMCh)

Figure 17. Chemical structures of quaternized chitosan derivatives, of carboxymethylchitosan and of carboxyethylchitosan.

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 109

A

B Figure 18. SEM micrographs of the nanofibers prepared from QCh/PVA solutions. Weight ratios QCh/PVA = 1/4 (A) and 1/1 (B). Total polymer concentration 8 wt. %, AFS 2.0 kV/cm, magnification: ×5000. Reproduced from Ignatova et al. [28] by permission of Elsevier.

It has been shown that electrospinning of QCh and PVP aqueous solutions at weight ratios of QCh/PVP (w/w) from 1/4 to 4/1 and total polymer concentration of 20 wt. % results in defect-free and cylindrically-shaped fibers with average diameters in the range from 1500 to 2800 nm (Figure 19) [32]. Both in the QCh/PVA system, and in this case, on increasing the ionogenic component (QCh) content, the fiber diameter significantly decreases and the fiber diameter distribution narrows. The observed changes have been attributed to the increased solution conductivity. The AFS changes have different effect on the fiber morphology for the two systems - QCh/PVA and QCh/PVP. In the case of QCh/PVA, a decrease in the average diameters is observed on increasing the AFS. Conversely, in the case of QCh/PVP, the increase in AFS leads to an increase in the average diameter of the nanofibers. The reasons for this difference remain unknown so far. In order to obtain bicomponent fibers from QCh and PLDLA an appropriate solvent system has been proposed, consisting of DMF/DMSO in volume ratio 60/40, which allows the mixed solutions to be obtained and successfully electrospun [246]. The continuous defectfree hybrid QCh/PLDLA nanofibers (average diameter 280 nm, Figure 20A) are randomly deposited when collected onto a stationary target.

110

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

A

B

Figure 19. SEM micrographs of the fibers prepared from QCh/PVP solutions. Weight ratios QCh/PVP = 2/3 (A) and 4/1 (B). Total polymer concentration 20 wt. %, AFS 2.2 kV/cm, magnification: ×1000. Reproduced from Ignatova et al. [32] by permission of Elsevier.

Partially aligned defect-free nanofibers from the same polymers are formed by electrospinning of QCh/PLDLA solution onto a rotating collection drum (Figure 20B).

A

B

Figure 20. SEM micrographs of the fibers from QCh/PLDLA = 30/70 w/w, collected onto stationary aluminum plate (A) or onto rotating drum (B). Total polymer concentration 5 wt.-% (DMF/DMSO = 60/40 v/v), AFS 1.4 kV·cm-1 and feeding rate of 1.3 mL·h-1. Ignatova et al. [246], Electrospun nonwoven nanofibrous hybrid mats based on chitosan and PLA for wound-dressing applications, Macromol. Biosci., 2009, 9, 102-111, Copyright Wiley-VCH Verlag GmbH and Co. KGaA. Reproduced with permission.

Attempts to electrospin concentrated aqueous solutions of CECh have proved to be unsuccessful. Formation of micro- and/or nanoparticles is observed depending on the АFS values [50,82]. CECh containing nanofibers can be prepared from its aqueous solutions in the presence of non-ionogenic water-soluble polymers with flexible chains, such as PААm [50] and PVA [50, 82]. CECh/PААm nanofibers obtained at different weight ratios of both polymers, are cylindrically-shaped and with average diameters in the range from 50 to 215 nm [50]. For the CECh/PААm system the composition of the spinning solution has a significant effect on the nanofiber morphology and the average fiber diameter. The decrease of the solution viscosity and the increase of the solution conductivity at higher CECh content favor the formation of nanofibers with smaller average diameter. Continuous defect-free

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 111 nanofibers (Figure 21А) are obtained by electrospinning of solutions with CECh content less than 50%. At CECh content 50% spindle-like defects along the nanofibers are formed (Figure 21ϥ). With the increase of the CECh content the number of defects increases and the average distance between two neighboring defects along the nanofiber is decreased (Figure 21ϥ). The formation of “tailed” beads non-connected with the fibers is observed in the case of nine-fold weight excess of CECh.

A

B

C

Figure 21. Effect of the composition of the spinning solution on the morphology of CECh/PAAm nanofibers. Weight ratio CECh/PAAm=1/4 (A), 1/1 (B) and 4/1 (C). Total polymer concentration 3%, AFS 1.1kV/cm. Reproduced from Mincheva et al. [50] by permission of SAGE.

The effect of the composition of spinning solution and the AFS on the fiber morphology and the average fiber diameter has also been studied in the case of the CECh/PVА system [82]. It has been shown that the average diameters of nanofibers decrease, and the shape of the PVАrichest fibers (weight ratio CECh/PVА = 1/75) changes from ribbon-like (Figure 22А) to cylindrical when the AFS value increases [82]. The ratio of polymer partners substantially affects the shape and the average diameters of CECh/PVA nanofibers [82]. The nanofibers of higher CECh content (weight ratio CECh/PVА = 1/8; 1/4; 1/3; 1/2 and 1/1) are cylindrical (Figure 22В). On increasing the CECh content in spinning solutions (at AFS = 1.6 kV/cm) the average nanofiber diameter decreases – e.g., from 420 nm to 100 nm in the case of CECh/PVA = 1/75 and 1/2, respectively. Both with the CECh/PААm system and in this case, on increasing the CECh content in spinning solutions, an increase of the amount of spindlelike defects is observed, as well as a decrease of the distance between two defects and narrowing of the fiber diameter distribution.

А

B

C

Figure 22. Effect of the composition of the spinning solution on the morphology of CECh/PVA nanofibers. Weight ratio CECh/PVA = 1/75 (A), 1/8 (B) and 1/2 (C); total polymer concentration 9.5%, AFS 1.2 kV/cm. Reproduced from Mincheva et al. [82] by permission of Elsevier.

In the same publication different types of collectors have been used (Figure 23) in order to obtain CECh/PVA nanofibers aligned in one or two directions [82]. Nanofibers, aligned in

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

112

one direction, have been obtained by using collector of the А1 type, consisting of two conductive strips. The preparation of transversely aligned nanofibers in two directions has been achieved by the consecutive connection of the conductive strips’ pairs 1а – 1b and 2а – 2b in collectors of А2 and А3 types (Figure 23), separated by an insulating material (polypropylene - PP, quartz – SiO2 or cardboard - C).

Figure 23. Schematic representation of the collectors used for aligned CECh/PVA nanofiber preparation. Reproduced from Mincheva et al. [82] by permission of Elsevier.

It has been shown that the degree of fiber alignment depends on the number and the configuration of the conductive strips-collector type (Figure 24А-C), as well as on the type of the insulating material used (Figure 24D-F) [82].

A) А1 ( d = 300 nm,

θ

= 15º)

B) А2 ( d = 280 nm,

θ

= 12º)

C) А3 ( d = 230 nm,

D) PP ( d = 760 nm,

θ

= 23º)

E) SiO2 ( d = 550 nm,

θ

= 17º)

F) C ( d = 300 nm,

θ

θ

= 10º)

= 15º)

θ is the angle at which the individual fiber deviates from the targeted direction, AFS 1.6 kV/cm, duration of electrospinning process 30 min. Reproduced from Mincheva et al. [82] by permission of Elsevier.

Figure 24. SEM micrographs of aligned CECh/PVA fibers, obtained on the different types of collectors.

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 113 Zhou et al. have followed the preparation of bicomponent CECh/PVA nanofibers from their aqueous solutions [269]. In agreement with the results obtained by Mincheva et al. [82] they have found that the morphology and diameter of nanofibers depend on the weight ratio between the polymer partners. Based on the DSC and XRD analyses it has been assumed that the crystalline microstructure of PVA in the fibers is not well developed. This has been attributed to the stretching of polymer chains in the process of fiber formation. As a result of the rapid solvent evaporation the polymer chains preserve such a strongly extended structure and majority of the chains are in the noncrystalline state. Nanofibers from CMCh have been electrospun with adding of water-soluble polymers – PEO, PAA, PAAm and PVA, to the spinning solution [270]. In the case of CMCh/PEO nanofibers with 30% CMCh can be obtained with average diameter being mostly 300 nm. Their shape is not a cylindrical one, and they merge in the fiber crossover points. Mixing CMCh with either PAAm or PAA enables the formation of nanofibers with a higher CMCh content of 50%, however the fibers contain a considerable number of bead-like defects. The most efficient formation of CMCh/PVA fibers is observed when using PVA, where the CMCh content may reach 80%. It has been found that continuous cylindrical defect-free CMCh/PVА nanofibers are obtained by electrospinning of solutions containing CMCh ≤ 50%. Moreover, the average fiber diameters decrease slightly from 210 to 170 nm on increasing the CMCh content from 20% to 50%.

5.2. Electrospinning of Chitosan Derivatives Chemical modification of chitosan into derivatives that are soluble in a wide variety of organic solvents is an alternative approach to facilitating chitosan electrospinning. Hexanoyl chitosan (Figure 25) is soluble in various common organic solvents and exhibits good blood compatibility [271], antithrombogenic activity and resistance to hydrolysis by lysozyme [260,271,272]; thus could be useful for biomedical applications.

OCO(CH2)4CH3 O O OC

O

N OC

CO

(CH2)4 (CH2)4 (CH2)4 CH3

CH3

Figure 25. Chemical structure of hexanoyl chitosan.

CH3

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

114

Continuous defect-free microfibers from hexanoyl chitosan with ribbon-like morphology have been prepared by electrospinning of hexanoyl chitosan solutions in chloroform at polymer concentration 14% w/v (Figure 26A) [260].

A

B

C

Figure 26. SEM micrographs of fibers prepared from: (A) 14% w/v hexanoyl chitosan solution in chloroform, (B) 8% w/v hexanoyl chitosan solution in chloroform without addition of pyridinium formate salt, and (C) with 7.5% w/v pyridinium formate salt addition. AFS 1.0 kV/cm, magnification: ×400. Reproduced from Neamnark et al. [260] by permission of Elsevier.

The electrospun fibers have average diameters in the range of 0.64 - 3.93 μm. It has been found that on increasing the hexanoyl chitosan concentration, the fiber diameters increase and the amount of bead-like defects decrease (Figure 26). The addition of an organic salt, pyridinium formate, to the spinning solution leads to increase of electrical conductivity of the spinning solution, which results in increase of the fiber diameter and and to a smaller number of defects (Figure 26). In addition to using chloroform as a solvent, DCM and THF have been used to electrospin hexanoyl chitosan and hexanoyl chitosan/PLA [273]. The electrospun hexanoyl chitosan fibers are ribbon-like, with smooth surface and large average diameters (0.91 µm and 0.50 µm from chloroform and DCM, respectively from 10% (w/v) hexanoyl chitosan solution). The use of THF solvent leads to low productivity. Fibers from hexanoyl chitosan/PLA obtained at different weight ratios hexanoyl chitosan/PLA in chloroform with the hexanoyl chitosan solution contents of less than or equal to 50% (w/w) are defect-free fibers with rough surface. The diameters of these fibers decrease with the increase of the hexanoyl chitosan content at a constant AFS value (1.06 kV/cm). When electrospinning hexanoyl chitosan/PLA mixed solutions in DCM, with varying the weight ratios between partners from 20/80 to 80/20, fibers with bead-like defects are obtained. The difference observed in the morphology of the hexanoyl chitosan/PLA fibers when applying different solvents in electrospinning has been attributed to the substantially lower viscosities of the mixed hexanoyl chitosan/PLA solutions in DCM than those in chloroform. The average diameter of the fibers obtained from mixture of 10% (w/v) hexanoyl chitosan solution and 24% (w/v) PLA solution in chloroform are in the range of 0.20 - 1.26 µm, while that from mixture of 10% (w/v) hexanoyl chitosan solution and 24% (w/v) PLA solution in DCM are in the range of 0.74 - 1.26 µm. PEG has good solubility in both water and organic solvents and possesses low toxicity, good biocompatibility and biodegradability [274]; and it finds a wide variety of applications in food, cosmetics and pharmaceutical industry [275].

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 115

OH O HO

O

NH CH2CH2(OCH2CH2)mOCH3

PEG-N-Ch OOCR O

O

NHCOR

RCOO

R = (OCH2CH2O)n OCH3

PEG-N,O-Ch Figiure 27. Chemical structures of PEGylated chitosans.

Because of these properties PEG chains have been grafted onto chitosan. PEGylated chitosans - PEG-N-chitosan (PEG-N-Ch) and PEG-N,O-chitosan (PEG-N,O-Ch) (Figure 27) have been synthesized via reductive amination and acylation of chitosan, respectively [261]. The electrospinning of solutions of PEGylated chitosan in distilled water has failed and only beads have been obtained [261]. In electrospinning of a 25% PEG-N,O-Ch solution in DMF fibers intermixed with beads-like defect are formed. To improve both the efficiency of fiber formation and fiber uniformity, it is necessary to add a cosolvent and a nonionic surfactant to the spinning solutions. Thus, the electrospinning of 15% PEG-N,O-Ch from 75/25 (v/v) THF/DMF with 0.5% Triton X-100TM leads to formation of cylindrical continuous defect-free nanofibers with an average diameter of 162 nm. It has been suggested that the improved fiber uniformity is most probably due to the higher number of chain entanglements which is a result of interactions between the chains of the surfactant and the graft PEG chains. In order to improve the mechanical properties of chitosan in its hydrated state with a view to its potential biomedical application, grafting of oligo(D,L-lactic acid) onto chitosan (Ch-goligo(D,L)LA) has been performed by dehydration of chitosan lactate (Figure 28) [176].

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

116

OH

OH

O

O O

HO NH3

O- + HO

O

80 0C

O

HO NH

- H 2O H

O

O

n

O

O Ch-g-oligo(D,L)LA Figure 28. Grafting of oligo(D,L-lactic acid) onto chitosan.

Only beads or beaded fibers have been obtained by electrospinning of a solution of this derivative in aqueous acetic acid using conventional electrospinning technique. Fibers have been produced via an electro-wet-spinning technique using a coagulation bath (Figure 29) [176]. The average diameters of the fibers are in the range 0,100 μm - 3 μm. The mats have various pore sizes ranging from 1 μm to less than 30 μm and different porosities up to 80%. The morphology and size of fibers, as well as their porosity depend on the concentration of the Ch-g-oligo(D,L)LA spinning solutions and the solution composition of the coagulation bath. The tensile strength and Young’s modulus of the obtained fibrous materials from Ch-goligo(D,L)LA in hydrated state are much higher compared to those of chitosan (793.4 ± 26.7 kPa and 18.1 ± 2.2 MPa, against 117.2 ± 28.9 kPa and 1.1± 0.2 MPa for Ch-g-oligo(D,L)LA and chitosan, respectively). Pump High voltage supply (1)

Polymer solution Electrode (+)

High voltage supply (2)

Liquid reservoir Electrode (-)

Figure 29. A schematic representation of an electro-wet-spinning set-up.

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 117 It has to be noted that in literature, most probably due to experimental difficulties, data on the mechanical properties of nanofibrous mats are scarce. Determination of such mat characteristics, especially in the hydrated state is of great interest, mainly with a view to their potential biomedical application. Another approach which keeps the chitosan amino groups unchanged is grafting L-lactide oligomers by ring opening polymerization of L-lactide in the presence of methanesulfonic acid which plays a dual solvent and catalyst role [276] (see Figure 30 for chemical structure of chitosan-g-poly(L-lactide) (Ch-g-PLLA).

OR O RO

O

NH2 OH

R = CO CH O CO CH O CO CH CH3 n CH3 CH3 Figure 30. Chemical structure of chitosan-g-poly(L-lactide) (Ch-g-PLLA).

The possibility to control the side chain length by varying the L-lactide/chitosan ratio allows the manipulation of the biodegradation rate and hydrophilicity of the material. Lowmolecular-weight (LMW) and high-molecular-weight (HMW) chitosan have been used however the molar mass of chitosan has not been specified, so it is difficult to discuss the differences observed in the fiber morphology. Microfibers have been prepared in all cases. Defect-free and ribbon-like- microfibers are formed by electrospinning of HMWCh-g-PLLA from its solution in ethyl acetate at polymer concentration 44 wt. % and of LMWCh-g-PLLA from its solution in 2-butanone at polymer concentration 50 wt. %.

5.3. Two-Step Imparting Water-Insolubility of Chitosan DerivativesContaining Nanofibers The possibilities for biomedical application of the bicomponent electrospun mats from water-soluble chitosan derivatives might be significantly enlarged if they are rendered stable in aqueous environment.

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

118

A

B

C

D

Figure 31. SEM micrographs of the QCh-containing fibers prepared: from QCh/PVA solutions without (А), and in the presence of 1 wt. % DMPA, 1 wt. % ammonium peroxydisulphate and 10.7 wt. % TEGDA (В), weight ratio QCh/PVA = 2/3, AFS 2,0 kV/cm, polymer concentration 10 wt. %; solvent H2O/DMSO=92/8 (w/w), magnification: ×10 000 (A and B); and from QCh/PVP solutions without (C), and in the presence of 1 wt. % DMPA, 4.5 wt. % TEGDA and 1.5 wt % DAS (D), QCh/PVP = 2/3, AFS 2.2 kV/cm, polymer concentration 20 wt. %; solvent H2O/DMSO=92/8 (w/w), magnification: ×2 000 (C and D). Reproduced from Ignatova et al. [28,32] by permission of Elsevier.

In order to retain the unique nano- and microfibrous structure two-step photo-mediated crosslinking of electrospun QCh/PVA and QCh/PVP mats in the solid state has been performed [28,32]. First, QCh/PVA and QCh/PVP solutions containing photo-crosslnking additives have been electrospun, and then UV irradiation of the fibers in the solid state has been performed. It has been shown that adding the photoinitiator 2,2-dimethoxy-2phenylacetophenone (DMPA), ammonium peroxydisulfate and triethylene glycol diacrylate (TEGDA) as crosslinking agent to the QCh/PVA mixed solution does not hamper the fabrication of fibers (Figure 31 A, B) [28]. The nanofiber diameter decreases (from 217 to 116 nm) and the size distribution of the crosslinked nanofibers is narrower on adding crosslinking additives. The observed effects have been attributed to increased conductivity of the solution (from 2.95 mS/cm to 4.3 mS/cm) when the inorganic salt - ammonium peroxydisulfate is added. Similar effects have been reported for other systems [45,50] and are explained by the higher charge density on the surface of ejected jet during electrospinning, thus imposing higher elongation forces to the jet. Cylindrical defect-free QCh/PVP microfibers have been electrospun at total polymer concentration 20 wt. %, using crosslinking agents DAS and TEGDA, and DMPA as

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 119 photoinitiator (Figure 31 C,D) [32]. In this case, due to the significant increase of the solution viscosity (from 8 200 cP to 9 800 cP) when adding the photo-crosslinking agents, fibers with a higher average diameter are obtained (up to 2 900 nm). Increased solution viscosity leads to an increase in the viscoelastic force which counteracts the Coulomb repulsion force that tries to stretch the charged jet, thus resulting in formation of fibers with higher average diameters and in decrease in the number of defects [20,232]. UV-irradiation for 10 h of the electrospun QCh-containing nano- and microfibrous mats containing crosslinking agents and photoinitiator results in stabilizing of the fibers against disintegration in aqueous medium [28,32]. When put in contact with water and water vapor the fibers keep their morphology and do not dissolve (Figure 32). The equilibrium swelling degree of photo-crosslinked QCh-containing micro- and nanofibers, in distilled water at 23 οC is 100% and 67% in the case of QCh/PVA system and QCh/PVP system, respectively [28,32]. In order to stabilize the electrospun bicomponent QCh/PLDLA [246] and CECh/PVA [269] mats against dissolving in water, a two-step process of crosslinking of fibers has been applied. QCh/PLDLA or CECh/PVA spinning solutions are first electrospun into nanofibers followed by covalent crosslinking with GA vapors.

А

B Figure 32. Effect of water on the morphology of the photo-crosslinked QCh/PVA and QCh/PVP fibers. Photo-crosslinked QCh/PVA mat after contact with water for 6 h (A), weight ratio QCh/PVA = 2/3, total polymer concentration 10 wt. % (H2O/DMSO=92/8 w/w), AFS 2,0 kV/cm, magnification: ×10 000. Photo-crosslinked QCh/PVP mat after contact with water for 6 h (B), weight ratio QCh/PVP = 2/3, total polymer concentration 20 wt. % (H2O/DMSO=92/8 w/w), AFS 2,2 kV/cm, magnification: ×2 000. Reproduced from Ignatova et al. [28,32] by permission of Elsevier.

It has been shown that after soaking in aqueous solutions for 10 h, the hybrid crosslinked QCh/PLDLA nanofibrous mats swell to some extent while maintaining their fibrous morphology and retaining their integrity [246] (Figure 33). The equilibrium swelling degree

120

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

(αeq) of crosslinked QCh/PLDLA nanofibers reaches 160% (distilled water, 25 °C). No weight loss has been detected after 24 h stay in water [246].

A

B Figure 33. Effect of water on the morphology of the crosslinked QCh/PLDLA nanofibers (QCh/PLDLA = 30/70 w/w). Non-treated mat crosslinked with GA vapor for 4 h (A), and after contact with water for 10 h (B). Ignatova et al. [246], Electrospun non-woven nanofibrous hybrid mats based on chitosan and PLA for wound-dressing applications, Macromol. Biosci., 2009, 9, 102-111, Copyright Wiley-VCH Verlag GmbH and Co. KGaA. Reproduced with permission.

In order to impart water insolubility to CECh-containing nanofibers with a view to their potential biomedical application, an alternative method of crosslinking CECh/PVA and CECh/PААm nanofibrous materials by heat treatment in solid state below the temperature of softening of the non-ionogenic polymers and below the decomposition temperature of the polyelectrolytes has been reported [50,82]. The conditions for heat treatment have been chosen by taking into account the thermal behavior of the polymer partners. Thus, for example the CECh/PVA mats have been heated at 100°C for 10 h [82], and those prepared from CECh/PААm have been heated at 100, 120 or 150°C for 5 h [50]. SEM analyses of the samples indicate that these temperatures cause no changes in the nanofiber morphology

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 121 [50,82]. The behavior of the nanofibers in contact with water vapor or water is dependent on their composition and on the heating conditions [50,82].

A

B

C

D

Figure 34. SEM micrographs of CECh/PAAm nanofibers. CECh/PAAm=4/1, heat-treated at 120°C for 5 h (A, ×10000) and after subsequent contact with water for 1h (B, ×10000). CECh/PAAm=9/1, heattreated at 120°C for 5 h (C, ×5000) and after subsequent contact with water for 1 h (D, ×5000), AFS 1.2 kV/cm. Reproduced from Mincheva et al. [50] by permission of SAGE.

Unlike the nanofibers from CECh/PAAm=1/1 that have been heat-treated at 120°C for 5 h are partially resistant to water vapor and water, the same treatment of nanofibers enriched in CECh (CECh/PAAm=4/1) leads to a considerable improvement of their resistance to water and water vapor (Figure 34B) [50]. In the case of the system with the highest CECh content (CECh/PAAm=9/1) on contact with water the nanofibers dissolve, while the defects remain unchanged (Figure 34D). This peculiar behavior is attributed to phase-separation during electrospinning at the large CECh excess. For the CECh/PAAm system the crosslinking of the nanofibers is due to interactions between the amino groups and carboxyl groups on CECh. Both with the CECh/PAAm system and in the case of CECh/PVA the resistance of the nanofibers against water depends on the composition of the spinning solution and increases on increasing CECh content [82]. A considerable crosslinking of the mats and retaining the fibrous structure of materials after contact with water for one week has been observed for the heat-treated CECh/PVA nanofibers in weight ratio CECh/PVA = 1/3 (Figure 35D), however the nanofibers with low CECh content (CECh/PVA = 1/8) (Figure 35A) strongly swell after contact with water for one week and the material loses its fibrous structure (Figure 35C).

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

122

A

B

C

D

Figure 35. SEM micrographs of nanofibers containing CECh and PVA at weight ratio CECh/PVA = 1/8 (A and B) and 1/3 (C and D) after heating at 100ºC for 10 h (A and C) and after subsequent contact with water for one week (B and D); A) ×10 000, B) ×10 000, C) ×5 000, D) ×2 500; AFS 1.6 kV/cm. Reproduced from Mincheva et al. [82] by permission of Elsevier.

It has been assumed that in the case of CECh/PVA the OH groups from the PVA chains also participate in the crosslinking reactions. The possible reactions with the participation of the both polymers - CECh and PVA, could proceed - amidation reaction (intra-molecular and inter-molecular) between carboxylic and amino groups of CECh or anhydride bond formation between carboxylic groups of CECh, and esterification reaction between the carboxylic groups of CECh and hydroxyl groups of PVA [277]. It has to be noted that the heat-induced crosslinking of the CECh/PAAm and CECh/PVA nanofibers is attained in the temperature range that is suitable for dry sterilization. This fact is important and it paves the way to shorten the production stages of sterile materials for potential biomedical application. Crosslinked water-resistent CMCh/PVA fibrous mats have been prepared by heatinduced crosslinking at 140 °C for 30 min [270]. The results from determination of weight losses and the SEM analyses on fiber morphology show that the crossliking proceeds more considerably for the electrospun mats containing N,O-CMCh (Mv = 405 kDa, DS = 1.14) with longer chain and higher substitution degree (Figure 36). These mats retain their fibrous structure after immersion in water for 1 h in a greater extent than that with much shorter and less substituted O-CMCh (Mv = 89 kDa, DS = 0.36) (Figure 36). It should be noted that the determined weight losses after dipping the mats in water for 3 h are too high – they reach 35.3% for the electrospun mats containing N,O-CMCh (Mv = 405 kDa, DS = 1.14) and 47.2% for those containing O-CMCh (Mv = 89 kDa, DS = 0.36). Although very important, the high weight losses have not been discussed by the authors.

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 123

A

B

C

D

E

F

Figure 36. SEM micrographs of nanofibers from CMCh and PVA at weight ratio CMCh (Mv = 405 kDa, DS = 1.14)/PVA = 50/50 w/w (A and B) and at weight ratio CMCh (Mv = 89 kDa, DS = 0.36)/PVA = 60/40 w/w (C and D) after heating at 140ºC for 30 min (A and C), and after subsequent contact with water for 1 h (B and D), and for 3h (C and F). Reprinted from IOP, Du, J., Hsieh, Y.-L. [270], Nanofibrous membranes from aqueous electrospinning of carboxymethyl chitosan, Nanotechnology, 2008, 19 (12), art. no. 125707, 1-9, with permission from IOP.

5.4. Biomedical Applications of Chitosan Derivatives-Containing Nanofibers As already mentioned in Section 4, the electrospun fibers of chitosan derivatives are the focus of intense study aimed at their biomedical applications, such as wound dressing materials, tissue engineering scaffolds and controlled drug delivery systems. The high specific surface area and small-size pores of electrospun mats are favorable for the adsorption of body fluids and for preventing bacteria penetration and thus provide good conditions for wound healing. It is of interest to incorporate a hydrophilic non-toxic polymer such as PVA [28], PVP [32], poly(ethylene-co-vinyl alcohol) [27], PEO [45,278-280] in the electrospun mats for wound healing applications. To prepare electrospun nanofibrous mats having woundhealing properties two routes have been reported: a) incorporation of a drug (e.g. antibacterial 8-hydroxyquinoline derivatives [45,50], cefazolin [281], itraconazole [282], heparin [280], or silver nanoparticles [283-288]) in the electrospinning solution, and b) electrospinning of polymers with inherent antibacterial and wound-healing properties, such as chitosan [45], QCh [28,32,246], PVP-iodine complex [32,289], sulfonated poly(vinyl phenol) [290], hyaluronic acid [291], collagen [292] and polyurethane [293]. A characteristic of choice of materials designed for wound dressing is its antimicrobial effect. For obtaining the nanofibrous materials with antimicrobial properties three-component spinning solutions containing CECh, PAAm and 7-iodo-8-hydroxyquinoline-5-sulphonic acid (SQ) - a model ionizable drug with broad-spectrum antmicrobial and antimycotic activity

124

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

have been used [50]. The presence of SQ has led to a more than two-fold decrease in the diameter of the nanofibers (from 200 to 80 nm), to narrow diameter distribution and to the formation of spindle-like defects but no “tailed” beads. The observed effects have been explained by the increase in the conductivity of the solution on adding a low-molecularweight ionizable compound. In order to estimate the activity of these nanofibrous mats containing a low-molecular-weight compound SQ with known antimicrobial and antimycotic activity, against Gram-negative bacteria (E. coli), Gram-positive bacteria (S. aureus) and the fungus C. albicans, a test by measuring the width of the sterile zone around the nanofibrous mat in Petri dishes with agar medium has been applied [50]. Unlike the blank controls of CECh/PAAm mats around which sterile zones have not appeared, well-defined wide sterile zones around the mats, containing SQ, have been observed. The nanofibrous mats containing high-molecular-weight component QCh with known inherent biocide activity that manifests itself during contact between the bioactive agent and the microorganism, are extremely perspective as wound healing materials. A microbiological test consisting of counting the viable bacterial cells [294] has been applied to assess the antibacterial activity of these mats against Gram-positive bacteria S. aureus and Gramnegative bacteria E. coli [28,32,246]. It has been found that the photo-crosslinked electrospun fibers of PVA, PVP and PLDLA do not inhibit the bacteria growth (Figures 37 and 38).

Log (survivors) cells/mL

6 5 4 control cross-linked PVP mat cross-linked QCh/PVP mat cross-linked QCh/PVP film

3 2 1 0

0

20

40

60

80

100

120

100

120

Exposure time, min

A

Log (survivors) cells/mL

6 5 4 3 control cross-linked PVP mat cross-linked QCh/PVP mat cross-linked QCh/PVP film

2 1 0 0

20

40

60

80

Exposure time, min

B Figure 37. Logarithm plot of the viable bacteria cell number versus the exposure time: for crosslinked PVP mats, for crosslinked QCh/PVP electrospun mats and for crosslinked QCh/PVP films prepared by solvent casting method (A, B). The tests have been carried out against S. aureus (A) and against E. coli (B). Reproduced from Ignatova et al. [32] by permission of Elsevier.

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 125

Log (survivors) cells/mL

5

control PDLLA mat Ch/PDLLA mat Ch/PDLLA film QCh/PDLLA mat QCh/PDLLA film

4

3

2

1

0 0

20

40

60

80

100

120

100

120

Exposure time, min

A

Log (survivors) cells/mL

5

4

3

2

control PDLLA mat Ch/PDLLA mat Ch/PDLLA film QCh/PDLLA mat QCh/PDLLA film

1

0 0

20

40

60

80

Exposure time, min

B Figure 38. Logarithm plot of the viable bacteria cell number versus the exposure time: for PLDLA electrospun mats, for crosslinked QCh/PLDLA electrospun mats, for crosslinked QCh/PLDLA films prepared by solvent casting method, for crosslinked Ch/PLA electrospun mats and for crosslinked Ch/PLA films prepared by solvent casting method (A, B). The tests have been carried out against S. aureus (A) and against E. coli (B). Ignatova et al. [246], Electrospun non-woven nanofibrous hybrid mats based on chitosan and PLA for wound-dressing applications, Macromol. Biosci., 2009, 9, 102111, Copyright Wiley-VCH Verlag GmbH and Co. KGaA. Reproduced with permission.

Unlike them, the crosslinked composite electrospun mats containing QCh or Ch are efficient in inhibiting growth of S. aureus and E. coli (Figures 37 and 38). The electrospun QCh/PVA and QCh/PVP photo-mediated crosslinked mats are found to kill S. aureus faster than E. coli at the same concentration of QCh (3000 µg/ml) (Figure 37) [28,32]. The QCh/PLDLA and Ch/PLDLA nanofibrous mats exhibit higher killing rate against S. aureus and E. coli than the solvent cast films with the same composition (Figure 38) [246]. Similar higher antibacterial efficacy of the electrospun mats has also been observed in other systems [289,290] and has been explained in terms of the high specific surface area of the mats which can result in an increased level of contact between the nanofibrous mats with antibacterial properties and the bacteria suspension, hence in higher bacteria kill rate of the electrospun mats. The obtained results show that the antibacterial activity of crosslinked electrospun QCh- and Ch-containing mats is due to the presence of chitosan or its quaternized derivatives and the effect against Gram-positive bacteria S. aureus and Gram-negative bacteria E. coli is mainly bactericidal. The adhesion of bacteria S. aureus on crosslinked QCh/PLDLA mats and on PLDLA mats has also been studied [246]. For this purpose S. аureus has been cultured for 24 h on the surfaces of electrospun crosslinked mats and after immersion into GA solution in a phosphate

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

126

buffer solution (PBS) for cell fixation and subsequent washing with PBS and freeze-drying, the mats have been analyzed by SEM. After 24 h a lot of cells of S. аureus are observed on partially oriented PLDLA nanofibers collected onto rotating drum (Figure 39A). In contrast, no bacteria adhesion is observed on the surface of bicomponent crosslinked QCh/PLDLA mats (Figure 39B).

A

B

Figure 39. SEM micrographs of mats that have been incubated in S. aureus cell culture (107 cells·mL-1) for 24 h at 37 °C. PLDLA mats (A) and crosslinked QCh/PLDLA mat (30/70 w/w) (B), spinning solution in DMF/DMSO = 60/40 (v/v). The mats were collected onto rotating drum. Magnification: × 2500. Ignatova et al. [246], Electrospun non-woven nanofibrous hybrid mats based on chitosan and PLA for wound-dressing applications, Macromol. Biosci., 2009, 9, 102-111, Copyright Wiley-VCH Verlag GmbH and Co. KGaA. Reproduced with permission.

This finding suggests that the tendency for adhesion or proliferation of bacteria has been prevented in great extent on the surfaces of mats containing QCh, which is known with its bactericidal effect. It has been concluded that the combination of adhesion-preventing properties towards pathogenic bacteria S. aureus and high bacteria kill rate render these mats promising candidates for wound healing applications. It is expected that tissue engineering scaffolds based on electrospun nanofibrous mats mimicking the architecture of the extracellular matrix should offer great advantages for tissue engineering. Nanofibrous scaffolds can be not only a substitute or a synthetic substrate for the natural extracellular matrix in the body, but they can provide a three-dimensional environment for better cell adhesion and proliferation stimulating the growing cells to organize into tissue. It has been reported that the fiber diameter may substantially affect the morphology and proliferation of cells grown on the scaffold, whereas nanofibrous scaffolds can facilitate cell attachment and support cell growth [295-297]. Electrospun nanofibrous scaffolds have a high specific surface area favoring protein absorption, thus providing more binding sites to the membrane receptors of the cells. Nanofibrous mats possess exceptionally high porosity which allows the exchange of gases and nutrients to support the growing tissue. The reported studies on the possibilities for use of electrospun nanofibers from chitosan derivatives as scaffolds in tissue engineering are still scarce and however, are directed to seeding the fibrous mats with specific cell types, observing cell attachment and proliferation on them over a period of time, as well as on carrying out cytotoxicity tests.

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 127 In vitro tests using rabbit fibroblasts seeded on chitosan and Ch-g-oligo(D,L)LA nanofibrous scaffolds have shown that the mats exhibit good capability to promote the adhesion and proliferation of fibroblast cells [176]. Moreover, no substantial difference in the size, density or distribution of fibroblasts grown on both materials has been observed. Citotoxicity tests performed using mouse fibroblasts (L929 cell line) on electrospun L-lactide modified chitosan fibers show that the electrospun mats at molar ratio of chitosan/L-lactide = 1/24 are nontoxic to the fibroblasts cells [276]. The potential use of the CECh/PVA electrospun fibrous mats as scaffolds for skin regeneration has been evaluated in vitro using mouse fibroblasts as reference cell lines [269]. Indirect cytotoxicity assessment of the fibrous mats shows that the mats are nontoxic to L929 fibroblast cells and do not release substances harmful to living cells. The L929 cells adhere well on the mat surface and possess a normal morphology. The biocompatibility of hexanoyl chitosan fibrous scaffolds has been assessed in vitro towards human keratinocytes (HaCaT) and human foreskin fibroblasts (HFF) (Figure 40) [298]. Hexanoyl chitosan fibrous scaffolds exhibit higher cell viability than the solvent-cast hexanoyl chitosan films. The electrospun hexanoyl chitosan fibrous scaffolds can support the attachment and the proliferation of both types of cells. In addition, the cells cultured on the hexanoyl chitosan fibrous scaffolds preserve their specific morphology and integrate well with surrounding fibers to form a 3D cellular network. Such matrices might be suitable as tissue engineering scaffolds for skin regeneration.

A

B

Figure 40. SEM micrographs of HaCaT cells that were cultured on hexanoyl chitosan fibrous scaffolds at 24 h (A) and 3 d (B) of cell culturing. Reproduced from Neamnark et al. [298] by permission of Elsevier.

Nanofibrous mats from chitosan derivatives have also shown potential for applications as carriers in controlled drug delivery systems. PLGA/PEG-g-Ch nanofibrous mats loaded with ibuprofen have been obtained by electrospinning aiming at materials suitable for treatment of atrial fibrillation [299]. Two approaches for the incorporation of ibuprofen in the mat have been used: i) dissolving of ibuprofen in the spinning solution where it is electrostatically conjugated to the PEG-g-Ch, and ii) covalent attachment of ibuprofen to the PEG-g-Ch prior to electrospinning. The solubility characteristics of PEG-g-Ch, i.e. solubility in organic solvents and insolubility in water at neutral Єϡ [300] have been considered as an advantage enabling the preparation of tri-component electrospun mats without any need of crosslinking. The electrospun mats are claimed to be mechanically robust and to have capability to conform

128

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

to motions and, thus, are compatible with atrial tissue. The performed studies have shown that ibuprofen release rate can be controlled by the presence of PEG-g-Ch and by the route of ibuprofen incorporation [299]. The release of ibuprofen from electrospun PLGA mats due to simple diffusion is very fast (burst release) however it has been slowed down by adding PEGg-Ch. This change in the release rate has been explained by electrostatic interactions between the dissociated carboxyl groups of ibuprofen and the protonated amino groups of chitosan molecules. Covalent attachment of ibuprofen to PEG-g-Ch has led to prolongation of its release more than two weeks.

6. REACTIVE ELECTROSPINNING OF CHITOSAN AND CHITOSAN DERIVATIVES The term “reactive electrospinning” has been recently introduced [301]. It denotes a new approach which combines electrospinning with a reaction that occurs during the spinning process. The reactive electrospinning is particularly attractive in terms of preparation of crosslinked fibrous materials. Such materials can be prepared applying two-step procedures. However, the concept the crosslinking to occur at a one stage during the electrospinning process is a particularly attractive one. The use of the former approach has allowed the preparation of crosslinked hydrogel fibers [301]. The hydrogel materials from crosslinked water-soluble polymers are considered as highly promising candidates for application in the biomedical field for design of wound healing devices, for controlled drug delivery, as scaffolds for tissue engineering, etc. [302]. It is expected that the combination of the properties of certain hydrogels (such as Єϡ- and temperature sensitivity, biocompatibility) with the high surface-to-volume ratio of the nanofibrous materials would be beneficial for the preparation of new materials with improved behavior. There are few data on successful reactive electrospinning of synthetic polymers. In addition, there are patented methods and original reactive electrospinning set-ups for one-step preparation of crosslinked polymer and polymer-nanocomposite nanofibers by in-line mixing of a polymer spinning solution with a solution containing the crosslinking agent [303]. Reactive azides have been added into the spinning solutions and functionalized PET fibers have been obtained [304]. Then the functionalized fibers have been crosslinked by heating, i.e. this is a two-step crosslinking procedure. Photo-crosslinking during the electrospinning process has been performed by electrospinning of poly(methyl methacrylate-co-2-hydroxyethyl acrylate) functionalized with cinnamoyl chloride [305]. Using a more easily feasible method - irradiation during the electrospinning of solution, nanofibers from crosslinked poly(2-hydroxyethyl methacrylate) (PHEMA) have been produced [301]. The solution contains the monomer, thermal initiator, photoinitiator and a crosslinking agent. First oligomers are synthesized by thermally initiated polymerization of HEMA, then electrospinning is performed under irradiation (Figure 41). The studies on the preparation of nanofibers by reactive electrospinning of natural polymers such as hyaluronic acid, alginates, gelatin, collagen, cellulose are much less in number [90,98].

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 129

Figure 41. Schematics of reactive electrospinning of crosslinked PHEMA-based nanofibers and the SEM image of the produced nanofibers. The scale bar in the SEM image is 500 nm. Reproduced from Kim et al. [301] by permission of ACS.

A dual-syringe mixing technique has been used in order to prepare electrospun materials based on hyaluronic acid [291]. It requires a complex system composed of HMW hyaluronic acid and LMW hyaluronic acid derivative modified with 3,3’-dithiobis(propanoic dihydrazide), and PEG diacrylate as an agent able to crosslink the hyaluronic acid derivative. Moreover, as previously discussed in this Chapter for electrospinning of polyelectrolytes, in this case addition of a non-ionogenic polymer (PEO) is necessary. Further in this Section the performed up to date studies on reactive electrospinning aimed at preparation of covalently crosslinked fibers from chitosan or its derivatives are discussed. The ability of chitosan to give physically crosslinked hydrogels by PEC formation with polyacids offers new possibilities for application of the reactive electrospinning targeted to design of new hydrogel nanofibrous materials from the natural polymer. Thus, the gained till now experience and knowledge on preparation of PEC based nanofibers by electrospinning is emphasized in the Section.

6.1. One-Step Preparation of Electrospun Crosslinked Chitosan Nanofibers As discussed in Sections 4 and 5 of the Chapter, data on two-step preparation of crosslinked nanofibers from chitosan and its derivatives are available. Concerning the twostep procedure, the formation of fibers is followed by crosslinking. One-step preparation of crosslinked chitosan nanofibers by electrospinning of its TFA solution containing the crosslinking agent GA has been reported [242]. The lack of data on the electrospun mat morphology after its stay in aqueous medium at different pH values does not allow the degree and the uniformity of the cross-linking to be assessed. TFA is a rather harsh solvent and it is preferable to be replaced by more readily available solvents with milder effect, i.e. less toxic and more suitable for the potential biomedical application of chitosan- and CECh-containing

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

130

nanofibrous materials, such as acetic and formic acid. Reactive electrospinning aimed at preparation of hydrogel nanofibers from semi-interpenetrating networks containing chitosan or CECh and HMW PEO has been applied using formic acid as a solvent and GA as a crosslinking agent [233]. Two mechanisms have been proposed for crosslinking of chitosan by GA. One of them is Schiff base formation (Figure 42, left) [182,185]. OH

OH

O

O

O

O

O

HO

O

HO

NH2

NH2

Base

H

O

O

O

H

H

H NH2

O

H

O

H

H

O

HO

O

O

OH

O O

OH

Base

H

O

O

O

HO N

O

H

H

OH O

O O

HO

O

N OH O

HO O

HO

O O

H

O

H

O

O

N

N

O OH

N O HO

O

O OH

H

O

O

H

Figure 42. GA crosslinks chitosan either by a Schiff base imine functionality (left) and/or by Michaeltype adducts with terminal aldehydes (right). Reproduced from Schiffman et al. [242] by permission of ACS.

The other mechanism takes into account the mechanism proposed earlier for crosslinking of proteins [307]. It is assumed that it holds also for the system chitosan/GA [306], in addition to Schiff base formation. It has been proven that GA aqueous solutions contain a significant amount of α,β-unsaturated aldehyde groups, obtained as a result of GA aldol condensation. Thus, it has been suggested that unsaturated bonds adjacent to -CHO groups would rather result in stable against hydrolysis Michael type amino adducts (Figure 42, right). This contributes to the suggestion that the mechanism of chitosan crosslinking with GA is based on imine structures (Schiff bases) formation and in a less extent of Michael type adducts formation [182]. Recent studies [308] on the crosslinking of chitosan with GA has made the assumption that the obtaining of stable in acidic conditions chitosan networks is due to the formation of resonantly stabilized imine groups as a result of reaction between the amino groups of chitosan and the unsaturated double bonds adjacent to the aldehyde groups in the GA oligomers (Figure 43).

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 131 Hydrogel fibers from chitosan or CECh, PEO and GA have been successfully prepared by reactive electrospinning when 85 % formic acid has been used instead of acetic acid as a solvent since at total polymer concentration of 2.0 wt. % HCOOH enables the rapid gel formation to be avoided [233]. Effective electrospinning in this case is feasible only within 90 min after the start of the process; after that the process transforms to electrospraying. The dependence of the dynamic viscosity of these spinning solutions on time (Figure 44) shows that after 90 min the viscosity of the solutions at total polymer concentration of 2.0 wt. % begins to increase significantly. This is an indication that at this polymer concentration gel formation occurs rapidly not allowing effective electrospinning and preparation of a nanofibrous material with a satisfying yield. The decrease of the polymer concentration up to 1.7 wt. % results in retaining of the dynamic viscosity value of the spinning solution as long as 6 h, i.e. to a significant delay of the crosslinking process. This is favorable for the electrospinning since it ensures a much longer operating regime. The hydrogel fibers, prepared at total polymer concentration of 1.7 wt. % have higher average diameters values ( d = 130 nm) than fibers obtained at polymer concentration of 2.0 wt. % ( d = 60 nm). This result can be attributed to fiber formation with the participation mainly of the sol-fraction of the system at polymer concentration of 2.0 wt. %. The latter contains smaller amounts of GA, thus the obtained fibers are loosely crosslinked. Evidence for this is given by studies on the stability of fibrous materials from chitosan/PEO crosslinked with GA

OH O O

HO N CH OHC (CH2)2 C

CH (CH2)2 C CH (CH2)3 CHO CH N OH

O O HO

Figure 43. Schematic representation of the resonantly stabilized imine bonds formation.

132

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

Figure 44. Dependence of the dynamic viscosity of a solution containing chitosan, PEO and GA using 85 wt. % HCOOH as a solvent on time, and SEM micrographs of the prepared fibers; total polymer concentration 1.7 or 2.0 wt. %; chitosan/PEO = 1/1 (w/w); molar ratio [aminoglucoside units]/[CHOgroup of GA] = 1/1; 25±0.1 °ϥ. Penchev et al. [233], Electrospun hybrid nanofibers based on chitosan or N-carboxyethylchitosan and silver nanoparticles, Macromol Biosci., 2009, 9, 000-000, on line, DOI: 10.1002/mabi.200900003, Copyright Wiley-VCH Verlag GmbH and Co. KGaA. Reproduced with permission.

against acidic medium. While the average diameter values of fibers prepared at total polymer concentration of 1.7 wt. % is slightly altered after a 24 h stay in aqueous medium at pH 4 (Figure 45А), the fibers obtained at total polymer concentration of 2.0 wt. % swell significantly in this medium, and their diameters increase from 60 nm to 210 nm. The use of 85 wt. % HCOOH as a solvent has enabled the first successful preparation of hydrogel nanofibers from CECh and PEO in the presence of GA as a crosslinking agent. For this purpose spinning solutions at weight ratio CECh/PEO = 1/1; total polymer concentration of 3.4 wt. % and molar ratio [aminoglucoside units of the initial chitosan]/[CHO-group of GA] = 1/1 have been used. Previously, at this ratio water-insoluble pH-sensitive hydrogels from CECh have been prepared [309]. The dynamic viscosity of the spinning solutions is not altered within 24 h, i.e. within this period the gel formation of CECh/PEO/GA system is delayed in a greater extent in the presence of 85 wt. % HCOOH as compared to chitosan/PEO/GA system. Thus, effective electrospinning of CECh/PEO/GA system can be performed within this period of time (Figure 45B).

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 133

A

B

C

D

Figure 45. Nanofibers prepared by reactive electrospinning of chitosan/PEO/GA (А, total polymer concentration 1.7 wt.%) and CECh/PEO/GA (B, total polymer concentration 3.4 wt. %) before and after a 24 h stay at pH 4 (0.3 Ϡ ϥϡ3ϥϢϢϡ) and deionized water, respectively. Penchev et al. [233], Electrospun hybrid nanofibers based on chitosan or N-carboxyethylchitosan and silver nanoparticles, Macromol. Biosci., 2009, 9, 000-000, on line, DOI: 10.1002/mabi.200900003, Copyright Wiley-VCH Verlag GmbH and Co. KGaA. Reproduced with permission.

Additional evidence for the fact that the gel formation is much more delayed in CECh/PEO/GA system has been obtained from the studies performed on the stability of hydrogel fibers after contact with aqueous medium. As seen from Figure 45 chitosan- and CECh-containing fibers swell in contact with acidic medium but do not dissolve. This is an indication that the crosslinking of chitosan under the action of GA during the electrospinning process has effectively proceeded. Unlike chitosan-containing fibers that retain almost unchanged their average diameters values, CECh-containing fibers under these conditions swell in a great extent and coalescence of the individual fibers is observed.

134

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

6.2. Electrospun Nanofibers Composed of Polyelectrolyte Complexes Based on Chitosan The formation of water-insoluble PEC from aqueous solutions containing oppositely charged polyelectrolytes is an efficient tool for preparation of novel hydrogel materials [310]. The application of this approach depending on the polymer partners’ nature and medium conditions (polymer concentration, pH value, ionic strength, temperature) allows the preparation of diverse in respect to sizes and morphology materials, such as nanoparticles, micro- and nanostructured gels, and multilayered films. Ionically crosslinked hydrogels prepared in this way are known as complex coacervates, polyion complexes or polyelectrolyte complexes (Figure 46).

polyanion

mixing

polycation

“Coacervate complex” or hydrogel from polyion complex

Figure 46. A schematic representation of the preparation of hydrogels from physical networks based on oppositely charged polyelectrolytes obtained by mixing of their aqueous solutions.

As reported in Section 3 of this Chapter, chitosan behaves as a polycation in aqueous solutions and is able to form complexes with weak and strong polyacids [198,199,311,312]. That is why the possibility of utilizing these properties of chitosan for preparation of nanofibrous materials applying the electrospinning process is highlighted in this Section. The combination of PEC formation and electrospinning has been applied for the first time using a two-step procedure: electrospinning of a polymer followed by formation of a multilayered coating of PEC applying the layer-by-layer technique [313]. After the development of the layer-by-layer (LbL) technique [314] this tool of formation of self-assembled structures has become one of the main approaches applied for preparation of functionalized thin films. It consists in consequent adsorption of oppositely charged polyelectrolytes on different in nature substrates. It is claimed that the thickness of thus prepared multilayer films can be controlled with accuracy within the nanoscale [315]. An exceptional advantage of this technique is the fact that it can be applied by using different types of polyelectrolytes, as well as different in shape, morphology and size substrates [316,317], such as metallic nanotubes [318], short inorganic fibers [319], metal nanoparticles [320], polymer microbeads [321], etc. In the case of using porous substrates the latter are usually obtained by applying phase inversion technique, whereas materials with non-uniform pore size distribution are prepared [322,323]. This structure type has two main disadvantages: a comparatively low porosity degree and a non-controlled pore size distribution [324]. Electrospinning is considered as an alternative

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 135 technology that can be applied for preparation of porous scaffolds with high degree of porosity, better pore size distribution and a high surface-to-volume ratio [17,313,325]. That is why the development of new solutions for combining the both technologies – electrospinning and the LbL technique, is of interest. Electrospun cellulose acetate nanofibrous membrane that has been coated by the LbL technique with a polycation (PAH) and polyanion PAA has been used [326]. Chain segments of the polyelectrolyte partners are shown on Figure 47. CH2 CH

CH2 CH n COOH PAA

CH2

n

NH3+ Cl

Poly(allylamine hydrochloride) (PAH)

Figure 47. Formulae of PAA and PAH.

Since ϤАА is a weak polyacid (ЄКа = 4.8, [327]) pH value of the polymer solutions has a substantial impact on the formation of multilayer coatings from PEC. Nanofibers have been prepared from cellulose acetate with a multilayer coating from chitosan/sodium alginate PEC or from chitosan/polystyrene sulfonate PEC [328] Hollow multilayered PEC fibers have been obtained by using polystyrene nanofibers as a template which, after depositing the PEC film, has been removed by selective dissolution (Figure 48) [329,330].

Figure 48. Schematic diagram illustrating the fabrication of hollow multilayer polyelectrolyte nanofiber via LbL coating and removal of template. Reproduced from Ge et al. [330] by permission of Elsevier.

PAH – a weak polybase, and poly(styrene sulfonate) – a strong polyacid, have been used to form PEC [330]. It is assumed that the obtained hollow fibers can find potential application as drug delivery systems, as filters and in the tissue engineering. Successful experiments for reactive electrospinning of solutions containing oppositely charged polyelectrolytes have been performed by using weak polyacids or weak polybases

136

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

[331-333]. At certain Єϡ values the ionogenic groups of the weak polyelectrolytes are not in an ionized form and they cannot form PEC with weak or strong oppositely charged polyelectrolytes. A common spinning solution of chitosan and collagen has been electrospun by using a HFIP/TFA solvent system at 90/10 (v/v) [331,333]. The use of this mixed solvent enables the obtaining of nanofibers of average diameter below 500 nm. Undoubtedly these results are interesting despite the lack of any experimental data proving the formation of water-insoluble PEC, e.g. by determination of the fiber stability in the pH range in which the complex exists. The preparation of nanofibers from aqueous solutions containing PAH and PAA (molar ratio PAH/PAA=1/2) is a successful one at pH 1.2 [332] since at this pH value the carboxylic groups of PAA are not ionized and they are not able to form a water-insoluble PEC. The solubility of the fibers in physiological solution imposes the necessity of application of subsequent thermal crosslinking at 140 ºϥ.

A

B Figure 49. SEM micrographs of nanofibers from PEC chitosan/PAA prepared by electrospinning before (A) and after a 24 h stay at pH 4 in 0.3 Ϡ ϥϡ3ϥϢϢϡ (B); magnification of ×5000. Penchev et al. [334], Novel electrospun nanofibers composed of polyelectrolyte complexes. Macromol. Rapid Commun., 2008, 29, 677-681, Copyright Wiley-VCH Verlag GmbH and Co. KGaA. Reproduced with permission.

A novel approach for preparation of nanofibers from PEC chitosan/PAA has been proposed [334]. Macrophase separation has not been detected and the solution of the weak polybase chitosan and the weak polyacid PAA is homogenous when the mixed solvent H2O/HCOOH (volume ratio of 1/3.4) was used (pH 1). Water-insoluble chitosan/PAA complex is not formed at pH < 3. Below this pH value the predominant part of the PAA

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 137 carboxylic groups are in non-ionized form and the polyacid is not able to participate in PEC formation with the protonated amino groups of chitosan. SEM micrographs of the prepared fibers are shown in Figure 49A. The fibers are cylindrical in shape with average diameter value of 100±40 nm. Water-insoluble PEC chitosan/PAA is formed in narrow pH range from 3 to 6. As seen from the SEM micrograph shown in Figure 49B, the fibers immersed for 24 h at pH 4 have retained their morphology since as known the complex is stable in the pH range from 3 to 6 [311]. The preparation of nanofibers from chitosan and the strong polyacid PAMPS is examined as well when the same solvent system is used [334].

A

B Figure 50. SEM micrographs of nanofibers from PEC chitosan/PAMPS prepared by electrospinning before (A) and after a 24 h stay at pH 4 in 0.3 Ϡ ϥϡ3ϥϢϢϡ (B); magnification ×5000. Penchev et al. [334], Novel electrospun nanofibers composed of polyelectrolyte complexes. Macromol. Rapid Commun., 2008, 29, 677-681, Copyright Wiley-VCH Verlag GmbH and Co. KGaA. Reproduced with permission.

The molar ratio [aminoglucoside units]/[AMPS units] has been selected to be 1/1 since at this ratio maximal amount of the complex chitosan/PAMPS is formed in the pH range from 1 to 6 [198]. The attempts a homogeneous solution from chitosan and PAMPS at this molar ratio to be obtained by mixing of their solutions in 85 % ϡϥϢϢϡ (volume ratio H2O/HCOOH = 1/6) are unsuccessful since phase separation occurs as a result of the PEC

138

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

formation. In the case of systems in which one of the polyelectrolytes is a strong polyacid or a strong polybase, homogeneous solutions may be obtained using a ternary solvent system [335]. The latter consists of water, polar organic solvent and ionizable low-molecular-weight salt at a certain ratio between the components. The role of the organic solvent is to disrupt the hydrophobic interactions since they significantly contribute to the PEC formation process. The role of the low-molecular-weight salt is to shield the charged ionized functional groups of the polyelectrolyte, thus hampering the formation of water-insoluble complexes. A homogeneous solution of chitosan and PAMPS is obtained using 85% HCOOH (volume ratio H2O/HCOOH = 1/6) in the presence of CaCl2. The electrospinning of this solution enables the preparation of fibers cylindrical in shape with diameters of 130±50 nm (Figure 50А). In acidic medium chitosan/PAMPS fibers swell without dissolution (Figure 50B). The stay of the fibrous mat in 0.3 Ϡ CH3COOH is accompanied by release of CaCl2 from the fibers in the aqueous medium and to the formation of water-insoluble PEC. It has been demonstrated by gravimetry that the entire amount CaCl2 is released after a 24 h stay of the mat in acidic medium. The determined loss of polymer from the mat is 19 %. Most probably this is due to extraction of the non-complexed polymer partners that are soluble in the aqueous medium at pH 4. The chitosan/PAMPS complex is stable up to pH 8 [198]. The immersion of the mat from PEC chitosan/PAMPS in a buffer solution of pH 9 is accompanied by swelling followed by its fragmentation after 48 h. The fraction of the mat insoluble at pH 9 is soluble in acidic medium. This confirms that the fibers consist of PEC chitosan/PAMPS which disintegrates at pH 9 to a soluble fraction (PAMPS) and an insoluble fraction (chitosan).

Figure 51. Yarns formation from self-assembled fibers during the electrospinning process from chitosan/PAMPS/HCOOH/CaCl2 system using a stationary collector [336].

In the case of the electrospinning of chitosan/PAMPS/CaCl2 system an interesting phenomenon of self-organization of the fibers during the electrospinning has been observed [336]. The initially formed thin fibers grow in height from the negatively charged collector to the positively charged capillary tip accompanied by of an intensive process of self-bundling of the fibers. This self-assembly leads to formation of yarns (Figure 51). This phenomenon has been observed during the electrospinning of non-ionogenic polymers in the presence of low-molecular-weight conductive salts [34]. It has been found that the self-assembling of the

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 139 respective fibers in bundles can occur when the conductivity of the spinning solutions is higher than 400 μS/cm. In the case of chitosan/PAMPS/HCOOH/CaCl2 the electrical conductivity of the spinning solution is 5600 μS/cm. Having in mind that the most recent trends in the field of micro- and nanofibrous yarns formation is the use of self-assembly of conductive nanofibers into yarns [34,337,338] the nanofibers from polyelectrolytes are highly promising in this respect. In addition, owing to their ionogenic nature and the presence of functional groups in their structure, this type of polymers and materials are potential candidates for design of new generation micro- or nanofibrous yarns that can find diverse application in variety of fields: for design of clothing of remarkably low weight and improved barrier properties against humidity and wind; for the design of novel highly effective membranes for bioreactors, as well as for new highly effective filters; for preparation of bioactive wound healing dressings; for diverse technological solutions related to advanced technologies such as design of biosensors; highly effective catalysts, new devices for the optoelectronics, etc. The one-step formation of waterinsoluble hydrogel fibers from PEC in combination with the self-bundling of the fibers is a substantial prerequisite for the successful preparation of new materials with desired properties. Additional advantage is the possibility natural polymers (e.g., chitosan) to be incorporated thus allowing the design of biocompatible and biodegradable polymer materials. Summarily, the reactive electrospinning enables the tailored preparation of hydrogel nanofibers from chitosan. Two approaches can be applied: covalent crosslinking of chitosan or PEC formation between chitosan and weak or strong polyacids. In addition, the formation of micro- and nanofibrous yarns during the reactive electrospinning broadens the possibilities for tailored preparation of novel electrospun materials that have properties prompted by the difference of the morphology and shape as compared to materials prepared by conventional techniques. As evidenced, a very attractive field for comprehensive research on the possibility for preparation of hydrogel fibrous materials by reactive electrospinning has arisen during the last years. It is to be expected that these materials could bring electrospinning closer to the industrial scale.

7. CONCLUSION In the past few years a large number of attempts to prepare chitosan-based nanofibers by electrospinning have been made using different approaches, e.g. use of suitable solvent, or addition of non-ionogenic polymer partner. At present, fibers from chitosan and chitosan derivatives having nanoscale diameters can be easily produced by electrospinning. The obtained nanofibrous materials have proven to be promising for application in diverse fields, and especially in biomedical field, mainly as scaffolds for cell and tissue engineering, and as wound-healing dressings. While progress in the preparation and characterization of the nanofibrous materials from chitosan and chitosan derivatives has been made, there are still efforts to be done to prepare materials with higher mechanical strength and diameter uniformity. Some studies on hybrid nanofibers prepared using a suitable reinforcing polymer partner have already been performed. Much deeper insight on the effect of the process parameters on the fiber morphology is needed. It may be anticipated that along with the conventional mode of electrospinning more attention will be focused on the reactive

140

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

electrospinning and on taking advantage of polyelectrolyte complex formation during electrospinning. The potential of electrospinning as a means to create nanofibrous materials based on chitosan is great, and gives reasons for the intensive development of this field of research.

ACKNOWLEDGEMENTS Financial support from the National Science Fund of Bulgaria (Grant DO-02-82/2008) is gratefully acknowledged.

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]

Allcock, H., Lampe, F F. Contemporary Polymer Chemistry; Prentice Hall: Engelwood Cliffs, NJ, 1990. Martin, C. R. Chem. Mater. 1996, 8, 1739-1746. Feng, L.; Li, S.; Li, H.; Zhai, J.; Song, Y.; Jiang, L. Angew. Chem. Int. Ed. 2002, 41, 1221-1223. Liu, G. J.; Ding, J. F.; Qiao, L. J.; Guo, A.; Dymov, B. P.; Gleeson, J. T. European J., 1999, 5, 274-2749. Whitesides, G. M.; Grzybowski, B. Science, 2002, 295, 2418-2421. Hongu, T., Phillips, G. O., Eds.; New Fibers; Woodhead Publishing Limited: Cambridge, England, 1997. Ma, P. X.; Zhang, R. J. Biomed. Mat. Res. 1999, 46, 60-72. Ondarcuhu, T.; Joachim, C. Europhys. Lett. 1998, 42, 215-220. Fabbricante, A.; Ward, G.; Fabbricante, T. US Patent 6,114,017, 2000. Torobin, L.; Findlow, R. US Patent 6,183,670, 2001. Huang, Z. M.; Zhang, Y. Z.; Kotaki, M.; Ramakrishna, S. Composites Sci. Techn. 2003, 63, 2223-2253. Teo, W.; Ramakrishna, S. Nanotechnology, 2006, 17, 89-106. Greiner, A.; Wendorff, J. H. Adv. Polym. Sci. 2008, 219, 107–171. Kriegel, C.; Arrechi, A.; Kit, K.; McClements, D. J.; Weiss, J. Crit. Rev. Food Sci. Nutrition, 2008, 48, 775–797. Cooley, J. F. US Patent 692,631, 1902. Morton, W. J. US Patent 705,691, 1902. Reneker, D. H.; Yarin, A. L. Polymer, 2008, 49, 2387-2425. McKee, M.; Elkins, C.; Long, T. Polymer, 2004, 45, 8705-8715. Shenoy, S. L.; Bates, W. D.; Frisch, H. L.; Wnek G. E. Polymer, 2005, 46, 3372-3384. Fong, H.; Chun, I.; Reneker, D. H. Polymer, 1999, 40, 4585-4592. Fong, H.; Reneker, D. H. J. Polym. Sci.: Polym. Phys. Ed., 1999, 37, 3488-3493. Spasova, M.; Stoilova, O.; Manolova, N.; Altankov, G.; Rashkov, I. J. Bioact. Comp. Polym. 2007, 22, 62-76. Gupta, P.; Elkins, C.; Long, T. E.; Wilkes, G. L. Polymer, 2005, 46, 4799-4810. Shenoy, S. L.; Bates, W. D.; Wnek, G. E. Polymer, 2005, 46, 8990-9004.

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 141 [25] Hager, B. L.; Berry, G. C. J. Polym. Sci., Polym. Phys. Ed. 1982, 20, 911-928. [26] McKee, M. G.; Wilkes, G. L.; Colby, R. H.; Long, T. E. Macromolecules 2004, 37, 1760-1767. [27] Kenawy, E. R.; Layman, J. M.; Watkins, J. R.; Bowlin, G. L.; Mattews, J. A.; Simpson, D. G.; Wnek, G. E.Biomaterials, 2003, 24, 907-913. [28] Ignatova, M.; Starbova, K.; Markova, N.; Manolova, N.; Rashkov, I. Carbohydr. Res. 2006, 341, 2098-2107. [29] Mincheva, R.; Manolova, N.; Rashkov, I. Eur. Polym. J. 2007, 43, 2809–2818. [30] Zong, X. H.; Kim, K.; Fang, D. F.; Ran, S. F.; Hsiao, B. S.; Chu, B. Polymer, 2002, 43, 4403-4412. [31] Jun, Z.; Hou, H.; Schaper, A.; Wendorff, J.; Greiner, A. e-Polymers, 2003, no. 009, 1-9. [32] Ignatova, M.; Manolova, N.; Rashkov, I. Eur. Polym. J. 2007, 43, 1609-1623. [33] McKee, M. G.; Hunley, M. T.; Layman, J. M.; Long, T. E. Macromolecules, 2006, 39, 575-583. [34] Wang, X.; Zhang, K.; Zhu, M.; Yu, H.; Zhou, Z.; Chen, Y.; Hsiao, B.S. Polymer, 2008, 49, 2755–2761. [35] Lin, T.; Wang, H. X.; Wang, H. M.; Wang, X. G. Nanotechnology, 2004, 15, 13751381. [36] Lee, K. H.; Kim, H. Y.; La, Y. M.; Lee, D. R.; Sung, N. H. J. Polym. Sci. Part B: Polym. Phys. 2002, 40, 2259-2268. [37] Wannatong, L.; Sirivat, A.; Supaphol, P. Polym. Int. 2004, 53, 1851-1859. [38] Bognitzki, M.; Czado, W.; Frese, T.; Schaper, A.; Hellwig, M.; Steinhart, M.; Greiner, A.; Wendorff, H. Adv. Mater. 2001, 13, 70-72. [39] Shin, Y. M.; Hohman, M. M.; Brenner, M. P.; Rutledge, G. C. Appl. Phys. Letters, 2001, 78, 1149-1151. [40] Shin, Y. M.; Hohman, M. M.; Brenner, M. P.; Rutledge, G. C. Polymer, 2001, 42, 9955-9967. [41] Fridrikh, S. V.; Yu, J. H.; Brenner, M. P.; Rutledge, G. C. Phys. Rev. Lett. 2003, 90, 144502, 1-4. [42] Jalili, R.; Hosseini, S. A.; Morshed, M. Iran. Polym. J. 2005, 14, 1074-1081. [43] Megelski, S.; Stephens, J. S.; Chase, D. B.; Rabolt, J. F. Macromolecules, 2002, 35, 8456-8466. [44] Lee, J. S.; Choi, K. H.; Ghim, H. D.; Kim, S. S.; Chun, D. H.; Kim, H. Y.; Lyoo, W. S. J. Appl. Polym. Sci. 2004, 93, 1638-1646. [45] Spasova, M.; Manolova, N.; Paneva, D.; Rashkov, I. e-Polymers, 2004, no. 056, www.e-polymers.org. [46] Takahashi, T.; Taniguchi, M.; Kawai, T. Japan. J. Appl. Phys. Part 2: Lett. Expr. Lett. 2005, 44, L860-L862. [47] Shukla, S.; Brinley, E.; Cho, H. J.; Seal, S. Polymer, 2005, 46, 12130-12145. [48] Gu, S. Y., Ren, J. Macromol. Mater. Eng., 2005, 290, 1097-1105. [49] Lee, K. H.; Kim, H. Y.; La, Y. M.; Lee, D. R.; Sung N. H. J. Polym. Sci. Part B: Polym. Physics 2002, 40, 2259-2268. [50] Mincheva, R.; Paneva, D.; Manolova, N.; Rashkov, I. J. Bioact. Compat. Polym., 2005, 20, 419-435. [51] Baker, S. C.; Atkins, N.; Gunning, P. A.; Granville, N.; Wilson, K.; Wilson, D.; Southgate, J. Biomaterials, 2006, 27, 3136-3146.

142

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

[52] Hong, Y.; Shang, T.; Li, Y.; Wang, L.; Wang, C.; Chen, X.; Jing, X. J. Membr. Sci. 2006, 276, 1-7. [53] Kidoaki, S.; Kwon, K.; Matsuda, T. J. Biomed. Mater. Res. Part B: Apll. Biomater. 2006, 76B(1), 219-229. [54] Buttafoco, L.; Kolkman, N. G.; Engbers-Buijtenhuijs, P.; Poot, A. A.; Dijkstra, P. J.; Vermes, I.; Feijen, J. Biomaterials, 2006, 27, 724-734. [55] Jeun, J. P.; Lim, Y. M.; Nho, Y. C. J. Industr. Eng. Chem. 2005, 11, 573-578. [56] Zhang, C.X.; Yuan, X. Y.; Wu, L. L.; Sheng, J. Eur. Polym. J., 2005, 41, 423-432. [57] Demir, M. M.; Yilgor, I.; Yilgor, E.; Erman, B. Polymer 2002, 43, 3303-3309. [58] De Vrieze, S.; Van Camp, T.; Nelvig, A.; Hagström, B.; Westbroek, P.; De Clerck, K.; J. Mater. Sci. 2009, 44, 1357-1362. [59] Kim, G.-T.; Lee, J.-S.; Shin, J.-H.; Ahn, Y.-Ch.; Hwang, Y.-J.; Shin, H.-S.; Lee, J.-K.; Sung, Ch.-M. Korean J. Chem. Eng., 2005, 22, 783-788. [60] Frenot, A.; Chronakis, I. S. Curr. Opin. Coll. Interface Sci., 2003, 8, 64-75. [61] Koombhongse, S.; Liu, W.; Reneker, D. H. J. Polym. Sci.: Polym. Phys. Ed., 2001, 39, 2598-2606. [62] Koski, A.; Yim, K.; Shivkumar, S. Mater. Lett., 2004, 58, 493-497. [63] Spasova, M.; Mincheva, R.; Paneva, D.; Manolova, N.; Rashkov, I. J. Bioactive Compatible Polym., 2006, 21, 465-479. [64] Bognitzki, M.; Hou, H.; Frese, M. I. T.; Hellwig, M.; Schwarte, C.; Schaper, A.; Wendorff, J. H.; Greiner, A. Adv. Mater., 2000, 12, 637-640. [65] Lee, Y. H.; Lee, J. H.; An, I.-G.; Kim, C.; Lee, D. S.; Lee, Y. K.; Nam, J.-D. Biomaterials, 2005, 26, 3165-3172. [66] Spasova, M.; Paneva, D.; Manolova, N.; Radenkov, Ph.; Rashkov, I. Macromol. Biosci., 2008, 8, 153-162. [67] Spasova, M. Novel Micro- and Nanofibrous Materials, Prepared by Electrospinning from Solutions of: Chitosan/Polyethylene oxide, Poly(L-lactide)/Polyethylene glycol and Polyethylene oxide /Ti(IV)isopropoxide, PhD Thesis, 2007, Technical University/Bulgarian Academy of Sciences, Sofia, Bulgaria. [68] Torres, B. Proc. Natnl. Conf. Undergrad. Res.; University of Kentucky, Lexington, Kentucky, 2001, p 1–5. [69] Liu, C.; Bai, R. J. Membr. Sci., 2005, 267, 68-77. [70] Geng, X.; Kwon, O. H.; Jang, J. Biomaterials, 2005, 26, 5427-5432. [71] Fennessey, S. F.; Farris, R. J. Polymer 2004, 45, 4217- 4225. [72] Li, D.; Wang, Y.; Xia, Y. Adv. Mater. 2004, 16, 361-366. [73] Park, S.; Park, K.; Yoon, H.; Son, J. G.; Min, T.; Kim, G. H. Polym. Int. 2007, 56, 1361-1366. [74] Chew, S.; Wen, J.; Yim, E.; Leong, K. Biomacromolecules, 2005, 6, 2017-2024. [75] Kim, K.; Lee, K.; Khil, M.; Ho, Y.; Kim, H. Fibers Polym. 2004, 5, 122-127. [76] Matthews, J.; Wnek, G.; Simpson, D.; Bowlin, G. Biomacromolecules, 2002, 3, 232238. [77] Theron, A.; Zussman, E.; Yarin, A. Nanotechnology, 2001, 12, 384-390. [78] Inai, R.; Kotaki, M.; Ramakrishna, S. Nanotechnology, 2005, 16, 208-213. [79] Sundaray, B.; Subramanian, V.; Natarajan, T. S.; Xiang, R.-Z.; Chang, C.-C.; Fann, W.S. Appl. Phys. Lett. 2004, 84, 1222-1224. [80] Teo, W.; Ramakrishna, S. Nanotechnology, 2005, 16, 1878-1884.

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 143 [81] [82] [83] [84] [85] [86]

Li, D.; Wang, Y.; Xia, Y. Nano Lett., 2003, 3, 1167-1171. Mincheva, R.; Manolova, N.; Rashkov, I. Eur. Polym. J. 2007, 43, 2809–2818. Zhou, F.-L.; Gong, R.-H.; Porat, I. Polym. Int. 2009, 58, 331-342. Gupta, P.; Wilkes, G. L. Polymer 2003, 44, 6353-6359. Ding, B.; Kimura, E.; Sato, T.; Fujita, S.; Shiratori, S. Polymer 2004, 45, 1895-1902. Gibson, Ph.; Schreuder-Gibson, H.; Rivin, D. Coll. Surf. A: Physicochem. Eng. Aspects, 2001, 187–188, 469–481. [87] Wang, X.; Kim, Y.-G.; Drew, Ch.; Ku, B.-Ch.; Kumar, J.; Samuelson, L.A. Nano Lett. 2004, 4, 331-334. [88] Zhou, F.-L.; Gong R.-H. Polym. Int. 2008, 57, 837-845. [89] Shaffer, M. S. P.; Sandler, J. K. W. In Nanofibers and Nanotechnology in Textiles, Brown, P. J.; Stevens K., Eds.; Woodhead Publishing Limited and CRC Press LLC, 2007; p 194-234. [90] Andrady, A.L. Science and Technology of Polymer Nanofibers; John Wiley and Sons, Inc.: Hoboken, New Jersey, Canada, 2008. [91] Tan, S.; Huang, X; Wu, B. Polym Int., 2007, 56, 1330–1339. [92] Sell, S.; Barnes, C.; Smith, M.; McClure, M.; Madurantakam, P.; Grant, J; McManus, M.; Bowlin, G. Polym Int., 2007, 56, 1349–1360. [93] Greiner, A.; Wendorff J. H. Adv. Polym. Sci., 2008, 219, 107–171. [94] Hunley M. T.; Long, T. E. Polym. Int., 2008, 57, 385–389. [95] Schiffman, J. D.; Schauer, C. L. Polym. Rev., 2008, 48, 317–352. [96] Agarwal, S.; Wendorff, J. H.; Greiner, A. Polymer 2008, 49, 5603-5621. [97] Jian, F.; HaiTao, N.; Tong, L.; XunGai, W. Chinese Sci. Bull. 2008, 53, 2265-2286. [98] Ramakrishna, S.; Fujihara, K.; Teo, W.-E.; Lim, T.-C.; Ma, Z. An Introduction to Electrospinning and Nanofibers; World Scientific Publishing Co. Pte. Ltd.: Ton Tuck Link, Singapore, 2005. [99] Tolaimate, A.; Desbrières, J.; Rhazi, M.; Alagui, M.; Vincendon, M.; Vottero, P. Polymer 2000, 41, 2463-2469. [100] Tharanathan, R. N.; Kittur, F. S. Crit. Rev. Food Sci. Nutr., 2003, 43, 61-87. [101] Harish Prashanth, K.V.; Tharanathan, R.N. Trends Food Sci. Technol. 2007, 18, 117131. [102] Synowiecki, J.; Al-Khateeb, N. Food Chem. 1997, 60, 605-610. [103] Muzzarelli, R. A. A. Chitin; Pergamon Press: Oxford, U.K., 1977. [104] Li, J.; Revol, J. F.; Marchessault, R. H. J. Appl. Polym. Sci. 1997, 65, 373-380. [105] Ravi-Kumar, M. N. V. React. Funct. Polymers 2000, 46, 1–27. [106] Domard, A. Int. J. Biol. Macromol. 1987, 9, 98-104. [107] Thanou, M.; Verhoef, J. C.; Junginger, H. E. Adv. Drug Deliv. Rev., 2001, 52, 117-126. [108] Muzzarelli, R.; Rocchetti, R.; Stanie, V.; Weckx, M.. In Chitin Handbook, Muzzarelli, R.; Peters, M., Eds.; Grottammare: Atee Edizioni, 1977; p 127. [109] Wu, A. C. M.; Bough, W. A. In: Proc. 1st Int. Conf. Chitin/Chitosan, Muzzarelli, R.A.A.; Pariser, E.R., Eds.; MIT-SG, Cambridge, 1978; p 88. [110] Sannan, T.; Kurita, K.; Iwakura, Y. Makromol. Chem. 1976, 177, 3589-3600. [111] Domszy, J. G.; Roberts, G. A. F.; Makromol. Chem. 1985, 186, 1671-1677. [112] Moore, G. K.; Roberts, G. A. F. Int. J. Biol. Macromol. 1980, 2, 115-116. [113] Muzzarelli, R. A. A.; Muzzarelli, C. Adv. Polym. Sci. 2005, 186, 151–209. [114] Prabaharan, M. J. Biomat. Appl. 2008, 23, 5-36.

144

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

[115] Agulló, E.; Rodríguez, M. S.; Ramos, V.; Albertengo, L. Macromol. Biosci. 2003, 3, 521-530. [116] Lim, S. H.; Hudson, S. M. J. Macromol. Sci., Part C, Polym. Rev., 2003, C43, 223-269. [117] Stoilova, O.; Paneva, D.; Manolova, N.; Rashkov, I. In Unit 2, Module 3 in Specific Content Issues Polymeric Materials for Food Packaging and Food Preservation; eReference Tools for VET Trainers in Food Industry, http://www.e-fost.org/en/SCI-A-32. [118] Steinbüchel, A.; Rhee, S. K., Eds., Polysaccharides and Polyamides in the Food Industry: Properties, Production, and Patents, Wiley-VCH Verlag GmbH and Co. KGaA: Weinheim, 2005. [119] Uragami, T.; Tokura, S., Eds., Material Science of Chitin and Chitosan., Springer, 2006. [120] Skryabin, K.G., Vikhoreva G.A., Varlamov V.P., Eds., Chitin and Chitosan: Production, Properties and Usage; Nauka, Russian Academy of Sciences: Moscow, 2002 (in Russian). [121] Dumitriu S., Ed., Polymeric Biomaterials, 2nd ed.; Marcel Dekker Inc.: New York., 2001. [122] Dumitriu S., Ed., Polysaccharides: Structural Diversity and Functional Versatility, 2nd ed., CRC press, 2004. [123] Kurakake, M.; Yo-u, S.; Nakagawa, K.; Sugihara, M.; Komaki, T. Current Microbiol., 2000, 40, 6-9. [124] Nogawa, M.; Takahashi, H.; Kashiwagi, A.; Ohshima, K.; Okada, H.; Morikawa, Y. Appl. Envir. Microbiol. 1998, 64, 890-895. [125] Ilyina, A.V.; Tatarinova, N. Yu.; Varlamov, V. P. Process Biochem. 1999, 34, 875-878. [126] Grigolon, L. B.; Azevedo, A.; Santos, R. R.; Franco, T. T. In Chitin Enzymology. Muzzarelli, R. A. A., Ed., Atec: Italy, 2001; p 78-87. [127] Shi, C.; Zhu, Y.; Ran, X.; Wang, M.; Su, Y.; Cheng, T. J. Surg. Res., 2006, 133, 185192. [128] Stoilova, O.; Koseva, N.; Petrova, Ts.; Manolova, N.; Rashkov, I. J. Bioact. Compat. Polym. 2001, 16, 379-392. [129] Paneva, D.; Mincheva, R.; Stoilova, O.; Manolova, N.; Rashkov, I. e-Polymers 2003, no 019. [130] Bonina, P.; Petrova, Ts.; Manolova, N.; Rashkov, I.; Naydenov, M. J. Bioact. Compat. Polym. 2004, 19, 197-208. [131] Bonina, P.; Petrova, Ts.; Manolov, N.; Rashkov, I. e-Polymers 2004, no 062. [132] Suzuki, S. Fragr. J., 1996, 15, 61-68. [133] Tokora, A.; Kobayashi, M.; Tatekawa, N.; Suzuki, K.; Okawa, Y.; Mikami, T. Microbiol. Immunol., 1989, 33, 357-367. [134] Vasyukova, N. I.; Zinov’eva, S. V.; Il’inskaya, L. I.; Perekhod, E. A.; Chalenko, G. I.; Gerasimova, N. G.; Il’ina, A. V.; Varlamov, V. P.; Ozeretskovskaya, O. L. Appl. Biochem. Microbiol., 2001, 37, 103-109. [135] Hirano, S.; Nagao, N. Agric. Biol. Chem., 1989, 53, 3065-3066. [136] Uchida, Y.; Izume, M.; Ohtakara, A. In Chitin and Chitosan Skjak-Braek, G.; Anthonsen, T.; Sandford, P., Eds.,. Elsevier Applied Science: New York, 1989; p 373382. [137] Chen, M.; Yeh, G. H.; Chiang, B. J. Food Process Preserv. 1996, 20, 379-390.

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 145 [138] Rao, S. B.; Sharma, C. P. J. Biomed. Mater. Res. 1997, 34, 21-28. [139] Muzzarelli, R. A. A.; Muzzarelli, B. In Polymeric Biomaterials, 2nd ed.; Dumitriu, S., Ed.; Marcel Dekker: New York, 2002. [140] Yalpani, M.; Johnson, F.; Robinson, L. E. In: Advances in Chitin and Chitosan, Sandford, P. A.; Zikakis, J. P., Eds.; Elsevier Applied Science: London, UK, 1992; p 543. [141] Sudardshau, N. R.; Hoover, D. G.; Knorr D. Food Biotechnol. 1992, 6, 257-272. [142] Li, Z.; Zhuang, X. P.; Liu, X. F.; Guan, Y. L.; De Yao K. Pоlymer 2002, 43, 15411547. [143] Guo, Z.; Chen, R.; Xing, R.; Liu, S.; Yu, H.; Wang, P.; Li, C.; Li, P. Carbohydr. Res. 2006, 341, 351-354. [144] Kim, C. H.; Choi, J. W.; Chun, H. J.; Choi, K. S. Polym. Bull. 1997, 38, 387-393. [145] Liu, X.F.; Guan, Y.L.; Yang, D.Z.; Li, Z.; Yao, K.D. J. Appl. Polym. Sci. 2001, 79, 1324-1335. [146] Hirano, S. Polym. Int. 1999, 48, 732-734. [147] Hadwiger, L. A.; Kendra, D. F.; Fristensky, B. W.; Wagoner, W. In Chitin in Nature and Technology. Plenum Press: New York, 1986, p 209-214. [148] Stone, C. A.; Wright, H.; Clarke, T.; Powell, R.; Devaraj, V. S. Br. J. Plast. Surgery 2000, 53, 601-606. [149] Mi, F. L.; Shyu, S. S.; Wu, Y. B.; Lee, S. T.; Shyong, J. Y.; Huang, R. N. Biomaterials, 2001, 22, 165-173. [150] Wang, X. H.; Zhu, Y.; Feng, Q. L.; Cui, F. Z. J. Bioact. Compat. Polym. 2003, 18, 135146. [151] Suzuki, M.; Ito, S.; Yamaguchi, I.; Takakuda, K.; Kobayashi, H.; Shinomiya, K.; Tanaka, J. J. Neurosci. Res. 2003, 72, 646-659. [152] Hirano, S. In Chitin and Chitosan. Environmental Friendly and Versatile Biomaterials; Stevens, W. F.; Rao, M. S.; Chandrakran-chang, S., Eds.; AIT: Bangkok, 1996, p 22. [153] Wang, Q. Z.; Chen, X. G.; Li, Z. X.; Wang, S.; Liu, C. S.; Meng, X. H.; Liu, C. G.; Lu, Y. H.; Yu, L. J. J. Mater Sci: Mater. Med. 2008, 19, 1371-1377. [154] Paneva, D.; Manolova, N.; Rashkov, I.; Danchev, D. J. Bioact. Compat. Polym. 2005, 20, 133-151. [155] Horton, D.; Just, E. Carbohydr. Res. 1973, 29, 173-180. [156] Whistler, R.; Kosik, M. Arch. Biochem. Biophys. 1971, 142, 106-110. [157] Vongchan, P.; Sajomsang, W.; Subyen, D.; Kongtawelert, P. Carbohydr. Res., 2002, 337, 1239-1242. [158] Tokura, S.; Nishi, N.; Itoyama, K.; Shirai, A.; Nishimura, S.; Azuma, I. In Chitin World, Karnicki, Z.; Wojtas, A.; Bykowski, P., Eds.; Bremershaven: Wirtschaftsverlag, NW, 1994; p 287. [159] Domard, A.; Rinaudo, M.; Terrassin, C. Int. J. Biol. Macromol. 1986, 8, 105-107. [160] Muzzarelli, R. A. A. Carbohydr. Polym. 1988, 8, 1-21. [161] Maresch, G.; Clausen, T.; Lang, G. In Chitin and Chitosan; Skjak-Braek, G.; Anthonsen, T.; Sandford, P., Eds.; Elsevier Appl. Sci.: London, New York, 1988; p 389. [162] Sashiwa, H.; Yamamori, N.; Ichinose, Y.; Sunamoto, J.; Aiba, S. Macromol. Biosci. 2003, 3, 231-233.

146

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

[163] Kurita, K. In Application of Chitin and Chitosan; Goosen, M. F. A., Ed.; Technomic Publishing Co.: Lancaster, Basel, 1997; p 297. [164] Shantha, K. L.; Bala, U.; Rao, K. P. Eur. Polym. J. 1995, 31, 377-382. [165] Yazdani-Pedram, M.; Retuert, J.; Quijada, R. Macromol. Chem. Phys. 2000, 201, 923930. [166] Yazdani-Pedram, M.; Retuert, J. J. Appl. Polym. Sci. 1997, 63, 1321-1326. [167] Blair, H. S.; Guthrie, J.; Law, T. K.; Turkington, P. J. Appl. Polym. Sci. 1987, 33, 641656. [168] Nud’ga, L. A.; Petrova, V. A.; Lebedeva, M. F.; Petropavlovskii, G. A. Russ. J. Appl. Chem. 1996, 69, 1058-1063. [169] Berkovich, L. A.; Timofeyeva, G. I.; Tsyurupa, M. P.; Davankov, V. A. Polym. Sci. U.S.S.R., 1980, 22, 2009-2018. [170] Lagos, A.; Reyes, J. J. Polym. Sci. Part A: Polym. Chem. 1988, 26, 985-991. [171] Retuert, J.; Yazdani-Pedram, M. Polym. Bull. 1993, 31, 559-562. [172] Yazdani-Pedram, M.; Lagos, A.; Retuert, J.; Guerrero, R.; Riquelme, P. J. Macromol. Sci. Chem. 1995, A32, 1037-1047. [173] Smirnova, L.A., Semchikov, Yu.D., Tikhobaeva, Ya.G., Pastukhova, N.V., Polymer Science – Series B 2001, 43, 33-36. [174] Najjar A. M. K.; Yunus, W. Z. W.; Ahmad, M. B.; Rahman, M. Z. J. Appl. Polym. Sci., 2000, 77, 2314-2318. [175] Qu, X.; Wirsen, A.; Albertsson, A.C. J. Appl. Polym. Sci. 1999, 74, 3193–3202. [176] Wan, Y.; Cao, X.; Zhang, S.; Wang, S.; Wu, Q. Acta Biomaterialia 2008, 4, 876-886. [177] Wei, Y. C.; Hudson, S. M.; Mayer, J. M.; Kaplan, D. L. J. Polym. Sci.: Part A: Polym. Chem. 1992, 30, 2187-2193. [178] Mayer, J. M.; Smithfield, R. I.; Kaplan, D. L.; Stow, M. US Patent 5,015,293, 1991. [179] Kawamura, Y.; Mitsuhashi, M.; Tanibe, H.; Yoshida, H. Ind. Eng. Chem. Res. 1993, 32, 386-391. [180] Juang, R.-S.; Ju, C.-Y. Ind. Eng. Chem. Res. 1997, 36, 5403-5409. [181] Knaul, J. Z.; Hudson, S. M.; Creber, K. A. M. J. Polym. Sci. Part B: Polym. Phys. 1999, 37, 1079-1094. [182] Roberts, G. A. F.; Taylor, K. E. Macromol. Chem. 1989, 190, 951-960. [183] Rorrer, G.L.; Hsien, T.-Y.; Way, J.D. Ind. Eng. Chem. Res. 1993, 32, 2170-2178. [184] Hsien, T.-Y.; Rorrer, G. L. Ind. Eng. Chem. Res. 1997, 36, 3631-3638. [185] Tual, C.; Espuche, E.; Escoubes, M.; Domard, A. J. Polym. Sci.: Part B: Polym. Phys., 2000, 38, 1521-1529. [186] Mi, F.-L.; Tan, Y.-C.; Liang, H.-F.; Sung H.-W. Biomaterials, 2002, 23, 181-191. [187] Kamiński, K.; Zazakowny, K; Szczubiałka, K; Nowakowska M. Biomacromolecules, 2008, 9, 3127-3132. [188] Remuñán-López, C.; Bodmeier, R. Drug Dev. Ind. Pharm. 1996, 22, 1201-1209. [189] Gupta, K. C.; Ravi-Kumar, M. N. V. Polym. Int. 2000, 49, 141-146. [190] Gupta, K. C.; Ravi-Kumar, M. N. V. Biomaterials 2000, 21, 1115-1119. [191] Koseva, N.; Manolova, N.; Markova, N.; Radoucheva, T.; Rashkov, I. Polym. Bull. 1999, 43, 101-107. [192] Koseva, N.; Stoilova, O.; Manolova, N.; Rashkov, I. J. Bioact. Compat. Polym. 2001, 16, 3-19.

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 147 [193] Mireles, C.; Martino, M.; Bouzas, J.; Torres, J. A. In Advances in Chitin and Chitosan; Brine, C. J., Sandford, P. A., Zikakis, J. P., Eds.; Princeton, 1991; p 506. [194] Cárdenas, A.; Argüelles-Monal, W.; Goycoolea, F.M.; Higuera-Ciapara, I.; Peniche, C. Macromol. Biosci. 2003, 3, 535-539. [195] Takahashi, T.; Takayama, K.; Machida, Y.; Nagai, T. Int. J. Pharm. 1990, 61, 35-41. [196] Peniche-Covas, C; Argüelles-Monal, W. Macromol. Symp. 2001, 168, 103-116. [197] Fong, H.; Chun, I.; Reneker, D. H. Polymer 1999, 40, 4585-4592. [198] Stoilova, O.; Koseva, N.; Manolova, N.; Rashkov, I. Polym. Bull. 1999, 43, 67-73. [199] Paneva, D.; Stoilova, O.; Manolova, N.; Rashkov I. e-Polymers, 2003, no. 033. [200] Kim, S. J.; Shin, S. R.; Lee, K. B.; Park, Y. D; Kim, S. I. J. Appl. Polym. Sci. 2004, 91, 2908- 2913. [201] Michaels, A. S. In Advances in Separations and Purifications, Peny, E. S., Ed., John Wiley and Sons Inc.: New York, 1968. [202] Schwarz, H. H.; Richau, K.; Paul, D. Polym. Bull. 1991, 25, 95-100. [203] Kim, K. Y.; Min, D. S. Trans. Soc. Biomater. 1988, 11, 658-661. [204] Torrado, S.; Prada, P.; de la Torre, P. M.; Torrado, M. Biomaterials, 2004, 25, 917-923. [205] Mi, F. L.; Her, N. L.; Kaun, C. Y.; Wong, T.; Shyu, S. J. Appl. Polym. Sci. 1997, 66, 2495-2502. [206] Duan, B.; Dong, C., Yuan, X.; Yao, K. J. Biomater. Sci. Polymer Edn., 2004, 15, 797– 811. [207] Ohkawa, K.; Cha, D.; Kim, H.; Nishida, A.; Yamamoto, H. Macromol. Rapid Commun., 2004, 25, 1600–1605. [208] De Vrieze, S.; Westbroek, Ph.; Van Camp, T.; Van Langenhove, L. J. Mater. Sci., 2007, 42, 8029-8034. [209] Greiner, A.; Wendorff, J.H. Angew. Chem. Int. Ed., 2007, 46, 5670–5703. [210] Torres-Giner, S.; Ocio, M.J.; Lagaron, J.M. Eng. Life Sci., 2008, 8, 303–314. [211] Klossner, R.R.; Queen, H.A.; Coughlin, A.J.; Krause, W.E. Biomacromolecules, 2008, 9, 2947-2953. [212] Nikolova, A.; Manolova, N.; Rashkov, I. Polym. Bull., 1998, 41, 115-121. [213] Mucha, M. React. Funct. Polymers, 1998, 28, 19-25. [214] Jia Y.-T.; Gong, J.; Gu, X.-H.; Kim, H.-Y.; Dong, J.; Shen X.-Y. Carbohydr. Polym., 2007, 67, 403–409. [215] Zhang, Y.; Huang, X.; Duan, B.; Wu, L.; Li, S.; Yuan, X.; Colloid Polym. Sci. 2007, 285, 855–863. [216] Zhang, Y.Z.; Su, B.; Ramakrishna, S.; Lim C.T. Biomacromolecules, 2008, 9, 136-141. [217] Lee, H.W.; Karim, M.R.; Park, J.H.; Ghim, H.D.; Choi, J.H.; Kim, K.; Deng, Y.; Yeum J.H. J. Appl. Polym. Sci., 2009, 111, 132–140. [218] Bhattarai, N.; Edmondson, D.; Veiseh, O.; Matsen, F.A.; Zhang, M. Biomaterials, 2005, 26, 6176–6184. [219] Li, L.; Hsieh Y.-L. Carbohydr. Res., 2006, 341, 374–381. [220] Son, B.; Yeom, B.-Y.; Song, S.H.; Lee, C.-S.; Hwang, T.S. J. Appl. Polym. Sci., 2009, 111, 2892–2899. [221] Yang, D.; Jin, Y.; Zhou, Y.; Ma, G.; Chen, X.; Lu, F.; Nie, J. Macromol. Biosci., 2008, 8, 239–246. [222] Yang, D.; Jin, Y.; Ma, G.; Chen, X.; Lu, F.; Nie, J. J. Appl. Polym. Sci., 2008, 110, 3328–3335.

148

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

[223] Subramanian, A.; Vu, D.; Larsen, G.F.; Lin H.-Y. J. Biomater. Sci. Polymer Edn, 2005, 16, 861–873. [224] Desai, K.; Kit, K.; Li, J.; Zivanovic, S. Biomacromolecules, 2008, 9, 1000-1006. [225] Lou, C.-W.; Lin, J.-H.; Yen, K.-C.; Lu, C.-T.; Lee, C.-Y. Textile Res. J., 2008, 78, 254– 257. [226] Vondran, J. L.; Sun, W.; Schauer, C. L. J. Appl. Polym. Sci., 2008, 109, 968–975. [227] Kriegel, C.; Kit, K. M.; McClements, D. J.; Weiss, J. Polymer, 2009, 50, 189–200. [228] Lin, T.; Fang, J.; Wang, H.; Cheng, T.; Wang, X. Nanotechnology, 2006, 17, 3718– 3723. [229] Zhou, Y.; Yang, D.; Nie, J. J. Appl. Polym. Sci., 2006, 102, 5692–5697. [230] Desai, K.; Kit, K. Polymer, 2008, 49, 4046–4050. [231] Duan, B.; Wu, L.; Li, X.; Yuan, X.; Li., X.; Zhang, Y.; Yao, K. J. Biomater. Sci. Polymer Edn, 2007, 18, 95–115. [232] Deitzel, J.M.; Kleinmeyer, J.; Harris, D.; Beck Tan. N.C. Polymer, 2001, 42, 261–272. [233] Penchev, H.; Paneva, D.; Manolova, N.; Rashkov, I. Macromol. Biosci., 2009, 9, 000000, on line, DOI: 10.1002/ mabi.200900003. [234] Sanpui, P., Murugadoss, A.; Prasad, P. V. D.; Ghosh, S. S., Chattopadhyay, A., Int. J. Food Microbiology, 2008, 124, 142-146. [235] Murugadoss, A.; Chattopadhyay, A. Nanotechnology, 2008, 19, 1-9. [236] Lu, S.; Gao, W.; Gu, H.-Y. Burns, 2008, 34, 623-628. [237] Twu, Y.-K.; Chen, Y.-W.; Shih, C.-M. Powder Techn., 2008, 185, 251-257. [238] Huang, C.J.; Lin, J.J.; Shieu F.-S. Jap. J. Appl. Phys., 2005, 44, 6332-6340. [239] Ojha, S. S; Stevens, D. R.; Hoffman, T. J.; Stano, K.; Klossner, R.; Scott M. C.; Krause, W.; Clarke, L. I.; Gorga, R. E. Biomacromolecules, 2008, 9, 2523-2529. [240] Ohkawa, K.; Minato, K.-I.; Kumagai, G.; Hayashi, S.; Yamamoto, H. Biomacromolecules, 2006, 9, 3291-3294. [241] Schiffman, J. D.; Schauer, C. L. Biomacromolecules, 2007, 8, 594-601. [242] Schiffman, J. D.; Schauer, C. L. Biomacromolecules, 2007, 8, 2665-2667. [243] Sangsanoh, P.; Supaphol P. Biomacromolecules, 2006, 7, 2710-2714. [244] Jung, K.-H.; Huh, M.-W.; Meng, W.; Yuan, J.; Hyun S. H.; Bae, J.-S.; Hudson, S. M.; Kang, I.-K. J. Appl. Polym. Sci., 2007, 105, 2816–2823. [245] Xu, J.; Zhang, J.; Gao, W.; Liang, H.; Wang, H.; Li, J. Materials Letters, 2009, 63, 658–660. [246] Ignatova, M.; Manolova, N.; Markova, N.; Rashkov, I. Macromol. Biosci., 2009, 9, 102–111. [247] Dinan, B.; Bhattarai, N.; Li, Z.; Zhang, M. J. Undergrad. Research Bioengineering, pp. 1-5 (www.uweb.engr.washington.edu/education/ pdf/Dinan2007.pdf) [248] Min, B.-M.; Lee, S. W.; Lim, J. N.; You, Y.; Lee, T. S.; Kang, P. H.; Park, W. H. Polymer, 2004, 45, 7137–7142. [249] Mi, F.-L.; Sung, H.-W.; Shyu, S.-S.; Su, C.-C.; Peng, C.-K. Polymer, 2003, 44, 6521– 6530. [250] Mi, F.-L. Biomacromolecules, 2005, 6, 975-987. [251] Butler, M. F.; Ng, Y.-F.; Pudney, P. D. A. J. Polym. Sci. Part A: Polym. Chem., 2003, 41, 3941–3953. [252] Ravi Kumar M. N. V., React. Funct. Polymers, 2000, 46, 1–27.

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 149 [253] Cho, J. H.; Kim, S.-H.; Park, K. D.; Jung, M. C.; Yang, W. I.; Han, S. W.; Noh, J. Y.; Lee J. W. Biomaterials, 2004, 25, 5743–5751. [254] Harris, L.G.; Mead, L.; Muller-Oberlander, E.; Richards, R.G. Eur. Cell. Mater., 2005, 10, 24. [255] Vårum, K. M.; Myhr, M. M.; Hjerde, R. J. N.; Smidsrød, O. Carbohydr. Res., 1997, 299, 99-101. [256] Ma, P. Materials Today, 2004, 7, 30-40. [257] Bhattarai, S. R.; Bhattarai, N.; Viswanathamurthi, P.; Yi, H. K.; Hwang, P. H.; Kim, H. Y. J. Biomed. Mater. Res., 2006, 78, 247–257. [258] Yu, S. H.; Mi, F. L.; Shyu, S. S.; Tsai, C. H.; Peng, C. K.; Lai, J. Y. J. Membr. Sci., 2006, 276, 68-80. [259] Engibaryan, L. G.; Chernukhina, A. I.; Gabrielyan, G. A.; Gal’braikh, L. S., Fibre Chem. 2005, 37, 285-288. [260] Neamnark, A.; Rujiravanit, R.; Supaphol, P. Carboh. Polym. 2006, 66, 298–305. [261] Du, J.; Hsieh, Y.-L. Cellulose 2007, 14, 543-552, [262] Chen, S. C.; Wu, Y. C.; Mi, F. L.; Lin, Y. H.; Yu, L. C.; Sung, H. W. J. Control. Release 2004, 96, 285-300. [263] Kogan, G.; Skorik, Y.A.; Žitňanová, I.; Križková, L; Ďuračková, Z; Gomes, C. A. R.; Yatluk, Y. G.; Krajčovi, J. Toxicol. Appl. Pharmacol. 2004, 201, 303–310. [264] Orienti, I.; Luppi, B.; Zecchi, V. J. Cosmet. Sci. 1999, 50, 307–313. [265] Jia, Z.; Shen, D.; Xu, W. Carbohydr. Res. 2001, 333, 1-6. [266] Kim, J. Y.; Lee, J. K.; Lee, T. S.; Park, W. H. Int. J. Biol. Macromol. 2003, 32, 23-27. [267] Chen, R. N.; Wang, G. M.; Chen, C. H.; Ho, H. O.; Sheu, M. T. Biomacromolecules 2006, 7, 1058–1064. [268] Chen, X. G.; Wang, Z.; Liu, W. S.; Park, H. J. Biomaterials 2002, 23, 4609–4614. [269] Zhou, Y.; Yang, D.; Chen, X.; Xu, Q.; Lu, F.; Nie, J. Biomacromolecules 2008, 9, 349354. [270] Du, J.; Hsieh, Y.-L. Nanotechnology 2008, 19, art. no. 125707. [271] Lee, K. Y.; Ha, W. S.; Park, W. H. Biomaterials 1995, 16, 1211–1216. [272] Hirano, S.; Noishiki, Y. J. Biomed. Mater. Res., 1985,19, 413–417. [273] Peesan, M.; Rujiravanit, R.; Supaphol, P. J. Biomater. Sci. Polymer Edn., 2006, 17, 547-565. [274] Harris, J. M., Ed. Poly(ethylene glycol) Chemistry: Biotechnical and Biomedical Applications. Plenum Press: New York, 1992. [275] Tirelli, N.; Lutolf, M. P.; Napoli, A.; Hubbell, J. A. Rev. Mol. Biotechnol., 2002, 90, 3– 15. [276] Skotak, M.; Leonov, A. P.; Larsen, G.; Noriega, S.; Subramanian, A. Biomacromolecules 2008, 9, 1902-1908. [277] Kumeta, K.; Nagashima, I.; Matsui, S.; Mizoguchi, K. J. Appl. Polym. Sci. 2003, 90, 2420-2427. [278] Huang, L.; Nagapudi, K.; Apkarian, R.; Chaikof, E. L. J. Biomater. Sci., Polym. Edn. 2001, 12, 979-993. [279] Jin, H. J.; Fridrikh, S.; Rutledge, G. C.; Kaplan, D. Abstr. Pap. Am. Chem. Soc. 2002, 224, 408-409. [280] Casper, C. L.; Yamaguchi, N.; Kiick K. L.; Rabolt, J. F. Biomacromolecules 2005, 6, 1998-2007.

150

D. Paneva, Ϡ. Ignatova, N. Manolova et al.

[281] Katti, D. S.; Robinson, K. W.; Ko, F. K.; Laurencin, C. T. J. Biomed. Mater. Res. - Part B Appl. Biomaterials, 2004, 70, 286-296. [282] Verreck, G.; Chun, I.; Rosenblatt, J.; Peeters, J.; Van Dijck, A.; Mensch, J.; Noppe, M.; Brewster, M. E. J. Controlled Release 2003, 92, 349-360. [283] Melaiye, A.; Sun, Z.; Hindi, K.; Milsted, A.; Ely, D.; Reneker, D. H.; Tessier, C. A.; Youngs, W. J. J. Am. Chem. Soc. 2005, 127, 2285-2291. [284] Yu, D.-G.; Teng, M.-Y.; Chou, W.-L.; Yang, M.-C. J. Membr. Sci. 2003, 225, 115-123. [285] Son, W. K.; Youk, J. H.; Lee T. S.; Park, W. H. Macromol. Rapid Commun. 2004, 25, 1632-1637. [286] Xu, X.; Yang, Q.; Wang, Y.; Yu, H.; Chen, X.; Jing, X. Eur. Polym. J. 2006, 42, 20812087. [287] Hong, K. H.; Park, J. L.; Sul, I. H.;. Youk, J. H; Kang, T. J. J. Polym. Sci. Part B: Polym. Physics. 2006, 44, 2468-2474. [288] Hong K. H., Polym. Engineer. Sci. 2007, 47, 43-49. [289] Ignatova, M.; Markova, N.; Manolova, N.; Rashkov, I. J. Biomater. Sci. Polymer Edn., 2008, 19, 373-386. [290] Kenawy, E.-R.; Abdel-Fattah, Y. R. Macromol. Biosci. 2002, 2, 261-266. [ 291] Ji, Y.; Ghosh, K.; Li, B.; Sokolov, J. C.; Clark R. A. F.; Rafailovich, M. H. Macromol. Biosci. 2006, 6, 811- 817. [292] Rho, K. S.; Jeong, L.; Lee, G.; Seo, B.-M.; Park, Y. J.; Hong, S.-D.; Roh, S.; Cho, J. J.; Park, W. H.; Min, B.-M. Biomaterials 2006, 27, 1452-1461. [293] Khil, M. S.; Cha, D. I.; Kim, H. Y.; Kim I. S.; Bhattarai, N. J. Biomed. Mater. Res. 2003, 67B, 675-679. [294] Franklin, T. J.; Snow, G. A. Biochemistry of Antimicrobial Action, Chapman and Hall: London, 1981; pp 58-60. [295] Teixeira, A. I.; Abrams, G. A.; Bertics, P. J.; Murphy, C. J.; Nealey, P F. J. Cell Sci. 2003, 116, 1881–1892. [296] Lee, C. H.; Shin, H. J.; Cho, I. H.; Kang, Y. M.; Kim, I. A.; Park, K. D.; Shin, J. W. Biomaterials 2005, 26, 1261–1270. [297] Li, W. J.; Laurencin, C. T.; Caterson, E. J.; Tuan, R. S.; Ko, F. K. J. Biomed. Mater. Res. 2002, 60, 613–621. [298] Neamnark, A.; Sanchavanakit, N.; Pavasant, P.; Rujiravanit, R.; Supaphol, P. Eur. Polym. J., 2008, 44, 2060-2067. [299] Jiang, H.; Fang, D.; Hsiao, B.; Chu, B.; Chen, W. J. Biomater. Sci., Polym. Ed. 2004, 15, 797–811. [300] Ouchi, T.; Nishizawa, H.; Ohya, Y. Polymer 1998, 39, 5171–5175. [301] Kim, S.H.; Kim, S.; Nair, S.; Moore, E. Macromolecules, 2005, 38, 3719-3723. [302] Hennink, W. E.; van Nostrum, C. F. Adv. Drug Delivery Rev., 2002, 54, 13-36. [303] Xu, X. US Patent WO 2005/026398 A2, 2004. [304] Baker, D. A.; Brown Ph. J. Polym. Preprints, 2003, 44, 118-119. [305] Gupta, R.; Trenor, S. R.; Long, T. E.; Wilkes, G. L. Macromolecules, 2004, 37, 92119218. [306] Muzarelli, R. A. A.; Barontini, G.; Rochetti, R. Biotechnolol. Bioeng., 1975, 18, 14451454. [307] Richards, F. M.; Knowles, J. R. J. Mol. Biol., 1968, 37, 231-233. [308] Monteiro Jr., O. A. C.; Airoldi C. Int. J. Biol. Macromol., 1999, 26, 119-128.

Novel Chitosan–Containing Micro- and Nanofibrous Materials by Electrospinning… 151 [309] Yancheva, E.; Paneva, D.; Maximova, V.; Mespouille, L.; Dubois, Ph.; Manolova, N.; Rashkov, I. Biomacromolecules, 2007, 8, 976-984. [310] Koetz, J.; Kosmella, S. Polyelectrolytes and Nanoparticles. Springer-Verlag Berh Heidelberg 2007. [311] Chavasit, V.; Klenzle-Sterzer, C.; Torres, J.A. Polym. Bull., 1988, 19, 223-230. [312] Dumitriu, S.; Chornet, E. Adv. Drug Deliv. Rev., 1998, 31, 223-246. [313] Wang, X.; Kim, Y.-G.; Drew, Ch.; Ku, B.-Ch.; Kumar, J.; Samuelson, L.A. Nano Lett., 2004, 4, 331-334. [314] Decher, G., Hong, J.D., Schmitt, J. Adv. Materials, 1991, 3, 617-619. [315] Decher, G. Science, 1997, 277, 1232-1237. [316] Crisp, M. T.; Kotov, N. A. Nano Lett., 2003, 3, 173-177. [317] Caruso, F.; Shi, X.; Caruso, R. A.; Susha, A. Adv. Materials, 2001, 13, 740-744. [318] Mayya, K. S.; Gittins, D. I.; Dibaj, A. M.; Caruso, F. Nano Lett., 2001, 1, 727-730. [319] Li B., Shen L., Liu X., Zhang S., Wu C., Liu W. Mater. Sci. Eng. A, 2004, 364: 324332. [320] Gittins, D. I.; Caruso, F. Adv. Mater., 2000, 12, 1947-1949. [321] Caruso, F.; Caruso, R. A.; Mohwald, H. Science, 1998, 282, 1111-1114. [322] Haack, J. M.; Lenk, W.; Lehmann, D.; Lunkwitz, K. J. Membr. Sci., 2001, 184, 233243. [323] Lajimi, R. H.; Abdallah, A. B.; Ferjani, E.; Roudesli, M.S.; Deratani, A. Desalination, 2004, 163,193-202. [324] Wang, X.; Fang, D.; Yoon, K.; Hsiao, B. S.; Chu, B. J. Membr. Sci., 2006, 278, 261268. [325] Wang, X.; Chen, X.; Yoon, K.; Fang, D.; Hsiao, B. S.; Chu, B. Environm. Sci. Techn., 2005, 39, 7684-7691. [326] Ding, B.; Fujimoto, K.; Shiratoroi, S. Thin Solid Films, 2005, 491, 23-28. [327] Zezin, A. B.; Lutsenko, V .V.; Rogacheva, V. B.; Aleksina, O. A.; Kalyuzhnaya, R. I.; Kabanov, V. A.; Kargin, V. A. Polymer Science U.S.S.R., 1972, 14, 857-865. [328] Ritcharoen, W.; Supaphol, P.; Pavasant, P. Eur. Polym J., 2008, 44, 3963-3968. [329] Müller, K.; Quinn, J. F.; Johnston, A. P. R.; Becker, M.; Greiner, A.; Caruso, F. Chem. Mater., 2006, 18, 2397-2403. [330] Ge, L.; Pan, C.; Chen, H.; Wang, X.; Wang, C.; Gu, Z. Colloids Surf. A, 2007, 293, 272-277. [331] Chen, Z.; Mo, X.; Qing, F. Matter. Lett., 2007, 61, 3490-3494. [332] Chunder, A.; Sarkar, S.; Yu, Y.; Zhai, L. Colloids Surf. B: Biointerf., 2007, 58, 172– 179. [333] Chen, Z.; Mo, X.; He, Ch.; Wang, H. Carbohydr. Polym., 2008, 72, 410–418. [334] Penchev, H.; Paneva, D.; Manolova, N.; Rashkov, I. Macromol. Rapid Commun., 2008, 29, 677-681. [335] Michaels, A. S.; Miekka, R. G. J. Chem. Phys., 1961, 65, 1765-1773. [336] Rashkov, I.; Penchev, H unpublished results. [337] Okuzaki, H.; Takahashi, T.; Miyajima, N.; Suzuki, Y.; Kuwabara, T. Macromolecules, 2006, 39, 4276-4278. [338] Okuzaki, H.; Takahashi, T.; Hara, Y.; Yan, H. J. Polym. Sci.: Part B: Polym. Phys., 2008, 46, 305–310.

In: Nanofibers: Fabrication, Performance, and Applications ISBN 978-1-60741-947-1 Editors: W. N. Chang © 2009 Nova Science Publishers, Inc.

Chapter 4

A NOVEL APPROACH FOR ANALYSIS OF PROCESSING PARAMETERS IN ELECTROSPINNING OF NANOFIBERS M. Ziabari1, V. Mottaghitalab1 and A. K. Haghi1,2 1

University of Guilan, Rasht, Iran 2 University of Ottawa, Canada

ABSTRACT The precise control of fiber diameter during electrospinning is very crucial for many applications. A systematic and quantitative study on the effects of processing variables enables us to control the properties of electrospun nanofibers. In this contribution, response surface methodology (RSM) was employed to quantitatively investigate the simultaneous effects of four of the most important parameters, namely solution concentration (C), spinning distance (d), applied voltage (V) and volume flow rate (Q) on mean fiber diameter (MFD) as well as standard deviation of fiber diameter (StdFD) in electrospinning of polyvinyl alcohol (PVA) nanofibers.

Keywords: Electrospinning, Nanofibers, Fiber diameter, Processing variables, Response surface methodology

INTRODUCTION Electrospinning is a novel and efficient method by which fibers with diameters in nanometer scale entitled as nanofibers, can be achieved. In electrospinning process, a strong electric field is applied on a droplet of polymer solution (or melt) held by its surface tension at the tip of a syringe's needle (or a capillary tube). As a result, the pendent drop will become highly electrified and the induced charges are distributed over its surface. Increasing the intensity of electric field, the surface of the liquid drop will be distorted to a conical shape known as the Taylor cone [1]. Once the electric field strength exceeds a threshold value, the repulsive electric force dominates the surface tension of the liquid and a stable jet emerges

154

M. Ziabari, V. Mottaghitalab and A. K. Haghi

from the cone tip. The charged jet is then accelerated toward the target and rapidly thins and dries as a result of elongation and solvent evaporation. As the jet diameter decreases, the surface charge density increases and the resulting high repulsive forces split the jet to smaller jets. This phenomenon may take place several times leading to many small jets. Ultimately, solidification is carried out and fibers are deposited on the surface of the collector as a randomly oriented nonwoven mat [2]-[5]. Figure 1 shows a schematic illustration of electrospinning setup.

Figure 1. Electrospinning setup [6].

Featuring various outstanding properties such as very small fiber diameters, large surface area per mass ratio [3], high porosity along with small pore sizes [7], flexibility, and superior mechanical properties [8], electrospun nanofiber mats have found numerous applications in biomedical (tissue engineering [9]-[11], drug delivery [12], [13], and wound dressing [14], [15]), protective clothing [7], filtration [16], reinforcement in composite materials [8], [17], and micro-electronics (battery [18], supercapacitors [19], transistors [20], sensors [21], and display devices [22]). Morphology of electrospun nanofibers such as fiber diameter, depend on many parameters which are mainly divided into three categories: solution properties (solution viscosity, solution concentration, polymer molecular weight, and surface tension), processing conditions (applied voltage, volume flow rate, spinning distance, and needle diameter), and ambient conditions (temperature, humidity, and atmosphere pressure) [23]. As mentioned earlier, electrospun nanofibers have numerous applications some of which have been commercialized. Most of these applications require nanofibers with desired properties suggesting the importance of the process control. This end may not be achieved unless having a comprehensive outlook of the process and quantitative study of the effects of governing parameters which makes the control of the process possible. In addition, qualitative description of the experimental observations are not adequate to derive general conclusions and either the equations governing behavior of the system must be found or appropriate empirical models need be presented. In Ziabicki's words, “in the language of science ‘to explain’ means to put forward a quantitative model which is consistent with all the known date and capable of predicting new fact” [24]. Employing a model to express the influence of electrospinning parameters will help us obtain a simple and systematic way for presenting the

A Novel Approach for Analysis of Processing Parameters…

155

effects of variables thereby enabling the control of the process. Furthermore, it allows us to predict the results under new combination of parameters. Hence, without conducting any experiments, one can easily estimate features of the product in unknown conditions. That's to say, a model tells us to what extent the output of a system will change if one or more parameters increased or decreased. This is very helpful and leads to a detailed understanding of the process and the effects of parameters. Despite the surge in attention to the electrospinning process, a few investigations have addressed the quantitative study of the effects of the parameters which has hindered the control of the process. Changing the behavior of materials in nano-scale, presence of electric field, branching of the jet, random orientation of fibers, etc. made the analysis of the process extremely complex and difficult that to date there has been no reliable theory capable of describing the phenomenon. Furthermore, the development of an empirical model has also been impeded due to the lack of systematic and characterized experimentations with appropriate designs. Adding to the difficulty is the number of parameters involving in the electrospinning process and the interactions between them which made it almost impossible to investigate the simultaneous effects of all variables. Affecting the characteristics of the final product such as physical, mechanical and electrical properties, fiber diameter is one of the most important structural features in electrospun nanofiber mats. Podgorski et al. [25] indicated that filters made of fibers with smaller diameters have higher filtration efficiencies. This was also proved by the work of Qin et al. [16]. Ding et al. [26] reported that sensitivity of sensors increase with decreasing the mean fiber diameter – due to the higher surface area. In the study on designing polymer batteries consisting of electrospun PVdF fibrous electrolyte by Kim et al. [27], it was demonstrated that lower mean fiber diameter results in a higher electrolyte uptake thereby increased ionic conductivity of the mat. Moroni et al. [28 found fiber diameters of electrospun PEOT/PBT scaffolds influencing on cell seeding, attachment, and proliferation. They also studied the release of dye incorporated in electrospun scaffolds and observed that with increasing fiber diameter, the cumulative release of the dye (methylene blue) decreased. Carbonization and activation conditions as well as the structure and properties of the ultimate carbon fibers are also affected by the diameters of the precursor PAN nanofibers [29]. Consequently, precise control of the electrospun fiber diameter is very crucial. Sukigara et al. [30] employed response surface methodology (RSM) to model mean fiber diameter of electrospun regenerated Bombyx mori silk with electric field and concentration at two spinning distances. They applied a full factorial experimental design at three levels of each parameter leading to nine treatments of factors and used a quadratic polynomial to establish a relationship between mean fiber diameter and the variables. Increasing the concentration at constant electric field resulted in an increase in mean fiber diameter. Different impacts for the electric field were observed depending on solution concentration. Trend of the effects of the two parameters on mean fiber diameter varied with changing the spinning distance which suggests the presence of interaction and coupling between the parameters. Gu et al. [31] and Gu et al. [32] also exploited the RSM for quantitative study of PAN and PDLA respectively. The only difference observed in the procedure was the use of four levels of concentration in the case of PAN. They included the standard deviation of fiber diameter in their investigations by which they were able to provide additional information regarding the morphology of electrospun nanofibers and its variations at different conditions.

156

M. Ziabari, V. Mottaghitalab and A. K. Haghi

Furthermore, they analyzed the significance of the factors in the models in order to understand the level of influence of each parameter. In the case of PAN, voltage as well as its interaction with concentration had no significant effects on both mean and standard deviation of fiber diameter. Hence, they eliminated the terms corresponding to these factors thereby obtained simplified quadratic models according to which mean and standard deviation of fiber diameter increased with polymer concentration. On the contrary, both voltage and its interaction with concentration were found to be significant in the case of PDLA. However, the effect of polymer concentration was more pronounced. Increasing voltage at constant concentration favored thinner fiber formation which gained momentum with increasing concentration. Fibers with more uniform diameters (less standard deviation) were obtained at higher applied voltage or concentration. In the most recent investigation in this field, Yördem et al. [33] utilized RSM to correlate mean and coefficient of variation (CV) of electrospun PAN nanofibers to solution concentration and applied voltage at three different spinning distances. They employed a face-centered central composite design (FCCD) along with a full factorial design at two levels resulting in 13 treatments at each spinning distance. A cubic polynomial was then used to fit the data in each case. As with previous studies, fiber diameter was very sensitive to changes in solution concentration. Voltage effect was more significant at higher concentrations demonstrating the interaction between parameters. Despite high reported R 2 values, the presented models seemed to be inefficient and uncertain. Some terms in the models had very high p-values. For instance, in modeling the mean fiber diameter, p-value as high as 0.975 was calculated for cubic concentration term at spinning distance of 16 cm, where half of the 2 values which were not reported in terms had p-values more than 0.8. This results in low Rpred their study and after calculating by us were found to be almost zero in many cases suggesting the poor prediction ability of their models. As it was mentioned by the previous authors, there are some interactions between electrospinning parameters. In the past studies, however, they only investigated the simultaneous effects of two variables; therefore they were unable to thoroughly capture the interactions which exist between the parameters. For instance, Sukigara et al. [30] and Yördem et al. [33] both agreed that spinning distance has a significant influence on fiber diameter and that this effect varies when solution concentration and/or applied voltage altered. However, they could not describe their findings in terms of quantitative relationships. Hence, the presented models suffer from lack of comprehensiveness. In addition, in every research where modeling of a process is targeted, the obtained models need to be evaluated with a set of test data which were not used in establishing the relationships. Otherwise, the effectiveness of the models will not be guaranteed and there will always be an uncertainty in the prediction of the models in new conditions. Hence, it is possible for a model very efficient in describing experimental data, to present unsatisfactory prediction results. In none of the previous works, however, the presented models were evaluated with a series of test data. Therefore, their models may not generalize well to new data and their prediction ability is obscure. In this contribution for the first time, the simultaneous effects of four electrospinning parameters (solution concentration, spinning distance, applied voltage, and volume flow rate) on mean and standard deviation of polyvinyl alcohol (PVA) fiber diameter were systematically investigated. PVA, the largest volume synthetic water-soluble polymer

A Novel Approach for Analysis of Processing Parameters…

157

produced in the world, is commercially manufactured by the hydrolysis of polyvinyl acetate. The excellent chemical resistance and physical properties of PVA along with non-toxicity and biodegradability have led to its broad industrial applications such as textile sizing, adhesive, paper coating, fibers, and polymerization stabilizers [34], [35]. Several patents reported process for production of ultrahigh tensile strength PVA fibers comparable to Kevlar® [36][38]. PVA has found many applications in biomedical uses as well due to its biocompatibility [39]. For instance, PVA hydrogels were used in regenerating articular cartilages [40], [41], artificial pancreas [42], and drug delivery systems [43], [44]. More recently, PVA nanofibers were electrospun and used as a protein delivery system [45], retardation of enzyme release [45] and wound dressing [46]. The objective of this paper is to use RSM to establish quantitative relationships between electrospinning parameters and mean and standard deviation of fiber diameter as well as to evaluate the effectiveness of the empirical models with a set of test data.

EXPERIMENTAL Solution Preparation and Electrospinning PVA with molecular weight of 72000 g/mol and degree of hydrolysis of >98% was obtained from Merck and used as received. Distilled water as solvent was added to a predetermined amount of PVA powder to obtain 20 ml of solution with desired concentration. The solution was prepared at 80°C and gently stirred for 30 min to expedite the dissolution. After the PVA had completely dissolved, the solution was transferred to a 5 ml syringe and became ready to electrospin. The experiments were carried out on a horizontal electrospinning setup shown schematically in Figure 1. The syringe containing PVA solution was placed on a syringe pump (New Era NE-100) used to dispense the solution at a controlled rate. A high voltage DC power supply (Gamma High Voltage ES-30) was used to generate the electric field needed for electrospinning. The positive electrode of the high voltage supply was attached to the syringe needle via an alligator clip and the grounding electrode was connected to a flat collector wrapped with aluminum foil where electrospun nanofibers were accumulated to form a nonwoven mat. The electrospinning was carried out at room temperature. Subsequently, the aluminum foil was removed from the collector. A small piece of mat was placed on the sample holder and gold sputter-coated (Bal-Tec). Thereafter, the morphology of electrospun PVA fibers was observed by an environmental scanning electron microscope (SEM, Phillips XL-30) under magnification of 10000X. For each specimen, fiber diameter distribution was determined from the SEM micrograph based on 100 measurements of random fibers. A typical SEM micrograph of electrospun nanofiber mat and its corresponding diameter distribution are shown in Figure 2.

158

M. Ziabari, V. Mottaghitalab and A. K. Haghi

(a)

(b)

Figure 2. (a) a typical SEM micrograph of electrospun nanofiber mat, (b) its corresponding diameter distribution.

CHOICE OF PARAMETERS AND RANGE As it was mentioned earlier in this paper, the number of variables which have the potential to alter the electrospinning process is numerous. Hence, investigating all of them in the framework of one single research would almost be impossible. However, some of these parameters can be held constant during experimentation. For instance, performing the experiments in a controlled environmental condition, which is concerned in this study, the ambient parameters (i.e. temperature, air pressure, and humidity) are kept unchanged. Solution viscosity is affected by polymer molecular weight, solution concentration, and temperature. For a particular polymer (constant molecular weight) at a fixed temperature, solution concentration would be the only factor influencing the viscosity. In this circumstance, the effect of viscosity could be determined by the solution concentration. Therefore, there would be no need for viscosity to be considered as a separate parameter. In this regard, solution concentration (C), spinning distance (d), applied voltage (V), and volume flow rate (Q) were selected to be the most influential parameters in electrospinning of PVA nanofibers as for the purpose of this study. The next step is to choose the region of interest – that is the ranges over which these factors are varied. Process knowledge, which is a combination of practical experience and theoretical understanding, is required to fulfill this step. The aim is here to find an appropriate range for each parameter where dry, bead-free, stable, and continuous fibers without breaking up to droplets are obtained. This goal could be achieved by conducting a set of preliminary experiments while having the previous works in mind along with utilizing the reported relationships. The relationship between intrinsic viscosity ( [η ] ) and molecular weight (M) is given by the well-known Mark-Houwink-Sakurada equation as follows: [η ] = KM a

(1)

A Novel Approach for Analysis of Processing Parameters…

159

where K and a are constants for a particular polymer-solvent pair at a given temperature [47]. For the PVA with molecular weight in the range of 69000 g/mol 9. Therefore, the appropriate range in this case could be found within 5