In Situ Synthesis of Polymer Nanocomposites

Edited by Vikas Mittal In-situ Synthesis of Polymer Nanocomposites Polymer Nano-, Micro- & Macrocomposites Series Mitt...

Author: Vikas Mittal | Editor

217 downloads 1520 Views 5MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Edited by Vikas Mittal In-situ Synthesis of Polymer Nanocomposites

Polymer Nano-, Micro- & Macrocomposites Series Mittal, V. (ed.)

Mittal, V. (ed.)

Surface Modiﬁcation of Nanotube Fillers

Characterization Techniques for Nanocomposites

Series: Polymer Nano-, Micro- and Macrocomposites (Volume 1) 2011 ISBN: 978-3-527-32878-9

Series: Polymer Nano-, Micro- and Macrocomposites (Volume 3) 2012 ISBN: 978-3-527-33148-2

Related Titles Mittal, V. (ed.)

Matyjaszewski, K., Müller, A. H. E. (eds.)

Optimization of Polymer Nanocomposite Properties

Controlled and Living Polymerizations

2010 ISBN: 978-3-527-32521-4

From Mechanisms to Applications 2010 ISBN: 978-3-527-32492-7

Mittal, V. (ed.)

Miniemulsion Polymerization Technology 2010 ISBN: 978-0-470-62596-5

Cosnier, S., Karyakin, A. (eds.)

Electropolymerization Concepts, Materials and Applications 2010 ISBN: 978-3-527-32414-9

Mittal, V. (ed.)

Polymer Nanotube Nanocomposites

Leclerc, M., Morin, J.-F. (eds.)

Synthesis, Properties, and Applications

Design and Synthesis of Conjugated Polymers

2010 ISBN: 978-0-470-62592-7

2010 ISBN: 978-3-527-32474-3

Pascault, J.-P., Williams, R. J. J. (eds.)

Epoxy Polymers New Materials and Innovations 2010 ISBN: 978-3-527-32480-4

Edited by Vikas Mittal

In-situ Synthesis of Polymer Nanocomposites

The Editor Dr. Vikas Mittal The Petroleum Institute Chemical Engineering Department Abu Dhabi UAE

All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate. Library of Congress Card No.: applied for British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliograﬁe; detailed bibliographic data are available on the Internet at . © 2012 Wiley-VCH Verlag & Co. KGaA, Boschstr. 12, 69469 Weinheim, Germany All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microﬁlm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not speciﬁcally marked as such, are not to be considered unprotected by law. Typesetting Toppan Best-set Premedia Limited, Hong Kong Printing and Binding Fabulous Printers Pte Ltd, Singapore Cover Design Graﬁk-Design Schulz, Fußgönheim Printed in the Federal Republic of Germany Printed on acid-free paper Print ISBN: 978-3-527-32879-6 ePDF ISBN: 978-3-527-64012-6 oBook ISBN: 978-3-527-64010-2 ePub ISBN: 978-3-527-64011-9 ISSN: 2191-0421

V

Contents Preface XIII List of Contributors XV 1 1.1 1.2 1.3

2

2.1 2.2 2.2.1 2.2.2 2.2.3 2.3 2.3.1 2.3.2 2.3.3 2.3.3.1 2.3.3.2 2.3.3.3 2.3.3.4 2.4

In-situ Synthesis of Polymer Nanocomposites 1 Vikas Mittal Introduction 1 Synthesis Methods 9 In-situ Synthesis of Polymer Nanocomposites 12 References 24 Polyamide Nanocomposites by In-situ Polymerization Anastasia C. Boussia, Stamatina N. Vouyiouka, and Constantine D. Papaspyrides Introduction 27 Manufacturing Processes of Commercially Important Polyamides 29 Poly(caproamide) (PA 6) 29 Poly(hexamethylene adipamide) (PA 6.6) 30 Low-Temperature Polymerization Processes 31 Polyamide Nanocomposites 34 Introduction 34 Lactam/Amino Acid-Based In-situ Intercalated PA Nanocomposites 36 Diamine- and Diacid-Based In-situ Intercalated PA Nanocomposites 41 Solution-Melt Polymerization Technique 41 Anhydrous Melt Polymerization Technique 43 Direct SSP Technique 44 Interfacial Polycondensation Technique 46 Conclusions 48 References 49

27

VI

Contents

3 3.1 3.2 3.2.1 3.2.2 3.2.3 3.2.4 3.2.4.1 3.2.4.2 3.3 3.3.1 3.3.2 3.3.3 3.3.4 3.3.4.1 3.3.5 3.3.5.1 3.3.5.2 3.3.5.3 3.3.6 3.3.7 3.3.7.1 3.3.7.2 3.3.7.3 3.3.7.4 3.3.7.5 3.3.8 3.3.8.1 3.3.8.2 3.3.8.3 3.3.8.4 3.3.8.5 3.3.9 3.3.10

4

4.1 4.2 4.3

Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization 53 Abolfazl Maneshi, João Soares, and Leonardo Simon Introduction 53 Clays 54 General Structure 54 Smectites 54 Clay Particle Morphological Hierarchy 56 Clay Chemical Reactions 58 Cation Exchange Reactions 58 Interaction with Organic Compounds 58 In-situ Polymerization of Oleﬁns with Coordination Catalysts Supported on Clays 59 Oleﬁn Polymerization with Coordination Catalysts 60 Polymerization Mechanism with Coordination Catalysts 60 Coordination Catalysts for in In-situ Polymerization 62 Catalyst Supporting 63 Catalyst Supporting Methods 63 Clay Surface Modiﬁcation Methods for In-situ Polymerization 64 Organic Modiﬁcation 64 Thermal Treatment 66 Treatment with Alkylaluminum Compounds 66 Particle Break-Up and Exfoliation 67 In-situ Polymerization Approaches 69 Clay as a Polymerization Additive 71 Clay as a Polymerization Catalyst Support 72 Clay as an Activator for Polymerization Catalysts 74 In-situ Production of Alkylaluminoxanes 76 Other Techniques 76 Factors Determining the Success of In-situ Polymerization 78 Clay Type 78 Swellability 79 Effect of Clay Surface Treatment 80 Catalyst : Clay Ratio 81 Effect of Polymerization Conditions 82 Clay Effect on the Polymerization Behavior and Polymer Molecular Structure 83 Future Approaches 84 References 85 Gas-Phase-Assisted Surface Polymerization and Thereby Preparation of Polymer Nanocomposites 89 Haruo Nishida, Yoshito Andou, and Takeshi Endo Introduction 89 In-situ Polymerization for Nanocomposite Preparation 89 Characteristics of GASP 91

Contents

4.3.1 4.3.2 4.3.3 4.4 4.4.1 4.4.2 4.4.3 4.4.4 4.5

Thin Layer Coating of Solid-Substrate Surfaces 91 Physically Controlled Polymerization Behavior 92 Photo-Induced Controlled Polymerization 93 Composite Preparation by GASP 95 Polymer/Clay Nanocomposites 95 Polymer/Inorganic Compound (Nano)composites 96 Polymer/Cellulose Fiber (Nano)composites 99 Polymer/Carbon Nanotube (Nano)composites 100 Outlook and Perspective 100 Abbreviations 101 References 101

5

PET Clay Nanocomposites by In-situ Polymerization 105 Hua Deng, Ke Wang, Qin Zhang, Feng Chen, and Qiang Fu Introduction 105 Preparation of PET/Clay Nanocomposites 106 Morphology of the Nanocomposites 108 Crystallization of the Nanocomposites 109 Properties of the Nanocomposites 112 Thermal Properties 112 Mechanical Properties 117 Barrier Properties 118 Conclusion and Outlook 121 References 122

5.1 5.2 5.3 5.4 5.5 5.5.1 5.5.2 5.5.3 5.6

6

6.1 6.2 6.2.1 6.2.1.1 6.2.1.2 6.2.2 6.2.2.1 6.2.2.2 6.3 6.3.1 6.3.1.1 6.3.1.2 6.3.1.3 6.3.1.4 6.3.1.5 6.3.2 6.3.2.1

Control of Filler Phase Dispersion in Bio-Based Nanocomposites by In-situ Reactive Polymerization 123 Lawrence A. Pranger, Grady A. Nunnery, and Rina Tannenbaum Introduction 123 Background 125 Polymer Matrix Nanocomposites 125 Cellulose Whisker Nanocomposites 128 Layered Silicate Nanocomposites 132 Reactive Molding Techniques for Composite Manufacture 133 Materials and Methods for Reactive Molding of Nanocomposites 134 Furfuryl Alcohol as a Precursor for Polymer Matrix Composites 135 Experimental Procedures 136 Reactive Molding of Cellulose Whisker Nanocomposites 136 Conceptual Approach 136 Preparation of CW 137 Resiniﬁcation of FA with CW 137 Curing of CW–PFA Composites 137 Characterization Techniques 138 Reactive Molding of MMT Nanocomposites 138 Conceptual Approach 138

VII

VIII

Contents

6.3.2.2 6.3.2.3 6.3.2.4 6.3.2.5 6.4 6.4.1 6.4.1.1 6.4.1.2 6.4.1.3 6.4.2 6.4.2.1 6.4.2.2 6.4.2.3 6.5

Types of MMT Clays Used 139 Resiniﬁcation of FA with MMT Clay 139 Curing of MMT–PFA Composites 139 Characterization Techniques 139 Results and Discussion 140 Reactive Molding of Cellulose Whisker Nanocomposites Morphology of CW 141 Resiniﬁcation of FA in the Presence of CWs 142 Thermal Resistance of CW–FA Nanocomposites 148 Reactive Molding of MMT Nanocomposites 149 Morphology of MMT Clay 150 Resiniﬁcation of FA in the Presence of MMT Clay 150 Thermal Resistance of MMT–FA Nanocomposites 161 Conclusions 164 Abbreviations 164 Acknowledgments 165 References 165

7

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties 169 Mo Song and Dongyu Cai Introduction 169 PU/Carbon Nanotube Nanocomposites (PUCNs) 170 Fabrication 170 Morphology and Characterizations of PUCNs 176 Physical Properties of PUCNs 183 PU/Clay Nanocomposites (PUCLN) 188 Fabrication 189 Exfoliation and Intercalation of Nanoclays in PU Matrix 189 Rheological Behavior of Polyol–Nanoclay Mixture 194 Morphology and Characterization 196 Physical Properties 200 Mechanical Properties 200 Scratch Resistance and Barrier Performance 204 Thermal Stability and Flame Retardancy 207 PU/Functionalized Graphene Nanocomposites (PUFGNs) 208 Fabrication 209 Morphology and Characterization 210 Physical Properties 214 Prospective of PUNs 217 References 218

7.1 7.2 7.2.1 7.2.2 7.2.3 7.3 7.3.1 7.3.1.1 7.3.1.2 7.3.2 7.3.3 7.3.3.1 7.3.3.2 7.3.3.3 7.4 7.4.1 7.4.2 7.4.3 7.5

8 8.1 8.2

140

In-situ Synthesis and Properties of Epoxy Nanocomposites 221 Vikas Mittal Introduction 221 Optimization of the Curing Conditions 222

Contents

8.3 8.4 8.5 8.6 8.7

Fillers, Surface Modiﬁcations, and Ion Exchange 224 Nanocomposite Synthesis 229 Morphology 231 Barrier Properties 238 Effect of Excess Surface Modiﬁcation Molecules 240 References 244

9

Unsaturated Polyester–Montmorillonite Nanocomposites by In-situ Polymerization 245 Michal Kedzierski Introduction 245 Nanocomposites with MMT Introduced into UP Prepolymer or Resin 246 Synthesis, Morphology, and Mechanical Properties 246 Rheology and Cure Properties 253 Flammability 258 Mixed-Resin and Filler Systems 259 Nanocomposites with MMT Introduced during the Synthesis of Prepolymer 260 Conclusions 263 References 265

9.1 9.2 9.2.1 9.2.2 9.2.3 9.2.4 9.3 9.4

10

Polymer Clay Nanocomposites by In-situ Atom Transfer Radical Polymerization 267 Hanying Zhao References 279

11

Polybutadiene Clay Nanocomposites by In-situ Polymerization Giuseppe Leone and Giovanni Ricci Introduction 283 Generalities 284 Clays 284 Polymer Nanocomposite Structures 286 Methods of Preparation of Polymer Nanocomposites 287 Polybutadiene Nanocomposites 287 1,3-Butadiene Polymerization Methods 287 In-situ Anionic Polymerization 289 In-situ Stereospeciﬁc Polymerization 293 Conclusions and Perspectives 299 Abbreviations 299 References 300

11.1 11.2 11.2.1 11.2.2 11.2.3 11.3 11.3.1 11.3.2 11.3.3 11.4

12

12.1

283

P3HT–MWNT Nanocomposites by In-situ Polymerization and Their Properties 303 Zhongrui Li and Liqiu Zheng Introduction 303

IX

X

Contents

12.2 12.3 12.4 12.4.1 12.4.2 12.4.3 12.4.4 12.4.5 12.4.6 12.5

13

13.1 13.2 13.3 13.3.1 13.3.2 13.3.3 13.4 13.4.1 13.4.1.1 13.4.1.2 13.4.1.3 13.4.2 13.4.2.1 13.4.2.2 13.4.2.3 13.4.3 13.5 13.5.1 13.5.1.1 13.5.1.2 13.5.1.3 13.5.1.4 13.5.1.5 13.5.2 13.5.2.1 13.5.2.2 13.5.2.3

Multiwall CNTs 305 In-situ Synthesis of P3HT–MWNT Composites 307 The Properties and Characterization of P3HT–MWNT Nanocomposites 310 The Dispersion and Morphology of the P3HT–MWNT Nanocomposites 310 HT Regioregularity 311 Mechanical Properties 311 Thermal Stability 313 Optical Properties 316 Charge Transportability 321 Conclusion and Outlook 325 References 326 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties 331 Ranya Simons, Greg G. Qiao, and Stuart A. Bateman Introduction 331 Morphology of Polymer–Clay Nanocomposites 331 Modiﬁcation of MMT 332 NonReactive Modiﬁcations 333 Reactive Modiﬁcations 343 Polymeric Initiator-Based Modiﬁcations 345 In-situ Polymerization Methods 346 Free Radical Polymerization Techniques 347 Bulk Polymerization 347 Emulsion Polymerization 348 Solution Polymerization 349 Controlled Polymerization Techniques 350 Atom Transfer Radical Polymerization 351 Reverse Addition-Fragmentation Transfer 351 Nitroxide-Mediated Polymerization 351 Dispersion of MMT in Styrene 352 Properties of PS–MMT Nanocomposites Prepared via In-situ Techniques 352 Mechanical Properties 353 Tensile 353 Impact and Flexural Properties 354 Dynamic Mechanical Thermal Analysis 354 Rheological Properties 355 Barrier Properties 355 Thermal Properties 356 Thermal Gravimetric Analysis 356 Dynamic Scanning Calorimetry (DSC) 358 Fire Performance 359

Contents

13.6

Summary 361 References 362

14

Aliphatic Polyester and Poly(ester amide) Clay Nanocomposites by In-situ Polymerization 367 Laura Morales-Gámez, Alfonso Rodríguez-Galán, Lourdes Franco, and Jordi Puiggalí Introduction: Biodegradable Polymers and Their Nanocomposites 367 Aliphatic Polyester Clay Nanocomposites by In-situ Polymerization 368 Poly(ε-Caprolactone)-Based Nanocomposites 368 Polylactide-Based Nanocomposites 375 PBS-Based Nanocomposites 380 PPDO-Based Nanocomposites 381 PEAs Clay Nanocomposites by In-situ Polymerization 382 Conclusion 384 Acknowledgments 384 References 384

14.1 14.2 14.2.1 14.2.2 14.2.3 14.2.4 14.3 14.4

Index

387

XI

XIII

Preface In-situ intercalation method was reported by Toyota researchers for the synthesis of polyamide nanocomposites in the early nineties which led to the exponential growth in the nanocomposites research. For generation of polymer nanocomposites by this method, the layered silicate mineral was swollen in liquid monomer. After swelling, the polymerization of the monomer was initiated in the presence of ﬁller. As monomer is present in and out of the ﬁller interlayers, the generated structure is exfoliated or signiﬁcantly intercalated. Since then, a large number of polymers like polyethylene, polypropylene (both cases have monomers as gases), PET, epoxy, polyurethane, and polystyrene have been synthesized in situ in the presence of ﬁller to generate nanocomposites. In-situ intercalative polymerization in the presence of ﬁller does provide distinct advantages as compared with other nanocomposite synthesis techniques like possibility to polymerize a large range of thermoplastic and thermosetting polymers, handling of gaseous or liquid monomers, handling of high-pressure polymerization, and easy control of heat of polymerization owing to dispersion medium present in the system. The current book also aims to highlight these advantages of in-situ polymerization in the light of generation of polymer nanocomposites with a large spectrum of polymer matrices. Chapter 1 provides an overview on the various synthesis technologies of polymer nanocomposites and presents a wide spectrum of polymer nanocomposites prepared with in-situ intercalative polymerization. Chapter 2 reviews the polyamide nanocomposites synthesis by in-situ polymerization. PA6 and PA6,6 matrices and various in-situ synthesis methods to generate nanocomposites (like solution-melt polymerization technique, anhydrous-melt polymerization technique, direct solidstate polymerization technique, interfacial polycondesation technique) are discussed. Chapter 3 focuses on polyoleﬁn nanocomposites generation. The effect of ﬁller on the polymerization behavior and the polymer molecular structure is also analyzed. Chapter 4 describes the technique of gas phase-assisted polymerization for the generation of polymer nanocomposites. Composites with various ﬁllers like clay, cellulose ﬁbers, and carbon nanotubes are described. PET nanocomposites synthesized with in-situ polymerization are described in Chapter 5. Bio-based polymer nanocomposites by in-situ polymerization are discussed in Chapter 6 and control of ﬁller-phase dispersion in the polymer matrix is reviewed. Chapter 7

XIV

Preface

describes synthesis and properties of thermoset polyurethane nanocomposites by in-situ polymerization, whereas another thermoset matrix of epoxy is discussed in Chapter 8 for the synthesis of polymer nanocomposites. Chapter 9 describes nanocomposites with unsaturated polyesters along with their properties. Chapter 10 reviews the use of controlled polymerization technique of atom transfer radical polymerization for the synthesis of polymer nanocomposites with in-situ polymerization. Polybutadiene-based polymer nanocomposites are discussed in Chapter 11 and synthesis and properties of P3HT nanocomposites with carbon nanotubes are detailed in Chapter 12. Chapter 13 describes one of the most common polymer polystyrene to synthesis nanocomposites with in-situ polymerization methods. Chapter 14 discusses clay-ﬁlled nanocomposites of aliphatic polyester and poly(esteramide) polymers. It gives me immense pleasure to thank Wiley-VCH publishers for their kind support throughout the whole process. I dedicate this book to my mother and to my wife Preeti for being my constant sources of support and inspiration. Vikas Mittal

XV

List of Contributors Yoshito Andou Kyushu Institute of Technology Eco-Town Collaborative R&D Center for the Environment and Recycling 2-4 Hibikino Wakamatsu-ku, Kitakyushu-shi Fukuoka 808-0196 Japan

Feng Chen Sichuan University College of Polymer Science and Engineering State Key Laboratory of Polymer Materials Engineering Chengdu 610065 People’s Republic of China

Stuart A. Bateman CSIRO Materials Science and Engineering Graham Road Highett VIC 3190 Australia

Hua Deng Sichuan University College of Polymer Science and Engineering State Key Laboratory of Polymer Materials Engineering Chengdu 610065 People’s Republic of China

Anastasia C. Boussia National Technical University of Athens School of Chemical Engineering Laboratory of Polymer Technology Zographou Campus 15780 Athens Greece Dongyu Cai Loughborough University Department of Materials Loughborough, LE11 3TU UK

Takeshi Endo Kinki University Molecular Engineering Institute 11-6 Kayanomori Iizuka, Fukuoka 820-8555 Japan Lourdes Franco Universitat Politècnica de Catalunya Departament d’Enginyeria Química Av. Diagonal 647 08028 Barcelona Spain

XVI

List of Contributors

Qiang Fu Sichuan University College of Polymer Science and Engineering State Key Laboratory of Polymer Materials Engineering Chengdu 610065 People’s Republic of China Michal Kedzierski Industrial Chemistry Research Institute Department of Polyesters, Epoxide Resins and Polyurethanes Rydygiera Street 8 01 793 Warsaw Poland Giuseppe Leone CNR-ISMAC Istituto per lo studio delle Macromolecole via E. Bassini 5 20133 Milano Italy Zhongrui Li University of Arkansas Nanotechnology Center Little Rock, AR 72204 USA Abolfazl Maneshi University of Waterloo Department of Chemical Engineering Waterloo, ON N2L 3G1 Canada Vikas Mittal The Petroleum Institute Chemical Engineering Department Abu Dhabi UAE

Laura Morales-Gámez Universitat Politècnica de Catalunya Departament d’Enginyeria Química Av. Diagonal 647 08028 Barcelona Spain Haruo Nishida Kyushu Institute of Technology Eco-Town Collaborative R&D Center for the Environment and Recycling 2-4 Hibikino Wakamatsu-ku Kitakyushu-shi, Fukuoka 808-0196 Japan Grady A. Nunnery Georgia Institute of Technology School of Materials Science and Engineering Atlanta, GA 30332 USA Constantine D. Papaspyrides National Technical University of Athens School of Chemical Engineering, Laboratory of Polymer Technology, Zographou Campus 15780 Athens Greece Lawrence A. Pranger Georgia Institute of Technology School of Materials Science and Engineering Atlanta, GA 30332 USA Jordi Puiggalí Universitat Politècnica de Catalunya Departament d’Enginyeria Química Av. Diagonal 647 08028 Barcelona Spain

List of Contributors

Greg G. Qiao University of Melbourne Department of Chemical and Bimolecular Engineering Polymer Science Group Parkville VIC 3010 Australia Giovanni Ricci CNR-ISMAC Istituto per lo studio delle Macromolecole via E. Bassini 5 20133 Milano Italy Alfonso Rodríguez-Galán Universitat Politècnica de Catalunya Departament d’Enginyeria Química Av. Diagonal 647 08028 Barcelona Spain Leonardo Simon University of Waterloo Department of Chemical Engineering Waterloo, ON N2L 3G1 Canada Ranya Simons University of Melbourne Department of Chemical and Bimolecular Engineering Polymer Science Group Parkville VIC 3010 Australia and CSIRO Materials Science and Engineering Graham Road Highett VIC 3190 Australia

João Soares University of Waterloo Department of Chemical Engineering Waterloo, ON N2L 3G1 Canada Mo Song Loughborough University Department of Materials Loughborough, LE11 3TU UK Rina Tannenbaum Georgia Institute of Technology School of Materials Science and Engineering Atlanta, GA 30332 USA Stamatina N. Vouyiouka National Technical University of Athens School of Chemical Engineering Laboratory of Polymer Technology Zographou Campus 15780 Athens Greece Ke Wang Sichuan University College of Polymer Science and Engineering State Key Laboratory of Polymer Materials Engineering Chengdu 610065 People’s Republic of China Qin Zhang Sichuan University College of Polymer Science and Engineering State Key Laboratory of Polymer Materials Engineering Chengdu 610065 People’s Republic of China

XVII

XVIII

List of Contributors

Hanying Zhao Nankai University Department of Chemistry Tianjin 300071 China

Liqiu Zheng University of Arkansas Nanotechnology Center Little Rock, AR 72204 USA

1

1 In-situ Synthesis of Polymer Nanocomposites Vikas Mittal

1.1 Introduction

It was the pioneering work of Toyota researchers toward the development of polymeric nanocomposites in the early 90s [1, 2], in which electrostatically held 1-nmthick layers of the layered aluminosilicates were dispersed in the polyamide matrix on a nanometer level, which led to an exponential growth in the research in these layered silicate nanocomposites. These nanocomposites were based on the in-situ synthesis approach in which monomer or monomer solution was used to swell the ﬁller interlayers followed by polymerization. Subsequently, Giannelis and coworkers [3, 4] also reported the route of melt intercalation for the synthesis of polymer nanocomposites. Montmorillonite has been the most commonly used layered aluminosilicate in most of the studies on polymer nanocomposites. The general formula of montmorillonites is Mx(Al4−xMgx)Si8O20(OH)4 [5, 6]. Its particles consist of stacks of 1-nmthick aluminosilicate layers (or platelets) with a regular gap in between (interlayer). Each layer consists of a central Al-octahedral sheet fused to two tetrahedral silicon sheets. In the tetrahedral sheets, silicon is surrounded by four oxygen atoms, whereas in the octahedral sheets, aluminum atom is surrounded by eight oxygen atoms. Isomorphic substitutions of aluminum by magnesium in the octahedral sheet generate negative charges, which are compensated for by alkaline-earth- or hydrated alkali-metal cations. Owing to the low charge density (0.25–0.5 equiv. mol−1) of montmorillonites, a larger area per cation is available on the surface that leads to a lower interlayer spacing in the modiﬁed montmorillonite after surface ion exchange with alkyl ammonium ions. On the contrary, the minerals with high charge density (1 equiv. mol−1) like mica have much smaller area per cation and can lead to much higher basal plane spacing after surface modiﬁcation; however, owing to very strong electrostatic forces present in the interlayers due to the increased number of ions, these minerals do not swell in water and thus do not allow the cation exchange. In contrast, aluminosilicates with medium charge densities of 0.5–0.8 equiv. mol−1 like vermiculite offer a potential of partial swelling in water and cation exchange that can lead to much higher basal plane spacing in In-situ Synthesis of Polymer Nanocomposites, First Edition. Edited by Vikas Mittal. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA.

2

1 In-situ Synthesis of Polymer Nanocomposites

the modiﬁed mineral if optimum ion exchange is achieved. Vaia et al. [7] also proposed further insight into the positioning of the surface modiﬁcation molecules on the surface of the ﬁller based on FTIR experiments. By monitoring frequency shifts of the asymmetric CH2 stretching and bending vibrations, they found that the intercalated chains exist in states with varying degrees of order. In general, as the interlayer packing density or the chain length decreases (or the temperature increases), the intercalated chains adopt a more disordered, liquidlike structure resulting from an increase in the gauche/trans conformer ratio. Nanocomposites with a large number of polymer matrices have been synthesized and signiﬁcant enhancements in the composite properties have been reported. The improvement in the mechanical properties of the nanocomposites is generally reported, though a synergistic enhancement in the other composite properties like gas barrier resistance is also generally achieved. Figure 1.1a demonstrates the decrease in oxygen permeation through the polyurethane, epoxy, and polypropylene nanocomposites as a function of inorganic ﬁller volume fraction [8–10]. Figure 1.1b also shows the improvement in mechanical properties of the polypropylene and polyethylene nanocomposites as a function of ﬁller volume fraction [11–13]. The polypropylene composites have been generated by using two different ﬁller surface modiﬁcations containing ammonium and imidazolium ions. The microstructure of the nanocomposites is also ideally classiﬁed as unintercalated (phase separated), intercalated, and exfoliated composites. The composite microstructure is classiﬁed as exfoliated when the ﬁller platelets are completely delaminated into their primary nanometer scale size and the platelets are far apart from each other so that the periodicity of this platelet arrangement is totally lost. When a single or sometimes more than one extended polymer chain is intercalated into the clay interlayers, but the periodicity of the clay platelets is still intact, such a microstructure is termed as intercalated. On the basis of the interfacial interactions and mode of mixing of the organic and inorganic phases, it is possible that both the phases do not intermix at all and a microcomposite or unintercalated composite is formed. Transmission electron microscopy (TEM) and X-ray diffraction (XRD) are the most commonly used methods to characterize the microstructure of the nanocomposites. Figure 1.2 shows the TEM micrographs depicting the various idealized morphologies of the polymer nanocomposite structures [15]. However, it should be noticed that these classiﬁcations of the composite microstructure as exfoliated and intercalated are not very realistic as generally in reality a mixture of different morphologies is present. Figure 1.3 also shows the three idealized morphologies of immiscible, intercalated, and exfoliated composites [16]. The presence or absence of diffraction peaks in the XRD of the composites is used to assess information about the microstructure of the composites. The intensity of the X-ray diffractograms is generally taken as a measure to classify the microstructure as intercalated or exfoliated. However, it should be noticed that the X-ray signal are very qualitative in nature and are strongly inﬂuenced by the sample preparation, orientation of the platelets, as well as defects present in the crystal structure of the montmorillonites. Therefore, the classiﬁcation of the nanocomposite microstructure just based on the intensity can be faulty. Also, the presence

1.1 Introduction a)

b)

Figure 1.1 (a) Relative oxygen permeation and (b) relative tensile modulus of various polymer

nanocomposites as a function of ﬁller volume fraction [8–13]. Reproduced from Ref. [14].

of diffraction signal in the diffractograms of the composite does not mean that 100% of the microstructure is intercalated and it is quite possible to have signiﬁcant amount of exfoliation present in the composite. Similarly, absence of diffraction signal also does not guarantee the complete exfoliation as small or randomly oriented intercalated platelets may still be present in the composite. Many factors inﬂuence the microstructure and hence the properties of the nanocomposites. The ﬁrst of such factors is the surface modiﬁcation of the ﬁller and its interaction with the polymer. The modiﬁcation is required to make the ﬁller organophilic and to push the ﬁller interlayers apart, thus providing possibilities for polymer intercalation. Table 1.1 details the various kinds of surface

3

4

1 In-situ Synthesis of Polymer Nanocomposites

a)

b)

25 nm

c)

25 nm

25 nm

Figure 1.2 TEM micrographs indicating various possible morphologies in the composites as

a function of the ﬁller distribution: (a) exfoliated, (b) intercalated, and (c) unintercalated. Reproduced from Ref. [15] with permission from Wiley.

Figure 1.3 XRD patterns of immiscible, intercalated, and exfoliated composites. Reproduced

from Ref. [16] with permission from Elsevier.

modiﬁcations commonly used to modify the ﬁller surface. The basal plane spacing of the ﬁller resulting after the surface modiﬁcation is also provided. The modiﬁcations differ in many aspects such as chain length, density of the chains in the surface modiﬁcation molecule, and chemical architecture. The modiﬁcation molecules have speciﬁc interactions with the matrix polymer and these interactions are responsible for the ability of the ﬁller to exfoliate or delaminate in the polymer matrix [8, 9]. Figure 1.4 shows an example of such interplay of interactions between

1.1 Introduction Table 1.1 Basal plane spacing values of various surface-modiﬁed montmorillonites and vermiculites [8–13].

Modiﬁcation

Substrate, CEC (μ.eq g−1)

Basal spacing (nm)

Octadecyltrimethylammonium

Montmorillonite, 880

1.84

Octadecyltrimethylammonium

Montmorillonite, 680

1.82

Octadecyltrimethylammonium

Montmorillonite, 900

1.85

Octadecyltrimethylammonium

Montmorillonite, 1000

2.14

Dioctadecyldimethylammonium

Montmorillonite, 880

2.51

Dioctadecyldimethylammonium

Montmorillonite, 680

2.45

Trioctadecylmethylammonium

Montmorillonite, 880

3.42

Trioctadecylmethylammonium

Montmorillonite, 680

3.29

Benzylhexadecyldimethylammonium

Montmorillonite, 880

1.88

Benzylhexadecyldimethylammonium

Montmorillonite, 680

1.85

Docosyltriethylammonium

Montmorillonite, 880

1.93

Decylmethyloctadecylimidazolium

Montmorillonite, 880

2.24

Didocyldimethylammonium/ dioctadecyldimethylammonium

Montmorillonite, 880

2.28

Didocyldimethylammonium/ dioctadecyldimethylammonium

Montmorillonite, 680

2.27

Benzylhydroxyethylmethyloctadecyl ammonium

Montmorillonite, 880

2.06

Benzldibutylhydroxyethylammonium

Montmorillonite, 880

1.52

Benzyldi(hydroxyethyl)butyl ammonium

Montmorillonite, 880

1.50

Benzyltriethanolammonium

Montmorillonite, 880

1.52

Benzylhydroxyethylmethyloctadecyl ammonium

Vermiculite, 1400

3.40

Benzylhexadecyldimethylammonium

Vermiculite, 1400

3.25

Reproduced from Ref. [14].

the polymer and surface modiﬁcation [8]. The oxygen permeation through polyurethane nanocomposites has been described in Figure 1.4a. Interestingly, the permeation decreasing with increasing ﬁller fraction for two of the surface modiﬁed ﬁllers, whereas it increased with the ﬁller fraction for one of the modiﬁed ﬁllers. The ﬁlers for which the permeation decreased in the composites had polar modiﬁcations with better match of polarity with the polyurethane matrix. Even the

5

1 In-situ Synthesis of Polymer Nanocomposites a) 1.6 O2 relative trans. rate (TO2/T0O2)

Nanofil 15

1.4 1.2 1.0 0.8

Nanofil 804

0.6 0.4

Nanofil 32

0

0.01

0.02 0.03 0.04 0.05 inorganic volume fraction

0.06

b) water vapor relative trans. rate (TH2O/T0H2O)

6

1.0

0.8

0.6 Nanofil 804 Nanofil 32

0.4

0.2

Nanofil 15

0

0.01

0.02 0.03 0.04 0.05 0.06 inorganic volume fraction

0.07

Figure 1.4 (a) Oxygen and (b) water vapor permeation through polyurethane composites.

Reproduced from Ref. [8] with permission from American Chemical Society.

reactive groups present in one of the modiﬁcations were expected to chemically react with the polyurethane matrix thus leading to chemical tethering of the polymer chains on the ﬁller surface. On the other hand, the ﬁller for which the permeation increased through the composite had a nonpolar surface modiﬁcation. This nonpolar modiﬁcation had polarity mismatch with the polyurethane polymer matrix leading to the increase of interfacial voids or free volume that helped to

1.1 Introduction

oxygen relative permeability (Pc/Pp)

increase the permeation as the extent of ﬁller increased in the composite. Another interesting observation was the behavior of water vapor transmission through the same polyurethane nanocomposites as shown in Figure 1.4b [8]. Here the permeation decreased for all the composites irrespective of the ﬁller modiﬁcation. This was a result of different interactions the permeant molecules have with the polymer matrix. Oxygen is noninteracting with the polymer matrices, whereas water molecules easily form clusters and hydrogen bonds with the polymer chains. Thus, apart from the interactions between the polymer and surface medications, speciﬁc interactions of permeant molecules with the polymer also affect the performance of nanocomposites. Second factor affecting the properties and performance of nanocomposites is the ﬁller volume fraction. The effect of ﬁller volume fraction on the permeation properties of polyurethane nanocomposites has already been shown in Figure 1.4 in conjunction with the interactions between the polymer and the surface modiﬁcation. Similarly, effect of ﬁller volume fraction in conjunction with ﬁller modiﬁcation–polymer interactions is shown in Figure 1.5 for epoxy nanocomposites [9]. Two different surface modiﬁcations benzyldimethylhexadecylammonium (BzC16) and benzyldibutyl(2-hydroxyethyl) ammonium (Bz1OH) were used. Owing to the increase in the ﬁller volume fraction, the oxygen permeation through the composites decreased. For BzC16, saturation in the permeation reduction was achieved at 4 vol% inorganic ﬁller fraction, whereas fro Bz1OH ﬁller, the permeation was observed to decrease further even at 5 vol% ﬁller fraction. The performance of two ﬁllers was signiﬁcantly different from each other owing to better compatibility of the Bz1OH-modiﬁed ﬁller with the epoxy matrix, whereas BzC16 modiﬁcation had a polarity mismatch with the epoxy polymer. The permeation through the epoxy nanocomposites was also compared with the different aspect ratio ﬁllers in Figure 1.5. In the case of BzC16 ﬁlled composites, a ﬁller aspect ratio near to 50 was observed, whereas for the Bz1OH composites, much higher

1.0

0.8

BzC16 a = 50

0.6 a = 150

a = 100

0.4 Bz1OH 0.2

0.00

0.02 0.04 inorganic volume fraction (f )

0.06

Figure 1.5 Oxygen permeation through epoxy nanocomposites as a function of ﬁller volume

fraction in the composites. Reproduced from Ref. [9] with permission from American Chemical Society.

7

8

1 In-situ Synthesis of Polymer Nanocomposites

aspect ratio of the ﬁller was observe indicating that aspect ratio of the ﬁller plays a signiﬁcant role in deﬁning the performance of the nanocomposites. Apart from aspect ratio, the ﬁller alignment or orientation is also important for certain properties of nanocomposites such as gas barrier properties. The aligned ﬁller would provide a better resistance to the ﬂow of the permeant molecules through the polymer matrix owing to the generation of higher extent of tortuity in the mean free path of the permeant molecules. However, in general, in most of the cases, the ﬁller is completely misaligned in the generated composites. Apart from misalignment, the ﬁller platelets are also observed to be bent and folded. Figure 1.6 presents one such example of polyurethane nanocomposites [8]. The ﬁller platelets are observed to be both intercalated and exfoliated. The orientation or alignment is totally absent and the bending and folding of the platelets is clearly observed. Figure 1.7 also shows the impact of the misaligned ﬁller on the barrier properties in comparison with the aligned ﬁller. Figure 1.7a is the theoretical model depicting the actual misaligned ﬁler in the composites and Figure 1.7b shows the predictions based on the model [17]. At an aspect ratio of 150 or higher, the misaligned ﬁller was observed to be only one-third effective in generating permeation resistance as compared to the aligned ﬁller. Synthesis methodology is another factor affecting the microstructure and properties of polymer nanocomposites. There are different methods available to manufacture polymer nanocomposites and have their own advantages and limitations. These are detailed in Section 1.2.

20 nm Figure 1.6 TEM micrograph of a PU nanocomposite indicating complete misalignment of the

ﬁller platelets. Reproduced from reference [8] with permission from American Chemical Society.

1.2 Synthesis Methods a)

relative transmission rate (Tc/Tp)

b) 1.0 0.8 0.6

misaligned

0.4 0.2

aligned

0.0 0

50

100 150 aspect ration (a)

200

Figure 1.7 (a) Theoretical model showing the complete misalignment of ﬁller platelets and

(b) inﬂuence of ﬁller misalignment on permeation properties. Reproduced from Ref. [17] with permission from Wiley.

1.2 Synthesis Methods

There are four general methods for the nanocomposite synthesis: 1) 2) 3) 4)

template synthesis intercalation of polymer from solution melt intercalation in-situ synthesis.

9

10

1 In-situ Synthesis of Polymer Nanocomposites

Template synthesis involves the synthesis of inorganic material in the presence of polymer matrix. Double-layer hydroxide-based nanocomposites have been synthesized by using this route [18, 19]. The polymer aids the nucleation and growth of the inorganic host crystals and gets trapped within the layers as they grow. Template synthesis technique is not widely used even though it presents potential to generate exfoliated nanocomposites. Drawbacks like use of high temperature and tendency of the generated ﬁller to aggregate are also to be considered. In intercalation of polymer from solution mode of nanocomposite synthesis, the organically modiﬁed silicate is dispersed in a solvent in which the polymer is also soluble. The polymer then adsorbs onto the delaminated sheets followed by the evaporation of the solvent. When the solvent is evaporated, the sheets reassemble, which also trap the polymer chains in between. Thus, an ordered multilayer structure is usually formed using this approach. The polymer chains loose entropy in the process of intercalation, which is compensated by the increase in the entropy of the solvent molecules due to their desorption from the ﬁller interlayers. The technique is mostly used for the intercalation of the water-soluble polymers like poly(vinyl alcohol), poly(ethylene oxide), poly(acrylic acid), poly(vinlypyrrolidone), etc. [20–24]. Later on, the use of this technique was also undertaken in organic solvents for polymers nonsoluble in water [25, 26]. Melt intercalation is one of the most commonly used techniques for the synthesis of polymer nanocomposites [27–30]. The high molecular weight polymer is melted at high temperature and the ﬁller is then blended with the polymer melt at high temperature under shear. Thus, this mode of composite generation does not require any chemical synthesis or solvent. But the intercalation of high molecular weight polymer chains in the ﬁller interlayers is still a challenge as both thermodynamic and kinetic factors inﬂuence the intercalation. Therefore, it is essential to modify the ﬁller in a way that it can be exfoliated in the polymer matrix by the action of shear. In contrast, low molecular weight compatibilizers can also be added to compatibilize both the organic and inorganic components so as to enhance polymer intercalation. Figure 1.8 shows one such example of melt intercalation method, in which modiﬁed ﬁller was ﬁrst mixed with low molecular weight compatibilizer, followed by the compounding of this hybrid with high molecular weight matrix polymer [27]. As melt intercalation method uses high temperature for mixing of the organic and inorganic components, therefore, thermal degradation of the ﬁller modiﬁcation and polymer poses a concern. As even a small extent of degradation can alter the ﬁller matrix interactions and hence affecting the microstructure of the composites, therefore, thermal degradation needs to be avoided. In-situ intercalation method was reported by Toyota researchers for the synthesis of polyamide nanocomposites that led to the exponential growth in the nanocomposites research. For generation of polymer nanocomposites by this method, the layered silicate mineral is swollen in monomer. After swelling, the polymerization of the monomer is initiated. As monomer is present in and out of the ﬁller interlayers, therefore, the generated structure is exfoliated or signiﬁcantly intercalated. As the rate or mechanism of polymerization in and out of the ﬁller interlayers

1.2 Synthesis Methods

Stearyl ammonium

Silicate Layer of Clay

PP-MA Oligomer Maleic Anhydride Group

PP

Figure 1.8 Schematic of polymer intercalation in the silicates using melt mixing approach.

Reproduced from Ref. [27] with permission from American Chemical Society.

11

12

1 In-situ Synthesis of Polymer Nanocomposites

layered clay n⭌11

ε-caprolactam (melt)

100 °C

25 °C n⬉8 ω-amino acid Figure 1.9 Intercalation of the modiﬁed montmorillonite with caprolactam. Reproduced from

Ref. [31] with permission from Elsevier.

may be different, therefore, it is important to control there intragallery and extragallery polymerization reactions for uniform polymerization. Toyota researchers reported the modiﬁcation of the montmorillonite with amino acids of different chain lengths which were subsequently swollen by caprolactam. The schematic of such a process has been demonstrated in Figure 1.9 [31]. Since then, a large number of polymers have been synthesized in situ in the presence of ﬁller. Section 1.3 provides a brief overview of these systems.

1.3 In-situ Synthesis of Polymer Nanocomposites

In the in-situ synthesis approach shown in Figure 1.9, caprolactam was used to swell the ﬁller, which was modiﬁed with amino acids of different chain lengths. The chain length of the modiﬁcation had a signiﬁcant effect on the extent of ﬁller swelling as higher chain lengths led to higher extent of monomer intercalation of the ﬁller. Figure 1.10 also shows the X-ray diffractograms of the ﬁller modiﬁed with modiﬁcations of different chain lengths [31]. In a typical synthesis step, ﬁller modiﬁed with 12-aminolauric acid (ALA) was swollen with caprolactam. The slurry was then heated at 250–270 °C for 48 h to polymerize caprolactam (ring-opening polymerization). The morphology of the generated composites was characterized using XRD and TEM, and it was observed that exfoliated nanocomposites were obtained when the ﬁller content in the composite was less than 15 wt%. On the contrary, when the ﬁller content was increased from 15 wt%, intercalated nanocomposites were obtained.

1.3 In-situ Synthesis of Polymer Nanocomposites

18

12

Intensity

11 8 6 5 4 3 n=2 1.0

5.0 2θ (Co-Kα)

10.0

Figure 1.10 X-ray diffractograms the surface modiﬁed ﬁller with modiﬁcations of different

chain lengths; n corresponds to the number of carbon atoms in the modiﬁcation. Reproduced from Ref. [31] with permission from Elsevier.

Nylon 12 nanocomposites were also reported following the in-situ synthesis approach [32]. ALA was used as both the layered silicate modiﬁer and the monomer. Cation exchange at the ﬁller surface by protonated ALA at low ALA concentration was observed. Further swelling of the ﬁller with zwitterionic ALA was observed when the ALA concentration was high as shown in Figure 1.11. The swelling was observed to be independent of the temperature used for swelling, the concentration of the ﬁller, and the type of acid used to protonate ALA. The composites morphology was partially exfoliated and intercalated. A number of other thermoplastic and thermosetting polymer systems have been employed to generate the polymer nanocomposites using in-situ synthesis approach. In one such example of polyurethane nanocomposites, the modiﬁed ﬁllers were ﬁrst swollen be solvent solution of prepolymer [8]. To this mixture was then added a crosslinker followed by polymerization and evaporation of solvent. The generated nanocomposites had exfoliated and intercalated morphology depending on the chemical architecture of the surface modiﬁcations. The modiﬁcation that had polymer groups such as OH groups had much better compatibility with the polymer matrix which led to the extensive exfoliation of the ﬁller in the

13

1 In-situ Synthesis of Polymer Nanocomposites 2.2 Interlayer distance (nm)

14

2.0 1.8 1.6 1.4 1.2 Cation exchange

1.0

swelling with an excess of ALA

0.8 0

10

20 30 40 50 ALA concentration (mmol/L)

60

Figure 1.11 Interlayer distance in the modiﬁed ﬁller as a function of the ALA concentration.

Reproduced from Ref. [32] with permission from Wiley.

20 nm

Figure 1.12 TEM micrograph of polyurethane nanocomposite containing ﬁller modiﬁed with

bis(2-hydroxyethyl) hydrogenated tallow ammonium. Reproduced from Ref. [8] with permission from American Chemical Society.

composite. Figure 1.12 shows the TEM micrograph of such nanocomposite containing ﬁller modiﬁed with bis(2-hydroxyethyl) hydrogenated tallow ammonium. The potential reaction of OH groups in the surface modiﬁcation with the polymer thus leading to the chemical tethering of the polymer chains on the ﬁller surface helps further to exfoliate the ﬁller. On the other hand, for other modiﬁcations that

1.3 In-situ Synthesis of Polymer Nanocomposites

oxygen relative permeability

2.0

1.5

1.0

0.5

0.0

1.5

2.0

2.5

3.0

3.5

d-spacing in composite Figure 1.13 Correlation between the relative oxygen permeability through the composites with

ﬁller basal plane spacing in the composite. Reproduced from Ref. [9] with permission from American Chemical Society.

did not have any polar groups such as dimethyl dihydrogenated tallow ammonium led to the generation of only intercalated nanocomposites. In another example of epoxy nanocomposites [9], a number of surface modiﬁcations differing in chemical architecture were used to modify the ﬁller. Similar to the polyurethane nanocomposites, the modiﬁed ﬁllers with better polarity match with the epoxy polymer were more exfoliated than intercalated, whereas for all other modiﬁed ﬁllers, only intercalated morphology of the composite was obtained. The inﬂuence of intercalated and exfoliated ﬁller on the composite properties was also demonstrated as shown in Figure 1.13. The oxygen permeation properties were plotted as a function of ﬁller basal plane spacing in the composite. The oxygen permeation through the composites was observed to roughly increase on increasing the basal plane spacing of the ﬁller, which is contrary to the expectation. Therefore, it signiﬁed no correlation between the oxygen permeation properties and the ﬁller intercalation. In other words, it is not the intercalated ﬁller, but the exfoliated ﬁller, which enhances the composite properties. Thus, only intercalation of polymer in the ﬁller interlayer is not enough to enhance the composite properties signiﬁcantly, the ﬁller needs to be signiﬁcantly exfoliated. In the case of exfoliated nanocomposites, impressive reduction in oxygen permeation was observed. Messersmith and Giannelis [33] reported the synthesis of epoxy nanocomposites by using in-situ synthesis approach. They used different curing agents and curing conditions for the generation of nanocomposites. The authors studied the swelling of the modiﬁed ﬁller with the epoxy prepolymer. Montmorillonite modiﬁed with bis(2-hydroxyethyl)methyl halogenated tallow ammonium was observed to be readily dispersible in diglycidyl ether of bisphenol A (DGEBA). As shown in Figure 1.14, the layer spacing increased when the ﬁller was added with DGEBA. The increased basal spacing indicated the intercalation of the prepolymer chains in the ﬁller interlayers. The temperature was also observed to have an impact on

15

1 In-situ Synthesis of Polymer Nanocomposites d-spacing (Å) 20

80 60 40

Counts (a. u.)

16

OMTS/DGEBA (90 °C)

OMTS/DGEBA (RT)

OMTS Powder

2

4

6 2Θ

8

10

Figure 1.14 X-ray diffraction patterns of the modiﬁed ﬁller and the mixture of epoxy prepoly-

mer with ﬁller at room temperature and at 90 °C. Reproduced from Ref. [33] with permission from American Chemical Society.

the extent of intercalation as it was also observed to increase when the swelling was performed at a temperature of 90 °C. Although the DGEBA prepolymer molecules intercalated the ﬁller interlayers indicating a better polarity match between them and the ﬁller surface, the ﬁller platelets were still not completely delaminated as conﬁrmed by the diffraction peak at lower angles. This was also observed in another study [9] in which the surface modiﬁcations of different chemical architectures were used and most of the modiﬁcations did not lead to the complete exfoliation of the ﬁller platelets when the ﬁller was swollen with epoxy prepolymer solution. Figure 1.15 also shows the XRD patterns indicating the generation of exfoliated morphology in the silicate/DGEBA/benzyldimethlyamine system as a function of temperature [33]. The scan temperature increased vertically from bottom to top of the ﬁgure. At lower temperature, complete exfoliation of the ﬁller was not achieved and a mix of both of intercalated and unintercalated silicate platelets was present. When the temperature was increased, disappearance of the diffraction peaks occurred indicating the ﬁller delamination during heating. The authors also observed signiﬁcant inﬂuence of the nature of the curing agent on the morphology of the composites. Only intercalated nanocomposites were

1.3 In-situ Synthesis of Polymer Nanocomposites d-spacing (Å) 20

10

Incr. Temp.

Counts (a. u.)

50 40 30

2

4

6 2Θ

8

10

Figure 1.15 X-ray diffraction patterns of ﬁller/DGEBA/benzyldimethlyamine system at

different temperatures from 20 to 150 °C. Reproduced from Ref. [33] with permission from American Chemical Society.

obtained when using diamines as curing agents. On the contrary, when benzyldimethlyamine was used as curing agent, exfoliated nanocomposites were obtained. The bridging of silicate layers (and hence degellation) by the amine curing agent was suggested as a reason for the generation of intercalated morphology in the composites when using diamines as curing agents. Another factor contributing to such degellation is the presence of excess modiﬁcation molecules on the surface of ﬁller. These molecules can also react with the curing agent or the epoxy polymer thus disturbing the interface between the ﬁller and polymer. Generation of PET nanocomposites by in-situ synthesis of PET (direct condensation reactions of diol and diacid) in the presence of clay was not successful [34] as only low molecular weight polymer was observed due to poor control on stoichiometry. Melt intercalation method was also employed for the generation of PET nanocomposites, but only intercalated nanocomposites were observed by this method. This was owing to the kinetic hindrance to the high molecular weight polymer chains to enter the ﬁller interlayers. As an alternative, ring-opening polymerization of ethylene terepthalate cyclic oligomers in the presence of organically modiﬁed montmorillonites was used. The schematic of the process leading to generation of PET nanocomposites is depicted in Figure 1.16. The ﬁller interlayers

17

18

1 In-situ Synthesis of Polymer Nanocomposites 1 alky1–

1

1 N+ 1

1 alky1– 1 N+

O

O

O (CH2)2 O

O

O

O

O

O

O

O

O

O

(CH2)2

O (CH2)2

O (CH2)2

O

O

O

O

O

O

O O (CH2)2

O

O

O O

n

O n

1

OR

1

1

O (CH2)2

O

n

layered silicates intercalated by cyclic oligomers

OR

n

n

H

O

(CH2)2

1

1

H

O

O

1

1

H

O

(CH2)2 O

1

1 O (CH2)2 O

H

O

–

O

(CH2)2

O –

ring-opening polymerization of cyclic oligomers causing increase of interlayer distance along with disintegration of layered silicates

OR

OR

1

exfoliated state of layered silicates

Figure 1.16 Ring-opening polymerization of cyclic oligomers to generate PET nanocompos-

ites. Reproduced from Ref. [34] by permission from Elsevier.

1.3 In-situ Synthesis of Polymer Nanocomposites d-spacings (Å) 30

o Cr3+

o

o Cr3+

10

o

o

Cr3+(o

o

) Cr3+ n

Counts (a. u.)

o

20

2

4

6 2 theta

8

10

Figure 1.17 X-ray diffraction patterns of the intercalated ﬁller (solid line) and the nanocom-

posite material (dashed line). Reproduced from Ref. [35] with permission from American Chemical Society.

were swollen with cyclic oligomers. These cyclic oligomers owing to their low molecular weight and hence lower viscosity could easily intercalate in the ﬁller interlayers. Ring-opening polymerization of the oligomers led to further increase in the interlayer distance followed by ﬁller delamination. In-situ synthesis of caprolactone in the interlayers of Cr3+-modiﬁed ﬂuorohectorite was reported by Messersmith et al. [35]. The microstructure development was studied by using XRD as shown in Figure 1.17. The unintecalated ﬁller had a basal plane spacing of 12.8 Å, which was increased to 14.6 Å for the intercalated ﬁller swollen with caprolactone. After the polymerization of caprolactone in and around of ﬁller interlayers, a ﬁnal basal plane spacing of 13.7 Å was observed for the nanocomposite. The observed basal plane spacing correlated well with 4-Å interchain distance in the crystal structure of poly(caprolactone). Polyoleﬁn nanocomposites have also been reported by using gas phase by in-situ polymerization [36, 37]. In this technique, a Zieler–Natta or any other coordination catalyst is anchored to the surface of the layered silicates. The anchoring of the catalyst is not achieved by the usual cation exchange, but it is achieved by the electrostatic interactions of the catalytic materials with MAO initially anchored to the ﬁller surface. The molecular weight of the generated polymer can also be controlled by the addition of chain transfer agent. It was also observed by the authors that in the absence of chain transfer agent, molecular weight of the polymer was too high to proceed further and hydrogen was added to improve

19

20

1 In-situ Synthesis of Polymer Nanocomposites 0.70 μm

Figure 1.18 TEM micrograph of the polyethylene nanocomposites prepared by in-situ

polymerization. Reproduced from Ref. [37] with permission from Elsevier.

the processability [36]. Figure 1.18 shows the TEM micrograph of the polyethylene nanocomposites prepared by the in-situ polymerization approach. Good dispersion of the ﬁller in the polymer matrix was achieved. Other studies on the in-situ synthesis of polyoleﬁn naoconpoites have also been reported [38–40]. Akelah et al. also reported vinylbenzyltrimethylammonium-modiﬁed montmorillonites [41]. The presence of vinyl groups was helpful in copolymerizing them with an external monomer like styrene in order to chemically tether or graft the polymer chains to the ﬁller surface. Fu and Qutubuddin also reported the synthesis of a polymerizable cationic surfactant, vinylbenzyldimethyldodecylammonium chloride with terminal monomer moiety [42]. The surfactant was ion exchanged on the clay surface and the modiﬁed clay was swollen with styrene monomer. Free radical polymerization of styrene with azo bis(iso-butyronitrile) as initiator led to the generation of exfoliated polystyrene–clay nanocomposites. Mittal [43] also reported the partial exchange of ﬁller surface with methacryloxyethyltrimethylammonium chloride and the remaining surface with nonreactive modiﬁcation as shown in Figure 1.19. The free radical polymerization of lauryl methacrylate in the presence of modiﬁed montmorillonite led to the tethering of poly(lauryl methacrylate) chains to the ﬁller surface. The successful grafting was conﬁrmed with increased organic weight loss in thermogravimetric analysis and increased basal plane spacing in XRD. Living or controlled polymerization techniques have also been frequently used for the in-situ synthesis of polymer nanocomposites. Figure 1.20 shows the example of styrene polymerization in the presence of modiﬁed ﬁller. The ﬁller modiﬁcation consisted of ammonium cation bearing a notroxide moiety [44]. Styrene was polymerized in bulk at 125 °C for 8 h. No diffraction peaks were observed in the XRD conﬁrming extensive exfoliation of the ﬁller. TEM micrographs also conﬁrmed the uniform distribution of ﬁller in the matrix. Generation

1.3 In-situ Synthesis of Polymer Nanocomposites a) O

O

O

O

O

N+

O

N+

N+

O O

N+

O

O

O

O

N+

N+

N+

N+

N+

N+

N+

b) CH3

CH3

H2C CO + O C12H25

C H2 O O N+

N+

C12H25 O CH3 CH3 CO C C C C0 nC H2 H CO 2 COH2 COH2 CH3 O O O C12H25 C12H25 CH3

N+

N+

CH3 C C+ n H2 CO CO O O C12H25 C12H25

O O N+

N+

CH3 H2C CO CH3 CH3 CH3 O C C C C0 n C12H25 H2 CO H2 COH2 CO O O O C12H25 C12H25 N+

N+

Figure 1.19 (a) Schematic of surface modiﬁcation and (b) schematic of lauryl methacrylate

polymerization on the ﬁller surface. Reproduced from Ref. [43] with permission from Elsevier.

21

22

1 In-situ Synthesis of Polymer Nanocomposites

1.26 nm

–

– Na+

– Na+ – Na+ –

Na+ –

O O

O

– Cl

NaCl – +

–

2.35 nm

O

O O

– O O

+ –

+

O

O

O

O O

+ –

+ N

O O

O

N

O O

+ –

Figure 1.20 Schematic of the nitroxyl-based organic cation modiﬁcation of montmorillonite

surface and generation of PS nanocomposite. Reproduced from Ref. [44] with permission from American Chemical Society.

1.3 In-situ Synthesis of Polymer Nanocomposites

–

– – + Na

+ Na

+ Na –

+

d = 1.23 nm

Na

(Na-MMT)

+

+

Na

Na

–

– O

+( ) O 11 – Br N

Br

NaBr

– + N( ) 11 O O

Br

O Br

– + N( ) 11 O O

d = 1.96 nm (MMT-A)

Br

– Br

Br O

Br

O ( ) N 11 +

– + N( ) 11 O O

O O ( ) N 11 +

O ( ) N 11 +

–

–

1. Styrene + CuBr + BPMODA ATRP 2. Butyl acrylate PBA block

PS block

Figure 1.21 Nanocomposites based on block copolymer generated by atom transfer radical

polymerization. Reproduced from Ref. [45] by permission from Elsevier.

of block copolymer of poly(styrene-block-butylacrylate) grafted from the clay surface by using ATRP approach was also reported by Zhao et al. [45]. Styrene was polymerized ﬁrst followed by polymerization of butyl acrylate. Figure 1.21 shows the schematic of the process. The ATRP initiator had a terminal ammonium moiety that was used for ion exchange on the ﬁller surface.

23

24

1 In-situ Synthesis of Polymer Nanocomposites

References 1 Yano, K., Usuki, A., Okada, A.,

2

3 4 5

6

7

8

9

10 11 12 13 14

15

16 17

18

Kurauchi, T., and Kamigaito, O. (1993) J. Polym. Sci., Part A: Polym. Chem., 31, 2493. Kojima, Y., Fukumori, K., Usuki, A., Okada, A., and Kurauchi, T. (1993) J. Mater. Sci. Lett., 12, 889. Vaia, R.A., Ishii, H., and Giannelis, E.P. (1993) Chem. Mater., 5, 1694. Mehrotra, V. and Giannelis, E.P. (1990) Mater. Res. Soc. Symp. Proc., 171, 39. Bailey, S.W. (1984) Reviews in Mineralogy, Virginia Polytechnic Institute and State University, Blacksburg, VA, USA. Brindley, G.W. and Brown, G. (1980). Crystal Structures of Clay Minerals and Their X-Ray Identiﬁcation (eds G.W. Brindley and G. Brown), Mineralogical Society, London, UK. Vaia, R.A., Teukolsky, R.K., and Giannelis, E.P. (1994) Chem. Mater., 6, 1017. Osman, M.A., Mittal, V., Mobridelli, M., and Suter, U.W. (2003) Macromolecules, 36, 9851. Osman, M.A., Mittal, V., Mobridelli, M., and Suter, U.W. (2004) Macromolecules, 37, 7250. Osman, M.A., Mittal, V., and Suter, U.W. (2007) Macromol. Chem. Phys., 208, 68. Mittal, V. (2007) J. Thermoplastic Comp. Mater., 20, 575. Osman, M.A., Rupp, J.E.P., and Suter, U.W. (2005) J. Mater. Chem., 15, 1298. Mittal, V. (2009) J. Thermoplastic Comp. Mater., 22, 453. Mittal, V. (2010). Advances in Polymer Nanocomposites Technology (ed. V. Mittal), Nova Science Publishers, New York. Mittal, V. (2009). Optimization of Polymer Nanocomposite Properties (ed. V. Mittal), Wiley-VCH Verlag GmbH, Weinheim. Beyer, G. (2002) Plast. Addit. Compound., 4, 22. Osman, M.A., Mittal, V., and Lusti, H.R. (2004) Macromol. Rapid Commun., 25, 1145. Wilson, O.C., Jr., Olorunyolemi, T., Jaworski, A., Borum, L., Young, D., Siriwat, A., Dickens, E., Oriakhi, E., and Lerner, M. (1999) Appl. Clay Sci., 15, 265.

19 Oriakhi, C.O., Farr, I.V., and

20

21 22 23 24 25

26 27

28 29

30

31 32

33 34 35 36

37

38

Lerner, M.M. (1997) Clays Clay Miner., 45, 194. Ruiz-Hitzky, E., Aranda, P., Casal, B., and Galvan, J.C. (1995) Adv. Mater., 7, 180. Ogata, N., Kawakage, S., and Ogihara, T. (1997) J. Appl. Polym. Sci., 66, 573. Parﬁtt, R.L. and Greenland, D.J. (1970) Clay Miner., 8, 305. Billingham, J., Breen, C., and Yarwood, J. (1997) Vib. Spectrosc., 14, 19. Levy, R. and Francis, C.W. (1975) J. Colloid Interface Sci., 50, 442. Jeon, H.G., Jung, H.-T., Lee, S.W., and Hudson, S.D. (1998) Polym. Bull., 41, 107. Krikorian, V. and Pochan, D. (2003) Chem. Mater., 15, 4317. Kawasumi, M., Hasegawa, N., Kato, M., Usuki, A., and Okada, A. (1997) Macromolecules, 30, 6333. Fornes, T.D., Yoon, P.J., Keskkula, H., and Paul, D.R. (2001) Polymer, 42, 9929. McNally, T., Murphy, W.R., Lew, C.Y., Turner, R.J., and Brennan, G.P. (2003) Polymer, 44, 2761. Davis, C.H., Mathias, L.J., Gilman, J.W., Schiraldi, D.A., Shields, J.R., Trulove, P., Sutto, T.E., and Delong, H.C. (2002) J. Polym. Sci., Part B: Polym. Phys., 40, 2661. Okada, A. and Usuki, A. (1995) Mater. Sci. Eng., C3, 109. Reichert, P., Kressler, J., Thomann, R., Mulhaupt, R., and Stoppelmann, G. (1998) Acta Polym., 49, 116. Messersmith, P.B. and Giannelis, E.P. (1994) Chem. Mater., 6, 1719. Lee, S.-S., Ma, Y.T., Rhee, H.-W., and Kim, J. (2005) Polymer, 46, 2201. Messersmith, P.B. and Giannelis, E.P. (1993) Chem. Mater., 5, 1064. Dubois, P., Alexandre, M., Hindryckx, F., and Jerome, R. (1998) J. Macromol. Sci. Rev. Macromol. Chem. Phys., C38, 511. Alexandre, M., Dubois, P., Sun, T., Garces, J.M., and Jerome, R. (2002) Polymer, 43, 2123. Bergman, J.S., Chen, H., Giannelis, E.P., Thomas, M.G., and Coates, G.W. (1999) J. Chem. Soc. Chem. Commun., 21, 2179.

References 39 Jin, Y.-H., Park, H.-J., Im, S.-S.,

Kwak, S.-Y., and Kwak, S. (2002) Macromol. Rapid Commun., 23, 135. 40 Heinemann, J., Reichert, P., Thomann, R., and Muelhaupt, R. (1999) Macromol. Rapid Commun., 20, 423. 41 Akelah, A. and Moet, A. (1996) J. Mater. Sci., 31, 3589.

42 Fu, X. and Qutubuddin, S. (2001)

Polymer, 42, 807. 43 Mittal, V. (2007) J. Colloid Interface Sci.,

314, 141.

44 Weimer, M.W., Chen, H., Giannelis,

E.P., and Sogah, D.Y. (1999) J. Am. Chem. Soc., 121, 1615. 45 Zhao, H., Farrell, B.P., and Shipp, D.A. (2004) Polymer, 45, 4473.

25

27

2 Polyamide Nanocomposites by In-situ Polymerization Anastasia C. Boussia, Stamatina N. Vouyiouka, and Constantine D. Papaspyrides

2.1 Introduction

Polyamides (PAs) are the polymers containing the amide repeating linkage –CONH– in the polymer backbone. They are tough, semicrystalline polymers, characterized in their majority by moderate production cost and easily manipulated melt processing. The sequence of amide bonds along the polymer chain deﬁnes two polyamide categories (Scheme 2.1): AB and AABB. In the ﬁrst category (AB), the polyamide bonds are formed through the polycondensation of lactam/amino acid, while in the second one (AABB), the polyamide structure is derived by the polycondensation of diamines with dicarboxylic acids. The groups R and R′ may be aliphatic, aromatic, or mixed hydrocarbon radicals, deﬁning the physical and chemical properties of polyamides [1, 2]. In the case of aliphatic repeating unit in a concentration higher than 15%, the polyamides are commonly referred to as nylons®, including semiaromatic structures, where one of the two monomers contains an aromatic ring. On the other hand, in case the concentration of aromatic rings adjacent to an amide group is higher than 85%, the pertinent polyamides are considered aromatic and referred to as aramids. The principal commercial polyamides of each category are presented in Table 2.1. Regarding polyamides nomenclature, the common practice is to call type AB or type AABB polyamides PA x or PA x.x, respectively, where x refers to the number of carbon atoms between the amide nitrogens. For type AABB polyamides, the number of carbon atoms in the diamine is indicated ﬁrst, followed by the number of carbon atoms in the diacid. For example, the polyamide formed from εcaprolactam/ε-aminocaproic acid is named PA 6, while PA 6.12 is formed from hexamethylenediamine and dodecanedioic acid [1]. In terms of properties, polyamides are characterized by advanced mechanical performance (tensile and impact strength, hardness, and toughness), high melting points (PA 6: 220 °C; PA 6.6: 265 °C), high chemical resistance, insulating electrical properties, good fatigue resistance – particularly under load, enhanced abrasion resistance, favorable appearance, low speciﬁc weight (−50% compared to aluminum), and potential for the formation of complex shaped parts by injection In-situ Synthesis of Polymer Nanocomposites, First Edition. Edited by Vikas Mittal. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA.

28

2 Polyamide Nanocomposites by In-situ Polymerization

O

O C

N

R

C

O R

N

C

H

R'

H

AB

N H

AABB

Scheme 2.1 Polyamide categories based on amide groups sequence.

Table 2.1 Principal commercial representatives of linear, semiaromatic, and aromatic

polyamides. Polyamides

Chemical structure

Trade name

Poly(hexamethylene adipamide) (PA 6.6)

[–HN(CH2)6NHCO(CH2)4CO–]x

Zytel® (Du Pont) Ultramid® (BASF)

Poly(caproamide) (PA 6)

[–HN(CH2)5CO–]x

Durethan® (Bayer)

Poly(hexamethylene sebacamide) (PA 6.10)

[–HN(CH2)6NHCO(CH2)8CO–]x

Akulon®, Nylatron® (DSM) Capron® (Honeywell) Novamid® (Mitsubishi engineering plastics) Technyl® (Rhone Poulenc) Amilan® (Toray)

Linear

Semiaromatic Poly(mxyleneadipamide (PA MXD6) Poly(hexamethylene terephthalamide) (PA 6.T)

HN CH2

HN

(CH2)6

CH2 NHCO (CH2)4

NHCO

x

Reny® (Mitsubishi gas) Glamide ® (Toyobo) Ixef® (Solvay)

x

Ultramid ® T (BASF) Glamide ® (Toyobo) Arlen® (Mitsui)

CO

CO

Aromatic Poly(p-phenyl terephthalamide) (PPTA) Poly(m-phenyl isophthalamide) (PMIA)

HN

NHCO

CO x

HN

NHCO

CO x

Twaron® (Akzo) Technora® (Teijin) Kevlar® (Du Pont) Amodel® (Solvay) Tejiconex (Teijin) Apial® (Unitika) Nomex® (Du Pont)

2.2 Manufacturing Processes of Commercially Important Polyamides

45%

Global demand of PAs 2007: 6300 ktons 2012: 6900 ktons

50% 40%

38%

4%

20% 10%

20%

16%

30%

15% 23%

18%

2012

15%

6% 2007

0%

Figure 2.1 Polyamides global demand for 2006 and prediction for 2012 per application

sector [3].

molding [1–3]. This material behavior renders them a particularly important class of engineering thermoplastics; PAs reﬂect almost 4% of the total polymer demand, 90% out of which is referred to PA 6.6 and PA 6 [1, 4], due to their low cost, raw materials and intermediates availability, and satisfactory cost–performance ratio. As a result, PAs cover a broad ﬁeld of applications ranging from ﬁbers to ﬁlms, while the use of shaped PA objects is encountered in several industrial sectors, such as automotive, aerospace, agricultural, packaging, and electrical and electronics industry, as well as in home equipment [1, 3–5]. Furthermore, despite the severe competition that they face by other polymers due to lower cost, such as poly(ethylene terephthalate) for ﬁbers and polypropylene for shaped parts, polyamides market has witnessed a constant increase in demand since the 1980s. Indeed, the annual market growth for the principal commercial polyamides, PA 6.6 and PA 6, is estimated to be 1.8% for the period 2007–2012. Especially, PAs use as thermoplastic resins and ﬁlms undertakes a consistently increasing market share, while the market of polyamide ﬁbers is almost constant (Figure 2.1) [6, 7]. This increase in polyamides demand as engineering thermoplastics is mainly attributed to the automotive industry; the expansion of the pertinent industry in Asian countries, in combination with the generalized trend of incorporating polyamide parts in automation for cost reduction (weight and energy demands) and recyclability reasons, strongly favors their expansion [3].

2.2 Manufacturing Processes of Commercially Important Polyamides 2.2.1 Poly(caproamide) (PA 6)

PA 6 is commercially produced by the ring-opening polymerization of ε-caprolactam (Scheme 2.2). The latter is accomplished by both the hydrolytic and anionic

29

30

2 Polyamide Nanocomposites by In-situ Polymerization

N

H

C

O

n R

-[HN-R-CO]n-

Scheme 2.2 Ring-opening polymerization of a lactam to polyamide.

mechanisms, with the former being by far the most predominantly used method due to its easier control and better adaption for large-scale production. The process generally involves the following steps: caprolactam and additives addition, hydrolysis, addition, condensation, pelletizing, leaching/extraction of monomers, drying, and packaging [8]. In a typical process [8–11], the caprolactam with the desired additives are charged to an autoclave with a small amount (2–4%) of water. A twostage polymerization process follows, with the temperature ranging from 80 to 260 °C. In the ﬁrst stage, water is held in the reactor, the pressure rises, and the hydrolysis and addition steps occur. After a predetermined time, pressure is released and the ﬁnal condensation reaction step occurs. Finally, extraction of the monomer and oligomers takes place. Anionic polymerization is also used in commercial processing. More speciﬁcally, stock or custom-shaped bulk polymer items can be fabricated directly by casting the rapidly polymerizing polymer or using reaction injection molding [12]. In other words, anionic polymerization is essentially a direct monomer-to-ﬁnished item process. In this process, two streams of caprolactam, one containing the catalyst and the other containing the activator, are mixed and then fed into a heated mold. In order to reduce the buildup of internal stresses as the polymer reacts and cools, the rate of reaction and the process temperatures are adjusted. The temperature of anionic polymerization is usually below the melting point of PA 6, at approximately 160 °C. The latter results in a signiﬁcantly lower concentration of residual caprolactam, than expected from the extrapolated concentration as a function of temperature in the melt [13]. As a consequence, the oligomers do not need to be extracted from the bulk plastic article, which renders this process economically feasible. 2.2.2 Poly(hexamethylene adipamide) (PA 6.6)

The commercial production of PA 6.6 constitutes a two- or three-step process, depending on the desired molecular weight of the resin (Figure 2.2). First, the aqueous solution of PA 6.6 salt (Figure 2.3) (70–90% wt) is reacted in an autoclave at temperatures in the range of 175–200 °C while increasing the pressure to minimize the loss of the volatile organic compounds (e.g., hexamethylenediamine), thus performing a solution polymerization step [1, 14–16]. Then, the temperature is further increased (250–270 °C) and the pressure is released to bleed off steam, thus proceeding with a melt polymerization step to drive the reaction toward polymerization (Figure 2.4) [16].

1

2

300 200 100

Pressure, psig

Hexamethylenediamine

Autoclave temperature, °C

2.2 Manufacturing Processes of Commercially Important Polyamides

250 200 150 100 50 0

31

Pressure Temperature

Reaction time

Processing to form product Further increase of molecular weight H2O

3

SSP PA 6.6 Figure 2.2 PA 6.6 production procedure. 1: PA 6.6 salt aqueous solution preparation, 2:

solution-melt polymerization, and 3: solid-state polymerization.

+

H2 N- CH2 6 –NH2 + HOOC- CH2 4 -COOH → +H3 N- CH2 6 –NH3 Figure 2.3

-

OOC– CH2 4 -COO-

PA 6.6 salt formation reaction.

n H2N–(CH2)6–NH2 + n HOOC–(CH2)4–COOH ↔ –[HN–(CH2)6–NHCO–(CH2)4–CO]n– + 2n H2O Figure 2.4 Polycondensation reaction for PA 6.6 formation.

2.2.3 Low-Temperature Polymerization Processes

The solution-melt polyamidation technique is interrupted at a moderate molecular weight product, due to problems arising from the high melt viscosity, the difﬁculties in dissipating heat transfer, and the oxidative degradation of polymer. Higher molecular weights, desired for injection and blow molding applications, improved processability, and end-product properties may be reached by solid-state polymerization (SSP). SSP is extensively applied on an industrial scale as a ﬁnishing stage (post-SSP), owing to the easy material handling, the low extent of product

32

2 Polyamide Nanocomposites by In-situ Polymerization

degradation, and the plain equipment requirement [17, 18]. It involves heating the starting material in an inert atmosphere or in vacuo, at a temperature higher than the glass transition point (Tg) but lower than the onset of melting (Tm). This way, the degree of polymerization of the polymer is substantially increased, while the material retains its solid shape. For example, a typical value of number-average molecular weight (Mn ) derived from polyamides melt technique is in the range of 15 000–25 000 g mol−1, meanwhile resins of Mn > 30 000 g mol −1 are required for injection and blow molding and prepared through solid state ﬁnishing [16]. Furthermore, polymerization of the monomers in a solid state has been proposed (direct SSP) in an attempt to completely overcome the main drawbacks of the melt technology, for example, energy consumption due to high temperatures and use of water as a solvent and heating medium, as well as polymer degradation [16–20]. Accordingly, the monomer crystals react at a temperature lower than the melting point (Tm) of both monomer and polymer under inert gas or vacuum or high pressure. In many cases, the direct SSP reactions are topotactic and the single monomer crystals can be converted into polycrystalline polymer aggregates, permitting the preparation of highly oriented polymers. Despite its laboratory scale, direct SSP presents considerable practical interest, since polymerization occurs from the beginning in the solid state and consequently all the problems mentioned above associated with the high temperatures of melt technology are avoided [20]. Papaspyrides et al. performed numerous studies on polyamide salts direct SSP [21–26], and observed that depending on the reaction conditions, a typical transition of the process from the solid to the melt state (solid–melt transition) occurs. In fact, the monomer is transformed to a polymer by a reaction that rarely takes place in a real solid phase: the salt grains start to melt even if the operating temperature is well below the initial melting point of the salt. It has been suggested that the water produced during the solid state polycondensation hydrates the polar groups of the reactant, and as the amount of water increases, the crystal structure of the salt is destroyed by the formation of highly hydrated, that is, melted regions. After the formation of these melt areas, the reaction proceeds mainly in the melt state, even at temperatures below the initial melting point of the salt and the reaction rate is considerably increased. The above behavior occurs more drastically at higher reaction temperatures. As the polymerization proceeds further, the molecular weight increases, the hygroscopicity of the reacting system decreases, and, eventually, the solid character of the process is restored. Based on the above, under conditions of reasonable reaction kinetics, the transition to the melt and the tendency to agglomerate become inevitable. Therefore, exploiting this phenomenon, an anhydrous melt prepolymerization process was proposed starting from polyamide salt, where the operating temperature is chosen a few degrees below or above the melting point of the salt [27–30]. In fact, operating even below the latter, due to the aforementioned solid–melt transition phenomenon, the melting point will be reduced and the salt grains will agglomerate and turn again to the melt state. In any case, however, the lower temperatures used are beneﬁcial, since undesirable side reactions are not apt to occur and deg-

Reacting mass Tm (°C)

2.2 Manufacturing Processes of Commercially Important Polyamides

Reaction zone

Reaction time

Figure 2.5 Melting point of the reacting mass versus reaction time and operating tempera-

ture zone suggested for the anhydrous prepolymerization of polyamide salts [27] (reproduced with permission from Wiley).

radation is minimized, in comparison to the conventional polyamidation processes, where the operating temperature is much higher. The suggested reaction temperature proﬁle is presented in Figure 2.5, in which the variation of the melting point of the reacting mass versus reaction time is also shown [27]. Furthermore, in order to treat the problem of hexamethylenediamine loss due to volatility reasons and achieve the production of a balanced prepolymer without the need for hexamethylenediamine recovery, an “autogenous” route is suggested: the use of a closed, intensively stirred, vessel, under pressure during the initial stages of the polymerization, is appropriate to ensure that a feed consisting of dry salt with balanced ends would result in a stable and balanced prepolymer. After all the HMD has reacted to form amide bonds, the pressure can be released to allow water removal and shift the reaction to a higher degree of polymerization. Finally, in situations where the reactants are sensitive to high temperature or the polymer degrades before the melting point, the acid chloride route is often used to produce the polyamide, by applying an interfacial polycondensation (IPC) process [1]. Because almost any diacid can be readily converted to the acid chloride, this reaction is quite versatile and several variations have been developed. In IPC, polymerization occurs at an interface between two immiscible, low molecular weight ﬂuids, each containing a different reactant. The dissolved monomers diffuse to the interface, where they undergo a polymerization reaction. The resulting polymer is usually incompatible with the liquid phases and a polymer ﬁlm grows at the interface [1]. This technique features several advantages compared to melt process: bulk stoichiometry is not necessary to produce high molecular weight polymers and short reaction times can be applied. Furthermore, operation at room temperature bypasses the problems associated with the elevated temperatures applied in the melt process. Finally, the applicability of simple lab-bench

33

34

2 Polyamide Nanocomposites by In-situ Polymerization

scale equipment renders the pertinent technique attractive for research purposes. However, there exist severe disadvantages of the process limiting its commercial application. These include the high cost of acid chloride reactants and the large amounts of solvents that must be used and recovered [31].

2.3 Polyamide Nanocomposites 2.3.1 Introduction

The demand for material properties to meet superior and more severe speciﬁcations has motivated vigorous research on polymer nanocomposites, that is, polymer matrices incorporated with ﬁllers with at least one dimension in the nanometer range. In a nutshell, these advanced materials exhibit enhanced thermal, mechanical, barrier, and ﬁre retardant properties over virgin polymers [32–37], while their performance depends on the level and the homogeneity of nanoﬁllers dispersion, as well as on the potential for interfacial bonding between the ﬁller and the matrix. Nanoﬁllers may be nanoclays, carbon nanotubes (single or multiwall) (CNTs), silica, layered double hydroxides (LDHs), metal oxides, etc., offering the promise of a variety of new composites, adhesives, coatings, and sealant materials with speciﬁc properties [32–37]. Among the ﬁllers mentioned, nanoclays have attracted most of the academia and industry interest, due to their abundance as raw materials and to the fact that their dispersion in polymer matrices has been studied for decades [38]. In fact, there are three major polymer nanocomposites categories in terms of nanoﬁller type that are expected to compile the global nanocomposites market in 2011: nanoclay-reinforced (24%), metal oxide-reinforced (19%), and CNTs-reinforced (15%) ones [39–41]. Many thermoplastic and thermosetting polymers of different polarities including, polyamide, polystyrene, polycaprolactone, polypropylene, poly(ethylene oxide), epoxy resin, polysiloxane, polyurethane, etc., have been used for the preparation of polymer nanocomposites [30, 32, 42–48]. More speciﬁcally, focusing on PA 6.6 and PA 6, mainly clay and CNT nanocomposites have been introduced owing to their excellent properties. The combination of exceptionally high interfacial area and small interparticle distance, obtained at low ﬁller loadings (5 wt% maximum), yields hybrids possessing unique properties, typically not shared by their more conventional microscopic counterparts. Improved properties in terms of strength, heat deﬂection temperature, dimensional stability, gas barrier, ﬂame retardancy, and electrical conductivity can be listed as highlighting features of polyamide nanocomposites [32–37]. Concerning the commercialization of polyamides nanocomposites, the ﬁrst product was launched in 1987 by Toyota, and was a 4 wt% PA 6–clay nanocomposite shaped as timing belt cover to be used in Starlet® (Figure 2.6). The pertinent

2.3 Polyamide Nanocomposites b)

a)

Figure 2.6 Examples of shaped products of PAs nanocomposite commercially available:

(a) timing belt cover in automotive industry and (b) barrier beverage container.

nanocomposite presented superior thermal stability and stiffness to that of the virgin material [41]. Another application of PA nanocomposites is represented by PA 6 and MXD6–clay nanocomposites, which are part of multilayer ﬁlm packaging. This consists of poly(ethylene terephthalate) and the nanocomposite system, to enhance the impermeablity of the packaging to O2 and CO2, for example, in beverage bottles (Figure 2.6) [33, 41]. On the other hand, indicative companies currently producing PA nanocomposites are Ube (PA 6, 6.6, and 12–clay nanocomposite) addressing automotive and fuel systems applications, Nanocor (MXD6– clay nanocomposite) for barrier beverage containers [33], Honeywell (PA 6–clay) for medium barrier bottles and ﬁlms, Unitika (PA 6–clay nanocomposite) for food barrier and automotive applications, and RTP Company (PA 6–clay nanocomposite) for extruded ﬁlm or sheet addressing packaging applications [33]. The most popular route to PA nanocomposites is the melt intercalation method, which consists of melt processing the polyamide and the nanoﬁller to obtain maximum dispersion level [32–35, 49–53]. However, this process bears the disadvantages of polymer thermal degradation due to this additional thermal processing stage [54, 55]. On the other hand, the in-situ intercalative polymerization, which consists of polymerizing the monomer in the presence of the nanoﬁller [32–35], seems particularly attractive; it bypasses the aforementioned disadvantages of melt intercalation, being cost-efﬁcient since there is no need of additional processing steps to form the nanocomposite, and exploits the advantages of in-situ intercalation technique in terms of nanoﬁller dispersion potential. In the following section, the major research ﬁndings on polyamide in-situ intercalative polymerization are summarized, presenting different approaches per

35

2 Polyamide Nanocomposites by In-situ Polymerization

matrix type, namely lactam/amino acid-based PAs (AB type) and diamine- and diacid-based PAs (AABB type). 2.3.2 Lactam/Amino Acid-Based In-situ Intercalated PA Nanocomposites

PA nanocomposites were formed by applying the in-situ intercalation method on PA 6 matrix, through the pioneering work presented by the Toyota research group [55]. Indeed, they were the ﬁrst to report the preparation of PA 6–clay nanocomposites through the ring-opening of ε-caprolactam in the presence of montmorillonite (MMT) organically modiﬁed with α,ω-amino acids (COOH-(CH2)n−1-NH3+, with n = 2, 3, 4, 5, 6, 8, 11, 12, and 18) [55, 56]. The role of the organic modiﬁcation was for the ﬁrst time indicated, since clay swelling was found to be strongly affected by the number of carbon atoms in α,ω-amino acids (Figure 2.7). More speciﬁcally, a high extent of intercalation requires a high number of carbon atoms in the ω-amino acid (>12), permitting caprolactam molecules to insert clay interlayer spacing and subsequently therein polymerize to obtain the nanocomposite [55, 57]. In this work, the more successful preparation route involved mixing 12-aminolauric acid-modiﬁed montmorillonite (12-MMT) with ε-caprolactam, and the mixture was heated at 250–270 °C for 48 h. Depending on the amount of 12MMT introduced, either exfoliated (for less than 15 wt%) or intercalated structures (from 15 to 70 wt%) were obtained, as evidenced by X-ray diffraction (XRD) and

8

Basal spacing (nm)

36

7

Basal spacing of montmorillonite without ε-caprolactam

6

Basal spacing with ε-caprolactam at 25°C

5

Basal spacing with ε-caprolactam at 100°C

4 3 2 1 0

0

5

10

15

20

Carbon number of w -amino acid Figure 2.7 Basal spacing of montmorillonite versus carbon number of amino acid [55]

(reproduced with permission from Wiley-VCH).

2.3 Polyamide Nanocomposites Nylon-6/clay Hybrid

Monomer

NCH polymerization

layered clay mineral blending + Polymer (nylon-6)

Nylon-6/clay Composite:NCC

Figure 2.8 Schematic illustration for the synthesis of PA 6–clay nanocomposite [55]

(reproduced with permission from Wiley-VCH).

transmission electron microscopy (TEM). A conceptual scheme for the synthetic method is presented in Figure 2.8 [55]. Further work demonstrated that intercalative polymerization of ε-caprolactam could be realized without the necessity to premodify the MMT surface, thus using the lactam both as monomer and as organic modiﬁer. Indeed, this monomer was able to directly intercalate the Na+-MMT in water in the presence of hydrochloric acid (HCl), as proved by the increase in interlayer spacing from 10 to 15.1 Å. At a high temperature (200 °C), in the presence of excess ε-caprolactam, the so modiﬁed clay can be swollen again, allowing for the ring-opening polymerization to proceed at 260 °C, when 6-aminocaproic acid is added as an accelerator. The resulting composite did not present any diffraction peak in XRD and TEM, observation agreeing with a molecular dispersion of the silicate sheets [55, 58]. However, the caprolactam-HCl-intercalated clay was ﬁltered with much difﬁculty due to its hydrophilic character, therefore the same researchers group proposed and optimized a process to avoid this step, by carrying the synthesis in “one pot” [55, 59]. For this purpose, acid was added directly to an aqueous suspension of ε-caprolactam and clay and heated to 250 °C to obtain the nanocomposite. They observed that the system was sensitive to the nature of the acid used to promote the intercalation of ε-caprolactam; the obtained results showed that, for unclear reasons, only phosphoric acid allowed for the preparation of a truly exfoliated nanocomposite. However, since many polymerization vessels are made of stainless steel, phosphoric acid might cause corrosion of the vessel, thus rendering this method unpractical.

37

38

2 Polyamide Nanocomposites by In-situ Polymerization

In an effort to evaluate the in-situ polymerization technique versus the melt intercalation for the preparation of PA 6–clay nanocomposites, Tung et al. [60] performed a comparative study in terms of rheological and mechanical behavior. Based on XRD and TEM measurements, the organoclay platelets were very well exfoliated in both alternative techniques applied. A slightly better dispersion in the case of the in-situ intercalated PA 6–clay nanocomposite was revealed by TEM, based on the absence of agglomerates. Meanwhile, a broader organoclay platelets length distribution was observed in nanocomposites prepared by melt blending, which was attributed to extensive damage of the silicate layers due to the high shearing forces generated during extrusion. The in-situ polymerized nanocomposite exhibited higher melt viscosity and higher tensile ductility than the meltblended one, which was related to improved dispersion and polymer–silicate interactions in this material. Furthermore, the slightly higher tensile modulus and degree of ductility, observed in the relevant grade compared to the melt-blended one, were related to better dispersion achieved by the in-situ method. Indeed, dispersion and adhesion between clay and the polymer matrix were highlighted as dominant issues in mechanical performance of the nanocomposites, since clay agglomerates act as stress concentrators in the matrix, initiating the failure under tensile load. Rothe et al. [61] studied the reactive extrusion process for the preparation of PA 6–clay nanocomposites. Their approach consisted of dispersing the silicates in caprolactam monomer and performing the anionic polymerization in a twin-screw extruder, leading to a good level of clay dispersion, based on TEM micrographs. The 0, 2, and 4 wt% PA 6–clay nanocomposites showed improved mechanical properties compared to the virgin PA 6. More speciﬁcally, 2 wt% nanoclay addition resulted in a 20% increase in Young’s modulus, while 4 wt% nanoclay addition raised the modulus up to 33%. Turning to other PA matrices, Reichert et al. [62] synthesized PA 12–clay nanocomposites. For this purpose, they used 12-aminolauric acid (ALA) both as monomer and as clay organic modiﬁer. In order to study the increase of clay interlayer spacing with respect to swelling conditions, they used different amounts of ALA in HCl solution, with the latter serving as ALA protonizing agent. First, they studied the effect of ALA concentration on clay swelling process; based on XRD, it was found that it can be separated in two regimes: a cation exchange of inorganic cations by protonated ALA at low ALA concentration and a further diffusion of zwitterionic ALA into the interlayer spacing, when the ALA concentration exceeds that of HCl in the medium (Figures 2.9 and 2.10). The swelling was found to be independent of the temperature, the layered silicate concentration, and the type of acid used to protonate ALA (HCl, H2SO4, and H3PO4). ALA was then polymerized with both types of swollen clay, and the resulting structures were partially exfoliated or intercalated nanocomposites, based on XRD, TEM, coupled with energy dispersive X-ray and atomic force microscopy. PA 11–clay nanocomposites have also been reported through the in-situ intercalation method, by using 11-aminoundecanoic acid both as monomer and as clay organic modiﬁer [63]. For less than 4 wt% of clay loading, the nanoﬁller was uni-

2.3 Polyamide Nanocomposites

Interlayer distance (nm)

2.2 2.0 1.8 1.6 1.4 1.2 Cation exchange

1.0

swelling with an excess of ALA

0.8 0

10

30 40 50 20 ALA concentration (mmol/L)

60

Interlayer distance of synthetic clay Somasif ME 100 in function of an increasing amount of ALA used as the organic modiﬁer [62] (reproduced with permission from Wiley-VCH).

Figure 2.9

0.95 nm

cation exchange with protonated ALA

swelling with an excess of zwitterionic ALA 1.70 nm over 5 nm

Schematic representation of the swelling behavior of the synthetic clay Somasif ME 100 in the presence of ALA [62] (reproduced with permission from Wiley-VCH).

Figure 2.10

formly dispersed in PA 11 matrix, based on XRD and TEM analysis. The presence of clay in PA 11 increased the crystallization temperature and the thermal stability of the polymer, while the latter was found to be dependent on the quality of the nanostructure achieved, being superior for the exfoliated nanocomposites (<4 wt% clay) as compared with the intercalated ones (>4 wt% clay). Furthermore, the nanocomposites showed much higher dynamic modulus and stronger shear thinning behavior. Another nanoﬁller type that has been reported to form PAs nanocomposites by in-situ intercalation are CNTs. Gao et al. [64] prepared PA 6-single-wall carbon nanotubes (SWNTs) by in-situ polymerizing caprolactam in the presence of carboxylic acid-functionalized SWNTs (Figure 2.11). They reported an efﬁcient dispersion of the nanoﬁller in the monomer and a subsequent grafting of the PA 6 chains to the CNTs, through a condensation reaction between the SWNTs carboxyl groups

39

40

2 Polyamide Nanocomposites by In-situ Polymerization H O

O

+ NH

NH + H2N(CH2)5COOH

H2N(CH2)5COO–

+

Protonated monomer O + H3N

Initiation 250 °C

C

H N

C

O

N H

COO–

n

O OH

+

Termination

O

O HN

C O

H N

C

N H

COO–

n

Figure 2.11 Synthesis of PA 6-SWNTs by in-situ polymerization of ring-opening polymerization

of caprolactam in the presence of COOH-functionalized CNTs [64] (reproduced with permission from ACS).

and the amine groups of PA 6. As a result of the grafting process, the nanocomposites Young’s modulus, tensile strength, and thermal stability were signiﬁcantly improved. Similarly, Saeed and Park [65] synthesized PA 6-multiwall carbon nanotubes (MWNTs) nanocomposites via the in-situ polymerization technique, using pristine and COOH-functionalized MWNTs. Based on SEM morphology analysis, it was shown that the COOH-functionalized MWNTs were better dispersed in the PA 6 matrix than the pristine ones, owing to the covalent attachment of PA 6 molecular chains to the side walls of MWNTs, which could act as in-situ compatibilizers in the nanocomposites and enhance the dispersion of MWNTs. In terms of physical properties, the crystallization of nanocomposites was increased compared to that of virgin PA 6, due to the nucleation effect of MWNTs, while the thermal stability under nitrogen of the nanocomposites was superior. Turning to the mechanical properties of the nanocomposites, the uniformly dispersed MWNTs improved the tensile properties because of the reinforcement effect. Finally, Yan et al. [66] utilized the in-situ anionic ring-opening polymerization to prepare monomer casting PA 6–MWNTs nanocomposites. The typical functionalization process of MWNTs by COOH groups was not therein applied, as the

2.3 Polyamide Nanocomposites

carboxyl groups usually have a severe inhibiting role on anionic ring-opening polymerization. Therefore, hydroxyl-functionalized MWNTs (OH-MWNTs) were chosen as the nanoﬁller to prepare PA 6–MWNTs nanocomposites, while in order to improve the dispersion of OH-MWNTs in ε-caprolactam monomer, water was used as an auxiliary dispersing agent. The OH-MWNTs were homogenously dispersed in the PA 6 matrix based on TEM micrographs, which was attributed to the strong hydrogen bonding between OH-MWNTs and the PA 6 chains. The well-dispersed OH-MWNTs effectively acted as a heterogeneous nucleation agent, increasing both the crystallization temperature and the degree of crystallization of PA 6. However, they noticed that the thermal stability of the nanocomposites under air environment diminished. The latter was related with the presence of Fe and Ni particles that were used as catalysts during the synthesis of MWNTs, and could easily form metal oxides of Fe or Ni which could catalyze the thermal degradation of PA 6. 2.3.3 Diamine- and Diacid-Based In-situ Intercalated PA Nanocomposites 2.3.3.1 Solution-Melt Polymerization Technique As discussed earlier, the predominant polymerization technique for PA 6.6 formation is the solution-melt process. However, despite the commercial importance of the latter process, few efforts have been reported to expand it on nanocomposites formation by in-situ polymerizing the monomers in the presence of the nanoﬁller. Notably, although the in-situ intercalation process has been extensively, successfully, and commercially applied for the preparation of lactam/amino acid-based nanocomposites, for example, PA 6, a different picture is presented in diamineand diacid-based nanocomposites, for example, PA 6.6, mainly characterized by the poor pertinent literature and lack of commercial application. In an effort to produce highly exfoliated PA 6.6–clay nanocomposites with superior properties by exploiting the advantages of in-situ intercalation technique, Goettler et al. [67] applied the solution-melt polymerization process. For this purpose, the aqueous solution of the diamine and diacid was polymerized in a ﬁrst stage and a melt stage was subsequently performed by evacuation of the reactor. A ﬁnal SSP stage was in some cases applied to increase the molecular weight of the resins. Clay, varying from organically modiﬁed to pristine (Na+), was introduced either in the solution stage of the process or in the melt one, following water removal. It was noted that various parameters affected the dispersability of clay in the PA matrix, such as the application of high shear (e.g., through sonication of the silicate before the polymerization process), the diamine–diacid salt concentration, the diamine–diacid ratio in the polymerization mixture, the pH of the relevant solution, clay concentration, and organic modiﬁcation type. Moreover, parameters associated with the PA matrix were considered to determine the dispersing potential of clay, such as PA molecular weight at clay incorporation time and PA composition, for example, homopolyamide (PA 6.6) or copolyamide (PA 6/6.6).

41

42

2 Polyamide Nanocomposites by In-situ Polymerization

Overall, a difﬁculty to ﬁnely disperse the silicate in the PA matrices was reported by Goettler et al. [67], especially in PA 6.6 case, in which clay aggregates were detected in the matrix. The latter was related with the ionic nature of the monomer diamine–diacid salt, the aqueous solution ionic conditions, or in a more generalized manner of speaking, as therein mentioned, with the “interactions between the organically modiﬁed clay and the inherently ionic environment dominating the polymerization process.” In order to overcome these limitations, the research group suggested the insertion of clay in the second stage of the process, that is, the melt one, where oligomers would have been already formed, so as to avoid clays exposure to the ionic conditions of the solution stage. Furthermore, in an optimum synthetic route toward a ﬁnely exfoliated nanocomposite, PA 6/6.6 (97.5/2.5% [w/w]) copolyamide formation was suggested, thus minimizing PA 6.6 presence in the ﬁnal nanocomposite. In this route, ALA-modiﬁed clay was applied, essentially resembling the Toyota process. In contrast, Boussia et al. [48] applied the traditional PA 6.6 production route, that is, a solution-melt polymerization process followed by extrusion, in an effort to evaluate the efﬁciency of the technique in terms of the nanostructure achieved and to highlight the effect of presence of nanoclays on the process itself. For this purpose, organoclays of different surfactant type and extent of cation exchange were tested. Some ﬂocculation of all organoclay types was encountered after the solution-melt process, as indicated by a diminishment of clay interlayer spacing at the same value of 13.6 Å, compared with the higher initial value of each one. The latter picture was improved after extrusion in some cases, leading to partially exfoliated nanocomposites. In the same work [48], a noteworthy acceleration of the polycondensation reaction by up to ≈75% was observed in the presence of the fully organically exchanged organoclays, which was attributed to a suggested chain extension mechanism based on the role of clay SiOH groups (Figure 2.12). On the contrary, the use of partially exchanged organoclay containing Na+ cations was proved nonbeneﬁcial for polymerization kinetics, due to presumable inorganic cations hydration and water clusters formation, thus affecting the reversible polyamidation reaction [68]. The latter ﬁnding highlighted for the ﬁrst time that when designing an in-situ intercalation approach for polyamide matrices, the choice of a suitable organoclay type should also be driven by its potential to affect polyamidation kinetics, apart from considering dispersability issues. Indeed, the inﬂuence of nanoclay presence in polyamidation processes has been evaluated in recent studies by Boussia et al. [69, 70] considering the SSP reaction of PA 6.6–clay nanocomposites. The SSP kinetics of 1 phr nanocomposite revealed a 53% reaction acceleration in clay presence, extending the nanoﬁllers incorporation not only for the modiﬁcation and/ or improvement of materials properties but also for their performance as “multifunctional” catalyst systems: this “nanocatalysis” was attributed to a synergistic effect of (i) nanocomposite crystals morphology that increased the end groups concentration in the amorphous regions and “forced” them to react, (ii) clay action as chain extender, as discussed earlier (Figure 2.12), and (iii) clay action as thermal stabilizer to the PA matrix [69].

2.3 Polyamide Nanocomposites H2N

O

H2N

C OH H2N

O OH Si

C OH

OH

OH

Si

Si

H2N

C

OH

OH

O

Si

Si

C NH

C OH

O

O

OH

OH

OH

Si

Si

Si

O

Figure 2.12 Mechanism of clay action during solution-melt polymerization process and

subsequent extrusion of PA 6.6 nanocomposites [48].

Finally, when jointly examining the performance of nanoclay with a phosphorousbased catalyst during SSP of PA 6.6 nanocomposite, Boussia et al. [70] demonstrated that the catalytic action was even more pronounced, resulting in tripling the reaction rate. However, in the presence of the phosphorous-based catalyst alone, the reaction rate was quintupled, implying signiﬁcant counteractions between the catalyst and the nanoﬁller, mainly attributed to the role of polycondensation water. The latter comprises one of the most important parameters in solidstate polyamidation reactions, as extensively studied by Papaspyrides et al. [21–26, 71–73]. Accordingly, clays hydrophilicity (5–10 wt%) was considered to act as a polycondensation water “trap,” thus hindering the escape of the by-product and affecting the reversible polyamidation reaction [70]. 2.3.3.2 Anhydrous Melt Polymerization Technique In an effort to investigate the incapability of clays to fully disperse in PA matrices derived from diamine and diacid monomers, during the solution-melt process [48, 67], Boussia et al. [30] studied the formation of PA n.6–clay nanocomposites (n = 2, 6, and 12) through applying an anhydrous melt polymerization process. The latter is characterized by the absence of water as solvent and benign time–temperature proﬁles, so as to minimize ionic conditions that have been considered to affect clays dispersability [67], as well as the risk of thermal degradation of organoclays surfactants. The ﬁndings of the pertinent work revealed that the resulting PA n.6–clay structures were found dependent on the diamine moiety length: in the case of shorter

43

44

2 Polyamide Nanocomposites by In-situ Polymerization

diamine PAs, that is, PA 2.6 and PA 6.6–clay nanocomposites, ﬂocculation of the silicate occurred, as evidenced by the collapse of its interlayer spacing at 13.3 and 13.6 Å, respectively, as compared to its initial value of 18.2 Å. On the contrary, in the longer diamine case, that is, PA 12–clay nanocomposite, an intercalated structure was obtained, as the interlayer spacing increased to 21.0 Å. These ﬁndings revealed an intrinsic interaction between the polyamide salt diamine cations and the organoclay surfactant ones. A pertinent mechanism was proposed, according to which an ion exchange occurs between these two competitive moieties, due to the bisfunctionality and small size of salt diamine, leading to the bridging of the silicate layers (Figure 2.13) [30]. As a result, in short-aliphatic-segment diamine PAs, such as PA 2.6 and PA 6.6, clay interlayer distance is locked at 13.3 and 13.6 Å, respectively, value insufﬁcient to allow intercalation of PA chains. On the contrary, in longer diamine salts, like in the PA 12.6, the resulting interlayer spacing after the cation exchange is adequately high to allow for the polyamide to intercalate and expand the silicate sheets, yielding intercalated nanocomposite structure. Another work on the formation of PA 6.6–clay nanocomposites by applying the anhydrous melt polycondensation of the relevant salt was performed by Song et al. [74]. They highlighted the importance of a suitable organoclay surfactant, namely aminocaproic acid, and that of using a low cation exchange capacity clay, so as to avoid inhibition of the surfactant spatial arrangement due to its high concentration, on the intercalation of the growing polymer chains. They reported on the formation of exfoliated nanocomposite structures, meanwhile, that clay ﬂocculation was encountered in some cases, which was therein attributed to thermal degradation of the organoclay surfactant and to the subsequent irreversible diminishment of the clay interlayer spacing. In terms of properties, the exfoliated nanocomposites were characterized by enhanced thermal stability and ﬂame retardancy compared with virgin PA 6.6. In contrast, Wu et al. [75] formed PA 10.12–clay nanocomposite through the anhydrous melt polycondensation of the relevant salt. In the derived nanocomposite, clay platelets were exfoliated, developing strong interaction with the polyamide – a fact that was considered to endow the nanocomposite with higher tensile strength and higher tensile modulus. Moreover, the nanocomposite exhibited a reduction in water absorption as a result of barrier properties enhancement. 2.3.3.3 Direct SSP Technique Although the process of direct SSP of PA salts is not commercially applied for the production of PAs, it presents considerable practical interest, since polymerization occurs from the beginning in the solid state and bypasses all the problems associated with the high temperatures of melt technology, for example, energy consumption and polymer degradation [20]. The up-to-date research on PA nanocomposites formation by applying the relevant polymerization technique is rare, and it concerns PA 6.6–LDH nanocomposites. More speciﬁcally, Zhu et al. [76] performed direct SSP on a dry PA 6.6–LDH nanocomposite salt, which had been formed at an earlier stage. The cationic LDH layers were ﬁrstly organically modiﬁed with adipate ions to ensure compatibility with the PA matrix, obtaining an interlayer

2.3 Polyamide Nanocomposites

Figure 2.13 Indicative schematic depiction

of cation exchange mechanism between the surfactant of organoclay and PA 6.6 salt diamine [30] (reproduced with permission

from Wiley). • = NH3+, 䊊 = COO−, = polyamide salt methylene groups, = surfactant tallow.

45

46

2 Polyamide Nanocomposites by In-situ Polymerization

spacing of ∼13 Å. In sequence, equimolecular portions of adipic acid and hexamethylenediamine were added to the adipate-LDH slurry, thus forming the LDHcontaining PA 6.6 salt. The latter was then polymerized at 190 °C for 200 min, proﬁle that has been proved to be efﬁcient in a relevant patent for direct SSP of PA 6.6 salt by Papaspyrides et al. [28]. It was found that LDH dispersed in the PA matrix at different degrees, yielding from aggregated to exfoliated structures, while it was noted that for higher nanoﬁller concentrations, aggregate formation was more dominant and the distribution of the LDH particles throughout the matrix appeared to be inhomogeneous. However, it is noted in this work that in most cases the occurrence of polymerization did not signiﬁcantly improve the dispersion of LDH, that is, nanocomposite salt structure was maintained after polymerization step. This is in agreement with the aforementioned work by Boussia et al. [30]: clay interlayer spacing collapsed at 13.6 Å during nanocomposite salt formation (2.5 wt% clay), which was proved irreversible during either subsequent anhydrous melt or solution-melt polymerization. The latter was attributed, as analyzed previously, to bridging of the silicate layers by hexamethylenediamine cations, which locked their distance at this speciﬁc value [30]. It is challenging to assume that something similar happens in the study of Zhu et al. [76]: in an analogous manner, the cationic LDH layers are modiﬁed with adipate ions, yielding an interlayer spacing of ∼13 Å, which could be considered low to allow for the growing PA 6.6 chains to intercalate. Further research on the synthesis of PA 6.6 nanocomposites with adipatemodiﬁed LDH by direct SSP was conducted by Herrero et al. [77], following the same polymerization process with Zhu et al., that is, directly polymerizing nanocomposite PA 6.6–LDH salts of different concentrations (0.1, 0.5, 1, 2, and 5 wt%) at 190 °C for 200 min. The LDH in this work was also modiﬁed with adipate ions. The XRD spectra of the derived nanocomposite salts indicated exfoliation of LDH layers at low concentrations (0.1 and 0.5 wt%), while at the highest one tested, that is, 5 wt%, formation of aggregates was suggested by the LDH peak presence, in agreement with the previous ﬁndings by Zhu et al. (Figure 2.14). The resulting PA 6.6–LDH nanocomposites showed that the best dispersion level is achieved in the low LDH-loaded nanocomposites, while some residual tactoids and particle agglomerates still exist at high concentration (5 wt%). Moreover, the presence of LDH enhanced the thermal stability of PA 6.6 matrix. 2.3.3.4 Interfacial Polycondensation Technique Despite the small-scale application of the IPC technique, several researchers have reported on the formation of PAs nanocomposites by the relevant process. Kalkan et al. [78] studied the formation of PA 6.6–clay nanocomposites, in the case of either organically modiﬁed or pristine (Na+) clay [79]. In the ﬁrst case [78], they found that in order to achieve better clay dispersion, stirring during the mixing process of the two phases, that is, the clay-containing adipoyl chloride toluene dispersion and hexamethylenediamine aqueous solution, is essential. Nevertheless, in the case of a polar treated organoclay, where anticipated favorable interactions between the surfactant and the polar polyamide chains should yield a

2.3 Polyamide Nanocomposites 2000

500

I/ cps

10 20 30 40 50 60 70 2θ/Cu Kα

Ad-LDH 5-A-PA6.6sal 0.5-A-PA6.6sal 0.1-A-PA6.6sal PA6.6sal 2

4

6 2θ/Cu Kα

Figure 2.14 Powder XRD patterns of adipate-

modiﬁed LDH (Ad-LDH) and nanocomposite PA 6.6-Ad-LDH salts containing 5, 0.5, 0.1, and 0 wt% LDH (5-A-PA 6.6 salt, 0.5-A-PA 6.6

8

10 salt, 0.1-A-PA 6.6, and PA 6.6 salt, respectively). Inset: full range diagram for sample 5PA 6.6sal [77] (reproduced with permission from Elsevier).

well-dispersed nanocomposite, poor dispersion was observed and attributed to the absence of an optimized organic dispersing solvent. Thus, it was demonstrated that an important parameter governing clay dispersion when applying the IPC method is that clay platelets should be well dispersed in one of the solvent phases before the initiation of polymerization. Similarly, the importance of the selection of a compatible solvent to prepare colloidal suspensions was highlighted also when using a pristine clay (Na+) [79]. Furthermore, the effect of reaction attributes, such as temperature and time, was studied, pointing that low temperatures with short reaction times lead to intercalated nanocomposite morphology, while longer mixing time promotes the dispersion of clay layers. Moreover, another parameter suggested to enhance clay dispersion is the method for adding the clay; it was found favorable to minimize its contact time with hexamethylenediamine, as it was shown that the ionic conditions of hexamethylenediamine aqueous solution led to ﬂocculation of the silicate layers. Another nanoﬁller type that has been used for PA nanocomposite formation through the IPC method are CNTs. Indeed, Haggenmueller et al. [80] adapted this method for the fabrication of PA 6.6–SWNTs nanocomposites. In their synthetic route, SWNTs were incorporated suspended either in water or in toluene. The quality of the nanoﬁller suspension prior to the in-situ polymerization, also in this case, was found to determine to a large extent the nanoﬁller dispersion in the

47

48

2 Polyamide Nanocomposites by In-situ Polymerization

resulting nanocomposite. Regarding the functionalizion of CNTs, it was revealed that the alkyl chains promote the suspension of SWNT and subsequently improve dispersion in the composites. With respect to the resulting nanocomposites properties, nanostructure was notably proved unstable under low shear forces causing nanotube agglomeration, which should be considered during subsequent nanocomposite processing. Furthermore, it was highlighted that the resulting composites properties (e.g., electrical conductivity and mechanical properties) are inﬂuenced by the SWNT characteristics, such as nanotube length, dispersion and type, and degree of the nanotube functionalization. In another approach, PA 6.10–MWNTs nanocomposites were successfully produced by Kang et al. [81] via the IPC of two liquid phases, one containing hexamethylenediamine in the presence of acid-treated MWNTs and the other containing sebacoyl chloride. Based on morphology analysis, the individual MWNTs were uniformly dispersed in the PA 6.10 matrix. The mechanical properties of the 1.5 wt% nanocomposite revealed signiﬁcant enhancement in MWNTs presence, since a 170% increase in Young’s modulus was observed, together with increases in the tensile strength and the elongation at break (about 40% and 25%, respectively). The thermal stability of the composite was also enhanced by the incorporation of MWNT into PA 610 matrix.

2.4 Conclusions

The in-situ polymerization process for the preparation of PA nanocomposites consists of polymerizing the monomer(s), that is, lactam/amino acid or diamine and diacid, in the presence of the nanoﬁller, with the latter including clays, CNTs, LDH, etc. Especially for PAs, the relevant technique is of particular importance as it bypasses the main drawbacks of the melt intercalation, that is, the need for additional melt processing to form the nanocomposite, which often leads to thermal degradation of the PA, except for adding to the cost of the whole process. In this chapter, the literature concerning the application of in-situ intercalation on PA matrices was reviewed, emphasizing on nanoﬁllers dispersability factors, interactions between nanoﬁller and host moieties, as well as highlighting properties enhancement. The compatibility between the coexisting species was featured as dominating parameter for efﬁcient dispersion and consequential properties enhancement, such as the potential for tethering the PA chains on the nanoﬁllers surface, through properly modifying the latter. A pioneering example was provided by the Toyota group, through modifying clay with an amino acid, in order to catalyze the ring-opening polymerization of ε-caprolactam simultaneously with achieving an efﬁcient tethering on the PA chains. Another typical approach in the same direction is to acid-functionalize CNTs, so as to provide the necessary tethering edge for PA chains to grow. On the other hand, when considering layered silicates incorporation in PA matrices based on diamine and diacid monomers, a critical

References

parameter has been found to be the length of the diamine, due to an intrinsic interaction between the diamine and the organoclay alkylammonium surfactant. A competitive cation exchange between them is favored due to prevailing ionic conditions and the bisfunctionality of the diamine, leading to bridging of clay layers and subsequent interlayer collapse, thus controlling the potential for exfoliated nanocomposite structures. Finally, the inﬂuence of nanoﬁllers on the polymerization process itself has been highlighted, indicating reaction kinetics enhancement due to clays presence.

References 1 Weber, J. (1996) Kirk–Othmer

2

3

4

5

6

7

8

9

10 11

12

13

Encyclopedia of Chemical Technology, John Wiley & Sons, Inc., Hoboken, NJ. Korshak, V. and Frunze, T. (1964) Synthetic Heterochain Polyamides, IPST, Jerusalem. Radici, P. (2008) Tecnon Orbichem Polyamide and Intermediates Conference, Congress Proceedings, Germany. Palmer, R. (2002) Encyclopedia of Polymer Science and Technology, John Wiley & Sons, Inc., Hoboken, NJ. De Bievre, B. (2000) 1st World Congress Polyamide 2000: The Polyamide Chain, Congress Proceedings, Switzerland. Krauch, T. (2003) 4th World Congress Polyamide 2003: The Polyamide Chain, Congress Proceedings, Switzerland. Charaf, F. (2000) 1st World Congress Polyamide 2000: The Polyamide Chain, Congress Proceedings, Switzerland. Rao, M.K.K. and Galanty, P.G. (1995) Nylon Plastics Handbook (ed. M.I. Kohan), Hanser Publishers, Munich. Komanicky, V., Cupak, F., and Kvarda, V. (1991) Lactam-Based Polyamides (ed. K. Sporka), CRC Press, Boca Raton, FL. Reimschuessel, H.K. (1977) J. Polym. Sci. Macromol. Rev., 12, 65. Klare, H., Fritzsche, E., and Groebe, V. (1963) Synthetische Fasern Aus Polyamiden, Akademie-Verlag, Berlin. Bongers, J. and Mooij, H. (1995), Nylon Plastics Handbook (ed. M.I. Kohan), Hanser Publishers, Munich. Sweeny, W. and Zimmerman, J. (1969), Encyclopedia of Polymer Science and

14

15

16

17

18

19

20

21 22 23

24 25 26 27

Technology, vol. 10 (ed. N. Bikales), John Wiley & Sons, Inc., New York. Hiemenz, P. (1984) Polymer Chemistry: The Basic Concepts, Marcel Dekker, New York. E.I. Du Pont de Nemours and Company (1984) Preparation of salt solution useful for making nylon. US 4,442,260. Vouyiouka, S.N. and Papaspyrides, C.D. (2009), Solid State Polymerization (eds C. Papaspyrides and S. Vouyiouka), John Wiley & Sons, Inc., Hoboken, NJ. Papaspyrides, C. and Vouyiouka, S. (2009) Solid State Polymerization, John Wiley & Sons, Inc., Hoboken, NJ. Vouyiouka, S., Karakatsani, E., and Papaspyrides, C. (2005) Prog. Polym. Sci., 30, 10. E.I. Du Pont de Nemours and Company (1998) Method for solid phase polymerization. WO Patent 98/23666. Papaspyrides, C.D. and Vouyiouka, S.N. (2009) Solid State Polymerization (eds C. Papaspyrides and S. Vouyiouka), John Wiley & Sons, Inc., Hoboken, NJ. Papaspyrides, C. and Kampouris, E. (1984) Polymer, 25, 791. Papaspyrides, C. (1988) Polymer, 29, 114. Papaspyrides, C. (1996), The Polymeric Materials Encyclopedia (ed. J.C. Salamone), CRC Press, Florida. Papaspyrides, C. (1992) Polym. Int., 29, 293. Papaspyrides, C. (1990) Polymer, 31, 490. Kampouris, E. and Papaspyrides, C. (1985) Polymer, 26, 413. Papaspyrides, C., Vouyiouka, S., and Bletsos, I. (2004) J. Appl. Polym. Sci., 92, 301.

49

50

2 Polyamide Nanocomposites by In-situ Polymerization 28 Tynan, D. and Papaspyrides, C. (E. I.

29

30

31

32 33 34 35 36 37 38 39

40

41 42 43 44 45

46 47

DuPont de Nemours and Company) (1999) Polymer mixing apparatus and method. US 5941634. Boussia, A.C., Vouyiouka, S.N., Porfyris, A.D., and Papaspyrides, C.D. (2010) Macromol. Mater. Eng., 295, 812. Boussia, A.C., Damianou, C.C., Vouyiouka, S.N., and Papaspyrides, C.D. (2010) J. Appl. Polym. Sci., 116, 3291. Odian, G. (2004) Principles of Polymerization, 4th edn, John Wiley & Sons, Inc., Hoboken, NJ. Pavlidou, S. and Papaspyrides, C.D. (2008) Prog. Polym. Sci., 33, 1119. Paul, D.R. and Robeson, L.M. (2008) Polymer, 49, 3187. Giannelis, E.P. (1996) Adv. Mater., 8, 29. Alexandre, M. and Dubois, P. (2000) Mater. Sci. Eng., 28, 1. Kiliaris, P. and Papaspyrides, C.D. (2010) Prog. Polym. Sci., 35, 902. Hatzigrigoriou, N. and Papaspyrides, C.D. (submitted) J. Appl. Polym. Sci. Bower, C.A. (1949) Iowa Agric. Exp. Stn. Res. Bull., 362, 39. BCC Research (2006) Nanocomposites, nanoparticles, nanoclays, and nanotubes. http://www.bccresearch.com/report/ NAN021C.html (accessed 4 February 2011). Global Industry Analysts Inc. (2011) Nanocomposites – a global strategic business report. http://www.strategyr. com/Nanocomposites_Market_Report.as (accessed 26 February 2011). Giordano, G., and Inman, H. (2010) Soc. Plastics Eng., 66, 6. LeBaron, P., Wang, Z., and Pinnavaia, T. (1999) Appl. Clay Sci., 15, 11. Fornes, T., Yoon, P., Keskkula, H., and Paul, D. (2001) Polymer, 42, 9929. Cho, J. and Paul, D. (2001) Polymer, 42, 1083. Manias, E. (2001) Origins of the materials properties enhancements in polymer/clay nanocomposites. http://www.plmsc.psu. edu/∼manias/PDFs/nano2001b.pdf (accessed 20 January 2011). Shelley, J., Mather, P., and DeVries, K. (2002) Polymer, 42, 5849. Chin, I.-J., Thurn-Albrecht, T., Kim, H.-C., Russell, T.P., and Wang, J. (2001) Polymer, 42, 5947.

48 Boussia, A.C., Vouyiouka, S.N., and

49 50 51

52 53

54 55 56

57

58

59

60

61 62

63

64

65 66 67

Papaspyrides, C.D. (in press) J. Appl. Polym. Sci. Chavarria, F. and Paul, D.R. (2004) Polymer, 45, 8501. Liu, X. and Wu, Q. (2002) Macromol. Mater. Eng., 287, 180. Kang, X., He, S., Zhu, C., Wang, L., Lu, L., and Guo, J. (2005) J. Appl. Polym. Sci., 95, 756. Mehrabzadeh, M. and Kamal, M. (2004) Polym. Eng. Sci., 44, 1152. Xie, W., Gao, Z., Pan, W., Hunter, D., Singh, A., and Vaia, R. (2001) Chem. Mater., 13, 2979. Fornes, T., Yoon, P., and Paul, D. (2003) Polymer, 44, 7545. Okada, A. and Usuki, A. (2006) Macromol. Mater. Eng., 291, 1449. Usuki, A., Kojima, Y., Kawasumi, M., Okada, A., Fukushima, Y., Kurauchi, T., and Kamigatio, O. (1993) J. Mater. Res., 8, 1179. Usuki, A., Kawasumi, M., Kojima, Y., Okada, A., Kurauchi, T., and Kamigaito, O. (1993) J. Mater. Res., 8, 1174. Kojima, Y., Usuki, A., Kawasumi, M., Okada, A., Kurauchi, T., and Kamigaito, O. (1993) J. Polym. Sci., Part A: Polym. Chem., 31, 983. Kojima, Y., Usuki, A., Kawasumi, M., Okada, A., Kurauchi, T., and Kamigaito, O. (1993) J. Polym. Sci., Part A: Polym. Chem., 31, 1755. Tung, J., Gupta, R.K., Simon, G.P., Edward, G.H., and Bhattacharya, S.N. (2005) Polymer, 46, 10405. Rothe, B., Elas, A., and Michaeli, W. (2009) Macromol. Mater. Eng., 294, 54. Reichert, P., Kressler, J., Thomann, R., Mulhaupt, R., and Stoppelmann, G. (1998) Acta Polymerica, 49, 116. Zhang, X., Yang, G., and Lin, J. (2006) J. Polym. Sci., Part B: Polym. Phys., 44, 2161. Gao, J., Itkis, M., Yu, A., Bekyarova, E., Zhao, B., and Haddon, R. (2005) J. Am. Chem. Soc., 127, 3847. Saeed, K. and Park, S.-Y. (2007) J. Appl. Polym. Sci., 106, 3729. Yan, D., Xie, T., and Yang, G. (2009) J. Appl. Polym. Sci., 111, 1278. Goettler, L., Joardar, S., Middleton, J., and Lysek, B. (Solutia Inc.) (2002)

References

68

69

70

71 72 73

74

Method for preparation of polyamide nanocomposite compositions by in situ polymerization. WO Patent 0,009,571. Vouyiouka, S., Papaspyrides, C., Weber, J., and Marks, D. (2007) Polymer, 48, 4982. Boussia, A.C., Konstantakopoulou, M.O., Vouyiouka, S.N., and Papaspyrides, C.D. (2011) Macromol. Mater. Eng., 296, 168. Boussia, A.C., Konstantakopoulou, M.O., Vouyiouka, S.N., and Papaspyrides, C.D. (in press) Macromol. Mater. Eng. Papaspyrides, C. and Kampouris, E. (1986) Polymer, 27, 1433. Papaspyrides, C. and Kampouris, E. (1986) Polymer, 27, 1437. Katsikopoulos, P. and Papaspyrides, C. (1994) J. Polym. Sci., Part A: Polym. Chem., 32, 451. Song, L., Hu, Y., He, Q., and You, F. (2008) Colloid Polym. Sci., 286, 721.

75 Wu, Z., Zhou, C., Qi, R., and Zhang, H.

(2002) J. Appl. Polym. Sci., 83, 2403. 76 Zhu, Y.D., Allen, G.C., Adams, J.M.,

77

78 79 80 81

Gittins, D., Herrero, M., Benito, P., and Heard, P.J. (2008) J. Appl. Polym. Sci., 108, 4108. Herrero, M., Benito, P., Labajos, F.M., Rives, V., Zhu, Y.D., Allen, G.C., and Adams, J.M. (2010) J. Solid State Chem., 183, 1645. Kalkan, Z.S. and Goettler, L.A. (2009) Polym. Eng. Sci., 49, 1491. Kalkan, Z.S. and Goettler, L.A. (2009) Polym. Eng. Sci., 49, 1825. Haggenmueller, R., Du, F., Fischer, J., and Winey, K. (2006) Polymer, 47, 2381. Kang, M., Myung, S.J., and Jin, H.-J. (2006) Polymer, 47, 3961.

51

53

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization Abolfazl Maneshi, João Soares, and Leonardo Simon

3.1 Introduction

Polyoleﬁns are a major class of commodity synthetic polymers. The technology for the production of these important polymers is well established, from catalyst synthesis to polymerization reactor technology. Despite constant advancements in polyoleﬁn production technology, applications of polyoleﬁns are still mainly limited to commodity products. The recent interest in the production of polyoleﬁn– clay nanocomposites extends the use of polyoleﬁns to specialty and engineering plastic applications. Polyoleﬁn–clay nanocomposites are lighter than conventional composites, but have thermal stability, barrier, and mechanical properties that are comparable to those of engineering plastics. Currently, the only commercial production method for polyoleﬁn–clay nanocomposites is melt mixing. In this method, clay layers are dispersed within the polymer matrix under high shear rates, allowing polymer chains to diffuse into the interlamellar spaces between the clay platelets. Usually, due to low compatibility between the polar clay surface and the nonpolar polymer chains, the clay must be subjected to organic modiﬁcation reactions to improve the stability of the ﬁnal nanocomposites. A common approach is to modify the clay through ion-exchange reactions with bulky ammonium cations prior to melt mixing. Functionalized polymers, such as polyethylene or polypropylene grafted with maleic anhydride may also be used to ensure good compatibility between the organic and inorganic nanocomposite phases. As an alternative to melt mixing, in-situ polymerization is an attractive technique for the preparation of polyoleﬁn–clay nanocomposites because it can promote better clay exfoliation and dispersion in the polymer matrix [1]. During in-situ polymerization, a coordination catalyst (such as Ziegler–Natta, metallocene, or late transition metal complex) is supported onto the clay interlayer surface to make polyoleﬁn chains directly between the clay layers, leading to their exfoliation and dispersion into the polymer phase. Oleﬁn in-situ polymerization methods for the production of polyoleﬁn–clay nanocomposites still face many challenges before becoming industrially relevant, In-situ Synthesis of Polymer Nanocomposites, First Edition. Edited by Vikas Mittal. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA.

54

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization

but they are also very promising routes for the production of novel materials. The references reviewed in this chapter constitute the building blocks for future developments in this important area. We will start the chapter with a short review on clay structures and properties, and will then move into our main topic, in-situ polymerization of oleﬁns with coordination catalyst for the production of polyoleﬁn–clay nanocomposites.

3.2 Clays

In mineralogy, the term clay is used for a variety of polycrystalline materials that are well described in clay science, mineralogy properties, and characterization textbooks [2–5]. Clays can be present in ﬁbrous, tubular, lath shaped, and planar geometries. In this chapter, however, our focus will be mainly on the planar clay varieties called smectites that include montmorillonites, the most commonly used clay for the production of polyoleﬁn–clay nanocomposites. In this section, we will focus on clay characteristics that are relevant to catalyst supporting and particle break-up during polymerization: clay chemistry, crystalline structure, and geometry. 3.2.1 General Structure

Planar clay minerals consist of two basic units: an octahedral sheet and a tetrahedral sheet. The octahedral sheet is comprised of closely packed oxygen atoms and hydroxyl groups, in which aluminum, iron, and magnesium atoms are arranged in the octahedral coordination. The second structural unit is the silica tetrahedral layer, having a silicon central atom and four oxygen atoms, or possibly hydroxyl groups, arranged in the form of a tetrahedron. These tetrahedra form a hexagonal network that is repeated inﬁnitely in two horizontal directions to form a silica tetrahedral sheet. The silica tetrahedral sheet and the octahedral sheet share the apical oxygen atoms or hydroxyl groups, forming 1 : 1 layers (kaolinite) or 2 : 1 layers (montmorillonite). The structures and compositions of these two major industrial clays (kaolins and smectites) are very different, even though they share the same octahedral and tetrahedral sheet patterns. The arrangement and composition of these octahedral and tetrahedral sheets account for most of the differences in their physical and chemical properties. Among different types of clays, smectites, including montmorillonite, are the most popular for the production of polyoleﬁn– clay nanocomposites [6]. 3.2.2 Smectites

The basic smectite structure if formed by an alumina octahedral layer sandwiched between two silica tetrahedral layers. These layers share the apical oxygen atoms

3.2 Clays

Exchangeable Cations n H 2O

Oxygens Hydroxyls Aluminum, Iron Magnesium Silicon, occasionally aluminum Figure 3.1 Smectite structure (reproduced from Ref. [5]).

of the silica tetrahedral sheets, as depicted in Figure 3.1 [5]. Water molecules and cations occupy the space between the layers. Their theoretical formula is (OH)4Si8Al4O20.nH2O and their theoretical composition without the interlayer material is SiO2 (66.7%), Al2O3 (28.3%), and H2O (5%). However, smectites may have substitutions in their octahedral and tetrahedral sheets that create a charge imbalance on the layer structure that needs to be balanced with alkali or earth alkali metal cations. Calcium montmorillonite (Ca2+ MMT), in which the layer charge deﬁciency is balanced by interlayer calcium cations and water molecules, is the most common smectite. The basal spacing of Ca2+ MMT is 14.2 Å. Sodium montmorillonite (Na+ MMT) occurs when the charge deﬁciency is balanced with Na+ ions and water; its basal spacing is 12.2 Å. Because of their small crystalline domains, smectite particles are considered to be structurally disordered; therefore, the X-ray diffraction data of smectites are sometimes difﬁcult to analyze [7]. Sodium montmorillonite, which is the focus of the major part of research on polyoleﬁn–clay nanocomposites, is made of small thin clay layers with a “cornﬂake texture,” having a high surface area of about 150–200 m2 g−1. Substitution of Al3+ by Fe3+, Fe2+, and Mg2+ in the octahedral sheets of smectites is a common occurrence. Silicon atoms may be substituted by aluminum atoms in tetrahedral sheets, which again creates a charge imbalance [5]. The isomorphous

55

56

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization

substitution of bivalent cations with trivalent cations on clay octahedral layers results in a positive charge deﬁciency and enhances the Lewis base behavior of the hexagonal cavity of the silicate layer to levels sufﬁcient to form complexes with dipolar molecules and cations [3]. 3.2.3 Clay Particle Morphological Hierarchy

The clay particle morphology may be divided in several structural levels, as illustrated in Figure 3.2. The elementary structure is the clay layer. As discussed previously for MMT, these layers are composed of tetrahedral and octahedral sheets. A particle is an assembly of layers. The tactoid stacking of clay layers, in which the clay layers lie on top of each other in a way that resembles a deck of cards, is a common type of particle assembly. An assembly of particles will be referred to as an aggregate. Aggregates may combine to form an assembly of aggregates, shown as superstructure D in Figure 3.2. In practice, clays are highly anisometric, often having irregular particle shapes and broad particle size distribution. The distribution of cation exchange capacity

a)

b)

Layer

c)

Aggregate

Particle

Interlayer Space (gallery) d)

Interparticle Space

Assembly of Aggregates

Interaggregate Space (pore) Figure 3.2 Clay structural hierarchy: (a) clay

interlayer and interparticle spaces; and layer; (b) a particle made up of stacked layers (d) an assembly of aggregates, enclosing (layer transition and deformation can give rise an interaggregate pore (reproduced from to lenticular pores); (c) an aggregate, showing Ref. [2]).

3.2 Clays

Figure 3.3 Schematic of curved montmorillonite particles (reproduced from Ref. [8]).

is not uniform and, therefore, a distribution of interlayer spacing exists in montmorillonites. The individual layers are ﬂexible and can be curved as depicted in Figure 3.3 [8]. Clay particles, in particular those of montmorillonites, are never crystals in the strict sense of the word [8]. In fact, a montmorillonite “crystal” is more equivalent to an assemblage of silicate layers than to a true crystal, as illustrated in Figure 3.3. Montmorillonite particles seen in the electron microscope never have the regular shape of real crystals, but rather look like paper torn into irregular pieces. The core of the particles is surrounded by disordered and bent silicate layers with frayed edges. Layers, or thin particles composed of a few layers, protrude from the packets and enclose wedge-shaped pores. The particles reveal many points of weak contact between the layer stacks. At these breaking points, the particles may easily disintegrate during interlayer reactions or as a result of mechanical forces that inﬂuence the rheological behavior. The electrostatic attractions between the layers and the interlayer cations increase the stacking order in more highly charged 2 : 1 clay minerals; the domains with equally spaced layers become thicker and the inﬂuence of the defects on the shape and position of the basal spacing (0 0 1) reﬂections, the characteristic of interlayer spacing in X-ray diffraction, decreases.

57

58

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization

3.2.4 Clay Chemical Reactions 3.2.4.1 Cation Exchange Reactions Clay interlayer cations can be exchanged with other cation types. The highest possible cation exchange extent on the clay surface is called cation exchange capacity (CEC). In addition to the interlayer cations, montmorillonites also have silanol and aluminol groups on their edges that can be used in cation exchange reactions, depending on the acidic properties of the reaction medium. The substitution within the lattice is responsible for about 80% of the total CEC; the hydroxyl groups on the layer edges account for the remaining 20% [5]. The reactivity of the latter fraction depends on the pH of the dispersion medium. Sodium montmorillonite has a high CEC, generally ranging from 80 to 130 meq/100 g. The high electrical charge on the lattice enables Na+ MMT to exchange the interlayer water and associated cations with more polar organic molecules, such as ethylene glycol, quaternary amines, and polyalcohols, which play important roles during clay organic modiﬁcation reactions. 3.2.4.2 Interaction with Organic Compounds Clays can interact with different types of organic compounds. The penetration of organic molecules into the clay interlayer space is called intercalation. Intercalated guest molecules can be displaced by other suitable molecules: water molecules in smectite interlayer spaces can be displaced by many polar organic molecules; neutral organic ligands can form complexes with the interlayer cations; interlayer cations can be exchanged with various types of organic cations. As will be discussed later, alkyl ammonium ions, mainly quaternary alkyl ammonium ions, are widely used to modify smectites for industrial applications. The silanol and aluminol groups on the montmorillonite layer edges can also be used for grafting reactions. The adsorption of neutral molecules on smectites is driven by various chemical interactions: hydrogen bonds, ion–dipole interactions, coordination bonds, acid– base reactions, charge transfer, and van der Waals forces. Several polar molecules, such as alcohols, amines, and acids, form intercalation complexes with montmorillonites. The intercalation can be performed from the vapor, liquid, or even solid state. In intercalation from solution, solvent molecules are generally coadsorbed in the interlayer space. Guest molecules may be intercalated in dried clay minerals or may displace the water molecules of hydrated montmorillonite. The displacement of interlayer water molecules depends on the hard -and -soft -acid -and -base (HSAB) character of the interlayer cations and the interacting groups of the guest molecules. HSAB, also known as the Pearson acid–base concept, is widely used in chemistry for explaining stability of compounds, reaction mechanisms, and pathways [9]. In the HSAB concept, chemical species are classiﬁed by their Lewis acidic and basic properties by the terms hard, which applies to small species which have high charge states and are weakly polarizable, or soft, which applies to large species which have low charge states and are strongly

3.3 In-situ Polymerization of Oleﬁns with Coordination Catalysts Supported on Clays a)

b)

Figure 3.4 Coordination of amines to the interlayer cations: (a) directly and (b) by water

bridges [9].

Figure 3.5 Ionic bond when a base is protonated in the acidic pH solution.

polarizable [10]. Water molecules near hard cations like Na+, Mg2+, and Ca2+ are displaced only by HO– or O= containing compounds, but not by amines. In contrast, amines as soft bases displace water molecules from soft interlayer cations such as Cu2+ and Zn2+ [9]. Aliphatic and aromatic amines can be directly coordinated to the interlayer cations (Figure 3.4a) or bound by water bridges (Figure 3.4b) [9]. The type of bonding is mainly determined by the hardness or softness of the cations due to the HSAB concept. While soft cations such as Zn2+, Cd2+, Cu2+, and Ag+ bind amines directly, water bridges are formed between amines and hard cations (alkali and earth alkali ions, Al3+).1) In addition to coordination bonds, ionic bonds are often present, especially in aqueous dispersions when the base is protonated due to the acidic pH solution [11, 12] or to the increased acidity of interlayer water molecules (Figure 3.5). The ratio of protonated base to unprotonated base in the interlayer clay spaces is much higher than the ratio in the homogeneous solution (outside interlayer clay space), mainly because of the increased acidity of water in the interlayer space. Protonation is also enhanced by the ability of the negatively charged clay mineral surface to lower the chemical potential of the protonated form of the base relative to the neutral form, and therefore drive the equilibrium toward protonation, as shown in Figure 3.5.

3.3 In-situ Polymerization of Oleﬁns with Coordination Catalysts Supported on Clays

Most commercial polyoleﬁns are produced by coordination polymerization catalysts. When compared to free radical processes used to make low-density polyethylene (LDPE), these catalysts work in comparatively gentle conditions, such as lower pressures and temperatures, while providing greater ﬂexibility in controlling the polyoleﬁn molecular structure. An understanding of the polymerization mechanism with coordination catalysts is essential for designing proper systems for the production of polyoleﬁn–clay nanocomposites and will be covered in the next section. 1)

These cations do not form stable amino complexes in water.

59

60

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization

3.3.1 Oleﬁn Polymerization with Coordination Catalysts

Coordination polymerization is a dominant method for the industrial production of polyoleﬁns. Ziegler–Natta and Phillips catalysts have been used since the early 1950s, and are responsible for most of the industrial production of polyoleﬁns. Metallocenes started being used in the late 1980s, and their impact on the industrial production of polyoleﬁns is increasing rapidly due to their better microstructural control [13]. Metallocenes are also very active and can be supported on a variety of inorganic and organic carriers, making the switch from Ziegler–Natta or Phillips technologies to metallocene-catalyzed polymerization relatively simple. Metallocenes are transition metal complexes in which a central metal atom from group IV is sandwiched between two cyclopentadienyl (or cyclopentadienylderivative) rings. An alkenyl group may connect the two cyclopentadienyl rings (the bridge). The hydrogen atoms on the ligands and the bridge can be substituted by other groups to change the reactivity and stereoselectivity of the catalyst. A typical metallocene is shown in Figure 3.6. The metal (M) is normally Ti, Zr, or Hf, X is normally Cl or a methyl group, and the R substituents are often H or methyl groups, but can also be other ring structures. 3.3.2 Polymerization Mechanism with Coordination Catalysts

Oleﬁn polymerization with metallocene catalysts includes a coordination step between the monomer molecule and the catalyst active site, followed by insertion into the metal–alkyl bond between the catalyst center and the growing polymer chain. These steps are repeated many times to form a polymer chain. Figure 3.7 shows that the active site for oleﬁn polymerization is a positivelycharged metal center. The active site is generated from a neutral metallocene

Figure 3.6 Chemical structure of a general metallocene.

3.3 In-situ Polymerization of Oleﬁns with Coordination Catalysts Supported on Clays

Figure 3.7

Coordination and insertion of an α-oleﬁn (reproduced from Ref. [14]).

Figure 3.8

General MAO structure.

precursor complex that is activated by a cocatalyst. The cocatalyst abstracts a ligand from the metallocene, generating an ion pair composed of a low-valence cation and a charge-balancing noncoordinating counter ion. Different types of cocatalysts may be used with metallocenes, but methylaluminoxane (MAO) is the most common one. MAO is an oligomer produced by the controlled hydrolysis of trimethylaluminum (TMA) [15, 16]. There is still some controversy regarding the molecular structure of MAO. It has been suggested that MAO has a linear, cyclic, or even a three-dimensional cage structure [15, 17, 18]. Figure 3.8 shows a possible MAO structure. As a cocatalyst, MAO generates and stabilizes the cationic species active for oleﬁn polymerization. The reaction between MAO and a zirconocene is shown in Figure 3.9. MAO is a Lewis acid and has high reactivity toward water and other oxygencontaining molecules that can poison the catalyst. Therefore, MAO is also believed to act as an impurity scavenger during polymerization. The high MAO concentrations required for oleﬁn polymerizations using metallocenes is considered one of the major drawbacks to its industrial use. Usually, MAO to metallocene ratios (Al/M) from 1000 to 10 000 are with unsupported metallocenes, but this ratio may decrease signiﬁcantly when metallocenes are supported on silica and other carriers. As an alternative, metallocenes can be stabilized by other bulky noncoordinating anions from molecules such as organoborate salts.

61

62

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization

Figure 3.9 Reaction between MAO and a zirconocene (reproduced from Ref. [19]).

Increasing the Al/M ratio results in higher polymerization activity up to a maximum value, after which a further increase will have little effect on the polymerization rate or may even cause the activity to decrease. Initially the polymerization rate increases with increasing Al/M ratio because the MAO molecules surrounding the active centers help stabilize them. However, it is supposed that after a given Al/M ratio, higher MAO concentrations may deactivate the active sites. 3.3.3 Coordination Catalysts for in In-situ Polymerization

The use of many different coordination catalysts has been reported in the increasing number of publications on in-situ polymerization for the production of polyoleﬁn–clay nanocomposites. In addition to metallocenes, which are the dominant systems in this area, the use of other catalyst types has also been investigated. Iron catalysts [20–24] have been used to make polyethylene–clay nanocomposites where the polyethylene had very broad molecular weight distribution (MWD). Ziegler–Natta [25, 26], organo-chromium (Phillips) [27], and bis(imino)pyridine iron and cobalt catalysts [28] have also been used to make polyoleﬁn–clay nanocomposites. Nickel-based late transition metal catalysts [29, 30], capable of producing highly branched polyethylene from only ethylene and of promoting the copolymerization of ethylene with polar comonomers, have also been applied to make polyoleﬁn–clay nanocomposites. The use of metallocenes for the production of polyoleﬁn–clay nanocomposites has several well-known advantages [31, 32]. Metallocenes can produce polyoleﬁns with narrow MWD and uniform comonomer incorporation. In addition, terminal groups, stereochemistry, short and long chain branching can be controlled depending on the metallocene structure employed [32].

3.3 In-situ Polymerization of Oleﬁns with Coordination Catalysts Supported on Clays

3.3.4 Catalyst Supporting

Metallocene catalysts are soluble in organic solvents and can be used directly in solution or in precipitation polymerization. In solution polymerization, the temperature is kept high enough so that both catalyst and polymer chains are soluble in the reaction medium; in precipitation polymerization, the catalyst is soluble in the reaction medium, but the polymer precipitates as it is formed. Unsupported metallocenes, however, are unsuitable for the production of polyethylene or isotactic polypropylene in gas phase or slurry processes [33]. Most commercial Ziegler–Natta oleﬁn polymerization reactors require a heterogeneous catalyst; therefore, homogeneous catalysts have to be supported onto a carrier to be used with these processes. A supported metallocene/MAO catalyst was ﬁrst suggested by Sinn et al. [34] and Kaminsky et al. [35]. Many types of supported metallocene catalysts have been reported since then and have been surveyed recently by Ribeiro et al. [36]. In order to be used in existing oleﬁn polymerization reactors, a supported catalyst needs to meet some requirements such as polymerization activity comparable to that of the homogeneous catalyst [33], low MAO requirements (low Al/M ratio) for economic viability, the ability produce polymer with high molecular weight at the temperatures commonly encountered in industrial polymerization reactors, and production of polymer particles with high bulk density and controlled powder morphology. Furthermore, the enhanced polymer microstructural control provided by metallocene catalysts should not be lost during catalyst supporting. For the case of polypropylene, the support should also not affect adversely the catalyst regio- and stereoselectivity. Similar requirements hold for the production of polyoleﬁn–clay nanocomposites by in-situ polymerization methods. Usually, the supporting techniques used for clays are adaptations of those commonly used to support metallocenes on silica particles, as described in the next section. 3.3.4.1 Catalyst Supporting Methods Several reviews have been published describing methods used to support metallocenes on silica and other inorganic and organic carriers [33, 37–40]. They are classiﬁed in three main types:

1) 2) 3)

absorption or in-situ production of MAO on the support surface, followed by metallocene impregnation (support/cocatalyst/catalyst); one step supporting of a preactivated metallocene/MAO complex ([catalyst + cocatalyst]/support); direct supporting of the catalyst onto the support surface and activation with a cocatalyst in the polymerization reactor (support/catalyst + cocatalyst).

These catalyst-supporting methods are illustrated in Figure 3.10. In the following section, we will see how these three basic approaches were modiﬁed for using clay as a catalyst support.

63

64

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization

Figure 3.10 Different metallocene supporting methods (reproduced from Ref. [41]).

3.3.5 Clay Surface Modiﬁcation Methods for In-situ Polymerization

Heterogeneous catalysts for oleﬁn polymerization are affected by the type of surface treatment used during supporting, including thermal treatment and reactions with organometallic compounds such as magnesium and aluminum alkyls, SiCl4, and SiMe2Cl2 [36]. Support treatment steps are required mainly to control the population of hydroxyl groups and surface acidity, and to scavenge impurities from the support surface before introduction of the catalyst precursor. In this section, we will review the most common clay modiﬁcation approaches used to make them more suitable as supports for in-situ polymerization processes. 3.3.5.1 Organic Modiﬁcation Organic modiﬁcation reactions are commonly applied to clays to enhance their efﬁciency as supports for polymerization catalysts. During supporting, the polymerization catalyst must achieve a uniform distribution within the clay layers so that interlayer polymerization may take place, leading to clay exfoliation and dispersion into the formed polymer phase. Therefore, the compatibility among the clay surface, organic solvent, catalyst, cocatalyst, and polymer has a great impact on the effectiveness of in-situ polymerization. Surface compatibility between the hydrophilic clay and the organic solvent that carries the reactants is enhanced through cation exchange reactions with alkyl ammonium ions, in which bulky

3.3 In-situ Polymerization of Oleﬁns with Coordination Catalysts Supported on Clays Table 3.1

Quaternary ammonium cations frequently used to prepare organoclays.

Quaternary cation

Abbreviation

Formula

Tetramethylammonium Trimethyl phenylammonium Benzyl trimethylammonium Hexadecyl pyridinium Benzyl dimethyl teradecylammonium Hexadecyl trimethylammonium Dioctadecyl dimethylammonium

TMA TMPA BTMA HDPY BDTDA HDTMA DODMA

(CH3)4N+ C6H5N+(CH3)3 C6H5CH2N+(CH3)3 C5H5N+(C16H33) C6H5CH2N+(C14H29)(CH3)2 C16H33N+(CH3)3 (C18H37)2N+(CH3)2

Reproduced from Ref. [70]).

anions, such as ammonium or phosphonium ions with long alkyl chains (C10–C18), substitute the small interlayer alkali ions to increase the clay basal spacing and render the hydrophilic clay layers more organophilic. Organically modiﬁed clays are sometimes called organoclays [9, 42]. Quaternary ammonium compounds containing alkyl, phenyl, benzyl, and pyridyl groups are the most common alkyl ammonium species used in the preparation of organoclays (Table 3.1). Several interlayer structures have been proposed for alkyl ammonium-exchanged montmorillonites. The alkyl chains are assumed to lie ﬂat on the clay surface as mono- or bilayers, but other arrangements, such as pseudotrimolecular and parafﬁn-type arrangements, have also been proposed [43]. To achieve acceptable activities for in-situ polymerization, the active sites immobilized on the clay surface should be shielded from side reactions that may lead to their deactivation. Pristine clays, such as Na+ MMT, are hydrophilic and have a relatively high water (up to 10 wt%), a fraction of which accompanying alkali metal cations. Water content and the type of water on the clay surface are important factors during supporting, since water is a strong catalyst poison. Depending on the clay structure, water may occur in different forms [9, 34, 44–48] that can be determined using thermal gravimetric analysis: (a) physically adsorbed water, also called zeolitic or free water, is generally released below 100 °C; (b) bound water is released below 300 °C; and (c) structural water is released around 600 °C. Since water is poison [23, 49] for most of oleﬁn polymerization catalysts, direct contact between pristine clays and polymerization catalysts produces supported catalysts with very low polymerization yields or, worse, completely inactive systems [50]. As a result of water displacement during organic modiﬁcation reactions, the water content in organoclays is considerably lower than in pristine clays (typically less than 0.2 wt%), lowering the risk of catalyst deactivation upon catalyst supporting on organoclays [51, 52]. Liu et al. [51] compared the polymerization of ethylene with Cp2ZrCl2 supported on Na+ MMT with montmorillonite modiﬁed with methyl glycinate hydrochloride. They found that the organically modiﬁed clay had a

65

66

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization

considerably higher catalyst loading and higher polymerization activity under similar conditions. There are several different types of commercially available organoclays. These organoclays are produced with different ammonium cations in different concentrations to cover a broad range of hydrophilicities. 3.3.5.2 Thermal Treatment During thermal treatment, physically adsorbed, bound, and structural water is removed at different temperature ranges speciﬁed in the previous section, and the clay porosity and acidity also change. In fact, acidity, porosity, and cation exchange capacity of clays are closely related to their water content [43]. Clays have both Brønsted and Lewis acid sites. Strong Brønsted acidity derives from dissociation of water that is directly coordinated to interlayer cations, and its strength increases with the polarizing power of the cations, that is, with decreasing ionic size and increasing charge. The smaller the amount of hydration water present on the clay surface, the greater the polarization of the remaining water molecules and, hence, their ability to donate protons. Dehydrated interlayer cations are the source of Lewis acidity in clays. A certain degree of Lewis acidity is beneﬁcial during the activation of metallocene catalyst for polymerization [23, 44, 50]. Both Brønsted and Lewis acidity vary greatly with the nature of the interlayer cation. Irrespective of the type of interlayer cation, maximum Brønsted acidity is attained at about 150 °C and decreases signiﬁcantly on further heating. Maximum Lewis acidity is achieved after thermal activation at 250–300 °C and does not change noticeably up to 500 °C [43]. 3.3.5.3 Treatment with Alkylaluminum Compounds Treatment with alkylaluminoxanes, MAO, or alkylaluminum compounds such as AlR3 (R = methyl [Me], ethyl [Et], or tertiary isobutyl [i-Bu]) can remove the residual water on the clay surface and protect the upcoming coordination catalyst from deactivation because alkylaluminum compounds react strongly with water due to the strength of Al–O bonds (−350 kJ mol−1) as compared to Al–C bonds (−255 kJ mol−1) [53]. The complete removal of water from clay requires thermal treatment at temperatures of about 400–600 °C, which leads to the partial collapse of the clay structure and CEC reduction. Even after thermal treatment, alkylaluminum compounds may still be used to scavenge residual water. For the case of organoclays, water content decreases considerably after organic modiﬁcation [45], but it is still large enough (in the form of hydrating water) to cause signiﬁcant catalyst deactivation if left unchecked. In this case, thermal treatment at high temperatures is not an option because it may degrade the organic modiﬁcation; therefore, treatment with alkylaluminum compounds and thermal treatment at moderate temperatures is required for further reduction of the adsorption of water. As an additional beneﬁt, treatment with alkylaluminum compounds will reduce the population of surface hydroxyl groups that may also lead to catalyst deactivation [23, 46, 47].

3.3 In-situ Polymerization of Oleﬁns with Coordination Catalysts Supported on Clays

3.3.6 Particle Break-Up and Exfoliation

Most published in-situ polymerization experiments for the production of clay– polyoleﬁn nanocomposites have been done in slurry polymerization reactors, with only a few taking place in gas-phase reactors [48, 54–56]. In polymerization with catalysts used for slurry or gas-phase production of polyoleﬁns (commonly supported for SiO2 or MgCl2), the ﬁnal particle morphology depends strongly on the particle break-up that takes place during the ﬁrst seconds of polymerization. Similarly, the dispersion of clay nanolayers on the polymer matrix in clay–polyoleﬁn nanocomposites made by in-situ polymerization is inﬂuenced by the initial clay particle break-up and nanolayer dispersion which result from the process of interlayer polymerization. This section presents a brief introduction on particle breakup and growth during oleﬁn polymerization with conventional heterogeneous Ziegler–Natta catalysts and supported metallocenes that is useful for the understanding of clay-supported catalysts. More detailed reviews on this topic have been published previously [57–60]. Laboratory scale oleﬁn slurry polymerizations typically start with the injection of catalyst particles into the reactor. The monomers can be gaseous or liquid, and a hydrocarbon diluent may also be present, forming the continuous phase of the reactor. The monomer is either dissolved in the diluent or, in the case of propylene, present as a liquid. Hydrogen is commonly used as a chain transfer agent to control the polymer molecular weight. When a supported catalyst is used, monomer must diffuse through the boundary layer around the catalyst particle and through its pores to reach the active sites where polymerization takes place. Upon polymerization, the formed polymer is deposited onto the catalyst surface, forming a layer around the active sites. For the polymerization to continue, monomer must be absorbed onto the surface of this polymer layer and diffuse through it until reaching the active sites. After ﬁlling up the catalyst support pores, the hydrodynamic pressure caused by the growing polymer breaks the support into many fragments that are known as micrograins, microparticles, or primary particles. As the polymerization proceeds, the initial catalyst support fragments and is dispersed within the growing polymer matrix. In the case of clay-supported catalysts, one needs to include another level of mass transfer to account for monomer diffusion and polymer formation between the clay platelets. A model to describe this process, called the multilayer model (MLM), is depicted in Figure 3.11. Extensive fragmentation and uniform particle growth are key indications that the replication process is proceeding as desired. Good replication requires the high support surface area, homogeneous distribution of active centers throughout the particle, and free access of the monomer to the innermost zones of the particle. Maneshi et al. [61] developed and used the MLM to investigate clay-supported polymerizations from a theoretical point of view and concluded that, when the active sites were uniformly distributed on the clay surface, a uniform monomer

67

68

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization

Figure 3.11 A schematic of the multilayer model (reproduced from Ref. [61]).

concentration proﬁle is achieved inside the galleries resulting in uniform polymerization rates and homogeneous exfoliation. Particle growth uniformity during in-situ polymerization can be detected by scanning electron microscopy (SEM). SEM images of two polyethylene–clay nanocomposite samples with uniform and nonuniform particle break-up behavior are shown in Figure 3.12 [62]. Polyethylene particles made with metallocene supported on Na+ MMT (Figure 3.12a) have nonuniform particle break-up; in contrast, Figure 3.12b shows that when the same metallocene is supported onto an organoclay, more uniform particles were obtained. The authors attributed this behavior to a better distribution of catalyst molecules in the organoclay. Transmission electron microscopy (TEM) is the main technique to detect intercalation and exfoliation for polymer–clay nanocomposites. Polyethylene–clay nanocomposites samples with poor (Figure 3.13a) and good (Figure 3.13b) exfoliation are shown in Figure 3.13 [62]. Uniform exfoliation and distribution of clay nanolayers is obtained by in-situ polymerization only when ethylene is polymerized with a metallocene supported on the organoclay in this case. A review of methods for the detailed characterization of polymer–clay nanocomposites is beyond the scope of this chapter. Valuable information on this area can be found in several literature reviews [13, 46, 63–68]. Fink et al. [60] have highlighted that the particle growth mechanism is associated with a polymerization kinetic proﬁle in which an initial induction period is followed by an acceleration stage after which, in the absence of chemical deactivation, a stationary polymerization rate is obtained, as indicated in Figure 3.14. This concept is useful for the interpretation of particle break-up during in-situ polymerization with clay-supported catalysts. Maneshi et al. [69] used ethylene

3.3 In-situ Polymerization of Oleﬁns with Coordination Catalysts Supported on Clays a)

b)

Figure 3.12 Morphology of a polyethylene particle made with (a) Cp2ZrCl2/Na+ MMT and

(b): Cp2ZrCl2/Cloisite 93A (reproduced from Ref. [62]).

uptake curves to evaluate particle break-up during the initial stages of slurry polymerization with clay-supported catalysts. Shin et al. [48] used a similar rationale to evaluate gas-phase polymerization with clay-supported catalysts. 3.3.7 In-situ Polymerization Approaches

Although all in-situ polymerization techniques take place as the monomer is polymerized in the presence of a catalyst supported on the clay interlayer spaces, they can be divided into three main categories:

69

70

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization a)

100 nm b)

100 nm Figure 3.13 TEM image of polyethylene–clay nanocomposite samples made with:

(a) Cp2ZrCl2/Na+ MMT and (b) Cp2ZrCl2/Cloisite 93A (reproduced from Ref. [62]).

1) 2) 3)

clay as a polymerization additive, clay as polymerization catalyst support, and clay as an activator for polymerization catalysts.

Each of these categories will be discussed below in more detail. In-situ production of alkylaluminoxanes can be used with any of the three methods and will be covered next. Finally, some alternative supporting methods will also be explored in the remaining of this section.

3.3 In-situ Polymerization of Oleﬁns with Coordination Catalysts Supported on Clays

Vp/Cperylene × 104 (S–1)

0

Polymerization (mn) 10 20 30 40 50 60 70 80 90 100 110

10 Polymer growth from 9 the outside to the inside 8 7 Initial period particle fragmentation 6 of low activity 5 4 3 2 1 0

Particle expansion

Figure 3.14 Schematic polymer growth and particle expansion from experimental analysis

(reproduced from Ref. [60]).

Me

Me Me Zr Si(CH3)2 Cl Cl Me

Me

N

N

+ CF3 –

B

Pd

Ni Br

MBI

N

N

Me

Br

DMN

Me

CF3

NC-Me

DMPN

4

–

BAr4

Figure 3.15 Catalyst components: MBI, DMN, and DMPN/borate (reproduced from Ref. [1]).

3.3.7.1 Clay as a Polymerization Additive In this method, the organoclay, the catalyst precursor, and the activator (alkylaluminum compounds or alkylaluminoxanes) are added to the reactor and the polymerization is started by introduction of the oleﬁn monomer. This is the simplest in-situ polymerization method, and only few reports [1, 51, 70, 71] have been published investigating this method for production of polyoleﬁn–clay nanocomposites. In comparison to the other in-situ polymerization methods, this method seems to have the lowest polymerization activities, although it is often difﬁcult to compare catalysts activities for polymerizations done at very different conditions and following distinct procedures. This method was ﬁrst reported by Heinemann et al. [1] for the polymerization of ethylene with the catalysts depicted in Figure 3.15 in the presence of clays with different types of organic modiﬁcations. Figure 3.16 compares the ethylene uptake curves with homogeneous and clay-supported MBI catalyst. Lower activity and more stable polymerization rate were observed for the clay-supported system. As water present on the clay surface acts as a catalyst poison [71], the low polymerization activity observed for the clay-supported system was attributed to water traces remaining on the organoclay surface. For polymerizations in the presence of MAO-modiﬁed clays, especially if additional MAO is added to the reactor, a considerable fraction of the active sites may be extracted from the support and become active in the solution. In this case,

71

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization Mass flow rate of ethylene (g/h)

72

70 65 60 55 50 45 40 35 30 25 20 15 10 5 0

A

B

0

10

20 Time (min)

Figure 3.16 Comparison of ethylene uptake for homogeneous polymerization (a) and

polymerization in presence of clay (modiﬁed with dimethyl stearyl benzyl ammonium chloride-DMDS) and (b) with MBI (reproduced from Ref. [1]).

polymerization takes place in two phases: on the clay-supported sites and in the bulk reactor phase. Usually, polymerization with homogeneous catalysts results in higher catalyst deactivation (albeit with higher initial polymerization rates, as illustrated in Figure 3.15), but leads to poor particle morphology and unacceptable reactor fouling. 3.3.7.2 Clay as a Polymerization Catalyst Support In this method, the catalyst is supported onto the clay surface before polymerization. The clay surface is commonly treated with an alkylaluminum compound before being impregnated with a catalyst solution. Generally, washing steps are included after each treatment step to avoid excess catalyst leaching from the support during the polymerization. Additional alkylaluminum compounds may be used during the course of polymerization. This catalyst supporting approach has been reported by several research groups [24, 48, 51, 52, 54, 56, 71–77]. In order to show the advantages of catalyst supporting on clay over simple addition of clay to the reactor, Kuo et al. [71] performed a comparative study between two different in-situ polymerization methods. In Method 1, they contacted Et(Ind)2ZrCl2, MAO, and an organoclay in the reactor and started the polymerization by introducing ethylene. In Method 2, they ﬁrst treated the organoclay with MAO, impregnated the MAO-treated clay with the catalyst solution, and then used the product to polymerize ethylene. They concluded that in-situ polymerization with Method 2 led to higher catalyst activities and was less sensitive to clay loading. In addition, a ﬁner and more homogeneous dispersion of polymer-clay particles was obtained when Method 2 was used. They also reported that extending the MAO treatment time from 1.5 to 2.5 h, and the catalyst impregnation time from 0.5 to 2 h, had no appreciable effect on polymerization activity.

3.3 In-situ Polymerization of Oleﬁns with Coordination Catalysts Supported on Clays

Ray et al. [24] treated Cloisite 20A (montmorillonite modiﬁed with dimethyl– ditallow – containing approximately 65% C18, 30% C16, and 5% C14 – ammonium cation chains) with a MAO solution, after vacuo-drying at 100 °C. The resulting MAO-treated clay was subsequently used for ethylene polymerization in the presence of a late transition metal catalyst (2,6-bis[1-(2,6-diisopropylphenylimino)ethyl] pyridine iron(II) dichloride) and additional MAO in a glass reactor. They compared the result with homogeneous polymerization with the same catalyst in the presence of Cloisite 20A and observed that the supported catalyst was more efﬁciently exfoliated than when only a mixture of catalyst and clay was used. This comparison led them to conclude that at least some of the active centers resided within the clay galleries. Inductively coupled plasma (ICP) measurements showed that all MAO and catalyst remained in the solid catalyst after drying. Lee et al. [52] supported Cp2ZrCl2 on Na+ MMT and on an organically modiﬁed montmorillonite, Cloisite 25A (the organic modiﬁcation is shown in Figure 3.17). During the supporting procedure, the clays were treated with modiﬁed MAO (MMAO), impregnated with a solution of metallocene in toluene, and then washed with toluene after each supporting step. By measuring the Al and Zr loading after each step for Na+ MMT and Cloisite 25A, the authors concluded that considerably higher Al and Zr loadings were obtained for Cloisite 25A, as shown in Table 3.2. Lee et al. also showed that the modiﬁer content in Cloisite 25A decreased after MMAO treatment from 34 wt% to 4 wt%, and then to 2.5 wt% after catalyst impregnation. Despite the extraction of about 90 wt% of the ammonium cation during catalyst supporting, X-ray diffraction (XRD) results (Figure 3.18) did not

Figure 3.17 Chemical structure of the quaternary ammonium cation used in Cloisite 25A. HT:

Hydrogenated tallow (∼65% C18; ∼30% C16; ∼5% C14).

Table 3.2

Comparison of Al and Zr loading for Na+ MMT and cloisite 25A.

Sample

Clay (g/L)

Added MMAO (mol L−1 g−1)

Supported/ added MMAO (%)

Added Cp2ZrCl2 (×104 mol L−1 g−1)

Supported/ added Cp2ZrCl2 (%)

Na+/MMAO 25A/MMAO Na+/MMAO/Cp2ZrCl2 25A/MMAO/Cp2ZrCl2

1.67 1.67 1.58 1.58

0.29 0.29 – –

0.65 0.80 – –

– – 2.1 2.1

– – 0.3 14.3

Reproduced from Ref. [52]).

73

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization

18.6 Å

25A Relative Intensity

74

25A-M

25A-MZ

2

4

6 2θ (degree)

8

10

Figure 3.18 X-ray diffraction patterns of Cloisite 25A before supporting (25A), after treatment

with MMAO (25A-M), and after catalyst supporting (25A-MZ) (reproduced from Ref. [52]).

reveal any clay structural collapse due to their removal. In fact, the XRD analysis showed that the clay was exfoliated, indicating that the active sites were likely ﬁxed on the surface of Cloisite 25A. As shown in Figure 3.18, the d-spacing peak in Cloisite 25A (25A) was reduced after MMAO treatment (25A-M) and ﬁnally disappeared after catalyst supporting (25A-MZ). The authors considered this organic modiﬁcation removal as a result of reactions between oxygen atoms in MMAO and ammonium ions. They also compared three different in-situ polymerization methods performed under atmospheric pressure of ethylene: 1) 2) 3)

polymerization in the presence of MMAO-modiﬁed clay (MMAO/clay + catalyst); polymerization with catalyst supported on MMAO-modiﬁed clay (or MMAO/ clay/catalyst), with additional MMAO; polymerization with catalyst supported on MMAO-modiﬁed clay (or MMAO/ clay/catalyst), without additional MMAO.

As expected, when Cp2ZrCl2/Cloisite 25A was used, high polymerization activities were observed, irrespectively of the polymerization method. The monomer consumption rate with Cp2ZrCl2/Cloisite 25A was also higher when MMAO was added during the polymerization. In contrast, the polymerization rate with Cp2ZrCl2/Na+ MMT was very low when no MMAO was added to the reactor. 3.3.7.3 Clay as an Activator for Polymerization Catalysts In many publications on the in-situ polymerization of oleﬁns with clay-supported catalysts, Brønsted acid sites located on clay edges have been considered to be

3.3 In-situ Polymerization of Oleﬁns with Coordination Catalysts Supported on Clays

the preferred sites for catalyst supporting [6, 30, 55, 78, 79]. Acid treatment was also suggested as a way to increase the catalyst supporting capacity of clays [42, 44, 80]. During the mid-1990s, a research group at Mitsubishi discovered that certain clays could be calcined and used to activate metallocenes [44, 81]. This activation ability was attributed to the natural acidity of clays, which were used as cracking catalysts in the past. Referring to the work of Japanese research groups [82], McDaniel et al. [44] proposed that the high polymerization activity could not be attributed only by the low population of acidic sites on the clay edges, but it might derive from the clay ability to conduct ion exchange reactions between metallocenes and interlayer cations. Cations spaced between the clays sheets are isolated because they balance negative charges within the interior of the clay sheets. Thus, the cation is separated from its balancing anion by an insulating silica layer. It is conceivable that metallocenes are activated by this separation, perhaps by ion exchange, as illustrated in Figure 3.19. As proposed in Figure 3.19, ion-exchange reactions might result in the formation of “clay anions” and metallocene cations (soft ions), and sodium and chloride ions (hard ions). The ion exchange of metallocenium cations with clay surfaces is also discussed by Mariott et al. [83, 84]. In polymerizations using metallocenes, one of the main roles of MAO is to function as a Lewis acid, helping to ionize – or at least to polarize – the metallocene compound into cationic species of the type (L2MtCH3)+·(MAO·X)−. Therefore, in principle, other Lewis acids may be able to substitute MAO. This is the role supposed to be played by the clay surface during oleﬁn polymerization.

Cl

Na+ +

Na

Na+

Cl

Zr

Outer Silica Layers (neutral) Vacancies (negative charge)

Na+

Cl Na+

Zr +

+ NaCl

Figure 3.19 Possible mechanism for metallocene activation on clay surface proposed by

McDaniel (reproduced from Ref. [44]).

75

76

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization

A number of reports have been published on oleﬁn polymerization with alkyl aluminum-treated clays in the presence of an alkylaluminum compound [6, 20, 85–88]. It has been shown that oleﬁn polymerization catalysts can be activated by acidic clay surfaces when combined with alkylaluminum compounds such as trimethylaluminum or triisobuthylaluminum, generating catalysts that are very active for polymerization. Note that alkylaluminum compounds alone cannot properly activate metallocenes to high polymerization activities. In this approach, clay surface is supposed to be the sole location where polymerization takes place. 3.3.7.4 In-situ Production of Alkylaluminoxanes In this activation method, water molecules present on the surface of pristine clay react with alkylaluminum compounds to produce MAO oligomers on the clay surface [6, 50, 62, 81, 89]. The modiﬁed clay can then be used directly as a polymerization cocatalyst, or impregnated with a catalyst solution prior to polymerization. The high temperature thermal treatment to remove surface-bonded water molecules is not required in this case, albeit it has been used by some researchers [6, 50]. Weiss et al. [6] tested kaolinite and montmorillonite as support materials, TIBA and TMA as cocatalysts, and different metallocenes (Cp2ZrCl2, Cp2ZrHCl, Cp2TiCl2) as catalysts. They found out that montmorillonite produces supported catalysts with considerably higher polymerization activities than kaolinite. TIBA-treated MMT, in particular, had higher polymerization activity for propylene compared to TMA-treated MMT. Novokshonova et al. [89] used two clay alkylaluminum treatment procedures. In the ﬁrst, they added an alkylaluminum compound drop-wise to the clay until the evolution of volatiles stopped (when adding TMA, CH4 will be formed due to the reaction with surface water or hydroxyl groups). In this procedure, the Al : H2O ratio was lower than unity. In the second procedure, a Al : H2O ratio equal to 1 was used, and the alkylaluminum was added in a single step to the clay suspension. The second procedure resulted in higher polymerization activities and extra MAO was not required during the polymerization. The authors suggested that higher degrees of alkylaluminum hydrolysis were obtained in the ﬁrst procedure and that fewer alkyl groups remained available for alkylation reactions of the metallocene. On the other hand, higher Al : H2O ratios resulted in partial hydrolysis of the alkylaluminum molecules and, consequently, more alkylaluminum molecules were available for metallocene alkylation. In the same study, Novokshonova et al. [89] showed that the MAO formed on the clay surface had similar structure to that of commercial MAO. 3.3.7.5 Other Techniques Despite the fact that the presence of organic modiﬁcations on the clay surfaces enhance the dispersion of the clay nanolayers in nonpolar polymer matrices, it has been found that they may also present some disadvantages on the ﬁnal physical and mechanical properties of the polymer [23]. For example, organic modiﬁca-

3.3 In-situ Polymerization of Oleﬁns with Coordination Catalysts Supported on Clays

Ni N

clay N

X

1: X = O 2: X = CH2

N

Ni +

N

X Al

–

Figure 3.20 Proposed mechanism of for nickel catalysts activation on the surface of clay

(reproduced from Ref. [23]).

tion molecules tend to degrade rapidly under the high temperatures required for polymer extrusion, leading to clay agglomeration and poor nanocomposites mechanical properties [90–93]. Scott et al. [23] proposed a technique to overcome this problem by avoiding the use of clay organic modiﬁers. By comparing different clay treatments for catalyst supporting and in-situ oleﬁn polymerization, they showed that montmorillonite treatment with mineral acid decreased its original stacking order and increased its Lewis acidity. Upon addition of a Ni catalyst ([N-(2,6-diisopropylphenyl)2-(2,6-diisopropylphenylimino) propanamidato-κ2-N,N]Ni(η3-CH2Ph)), polymerization activity was observed even in the absence of TMA as a scavenger. They proposed that the nickel complex interacted via a Lewis basic site on the ligand backbone with a Lewis acid site on the clay surface was particularly effective at promoting polymerization solely on the clay surface, leading to effective clay dispersion in the polymer matrix, while not needing any cocatalyst or scavenger (Figure 3.20). Unfortunately, this procedure was reported to be inadequate for metallocenes. When a Cp2ZrMe2 solution was added to the acid-treated clay, a very low ethylene uptake was observed and the majority of the gel-like polyethylene seemed to be produced by soluble catalyst molecules that were not supported on the clay surface. The authors speculated that the low activity of the clay-supported metallocene was due to severe decomposition of the catalyst in contact with strong Brønsted acidic surface of the clay. Scott et al. also showed that the clay dispersion in the polyoleﬁn matrix is stable during annealing at 170 °C for 30 min and related this behavior to the high molecular weight and high viscosity at the test temperature (170 °C). The retarded structural collapse due to the high molecular weight of the polymer matrix is also mentioned elsewhere [46]. Considering hydroxyl groups as sites for catalyst supporting on the clay surface, Tang et al. [94] and Wei et al. [95] proposed an indirect supporting method in which a common support, such as SiO2 or MgCl2, is deposited onto the clay surface from a solution or by the sol–gel method to increase the hydroxyl population on the clay surface, and the catalyst is then ﬁxed on the top of this layer. Tang et al. [94] developed a method in which silica or titanium oxide nanoparticles were ﬁxed on the surface of organically modiﬁed clays. This modiﬁed clay was subsequently treated

77

78

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization

Catalyst

TEOS

C16-MT solvated by dodecylamine : C16-MT

MT-Si

PE

cat cat Ethylene cat cat cat MT-Si-cat

: Silica nanoparticle

: H2O

PE/clay-silica nanocomposite : Dodecylamine

Figure 3.21 Proposed mechanism for formation of MT-Si and the PE/clay–silica

nanocomposites (reproduced from Ref. [95]).

with MAO and loaded with a metallocene catalyst. The procedure reported by Wei et al. [95] and Tang et al. [94, 95] was similar: they supported silica nanoparticles on the clay layers using tetraethoxysilane (TEOS), and supported metallocene catalyst after MAO treatment (Figure 3.21). The authors reported the production of polymer powder with granular morphology and of higher bulk density (0.2 g cm−3, instead of 0.07 g cm−3 for polyethylene made with a soluble catalyst). Although a higher amount of catalyst was supported on the silica-treated clay, no improvement in activity was observed, indicating that either the activity of the sites decreases or that part of the sites is deactivated during supporting. In all heterogeneous cases (with or without silica intermediate) the polymerization activities were lower than those for homogeneous polymerization. 3.3.8 Factors Determining the Success of In-situ Polymerization

Many parameters may inﬂuence the behavior of in-situ polymerizations: clay structure, type of organic modiﬁcation (if any), thermal and alkylaluminum treatment conditions, type of catalyst and cocatalyst, catalyst supporting methods, and polymerization conditions. This section reviews some of the most important parameters necessary to achieve a successful in-situ polymerization. 3.3.8.1 Clay Type A vast variety of clay materials are available that, despite their common structural unit, have different surface charges, level and type of isomorphous substitutions, ion exchange capacities, hydroxyl group densities, structural unit arrangements in the crystalline layer (2 : 1 or 1 : 1), and layer dimensions. Some of these parameters can determine how adequate a clay type is for the in-situ production of polyoleﬁn– clay nanocomposites. As an example, kaolinites are composed of crystalline nanolayers with higher aspect ratios and hydroxyl group densities than montmo-

3.3 In-situ Polymerization of Oleﬁns with Coordination Catalysts Supported on Clays

rillonites, which may initially look as attractive properties for catalyst supporting; however, due to their lower compatibility with organic solvents and limited ion exchange ability, they cannot easily host polymerization active sites between their layers and, therefore, are not the most suitable candidates as supports for in-situ polymerizations, although in-situ polymerizations with kaolinites have been reported [46]. On the other hand, due to their structural characteristics, montmorillonites are much better choices, and consequently most frequently used for in-situ polymerizations. Another main source of difference in clay performance is their surface acidities. Jeong et al. [50] compared two clays with different acidities from two different suppliers. They found out that clay acidity played a signiﬁcant role in polymerization activity; for the acidic clay sample, irrespective of its water content, polymerization activity was always observed, while no activity was found when the catalyst was supported on the basic clay sample. It was reported [20, 96] that the counter-ion type on mica inﬂuences the polymerization behavior of the supported catalyst. Even though micas do not have exactly the same structure of montmorillonites, they are considered as clay materials and some of their behavior may be translated to montmorillonites. Hiyama et al. [20] compared ethylene polymerization with an iron catalyst supported on micas with different counterions. They observed that when the polymerization catalyst was supported on Mn+ mica (where Mn+ = Mg2+, Zn2+, and Fe3+), polymerization activities were approximately 10-fold higher than those obtained when Na+ mica was used as a support. Kurokawa et al. [96] supported Cp2ZrCl2 on ﬂuorotetrasilicic micas with different counter ions. During ethylene polymerization, it was observed that the polymerization activity of the catalyst supported on Na+ mica (Cp2ZrCl2/ Na+ mica) was very low, but increased dramatically when the zirconocene was supported on Mn+ mica. By comparing the polymerization activities of catalysts supported on mica with different metal ions, they concluded that the ability for activation of the metallocene complex is strongly dependent on the nature of the exchanged cations. 3.3.8.2 Swellability Organically modiﬁed clays can swell when exposed to organic solvents [97–100], depending on the interaction between the clay surface and solvent, in different scales. The level of dispersion can be correlated to the solubility parameters of solvent and organic modiﬁcation of the organoclay [100, 101]. In-situ polymerization methods can also be classiﬁed according to their exfoliation mechanism: (a) in-situ intercalative polymerization, in which polymerization inside the galleries of the already modiﬁed clay (by organic modiﬁcation, alkylaluminum treatment, or catalyst supporting) results in the exfoliation of clay aggregates, and (b) ﬁlling polymerization, in which the organoclay is introduced in the polymerization system where it may be intercalated by solvent molecules during polymerization and polymer chains are formed on the surface of the exfoliated clay. In the latter case, the dispersion quality of the organoclay in the solvent is

79

80

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization

the main factor determining the ﬁnal clay platelet dispersion in the nanocomposites [97, 102], so that if the modiﬁed clay is exfoliated before polymerization, the resulting polymer–clay nanocomposite is surely exfoliated [102]. Swellability can be manipulated by the type of organic modiﬁcation and the type of solvent used during polymerization. Besides their role of enhancing solvent-clay surface compatibility already discussed above, some organic modiﬁer molecules may provide functional groups that participate in catalyst supporting reactions [62, 103] or even become copolymerized with the polymer during in-situ polymerization [70]. Commercially available organoclays are mostly modiﬁed with quaternary ammonium cations that are efﬁcient to produce nanocomposites by melt mixing. Quaternary ammonium cations seem to act merely as surfactants and/or expand the clay interlayer spacing. In contrast, other types of ammonium cations may be involved in chemical reactions during catalyst supporting and affect the in-situ polymerization efﬁciency. Maneshi et al. [62] showed that montmorillonites modiﬁed with tertiary ammonium cations had high supporting efﬁciencies, high polymerization activities, and better exfoliation than those modiﬁed with quaternary ammonium cations. They concluded that the latter failed because TMA could not react with quaternary ammonium cations. On the other hand, Lee et al. [52] and Leone et al. [104] obtained active metallocene catalysts supported on montmorillonites modiﬁed with quaternary ammonium cations when MMAO was used for the surface treatment. Besides affecting the swellability of the organoclay during polymerization, solvent type has been reported to affect the supporting efﬁciency of metallocenes on clays [46, 105]. For Cp2ZrCl2 supported on Cloisite 93A (MMT modiﬁed with a tertiary ammonium cation), Maneshi [105] showed that changing the solvent from toluene to hexane during supporting signiﬁcantly increased polymerization activity, particle break-up behavior, and led to the production of higher molecular weight polymer. They attributed this behavior to the weaker interaction between hexane and the clay surface modiﬁcation. Dubois et al. [46] also compared polymerization activities when montmorillonite was treated with MAO in heptane or toluene, and reported treatment in heptane was more efﬁcient. 3.3.8.3 Effect of Clay Surface Treatment Clay surface treatment can inﬂuence its reaction with the catalyst and in turn determine the stability of polymerization active site on the clay surface. As mentioned earlier, there are two main types of clay surface treatments: thermal and chemical. Thermal treatment at high temperatures may result in loss of swellability [96] and reduced efﬁciency for in-situ polymerization applications. The type of alkylaluminum and the procedure of addition have a great inﬂuence on the success of in-situ polymerization. Lee et al. [52] used MMAO to treat Cloisite 25A for catalyst supporting, obtaining a highly active catalyst system, while Maneshi et al. [62] showed that using TMA for surface treatment of Cloisite 25A before catalyst supporting did not result in any polymerization activity. Lee et al. [52] attributed their positive results to reactions between quaternary ammonium

3.3 In-situ Polymerization of Oleﬁns with Coordination Catalysts Supported on Clays

cations and oxygen atoms on MMAO. This may explain different reactivities of MMAO and TMA toward quaternary ammonium cations. Using the same technique applied by Maneshi et al. [62], Ushakova et al. [106] showed that treating Na+ MMT with TIBA led to a catalyst that had a higher propylene polymerization activity than when the same clay was treated with TMA. 3.3.8.4 Catalyst : Clay Ratio Theoretically, the maximum catalyst loading capacity onto clay should be comparable to their CEC, which for most montmorillonites is about 1−1.2 × 10−3 mol g−1 clay. Catalyst loadings that are substantially below the CEC may result in the nonuniform distribution of active sites on the clay layers. Due to mass transfer limitations, one might imagine that most of the sites would be preferentially supported on more external surfaces, around the layer edges or closer to the clay gallery openings. A homogeneous distribution of active sites over the entire clay surface is necessary to guarantee uniform exfoliation and even distribution of clay nanolayers in the polymer matrix after in-situ polymerization. Maneshi et al. [62] compared in-situ polymerization using pristine montmorillonite and an organically modiﬁed clay (Cloisite 93A), and observed that, despite reaching near 100% supporting efﬁciencies for both clays, the organically modiﬁed clay produced clay–polyethylene nanocomposites with more uniform clay nanolayers distribution in the polymer matrix, supposedly because a more uniform catalyst distribution was obtained with the organoclay. In a subsequent publication, Maneshi et al. [69] compared two different levels of catalyst loading from two different supporting series on Cloisite 93A and found out that the clay with the highest catalyst loading generated a supported catalyst with higher polymerization activity and more uniform clay distribution in the nanocomposites. Maneshi et al. also compared the effect of precontacting the catalyst with the organoclay before the polymerization with simply adding the homogeneous catalyst and the TMA-treated organoclay directly to the reactor. They reported signiﬁcantly higher polymerization rates for the latter case and concluded that a signiﬁcant fraction of active sites was deactivated during the precontact stage, likely by impurities on the clay surface. A similar behavior was reported by Hiyama et al. [20]. They compared the polymerization behavior of bisimino pyridine iron(II) supported on Mg2+ MMT, saponite, and Mg2+ mica. They observed that catalysts supported on saponite and Mg2+ MMT had higher polymerization activities than catalysts supported on Mg2+ mica; they also correlated this observation to the higher population of hydroxyl groups, which is known to enhance the supporting efﬁciency, in the former supports. Besides the higher polymerization activities of the catalysts supported on Mg2+ MMT and saponite, the rate of deactivation during catalyst storage were considerably higher for these systems. The activity of the saponite-supported catalyst decreased to one-third of the fresh catalyst during storage for 48 h, and the activity of the montmorillonite-supported catalyst decreased to half its original value after one month. In contrast, the mica supported catalyst maintained its original activity

81

82

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization

even after 3 months of storage. Considering the absence of OH groups on the ideal mica structure and a much higher population of OH groups on Mg2+ MMT and saponite, the deactivation of active sites was attributed to the presence of hydroxyl groups on the clays. For metallocenes, reaction with surface OH groups form inactive μ-oxo species [107]. Hiyama et al. [20] also related the rate of catalyst deactivation to the available surface area of the clay support. They reported the speciﬁc areas of Mg2+ saponite (207 m2 g−1) was much higher than those of Mg2+ MMT (17 m2 g−1), and Mg2+ mica (12 m2 g−1). 3.3.8.5 Effect of Polymerization Conditions Aluminoxane or alkylaluminum loading on the clay surface is one the important factors that inﬂuences the polymerization activity with clay-supported metallocenes. Jeong et al. [50] synthesized in-situ MAO on the clay surface before catalyst supporting and showed that without addition of TMA, the polymerization activity was very small and the catalyst deactivated rapidly. They showed that addition of free TMA up to Al/Zr ratio of 500 increased the polymerization activity. They suggested that the amount of MAO on the clay surface is enough for catalyst activation, but not enough to maintain high polymerization activities. According to the authors, that during polymerization and particle break-up the contact between zirconocene and newly exposed hydroxyl groups that had not reacted with TMA during supporting deactivated the polymerization catalyst. Similar observations were reported by Mansehi et al. [69] It has been reported that polymerization temperature has a considerable effect on particle break-up and, therefore, on the ﬁnal particle morphology [108, 109]. Maneshi et al. [69] observed that higher polymerization temperatures, up to a certain upper limit, enhanced clay exfoliation. Above this upper limit (which varies depending on the polymer and solvent type) active site and/or polymer chain start being extracted from the clay surface, resulting in poor exfoliation, inadequate powder morphology, and severe reactor fouling. Polymerization time is also an important factor regulating clay exfoliation and dispersion in the polymer. Particle break-up and exfoliation during polymerization can be studied by performing polymerizations at different reactor residence times and comparing the microstructures and morphologies of the resulting nanocomposites. For example, Ren et al. [70] performed in-situ polymerization of ethylene at increasing times (15, 30, 45, and 60 min), producing nanocomposites with the decreasing clay contents (3.7, 3.1, 2.2, and 1.6 wt%, respectively). By comparing their WAXD proﬁles, they observed no shift for the (0 0 1) diffraction peak with polymerization time (Figure 3.22), which implies that the metallocene catalyst was most likely not intercalated in the clay layers, but rather only located inside the clay particle aggregates. In some investigations, it has been shown that the clay content in nanocomposites can be tuned by setting the polymerization time [30, 52]. However, it should be remembered that, depending on the in-situ polymerization system, the particle morphology may undergo noticeable changes during polymerization [52, 70].

3.3 In-situ Polymerization of Oleﬁns with Coordination Catalysts Supported on Clays 2000 15 min 1800

30 min 45 min

1600

60 min 1400

Intensity (a.u.)

1200 1000 800 600 400 200 0 5

6

7

2q (degree) Figure 3.22 WAXD proﬁles of PE/P-MMTs samples prepared at different polymerization time.

3.3.9 Clay Effect on the Polymerization Behavior and Polymer Molecular Structure

As a catalyst support material, clays possess unique electronic and structural properties. They may be considered as solid acids, with electrostatic properties that can affect catalytic behavior and, possibly, the microstructure of polymers made with clay-supported catalysts. The layered structure of these aluminosilicates has also been reported to affect polymerization kinetics. Studies in this area have focused on three main effects: polymerization activity and stability, polymer molecular weight, and polymer tacticity. It has been observed by most researchers that polymerization activity decreases substantially upon catalyst supporting on clay. For butadiene polymerization with cobalt catalysts (CoCl2(PMePh2)2-MAO and CoCl2(PiPrPh2)2-MAO), Leone et al. [110] showed that clay supporting reduced the polymerization activity. Junges et al. [77] reported that ethylene polymerization activity decreased when a titanium catalyst was supported on the surface of organically modiﬁed clay (cloisite 30B).

83

84

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization

Marques et al. [76] also showed that when SiMe2(2-Me-Ind)2ZrCl2 was supported on Cloisite 15A, the propylene activity was remarkably decreased. This behavior is in agreement with what has been reported for metallocene catalysts supported on other carriers, such as SiO2 and MgCl2. The effect of the support on catalyst tacticity and stereoselectivity has also been reported in a few studies. Ushakova et al. [106] supported different zirconocene catalysts on monotmorillonite and compared the streoregularities of polypropylenes made with those supported catalysts with their homogeneous counterparts. They observed that heterogenization affected catalyst stereoselectivity, depending on the type of polymerization catalyst. Interestingly, rac-Me2Si(Ind)2ZrCl2, having C2 symmetry, was more stereospeciﬁc when supported on clay, while the Cs-symmetric rac-[1-(9-η5-Flu)2-(5,6-Cp-2-Me-1-η5-Ind)Et]ZrCl2 became less stereospeciﬁc. On the other hand, during butadiene polymerization with cobalt catalysts (CoCl2(PMePh2) 2-MAO and CoCl2(PiPrPh2)2-MAO), Leone et al. [110] showed that supporting on clay had no effect on the polymer microstructure, either on 1,2-insertion frequency, or on polymer tacticity. Metallocenes supported on SiO2, usually (but not always) make polyoleﬁns with molecular weights that are higher than those obtained with homogeneous catalyst [32, 111]. This has been attributed to blocking one of active sites sides by the support, hindering β-elimination [111]. A similar behavior is observed for claysupported metallocenes [74, 106, 110], but should not be considered a general rule. For instance, the molecular weight of ultrahigh molecular weight polyethylene made with a clay-supported catalyst is lower than that made with the equivalent homogeneous catalyst [77]. 3.3.10 Future Approaches

The combination of a layered structure, high surface area, electrostatic properties, surface chemistry, swellability, and high-cation exchange capacity, have given clays their unique capability for the production of polyoleﬁn–clay nanocomposites by in-situ polymerization methods. Considering the high number of parameters that inﬂuence in-situ polymerization with clay-supported catalysts, there is a great deal of ﬂexibility for tuning the properties of nanocomposites made with this powerful technique. Regarding synthesis of polyoleﬁn nanocomposites with wider applications, copolymerization of ethylene with other oleﬁnic monomers, including higher αoleﬁns and polar comonomers (with late transition metal catalysts) still needs to be investigated in more detail. For instance, exfoliated polyoleﬁn–clay nanocomposites using chain end functionalized polyoleﬁns can potentially be used as the polymeric surfactants [112]. In order to be used in industrial scale, some problems such as low bulk density and ﬁnal product cost should further be investigated. Generally, the fully fragmented clay particles that are generated during polymerization appear as highly

References

porous macroparticles composed of leaf-like smaller particles. This morphology is responsible for low bulk density that needs to be addressed before these systems can be used in conventional industrial processes.

References 1 Heinemann, J., Reichert, P., Thomann,

2

3 4

5

6

7

8

9

10

11

12

13

R., and Mülhaupt, R. (1999) Macromol. Rapid Commun., 20, 423. Bergaya, F., Theng, B.K.G., and Lagaly, G. (2006) Handbook of Clay Science, Elsevier, Amsterdam, Holland. Meunier, A. (2005) Clays, Springer, Berlin, Germany. Moore, D.M., and Reynolds, R.C. (1997) X-Ray Diffraction and the Identiﬁcation and Analysis of Clay Minerals, Oxford University Press, Oxford, UK. Murray, H.H. (2007) Occurrences, Processing and Applications of Kaolins, Bentonites, Palygorskitesepiolite, and Common Clays, Elsevier, Amsterdam, Holland. Weiss, K., Wirth-Pfeifer, C., Hofmann, M., Botzenhardt, S., Lang, H., Brüning, K., and Meichel, E. (2002) J. Mol. Catal. A, 182–183, 143. Brindley, G.W., and Brown, G. (1980) Crystal Structures of Clay Minerals and Their X-Ray Identiﬁcation, Mineralogical Society, London, UK. Lagaly, G. (2006). Developments in Clay Science (eds F. Bergaya, B.K.G. Theng, and G. Lagaly), Elsevier, Amsterdam, Holland. Lagaly, G., Ogawa, M., and Dékány, I. (2006). Developments in Clay Science (eds F. Bergaya, B.K.G. Theng, and G. Lagaly), Elsevier, Amsterdam, Holland. Jolly, W.L. (1984) Principles of Inorganic Chemistry, McGraw-Hill, New York, USA. Ainsworth, C.C., Zachara, J.M., and Schmidt, R.L. (1987) Clays Clay Miner., 35, 121. Slade, P.G., Dean, C., Schultz, P.K., and Self, P.G. (1987) Clays Clay Miner., 35, 177. White, J.L. and Choi, D.D. (2005) Polyoleﬁns – Processing, Structure

14 15

16

17

18

19

20

21

22

23

24

25

Development, and Properties, Hanser Publishers, Munich, Germany. Bochmann, M. (2004) J. Organometallic Chem., 689, 3982. Fink, G., Mulhaupt, R., and Brintzinger, H.H. (1995) Ziegler Catalysts: Recent Scientiﬁc Innovations and Technological Improvements, Springer-Verlag, Berlin, Germany. Reddy, S.S., Shashidhar, G., and Sivaram, S. (1993) Macromolecules, 26, 1180. Mason, M.R., Smith, J.M., Bott, S.G., and Barron, A.R. (1993) J. Am. Chem. Soc., 115, 4971. Ystenes, M., Eilertsen, J.L., Liu, J., Ott, M., Rytter, E., and Støvneng, J.A. (2000) J. Polym. Sci., Part A: Polym. Chem., 38, 3106. Kaminsky, W. (2000) MetalloceneCatalysed Polymerisation, Rapra Technology Ltd., Shropshire, UK. Hiyama, Y., Kawada, Y., Ishihama, Y., Sakuragi, T., Ohshima, M.-A., Kurokawa, H., and Miura, H. (2009) Bull. Chem. Soc. Jpn., 82, 624. Gong, D., Dong, W., Hu, Y., Bi, J., Zhang, X., and Jiang, L. (2009) Polymer, 50, 5980. Gibson, V.C. and Solan, G.A. (2010) Iron-Based and Cobalt-Based Oleﬁn Polymerisation Catalysts, Springer, Berlin, Germany. Scott, S.L., Peoples, B.C., Yung, C., Rojas, R.S., Khanna, V., Sano, H., Suzuki, T., and Shimizu, F. (2008) Chem. Commun., 4186. Ray, S., Galgali, G., Lele, A., and Sivaram, S. (2005) J. Polym. Sci., Part A: Polym. Chem., 43, 304–318. Liu, X., He, A., Du, K., and Han, C.C. (2009) Isothermal crystallization behavior of exfoliated-PP/IMMT nanocomposites via in situ

85

86

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization

26

27

28

29

30

31 32 33 34

35

36

37 38 39

40 41 42

43

44

polymerization. J. Polym. Sci., Part B: Polym. Phys., 47, 2215. Du, K., He, A.H., Liu, X., and Han, C.C. (2007) Macromol. Rapid Commun., 28, 2294. Yamamoto, K., Ishihama, Y., Isobe, E., and Sugano, T. (2009) J. Polym. Sci., Part A: Polym. Chem., 47, 2272. Fujii, K., Ishihama, Y., Sakuragi, T., Ohshima, M.-A., Kurokawa, H., and Miura, H. (2008) Catal. Commun., 10, 183. He, F.-A., Zhang, L.-M., Jiang, H.-L., Chen, L.-S., Wu, Q., and Wang, H.-H. (2007) Compos. Sci. Tech., 67, 1727. Shin, S.-Y.A., Simon, L.C., Soares, J.B.P., and Scholz, G. (2003) Polymer, 44, 5317. Kaminsky, W. (2000) Catal. Today, 62, 23. Kaminsky, W. and Laban, A. (2001) Appl. Catal. A, 222, 47. Chien, J. (1999) Top. Catal., 7, 23. Sinn, H.W.H., Kaminsky, W.O., Vollmer, H.-J.C., and Woldt, R.O.H.H. (1983) BASF Aktiengesellschaft (DE), United States. Kaminsky, W.P., Hahnsen, H., Kulper, K., and Woldt, R. (1985) Hoechst Aktiengesellschaft (DE), United States. Ribeiro, M.R., Defﬁeux, A., and Portela, M. (1997) Ind. Eng. Chem. Res., 36, 1224. Estenoz, D.A. and Chiovetta, M.G. (2001) J. Appl. Polym. Sci., 81, 285. Kristen, M. (1999) Top. Catal., 7, 89. Fink, G., Steinmetz, B., Zechlin, J., Przybyla, C., and Tesche, B. (2000) Chem. Rev., 100, 1377. Hlatky, G.G. (2000) Chem. Rev., 100, 1347. Kaminsky, W. and Winkelbach, H. (1999) Top. Catal., 7, 61. Ruitz-Hitzky, E. and Meerbeek, A.V. (2006) Developments in Clay Science (eds F. Bergaya, B.K.G. Theng, and G. Lagaly), Elsevier, Amsterdam, Holland. Heller-Kallai, L. (2006) Developments in Clay Science (eds F. Bergaya, B.K.G. Theng, and G. Lagaly), Elsevier, Amsterdam, Holland. McDaniel, M.P., Jensen, M.D., Jayaratne, K., Collins, K.S., Benham, E.A., McDaniel, N.D., Das, P.K.,

45 46 47

48

49

50

51

52

53

54

55 56 57 58

59 60

Martin, J.L., Yang, Q., Thorn, M.G., and Masino, A.P. (2008) Tailor-Made Polymers Via Immobilization of AlphaOleﬁn Polymerization Catalysts (eds J.R. Severn and J.C. Chadwick), Wiley-VCH Verlag GmbH, Weinheim, Germany. Hedley, C.B., Yuan, G., and Theng, B.K.G. (2007) Appl. Clay Sci., 35, 180. Dubois, P., Alexandre, M., and Jérôme, R. (2003) Macromol. Symp., 194, 13. Novokshonova, L., Ushakova, T., Krasheninnikov, V., Kovaleva, N., Ladygina, T., Lysova, M., Leipunskii, I., Zhigach, A., and Kuskov, M. (2006) Kinet. Catal., 47, 251. Shin, S.-Y.A., Simon, L.C., Soares, J.B.P., Scholz, G., and McKenna, T.F.L. (2009) Macromol. Symp., 285, 64. Rong, J., Li, H., Jing, Z., Hong, X., and Sheng, M. (2001) J. Appl. Polym. Sci., 82, 1829. Jeong, D.W., Hong, D.S., Cho, H.Y., and Woo, S.I. (2003) J. Mol. Catal. A, 206, 205. Liu, C., Tang, T., Zhao, Z., and Huang, B. (2002) J. Polym. Sci., Part A: Polym. Chem., 40, 1892. Lee, D.-H., Kim, H.-S., Yoon, K.-B., Min, K.E., Seo, K.H., and Noh, S.K. (2005) Sci. Tech. Adv. Mater., 6, 457. Mitra, A., and Atwood, D.A. (2007) Comprehensive Organometallic Chemistry III (eds D. Michael, P. Mingos, and R.H. Crabtree), Elsevier, Amsterdam, Holland. Shin, S.-Y.A., Simon, L.C., Soares, J.B.P., Scholz, G., and McKenna, T.F. (2009) Macromol. React. Eng., 3, 543. Liu, C., Tang, T., and Huang, B. (2004) J. Catal., 221, 162. Reddy, C.S., Ratna, D., and Das, C. (2008) Polym. Int., 57, 282. McKenna, T.F. and Soares, J.B.P. (2001) Chem. Eng. Sci., 56, 3931. Hutchinson, R.A., Chen, C.M., and Ray, W.H. (1992) J. Appl. Polym. Sci., 44, 1389. McKenna, T. and Mattioli, V. (2001) Macromol. Symp., 173, 149. Alexiadis, A., Andes, C., Ferrari, D., Korber, F., Hauschild, K., Bochmann, M., and Fink, G. (2004) Macromol. Mater. Eng., 289, 457.

References 61 Maneshi, A., Soares, J.B.P., and

62

63 64

65 66 67

68

69

70

71

72 73

74

75

76 77

78

Simon, L.C. (2006) Macromol. Symp., 243, 277. Maneshi, A., Soares, J.B.P., and Simon, L.C. (2010) Macromol. Chem. Phys., 212, 1521–3935. Pavlidou, S. and Papaspyrides, C. (2008) Prog. Polym. Sci., 33, 1119. Ke, Y.C., and Stroeve, P. (2005) Polymer-Layered Silicate and Silica Nanocomposites, Elsevier Science, Amsterdam, Holland. Alexandre, M., and Dubois, P. (2000) Mater. Sci. Eng. R, 28, 1. Sinha Ray, S. and Okamoto, M. (2003) Prog. Polym. Sci., 28, 1539. Ciardelli, F., Coiai, S., Passaglia, E., Pucci, A., and Ruggeri, G. (2008) Polym. Int., 57, 805. Bhattacharya, S.N., Kamal, M.R., and Gupta, R.K. (2008) Polymeric Nanocomposites: Theory and Practice, Carl Hanser Publishers, Munich, Germany. Maneshi, A., Soares, J.B.P., and Simon, L. (2011) Macromol. Chem. Phys. doi: 10.1002/macp.201100189 Ren, C., Du, X., Ma, L., Wang, Y., and Tang, T. (2010) J. Appl. Polym. Sci., 117, 1646. Kuo, S.-W., Huang, W.-J., Huang, S.-B., Kao, H.-C., and Chang, F.-C. (2003) Polymer, 44, 7709. Huang, Y., Yang, K., and Dong, J.-Y. (2007) Polymer, 48, 4005. Choi, Y., Shin, S.Y.A., and Soares, J.B.P. (2010) Macromol. Chem. Phys., 211, 1026. Li, W., Jiang, B., Buda, A., Wang, J., Blümich, B., Yang, Y., and Zheng, J. (2010) J. Polym. Sci., Part B: Polym. Phys., 48, 1363. Guo, C.-Y., Ke, Y., Liu, Y., Mi, X., Zhang, M., and Hu, Y. (2009) Polym. Int., 58, 1319. de Fátima Vieira Marques, M. and de Oliveira, M. (2010) Polym. Bull., 64, 221. Junges, F., Beauvalet, M.S., Leal, B.C., Casagrande, A.C.A., Mota, F.F., Mauler, R.S., and Casagrande Jr, O.L. (2009) J. Braz. Chem. Soc., 20, 472. Alexandre, M., Dubois, P., Sun, T., Garces, J.M., and Jérôme, R. (2002) Polymer, 43, 2123.

79 Yang, F., Zhang, X., Zhao, H., Chen, B.,

80

81

82

83

84 85

86 87

88

89

90

91

92 93

94

Huang, B., and Feng, Z. (2003) J. Appl. Polym. Sci., 89, 3680. Lagaly, G. and Fahn, R. (1983) Ullmanns Encyklopädie Der Technischen Chemie, Wiley-VCH, Weinheim, Germany. Suga, Y.M., Maruyama, Y., Isobe, E., Suzuki, T., and Shimizu, F. (1994) Mitsubishi Kasei Corporation (Tokyo, JP), United States. Takashashi, T., Nakano, H., Hideshi, U., Tayano, T., and Sugan, T. (2002) Polym. Preprints, 43, 1259. Mariott, W.R., Escudé, N.C., and Chen, E.Y.-X. (2007) J. Polym. Sci., Part A: Polym. Chem., 45, 2581. Mariott, W.R. and Chen, E.Y.-X. (2003) J. Am. Chem. Soc., 125, 15726. Nakano, H., Takahashi, T., Uchino, H., Tayano, T., and Sugano, T. (2006). Studies in Surface Science and Catalysis (eds K.N. Takeshi Shiono and T. Minoru), Elsevier, Amsterdam, Holland Sun, T. and Garcés, J. (2003) Catal. Commun., 4, 97. Suga, Y.Y., Uehara, Y., Maruyama, Y., Isobe, E., Ishihama, Y., and Sagae, T. (1999) Mitsubishi Chemical Corporation (Tokyo, JP), United States. Scott, S.G., Peoples, B., Rojas, R., Tanna, A., and Shimizu, F. (2008) United States. Novokshonova, L., Kovaleva, N., Ushakova, T., Meshkova, I., Krasheninnikov, V., Ladygina, T., Leipunskii, I., Zhigach, A., and Kuskov, M. (2005) Kinet. Catal., 46, 853. Xie, W., Gao, Z., Pan, W.-P., Hunter, D., Singh, A., and Vaia, R. (2001) Chem. Mater., 13, 2979. Frankowski, D.J., Capracotta, M.D., Martin, J.D., Khan, S.A., and Spontak, R.J. (2007) Chem. Mater., 19, 2757. Mittal, V. (2007) Eur. Polym. J., 43, 3727. Araujo, E., Barbosa, R., Oliveira, A., Morais, C., de Melo, T., and Souza, A. (2007) J. Therm. Anal. Calorimetry, 87, 811. Tang, T.C., Wei, L., and Huang, B. (2003) Institute Of Applied Chemistry Chinese Academy Of Science (Jilian, CN), United States.

87

88

3 Polyoleﬁn–Clay Nanocomposites by In-situ Polymerization 95 Wei, L., Tang, T., and Huang, B. (2004)

96

97

98 99 100

101 102 103

104

J. Polym. Sci., Part A: Polym. Chem., 42, 941. Kurokawa, H., Morita, S., Matsuda, M., Suzuki, H., Ohshima, M.-A., and Miura, H. (2009) Appl. Catal. A, 360, 192. Burgentzlé, D., Duchet, J., Gérard, J.F., Jupin, A., and Fillon, B. (2004) J. Colloid Interface Sci., 278, 26. Moraru, V.N. (2001) Appl. Clay Sci., 19, 11. Lagaly, G., and Malberg, R. (1990) Colloids Surf., 49, 11. Connolly, J., van Duijneveldt, J.S., Klein, S., Pizzey, C., and Richardson, R.M. (2006) Langmuir, 22, 6531. Ho, D.L. and Glinka, C.J. (2003) Chem. Mater., 15, 1309. Owusu-Adom, K. and Guymon, C.A. (2008) Macromolecules, 42, 180. Ren, C., Du, X., Ma, L., Wang, Y., Zheng, J., and Tang, T. (2010) Polymer, 51, 3416. Leone, G., Bertini, F., Canetti, M., Boggioni, L., Conzatti, L., and Tritto, I. (2009) J. Polym. Sci., Part A: Polym. Chem., 47, 548.

105 Maneshi, A. (2010) In-situ ethylene

106

107 108

109

110

111

112

polymerization with organoclaysupported metallocenes for the preparation of polyethylene-clay nanocomposites. PhD dissertation. University of Waterloo. Ushakova, T., Lysova, M., Kudinova, O., Ladygina, T., Kiseleva, E., Novokshonova, L., Lyubimtsev, A., and Dybov, A. (2007) Kinet. Catal., 48, 669. Nicholas, C.P., Ahn, H., and Marks, T. (2003) J. Am. Chem. Soc., 125, 4325. Weickert, G., Meier, G.B., Pater, J.T.M., and Westerterp, K.R. (1999) Chem. Eng. Sci., 54, 3291. Fink, G., Tesche, B., Korber, F., and Knoke, S. (2001) Macromol. Symp., 173, 77. Leone, G., Boglia, A., Bertini, F., Canetti, M., and Ricci, G. (2009) Macromol. Chem. Phys., 210, 279. Kaminsky, W. and Renner, F. (1993) Die Makromolekulare Chemie, Rapid Commun., 14, 239. Chung, T.-C.S.C., Manias, E., and Wang, Z. (2007) The Penn State Research Foundation (University Park, PA, US), United States.

89

4 Gas-Phase-Assisted Surface Polymerization and Thereby Preparation of Polymer Nanocomposites Haruo Nishida, Yoshito Andou, and Takeshi Endo

4.1 Introduction

The gas-phase-assisted surface polymerization (GASP) technique, which uses gaseous monomers including monomers with low boiling point and vaporized monomers, has been developed as the simplest method for constructing solventfree and precise microarchitectures on solid substrate surfaces (Figure 4.1). Gaps and spaces within the solid substrate give the GASP technique additional advantages, besides simplicity and precision, commending it for the construction of ﬁne structured composites [1–5] and coatings [6, 7]. This is because these gaps and spaces, when wider than the size of monomer molecule, allow the gaseous monomer to diffuse and penetrate interstitially within the solid substrates. After the diffusion and adsorption on the interstitial surfaces, the monomers are polymerized in a manner of “pseudografting” from the substrate surfaces. Polymer chains then grow on the surfaces by ﬁlling the gaps and spaces to give a superior anchoring effect, and consequent excellent binding strength at the polymer/ substrate interface. Moreover, the growing chain ends, which show a living nature even when a convenient free radical initiator 2,2′-azobis(isobutyronitrile) (AIBN) is used, enable block copolymers to form [8–17]. This GASP technique, which proceeds in the simplest manner without any solvent, has great potential in the construction of nanocomposites by combinations with inorganic/organic materials consisting of nanoscale structures.

4.2 In-situ Polymerization for Nanocomposite Preparation

Over the last two decades, the polymer–clay nanocomposites have been widely investigated as materials exhibiting superior properties, such as high modulus, increased thermal stability, and good gas-barrier characteristics [18–20]. Their development started from the nylon/clay hybrid found by Kamigaito et al. [21, 22] and has extended to various combinations of monomer/nanoﬁller using more In-situ Synthesis of Polymer Nanocomposites, First Edition. Edited by Vikas Mittal. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA.

90

4 Surface Polymerization and Thereby Preparation of Polymer Nanocomposites

Figure 4.1 GASP and conventional polymerization in the liquid phase.

controllable methods [23–28]. Their unique properties, with applications to automotive, mechanical, and biomedical devices, have attracted much attention from many researchers and companies [18–20, 29]. Since natural clay montmorillonite (MMT) is hydrophilic, it is very important to improve its organophilicity so that it can be compatible with organic monomers/ polymers. MMTs have been treated with quarternized alkyl ammonium ions to obtain organophilic surfaces with an accompanying increase in the distance between silicate layers. In order to improve MMTs compatibility with polymers and ensure exfoliation of the silicate platelets, various compatibilizing agents such as functional surfactants capable of copolymerizing [30, 31] or initiating [32–34] the intergallery polymerization, have been used for the modiﬁcation of MMT gallery surfaces. The preparation of polymer/clay nanocomposites utilizes bulk, solution, emulsion, and suspension polymerization methods of vinyl and cyclic monomers [35– 37], together with solution [38] and melt blendings [39]. Resulting nanocomposites have two types of structure: intercalated and exfoliated. In order to achieve a desirable structure through the exfoliation of silicate layers, various effective physical conditions for the in-situ polymerization have been investigated. Uthirakumar et al. [33] reported that in the preparation of high-impact polystyrene (PS)/MMT nanocomposites by solution polymerization, the low extra-gallery viscosity facilitated the diffusion of monomers into the clay layers to obtain exfoliated nanocomposites. Choi et al. [40] reported that monomers with high dipole moments showed an increase in the basal spacings of MMT before polymerization and produce exfoliated polymer–MMT nanocomposites. In contrast, those with low dipole moments showed smaller basal spacings and produced intercalated polymer–MMT nanocomposites. ¸ Sen et al. [41] examined effects of mixed solvents of tetrahydrofuran and water on the preparation of modiﬁed MMT by quarternized low molecular weight block copolymers of styrene and 4-vinylpyridine, following in-situ polymerization of styrene. They found that higher interlayer distances of the modiﬁed MMT and the desired exfoliated nanocomposite structures were achieved when the MMT modiﬁcation was conducted in a mixture solvent including 50 or 66 wt%

4.3 Characteristics of GASP

THF. Yan et al. [42] studied the preparation of PS/MMT nanocomposites in supercritical CO2, which functions both as a diffusion carrier of styrene and as a precipitation solvent for formed polystyrene. All the above-stated studies were intended to facilitate the approach of monomers to the intergallery spaces, in-situ polymerization, and consequent exfoliation of MMT platelets. In spite of the remarkable developments achieved in the preparation techniques of nanocomposites, there are still some problems associated with the liquid processes involved in nanocomposite preparation, such as the requirement of large amounts of aqueous/organic solvents, the aggregation of clay due to its insufﬁcient afﬁnity with monomers, and the lower level of homogeneous dispersibility in matrix polymers [34, 43–45]. The GASP method may provide a solution to these problems, because of its solventless nature and the superior diffusing ability of gaseous monomers.

4.3 Characteristics of GASP 4.3.1 Thin Layer Coating of Solid-Substrate Surfaces

Gaseous and vaporized monomers have obvious advantages resulting from their ability to diffuse onto various types of surfaces, such as extremely wide surfaces, very narrow spatial surfaces, and complex 3D-shaped surfaces. Even when the surface to be covered is chemically incompatible, the gaseous monomers easily diffuse into spaces found in narrow gaps and pores allowing the following GASP to produce a thinly layered coating on the surface or to ﬁll in the narrow spaces. Moreover, GASP is able to conﬁne ﬁne substrate particles in capsules. Polymerization of adsorbed monomers on solid surfaces results in the formation of ultrathin coating layers. Obtained thin polymer layers are attractive coating materials for many applications due to their wide range of chemical, mechanical, electrical, and optical properties that can be engineered to ﬁt speciﬁc needs. Therefore, the GASP method is very suited to the modiﬁcation of surface properties of bulk materials. Many technologies for the gaseous monomer treatment of solid surfaces have been developed, for example, chemical vapor deposition (CVD) [11, 17, 46–51], laser-generated gas-phase polymerization [52–54], and plasma polymerization [55–57]. These methods involve condensation and reactions between reactive species, resulting in deposition of polymers onto target surfaces. However, such methods often gave complex polymeric products such as insoluble networks [58]. GASP technique has been developed originally as the simplest method for constructing micro architectures on/with solid substrates [2, 3, 14–16]. Unlike plasma polymerization, the use of initiators immobilized on substrate surfaces avoids unnecessary reactions in the gas phase and leads to polymer layers with welldeﬁned chemical structures that resemble their bulk counterparts. For example,

91

92

4 Surface Polymerization and Thereby Preparation of Polymer Nanocomposites

Bartlett et al. [59] reported a solventless deposition polymerization of ﬂuorostyrene monomers with a peroxide initiator to prepare thin ﬁlms. Hsieh and Zellers [60] also prepared thin ﬁlms of silylated methacrylate monomers on substrates by UV photopolymerization in the vapor phase. Wang and Chang [61] reported grafting of poly(γ-benzyl L-glutamate) on silicon native oxide surfaces by a surface-initiated vapor deposition polymerization and suggested the preparation of block copolypeptides. Fu et al. [8] and Gu et al. [10] also demonstrated the thin-ﬁlm coating of silicon substrates by the ring-opening metathesis polymerization of vaporized monomers such as norbornene. 4.3.2 Physically Controlled Polymerization Behavior

GASP fundamentally proceeds in a similar manner to conventional free radical polymerization via the four elementary reaction steps of initiation, propagation, chain transfer, and termination. It was suggested by El-Shall and Reiss [62] that the termination step could be eliminated in the case of a homogeneous gasphase polymerization using a low concentration of initiating free radicals. GASP is a solvent-free process on substrate surfaces, on which the growing chain ends are immobilized. This restriction on growing chain ends induces a speciﬁc reactivity compared to the conventional free radical polymerization in bulk and solution. Gleason and coworkers demonstrated an initiated CVD method, in which the initiator and monomer are vaporized and introduced into a chamber simultaneously [63]. The initiator was thermally dissociated by heated ﬁlament wires, which were suspended in the chamber, to generate radical species. In this method, no pretreatment for initiator immobilization on substrate surfaces is necessary. The generated radicals initiate radical polymerization, which is followed by propagation and termination. Therefore, the elementary reactions, initiation, propagation, and termination, may occur in the gas phase, on the surface, or both. By analysis of the relationships among deposition rate, number-average molecular weight of produced polymers, and equilibrium monomer surface concentration, it was concluded that the chain propagation occurs predominantly on the surface. Yasutake et al. [11, 17] have demonstrated the physically controlled radical polymerization behavior of vinyl monomer vapors with conventional free radical initiators during GASP. The polymerization of methyl methacrylate (MMA) and styrene (St) by a free radical initiator AIBN on substrate surfaces resulted in the deposition of high molecular weight polymers, in which the proportional relationship between the number average molecular weight (Mn) and polymer yield was successfully obtained (Figure 4.2). The consecutive copolymerization of MMA and St produced a block copolymer, poly(MMA-b-St). These results evidence the living nature of the free radical polymerization on surfaces in the presence of monomer vapors. The SEM observations of the deposits revealed that the polymerization proceeded with a continuous polymer deposition, reﬂecting a predesigned pattern on the substrate surface.

4.3 Characteristics of GASP

Figure 4.2 Plots of Mn versus polymer yield on GASP of MMA with different concentrations of AIBN at 55 °C on the Al plate.

Figure 4.3

Photo-induced GASP of the vinyl monomer with a photoinitiator.

4.3.3 Photo-Induced Controlled Polymerization

Photo-induced GASP (Figure 4.3) of vinyl monomers MMA and St was carried out on solid surfaces precoated with a photoiniferter, 2-cyanoprop-2-yl-N,N′dimethyldithiocarbamate, under UV-irradiation, resulting in the formation of polymers on the irradiated parts of substrate surfaces [16]. Obtained polymers showed a proportional relation between Mn and polymer yield. Consecutive copolymerization of MMA and St led to the formation of block copolymers. These results demonstrate that photo-controlled radical polymerization of vaporized monomers occurred on solid surfaces. Andou et al. investigated the consecutive photo-induced GASP of MMA and St to build up a ﬁnely designed deposition of 2D and 3D patterns on solid surfaces (Figure 4.4) [14, 15]. Under UV-irradiation through a stripe-patterned photo mask on a Si-wafer or Au-plate surface, the photo-induced GASP resulted in the

93

94

4 Surface Polymerization and Thereby Preparation of Polymer Nanocomposites

Figure 4.4 2D and 3D patterning by photo-induced GASP.

reproduction of the designed and multilayered accumulative patterns made of block copolymer chains grafted from the surfaces. Based on the above examples, the characteristics of the GASP method can be summarized in the following points: 1)

Solvent free: no solvents are required during polymerization; postpuriﬁcation free: no need to purify products after the polymerization, and efﬁcient utilization of monomer feed stocks.

2)

Allows gaseous monomers to diffuse onto various surfaces, even if these are narrow spaces between silicate platelets, and to polymerize in situ.

3)

Capable of producing polymer coatings having well-deﬁned structures.

4)

Able to achieve controlled polymerization to obtain block copolymers by consecutive reactions due to the restricted mobility of growing species, even when conventional-free radical initiators are used.

5) Able to produce tailor-made polymer patterns on substrate surfaces using photo-induced GASP and photolithography techniques. 6)

Allows the selection of pre- and cointroduction methods of initiators and monomers in conformity to needed product structures.

7)

For polyoleﬁns and their composites, is the preferred method for large-scale industrial production with metallocene catalysts. Successful in-situ preparation of polypropylene nanocomposites by GASP technique has been mentioned in earlier reports [64].

These advantages illustrate the great potential of the GASP method in the preparation of nanocomposites by combinations of many kinds of monomers with (pre) nanostructured materials.

4.4 Composite Preparation by GASP

4.4 Composite Preparation by GASP 4.4.1 Polymer/Clay Nanocomposites

A small amount of clay dispersed in a polymer matrix gives better ﬁnal properties. The dispersion is controlled by various interactions between ﬁller particles and polymer chains; however, it is difﬁcult to achieve a homogeneous dispersion on a nanoscale without the pretreatment of clay surfaces, since, in many cases, the hydrophilic clay surface and hydrophobic polymer chains are incompatible. Polymethylmethacrylate (PMMA) is a transparent and rigid material with excellent ultraviolet stability, low water absorption, and outstanding outdoor weathering properties. In previous reports, PMMA–clay nanocomposites have been prepared by in-situ intercalative polymerization [65], emulsion polymerization [66, 67], solution polymerization [67], suspension polymerization [30, 36], and bulk polymerization [30, 35, 36, 68]. To achieve the exfoliation of silicate platelets, silicate-anchored initiators [23–28, 36, 44, 69, 70] and comonomers having ammonium groups [30, 71, 72] have been used to tether the chain ends or internal units to the silicate surfaces. In comparison with PMMA, PS is a more hydrophobic polymer, such that to overcome this and prepare the PS-based nanocomposites, various methods have been attempted including (i) procedures using initiator-MMT hybrids [24, 27, 32, 44], (ii) protonated amine and carboxyl terminated PSs [73, 74], (iii) in-situ bulk and solution polymerizations of St using coreactive organophilic MMT [75, 76], and (iv) melt compounding of organophilic layered silicates and PS in the presence of poly(styrene-co-vinyloxazoline) [77]. Even though these ingenious procedures have induced the successful exfoliation of the layered silicates, the processes are troublesome and still require large amounts of solvents [36, 44, 69, 70]. Andou et al. [4, 5] applied the GASP method for the preparation of PMMA and PS/clay nanocomposites with modiﬁed MMTs, whose surfaces were changed to possess an organophilic property using a trimethylstearyl ammonium salt followed by the intercalation of a free radical initiator AIBN (Figure 4.5). As a result of GASP, exfoliated polymer/clay nanocomposites with high clay contents (∼19 wt%) were easily obtained (Figure 4.6). PS-based copolymer/MMT nanocomposites were prepared by simultaneous and consecutive GASP of St and MMA on modiﬁed MMT, resulting in the accumulation of poly(MMA-ran-St) and poly(MMAblock-St), respectively, on the MMT surface with accompanying effective exfoliation of silicate layers (Figure 4.7). An important contributor to the exfoliation was the particular interaction of silicate surfaces with a small amount of MMA units in the copolymers. The GASP method allows for a wide range of vaporizable monomers to be utilized in the preparation of nanocomposites. Moreover, the GASP method does not require prehomogenizing and postpolymerization stages, which are often performed during the liquid process. The obtained polymer/clay nanocomposites

95

96

4 Surface Polymerization and Thereby Preparation of Polymer Nanocomposites

Figure 4.5 Random copolymerization of MMA and St by GASP on organically modiﬁed MMT

with AIBN.

Figure 4.6 XRD patterns of organically modiﬁed MMT with AIBN (C18MMT) and products

after GASP of MMA and St.

contain a high amount of clay and are useful not only for direct moldings, but also as a master batch for use by diluting with other polymers. 4.4.2 Polymer/Inorganic Compound (Nano)composites

The synthesis of polymer/inorganic compound nanocomposites, accomplished by multistep reactions in solutions, has been bothered by the serious problem of coagulating inorganic particles, especially, ﬁne metal particles in the polymer matrices [78–81]. In order to prevent coagulation, El-Shall and coworkers utilized

4.4 Composite Preparation by GASP

Figure 4.7 TEM image of melt-pressed poly(MMA-ran-St)/C18MMT nanocomposite

thin ﬁlm.

interactions between monomer vapors and laser-induced plasma generated from metal targets by laser vaporization/ionization. In experiments, the laser vaporization of metal targets induced the generation of ultraﬁne inorganic particles (silicon nanocrystal; ∼4 nm; SiO2: 10–20 nm in diameter) and cations, which were capable of catalyzing the cationic polymerization of vaporized vinyl monomers such as isobutylene. High molecular weight polymers containing submicron metal particles, in the form of polymer/metal nanocomposites were obtained successfully [82, 83]. This combination of laser vaporization of metals and gas-phase polymerization offers great advantages including less complicated processing and a onestep synthesis of polymer/metal nanocomposites. Inorganic nanoparticles such as metal/semiconductors (M/SC) immobilized in polymer matrices have attracted considerable interest in recent years due to their distinct individualistic and cooperative properties [84]. Although the control of size and shape of M/SC nanoparticles has been widely investigated, the fundamental mechanism of nanostructural formation and evolution is still poorly understood. A novel cryochemical solid-state synthesis technique has been developed to produce M/SC nanocomposites [85]. This method is based on the low-temperature cocondensation of M/SC and monomer vapors, followed by the low-temperature solid-state polymerization of the cocondensates. As a result of the method of stabilizing the metal particle without requiring any speciﬁc coordination bonds between the particle surface and the polymer matrix, generated nanoparticles (Ag´) were embedded in the polymer matrix with wellnanocrystal: mean size 50 Å controlled shapes and a narrow size distribution [86]. The polymer/metal particle composites have been synthesized by utilizing ﬁne metal surfaces as effective initiation sites for radical polymerization of vaporized vinyl monomers. On the metal surfaces, GASP of vinyl monomers is initiated and induces the formation of polymer thin-ﬁlm coatings of the ﬁne metal particles. Andou et al. demonstrated that GASP of MMA on a zero-valent iron (Fe(0))

97

98

4 Surface Polymerization and Thereby Preparation of Polymer Nanocomposites

Figure 4.8 GASP of vinyl monomer on a metal particle surface.

particles (<45 μm in diameter)/alkyl halide (RX; X = Cl, Br, I) initiator system resulted in successful high polymer production on the metal particle surfaces [2, 3]. GASP was initiated by radical species generated via expected redox reactions among Fe(0), Fe(II), Fe(III) species, and RX on the Fe(0) particle surfaces (Figure 4.8). It was also found that the physically controlled polymerization proceeded on the metal particle surface to produce polymers having well-deﬁned structures. Ring-opening GASP of β-propiolactone (βPL) was investigated using substratesupported anionic initiators such as tetramethylammonium acetate to produce a strongly surface bonded polypropiolactone (PPL)/CaO particle (primary particle size 3–5 μm) composite. A novel PPL crystalline with a high melting-point value was detected in the deposits on Al plates [1]. The polymerization of βPL smoothly proceeded at the interface between the gas phase and CaO particles to give high molecular weight PPLs having wide PDI values. A linear relationship between the Mn value and the incremental increase in the deposit was found. This suggests the living nature of the GASP of βPL. The obtained PPL/CaO composite showed a crazing of the polymer matrix because of a strong interaction at the organic/ inorganic interface (Figure 4.9). The absence of fracturing at organic/inorganic interfaces after tensile testing suggests that GASP is an effective method for preparing the strong adhesive coating of inorganic substrates. Potential applications of polymer/inorganic compound (nano)composites exist in the ﬁeld of polymer additives. Here, inorganic compound (nano)particles are

4.4 Composite Preparation by GASP a)

b)

10μm Figure 4.9

10μm

SEM images of uniaxially drawn PPL/CaO films. (a) GASP, (b) blend.

used to control mechanical, optical, and electrical properties, and to increase the abrasion resistance of matrices, at the same time as ensuring the transparency of the bulk material in cases of homogeneously dispersed nanoparticles. Furthermore, the interface adhesion between inorganic particle and polymer matrices is very important, because it directly inﬂuences almost all properties of the materials. To prepare polymer/inorganic compound (nano)composites having expected properties, employing well-known controlled polymerization methods, such as the atom transfer radical polymerization (ATRP) system, is practically useful. GASP is able to effectively use these controlled polymerization methods. 4.4.3 Polymer/Cellulose Fiber (Nano)composites

Cellulose nanoﬁbers (CNFs) [87, 88] and nanowhiskers (CNWs) [89] derived from renewable biomass have attracted much interest as alternatives to microsized and glass-ﬁber reinforcements in composite materials. Jonoobi et al. [87] developed CNF-reinforced polylactic acid (PLA) by twin-screw extrusion. Obtained PLA/CNF nanocomposites with 5 wt% CNF showed improved tensile modulus and strength. Ljungberg et al. [89] prepared nanocomposite ﬁlms of isotactic-PP reinforced with cellulose whiskers, which were highly dispersed in the PP matrix with a surfactant. The surfactant-modiﬁed whiskers acted as nucleating agents for isotactic-PP and the obtained nanocomposite displayed an increased tensile strength and strain at break as compared to the neat isotactic-PP. Although the biomass-originated materials are interesting additives, pretreatments of the materials with organic agents or in solutions are apt to result in some damage to the original properties of ﬂexibility, softness, and unique tactile nature. To avoid such damage on the material surfaces after pretreatment, Andou et al. [90] demonstrated a continuous introduction of vaporized initiator and monomer species. GASP of 2,2,3,3,3-pentaﬂuoropropyl methacrylate (FMA) was carried

99

100

4 Surface Polymerization and Thereby Preparation of Polymer Nanocomposites

Figure 4.10 Microscopic images of PFMA-coated cellulose ﬁbers by GASP (bar 50 mm) with a

water droplet to illustrate hydrophobicity.

out on cellulose ﬁbers using two vaporizable photoinitiators: 2-hydroxy-2methylpropiophenone and benzoin isobutyl ether, which were adsorbed on the cellulose surfaces. The cellulose surfaces were consecutively exposed to vaporized initiator and monomer under UV irradiation to initiate photopolymerization. The cellulose ﬁber surfaces were not only homogeneously coated by a thin polymer layer, but had obviously also retained their original tactile nature, showing a superior water-repellency evidenced by the controlled static contact angle being greater than 130 ° (Figure 4.10). This result demonstrated that GASP not only signiﬁcantly changes surface properties, but also avoids the morphological changes often associated with liquid processes. 4.4.4 Polymer/Carbon Nanotube (Nano)composites

Gleason and coworkers have demonstrated that the surface coating of carbon nanotubes (CNTs) by hot ﬁlament CVD of tetraﬂuoroethylene and by plasma enhanced CVD of MMA provide surface-modiﬁed CNTs having a stable/ superhydrophobic surface and good compatibility with PMMA, respectively [91, 92]. The superhydrophobic CNT indicated superior water repellency, whereby essentially spherical and micrometer-sized water droplets were suspended on top of the nanotube forest. The multiwall-CNT/PMMA nanocomposites were dispersed into PMMA via melt mixing. The orientation of CNT in the blend was achieved by melt drawing. The dispersed CNT in the blend indicated a signiﬁcant effect on mechanical properties of the blend even at a 1% concentration of CNT.

4.5 Outlook and Perspective

Technologies of the GASP method have been progressing day by day along with advances in nanocomposite chemistry. The key advantages of the GASP method

References

are its properties of being solvent free, postpuriﬁcation free, superior diffusion ability of its monomers, and its controlled propagation behavior. It is expected that the GASP method will continue to be used to exploit various potentialities functioning at the interface between gas-phase and solid substrates. For example, in such ﬁelds as materials, electronics, and photonics, the in-situ composite preparation using the GASP method can be achieved with various substrates, especially when characterized by the small size, wide surface area, complex shape, delicate surface, etc. Much effort has so far been directed at the ﬁelds of technology and science. In the future, with the excellent technologies and products developed with more expansions into novel ﬁelds are likely to occur. The abovementioned advantageous properties not only engender promising processes saving energy and resources, but also enable the production of variously designed nanocomposites relevant to an expanding ﬁeld of application.

Abbreviations

ATRP AIBN CNT CNF CNW CVD GASP M/SC MMA MMT Mn FMA PMMA PFMA PLA PPL PS βPL St

atom transfer radical polymerization 2,2′-azobis(isobutyronitrile) carbon nanotube cellulose nanoﬁber cellulose nanowhisker chemical vapor deposition gas-phase-assisted surface polymerization metal/semiconductor methyl methacrylate montmorillonite number average molecular weight 2,2,3,3,3-pentaﬂuoropropyl methacrylate polymethylmethacrylate poly(2,2,3,3,3-pentaﬂuoropropyl methacrylate) polylactic acid polypropiolactone polystyrene β-propiolactone styrene

References 1 Nishida, H., Yamashita, M., Andou, Y.,

3 Andou, Y., Yasutake, M., Jeong, J.M.,

Jeong, J.-M., and Endo, T. (2005) Macromol. Mater. Eng., 290, 848. 2 Andou, Y., Yasutake, M., Jeong, J.M., Nishida, H., and Endo, T. (2005) Macromol. Chem. Phys., 206, 1778.

Kaneko, M., Nishida, H., and Endo, T. (2007) J. Appl. Polym. Sci., 103, 1879. 4 Andou, Y., Jeong, J.M., Hiki, S., Nishida, H., and Endo, T. (2009) Macromolecules, 42, 768.

101

102

4 Surface Polymerization and Thereby Preparation of Polymer Nanocomposites 5 Andou, Y., Jeong, J.M., Nishida, H.,

6

7

8 9 10 11

12 13 14 15

16

17

18 19

20 21

22

23

24

and Endo, T. (2009) Macromolecules, 42, 7930. Lau, K.K.S., Mao, Y., Lewis, H.G.P., Murthy, S.K., Olsen, B.D., Loo, L.S., and Gleason, K.K. (2006) Thin Solid Films, 501, 211. Gupta, M., Kapur, V., Pinkerton, N.M., and Gleason, K.K. (2008) Chem. Mater., 20, 1646. Fu, D., Weng, L.-T., Du, B., Tsui, O.K.C., and Xu, B. (2002) Adv. Mater., 14, 339. Wang, Y. and Chang, Y.C. (2003) Adv. Mater., 15, 290. Gu, H., Xu, C., Weng, L.-T., and Xu, B. (2003) J. Am. Chem. Soc., 125, 9256. Yasutake, M., Hiki, S., Andou, Y., Nishida, H., and Endo, T. (2003) Macromolecules, 36, 5974. Chan, K. and Gleason, K.K. (2005) Langmuir, 21, 8930. Lau, K.K.S. and Gleason, K.K. (2006) Adv. Mater., 18, 1972. Andou, Y., Nishida, H., and Endo, T. (2006) Chem. Commun., 2006, 5018. Andou, Y., Yasutake, M., Nishida, H., and Endo, T. (2007) J. Photopolym. Sci. Technol., 20, 523. Yasutake, M., Andou, Y., Hiki, S., Nishida, H., and Endo, T. (2004) Macromol. Chem. Phys., 205, 492. Yasutake, M., Andou, Y., Hiki, S., Nishida, H., and Endo, T. (2004) J. Polym. Sci., Part A: Polym. Chem., 42, 2621. Alexandre, M. and Dubois, P. (2000) Mater. Sci. Eng., R28, 1. Giannelis, E.P., Krishnamoorti, R., and Manias, E. (1999) Adv. Polym. Sci., 138, 107. Kawasumi, M. (2004) J. Polym. Sci., Part A: Polym. Chem., 42, 819. Kojima, Y., Usuki, A., Kawasumi, M., Okada, A., Kurauchi, T., and Kamigaito, O. (1993) J. Polym. Sci., Part A: Polym. Chem., 31, 983. Usuki, A., Kawasumi, M., Kojima, Y., Okada, A., Kurauchi, T., and Kamigaito, O. (1993) J. Mater. Res., 8, 1174. Messersmith, P.B. and Giannelis, E.P. (1995) J. Polym. Sci., Part A: Polym. Chem., 33, 1047. Weimer, M.W., Chen, H., Giannelis, E.P., and Sogah, D.Y. (1999) J. Am. Chem. Soc., 121, 1615.

25 Ma, J., Qi, Z. and Hu, Y. (2001) J. Appl.

Polym. Sci., 82, 3611. 26 Manias, E., Touny, A., Wu, L.,

27 28 29 30 31

32

33

34

35 36 37

38

39

40 41 42 43

44 45

46

Strawhecker, K., Lu, B., and Chung, T. (2001) Chem. Mater., 13, 3516. Di, J. and Sogah, D.Y. (2006) Macromolecules, 39, 5052. Shi, Y., Peterson, S., and Sogah, D.Y. (2007) Chem. Mater., 19, 1552. Haraguchi, K. and Takehisa, T. (2002) Adv. Mater., 14, 1120. Wang, D., Zhu, J., Yao, Q., and Wilkie, C.A. (2002) Chem. Mater., 14, 3837. Nazarenko, S., Meneghetti, P., Julmon, P., Olson, B.G., and Qutubuddin, S. (2007) J. Polym. Sci., Part B: Polym. Phys., 45, 1733. Uthirakumar, P., Nahm, K.-S., Hahn, Y.-B., and Lee, Y.-S. (2004) Eur. Polym. J., 40, 2437. Uthirakumar, P., Kim, H.-J., Hong, C.-H., Suh, E.-K., and Lee, Y.-S. (2008) Polym. Compos., 29, 142. Nese, A., Sen, S., Tasdelen, M.A., Nugay, N., and Yagci, Y. (2006) Macromol. Chem. Phys., 207, 820. Lee, D. and Jang, L. (1996) J. Appl. Polym. Sci., 61, 1117. Huang, X. and Brittain, W. (2001) Macromolecules, 34, 3255. Bandyopadhyay, S., Giannelis, E.P., and Hsieh, A.J. (2000) Polym. Mater. Sci. Eng., 82, 208. Zha, W., Han, C.D., Moon, H.C., Han, S.H., Lee, D.H., and Kim, J.K. (2010) Polymer, 51, 936. Okamoto, M., Morita, S., Taguchi, H., Kim, Y., Kotaka, T., and Tateyama, H. (2000) Polymer, 41, 3887. Choi, Y.S., Ham, H.T., and Chung, I.J. (2004) Chem. Mater., 16, 2522. ¸ Sen, S., Nugay, N., and Nugay, T. (2009) J. Appl. Polym. Sci., 112, 52. Yan, C., Ma, L., and Yang, J. (2005) J. Appl. Polym. Sci., 98, 22. Kim, Y.K., Choi, Y.S., Wang, K.H., and Chung, I.J. (2002) Chem. Mater., 14, 4990. Di, J. and Sogah, D.Y. (2006) Macromolecules, 39, 1020. Du, K., He, A.H., Liu, X., and Han, C.C. (2007) Macromol. Rapid Commun., 28, 2294. Isoda, S. (1984) Polymer, 25, 615.

References 47 Macchi, E.M. (1972) J. Polym. Sci. Part 48

49 50 51

52 53 54

55

56

57

58 59

60 61 62 63 64 65

66 67 68

A – 1, 10, 45. Behbahani, H.F. and Inoue, H. (1988) J. Polym. Sci., Part B: Polym. Phys., 26, 1519. Chang, Y.-C. and Frank, C.W. (1998) Langmuir, 14, 326. Vaeth, K.M. and Jensen, K.F. (2000) Macromolecules, 33, 5336. Meier, C.B., Weickert, G., and Van Swaaij, W.P.M. (2001) J. Polym. Sci., Part A: Polym. Chem., 39, 500. Pola, J. (1997) Radiat. Phys. Chem., 49, 151. Ouchi, A., Koga, Y., Bastl, Z., and Pola, J. (1999) Appl. Organomet. Chem., 13, 643. Pola, J., Urbanova, M., Bastl, Z., and Morita, H. (2000) Macromol. Rapid Commun., 21, 178. Pan, Y.V., Barrios, E.Z., and Denton, D.D. (1998) J. Polym. Sci., Part A: Polym. Chem., 36, 587. Shirafuji, T., Miyazaki, Y., Nakagami, Y., Hayashi, Y., and Nishino, S. (1999) Jpn. J. Appl. Phys., 38, 4520. Groenewoud, L.M.H., Engbers, G.H.M., Terlingen, J.G.A., Wormeester, H., and Feijen, J. (2000) Langmuir, 16, 6278. Casserly, T.B. and Gleason, K.K. (2006) Chem. Vap. Deposition, 12, 59. Bartlett, B., Buckley, L.J., Godbey, D.J., Schroeder, M.J., Fontenot, C., and Eisinger, S. (1999) J. Vac. Sci. Technol. B, 17, 90. Hsieh, M.-D. and Zellers, E.T. (2002) Sens. Acutuators, B 82, 287. Wang, Y. and Chang, Y.-C. (2002) Langmuir, 18, 9859. El-Shall, M.S. and Reiss, H. (1988) J. Phys. Chem., 92, 1021. Chan, K. and Gleason, K.K. (2006) Macromolecules, 39, 3890. Reddy, C.S. and Das, C.K. (2007) J. Polym. Res., 14, 129. Krishnamoorti, R., Vaia, R.A., and Giannelis, E.P. (1996) Chem. Mater., 8, 1728. Li, Y., Zhao, B., Xie, S., and Zhang, S. (2003) Polym. Int., 52, 892. Xie, T., Yang, G., Fang, X., and Ou, Y. (2003) J. Appl. Polym. Sci., 89, 2256. Qu, X., Guan, T.G., Liu, G., She, Q., and Zhang, L. (2005) J. Appl. Polym. Sci., 97, 348.

69 Fan, X., Xia, C., and Advincula, R.C.

(2003) Langmuir, 19, 4381. 70 Fan, X., Xia, C., and Advincula, R.C.

(2005) Langmuir, 21, 2537. 71 Fu, X. and Qutubuddin, S. (2001)

Polymer, 42, 807. 72 Zeng, C. and Lee, L.J. (2001)

Macromolecules, 34, 4098. 73 Hoffmann, B., Dietrich, C., Thomann, R.,

74 75 76

77

78 79

80 81

82 83 84

85

86

87

88

89

Friedrich, C., and Muehaupt, R. (2000) Macromol. Rapid Commun., 21, 57. Zhao, J., Morgan, A.B., and Harris, J.D. (2005) Polymer, 46, 8641. Akelah, A. and Moet, M. (1996) J. Mater. Sci., 31, 3589. Camerani, M., Lelle, M., Sparanacci, K., Sandrolini, F., and Franxescangeli, O. (1998) J. Mater. Sci., 33, 2883. Park, C.I., Choi, W.M., Kim, M.H., and Park, O.O. (2004) J. Polym. Sci., Part B: Polym. Phys., 42, 1685. Collins, D.E. and Slamovich, E.B. (1999) Chem. Mater., 11, 2319. Govindaraj, B., Sastry, N.V., and Venkataraman, A. (2004) J. Appl. Polym. Sci., 93, 778. Du, H., Xu, G.Q., Chin, W.S., Huang, L., and Ji, W. (2002) Chem. Mater., 14, 4473. Burke, N.A.D., Stoever, H.D., and Dawson, F.P. (2002) Chem. Mater., 14, 4752. El-Shall, M.S. (1996) Appl. Surf. Sci., 106, 347. Li, S. and El-Shall, M.S. (1998) Appl. Surf. Sci., 127, 330. Biswas, A., Aktas, O.C., Kanzow, J., Saeed, U., Strunskus, T., Zaporojtchenko, V., and Faupel, F. (2004) Mater. Lett., 58, 1530. Song, H., Ilegbusi, O.J., and Trakhtenberg, L.I. (2005) Thin Solid Films, 476, 190. Trakhtenberg, L.I., Axelrod, E., Gerasimov, G.N., Nikolaeva, E.V., and Smirnova, E.I. (2002) J. Non-Cryst. Solids, 305, 190. Jonoobi, M., Harun, J., Mathew, A.P., and Oksman, K. (2010) Compos. Sci. Technol., 70, 1742. Zimmermann, T., Pöhler, E., and Schwaller, P. (2005) Adv. Eng. Mater., 7, 1156. Ljungberg, N., Cavaillé, J.-Y., and Heux, L. (2006) Polymer, 47, 6285.

103

104

4 Surface Polymerization and Thereby Preparation of Polymer Nanocomposites 90 Andou, Y., Jeong, J.M., Kaneko, M.,

Nishida, H., and Endo, T. (2010) Polym. J., 42, 519. 91 Lau, K.K.S., Bico, J., Teo, K.B.K., Chhowalla, M., Amaratunga, G.A.J.,

Milne, W.I., McKinley, G.H., and Gleason, K.K. (2003) Nano Lett., 3, 1701. 92 Gorga, R.E., Lau, K.K.S., Gleason, K.K., and Cohen, R.E. (2006) J. Appl. Polym. Sci., 102, 1413.

105

5 PET Clay Nanocomposites by In-situ Polymerization Hua Deng, Ke Wang, Qin Zhang, Feng Chen, and Qiang Fu

5.1 Introduction

Research activities in nanomaterials have increased dramatically over the last decade. Nanocomposites based on polymers have been considered as one of the most important methods to modify polymer matrices at the nanometer level [1]. These materials differ from conventional materials not only thanks to the properties of the nanoﬁllers, but also to their extreme small size and large surface area which allow them to have much more interaction with the polymer matrix than conventional composites. In geology, the term clay includes particles <2 μm in size, the morphology of the clay-mineral components being a distinctive property of a particular clay. For instance, kaolinite usually shows hexagonal ﬂake-shaped unites with a ratio of areal diameter to thickness (aspect ratio) of 2–25 : 1, while most of smectite mineral particles have an irregular ﬂake shape but with a much higher aspect ratio of 100–300 : 1. Halloysite minerals show an elongated tubular shape, while the family of attapulgite/sepiolite/palygorskite is characterized by an elongated or a ﬁber shape [2]. The research area of polymer/clay nanocomposites has attracted a great deal of interest in the last two decades. Outstanding improvements in the physical properties of polymeric matrices (e.g., stiffness, strength, heat distortion temperature, reduced permeability to gas and liquids, ﬁre retardancy, etc.) have been obtained with only few weight percents of nanoﬁller, therefore promising to eliminate the typical compromise between the properties and processability of composite materials. Since the ﬁrst work on polyamide 6 (PA6)/montmorillonite (MMT) nanocomposites obtained via in-situ intercalative polymerization [3], a wide range of polymers have been studied as hosting matrices for nanoclays and a rich literature is available [2, 4–11]. Nevertheless, only a small number of papers has been reported on ﬁber-shaped clay/polymer nanocomposites [9, 10, 12, 13]; most of the work on polymer/clay nanocomposites reported in the literature are based on MMT/polymer. Therefore, nanocomposites based on MMT will be mainly reviewed in this chapter.

In-situ Synthesis of Polymer Nanocomposites, First Edition. Edited by Vikas Mittal. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA.

106

5 PET Clay Nanocomposites by In-situ Polymerization

Polyethylene terephthalate (PET) as one of the most important engineering thermoplastics has a wide range of applications in the areas of ﬁbers, clothes, soft drink bottles, etc., due to its low cost and high performance. Thus, any improvement for PET in terms of its crystallization, antistatic, ﬁre retardant, barrier, and mechanical properties could enhance its potential for industrial application [14].

5.2 Preparation of PET/Clay Nanocomposites

Several methods including melt compounding, solution-based process, and in-situ polymerization could be used for the preparation of PET/clay nanocomposites. In-situ polymerization involves swelling of clays in a liquid monomer, or a monomer solution, followed by polymerization initiated thermally or by the addition of a suitable compound in the presence of intercalated/exfoliated clays. It is considered as one of the best methods to produce polymer nanocomposites based on clay because relatively small-sized monomers could be easily intercalated into clay layers before polymerization. Then, further polymerization between clays could easily lead to well-dispersed clay in the polymer matrix. Furthermore, clay could be added into the recipe for the polymerization process; thus, no further process is needed to realize industrial application. The most important issue regarding the preparation of polymer nanocomposites based on clays is how to exfoliate the clay uniformly in the polymer matrix into individually layers. An interesting method has been reported by Tsai et al. [15], where a catalyst is intercalated into the gallery spaces of the clay, and this step is combined with the novel polymerization process to disperse clay in the polyester (see Figure 5.1). It is believed that the preintercalated catalyst could help exfoliating clay during polymerization, as the polymerization process is more likely to occur between clays compared with the conventional method. A similar clay-supported catalyst-based method has also been reported by Choi et al. [16], where chlorotitanium triisopropoxie was used as a catalyst. In the study reported by Yin et al. [17], a novel organic MMT, which could act as both polycondensation catalyst for PET and ﬁller of PET/clay nanocomposites, was prepared. The original MMT was ﬁrstly treated with different amounts of poly(vinylpyrrolidone) (PVP) and then intercalated by TiO2/SiO2 sol to gain polycondensation catalytic activity. Furthermore, Chen et al. [14] developed a simple method to prepare PET/MMT nanocomposites without an organo-modiﬁer. Bis(hydroxyethyl) terephthalate (BHET), monomer of PET, was ﬁrstly solution blended with MMT to produce the clay gallery. The as-prepared BHET-MMT powder was used as the nanoﬁller for the preparation of hybrids. Then, these powders were mixed with ethylene glycol (EG) and other additives to carry out conventional polymerization. Martinez-Gallegos et al. [18] have compared the difference between PET/clay composites prepared with in-situ polymerization and the mechanical grinding method. As shown in Figure 5.2, it can be observed that the chemical nature of the specimens is the same in both cases. However, the spectrum of the specimen produced by the in-situ polymerization method (PET-LDH) shows broader bands

5.2 Preparation of PET/Clay Nanocomposites

Figure 5.1 The proposed mechanisms for the (d) Polymerization occurs between the

formation of polymer/clay nanocomposites by the driving force concept. (a) Sodium-type smectite clays. (b) Intercalation of catalyst/ initiator and modiﬁed agent leads to inter-layer spacing expansion. (c) Monomers/ oligomers were driven by the catalyst/initiator to swell into the gallery of clay lamellar.

adjacent silicate layers and the growth of polymer molecule breaks the lamellar structure of the clays into individually disordered exfoliated layers in the polymer matrix. Reproduced from Ref. [15] with permission.

Figure 5.2 FT-IR spectra of composites prepared by (a) in-situ polymerization (PET-LDH

series), and (b) hand grinding in an agate mortar (M-PET-LDH series). Reproduced from Ref. [18] with permission.

107

108

5 PET Clay Nanocomposites by In-situ Polymerization

than the specimens prepared by mechanical grinding. This difference indicates the existence of weak–medium interaction between polymer and layered material in specimens produced by the in-situ polymerization method, and the lack of any interaction in those prepared by mechanical grinding. Kim et al. [19] fabricated PET/clay composites using a two-step in-situ polymerization method. In the ﬁrst step, a slurry mixture of monomer (puriﬁed terephthalic acid and ethylene glycol), polycondensation catalyst, clays, and some additives was kept at 250 °C for 5–6 h in the esteriﬁcation step. Then, it was transferred to a polycondensation reactor until the intrinsic viscosity (IV) value reached 0.6 dl g−1. Then, the materials were pelletized. Furthermore, a solid-state polymerization (SSP) process is carried out to conduct the polymerization process further. SSP was carried out at between 220 and 145 °C for around 8 h until the IV reached 0.8 dl g−1.

5.3 Morphology of the Nanocomposites

The morphology of the nanocomposites plays a signiﬁcant role in different properties of the composites, and this has been investigated extensively in the literature. As one of the most powerful tools to investigate the morphology of polymer/MMT composites, powder X-ray diffraction (XRD) has been widely used to study the spacing between clays to reﬂect their dispersion status. Figure 5.3 shows the XRD pattern of PET and some PET/clay nanocomposites [15]. It can be observed that the dispersion of modiﬁed clay in the PET/clay nanocomposites varies while different clays are used. Figure 5.3a shows the characteristic peaks of neat PET ranges from 15 ° to 35 °. Figure 5.3b demonstrates the XRD peak of the composites produced with a method without intercalating a catalyst into the gallery spaces of the clays. A clear peak from clay (0 0 1) reﬂection around 6.1 ° is observed indicating a d-spacing of around 14.48 Å is obtained between clays. The clay (0 0 1) peak disappeared in Figure 5.3c, which indicates the complete exfoliation of clay in the PET/ clay composites produced with the catalyst preintercalated. In the work reported by Chang et al. [20], they observed that drawing PET/clay ﬁber could enhance the dispersion of clay in the polymer. In the XRD pattern they showed, a d-spacing of 17.25 Å is observed before drawing and the characteristic peak of clay disappeared after drawing. This demonstrates that a good dispersion of clay has been achieved. Nevertheless, XRD is a useful tool to study the d-spacing of ordered intercalated polymer nanocomposites, it might not be sufﬁcient for exfoliated nanocomposites with no peak. It is generally accepted that transmission electron microscopy (TEM) is the best direct method to demonstrate the dispersion of nanoﬁllers in a polymer matrix. It has been widely used to investigate the dispersion of clay in a PET matrix. Figure 5.4 shows the TEM study of PET/MMT composites [14]. It can be observed that the dispersion of clay in the composites varies due to different processing procedures. It can be either labeled as nanocomposites or conventional

5.4 Crystallization of the Nanocomposites

X-ray Intensity

a)

b)

c)

0

5

10

15 20 25 2θ (degrees)

30

35

40

Figure 5.3 Powder XRD patterns of polyester/clay nanocomposites from different modiﬁed

clays: (a) Pure PET. (b) PET/PK-805/SB. (c) PET/PK-805/Sb/SB. Reproduced from Ref. [15] with permission.

microcomposites according to the status of ﬁller dispersion. The better dispersion (Figure 5.4b) achieved here is obtained by a precompounding monomer with clay as discussed in the above section, and the poor dispersion is achieved while Na+MMT is used without the preblending procedure.

5.4 Crystallization of the Nanocomposites

As well known, nanoﬁllers often have signiﬁcant inﬂuence on the crystal morphology of a semicrystalline polymer. As shown in Figure 5.5, various spherulite morphologies of PET and PET/clay nanocomposites have been characterized with polarized optical microscopy [21]. Neat PET demonstrates a well-deﬁned spherulite texture with a ﬁbril pattern and a maltese cross at 200 °C. However, the maltese cross is not observed at higher crystallization temperature. PET/clay nanocomposites exhibited typical crystalline morphologies indicating that clay exhibits a

109

110

5 PET Clay Nanocomposites by In-situ Polymerization a)

b)

0.5 μm

0.5 μm

Figure 5.4 TEM photographs for PET/MMT hybrids: (a) PETNA and (b) PETBM. Reproduced

from Ref. [14] with permission. a)

b)

c)

e)

f)

10.0 μm d)

Figure 5.5 Micrographs showing spherulite

growth from samples held at different crystallization temperatures for 5 min: (a) PET at 200 °C, (b) PET at 220 °C, (c) PET/Na+-MMT 0.5 wt% at 220 °C,

(d) PET/Na+-MMT 2.0 wt% at 220 °C, (e) PET/A10-MMT 2.0 wt% at 220 °C, and (f) PET/A10-MMT 5.0 wt% at 220 °C. Reproduced from Ref. [21] with permission.

5.4 Crystallization of the Nanocomposites a)

b) 2.0 wt% 1.0 wt%

2.0 wt%

PET

Endo

Endo

0.5 wt%

1.0 wt% 0.5 wt% PET

0 0

50

100 150 200 Temperature (°C)

250

300

d) 5.0 wt% 2.0 wt% 1.0 wt% 0.5 wt% PET

50 100 150 200 250 300 Temperature (°C)

Endo

Endo

c)

0

50 100 150 200 250 300 Temperature (°C)

0

5.0 wt% 2.0 wt% 1.0 wt% 0.5 wt% PET 50 100 150 200 250 300 Temperature (°C)

Figure 5.6 DSC thermograms of PET/

Na1-MMT, (c) cooling run of PET/A10-MMT, Na1-MMT and PET/A10-MMT nanocomposite and (d) 2nd heating thermogram of PET/ materials: (a) a cooling run of PET/Na1-MMT, A10-MMT. Reproduced from Ref. [21] with (b) 2nd heating thermogram of PET/ permission.

nucleating agent effect. As a result, the crystallization rate of PET is increased. The spherulites in the composites are smaller compared with the ones in neat PET. It is interesting to note that different spherulite forms are obtained with increasing clay content. Compressed peanuts-shaped spherulites are obtained in composites containing 0.5 wt% clay (Figure 5.5c), and star-shaped spherulites are obtained for composites containing 5.0 wt% clay (Figure 5.5f ). Differential scanning calorimetry (DSC) is often used as an effective method to study the crystallization behavior of polymer or polymer composites. Figure 5.6 shows the thermograms of PET/Na+-MMT and PET/A10-MMT. The results are summarized in Table 5.1. As shown in Figure 5.6a, the crystallization temperature of PET/Na+-MMT during cooling increases with increasing clay content. This indicates that Na+-MMT acted as a nucleating agent for the PET matrix. In the following heating process, the composites demonstrate two melting peaks compared with a single peak from neat PET (see Figure 5.6b). The lower melting peak is thought to come from the melting of imperfect crystals, which was formed driven by the nucleation effect of Na+-MMT during cooling. The higher melting peak comes from the melting of the reorganized crystal during melting [22].

111

112

5 PET Clay Nanocomposites by In-situ Polymerization Table 5.1 Thermal properties of PET/Na1-MMT and PET/A10-MMT nanocomposites.

Sample name

ηinha)

Tc (°C)b)

Tm (°C)c)

ΔT d)

Hc (J g−1)e)

Hm (J g−1)e)

PET Na+-MMT-0.5 wt% Na+-MMT-1.0 wt% Na+-MMT-2.0 wt% 0A-MMT-0.5 wt% 0A-MMT-1.0 wt% 0A-MMT-2.0 wt% 0A-MMT-5.0 wt%

0.76 0.72 0.73 0.74 0.74 0.75 0.72 0.70

184 206 210 214 195 192 187 187

253 248/256f) 249/256f) 249/256f) 255 255 255 253

69 50 46 42 60 63 68 66

−38 −40 −42 −45 −39 −45 −43 −40

36 38 40 41 46 42 40 41

Reproduced from Ref. [21] with permission. a) Intrinsic viscosity (IV) were measured by using concentration of 0.2 g dl−1 in phenol/1,1,2,2etrachloroethane (6/4 v/v) solution. b) Crystallization temperature from melt. c) Melting temperature. d) The degree of supercooling (Tm − Tc = ΔT ). e) Enthalpy change of fusion. f ) The standard of 2nd Tm.

Nevertheless, the DSC curves from 0.5 wt% PET/A10-MMT shown in Figure 5.6c demonstrate increased peak intensity during cooling. The intensity of such a peak decreases with increasing clay loading, but it remains higher than the peak from neat PET. This shows that the nucleation effect of A10-MMT is lower than that of Na+-MMT. Such an effect becomes more obvious at higher A10-MMT contents. The authors believe that this might be caused by the interference from the alkyl groups on the A10-MMT surfaces with secondary nucleation and diffusion of PET chains, which lead to a decrease in the intensity of the DSC crystallization peak. The melting curve for PET/A10-MMT is shown in Figure 5.6d. A small shoulder peak is observed. It is attributed to the melting of imperfect crystals obtained during cooling. However, the melting peak decreases slightly with increasing A10MMT content as shown in Table 5.1.

5.5 Properties of the Nanocomposites 5.5.1 Thermal Properties

The aim of adding clay into polymer matrix is to improve various properties (mechanical, thermal, barrier, etc.) of the composites. The thermal stability of neat polymer is rather poor and improvement is required for a wide range of applications. As well known, nanoclays are thought to play important roles in the

5.5 Properties of the Nanocomposites

Figure 5.7 TGA curves of clay, organoclay, and PET hybrid ﬁbers with different organoclay

contents. Reproduced from Ref. [23] with permission.

thermal stability of the composites. It is widely believed that the formation of a char layer, obtained due to the collapse of intercalated/exfoliated structures during polymer decomposition, can act as a barrier to both mass and energy transportation. Therefore, the thermal stability of the composites could be improved by adding nanoclay into the polymer matrix. The thermal stability of in-situ polymerized PET/clay composites is often primarily characterized with TGA as shown in Figure 5.7 [23]. The initial degradation temperature of the composites increased with increasing clay content. It is shown that the TD (temperature at 2% weight loss) for neat PET and its composites range from 370 to 389 °C. The composites containing 5 wt% of clay demonstrates the highest TD. It is believed that the increase in TD with clay content could be caused by the high thermal stability of clay and the interaction between the clay particles and the polymer matrix. Furthermore, the mass-transport barrier introduced by clay to the volatile products generated during decomposition is thought to play an important role as well. The weight residue at 600 °C increases from 1% to 22% while the clay content is increased from 0% to 5%. Such an enhancement is caused by the high heat resistance of the clay. The thermal degradation of PET with different concentrations of OMMT and SiO2 was investigated with TGA [24]. As shown in Figures 5.8 and 5.9, the mass (%) and derivative mass (DTG) trends of all specimens are plotted. It is noted that the degradation of OMMT can be divided into three stages. A small mass loss (less than 5 wt%) is observed below 300 °C, such a weight lost mainly originates from water content in MMT. The highest rate observed in the second degradation stage occurred at around 325 °C. It is well above the temperature at which PET is polymerized (250 °C). Finally, the third degradation is observed at around 470 °C. These TGA results from OMMT conﬁrm that a thermal stable organically modiﬁed MMT has been obtained. Regarding the TGA study of these nanocomposites, there is no obvious degradation observed for temperature up to 330 °C. For PET/OMMT and PET/SiO2 composites, there is no signiﬁcant difference between these degradation curves as shown in Figures 5.8 and 5.9.

113

5 PET Clay Nanocomposites by In-situ Polymerization a) 100

PET + 2.0 wt% OMMT PET

mass (%)

80 OMMT 60 PET + 1.0 wt% OMMT

40 20

PET + 0.5 wt% OMMT

100

200

b)

300 400 Temperature (°C)

500

600

500

600

OMMT

DTG

114

PET + 0.5 wt% OMMT PET + 1.0 wt% OMMT PET + 2.0 wt% OMMT PET 100

200

300 400 Temperature (°C)

Figure 5.8 (a) Mass (%) versus temperature and (b) derivative mass (DTG) versus tempera-

ture at a heating rate of 10 °C/min for PET nanocomposites with OMMT. Reproduced from Ref. [24] with permission.

In order to investigate the degradation mechanism of PET and its nanocomposites, studies on the kinetic parameters (activation energy E and pre-exponential factor A) and conversion function f(α) have been carried out [24]. Vassiluou et al. used the method proposed by Ozawa, Flynn, and Wall (OFW) [25, 26] to calculate the activation energy E. Such a method involves measuring the temperatures corresponding to ﬁxed values of the degree of mass conversion at different heating rates using Eq. (5.1): ⎡ Af (α ) ⎤ E ln β = ln ⎢ ⎥− ⎣ da / dT ⎦ RT

(5.1)

Therefore, the activation energy can be determined by plotting ln β against 1/T according to Eq. (5.1), as the slope of the line is proportional to the activation

5.5 Properties of the Nanocomposites a) 100

Mass (%)

80

60 PET + 2.0 wt% SiO2

40 20 300 b)

PET + 1.0 wt% SiO2 PET + 0.5 wt% SiO2 PET 350 400 450 500 Temperature (°C)

550

DTG

PET + 0.5 wt% SiO2

PET + 1.0 wt% SiO2 PET + 2.0 wt% SiO2 300

350

PET

400 450 Temperature (°C)

500

550

Figure 5.9 (a) Mass (%) versus temperature and (b) derivative mass (DTG) versus tempera-

ture at a heating rate of 10 °C/min for PET nanocomposites with SiO2. Reproduced from Ref. [24] with permission.

energy. A single-step reaction can be concluded if the calculated activation energy is the same for various α. In contrast, a complex reaction mechanism invalidates the separation of variables considered in the OFW analysis if a change in E is observed with increasing degree of conversion [24, 27]. As shown in Figure 5.10, the dependence of the activation energy (E) versus conversion α for PET, 2.0 wt% OMMT/PET, and 2.0 wt% SiO2/PET composites are plotted. It can be noted that the OMMT improves the thermal stability slightly compared with SiO2. The reaction mechanism of polymer degradation is indeed a very complicated, as initiation, propagation, and termination reactions are included. To determine the conversion function f(α), the “model-ﬁtting method” has been used [24]. The results of the ﬁtting for 2 wt% OMMT/PET composites are shown in Figure 5.11

115

5 PET Clay Nanocomposites by In-situ Polymerization

PET PET + 2.0 wt% SiO2 PET + 2.0 wt% OMMT

activation energy (kJ mol–1)

240

230

220

210 0.0

0.2

0.4 0.6 mass conversion (a)

0.8

1.0

Figure 5.10 Dependence of the activation energy (E ) on the degree of the conversion (α) of

the mass loss, as calculated with the OFW method for the different samples (lines are guide to eye). Reproduced from Ref. [24] with permission.

100 3

80 Mass (%)

116

60 2 4

40 1 20 350

375

400 425 450 Temperature (°C)

475

500

Figure 5.11 Fitting (red line) and experimental mass (%) (black symbols) curves for all the

different heating rates for PET containing 2.0 wt% OMMT. (1) 5 K min−1, (2) 10 K min−1, (3) 15 K min−1, (4) 20 K min−1. Reproduced from Ref. [24] with permission.

as an example. The obtained form of the conversion function is the mechanism of nth-order-auto-catalysis for all the specimens studied, as described in Eq. (5.2): f (α ) = (1 − α )n (1 + K cat X )

(5.2)

where Kcat is a constant and X is a product in the complex model. The calculated parameters can be found in the work reported by Vassiliou et al. [24], where the

5.5 Properties of the Nanocomposites 1.8

2400 2000 1800 1600

1.4 1.2

1400

Tanδ

Storage modulus (MPa)

1.6

NPET PETNA2 PETBM1 PETBM2 PETBM3

2200

1200 1000

0.8 0.6

600

0.4

400

0.2

200

0.0

0

30

NPET PETNA2 PETBM1 PETBM2 PETBM3

1.0

800

–200

117

40

50 60 70 80 Temperature (°C)

90

–0.2 100 30

40

50 60 70 80 Temperature (°C)

Figure 5.12 DMA curves for neat PET and PET/MMT hybrids. Reproduced from Ref. [14] with

permission.

activation energy (E, kJ mol−1) for neat PET, 2.0 wt% OMMT/PET, and 2.0 wt% SiO2/PET composites is 223.5, 228.3, and 222.1, respectively. The activation energy of OMMT/PET is higher than that of the composites containing SiO2/PET, indicating that a better stability is obtained while OMMT is added into the matrix. 5.5.2 Mechanical Properties

Mechanical properties are largely inﬂuenced by the addition of clay in polymer matrix thanks to the outstanding properties of clay and the strong interaction with the polymer matrix due to its huge speciﬁc surface area. As shown in Figure 5.12, the dynamic mechanical property of PET/clay composites is investigated [14]. It is noted that the storage modulus and glass transition temperature of the composites increased with increasing clay content, while the monomer of PET was ﬁrstly solution blended with MMT to produce the clay gallery and used as the nanoﬁller for the preparation of the hybrids (labeled as PETBM). Nevertheless, there is almost no change for those composites produced by direct blending followed by in-situ polymerization (labeled as PETNA). The improvement in storage modulus has been widely reported [28]. However, an increase in Tg is rarely reported in the literature because the hybrids always have much lower molecular weight and the plasticizer effect of organic-modiﬁed clay. Such an increase in Tg in PETBM is related to the reduced polymer chain mobility caused by improved interaction between clay and polymer thanks to well-dispersed clay. Figure 5.13 shows the tensile behavior of PET/clay composites [14]. The tensile behavior of PETNA2 is similar to neat PET, except the yielding of PETNA rapidly collapses. This is caused by the rather poor dispersion of clay in PETNA2 as shown in Figure 5.4a. The tensile behavior of PETBM differs from neat PET and PETNA,

90

100

5 PET Clay Nanocomposites by In-situ Polymerization 50 PETBM2

45 40 35 Stress (MPa)

118

30 PETBM3

NPET

25 20 15 PETNA2

10 5 0 0

2

4

6

8

10 12 Strain (%)

14

16

18 20

Figure 5.13 Stress–strain curves for neat PET and PET/MMT hybrids. Reproduced from Ref.

[14] with permission.

a brittle rupture behavior is observed. Furthermore, a sample from PETBM3 fails before it reaches the elastic deformation limit. This is caused by the agglomerations at a relatively high ﬁller content (3 wt%). It is also shown in Table 5.2 [15] that the ﬂexural strength and ﬂexural modulus improved with the addition of clay in PET. A more notable increase from 730 to 1252 kg cm−2 is observed in ﬂexural strength while 2.5 wt% of clay is contained in the PET matrix. Meanwhile, a 30 °C increase in heat distortion temperature (HDT) is obtained. Complete exfoliated clay in the polymer matrix is thought to be responsible for the improvement in these properties. 5.5.3 Barrier Properties

The barrier properties of PET play very important roles for applications such as: beverage bottle, food packaging, etc. The high aspect ratio of layered silicates has been found to signiﬁcantly reduce the gas permeability in exfoliated nanocomposites ﬁlms, by the creation of a “tortuous path” (Figure 5.14) that reduces the diffusiveness of gas molecules. The plate-like shape of the nanoclays is particularly efﬁcient in maximizing the path length that a diffusing molecule must travel, because of the high aspect ratio, compared to others ﬁllers shapes (i.e., sphere or cube). As shown in Figure 5.15, the oxygen transmission rate (OTR) of PET/MMT composites is investigated by Kim et al. [19]. The OTR of the composites is signiﬁcantly improved with the addition of MMT into the PET matrix. The value decreases from 3.41 to 6.48 while 1 wt% of MMT is incorporated into PET.

5.5 Properties of the Nanocomposites Table 5.2

119

Properties of PET/clay nanocomposites.

Samplea)

Clay content (wt%)

IVc)

Pure PET

0

0.523

730

23000

74

PET/ PK-802/Sb/ SB

0.66

0.604

1200

34069

100.5

PET/CWC/ Sb/SB

1.22

0.744

1215

34391

PET/ PK-805/Sb/ SB

2.5

0.744

1252

2.5

0.572

1061

Flexural strength

UVd) trans. (%)

CO2 Barrier

Clarity

75

0.304

Transparent

2.23

0.04

Transparent

102

1.42

0.01

Transparent

34664

104

0.33

–

Hazy

30301

82

0.225

Hazy

Flexural modulus

HDTb) (18.6 kg f cm−2)

Exfoliated

Intercalated PET/ PK-805/SB

18.6

Reproduced from Ref. [15] with permission. a) The PET/clay nanocomposites samples were prepared in the same reactor under the same condition with different ratios and types of clays. b) HDT: heat distortion tests. c) IV: intrinsic viscosity. d) UV transmission was measured at 375 nm for each sample.

Figure 5.14 Formation of a “tortuous path” in polymer–clay nanocomposites.

5 PET Clay Nanocomposites by In-situ Polymerization 7 OTR (mm cc/m2 day)

120

6

Pure PET

5 MMT-PCN-1.0%

4 3

MDEA III-PCN-1.0% 2 1

5

0

10 Time (h)

15

20

Figure 5.15 The oxygen transmission rate (OTR) graph of a pure PET sheet and PET/clay

nanocomposite sheets. Reproduced from Ref. [19] with permission.

As mentioned above, the plate-like-shaped clay is indeed efﬁcient in maximizing the path length of a diffusing molecule. According to Nielsen [29], the tortuousity factor τ is deﬁned as the ratio of the actual distance d′, that a diffusient must travel, to the shortest distance d, that it would travel in the absence of obstacles. τ can be expressed in terms of the length L, width W, and volume fraction φ of the sheetlike particle as follows:

τ=

d′ L =1+ ϕ d 2W

(5.3)

The effect of tortuosity on the permeability is expressed as PS 1 − ϕ = PP τ

(5.4)

where PS and PP represent, respectively, the permeability of the nanocomposite and the pure polymer. Supposing that the layered clays are arranged perpendicular to the direction of diffusion, a key role is played by the aspect ratio. This simple model has been usefully employed by different authors to interpret experimental data. The UV barrier property is also one of the most important issues for food and beverage packaging as UV radiation could induce photo-oxidation, which has signiﬁcant inﬂuence on the quality of some food product [30]. As shown in Figure 5.16, a 4–9% reduction in the UV transmission rate is obtained at a wavelength of 360 nm for PET/clay composites compared with neat PET. A similar study has also been carried out by Tsai et al. [15] as shown in Table 5.2. It is interesting to note that a signiﬁcant decrease in UV transmission (from 75% to 0.33%) has been observed. Meanwhile, the CO2 barrier property decreases from 0.304 to 0.01 while 1.22 wt% of clay is added into the PET matrix. By comparing the results shown in Table 5.2 with those in Figure 5.16, it can be noted that much better UV barrier

5.6 Conclusion and Outlook

Transmission rate (%)

80 70 60 50 40 30 20

Pure PET MDEA III-PCN MDEA IV-PCN

10 0 300 310 320 330 340 350 360 370 380 390 400 Wavelength (nm)

Figure 5.16 UV transmission spectra of pure PET, MDEA III–PCN, and MDEA IV–PCN.

Reproduced from Ref. [19] with permission.

property was achieved by Tsai et al. It is believed that this is directly related to the poorer dispersion status of clay in the latter. Furthermore, the clarity/transparency of PET/clay nanocomposites is also one of the most important issues for some applications, such as packaging, beverage bottles, etc. It is noted that the transparency can be maintained at a clay content of less than 2 wt% (see Table 5.2). At the mean time, reasonable barrier properties (CO2 and UV) are achieved. This demonstrates that a balance could be achieved between clarity and good barrier properties for these nanocomposites.

5.6 Conclusion and Outlook

The in-situ polymerized PET composites containing clay are reviewed in this chapter. Issues including the preparation of PET/clay nanocomposites, morphology of the nanocomposites, crystallization of the nanocomposites, and different properties of the nanocomposites are discussed in detail. It is shown that clays could be fully exfoliated in the PET matrix through different in-situ polymerization methods. The improvement in mechanical properties and barrier properties by adding clay into the PET matrix is more noticeable than other properties (such as thermal stability). Nevertheless, the dispersion status of clay is shown to play a vital role in different properties of the ﬁnal composites. It is shown that an in-situ polymerization method dose not guarantee good ﬁller dispersion, careful processing control, or pretreatment needs to be carried out to fully exfoliate clay into a polymer matrix. The potential of clay/PET composites lies in their industrial application such as: barrier materials for barrage bottles, ﬁlms, ﬁbers, etc. More work is needed to further improve their properties to replace their potential competitor in the market. Moreover, simpler methods are needed to produce these materials at lower cost.

121

122

5 PET Clay Nanocomposites by In-situ Polymerization

References 1 Ishida, H., Campbell, S., and Blackwell, J. 2

3

4 5 6

7 8 9

10

11 12

13

14 15

(2000) Chem. Mater., 12, 1260. Bilotti, E. (2010) Polymer/sepiolite clay nanocomposites. Ph.D Thesis, Queen Mary University of London. Usuki, A., Kawasumi, M., Kojima, Y., Okada, A., Kurauchi, T., and Kamigaito, O. (1993) J. Mater. Res., 8, 1174. Alexandre, M. and Dubois, P. (2000) Mater. Sci. Eng. R, 28, 1. Pavlidou, S. and Papaspyrides, C.D. (2008) Prog. Polym. Sci., 33, 1119. Pinnavaia, T.J. and Beall, G.W. (2000) Polymer-Clay Nanocomposites, John Wiley & Sons, Inc., New York. Tjong, S.C. (2006) Mater. Sci. Eng. R, 53, 73. Zanetti, M., Lomakin, S., and Camino, G. (2000) Macromol. Mater. Eng., 279, 1. Bilotti, E., Deng, H., Zhang, R., Lu, D., Bras, W., Fischer, H.R., et al. (2010) Macromol. Mater. Eng., 295, 37. Bilotti, E., Zhang, R., Deng, H., Quero, F., Fischer, H.R., and Peijs, T. (2009) Compos. Sci. Technol., 69, 2587. Paul, D.R. and Robeson, L.M. (2008) Polymer, 49, 3187. Yuan, X.P., Li, C.C., Guan, G.H., Liu, X.Q., Mao, Y.N., and Zhang, D. (2007) J. Appl. Polym. Sci., 103, 1279. Yuan, X.P., Li, C.C., Guan, G.H., Mao, Y.N., and Zhang, D. (2008) Polym. Degrad. Stabil., 93, 466. Chen, Z.J., Luo, P., and Fu, Q. (2009) Polym. Adv. Technol., 20, 916. Tsai, T.Y., Li, C.H., Chang, C.H., Cheng, W.H., Hwang, C.L., and Wu, R.J. (2005) Adv. Mater., 17, 1769.

16 Choi, W.J., Kim, H.J., Yoon, K.H.,

17

18

19 20 21

22 23 24

25 26 27 28

29 30

Kwon, O.H., and Hwang, C.I. (2006) J. Appl. Polym. Sci., 100, 4875. Yin, M., Li, C.C., Guan, G.H., Yuan, X.P., Zhang, D., and Xiao, Y.N. (2009) Polym. Eng. Sci., 49, 1562. Martinez-Gallegos, S., Herrero, M., Barriga, C., Labajos, F.M., and Rives, V. (2009) Appl. Clay Sci., 45, 44. Kim, S.H. and Kim, S.C. (2007) J. Appl. Polym. Sci., 103, 1262. Chang, J.H., Kim, S.J., Joo, Y.L., and Im, S. (2004) Polymer, 45, 919. Hwang, S.Y., Lee, W.D., Lim, J.S., Park, K.H., and Im, S.S. (2008) J. Polym. Sci., Part B: Polym. Phys., 46, 1022. Yoo, E.S. and Im, S.S. (1999) J. Polym. Sci., Part B: Polym. Phys., 37, 1357. Chang, J.H., Mun, M.K., and Lee, I.C. (2005) J. Appl. Polym. Sci., 98, 2009. Vassiliou, A.A., Chrissaﬁs, K., and Bikiaris, D. (2010) Thermochim. Acta, 500, 21. Flynn, J. and Wall, L.A. (1966) Polym. Lett., 4, 323. Ozawa, T. (1965) Bull. Chem. Soc. Jpn., 38, 1881. Ozawa, T. (1970) J. Therm. Anal. Calorimetry, 2, 301. Hao, J.Y., Lu, X.H., Liu, S.L., Lau, S.K., and Chua, Y.C. (2006) J. Appl. Polym. Sci., 101, 1057. Nielsen, L. (1967) J. Macromol. Sci., Part A, 1, 929. Coltro, L., Padula, M., Saron, E.S., Borghetti, J., and Buratin, A.E.P. (2003) Packaging Technol. Sci., 16, 15.

123

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites by In-situ Reactive Polymerization Lawrence A. Pranger, Grady A. Nunnery, and Rina Tannenbaum

6.1 Introduction

Composites are an important class of materials deﬁned as having a matrix, which can be a metal, a ceramic, or a polymer that has been modiﬁed with a particulate phase of ﬁbers, whiskers, ﬂakes, or conductive ﬁllers such as carbon black. Composites are designed to take advantage of the most desirable characteristics of each constituent material while negating their undesirable properties. For example, polymer matrix composites offer a higher strength-to-weight ratio compared to a pure polymer [1]. Polymer matrix nanocomposites (PNCs) are an attractive new class of polymer matrix composites under intense development. A PNC consists of a polymer matrix, in which nanoparticles which are deﬁned as having at least one characteristic dimension, that is, length, width, or thickness in the range of 1–100 nm have been dispersed. Ideally, the nanoparticle phase is evenly dispersed in the matrix. A few examples of nanoparticles used in PNCs are metal nanoclusters, carbon nanotubes, nanoclays, and nanoﬁbers of cellulose, often called “whiskers” [2, 3]. Depending on the type of nanoparticles used, the physical properties of the resulting PNC may be superior to those achieved in corresponding conventional polymer composites or a pure polymer. Remarkable increases in tensile modulus and strength, glass transition temperature, ﬂame retardancy, and barrier properties (permeability) have been observed in PNCs with nanoparticle loadings of less than 5.0% wt [4, 5]. In some cases, the small size of the nanoparticles also affords improved optical properties, for example, transparency, or antibacterial properties [6, 7]. This makes PNCs attractive for a wide range of applications including automotive parts, barrier ﬁlms for food packaging, and materials where transparency or ﬂame retardancy is required [8]. In many cases, the enhanced properties of PNCs can be attributed to the high surface area of the nanoparticle phase. For example, nanoparticles of cellulose (cellulose “whiskers”) have a surface area in the range of 150–170 m2 g−1, while nanoclays (layered silicates) may have surface areas as high as 750 m2 g−1 [2, 9–11]. When well dispersed in the composite matrix, the high surface area of the nanoparticle phase leads to a signiﬁcant interfacial area between matrix and ﬁller, and this in turn In-situ Synthesis of Polymer Nanocomposites, First Edition. Edited by Vikas Mittal. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA.

124

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

leads to the immobilization of the matrix polymer at the nanoparticle surface [12]. This largely explains the increase in thermomechanical properties compared to an unﬁlled polymer [13]. Unfortunately, the high surface area of the nanoparticles also poses a signiﬁcant processing challenge during PNC fabrication. High particle surface area translates to high surface energy and low thermodynamic stability, and therefore, there is a strong tendency for the nanoparticles to agglomerate so the surface energy of that phase can be reduced. This, in conjunction with the high viscosity of the matrix polymer, makes it difﬁcult to produce stable, uniform dispersions of the nanoparticles throughout the matrix [12]. A uniform dispersion of the particulate phase is especially important for those PNCs in which the formation of a percolating network of nanoparticles is necessary for achieving desired mechanical, electrical, or thermal properties [5]. Inhomogeneities in particle size and particle dispersion reduce the mechanical properties of the composite and negate the unique advantages of PNCs. A second challenge is that the small size of the nanoparticles makes the characterization of these particles and their dispersion in the matrix much more difﬁcult compared to conventional ﬁllers. Hence, the high surface area of the nanoparticle phase can be regarded as the most desirable structural feature in PNCs, but at the same time also the least desirable processing parameter. This chapter explores the use of furfuryl alcohol (FA) both as the initial medium for dispersing nanoparticles of cellulose whiskers (CW) or montmorillonite (MMT) nanoclays and as the monomer precursor for in-situ polymerization of the PNC matrix. In general, nanoparticles prepared from biomass or from minerals possess an abundance of “built-in” surface functionality, and this can be exploited in the reactive molding approach to achieve their dispersion in PNCs. The premise underlying the reactive molding approach is that nanoparticles reﬁned from wood and clay naturally contain an abundance of surface functional groups. For example, the surfaces of nanoparticles of CW and MMT are richly hydroxylated [14, 15]. The natural surface functionality of these types of nanoparticles can be exploited to produce a stable dispersion of the nanoparticles in a polymerizable solvent medium, followed by in-situ polymerization, thus preserving their uniform dispersion in the PNC after matrix consolidation. A signiﬁcant advantage of this approach is that it avoids the use of surfactants. This is an important feature, because while the introduction of a surfactant may help to initially stabilize the nanoparticles against agglomeration, surfactant molecules may weaken the interface between particulate phase and PNC matrix, and hence lower composite strength in the consolidated PNC. Another major beneﬁt of the reactive molding approach is that it avoids the use of inert solvents, because the matrix precursor itself provides a suitable medium for the dispersion of the nanoparticles. This is important because by eliminating the need for solvent removal, closed molding processing can be used, and the molding process can be made more economical and environmentally sound. For this approach to work, monomers must be identiﬁed such that they would offer the right combination of (i) low viscosity, (ii) strong chemical afﬁnity for the functional groups at the nanoparticle surface, and (iii) reactivity toward in-situ polymerization. A fourth

6.2 Background

criterion, which is garnering increasing importance, is the preference for biobased monomer – nanoparticle systems which can be sourced completely from natural materials outside the petrochemical supply chain [16]. Therefore, this chapter explores the use of FA as a polymerizable liquid dispersant for nanoparticles of CW and MMT. FA, the hydrophilic monomer precursor of polyfurfuryl alcohol (PFA), is bio-based and available on an industrial scale [17]. CW nanoparticles were produced by acid hydrolysis of microcrystalline cellulose (MCC) and the whisker morphology and structure was characterized by atomic force microscopy (AFM) and by their birefringent optical properties when dispersed in a solvent. PNCs were then successfully produced by reactive molding of the CW–FA and MMT–FA systems. A combination of differential scanning calorimetry (DSC) and Fourier-transform infrared (FTIR) spectroscopy was employed to characterize the reaction chemistry of the MMT–FA and CW–FA systems, as well as their curing behavior. The thermal stability of cured CW–PFA and MMT– PFA nanocomposites was characterized by DSC and thermogravimetric analysis (TGA). The reactive molding of MMT clay nanocomposites was achieved by in-situ intercalative polymerization of FA inside the interlayer galleries of the MMT. The process of intercalation and exfoliation was studied using X-ray diffraction (XRD). Finally, the oxidative and nonoxidative degradation behaviors of CW–PFA and MMT–PFA nanocomposites were compared using TGA.

6.2 Background 6.2.1 Polymer Matrix Nanocomposites

Composites are an important class of materials deﬁned as having a matrix, which can be a metal, a ceramic, or a polymeric material, that has been modiﬁed with a particulate phase of ﬁbers, whiskers, ﬂakes, or ﬁllers such as talc, wood ﬂour, or carbon black [18]. Composites are designed to take advantage of the most desirable characteristics of each constituent material while reducing or eliminating their undesirable properties. For example, ﬁber-glass reinforced plastics exhibit a higher strength-to-weight ratio compared to a pure polymer, because the relatively low density of the polymer matrix is combined with the relatively high modulus and thermal stability of the glass ﬁber reinforcement. Conductive plastics can be produced by embedding a conductive ﬁller material such as carbon black or particles of a conductive polymer in an otherwise insulating polymer matrix [19]. In some cases, the main purpose of the ﬁller is to increase bulk at low cost. With the advent of PNCs, further increases in composite properties have been made possible. PNCs represent a new class of polymer matrix composites, which consist of a conventional polymer matrix in which nanoparticles have been dispersed to provide the desired increase in performance. Nanoparticles are deﬁned as particles having at least one characteristic dimension, that is, length, width, or

125

126

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

thickness in the range of 1– 100 nm [3, 10]. A few examples of nanoparticles used in PNCs are metal nanoclusters, carbon nanotubes, nanoclays, for example, MMT, and nanoﬁbers of cellulose, often called “whiskers” [7, 20–22]. The physical properties of PNCs are often found to be much superior to those achieved in a pure polymer or in the corresponding conventional polymer composites at equal ﬁller concentration. For example, PNCs reinforced with various nanoclays, nanotubes, or nanoﬁbers show remarkable increases in tensile modulus and strength (strength-to-weight ratio), heat deﬂection temperature, onset of degradation, glass transition temperature, ﬂame retardancy, and barrier properties (permeability) [23]. In some PNCs, the small size of the nanoparticles affords increased transparency or antibacterial properties. These enhanced features of PNCs are in addition to the corrosion resistance, noise dampening, and parts consolidation advantages offered by polymer matrix composites in general. This makes PNCs very attractive for a wide range of general applications, such as car parts, sporting goods, and ﬁlms for packaging. PNCs ﬁlled with conductive nanoparticles target a number of specialty applications, including antistatic coatings and packaging, electromagnetic shielding, and self-regulating heaters [24, 25]. In some PNCs, the nanoparticles are characterized by a high aspect (length-todiameter) ratio. For example, the aspect ratio of CWs is in the range of 50–100 [2, 5]. Such nanoparticles are able to form of a percolating network of rigid particles within the matrix above the percolation threshold concentration, which can be achieved at 1 wt% of ﬁller [5]. Signiﬁcant increases in thermomechanical or conductive properties observed in PNCs ﬁlled with high aspect ratio nanoparticles have been attributed to this percolation mechanism. In all PNCs, the nanoparticles are characterized by higher speciﬁc surface area as compared to conventional ﬁllers. Nanoparticles of cellulose (cellulose “whiskers”), have a surface area in the range of 150–170 m2 g−1, while nanoclays (layered silicates) may have surface areas as high as 750 m2 g−1 when exfoliated into individual platelets [9, 10, 26, 27]. This affords increased interaction with the polymer matrix, while also minimizing the weight fraction of a ﬁller which typically has a higher density than the matrix polymer. When well dispersed in the composite matrix, the high speciﬁc surface area of the nanoparticle phase leads to high interfacial area between matrix and ﬁller, and this in turn leads to the immobilization of the matrix polymer at the nanoparticle surface. The enhancement in thermomechanical properties observed in PNCs is generally attributed to this phenomenon. In this case, the thermomechanical performance of a PNC is governed by the strength and the nature of the interaction between the matrix and the surface of the particulate phase. As with conventional polymer matrix composites, efﬁcient shear transfer from the matrix to the particulate phase depends on strong chemical and/or physical adhesion between the polymer matrix and the particulate surface. At the particulate surface, a layer of an immobilized polymer is produced as a function of the work of adhesion [1, 13]. The thickness of this layer in turn inﬂuences the thickness of the interphase region. The interphase is deﬁned as the region that develops between the particle surface and those regions of the polymer matrix exhibiting bulk properties, as shown schematically in Figure 6.1.

6.2 Background

Interphase

Polymer Matrix

Nanoparticle

Figure 6.1 Schematic representation of the interphase in a polymer nanocomposite.

The morphology of the interphase in PNCs depends on the ﬂexibility of the matrix polymer chain and on the functional groups along the chain backbone. The interphase may occupy a high volume fraction of the PNC and, accordingly, dominate the thermomechanical properties of a PNC. For a well-dispersed particulate phase, the average inter-particle distance may approach the radius of gyration of a single polymer chain [12]. While the intricate interplay between interface and interphase is still not well understood and will likely be the topic of intense research for years to come, it is clear that the high speciﬁc surface area of the nanoparticles in a PNC highly accentuates the need to control the interface of the particulate phase in PNCs. This in turn calls for processing techniques capable of tailoring the surface chemistry and the physical structure at the interface between the particulate phase and the polymer matrix. The high speciﬁc surface area of the nanoparticles poses a signiﬁcant processing challenge during PNC fabrication. High particle surface area translates to high surface energy and low thermodynamic stability, and therefore there is a strong tendency for the nanoparticles to agglomerate so the surface energy of that phase can be reduced [12]. This, in conjunction with the high viscosity of the matrix polymer, makes it difﬁcult to produce a uniform dispersion of the nanoparticles in the polymer phase, and to maintain this dispersion during consolidation of the PNC matrix. Inadequate particle dispersion has the potential to reduce the high mechanical properties of the composite thereby negating one of the most sought-after advantages of PNCs. A uniform dispersion of the particulate phase is especially important for those PNCs in which the formation of a percolating network of nanoparticles is necessary for achieving desired mechanical, electrical, or thermal properties [2]. An additional challenge presented by PNCs at the processing stage is that the small size of the nanoparticles makes the characterization of these particles, and their dispersion in the matrix, much more difﬁcult compared to conventional ﬁllers. The need to neutralize the agglomeration tendency of nanoparticles in order to better solubilize, disperse, and stabilize the particulate phase in the polymer matrix

127

128

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

during PNC fabrication can be addressed in several different ways. For example, the surface chemistry of the nanoparticles can be modiﬁed to aid their dispersion. This can be achieved by adding surfactants to break up aggregates of nanoparticles [28]. This approach has a major drawback: while the use of surfactants may help to initially stabilize the nanoparticles against agglomeration, the surfactant molecules may weaken the interface between particulate phase and PNC matrix, and hence lower composite strength in the consolidated PNC. An alternate surface chemistry approach involves grafting polymers onto the particle surface ﬁller to increase compatibility with a matrix. Another approach that has been used to achieve the desired dispersion in cellulose whisker PNCs is the suspensions mixing technique [29, 30]. This technique is based on diluting the whiskers in an aqueous suspension, which is then combined with a polymer suspension, followed by ﬁlm casting. An intimate nanoparticle–polymer mixture can be achieved in this manner, which is preserved as the water is evaporated during matrix consolidation. A major drawback with this approach is that is only suitable for forming ﬁlms, and moreover, this technique requires solvent (water) removal, which may leave defects (voids) in the ﬁnal composite. A third approach, which has been used to produce with polymer-layered silicate clay PNC is the in-situ intercalative polymerization technique [4, 23]. This approach uses in-situ polymerization of a monomer or a oligomer inside the interlayer galleries of a 2 : 1 layered silicate to achieve exfoliation of the clay aggregates. One of the attractive features of this approach is that it does not require the use of any surfactants or solvents that must be evaporated during the matrix consolidation process. Rather the matrix precursor itself provides a suitable medium for the dispersion of the clay nanoparticles. The suspensions mixing and in-situ intercalative polymerization techniques are discussed in more detail in the following sections, in the context of cellulose whisker and layered silicate nanocomposites, respectively. 6.2.1.1 Cellulose Whisker Nanocomposites Polymer nanocomposites can be synthesized using cellulose in the form of cellulose nanocrystals or CWs. Pure cellulose is a biopolymer, speciﬁcally the polysaccharide of d-anhydroglucose units connected through the β-1,4-glycosidic ether bond [31], as shown in Figure 6.2. Wood is by far the primary source of cellulose, though it also occurs in plant ﬁbers and in the shells of tunicates (a sea animal), and is also produced by certain bacteria. Wood ﬁbers exhibit a complex, hierarchical composite structure, consist-

6 HOH2C

HO

1

5 2

O 4 3

HO

O OH

O HOH2C

Figure 6.2 Molecular structure of cellulose.

OH O O

n

6.2 Background

Figure 6.3 Hierarchical supramolecular structure of cellulose [33, http://www.bio.miami.edu/

dana/226/226F09_3.html].

ing of a matrix of amorphous lignin and hemicellulose, reinforced with semicrystalline ﬁbers of cellulose, as shown in Figure 6.3 [15, 32]. In the pulping process, most of the lignin and hemicellulose content is dissolved and removed, to allow the extraction of the cellulose ﬁbers from the wood. Cellulose ﬁbers extracted from wood by means of traditional pulping processes are suitable for papermaking, but they exhibit relatively poor mechanical properties compared to man-made ﬁbers, due to a high concentration of defects, and variability depending on the ﬁber source. The wall of a cellulose ﬁber in turn exhibits a complex, laminar structure, wherein each layer consists of smaller, unidirectional ﬁbers, or microﬁbrils, in the range of 5 to 50 nm wide, and anywhere from 100 nm to several microns long, depending on the source [33]. Each layer of microﬁbrils varies with respect to ﬁbril orientation (microﬁbril angle). Microﬁbrils in turn have a composite structure, consisting of slender cellulose crystallites, or “whiskers,” with diameters on the order of 5 nm, which are threaded together and embedded in the microﬁbrils between amorphous regions of cellulose and hemicellulose [34]. Hence, cellulose can be viewed as a composite material from the nanoscale perspective (whiskers in microﬁbrils), the microscopic perspective (microﬁbrils in ﬁbers), and the macroscopic perspective (ﬁbers in wood). Historically, the use of cellulose ﬁbers in polymer matrix composites has been mainly as a cheap ﬁller in the form of wood ﬂour [2]. Although wood ﬂour does increase the dimensional stability of a polymer composite, it does not signiﬁcantly increase the strength of the composite, due to its relatively low degree of cellulose ﬁber reﬁnement. In addition, cellulose is hydrophilic because of its many hydroxyl groups and ether links, and this leads to high moisture uptake. Moisture adsorption tends to lower the mechanical strength of cellulose composites over time,

129

130

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

because water has a plasticizing effect at the ﬁber–matrix interface [35, 36]. By chemically modifying the cellulose surface, moisture uptake can be decreased. For example, the hydroxyl groups at the cellulose surface can react with acetyl groups, thereby rendering the cellulose surface hydrophobic. The hydroxylated surface of cellulose also makes it incompatible with many low cost thermoplastic matrices such as polyethylene (PE), and polypropylene (PP), which are hydrophobic. However, the compatibility of cellulose with these polyoleﬁnes can be increased by reacting the hydroxyl groups with coupling agents, for example, diamines, diisocyanates, or silanes [36–39]. Grafting of matrix-compatible side chains, for example, maleic anhydride-modiﬁed polypropylene, is also effective for compatibilizing the cellulose ﬁber surface with polyoleﬁn matrices [38, 40, 41]. The extraction of microﬁbrils or whiskers from cellulose ﬁbers enables their use as a highly effective reinforcement in polymer matrix composites. Depending on the source and on the chemical treatment during extraction, the elastic modulus of microﬁbrils varies in the range of 70 to 140 GPa [2, 42, 43]. This compares favorably with other ﬁbers commonly used as reinforcement in high performance composites, for example, glass and aramid ﬁbers, and represents a dramatic improvement compared to regular wood and cellulose ﬁbers, the moduli of which typically does not exceed 40 GPa [32]. However, defects are still present in microﬁbrils in the form of regions of amorphous cellulose. Cellulose whiskers constitute the reinforcing element in the cellulose microﬁbrils. By carefully reﬁning the microﬁbrils, such as by acid hydrolysis, the amorphous cellulose can be removed and the remaining high aspect ratio, crystalline cellulose nanoﬁbers can be recovered. A suspension of CWs is shown in Figure 6.4. Strong intermolecular hydrogen bonding in CWs gives them a high degree of crystallinity and a high elastic modulus, making them highly suitable for their use

Figure 6.4 TEM image of a suspension of cellulose whiskers. Magniﬁcation factor is ×21 000. Horseshoe shaped border is the lacey carbon backing of the TEM grid.

6.2 Background

as a PNC reinforcement. The elastic modulus of a defect-free monocrystal of cellulose has been calculated to be at least 100 GPa with some estimates as high as 250 GPa [15, 42]. Cellulose whiskers are also characterized by a high aspect ratio, estimated in the range of 50–200 [2, 9, 44, 45], which enables them to attain a percolation threshold of as low as 1 wt% [2]. Above the critical percolation threshold concentration, CWs form a rigid, interpenetrating network throughout a polymer matrix. This mode of reinforcement is complementary to the effect of large surface area, which leads to a large interfacial area in the composite, typically 150–170 m2 g−1 [2, 9]. As discussed above, a general characteristic of PNCs is that immobilization of the matrix polymer at the nanoparticle surface leads to signiﬁcant increases in the thermomechanical performance of the PNC. A third mode of composite reinforcement which has been observed in PNCs both with polar matrices, for example, polyhydroxybutyrate (PHB) and starch, and nonpolar matrices, for example, PP and high density polyethylene (HDPE), is the ability of CWs to induce transcrystallinity. Transcrystallinity involves creating a crystalline interphase between the ﬁber surface and matrix [2, 9]. The cellulose ﬁber acts as a nucleation site for crystallization of the matrix polymer. Transcrystallinity enables signiﬁcant improvements in ﬁber–matrix bonding and hence interfacial shear strength and toughness. Several studies have focused on achieving a better understanding these different reinforcement mechanisms in cellulose whisker PNCs [9, 15, 38]. The modulus and strength of PNCs reinforced with whiskers compares favorably with composites reinforced with glass and aramid ﬁbers, aluminum, and magnesium alloy [32, 42]. At a cellulose whisker concentration of 6 wt%, high increases in modulus and elongation at break have been obtained for polystyrene– butylacrylate copolymer matrix composites [5]. PNCs have been produced using CWs both with thermoplastic matrices, for example, polyethylene oxide (PEO), plasticized polyvinyl chloride (PVC) and polycaprolactone (PCL), and with thermosetting phenolic and epoxy matrices [46, 47]. Signiﬁcantly, whiskers can be incorporated into biodegradable and biocompatible matrix materials, for example, starch, silk-ﬁbroin, and bacterial polyesters such as PHB [27, 42, 48, 49]. However, strong hydrogen bonding interactions between cellulose molecules tends to make the homogeneous dispersion of cellulose ﬁbers difﬁcult, especially in a nonpolar matrix. Dispersion of cellulose ﬁbers through melt processing is not an option because decomposition of the polymer occurs prior to melting [50, 51]. To date, good dispersion of cellulose microﬁbrils has been achieved only by the “dispersions mixing” method (also called “suspensions mixing”) [15, 30]. This involves mixing a dispersion of whiskers in a dilute aqueous suspension with a second suspension of matrix polymer, followed by ﬁlm casting. This technique produces an intimate mixture of cellulose and matrix polymer, which is preserved as the water is evaporated during matrix consolidation. High strength composite ﬁlms have been produced. A major drawback with this approach is that it is only suitable for forming composite ﬁlms, and that the consolidation of the polymer matrix requires volatilization and removal of the solvent phase, which may create defects (voids) in the ﬁnal product and poses economic and environmental concerns.

131

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

Tetrahedral Silica Layer Octahedral Alumina Layer Tetrahedral Silica Layer Interlayer H2O and Cations Layer Spacing ~2 nm

132

Figure 6.5 Structure of montmorillonite clay [53, http://mrsec.wisc.edu/Edetc/pmk/pages/

montmorillonite.html].

6.2.1.2 Layered Silicate Nanocomposites Another area within the ﬁeld of PNCs is the preparation of polymer layered nanocomposites using layered silicates, for example, MMT clay, as the nanoparticle phase. For MMT, the aspect ratio is in the range of 20–100, and the elastic modulus of a platelet has been estimated as approximately 270 GPa [52]. MMT is a layered silicate, belonging to the 2 : 1 phyllosilicate family, and consists of stacks of thin platelets [10]. Each individual platelet consists of an octahedral alumina sheet, sandwiched between two tetrahedral silica sheets (hence the 2 : 1 ratio). This is shown in Figure 6.5. The platelets have a net negative charge due to substitution of some of the Al3+ cations with Mg2+ ions. This charge is counterbalanced by inorganic cations, for example, Na+ conﬁned to interlayer galleries between platelets. This causes platelets to form stacks (“tactoids”) held together by electrostatic forces. In PNCs, the goal is to produce a structure in which the tactoids are intercalated and expanded like an accordion, or exfoliated to individualize the platelets. The lamellar morphology of MMT clay makes it an ideal additive for bulk composites and ﬁlms where high barrier properties are required. The morphology of exfoliated MMT PNCs is often modeled as a labyrinth, providing an excellent barrier for mass transport, retarding diffusion through the matrix. To date, the in-situ intercalative polymerization technique is one of the most successful techniques for producing MMT nanocomposites on an industrial scale. The ﬁrst step of this process, as shown in Figure 6.6, involves the intercalation of a monomer or low-molecular-weight precursor into the galleries of the silicates.

6.2 Background

Initial Mixing

Intercalation

In-situ polymerization

Exfoliation

Figure 6.6 Steps in the in-situ intercalative polymerization technique.

After the intercalation step, in-situ polymerization of the intercalated species inside the galleries gradually forces apart the individual silicate platelets. The ﬁnal result of in-situ intercalative polymerization is an exfoliated structure in which the individual silicate layers are sequestered and surrounded by the matrix polymer, or an intercalated structure in which the polymer chains are partially immobilized within the galleries of the silicate. Often, mixed structures are obtained [4]. In in-situ intercalative polymerization, the surface chemistry of the clay surface is selected for compatibility with the intercalating monomer. Naturally occurring MMT is hydrophilic. However, the silicate surface can be rendered organophilic by ion exchange with, for example, onium ions with aliphatic or aromatic functions, for compatibility with an organophilic intercalate. A range of organomodiﬁed MMT is commercially available. PNCs modiﬁed with nanoclays show remarkable increases in thermomechanical properties. The clay–nylon PNC Toyota Research produced for under-the-hood applications in the Toyota Camry is a good example. By incorporating a small amount of MMT clay into a nylon matrix, the heat distortion temperature of the composite increased by 87 °C, allowing its use in under-the-hood automobile components [54]. 6.2.2 Reactive Molding Techniques for Composite Manufacture

The main distinctive feature of reactive molding is that the complete polymerization of the entire polymer matrix is effected in the mold, starting from a system of monomers or other low-molecular-weight precursors. Reactive molding differs from conventional molding of thermoplastics where a polymer is introduced to a mold, after a process of melt blending at high viscosity, and simply allowed to solidify in the mold. It also differs from molding of thermosetting materials where a resin is mixed with a monomer like styrene, which acts at ﬁrst as a solvent to reduce viscosity, and later as a crosslinking agent during curing in the mold. The most common reactive molding process today is reaction injection molding (RIM). The RIM process is typically a two-component system based on

133

134

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

the polymerization of a diisocyanate with a diol, plus a small amount of polyol for crosslinking, and a catalyst [55]. The components are mixed together by impingement in a mixing head, followed by injection into a mold. In the mold, polymerization occurs very rapidly as the hydroxyl groups of the polyol component react with isocyanide groups to form urethane linkages. To extend the use of RIM to composite applications, structural reaction injection molding (SRIM) and reinforced reaction injection molding (RRIM) processes have been developed [56, 57]. RIM chemistry is suitable for incorporation of cellulosic ﬁllers, and to an extent, cellulose ﬁbers are used in RRIM composites [58]. The low abrasion of cellulose ﬁbers compared to inorganic ﬁllers, for example, chopped glass ﬁber, makes them especially suitable for RRIM processing [38]. In addition, cellulose is light weight, affordable, renewable, environmentally benign, and biodegradable. In general, one of the main limitations for the use of cellulose in polymer matrix composites is the strongly hydrophilic nature of cellulose, which causes problems with moisture absorption, as discussed above [2, 35, 36]. However, when using cellulose as a ﬁller for RRIM composites, the hydroxyl groups at the cellulose surface can be exploited to increase the ﬁber–matrix interaction and enhance composite strength. Cellulose ﬁber reinforcement has been shown to double the modulus of SRIM composites at only 4 wt%, due to its ability to provide hydroxyl groups for crosslinking reactions with isocyanate groups in the matrix [59]. 6.2.2.1 Materials and Methods for Reactive Molding of Nanocomposites For processing of PNCs, reactive molding offers several important advantages over competing processes. For example, in conventional melt blending of thermoplastics, it is difﬁcult to disperse the particulate phase in the polymer, and to completely wet the surface of the particulate phase due to the high viscosity of the polymer melt. In contrast, the low initial viscosity of a reactive molding system is conducive to achieving a dispersion of nanoparticles in the monomer phase, without having to add any solvents to lower the viscosity of the system. Moreover, reactive molding processes are closed-mold processes, fact which avoids the environmental concerns associated with open mold processes. In-situ intercalative polymerization of layered silicates is perhaps the best example of reactive molding of nanocomposites today. In-situ interactive polymerization of layered silicates, which was discussed above, can be achieved either with thermosetting matrices, such as polyurethane and epoxy, or with thermoplastic systems, such as nylon-6 [4, 23]. A general requirement for reactive molding of nanocomposites is that the particulate phase of a PNC is compatible with the monomer phase of the reactive molding system, which acts as a polymerizable solvent. This makes it possible to achieve and maintain a ﬁne dispersion of the particulate phase in the monomer during matrix consolidation, resulting in excellent particle distribution in the ﬁnal PNC. Above, it was noted that the hydroxylated surface of cellulose makes it reactive to isocyanate. Cellulose whiskers may therefore represent the ideal particulate phase for a “nano-RIM” process. For this to be achieved, the whisker–polyurethane system needs to be better characterized, so that the RIM process can be adapted to fabrication of cellulose whisker PNCs.

6.2 Background

Stereolithography, which has also been referred to as “UV-RIM” (i.e., ultraviolet light induced reaction injection molding), is a second reactive molding process, which has found application in rapid prototyping [60, 61]. The low viscosity of a UV-RIM resin should also be conducive to achieving a dispersion of nanoparticles in the PNC matrix. Although a range of reactive monomers could be selected for reactive molding of polymer nanocomposites, bio-based monomers with the potential to replace materials traditionally derived from the petrochemical supply chain are of particular interest today. On the one hand, the Technology Roadmap for Plant/CropBased Renewable Resources 2020 (sponsored by the U.S. Department of Energy), calls for 10% of basic chemical building blocks to be plant-derived by 2020 [16]. On the other hand, the Forest Products Industry’s Agenda 2020 identiﬁes the critical importance of nanotechnology in the development of new generations of high-value, high-performance materials from forest-based products that would compete favorably with the properties of conventional, petroleum-based materials. One such polymerizable solvent – monomer, which satisﬁes these various requirements for a reactive molding system, is FA. 6.2.2.2 Furfuryl Alcohol as a Precursor for Polymer Matrix Composites FA is a bio-based material, produced by hydrogenation of furfural on an industrial scale. Furfural has been prepared in commercial quantities for many decades from pentose-rich agricultural residues, including rice hulls, bagasse, oat hulls, and corn cobs. Furfural can also be derived from wood and wood products, which represent a second natural storehouse for furfural [62]. The resiniﬁcation of FA was reported on as early as 1873 [63]. The resonance structures of the furan ring make it susceptible to substitution at the C2 and C5 “alpha” positions, and hence FA behaves like a difunctional monomer. In the early stages of resiniﬁcation, FA polymerizes predominantly through “head-to-tail” reactions, in which the hydroxymethylene function of one molecule reacts with the hydrogen at the C5 position of the furan ring of a second molecule, yielding a methylene bridge. This is shown in Figure 6.7. “Head-to-head” condensation between hydroxymethylene groups, yielding dimethylene ether bridges, also occurs at this stage. In addition, levulinic acid

(n+1) O

H2 C OH

H+ O

H+

H2 C

O

H2 C

O (n-1)

H2 C OH

Figure 6.7 Initial condensation step in the homopolymerization of furfuryl alcohol.

135

136

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

and lactone byproducts may be produced as a result of hydrolytic ring cleavage of furan rings [63]. Crosslinking occurs in the later stages of resiniﬁcation. Though still not well understood, the crosslinking mechanism during FA polymerization, most likely involves condensation between the methylol group at the end of one chain and the methylene bridge of second chain. Reactive hydrogen in unsaturated sequences along the chain backbone may play a mediating role. A second crosslinking mechanism involves formaldehyde, which is split off when dimethyl ether bridges decompose to methylene bridges [64–66]. The polymerization of FA to PFA is typically catalyzed using a Brönstedt acid, for example, sulfuric, hydrochloric, or para-toluene sulfonic acid (PTSA) [66]. When using strong mineral acids such as these, overcatalysis may lead to an explosive reaction. To avoid this hazard, polymerization can alternately be effected in the presence of γ-alumina, owing to its Lewis acidity. Common applications of PFA include adhesives, corrosion-resistant coatings and composites, and a binder material in foundry cores. Many applications of PFA owe to the thermal stability of the aromatic furan ring, which gives it high inherent ﬂame retardancy, low smoke release and high char yield. While PFA is hydrophobic, its precursor, FA, is hydrophilic and completely soluble in water, due to the hydroxyl group of the side chain and the oxygen heteroatom of the furan ring. When FA is polymerized in the presence of wood pulp, the result is that FA mostly phase separates from the cellulose phase, producing a mixture of free cellulose and crosslinked PFA [67]. However, grafting of PFA to pulp and ﬂax ﬁbers has been achieved through a process involving free radical polymerization [66, 67]. When polymerized by acid catalysis in the presence of ﬁne silica particles, FA forms a crosslinked polymer that is attached covalently to the silica surface. The PFA coverage is dense enough to render the silica hydrophobic [68]. Finally, it is also worth noting that the polymerization of FA to PFA can be achieved using UV radiation. The UV-photopolymerization of FA has been investigated for its potential application as a resin for stereolithography [69, 70]. As with conventional RIM, UV-RIM requires low viscosity precursors, and hence FA, which is a clear liquid of low viscosity at room temperature, lends itself to this application.

6.3 Experimental Procedures 6.3.1 Reactive Molding of Cellulose Whisker Nanocomposites 6.3.1.1 Conceptual Approach The overall objectives of the reactive molding of cellulose whisker nanocomposites were to (i) disperse CWs in a FA monomer, and (ii) to achieve in-situ polymerization of the CW–FA dispersion using FA as a polymerizable solvent medium.

6.3 Experimental Procedures

Due to the richly hydroxylated surface of CW, FA was chosen for its ability to compete with hydrogen bonding between whiskers, and because it is bio-based. Cellulose whiskers were prepared by hydrolysis with sulfuric acid, and hence, the surface of such whiskers contains sulfonic acid groups, that is, –HSO3 [9, 11, 71]. These residual acid groups represent catalytic sites for polymerization of FA in close contact with the whiskers surface. The sulfonic acid groups on the whiskers surface also help to electrostatically stabilize the CW against agglomeration while in suspension during nanocomposite consolidation. To characterize the curing behavior and to investigate how the presence of CW inﬂuences the polymerization of FA, FTIR, and DSC, spectra were collected before and during the resiniﬁcation process. As a qualitative measure of CW dispersion in the cured PNC, coded CW–PFA, TGA was performed to compare the thermal stability of the PNC compared to a pure polymer. 6.3.1.2 Preparation of CW Cellulose whiskers were prepared starting from MCC precursor (Avicel, Aldrich). The main steps in the whiskers preparation are as follows: (i) acid hydrolysis (62% H2SO4, 12.5 ml g−1 cellulose, 55 °C, 2 h) which dissolves most of the amorphous cellulose, leaving only the crystalline whiskers intact; (ii) reﬁning and puriﬁcation of the solid residue remaining after the hydrolysis by repeated cycles of ultracentrifuging followed by resuspension and washing of the solids in distilled water until a turbid supernatant is obtained; (iii) dialysis against distilled water to pH 5–6, and (iv) freeze drying. The freeze-dried whiskers can then be transferred to the FA monomer phase. To conﬁrm that this procedure yields CW with the desired nanoscale morphology, samples of the aqueous suspensions were slowly concentrated by open evaporation to the range 2.7–3.5% and viewed through crossed polars. The concentrated suspension initially showed ﬂow birefringence, and with increased concentration, the suspension exhibited birefringence also at rest [71]. 6.3.1.3 Resiniﬁcation of FA with CW To determine whether CW has a catalytic effect of the resiniﬁcation of FA, in a typical experiment, 0.75–1.0 phr freeze-dried CW was redispersed in FA by means of a brief ultrasonication treatment (Fisher 500 W, 25–50% amplitude, 15 min bursts) followed by heating to 50 °C for additional dispersion. At 50 °C, the sulfonic acid residues are stable and de-esteriﬁcation is not expected. After 1 h at 50 °C, the mixture was heated to the target reaction temperature until resiniﬁcation of FA to PFA occurred. To characterize the curing behavior and to investigate how the presence of CW inﬂuences the polymerization of FA, FTIR, and DSC, spectra were collected before and during the resiniﬁcation process. 6.3.1.4 Curing of CW–PFA Composites Oven-curing of the CW–PFA resins was performed in two stages, ﬁrst at 130 °C for 75 min and then at 210 °C for 105 min. Oven-cured samples showed no residual

137

138

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

cure exotherm when scanned by DSC, but only a broad decomposition peak above 250 °C with a peak maximum at 360 °C. 6.3.1.5 Characterization Techniques To estimate the aspect ratio of the whiskers, AFM (close contact mode) was performed on dried ﬁlms of dilute CW suspensions. Imaging was performed in close contact mode using silicon tips, on a Paciﬁc Nanotechnology scanner, varying the scanning frequency between 0.5 and 1 Hz, and z-setpoint as required for each scan. Conventional transmission electron microscope (TEM) analysis was not successful due to low contrast. However, the dispersion of CW in its initial aqueous suspension after neutralization and dialysis against distilled water was successfully imaged by cryo-TEM. Cryo-TEM samples were prepared by ultrafast cooling with liquid N2 to the vitreous state, and maintained in that state during the TEM imaging process with the use of a cooling-holder system [72]. The samples were imaged in bright ﬁeld mode using low electron dose in a Technai (formerly Philips) G2 TEM, at a 120 kV accelerating voltage. To characterize the curing behavior and to investigate how the presence of CW inﬂuences the polymerization of FA, FTIR spectra were collected on a Nexus 870 (Thermo Electron) in transmission mode. An average of 50 scans per analysis was collected, with a resolution of 2 cm−1. In FTIR, the polymerization of FA is normally accompanied by signiﬁcant increases in the intensity of the peak at 1562 cm−1, assigned to the skeletal vibration of 2,5 disubstituted furan rings [63, 65]. At the same time, there is a decrease in peak intensity of the broad peak in the hydroxyl stretching region (3200–3600 cm−1) and of the peak at 3120 cm−1, assigned to the in plane stretch of the hydrogen at the C5 position of the furan ring [73]. The intensity of the sharp peak at 1504 cm−1 is assigned to the ring stretching of the furan ring and is prominent both in FA and PFA spectra. To characterize the thermal stability of PNCs, TGA data was collected on a TA Instruments Q50, at a heating rate of 10 °C min−1. Samples were heated up to 800 °C under a ﬂow of 25 ml min−1 nitrogen to study nonoxidative degradation, and under a ﬂow of 25 ml min−1 air to study oxidative degradation. 6.3.2 Reactive Molding of MMT Nanocomposites 6.3.2.1 Conceptual Approach The overall objectives of the reactive molding of MMT nanocomposites are: (i) to disperse MMT clay in a FA monomer and (ii) to achieve in-situ intercalative polymerization of FA inside the interlayer galleries of the MMT clay to effect complete exfoliation of the MMT platelets. Due to the richly hydroxylated surface of MMT, FA was chosen for its ability to compete with hydrogen bonding between platelets, and because it is bio-based. In addition, FA has water-like viscosity at room temperature, which facilitates penetration of FA into the interlayer galleries of MMT. The MMT surface contains an abundance of coordinatively unsaturated Al3+ edge sites which act as Lewis acids and a surface high in Brönstedt acidity. Because the polymerization of FA to PFA

6.3 Experimental Procedures

is catalyzed by acids, the acidic sites on the surface of MMT can be exploited to achieve in-situ interactive polymerization of FA to PFA in close proximity to the MMT surface facilitating exfoliation and platelet dispersion [14]. The polymerization of FA is typically catalyzed using a Brönstedt acid for example, sulfuric, hydrochloric, or PTSA. When using strong mineral acids such as these, overcatalysis may lead to an explosive reaction. This hazard is avoided by using MMT, since MMT-catalyzed polymerization of FA is mild. Polymerization of FA to PFA can also be effected in the presence of γ-alumina, which has Lewis acid sites in the form of highly acidic hydroxyl groups [74]. 6.3.2.2 Types of MMT Clays Used Two types of MMT – sodium montmorillonite (Cloisite® Na, Southern Clay) and an organomodiﬁed montmorillonite (Cloisite® 30B) – were used to prepare MMT– PFA nanocomposites, denoted as NaMMT–PFA and 30BMMT–PFA, respectively. The organic modiﬁer in Cloisite® 30B is methyl, tallow, bis-2-hydroxyethyl, quaternary ammonium, where tallow is 65% C18, 30% C16, and 5% C14. 6.3.2.3 Resiniﬁcation of FA with MMT Clay MMT–PFA resins were produced by charging FA (99% pure, Sigma-Aldrich), and 10 phr MMT (actually, in this case, parts per hundred FA) to a round bottom ﬂask, homogenizing the mixture by vigorous agitation with a magnetic stir bar, and slowly heating to the target reaction temperature of 150 °C. Preliminary DSC surveys indicated that MMT possesses sufﬁcient catalytic activity (attributed to Lewis acidity) to effect the polymerization of FA to PFA above approximately 140 °C. Therefore, the reaction mixture was maintained at a reaction temperature of 150 °C until resiniﬁcation of FA to PFA occurred. To provide a basis for comparison with PFA modiﬁed with CW and MMT, PFA resin without CW or MMT was prepared using 3 phr γ-alumina (nanopowder, Aldrich) as a catalyst. The resin obtained after reaction at 100 °C for 12 h (denoted γ-Al-PFA) was stable during storage (at room temperature) for several months. To characterize the curing behavior and to investigate how the presence of MMT inﬂuences the polymerization of FA, FTIR, and DSC, spectra were collected before and during the resiniﬁcation process. To monitor the exfoliation and polymerization process, XRD data was collected before and during the resiniﬁcation process. 6.3.2.4 Curing of MMT–PFA Composites Oven-curing of the MMT–PFA resins was performed in two stages, ﬁrst at 130 °C for 75 min and then at 210 °C for 105 min. Oven-cured samples showed no residual cure exotherm when scanned by DSC, but only a broad decomposition peak above 250 °C with a peak maximum at 360 °C. 6.3.2.5 Characterization Techniques The morphology of the MMT used in the preparation of MMT–PFA nanocomposites was examined by AFM, by drying dilute aqueous suspensions on a smooth silicon wafer. Imaging was done in close contact mode using silicon tips, on a

139

140

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

Paciﬁc Nanotechnology scanner, varying the scanning frequency between 0.5 and 1 Hz and z-setpoint as required for each scan. To monitor the degree of intercalation and exfoliation of the MMT at various stages of resiniﬁcation, XRD was used. XRD data was collected on a Rigaku Miniﬂex diffractometer, at a scanning rate of 0.5 ° per second using Cu Kα radiation with a wavelength of λ = 1.54 Å. The change in the d001 spacing (i.e., the basal spacing) of the MMT tactoids was calculated from the position of the diffraction peak using Braggs law, 2dsinθ = λ. The interlayer gallery spacing was then estimated by subtracting 10 Å (the thickness of a single platelet) from the calculated basal spacing [10]. FTIR data was collected on a Thermo Nicolet Nexus 870 unit, using ZnSe windows. A minimum of 50 scans were collected at a resolution of 2 cm−1. In FTIR, the polymerization of FA is normally accompanied by signiﬁcant increases in the intensity of the peak at 1562 cm−1, assigned to the skeletal vibration of 2,5-disubstituted furan rings. At the same time, there is a decrease in peak intensity for the broad peak in the hydroxyl stretching region (3200–3600 cm−1) and for the peak at 3120 cm−1, assigned to the in plane stretch of the hydrogen at the C5 position of the furan ring. The intensity of the sharp peak at 1504 cm−1 is assigned to the ring stretching of the furan ring, and is prominent both in FA and PFA spectra. The Al–O stretching vibration from MMT can also be observed in the FTIR spectrum at approximately 520 cm−1. TGA data was collected on a TA Instruments Q50, at a heating rate of 10 °C min−1. Samples were heated up to 800 °C under a ﬂow of 25 ml min−1 nitrogen to study degradation under nonoxidative conditions. A second series of samples were heated under a ﬂow of 25 ml min−1 air to study degradation under oxidative conditions.

6.4 Results and Discussion 6.4.1 Reactive Molding of Cellulose Whisker Nanocomposites

This section discusses the results of reactive molding of CW nanocomposites using FA as a polymerizable solvent medium to produce CW–PFA nanocomposites. Cellulose whiskers are not commercially available, and therefore, they were prepared by hydrolysis of MCC with sulfuric acid. The preparation of the CW was followed by their thorough morphology characterization, and ﬁnally, by the polymerization of FA to PFA in their presence. To characterize the polymerization behavior and to investigate how the presence of CW inﬂuences the polymerization of FA, FTIR spectra were collected before and during the resiniﬁcation process. Finally, characterization of the thermal stability of the CW–PNC, as measured by TGA, is discussed and compared to the pure polymer. The results provide a useful qualitative measure of the CW dispersion in the cured PNC.

6.4 Results and Discussion

6.4.1.1 Morphology of CW The CW exhibited birefringence when viewed through crossed polars. Figure 6.8 shows the ﬂow birefringence (top) and the birefringence at rest (bottom) of a concentrated suspension of CWs [71]. The birefringence of the CW suspensions is attributed to their characteristic liquid crystalline organization in suspension, and proves that whiskers have been successfully reﬁned from the MCC precursor. The dispersion of CW is also evident from the gallery of TEM and AFM images in Figures 6.9–6.13. Figures 6.9–6.10 a)

b)

Figure 6.8 Birefringence of suspensions of cellulose whiskers, viewed through crossed polars:

(a) ﬂow bifringence, and (b) birefringence at rest.

Figure 6.9 Cryo-TEM image of a suspension of cellulose whiskers. Magniﬁcation factor is

×30 000.

141

142

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

Figure 6.10 Cryo-TEM image of a suspension of cellulose whiskers. Magniﬁcation factor is

×21 000.

are TEM images of a cellulose suspension, performed by the cryo-TEM method. This method used liquid N2 to freeze a sample of the CW suspension and maintain the frozen state throughout the TEM imaging process [72]. The results therefore prove that the CWs are well dispersed and stable in suspension. Figures 6.11–6.13 show AFM images of a ﬁlm of dried cellulose whiskers. Like the TEM images, the AFM images show that the CWs are well dispersed. From these ﬁgures, it can be seen that the average diameter of the CWs is around 10 nm, and that the aspect ratio is in the range of 50–100. This is consistent with the literature [9, 42]. 6.4.1.2 Resiniﬁcation of FA in the Presence of CWs CW naturally has a richly hydroxylated surface, but because sulfuric acid is used in the CW preparation, the CW surface subsequently acquires sulfonic acid residues originating from the hydrolysis step. The sulfonic acid groups carry a negative charge, which is critical for providing electrostatic stabilization of the CW dispersion in the FA [9, 11]. However, the esteriﬁcation reaction is reversible, and the deesteriﬁcation of sulfonic acid groups from the CW surface at elevated temperature can be exploited to catalyze in-situ polymerization of FA in close proximity to the CW [75]. By wrapping PFA around the CW, the initial dispersion can be preserved in the PFA resin and in the cured nanocomposite. To explore this approach, 1.0 phr freeze-dried CW was dispersed in FA by means of a brief ultrasonication treatment (Fisher 500 W, 25–50% amplitude, 15 min.) followed by heating to 50 °C for 1 h. At this temperature, the CW–FA mixture was stable, and

6.4 Results and Discussion

20.0

Line Scan Height (nm)

18.0 16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5

Line Scan Length (um)

Figure 6.11 AFM phase contrast images of cellulose whiskers from acid hydrolysis of

microcrystalline cellulose.

the rate of polymerization was negligible. However, increasing the reaction temperature to the 60–100 °C range triggered the polymerization of FA. Within 2–3 min at 100 °C, the mixture turned from yellow to dark brown due to rapid accumulation of PFA particles and within 2 h the gel point was reached. After 9 h at 60 °C or 4 h at 75 °C, a ﬂuid resin was obtained. Under identical experimental conditions, neither MCC nor silica gel induced the polymerization of FA, demonstrating that

143

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

16.0

16.0

14.0

14.0 Line Scan Height (nm)

Line Scan Height (nm)

144

12.0 10.0 8.0 6.0 4.0 2.0 0.0 0.0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 3.2 3.6 4.0 4.4 4.8 Line Scan Length (um)

12.0 10.0 8.0 6.0 4.0 2.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 Line Scan Length (um)

Figure 6.12 AFM phase contrast images of cellulose whiskers from acid hydrolysis of

microcrystalline cellulose.

neither the ultrasonication treatment, nor the presence of a high, hydroxyl-rich surface area is sufﬁcient to explain the rapid polymerization of FA in the presence of CW. Rather, it is the sulfonic acid residues at the CW surface that are responsible for triggering the polymerization. This mechanism is shown schematically is Figure 6.14. At room temperature and up to 50 °C, the sulfonic acid groups at the CW surface are stable. However, increasing the temperature to the 60–100 °C range leads to the de-esteriﬁcation of the sulfonic acid groups, detaching them from the CW surface, and allowing them to combine with H2O from the condensation polym-

14.0

14.0

12.0

12.0

10.0 8.0 6.0 4.0 2.0 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 Line Scan Length (um)

Line Scan Height (nm)

Line Scan Height (nm)

6.4 Results and Discussion

145

10.0 8.0 6.0 4.0 2.0 0.0 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 Line Scan Length (um)

Figure 6.13 AFM phase contrast images, with line scans, of cellulose whiskers from acid

hydrolysis of microcrystalline cellulose.

erization of FA and reform sulfuric acid [75–78]. The sulfuric acid then catalyzes the polymerization of FA in close proximity to, and around the CW, trapping the ﬁne CW dispersion in the PFA resin. The rapid progress of resiniﬁcation is evident from the FTIR spectra, shown in Figure 6.15, for FA with 0.75 phr CW after 1 h at 50 °C (spectrum a) and after reacting the CW–FA mixture at 100 °C for 30 min, 1 h, and 2 h (spectra b, c, and d, respectively). The spectrum for γ-Al-PFA after 12 h resiniﬁcation at 100 °C (spectrum e) is included for comparison. The polymerization of FA is typically accompanied by signiﬁcant increases in the intensity of the peak at 1562 cm−1, assigned

146

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

CW

FA

O-SO3Initial CW – FA Mixing

>50oC O-SO3-

OH + H2SO4

Desulfonation at CW surface, forming H2SO4

PFA

Polymerization of FA catalyzed by H2SO4 from CW surface Figure 6.14 Proposed mechanism for polymerization of furfuryl alcohol in the presence of

cellulose whiskers (Adapted with permission from Macromolecules 2008, 41(22), 8682–8687. Copyright 2008, American Chemical Society).

to the skeletal vibration of 2,5 disubstituted furan rings, and at 1712 cm−1, assigned to the C=O stretch of γ-diketones formed from hydrolytic ring opening of some of the furan rings along the PFA chain [63, 65]. At the same time, there is a decrease in peak intensity (not shown) for the broad peak in the hydroxyl stretching region (3200–3600 cm−1) and for the peak at 3120 cm−1, assigned to the in plane stretch of the hydrogen at the C5 position of the furan ring [73].

6.4 Results and Discussion

Rel. Absorbance (a.u.)

e)

d) c) b) a)

1800

1750

1700

1650

1600 1550 1500 Wavenumbers (cm–1)

Figure 6.15 FTIR spectra for the

polymerization of FA in the presence of CWs: (a) suspension of CW in FA monomer at 50 °C for 1 h; (b) CW–PFA resin after reaction at 100 °C for 30 min.; (c) CW–PFA resin after reaction at 100 °C for 1 h; (d) CW–PFA resin

Table 6.1

1450

1400

1350

1300

after reaction at 100 °C for 2 h; and for comparison, (e) resiniﬁcation of FA with γ-alumina (γ-Al) at 100 °C for 12 h (Adapted with permission from Macromolecules 2008, 41(22), 8682–8687. Copyright 2008, American Chemical Society).

Internal referencing of the 1562 and 1712 cm−1 peaks from Figure 6.15.

Peak height ratio I/Iref

γ-Al-PFA 12 h/100 °C

CW–PFA 0.5 h/100 °C

CW–PFA 1 h/100 °C

CW–PFA 2 h/100 °C

I1562/I1500 I1712/I1500

0.527 0.424

0.465 0.455

0.663 0.697

1.107 1.349

Reprinted with permission from Macromolecules 2008, 41(22), 8682–8687. Copyright 2008, American Chemical Society).

The intensity of the sharp peaks at 1504 cm−1, assigned to ring stretching of the furan ring serves as an internal reference peak for semiquantitative analysis. Table 6.1 shows the results of applying the internal referencing method to the peaks at 1562 and 1712 cm−1 for the spectra in Figure 6.15. Internal referencing of the 1562 cm−1 peak to the 1500 cm−1 peak indicates that in less than 1 h at 100 °C, the degree of polymerization of the CW–PFA resin is comparable to γ-Al-PFA resin after 12 h resiniﬁcation at 100 °C. This is also apparent from internal referencing of the 1712 cm−1 peak to the 1500 cm−1 peak. Figure 6.16 compares the FTIR spectra for CW–PFA with pure FA and with γAl-PFA. In the 1300–1800 cm−1 spectral region, it can be seen that the growth of the γ-diketone peak at 1710 cm−1 with increasing polymerization is characteristic both in the γ-Al-PFA system catalyzed by a Lewis acid (in this case, γ-Al ) and the CW–PFA system catalyzed by the sulfonic (Brönstedt) acid residues originating from the whiskers.

147

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

Rel. Absorbance (a.u.)

a)

1800

c) b)

a)

1750

1700

1650

b)

Rel. Absorbance (a.u.)

148

1600 1550 1500 Wavenumbers (cm–1)

1450

1400

1350

1300

c) b) a)

900

850

800

750 700 650 Wavenumbers (cm–1)

600

550

500

Figure 6.16 FTIR spectra comparing the polymerization of FA to PFA in the 1300–1800 cm−1

(top) and the 500–900 cm−1 (bottom) spectral regions for (a) pure FA; (b) γ-al-PFA resin; (c) CW–PFA resin.

6.4.1.3 Thermal Resistance of CW–FA Nanocomposites When dispersed in a composite matrix, the high surface area of the CW ﬁller translates to a high interfacial area between matrix and ﬁller, leading to the immobilization of the matrix polymer at the nanoparticle surface. As a result, signiﬁcant enhancements in thermal and mechanical performance are expected for CW PNCs as compared to the pure polymer matrix. Therefore, TGA provides a useful qualitative measure of CW dispersion in the cured PNC. Figure 6.17 shows the thermal proﬁles for nonoxidative degradation in a nitrogen ﬂow of γ-Al-PFA (proﬁle a) and CW–PFA (proﬁle b) composite samples. These were prepared by oven-curing at 130 °C for 75 min and then at 210 °C for 105 min. For comparison, the spectra for the MMT–PFA nanocomposites (MMT–PFA PNCs will be discussed in detail in Section 4.2) are also shown. These are coded 30B-MMT PFA (proﬁle c) and NaMMT–PFA (proﬁle d).

6.4 Results and Discussion 100 b))

95 90

Weight

85

a)

80 75 70 65

c)

60

d)

55 50 50 100 150 200 250 300 350 400 450 500 550 600 650 700 Temperature (°C)

Figure 6.17 TGA scans collected at 10 °C

min−1 showing onset of degradation for the case of nonoxidative degradation in N2 for cured PNCs of (a) γ-Al-PFA; (b) CW–PFA;

(c) 30BMMT–PFA; and (d) NaMMT–PFA (Adapted with permission from Macromolecules 2008, 41(22), 8682–8687. Copyright 2008, American Chemical Society).

Table 6.2 Temperature at the onset of decomposition and weight retention at 500 and 800 °C for cured PFA nanocomposites.

γ-Al-PFA

CW–PFA

Onset of degradation (5% weight loss) in N2 ﬂow 246 °C 323 °C

NaMMT–PFA

30BMMT–PFA

302 °C

295 °C

Weight retained after nonoxidative degradation in N2 500 °Ca) 59% 65% 67% 800 °Cb) 48% 52% 55%

70% 60%

Reprinted with permission from Macromolecules 2008, 41(22), 8682–8687. Copyright 2008, American Chemical Society). a) Standard deviation for residual weights is 1.0%. b) Standard deviation for residual weights is 1.3%.

In CW–PFA, the onset of degradation (temperature at 5% weight loss) is at 323 °C, which is 77 °C higher compared to γ-Al-PFA, and remarkably, also 20–30 °C higher compared to the MMT–PFA nanocomposites. The residual weight of CW–PFA is 6 wt% (units) higher at 500 °C and 4 wt% (units) higher at 800 °C compared to pure PFA. Table 6.2 summarizes the main TGA results. 6.4.2 Reactive Molding of MMT Nanocomposites

This section discusses the results of reactive molding of nanocomposites from naturally occurring MMT clay. This was achieved by in-situ polymerization, using FA as a polymerizable solvent medium to produce MMT–PFA nanocomposites.

149

150

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

Surface-modiﬁed MMT clay is commercially available, as well as natural sodium MMT. Since FA is hydrophilic, in contrast to the hydrophobic PFA, the intercalation process was studied both with naturally occurring hydrophilic sodium MMT (NaMMT) and with an organomodiﬁed MMT (30BMMT), which is hydrophobic. To characterize the polymerization behavior of FA and to investigate how the presence of MMT inﬂuences this polymerization, FTIR spectra were collected before and during the resiniﬁcation process. The dispersion of the MMT in the PFA matrix is shown both directly and indirectly. The direct evidence consists of the XRD patterns of the FA-MMT suspension, which was used to monitor the process of intercalation and exfoliation of the MMT at various stages of resiniﬁcation. The dispersion is indirectly evidenced in increased thermal stability of the MMT–PFA nanocomposite, as measured by TGA. The thermal stability is discussed and compared to the pure polymer and to the CW–PFA nanocomposites. In addition, the important differences between oxidative and nonoxidative degradation of the NaMMT–PFA nanocomposite is discussed, and a mechanism is proposed to explain the difference in terms of acid-catalyzed degradation. 6.4.2.1 Morphology of MMT Clay The morphology of the MMT used in the preparation of MMT–PFA nanocomposites was examined with AFM, with samples prepared by drying dilute aqueous suspensions on a smooth silicon wafer. Figures 6.18 and 6.19 highlight the stacked lamellar morphology of the MMT. The line scan shows that large tactoids spanning 0.6–1.0 μm in length. Because the thickness of an individual platelet is approximately 1 nm, the aspect ratio of these plates is 20–100 upon complete exfoliation [10, 79]. The tactoid height varies between 150 and 200 nm, hence the largest tactoids (bottom right hand image) contain approximately 150 and 200 silicate layers prior to exfoliation. As shown in Figures 6.18–6.19, large tactoids tend to have a regular “arrowhead” shape. Figure 6.20 shows several tactoids in the process of exfoliation, such as, for example, the tactoid that is bisected by the line scan in the top left image. Previous studies have noted that tactoids disperse as a deck of cards thrown on a table, because they cleave along the (0 0 1) basal surface, that is, they exfoliate as platey crystals. Note from the line scan of top right hand image that several tactoids are actually highly tilted. Figures 6.21 and 6.22 show several tactoids that have been almost completely exfoliated. As the tactoids exfoliate, the platelets tend to fracture, and the regular arrowhead morphology predominant among larger tactoids disintegrates to yield an array of irregular shapes [80, 81]. As can be seen from the line scans, some tactoids are only three to four platelets high. 6.4.2.2 Resiniﬁcation of FA in the Presence of MMT Clay Preliminary DSC surveys indicated that MMT possessed sufﬁcient catalytic activity (attributed to its Lewis acidity) to effect the polymerization of FA to PFA above approximately 140 °C. Figure 6.23 shows typical DSC proﬁles for FA with 10 phr NaMMT, FA with 10 phr 30BMMT, and pure FA. In the presence of MMT clay, exothermic curing is observed in the temperature range of 130–210 °C. The exo-

6.4 Results and Discussion

Line Scan Height (nm)

Line Scan Height (nm)

160.0 140.0 120.0 100.0 80.0 60.0 40.0 20.0 0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5

120 100 80 60 40 20 0 0.0

0.5

Line Scan Length (um)

Figure 6.18

151

1.0

1.5

2.0

2.5

3.0

3.5

Line Scan Length (um)

AFM phase contrast images of large tactoids of MMT.

therm is interrupted by a strong endothermic peak above 160 °C, which can be attributed to the loss of water from the polymerization process and volatiles (the DSC pans could not be sealed tightly enough to avoid through-leakage). In contrast, for uncatalyzed, pure FA, no exotherm is observed, but only endothermic loss of volatiles. Therefore, the MMT–FA system was heated to 150 °C temperature, at which both NaMMT–FA and 30BMMT–FA mixtures resiniﬁed, reaching the gel point within 1–2 h. Intercalation and exfoliation of MMT leads to an increase in the d001 spacing (i.e., the basal spacing) of the MMT tactoids, which can be calculated from the position of the main diffraction peak in XRD and from Braggs law, nλ = 2d · sin θ (with n = 1). Therefore, XRD was employed to monitor the degree of intercalation and exfoliation of the MMT at various stages of resiniﬁcation. The change in the d001 spacing (i.e., the basal spacing) of the MMT tactoids was calculated from the position of the diffraction peak, and the interlayer gallery spacing was then estimated by subtracting 10 Å (the thickness of a single platelet) from the calculated basal spacing.

4.0

4.5

40 35 30 25 20 15 10 5 0 0.0

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

Line Scan Height (nm)

Line Scan Height (nm)

152

0.2

0.4

0.6

0.8

1.0

Line Scan Length (um)

1.2

1.4

250 200 150 100 50 0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 Line Scan Length (um)

Figure 6.19 AFM phase contrast images of large tactoids of MMT (Adapted with permission

from Macromolecules 2008, 41(22), 8682–8687. Copyright 2008, American Chemical Society).

Figure 6.24 (patterns a–c) shows the diffraction patterns collected for the 30BMMT–PFA resin. Prior to intercalation, 30B MMT powder (pattern a) has a basal spacing of d001 = 17.2 Å (2θ = 5.1 °) and hence an interlayer gallery spacing of 7.2 Å. This spacing is due to the organic modiﬁer being oriented roughly parallel to the platelets. In the initial 30BMMT–FA mixture, the basal spacing increases by 20 Å, evidenced by a dramatic shift in the main diffraction peak from 5.1 ° to 2.4 °, showing that 30BMMT readily solvates FA. Interestingly, this basal spacing remains nearly constant as the intercalated FA undergoes resiniﬁcation. At the gel point of the 30BMMT–PFA resin (pattern b), some 30BMMT tactoids remain intact albeit with a slightly increased basal spacing of 18.3 Å (2θ = 4.8 °). However, most of the 30BMMT is intercalated with PFA, as evidenced by the intense diffraction peak at 37.1 Å (2θ = 2.4 °), which corresponds to an interlayer gallery spacing of 27 Å. 30BMMT–PFA resin produced using 2 phr PTSA/acetic acid as a cocatalyst was also found by XRD to be intercalated, with nearly the same basal spacing (37 Å). At this stage, the organic modiﬁer is envisioned as forming a par-

6.4 Results and Discussion

140 120 100 80 60 40 20 0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 Line Scan Length (um)

60 Line Scan Height (nm)

Line Scan Height (nm)

160

153

50 40 30 20 10 0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 Line Scan Length (um)

Figure 6.20 AFM phase contrast images of medium size tactoids of MMT in the process of

exfoliation.

afﬁnic monolayer, in which the tallow chain of the organomodiﬁer is extended away from the clay surface, to optimize salvation by the intercalating a monomer or a polymer [23]. The observed basal spacing is very close to that described for 30BMMT intercalated with epoxy precursor (DGEBA), and for octadecyl ammonium – modiﬁed MMT intercalated with polyols with molecular weights ranging from 700 to 3000 g mol−1 [22, 82]. In the latter case, the interlayer gallery expansion was shown to depend on the chain length of the organic modiﬁer, rather than the molecular weight of the intercalating polyol. The diffraction peaks at 2θ = 2.4 ° and 4.8 Å were still observed in the 30BMMT PNC after the ﬁrst stage of curing at 130 °C (pattern c); however in the ﬁnal cured 30BMMT PNC (pattern d), no diffraction peaks are observed in the XRD patterns, showing that the 30BMMT has been completely exfoliated throughout the matrix. The exfoliation has presumably been effected by stiffening of the PFA due to extended crosslinking and molecular weight increase which characterize the second stage of curing.

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

6.0 5.0 4.0 3.0 2.0 1.0 0.0 0.0 0.1 0.2 0.30.4 0.5 0.60.7 0.80.9 1.01.11.2 1.3 1.4 1.51.6 1.7 1.8 Line Scan Length (um)

Line Scan Height (nm)

Line Scan Height (nm)

154

6.0 5.0 4.0 3.0 2.0 1.0 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 Line Scan Length (um)

Figure 6.21 Exfoliated tactoids of montmorillonite. Note the irregular shapes of exfoliated

platelets compared to the larger tactoids in Figures 6.18–6.19.

In the case of NaMMT, exfoliation is achieved already upon resiniﬁcation. Figure 6.25 shows that prior to intercalation, NaMMT powder (pattern a) has a basal spacing of d001 = 11.7 Å (2θ = 7.5 °) corresponding to an interlayer gallery spacing of approximately 1.7 Å. NaMMT, like 30BMMT, readily solvates FA, and after initial mixing, it undergoes an interlayer gallery expansion to approximately 15.5 Å (2θ = 5.7 °). At this stage, a monolayer of FA is most likely coordinated to the surface of the NaMMT by hydrogen bonding. As the reaction temperature increases to 130–140 °C (patterns b–c), resiniﬁcation proceeds more rapidly, the diffraction peak broadens, its intensity decreases, and by the time the temperature reaches 150 °C (pattern e), the peak vanishes, indicating complete exfoliation of NaMMT resin (pattern f ). The in-situ intercalative polymerization process, shown schematically in Figure 6.26, is especially favorable for the FA–MMT combination because MMT has the ability to catalyze the polymerization of FA to PFA which in turn drives the exfoliation process observed in XRD as discussed above.

6.0

Line Scan Height (nm)

Line Scan Height (nm)

6.4 Results and Discussion

5.0 4.0 3.0 2.0 1.0 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 Line Scan Length (um)

Heat flow (exo up)

Figure 6.22

4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 Line Scan Length (um)

Exfoliated tactoids of montmorillonite.

Uncatalyzed FA FA with 10phr 30BMMT FA with 10phr NaMMT 50

100

150 Temp (°C)

200

DSC spectra collected at 10 °C min−1, using sealed DSC pans, comparing uncatalyzed FA with FA catalyzed with 10 phr MMT. Figure 6.23

155

250

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

Rel. Intensity

d)

c) b) a)

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

10.0

2θ Figure 6.24 XRD patterns showing intercala-

tion and exfoliation of MMT as a function of polymerization of PFA: (a) 30BMMT; (b) intercalated 30BMMT–PFA resin; (c) 30BMMT–PFA resin after ﬁrst stage curing

process; (d) fully cured and exfoliated 30BMMT PNC (Adapted with permission from Macromolecules 2008, 41(22), 8682–8687. Copyright 2008, American Chemical Society).

f) e) Rel. Intensity

156

d) c) b) a)

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

10.0

2θ Figure 6.25 XRD patterns showing intercala-

tion and exfoliation of MMT as a function of polymerization of PFA: (a) NaMMT; (b) after initial mixing with FA; (c) intercalated NaMMT–PFA resin at 130 °C, (d) intercalated NaMMT–PFA resin at 140 °C; (e) intercalated

NaMMT–PFA resin at 150 °C; and (f) exfoliated NaMMT–PFA resin (Adapted with permission from Macromolecules 2008, 41(22), 8682–8687. Copyright 2008, American Chemical Society).

FTIR analysis shows that the polymerization of FA is normally accompanied by signiﬁcant increases in the intensity of the peak at 1562 cm−1, which is assigned to the skeletal vibration of 2,5 di-substituted furan rings. At the same time there is a decrease in peak intensity for the broad peak in the hydroxyl stretching region (3200–3600 cm−1) and for the peak at 3120 cm−1, assigned to the in-plane stretch of the hydrogen at the C5 position of the furan ring. The intensity of the sharp peak

6.4 Results and Discussion

FA Initial Mixing

Intercalation

PFA In-situ polymerization and exfoliation

Figure 6.26 Schematic representation of in-situ intercalative polymerization.

at 1504 cm−1 is assigned to the ring stretching of the furan ring, and is prominent both in FA and PFA spectra. The resiniﬁcation of with NaMMT is shown in Figure 6.27, which compares the FTIR spectra for FA with 10 phr NaMMT after initial mixing (spectrum a), at 150 °C (spectrum b), after 1 h at 150 °C (spectrum c), and the ﬁnal gelled resin after 1.25 h at 150 °C and cooling to room temperature (spectrum d). The resiniﬁcation of FA with 30BMMT is shown in Figure 6.28, which compares the FTIR spectra for FA with 10 phr 30BMMT after initial mixing with FA (spectrum a), at 150 °C (spectrum b), after 1 h at 150 °C (spectrum c) and the

157

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites a)

Rel. Absorbance (a.u.)

NaMMT PFA resin

d) c) b) a) 1800

1750

1700

1650

1600

1550

1500

1450

1400

1350

1300

b)

Rel. Absorbance (a.u.)

158

d) c) b) a) 900

850

800

750

700

650

600

550

500

Figure 6.27 FTIR spectra in the 1300–1800 cm−1 (top) and the 500–900 cm−1 (bottom) spectral

regions for (a) NaMMT–FA after initial mixing; (b) at 150 °C; (c) after 1 h at 150 °C; (d) gelled resin after 1.25 h at 150 °C and cooling to room temperature.

ﬁnal gelled resin after 2 h at 150 °C and cooling to room temperature (spectrum d). From Figures 6.28 to 6.29, several spectral features are worth noting. Typically, a strong peak at 1711 cm−1, assigned to the C=O stretch of γ-diketones formed from hydrolytic ring cleavage of some of the furan rings along the PFA chain, is observed in PFA resins. However, Figures 6.28 and 6.29 show that in the 1300– 1800 cm−1 region, for both NaMMT–PFA and 30BMMT–PFA, the intensity of the peak at 1710 cm−1 is remarkably weak, especially for NaMMT–PFA. The relatively high intensity of the peak at 1420 cm−1, assigned to C–O–C stretching of the furan ring is consistent with the lack of ring cleavage. In the 500–900 cm−1 region, the two broad peaks centered at 815 and 744 cm−1 evolve into a single band, as a polymerization progresses. This is due to a new peak forming at approximately 800 cm−1. The peak at 730 cm−1 was in the literature originally assigned to the out of plane vibration of the C–H groups at the C5 position of the furan ring, because in the transmission spectrum, it appears to decrease with increasing polymerization relative to the peak around 800 cm−1 [63, 74].

6.4 Results and Discussion a)

Rel. Absorbance (a.u.)

30BMMT PFA resin

d) c) b) a)

1800

1750

1700

1650

1600

1550

1500

1450

1400

1350

1300

Rel. Absorbance (a.u.)

b)

d) c) b) a)

900

850

800

750

700

650 −1

600

550

500 −1

Figure 6.28 FTIR spectra in the 1300–1800 cm (top) and the 500–900 cm (bottom) spectral

regions for (a) 30BMMT–FA after initial mixing; (b) at 150 °C; (c) after 1 h at 150 °C; (d) ﬁnal gelled resin after 2 h at 150 °C and cooling to room temperature.

However, in the absorbance spectra of Figures 6.27 and 6.28, it can be seen that the intensity of the 730 cm−1 peak does not decrease with increasing polymerization. Hence, the assignment of this peak to the C–H groups at the C5 position of the furan ring is questionable. Assigning the peak at 730 cm−1 to a ring stretching mode of the furan ring, as has been done in recent papers, is more reasonable [73, 83]. A third feature of interest in Figure 6.28 is the peak at 520 cm−1 assigned to the Al–O stretch in MMT [84]. The intensity of this peak increases with increasing polymerization as expected, given that dispersion of MMT is occurring simultaneously. For NaMMT–PFA, the intensity of this peak increases dramatically upon ﬁnal resiniﬁcation and exfoliation, and the peak can be seen to split into several peaks, at 533, 518, and 508 cm−1. This feature of the MMT–PFA spectra is useful, as it serves to differentiate between intercalated and exfoliated MMT morphology and hence to corroborate the XRD analysis. The intensity of the sharp peak at 1504 cm−1, assigned to the ring stretching of the furan ring, serves as an internal reference for semiquantitative analysis.

159

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites 100.0

b))

95.0 90.0 85.0

Weight

160

a)

80.0 75.0 70.0 65.0

c)

60.0

d)

55.0 50.0 50

100

150

200

250

300

350

400

450

500

550

600

650

700

o

Temperature ( C) Figure 6.29 TGA scans collected under N2 ﬂow, at 10 °C min−1, showing the onset of degradation for the case of nonoxidative degradation of cured PNCs consisting of (a) γ-Al-PFA; (b) CW–PFA; (c) 30BMMT–PFA; and (d) NaMMT–PFA.

Table 6.3 Internal referencing of the 1562 and 1711 cm−1 peaks from Figures 6.28–6.29.

Peak height ratio I/Iref

CW–PFA resina)

30BMMT resinb)

NaMMT resinb)

γ-Al-PFA resinc)

I1560/I1500 I1710/I1500

1.115 1.349

0.423 0.195

0.825 0.177

0.527 0.424

Reprinted with permission from Macromolecules 2008, 41(22), 8682–8687. Copyright 2008, American Chemical Society). a) 2 h reaction at 100 °C. b) 1–2 h reaction at 150 °C. c) 12 h reaction at 100 °C.

Table 6.3 shows the results of applying the internal referencing method to the peaks at 1562 and 1711 cm−1 for γ-Al-PFA (12 h reaction at 100 °C) and MMT–PFA (1–2 h resiniﬁcation at 150 °C). For comparison, the results for CW–PFA (2 h resiniﬁcation at 100 °C) are also included. Based on the 1710 cm−1/1500 cm−1 ratios, it can be seen that the structures of CW–PFA and MMT–PFA differ in that CW–PFA is characterized by very strong hydrolytic furan ring cleavage to γ-diketone, while MMT suppresses the ring cleavage. As shown in Figure 6.16, catalysis with γ-Al leads to furan ring cleavage to γ-diketone. Hence, the difference cannot simply be attributed to the difference between Lewis and Brönstedt acid catalysis. An explanation for this structural difference awaits further investigation in future studies.

6.4 Results and Discussion

6.4.2.3 Thermal Resistance of MMT–FA Nanocomposites MMT is often used to enhance the thermal properties of a composite material. Signiﬁcant increases in the onset of decomposition and char retention have been described for a range of PNCs using MMT [4, 8]. This is generally attributed to the restricted thermal motion of polymer chains at the MMT surface. The key role of the latter mechanism helps to explain cases where PNCs in which MMT is intercalated exhibit higher thermal stability compared to corresponding PNCs in which MMT is completely exfoliated, even though the exfoliated structure provides a more tortuous path for out-diffusion. Figure 6.29 compares the nonoxidative degradation of MMT–PFA with that of CW–PFA and GAl-PFA. As discussed previously, the largest increase in the onset of degradation (temperature at 5% weight loss) is seen for CW–PFA. In the case of MMT–PFA nanocomposites, onset of degradation is 50–60 °C higher compared to the pure PFA polymer. Above 400 °C, 30B-MMT PFA shows the highest thermal stability of all three PNCs. Table 6.4 shows weight retention at 500 and 800 °C for cured PFA nanocomposites. At 800 °C, the residual mass of 30BMMT–PFA is 12 wt% (units) higher compared to γ-Al-PFA after nonoxidative degradation. The increased thermal resistance of the MMT–PFA nanocomposites compared to CW–PFA nanocomposites can be attributed to the retarded out-diffusion of decomposition products thanks to the “labyrinth” morphology of exfoliated MMT in the matrix [23, 85]. Above 450 °C, H2O evolved from the decomposing PFA tends to oxidize methylene bridges between furan rings to carbonyl functions [86, 87]. Hence, the longer diffusion path in MMT–PFA PNCs translates to a greater likelihood of oxidation as opposed to mass loss. The explanation for the increased thermal resistance of 30BMMT–PNC over NaMMT–PNC is less straightforward. However, since exfoliation of NaMMT occurs while the PFA is still resinous, this allows for a small amount of phase separation of the NaMMT prior to ﬁnal matrix consolidation. NaMMT has a hydrophilic surface, and is less compatible with the organophilic PFA matrix compared to 30MMMT. Phase separation of the NaMMT may therefore result in gaps in the labyrinth morphology through which diffusion is Table 6.4 Temperature at the onset of decomposition and weight retention at 500 and 800 °C for cured PFA nanocomposites under nonoxidative degradation in N2.

γ-Al-PFA

CW–PFA

NaMMT–PFA

30BMMT–PFA

Onset of degradation (5% weight loss) in N2 ﬂow 246 °C 323 °C

302 °C

295 °C

Weight retained after nonoxidative degradation in N2 500 °Ca): 59% 65% 800 °Cb): 48% 52%

67% 55%

70% 60%

Reprinted with permission from Macromolecules 2008, 41(22), 8682–8687. Copyright 2008, American Chemical Society). a) The standard deviation for the residual weights is 1.0%. b) The standard deviation for the residual weights is 1.3%.

161

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

more rapid. In contrast, organophilic 30BMMT exfoliates during the late cure stage, and the exfoliated morphology of the 30BMMT is locked in place by the low mobility of the highly branched and crosslinked PFA matrix. Figure 6.30 shows the oxidative degradation in an air ﬂow of γ-Al-PFA (spectrum a), CW–PFA (spectrum b), 30B-MMT PFA (spectrum c), and NaMMT–PFA (spectrum d) after oven-curing at 130 °C for 75 min, and then at 210 °C for 105 min. As in the case of nonoxidative degradation, CW–PFA shows the largest increase in the onset of degradation of all PNCs. For CW–PFA, the onset of degradation is 350 °C, which is 92 °C higher compared to γ-Al-PFA. The behavior of 30BMMT is also similar compared to the case of nonoxidative degradation in that above 370 °C, 30BMMT–PFA shows the highest thermal stability of all three PNCs. Table 6.5 shows weight retention at 500 and 800 °C for cured 100.0

b))

90.0

a)

80.0 70.0

Weight

162

c)

60.0 50.0 40.0 30.0

d)

20.0 10.0 0.0 50

100

150

200

250

300

350

400

450

500

550

600

650

700

o

Temperature ( C) Figure 6.30 TGA scans collected under air ﬂow, at 10 °C min−1 showing the onset of degradation for the case of oxidative degradation of cured PNCs of (a) γ-Al-PFA; (b) CW–PFA; (c) 30BMMT–PFA; and (d) NaMMT–PFA. Table 6.5 Temperature at the onset of decomposition and weight retention at 500 and 800 °C

for cured PFA nanocomposites under oxidative degradation in air. γ-Al-PFA

CW–PFA

NaMMT–PFA

30BMMT–PFA

Onset of degradation (5% weight loss) in N2 ﬂow 258 °C 350 °C

288 °C

327 °C

Weight retained after nonoxidative degradation in N2 500 °Ca): 67% 70% 800 °Cb): 2.1% 4.8%

39% 14%

75% 32%

a) The standard deviation for the residual weights is 7% for NaMMT–PFA, but <1% otherwise. b) The standard deviation for the residual weights is 1%.

6.4 Results and Discussion

Residual Weight (%)

PFA nanocomposites. At 800 °C, the residual mass of 30BMMT–PFA is 30 wt% (units) higher compared to γ-Al PFA. However, for NAMMT–PFA, there is a signiﬁcant difference in thermal resistance compared to the case of nonoxidative degradation. The residual weight of NaMMT–PFA drops rapidly above 450 °C and levels out at approximately 14 wt% (units) above 650 °C. In this case, the NaMMT appears to accelerate, rather than retard the decomposition of PFA. The main steps in the oxidative degradation of PFA are oxidation of methylene bridges to carbonyl functions, followed by chain scission [63, 86, 87]. Degradation under HCl gas ﬂow has been shown to accelerate the decomposition of PFA compared to degradation under argon ﬂow [88]. Since the surface of NaMMT contains highly acidic sites, with acidity comparable to HCl, these sites most likely accelerate the degradation of PFA. The fact that 30BMMT does not accelerate degradation of PFA under the same conditions may be due to blocking of the acidic sites at the 30BMMT surface, perhaps by decomposition products originating from the organic modiﬁer. Regardless of the nature of the mechanism, this result highlights the importance of studying degradation under MMT-modiﬁed PNCs both under oxidative and nonoxidative conditions. Since the MMT–PFA, CW–PFA, and γ-Al-PFA (control) nanocomposites have signiﬁcantly different ﬁller content of 10, 0.75, and 5 phr, respectively, it is also interesting to compare the residual weights after accounting for the difference in ﬁller loading. Therefore, the ﬁller content was normalized from phr to percent, and the weight percentages of PFA matrix and ﬁller (i.e., CW, γ-Al, or MMT) before decomposition were calculated. The residual weight obtained when decomposing the ﬁller alone was determined by TGA under the same conditions as for the PFA nanocomposites. For each nanocomposite, the residual weight attributable to the ﬁller was then subtracted from the nanocomposite residual weight, yielding residual weight of the PFA matrix. Finally, the PFA matrix residual weight was normalized with respect to the PFA matrix initial weight. Figure 6.31 shows the normalized residual weight of PFA at 800 °C for PFA nanocomposites after nonoxidative and

60 40 20 0

Oxidative degradation under nitrogen Nonoxidative degradation under air flow

g-Al-PFA

NaMMTPFA

30BMMTPFA

CW-PFA

0.3

6.8

28.1

0.2

46.8

51.8

59.5

52.7

Figure 6.31 Residual weight at 800 °C for PFA nanocomposites after nonoxidative and

oxidative degradation, normalized to account for difference in ﬁller loading.

163

164

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites

oxidative degradation. The tall bars represent normalized residual weight of the PFA after nonoxidative degradation and the short bars represent normalized residual weight of PFA after oxidative degradation, both at 800 oC. Compared to the control, that is, γ-Al-PFA, the normalized residual weight of the PFA matrix in the 30BMMT–PFA nanocomposite is 12.7 wt% (units) under nonoxidative degradation and 27.8 wt% units higher under oxidative degradation. This conﬁrms that the 30BMMT–PFA system exhibits the highest thermal stability of all three PNCs, both under oxidative and nonoxidative conditions.

6.5 Conclusions

The reactive molding of bio-based PNCs has been achieved by using in-situ polymerization techniques to disperse nanoparticles of CW or MMT clay in a thermosetting PFA matrix. Nanoparticles of CW and MMT play a dual role in the in-situ polymerization process, by ﬁrst catalyzing the polymerization, conveniently eliminating the use of strong mineral acid catalysts, and then enhancing the thermal stability of the consolidated PNCs. In the case of CW–PFA nanocomposites, TGA analysis shows a signiﬁcant increase in the temperature at onset of degradation (nonoxidative) of nearly 80 °C compared to the pure PFA matrix, at a loading of only 1.0 wt% CW. Similarly, in the case of MMT–PFA nanocomposites, the onset of degradation (nonoxidative) is 50–60 °C higher compared to a pure PFA polymer. However, the most remarkable increase in thermal properties achieved with MMT–PFA nanocomposites is the increase in residual weight above 400 °C under oxidative degradation. Organomodiﬁed MMT increases residual weight at 800 °C by 30 wt% (units) compared to pure PFA. For high-temperature applications, organomodiﬁed MMT should be selected rather than sodium MMT, because above 450 °C NaMMT appears to accelerate, rather than retard the decomposition of PFA. This shows the importance of studying both oxidative and nonoxidative degradation of PNCs. The increased thermal stability of CW–PFA and MMT–PFA is attributed to the nanoparticles restricting the thermal motion of the matrix polymer because of the high interfacial area between the matrix and the nanoparticle phase. In conclusion, in-situ reactive polymerization with CW or MMT offers an attractive processing route for producing PFA matrix nanocomposites without the use of strong mineral acids, solvents, or surfactants. Moreover, this approach simultaneously fulﬁls the objectives of increasing the use of of bio-based materials while realizing advanced composite materials with nanoscale ﬁllers. Future studies will focus on mechanical testing of CW–PFA and MMT–PFA nanocomposites.

Abbreviations

CW FA

cellulose whiskers furfuryl alcohol

References

MCC MMT 30BMMT phr PFA PNC NaMMT

microcrystalline cellulose montmorillonite organomodiﬁed MMT Cloisite 30B® parts per hundred resin polyfurfuryl alcohol polymer matrix nanocomposite sodium MMT Cloisite Na+®

Acknowledgments

This work was sponsored by a grant from the Institute of Paper Science and Technology (IPST) at Georgia Tech and by Paper Science and Engineering (PSE) graduate fellowships to both Dr Larry Pranger and Dr Grady Nunnery, also from the IPST at Georgia Tech. We would like to thank Ms Yehudit Schmidt from the Department of Chemical Engineering at the Technion in Israel for her expertise in obtaining the cryo-TEM images of the CW. We are indebted to Professors Hamid Garmestani, Preet Singh, Karl Jacob, and Tim Patterson from Georgia Tech for stimulating discussions on various aspects of this work.

References 1 Mallick, P.K. (1993) Fiber-Reinforced

2 3

4 5 6

7 8 9

Composites Materials, Manufacturing, and Design, Mechanical Engineering Series 83, Marcel Dekker, New York. Chazeau, L., et al. (1999) J. Appl. Polym. Sci., 71, 1797. Krishnamoorti, R. and Vaia, R.A. (2002) Polymer Nanocomposites: Synthesis, Characterization, and Modeling, ACS Symposium Series, American Chemical Society, Cary, NC. Alexandre, M. and Dubois, P. (2000) Mater. Sci. Eng. R, 28, 1. Brechet, Y., et al. (2001) Adv. Eng. Mater., 3, 571. Ruan, D., et al. (2004) J. Polym. Sci., Part B: Polym. Phys., 42, 367. Suvorova, E.I., et al. (2005) J. Cryst. Growth, 275, 2351. Ray, S.S. and Okamoto, M. (2003) Prog. Polym. Sci., 28, 1539. Dufresne, A., Kellerhals, M.B., and Witholt, B. (1999) Macromolecules, 32, 7396.

10 Luo, J.-J. and Daniel, I.M. (2003) Compos.

Sci. Technol., 63, 1607. 11 Terech, P., Chazeau, L., and Cavaille, J.

(1999) Macromolecules, 32, 1872. 12 Wu, C.L., et al. (2002) Compos. Sci.

Technol., 62, 1327. 13 Kovacevic, V., Lucic, S., and Leskovac, M.

(2002) J. Adhes. Sci. Technol., 16, 1343. 14 Balogh, M.L.P. and Laszlo, P. (1993)

15 16 17 18

19

Organic Chemistry Using Clays, Reactivity and Structure: Concepts in Organic Chemistry, vol. 29, Springer-Verlag, Berlin. Eichhorn, S.J., et al. (2001) J. Mater. Sci., 36, 2107. Mohanty, A.K., Misra, M., and Drzal, L.T. (2002) J. Polym. Environ., 10, 19. Zeitsch, K.J. (2000) Sugar Series, vol. 13, Elsevier, New York. Agarwal, B. (1980) Analysis and Performance of Fiber Composites, SPE Monographs, John Wiley & Sons, Inc., New York. Pron, A., et al. (1997) Synthetic Metals, 84, 89.

165

166

6 Control of Filler Phase Dispersion in Bio-Based Nanocomposites 20 Ajayan, P.M. (1999) Chem. Rev., 99, 1787. 21 Araki, J., Wada, M., and Kuga, S. (2001) 22 23 24 25 26

27

28 29 30 31 32 33

34

35 36 37 38 39 40 41 42 43

44

Langmuir, 17, 21. Park, J.H. and Jana, S.C. (2003) Macromolecules, 36, 2758. LeBaron, P.C., Wang, Z., and Pinnavaia, T. (1999) Appl. Clay Sci., 15, 11. Jia, W., et al. (2003) Synthetic Metals, 132, 269. Yu, G., Zhang, M.Q., and Zeng, H.M. (1999) Composite Interfaces, 6, 275. Chazeau, L., Paillet, M., and Cavaille, J.Y. (1999) J. Polym. Sci., Part B: Polym. Phys., 37, 2151. Dufresne, A., Dupeyre, D., and Vignon, M.R. (2000) J. Appl. Polym. Sci., 76, 2080. Matarredona, O., et al. (2003) J. Phys. Chem. B, 107, 13357. Dubief, D., Samain, E., and Dufresne, A. (1999) Macromolecules, 32, 5765. Samir, M., et al. (2004) Macromolecules, 37, 1386. Klemm, D., et al. (2005) Angew. Chem., Int. Ed., 44, 3358. Bledzki, A.K. and Gassan, J. (1999) Prog. Polym. Sci., 24, 221. Huttermann, A., Mai, C., and Kharazipour, A. (2001) Appl. Microbiol. Biotechnol., 55, 387. Matsumura, H., Sugiyama, J., and Glasser, W.G. (2000) J. Appl. Polym. Sci., 78, 2242. Belgacem, M.N., Bataille, P., and Sapieha, S. (1994) J. Appl. Polym. Sci., 53, 379. Espert, A., Vilaplana, F., and Karlsson, S. (2004) Compos. Part A, 35A, 1267. Joly, C., Gauthier, R., and Chabert, B. (1996) Compos. Sci. Technol., 56, 761. Qiu, W.L., Endo, T., and Hirotsu, T. (2004) J. Appl. Polym. Sci., 94, 1326. Urreaga, J.M., et al. (2000) Polym. Eng. Sci., 40, 407. Amash, A. and Zugenmaier, P. (1998) Polym. Bull., 40, 251. Tjong, S.C., Xu, Y., and Meng, Y.Z. (1999) J. Appl. Polym. Sci., 72, 1647. Noishiki, Y., et al. (2002) J. Appl. Polym. Sci., 86, 3425. Svagan, A.J., Samir, M.A.S., and Berglund, L.A. (2008) Adv. Mater., 20, 1263. Angles, M.N. and Dufresne, A. (2001) Macromolecules, 34, 2921.

45 Chazeau, L., Cavaille, J.Y., and Perez, J.

46 47 48 49 50

51 52 53 54 55 56 57 58 59 60 61 62

63 64

65 66 67 68 69

(2000) J. Polym. Sci., Part B: Polym. Phys., 38, 383. Nakagaito, A.N. and Yano, H. (2004) Appl. Phys. A Mater. Sci. Proc., 78, 547. Samir, M., et al. (2004) J. Phys. Chem. B, 108, 10845. Angles, M.N. and Dufresne, A. (2000) Macromolecules, 33, 8344. Dufresne, A. (2000) Compos. Interfaces, 7, 53. Al-Ahmed, A., Mohammad, F., and Rahman, M.Z.A. (2006) J. Appl. Polym. Sci., 99, 437. Nishino, T., Matsuda, I., and Hirao, K. (2004) Macromolecules, 37, 7683. Podsiadlo, P., et al. (2007) Science, 318, 80. Yalcin, B. and Cakmak, M. (2004) Polymer, 45, 6623. Shelley, J.S., Mather, P.T., and DeVries, K.L. (2001) Polymer, 42, 5849. Dupuy, J., et al. (2002) Macromol. Symp., 184, 249. Polushkin, E.Y., et al. (2002) Polym. Eng. Sci., 42, 846. Grande, J.A. (1999) Modern Plastics, 76, 30. El-Meligy, M.G. (2004) Polym. Plastics Technol. Eng., 43, 981. Collier, J.R., et al. (1996) J. Appl. Polym. Sci., 61, 1423. Koh, W.G. and Pishko, M. (2003) Langmuir, 19, 10310. Pﬂeging, W., et al. (2002) Microsyst. Technol., 9, 67. Dunlop, A.P. (1953) The Furans, American Chemical Society, Monograph Series, no. 119, Reinhold Pub. Corp., New York. Conley, R.T. and Metil, I. (1963) J. Appl. Polym. Sci., 7, 37. Choura, M., Belgacem, N.M., and Gandini, A. (1996) Macromolecules, 29, 3839. Chuang, I.S., Maciel, G.E. and Myers, G.E. (1984) Macromolecules, 17, 1087. Principe, M., Ortiz, P. and Martinez, R. (1999) Polym. Int., 48, 637. Nguyen, C. and Zavarin, E. (1986) J. Wood Chem. Technol., 6, 15. Spange, S., et al. (1992) Makromol. Chem. Rapid Commun., 13, 511. Sthel, M., Rieumont, J., and Martinez, R. (1999) Polym. Test., 18, 47.

References 70 Sthel, M., Rieumont, J., and Martinez, R. 71 72 73 74 75 76

77 78 79

(1999) J. Appl. Polym. Sci., 71, 1749. Araki, J., et al. (2000) Langmuir, 16, 2413. Bitton, R., et al. (2005) Langmuir, 21, 11888. Yao, J., et al. (2005) Carbon, 43, 1709. Wewerka, E.M. (1971) J. Polym. Sci., Part A – 1: Polym. Chem., 9, 2703. Gohdes, M. and Mischnick, P. (1998) Carbohydr. Res., 309, 109. Beck-Candanedo, S., Roman, M., and Gray, D.G. (2005) Biomacromolecules, 6, 1048. Cheng, R.-S., et al. (2007) Huanan Ligong Daxue Xuebao, Ziran Kexueban, 35, 91. Wang, N., Ding, E., and Cheng, R. (2007) Polymer, 48, 3486. Piner, R.D., et al. (2003) Langmuir, 19, 7995.

80 Ho, D.L., Briber, R.M., and Glinka, C.J.

(2001) Chem. Mater., 13, 1923. 81 Tournassat, C., et al. (2003) Am. Mineral.,

88, 1989. 82 Wang, Z. and Pinnavaia, T.J. (1998)

Chem. Mater., 10, 3769. 83 Gonzalez, R., Figueroa, J.M., and

84 85

86 87 88

Gonzalez, H. (2001) Eur. Polym. J., 38, 287. Chen, G., et al. (2001) Macromol. Chem. Phys., 202, 1189. Yano, K., Usuki, A., and Okada, A. (1997) J. Polym. Sci., Part A: Polym. Chem., 35, 2289. Fitzer, E. and Schaefer, W. (1970) Carbon, 8, 353. Gaefke, C.B., et al. (2007) J. Appl. Polym. Sci., 106, 2274. Shindo, A. and Izumino, K. (1994) Carbon, 32, 1233.

167

169

7 Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties Mo Song and Dongyu Cai

7.1 Introduction

Polyurethane (PU) has been considered as one of most versatile polymeric materials that are widely utilized as elastomers, coatings, and adhesives. PU chemistry basically refers to the formation of urea group (–NHCO–) via the reaction between hydroxyl groups (–OH) in polyols and isocyanate group (–NCO) in polyisocyanates. Morphologically, PU is a typical segmented copolymer consisting of soft segment and hard segment which is self-assembled by polyols and polyisocyanate, respectively. The structure and properties of PU can be ﬂexibly tailored to meet industrial demands. For instance, linear or crosslinked PU can be designed based on the functionalities of polyols and polyisocyanates. The PU containing polyester polyols is stronger than polyether-based one due to stronger intermolecular force between ester groups than ether groups. Aliphatic and aromatic isocyanate-based PU exhibits individual advantages in different industrial applications. Incorporation of nanoﬁllers into polymeric materials has been generated a great deal of interest in developing lightweight and multifunctional nanocomposites for broader industrial applications. It has been commonly recognized that nanoparticles are capable of enhancing the mechanical, electrical, and thermal properties of polymers. The enhancement closely relates to the dispersion level of nanoﬁllers and the interfacial strength between nanoﬁllers and polymeric matrices. Up to date, a variety of methods have been developed to fabricate polymer nanocomposites (PNCs) with homogenous dispersion of nanoﬁllers, in which melt and solution processing, in-situ polymerization are most active [1]. Melt processing is particularly suitable for dispersing nanoﬁllers in thermoplastic polymers in melt state [2, 3]. Solution processing mainly involves the incorporation of nanoﬁller into polymers with the assistance of solvents. It provides better dispersion than melt processing as nanoﬁllers are easier to be dispersed in solvents using sonication [4]. In-situ polymerization is a low-energy consumable method in which liquid organic monomers are used to carry nanoﬁllers and the polymerization of monomers takes place in the presence of nanoﬁllers [5]. Other techniques such

In-situ Synthesis of Polymer Nanocomposites, First Edition. Edited by Vikas Mittal. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA.

170

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties

as layer-by-layer deposition [6] and latex technology [7] are also reported to uniquely fabricate high-performance PNCs. PU can exist in many forms such as thermoplastic, solvent-based and WPU. It makes PU very welcomed to host nanoﬁllers using the methods introduced above. Melt or solution processing is normally applied to thermoplastic PU. Latex technology is speciﬁcally available for water-based PU [8]. In-situ polymerization is the most popular method to fabricate PUNs. It has great potential to be scaled-up to manufacture commercial PUNs due the advantages of low cost and ﬂexibility. In this chapter, our focus will be placed on reviewing the progress on the in-situ polymerization of PUNs. The fabrication, morphology, and properties of PUNs will be comprehensively discussed in the following sections.

7.2 PU/Carbon Nanotube Nanocomposites (PUCNs) 7.2.1 Fabrication

Carbon nanotubes (CNTs) are one-dimensional nanomaterials with cylindrical structures consisting of sp2 carbon bonds [9]. It is believed that CNTs are ideal nanoﬁllers due to its super mechanical [10], thermal [11, 12], and electrical properties. Two main commercial players for PNCs are multiwalled carbon nanotubes (MWCNTs) with a diameter range from 4 to 30 nm and single-walled carbon nanotubes (SWCNTs) with a diameter range from 0.4 to 2–3 nm. This section will follow the in-situ approaches to incorporate CNTs into PU elastomer, solvent-based and waterborne PU. For the preparation of PUCN elastomer, the exfoliation of CNTs into polyols is the ﬁrst and key step to achieve homogenous dispersion before the synthesis of PU. Physical or chemical pretreatments of CNTs are essential to enhance the compatibility between CNTs and polyols and reduce the aggregation of CNTs in polyols. Xia et al. [13] reported that a BYK branded dispersant could effectively wet SWCNTs and facilitate the dispersion of the nanotubes in a trifunctional polyether polyol, and also found that ball milling is a more efﬁcient mechanical tool than high-speed stirring to disperse the nanotubes in the polyol. Although the physical method is a simple and convenient to functionalize CNTs, more attention is generated on chemical modiﬁcation of CNTs in which treating agents are attached to CNTs via chemical bonding. The chemical modiﬁcation starts from the oxidization of CNTs using strong acids or other oxidants which can introduce hydroxyl and carboxylic groups onto the surface of CNTs. With these functional groups, PU chains can be chemically attached to the surface of CNTs taking advantage of the reaction between hydroxyl/carboxylic groups and polyols ended up with isocyanate groups, as shown in Scheme 7.1 [14]. In comparison with the functional groups, long PU chains can lead to stronger steric hindrance preventing the aggregation of CNTs in polyols. The oxidation generates relative low amount of hydroxyl and

7.2 PU/Carbon Nanotube Nanocomposites (PUCNs)

Scheme 7.1

Functionalization of MWCNTs with PU chains. Reproduced from Ref. [14] with

permission.

carboxyl groups on CNTs, which results in the low efﬁciency of grafting the nanotubes with PU chains in the following step. It was also reported that the degree of carboxylation could be enhanced by a free radical addition of alkyl groups terminated with carboxylic groups onto the MWCNTs [14]. After converting MWCNTs–COOH to MWCNT–COCl and further to MWCNT–NH2, the PU chains were covalently grafted to the nanotubes via the formation of –HN–CO–NH– linkage. Characterization of functionalized CNTs takes a big part of work to convince the achievement of the chemical linkage. Purifying functionalized CNTs is essential to obtain trustable results because the agents used for the functionalization of CNTs can be physically attached to CNTs. Removal of these noncovalently attached

171

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties

172

a)

c) C1s C Auger O Auger

b

N1s

a

a: MWCNTs-R2-NH2 b: MWCNTs-PU

1200

1000

b

800 600 400 Binding Energy (eV)

b)

a: MWCNTs-R2-NH2 b: MWCNTs-PU

O1s Counts/s

Counts/s

a

O1s

200

0

545 d)

540

535 530 525 Binding Energy (eV)

C1s

520

N1s a: MWCNTs-R2-NH2 b: MWCNTs-PU

Counts/s

Counts/s

a a: MWCNTs-R2-NH2 b: MWCNTs-PU

a

b

b 300

295

290

285

Binding Energy (eV)

280

410

405

400

395

390

Binding Energy (eV)

Figure 7.1 (a) XPS spectrum of (a) MWCNTs-R2-NH2 and (b) MWCNTs-PU). (b–d) enlarged

sections from (a) for C(1s), O(1s), and N(1s), respectively. Reproduced from Ref. [14] with permission.

agents can be performed by repeating washing and ﬁltering using polar organic solvents in a Soxhlet extraction kit [15]. Spectroscopic techniques including FTIR and XPS are directly used to identify the functionalities on the surface of CNTs. The existence of carboxylic groups (-COOH) is normally conﬁrmed by observing the bands at near 1730 cm−1 and 3440 cm−1 in FTIR spectra. Moreover, the XPS technique can quantitatively measure carbon, oxygen, and nitrogen atoms in the CNTs grafted with PU chains. Figure 7.1 shows the XPS spectrum of functionalized MWCNTs with NH2 groups and PU chains. The shifts in the binding energy of the atoms indicate the change of chemical environments around the atoms as urea linkages are formed. The MWCNT–NH2 contains 78.86% carbon, 18.56% oxygen, and 2.5% nitrogen atoms and the MWCNT–PU contains 74.39% carbon, 22.1% oxygen, and 3.51% nitrogen atoms. Thermogravimetric analysis (TGA) is an effective tool to measure relative contents of grafted PUs and other functional groups onto CNTs according to the weight loss of functionalized CNTs against the

7.2 PU/Carbon Nanotube Nanocomposites (PUCNs)

Figure 7.2 TGA curves for (a) MWCNTs–COOH, (b) MWCNTs-R1-COOH, (c) MWCNTs-R2NH2, and (d) MWCNTs-PU. Reproduced from Ref. [14] with permission.

Table 7.1 The viscosities of the polyether polyol–CNT mixtures at a low shearing rate of 4.45 s−1(ηL) and a high shear rate of 159.8 s−1(ηH), and shear thinning parameter (n) at room temperature.

CNT concentration (%)

0 0.5 1 2

ηL(mPa s)

ηH (mPa s)

n

MWCNT

SWCNT

MWCNT

SWCNT

MWCNT

SWCNT

1.59 2.31 3.32 12.50

1.59 1.96 3.35 8.00

1.58 1.99 2.44 4.45

1.58 1.79 2.15 2.95

1.000 0.968 0.908 0.751

1.000 0.975 0.928 0.830

Reproduced from Ref. [13] with permission..

increment of temperature. Figure 7.2 shows typical TGA curves of the CNTs with different functionalities. The weight loss below 600 °C is attributed to the decomposition of the functionalities and sharp drop in weight above 600 °C is due to the decomposition of CNTs. Therefore, the content of grafted PU can be roughly estimated to be 34%. The rheological behavior of liquid polyols with CNTs is crucial in the processing of ﬁnal PUCN products. Xia et al. [13] used the Herschel–Bulkley equation to investigate the rheological behavior of several polyol–CNT mixtures, which generally shows that the rheological behavior of polyols turns from Bingham plastic behavior independent of shear rates to shear-thinning behavior. Table 7.1 presents the viscosity of a trifunctional polyether polyol with pristine nanotubes at a low shearing rate of 4.45 s−1 and a high shear rate of 159.8 s−1 at room temperature. It can be seen that the addition of 2 wt% CNTs can signiﬁcantly increase the viscosity

173

174

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties

of the polyether polyol and reduce its liquidity. MWCNTs exhibit stronger thickening effects than SWCNTs. The polyol–MWCNT mixtures present stronger shearthinning behavior indicated by lower shear thinning parameters. The increase in the viscosity indicates restricted mobility of polyol chains resulting from the interaction between the nanotubes and polyol chains. This interaction can be gradually reduced with increasing shear force. It was believed that the viscosity for the polyol–CNT mixture with better dispersion could be higher because more exfoliated nanotubes interact with polyol chains. The equation as follows was used to quantitatively measure the relationship between the viscosity and dimension and concentration of nanoﬁllers: η = η0 (1 + 0.67 fC + 1.62 f 2C 2 ) in which f is the shape factor or the aspect ratio of nanoﬁllers, η0 is the viscosity of polyols, and C is the concentration of nanoﬁllers. By ﬁtting the viscosity data into the equation, it was found that the f value of polyol–MWCNT is much higher than that of polyol–SWCNT, which indicated that SWCNTs presented more nanoaggregates with lower aspect ratio in polyols than MWCNTs. In-situ polymerization of PUCNs can also be performed in organic solvents. Solvent-based PUCNs might have better dispersion than solvent-free ones since the exfoliation of CNTs is easier in organic solvents than viscous polyols. A typical way to prepare solvent-based PUCNs starts from the exfoliation of CNTs in polar organic solvents with ultrasonic treatment. The following synthesis of PUs in organic solvents with exfoliated CNTs can be ﬂexibly designed to achieve desirable structure and properties. Water-based PU (WPU) has been regarded as environmentally friendly member in the family of PU products, which are widely used as adhesive and coatings. Preparation of WPU generally involves two steps including (i) synthesis of PU with hydrophilic functionality in a solvent-free environment or in the presence of small quantity of less-toxic solvents such as actone; (ii) dissolution of hydrophilic PU into water by high-speed mixing. Dimethylol-propionic acid is a typical chemical used to introduce hydrophilic groups into the main chain of PU. The incorporation of CNTs into WPU can be simply performed by solution blending/casting, in which chemically functionalized or surfactant-wrapped CNTs are exfoliated into water, and then mixed with prepared PU latex in water [8]. Kwon and Kim [16] reported an in-situ polymerization route to prepare CNT/WPU nanocomposites shown in Scheme 7.2. First, poly(tetramethylene oxide) glycol (PTMG, Mn = 2000) was reacted with dimethylol propionic acid (DMPA) and isophorone disocyanate (IPDI) in the presence of N-methyl-2-pyrrolidone (NMP) at 80 °C until the theoretical NCO content was achieved. Afterward, the addition of triethylamine (TEA) was carried out to neutralize the carboxyl groups of the NCO-terminated PU prepolymer to form ionic couples on the side of PU chains. Distilled water was added into hydrophilic PU to form stable PU latex in water, which is normally called emulsiﬁcation. The addition of CNTs usually takes place in this step because CNTs can be exfoliated in the water with the assistance of surface modiﬁcation and strong ultrasonic treatment. Finally, the ethylene amine (EDA) was dropped

7.2 PU/Carbon Nanotube Nanocomposites (PUCNs)

Scheme 7.2 In-situ polymerization process for WPU/CNT nanocomposites. Reproduced from Ref. [16] with permission.

into PU latex to carry on the chain extension of a PU chain in the core of the PU latex. Therefore, the key step for the in-situ polymerization of PUCNs is the dispersion of CNTs in macromolecular polyols. In order to reduce the aggregations, it is necessary to physically or chemically modify the surface of CNTs to reduce the van der Waals force among the nanotubes. Strong mechanical tools such as ball milling and ultrasonic treatment can be used to help break down the aggregation of CNTs. The kinetics of PU chain growth should be taken into account although it is rarely reported in up-to-date publications.

175

176

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties a)

b)

c)

TEM images of PU/1%MWCNTs (a, b) and PU/1%SWCN (c) nanocomposites. Reproduced from Ref. [13] with permission.

Figure 7.3

7.2.2 Morphology and Characterizations of PUCNs

The changed morphology of the PU matrix resulting from CNTs is an interesting topic being widely investigated, which is the driving force for the enhanced physical properties of PUCNs. The key issues should concern the exfoliation of CNTs, the interface between the CNT and PU matrix, and the soft–hard segmented structure. Electron microcopies are common tools to examine the dispersion status including scanning and transmission electron miscopy. As shown in Figure 7.3, TEM images (a and b) exhibit the exfoliated MWCNTs in the PU matrix with a diameter of 10–20 nm. However, TEM image (c) reveals the aggregation of SWCNTs at a submicron scale, which indicates the difﬁculty to obtain fully exfoliated SWCNTs because SWCNTs have much stronger van der Waals forces than MWNCTs. Figure 7.4 shows a typical SEM image of PUCNs, which can observe CNTs (bright dots) well dispersed in the PU matrix. In some cases, optical microscopy (OM) is used to overcome the argument that electron microscopic images only can cover small and selective area. The thin ﬁlms of PUCNs under OM should be “clean” without many micro-sized black dots if CNTs are well dispersed in the PU matrix. It has been generally accepted that the interface between nanoﬁllers and polymeric matrices is a key player determining the mechanical reinforcing effects of nanoﬁllers. The interface can be enhanced by introducing functional groups or grafting polymeric chains onto the surface of nanoﬁllers. The strong interface can be engineered via the formation of covalent bonds or nucleated polymeric crystalline layers bridging nanoﬁllers and polymeric matrices. CNT–PU interactions can be measured by Raman spectroscopy and thermal analysis. The principle of the measurements is based on the level of change to the structural characteristics of

7.2 PU/Carbon Nanotube Nanocomposites (PUCNs)

Figure 7.4 FESEM images of the MWCNT(3 wt%)/PU nanocomposite. Reproduced from Ref.

[17] with permission.

each component in PUCNs after the mutual interaction. Many literatures have indicated that Raman spectra can give characteristic details of typical molecular vibration of CNTs, in which (i) a radial breathing mode (RBM) is observed in the region of 160–300 cm−1 associated with a symmetric movement of all carbon atoms in the radial direction; (ii) disordered mode (D-mode) resulting from the defects in the nanotubes is located between 1330 and 1360 cm−1; and (iii) graphitic mode (G-mode) peaking around 1580 cm−1 connects with the stretching mode in the graphite plane. The shifts of these vibrational peaks in Raman spectra can be indicative of the level of the CNT–polymer interactions. Figure 7.5 presents the Raman spectra of SWCNTs, SWCNTs-g-PU, and their PU nanocomposites in three regions of (a) 100–450 cm−1, (b) 1500–1750 cm−1, and (c) 2400–2800 cm−1. The Raman shift of SWNT did not result from the molecular interaction between PU and SWNT. The upshift of SWNT in PCL or PU at different modes could be attributed to the hydrostatic pressure exerted by the surrounding polymer matrix. It is believed that this shift is related to the cohesive energy density (CED) and a higher CED will lead to a larger Raman shift. FTIR spectroscopy was used by Jung et al. [19] to track the formation of covalent bonds (urethane groups) between MWCNTs with carboxylic acid groups (MWCNTs–COOH) and isocyanate groups on the ends of PUs. Figure 7.6 shows the variation of FTIR peak (2272 cm−1) belonging to isocyanate groups (N=C=O) on the end of PUs against time after mixing MWCNTs–COOH into NCOterminated PUs. The peak at 2272 cm−1 gradually deceases with reaction time and nearly disappears after 300 min, which indicates the formation of urethane bonds between then nanotubes and PU. The swelling behavior of PUCNs ﬁlms in organic solvents was an indirect evidence to conﬁrm the crosslinking of PU in the presence of the nanotubes. It was shown in Figure 7.7 that the crosslinked PUs never can be dissolved in N,N-dimetylformamide (DMF) and the PUs with pristine MWCNTs readily dissolved in DMF.

177

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties a) e: PU f: 0.35% SWNT/PU g: 0.35% SWNT-g-PU/PU h: 0.7% SWNT/PU l: 0.7% SWNT-g-PU/PU

a: SWNT b: SWNT-g-PU c: 0.5% SWNT-g-PU/PCL d: 1% SWNT-g-PU/PCL

Relative Intensity

l h g f e d c b a 100

150

200 250 300 350 Wavenumbers (cm–1)

400

450

b)

Relative Intensity

a: SWNT b: SWNT-g-PU c: 0.5% SWNT-g-PU/PCL d: 1% SWNT-g-PU/PCL

e: PU f: 0.35% SWNT/PU g: 0.35% SWNT-g-PU/PU h: 0.7% SWNT/PU l: 0.7% SWNT-g-PU/PU

l h g f e d c b a 1550

1600 1650 1700 Wavenumbers (cm–1)

1750

c) l h g Relative Intensity

178

f e d c b a 2450 2500 2550 2600 2650 2700 2750 2800 Wavenumbers (cm–1)

Figure 7.5 Raman spectra

of CNT, CNT-polyol dispersion, and PUCNs. Reproduced from Ref. [18] with permission.

7.2 PU/Carbon Nanotube Nanocomposites (PUCNs)

N=C=O 2272 cm–1

Absorbance (a. u.)

20 min 40 min 70 min 110 min 150 min 300 min

2700 2600 2500 2400 2300 2200 2100 2000 Wavenumber (cm –1) Figure 7.6 FTIR spectra of samples obtained during the reaction of NCO-terminated PU with

MWCNTs–OH. Reproduced from Ref. [19] with permission.

a)

b)

Figure 7.7 Two possible conﬁgurations for MWCNT-PU nanocomposite molecules: (a)

MWCNT-reinforced mixtures dissolved in DMF and (b) insoluble nanotube-crosslinked PU ﬁlm. Reproduced from Ref. [19] with permission.

As stated above, a strong physical interaction can also be engineered by the formation of polymeric crystalline layers around CNTs in semicrystalline PNCs. Coleman and his co-workers [20, 21] ﬁrst proposed the stress transferring mechanism across the crystalline layers and found that nucleating ability of CNTs led to the formation of the crystalline layers in a natural way. The evidences to prove the existence of the crystalline layers are usually sourced by DSC and high-resolution electron microscopy. This kind of crystalline layers were also found in the semicrystalline PUCNs prepared using semicrystalline soft segment such as polycarprolactone (PCL) [18]. DSC curves in Figure 7.8 clearly show the nonisothermal melting and crystallization of PCL-based PUCNs with SWCNTs grafted with PU

179

180

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties

Figure 7.8 DSC heat ﬂow curves of PU and PU composites (lower: ﬁrst heating curves at a

heating rate of 10 °C min−1; upper: ﬁrst cooling curves at a cooling rate of 10 °C min−1; inset: after 14 days). Reproduced from Ref. [18] with permission.

(SWCNTs-g-PU). All the samples were treated at 120 °C to remove previous thermal history and kept at room temperature to crystalline for 30 days due to slow crystallization rate of the PCL-based PU. No crystallization peak found during cooling program indicated the slow crystallization of the soft segment in the PCL-based PU and its nanocomposites. The inset ﬁgure reveals the reformation of crystals in the soft segment after all the melted samples were kept for 14 days at room temperature. According to the melting peaks around 40 °C, the crystallinity of the PCL-based PUs was found to be increased with the increasing incorporation of the functionalized SWCNTs. It should be resulting from the nucleating effects of CNTs. In contrast, pristine SWNCTs failed to show the nucleating effect and negatively affected the crystallinity of PCL-based PU. X-ray diffraction (XRD) could be used to detect the changed crystalline structure of the soft segment in the presence of CNTs and support the conclusion made from DSC results. Figure 7.9 shows the XRD patterns of the PCL-based PUCNs with SWCNTs and SWCNTsg-PU, respectively. The peak at 20 ° belongs to the crystalline structure of PCL soft segment, and the increase in the integrated area of the peak is associated with increased crystallinity of PCL soft segment with SWNTs-g-PU instead of SWCNTs. Here, it could be postulated that bad dispersion for SWCNTs was the main reason for their negligible nucleating effect in comparison with SWCNTs-g-PU.

7.2 PU/Carbon Nanotube Nanocomposites (PUCNs)

Figure 7.9 XRD patterns for the PCL-based PU and its nanocomposites with SWCNTs and

SWCNTs-g-PU, respectively. Reproduced from Ref. [18] with permission.

Not only is the interfacial structure brought into PUs by CNTs, the signiﬁcant change is also found in the morphology of PUCNs. It can be reﬂected by the change of the Tg of the soft segment (Tg,soft). The characterization of Tg,soft is normally performed in dynamic mechanical analysis (DMA) and standard differential scanning calorimetry (DSC). The effect of CNTs on the Tg,soft seems very complex. The DMA study by Xiong et al. [22] discloses that the Tg,soft of a polyoxytetra methylene (PTMO)-based PU clearly observes the increase of ∼11 °C with the addition of 2 wt% functionalized MWCNTs according to the peaking position of tan δ associated with glass transition. In this system with covalent bonding between CNTs and PU, it was considered that the nanotubes could have more restriction on mobility of the soft segment. The increase in the damping capacity suggesting the improved mobility of PU chains might be caused by the decrease in crosslinking density due to the consumption of some NCO groups by functionalized nanotubes. Another case studied by Xia et al. [13] showed that both of SWCNTs and MWCNTs had weak effects on the Tg,soft and damping capacity of a tripolyetherbased PU, which might be due to the weak interaction between pristine CNTs and PU matrix. The DSC curves in Figure 7.10 show that 1 wt% MWCNTs decrease the Tg,soft of a polybutylene adipate (PBA)-based PU by ∼8 °C. Similar to Xiong’s study, it was stated that the covalent bonding was formed between CNTs and the hard segment of the PU. But the effect of CNTs on Tg.soft pointed to a totally different direction. PBA is a semicrystalline polyester polyol. A new peak of 54 °C appearing in the DCS curves of the nanocomposites indicated the formation of a new crystalline structure in the soft segment due to the nucleating effects of the nanotubes (similar to Xia’s study). As a result, microphase separation was enhanced with the less hard segment existing in the soft segment and Tg,soft was decreased accordingly.

181

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties 2 a:0 b:0.33 wt.% c:0.67 wt.% d:1.0 wt.% e:1.33 wt.% f:1.67 wt.% g:2.0 wt.%

0 Heat Flow (mW)

182

Tg –2 g f

–4

e d c b a

–6 –8

–50

0

50

100

150

Temperature (°C) Figure 7.10 DSC curves of pure PU and MWCNT–PU nanocomposites. Reproduced from Ref.

[23] with permission.

Figure 7.11 ATR-FTIR spectra of PU, 0.7 wt% pristine SWNT/PU, and 0.7 wt% SWNT-g-PU/PU

composites. Reproduced from Ref. [18] with permission.

Microphase separation between the soft and hard segments is the main characteristic of PU morphology. Cooper and his co-workers [24] ﬁrst proposed a FTIR method to measure the degree of phase separation in segmented PU. It is noted that hydrogen bonding could be formed between the –NH groups serving as proton donors and the –C=O group in the hard segment acting as proton acceptors. Hydrogen bonded –C=O and free –C=O can result in split peak of –C=O in FTIR spectra as shown in Figure 7.11 into ∼1705 cm−1 and ∼1728 cm−1, respectively. The intensity ratio of these two peaks (A1705/A1728) can be used to estimate the degree of hydrogen bonding, which is named as hydrogen bonding index R. The

7.2 PU/Carbon Nanotube Nanocomposites (PUCNs)

degree of phase separation (DPS) is deﬁned to be A1705/(A1705 + A1728) based on the understanding that the formation of the hydrogen bonding inside urethane (–NH– CO) group facilitates the enhancement in the phase separation. The calculated value of DPS for PU was increased from 0.49 to 0.68 with 0.7 wt% SWCNTs-g-PU, whereas 0.7 wt% SWCNTs exhibited a negligible effect on the DPS of PU. It is clearly seen that the microphase separation of PU is increased with increasing addition of SWCNT-g-PU, in which SWCNTs-g-PU shows stronger effect on phase separation than SWCNTs. Many conventional techniques are introduced in this section to gain the understanding of the nanoeffects on the morphology of PUCNs. The relationship between the complex morphology and properties should be necessarily investigated to achieve the ﬁnal application of PUCNs in real world from fundamental understanding. The following section will address detailed introduction to this issue. 7.2.3 Physical Properties of PUCNs

“Lightweight, strength, and function” are the basic goals of developing PNCs. CNTs are expected to take an important part in doing these jobs due to their super intrinsic mechanical, electrical, and thermal properties. Loads of studies show that the physical properties of polymers can be signiﬁcantly improved by CNTs including mechanical, electrical, and thermal properties. High cost of CNTs is the barrier for the commercialization of CNTs-based nanocomposites. How to maximally transfer the super properties of CNT to polymeric materials is the challenge ahead of us to reduce the unnecessary cost. It has been understood that it needs different ways to maximally deliver each super property of CNTs to polymeric matrices. The ﬂexible forms of PU provide the possibility to smartly tailor a unique morphology for the speciﬁc delivery. First of all, our discussion goes for the mechanical reinforcing effect of CNTs in the PU matrix. The dispersion of CNTs and the strength of CNT-PU interface are two key issues determining the level of the reinforcement. The approaches to engineer good dispersion and strong interface have been discussed above. The chemical bonding or crystalline layers between CNTs and PU are the ideal interfaces for transferring mechanical properties of CNTs. Here, we presented selected results to discuss the role of the interface in the mechanical reinforcement. Figure 7.12 shows stress–strain curves of the PUCNs with raw MWCNTs and modiﬁed MWCNTs. The modiﬁed MWCNTs can form the covalent interaction with the PU matrix. Clearly, 4 wt% modiﬁed nanotubes signiﬁcantly increase Young’s modulus and stress at break by 3 and 1.3 times, respectively. Without covalent interface, the same percentage of raw MWCNTs presented much less good reinforcing effects due to poor stress transfer via the weak interface. With more modiﬁed nanotubes, deteriorated dispersion may be the reason for the achievement of less good reinforcements. The contribution of the crystalline interface to effective stress transfer was also found in the system of semicrystalline PU. Table 7.2 summarizes the

183

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties 30 25 PU-M4 Stress (MPa)

184

20 Pure PU 15 PU-R4 PU-M6

10 5 0 0

2

4

6

8 Strain

10

12

14

Figure 7.12 Stress–strain curves of pure PU

number after R or M indicates the percentage and PU/MWCNT nanocomposites. Additional of the MWCNTs. Reproduced from Ref. [19] note: R and M means raw MWCNTs and with permission. modiﬁed MWCNTs, respectively, and the

Table 7.2 Tensile properties of PU and PU/SWCNT nanocomposites.

Maximum stress 50% 100% 200% 300% (MPa) strain strain strain strain

Extraction at Breat %

Young’s modulus (MPa)

3.46 3.89 4.12 4.59 6.63

872 969 916 862 852

7.94 9.62 11.77 14.44 22.10

Stress (MPa)

PU PU/0.35 wt% SWNT PU/0.7 wt% SWNT PU/0.35 wt% SWNT-g-PU PU/0.7 wt% SWNT-g-PU

4.34 4.89 5.19 5.38 7.21

6.01 6.57 6.80 7.10 8.87

7.67 8.34 8.74 9.17 11.18

31.00 45.00 39.49 37.13 38.09

Reproduced from Ref. [18] with permission.

mechanical properties of PCL-based PUCNs. Young’s modulus of PCL-based PU is increased by ∼178% with the incorporation of 0.7 wt% SWCNTs-g-PU, whereas the same percentage of pristine SWCNTs can only improve Young’s modulus by ∼48%. There map to strong PUCNs is almost clear, in which the achievement of good dispersion and strong interface are two big milestones. However, it seems that there is another map to transfer the electrical properties of CNTs to polymers. In principle, the formation of conductive pathway by conductive nanoﬁllers is the condition for insulating-conductive transition in PNCs. Percolation threshold is

7.2 PU/Carbon Nanotube Nanocomposites (PUCNs)

log (Electrical conductivity) (S/m)

the parameter to evaluate the minimum amount of conductive nanoﬁllers needed to generate this transition. Conductive nanoﬁllers with higher aspect ratio should have lower percolation threshold. Geometrically, CNTs have a tubular structure in one dimension, which has higher aspect ratio than other conductive carbon nanoﬁllers. With the advantages of high aspect ratio and super conductivity, it is initially expected that highly conductive PNCs can be achieved by very low incorporation of CNTs. The publications show that the percolation threshold can reach down to around 2–4 wt% for PUCNs with randomly dispersed CNTs, which is much lower than other PUNs with such as carbon black or silver particles. From the side of polymer matrices, reducing free volume for location of CNTs in polymer matrices has been reported to be another route to lower down the percolation threshold. In comparison with the PNCs prepared by melt and solution blending, the system of polymer latex can provide limited space for CNTs to form a conductive pathway since both of CNTs and latex are exclusive to each other. For the mixture of polymer latex and CNTs, all CNTs are pushed to the interstitial space between latex particles to form a continuous conductive pathway in the polymer matrix during the drying process [7, 25]. This idea should work out in the system of PU latex. Figure 7.13 shows the electrical conductivity of the PUCNs prepared from the mixture of PU latex and MWCNTs. The percolation threshold is observed to be around 0.2 wt%. It was also discovered that the PU foam with porous structures is another matrix to supply limited space for the accommodation of CNTs. Xu et al. [26] fabricated PU foams in the presence of the CNTs using water as a blown agent. SEM images in Figure 7.14 show that the CNTs are pushed to the walls of each pore in the PU foam. In this speciﬁc case, the percolation threshold was regarded as a function of the density of the PU foam instead of CNT concentration. The electrical conductivity sharply increased as the density increased from 0.03 to 0.05 g cm−3. Figure 7.15 discloses the alternation to the morphology of the PU foams with decreasing density, in which the wall of pores turns to be

–4 –5 –6 –7 –8 –9

–10

0

1

2

3 4 5 6 MWCNT content (wt%)

7

8

Figure 7.13 Electrical conductivity of MWCNT/WPU nanocomposites. Reproduced from Ref.

[8] with permission.

185

186

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties

a)

b)

Figure 7.14 SEM micrographs of the fracture

and cell struts therein indicated by the arrows). (b) High-magniﬁcation micrograph of the cross-section of the cell strut in a. Reproduced from Ref. [26] with permission.

surface of the ultralightweight CNT/PU foam composite with a density of 0.05 g cm−3. (a) Low magniﬁcation (with some cell walls

a)

b) Cell

Cell

c) d) Cell

Cell Cell

Cell Cell

The cross section of the cell strut Cell

Figure 7.15 Schematic diagram of the

microstructural changes in the CNT/PU foam composites with the decrease of density. The thin black lines represent the CNTs and the wide lines represent the boundaries of the

cells. The density decreases gradually from (a) to (d). The arrows in the cells show the growth directions of the air bubbles during foaming. Reproduced from Ref. [26] with permission.

7.2 PU/Carbon Nanotube Nanocomposites (PUCNs)

thinner and pore size becomes bigger. The thinner walls for less settlement of the nanotubes led to the damage to the continuity of conductive CNT network and decrease in the electrical conductivity. Many communications have disclosed that CNTs are super thermal conductors according to their nearly perfect crystalline lattice structure and free length of path for phonon and electron transport. The thermal conductivity of SWCNTs has been theoretically demonstrated at ∼6000 W m−1 K−1 at room temperature by molecular dynamics simulations. The thermal conductivity of isolated MWCNTs was experimentally measured at ∼3000 W m−1 K−1. Due to their outstanding thermal conductivity, CNTs have been considered as potential candidates to improve the thermal conductivity of polymers and develop thermal management materials. Researchers initially believed that the percolated CNT network in the polymer matrix should facilitate the phonon transport as well as the electron transport. Theoretically, the thermal conductivity of the composites ﬁlled with 0.1 wt% MWCNTs can be six times that of pure polymers. However, most of experimental results were published to release a disappointing message indicating the thermally transporting performance of the CNTs failed to match with the theoretical prediction. Strong interfacial phonon scattering has been commonly considered as the biggest bottleneck limiting the thermally transporting performance of CNTs in polymer composites. It is found that the thermal conductivities of PUCN elastomers exhibited slight increase which was also far away from the prediction [13]. Although the relative high thermal conductivity of 0.5 W m−1 K−1 was reported in the PUCN with 20 vol% MWCNTs, it was very difﬁcult to retain the mechanical properties and the fabricating cost was very expensive [27]. Novel technologies are being explored to effectively reduce interfacial phonon scattering in order to get out most thermal conductivity of CNTs and develop affordable thermally conductive composites for thermal management. The promising way is to achieve a continuous nanotuberich phase in polymeric matrices, by which the nanotube-polymer interfacial area can be signiﬁcantly reduced. Du et al. [28] used nitrogen gasiﬁcation to remove the polymer matrix from the composites to form a freestanding nanotube framework, and then reinjected epoxy to the nanotube framework which was considered a continuous nanotube-rich phase facilitating the heat transport through the epoxy matrix. This unique structure allowed 220% improvement of the thermal conductivity by the incorporation of 2.3 wt% SWCNTs. Huang et al. [29] designed a novel composite structure for effectively reducing the interfacial phonon scattering by using in-situ injection of polymers into CNT arrays. The aligned CNT arrays vertically stood from the bottom to the top of the composite ﬁlms with all CNT tips protruding out of the surface to form an ideal thermally conducting pathway from one side to another. The thermal conductivity of the low-thermal-conductivity silicone elastomer was improved by about 280% with the incorporation of 0.4 vol% aligned CNTs. In both systems, the contribution of randomly dispersed nanotube bundles to the enhancement of the thermal conductivity signiﬁcantly is much less than the continuous nanotube-rich phase. The advantages for the latex technology in the preparation of electrically conductive PNCs have been introduced above. The limited free volume available for the settlement of CNTs allows formation of

187

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties 0.30 Thermal conductivity (W/mK)

188

0.27 0.24 0.21 0.18 0.15 0.12

0

2

4 6 MWCNT content (wt%)

8

Figure 7.16 Thermal conductivity of MWCNT/polyether-based WPU nanocomposites as a

function of MWCNT content. Reproduced from Ref. [8] with permission.

a conductive CNT network with very low percolation threshold. It was found that the created continuous nanotube-rich phase going through the boundary area between latex particles could be helpful for the reduction in interfacial phonon scattering. Figure 7.16 shows the thermal conductivity of polyether-based WPUs ﬁlled with MWCNTs. The enhancement in the thermal conductivity is much higher than those with randomly dispersed nanotubes. A phonon is a quantum of crystal vibrational energy, which has two fundamental lengths: wavelength and mean free path. An amorphous phase without long-range crystalline structure limits the free length of path for phonon transport in solid. It was considered that, in PNCs, an amorphous interface is inclined to arise the stronger interfacial phonon scattering than the crystalline interface. “Repairing” amorphous interface is also considered as a possible way to reduce the interfacial phonon scattering. Nucleating ability of CNTs in semicrystalline polymer has been practically applied to form crystalline layers around CNT acting as a crystalline interface. It was found that the crystallites formed at the CNT–polyethylene interface could lift the performance of CNTs in improving the thermal conductivity of polyethylene. In the system of PU latex, a semicrystalline PU dispersion that was synthesized from PCL was used as latex host to accommodate CNTs. As shown in Figure 7.17, a much higher improvement of the thermal conductivity was found in comparison with the polyether-based PUCNs. The thermal conductivity of the PCL-based PU nearly tripled with the incorporation of 3 wt% MWCNTs.

7.3 PU/Clay Nanocomposites (PUCLN)

Nanoclay is low-cost nanoﬁller for PNCs since it can be richly sourced from natural clay minerals. Nanoclay has a layered structure consisting of tetrahedral and octahedral sheets [30]. It is commonly referred to 1 : 1 or 2 : 1 clay-based ratio of tetrahedral and octahedral sheets. The 1 : 1 clay includes the typical examples of

Thermal conductivity (Wm–1 k–1)

7.3 PU/Clay Nanocomposites (PUCLN) 0.6 0.5 0.4 0.3 0.2 0.1 0

1

2 3 4 5 MWCNT content (wt%)

6

7

Figure 7.17 Thermal conductivity of the

samples (•) were prepared with consistent MWCNT/PCL-based PU nanocomposites as a weight ratio (1 : 1) of SDS to MWCNTs. function of MWCNT content. The samples (䊐) Reproduced from Ref. [17] with permission. were consistently loaded with 1 wt% SDS. The

kaolinite and serpentine. Smectite particularly known as montmorillonite (MMT) is a 2 : 1 clay which has an octahedral sheet in center with two tetrahedral sheets in neighborhood. A layered structure of MMT consists of 1-nm-thick silicate layer and inter layer [30]. The silicate layers are bonded to each other by van der Waals and ionic forces. The space of the inter layer can be expanded by the intercalation of other chemical components. It has been well-known that polymer/clay nanocomposites can be classiﬁed into “intercalated,” “ﬂocculated,” and “exfoliated” types [31]. In the intercalated type, the structure of nanoclay is maintained and the inter layers are ﬁlled with polymer chains. In the exfoliated type, the silicate layers are forced to be homogenously dispersed in polymer matrices as the interlayer forces are completely destroyed to allow free movement of the silicate layers away from each other. The ﬂocculated type is conceptually the same as intercalated one. The difference is silicate layers are sometimes ﬂocculated due to the hydroxylated edge–edge interaction of the silicate layers. Since the ﬁrst discovery of nylon/clay nanocomposites, it has been >20 years for the development of high-performance polymer/clay nanocomposites for industrial and scientiﬁc interests. The core work is persistently focused on the production of “intercalated” or “exfoliated” structure using a variety of techniques to beneﬁt the mechanical, thermal, and barrier properties of polymers. It this section, it is our interest to discuss this beneﬁt in PU materials. 7.3.1 Fabrication 7.3.1.1 Exfoliation and Intercalation of Nanoclays in PU Matrix The idea for the in-situ polymerization of PUNs is similar, in which nanoﬁllers should be dispersed in macromolecular polyols ﬁrstly for further polymerization

189

190

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties

in the presence of nanoﬁllers. It was found that high-speed mixing for several hours at a temperature (50–80 °C) could effectively enforce the homogenous dispersion of nanoclay in a trifunctional polyether polyol [32]. The formation of intercalated or exfoliated structures heavily depends on the types of nanoclay [33]. Inorganic nanoclay generally has inorganic cations such as Na+ or Ca2+ in the interlayer to counterbalance the negatively charged silicate layers. The interlayer space with a very small d-spacing of ∼1.1 nm is hostile to the intercalation of polymer chains. However, these inorganic cations are exchangeable with cationic surfactants such as quaternary ammonium to form so-called organoclay. Organoclay is hydrophobic and has enlarged d-spacing due to the existence of long-chain surfactants in the interlayer space. It enhances the possibility for the intercalation or exfoliation of nanoclay. As a kind of effective additive for polymers, many types of commercial organoclays have been supplied from many companies such as Southern Clay Products, Inc. An intercalated PU/MMT nanocomposite was synthesized following an intercalated polyol/MMT mixture was ﬁrstly achieved with 3-h high-speed mixing at 50 °C. Figure 7.18 shows the wide angle X-ray diffraction (WAXD) patterns of MMT, the PU, and the PU/(21.5 wt%)MMT nanocomposite [32]. The d-spacing of MMT was calculated to be 1.1 nm according to Bragg’s equation. The shift in the diffraction angle indicates that the PU chains were intercalated into the interlayer space and the d-spacing was expanded to 1.6 nm. However, exfoliation of MMT in PU matrices is hardly produced. Organoclay has the chance to be intercalated even exfoliated in PU matrices. Cloisite® 20A (C20A) modiﬁed from natural MMT is an organoclay supplied by Southern Clay Products, Inc. It can be intercalated in a trifunctional polyether polyol (Lupranol 2090, molecular weight = 6000, function = 3) with 1-h high-speed

Figure 7.18 Wide angle X-ray diffraction patterns of the PU/MMT nanocomposites: (1) PU,

(2) MMT, and (2) PU nanocomposite with 21.5% MMT. Reproduced from Ref. [32] with permission.

7.3 PU/Clay Nanocomposites (PUCLN)

Figure 7.19 Wide-angle X-ray diffraction patterns of PU/C20A nanocomposites with different

intercalation times (10 wt% C20). Reproduced from Ref. [34] with permission.

mixing at 60 °C. Figure 7.19 illustrates the WAXD patterns of the PU/C20(10 wt%) nanocomposite for different mixing time. The diffraction peak shifts to a lower diffraction angle with 1-h mixing, which indicates the expansion of the interlayer space of C20A with the intercalation of polyether polyol chains. The diffraction peaks remain unchanged with increasing mixing time. Two diffraction peaks are observed in the WAXD pattern of C20A, which can be attributed to the d-spacing of 1.1 and 2.3 nm, respectively. C20A should be a mixture of MMT with a d-spacing of 1.1 nm and organically modiﬁed MMT having a d-spacing of 2.3 nm. The dspacing of MMT is expanded from 1.1 nm to 1.7 nm in the PU/(10 wt%)C20A nanocomposite, and the d-spacing of modiﬁed MMT increases from 2.3 nm to 3.8 nm. This result reveals that the polyether polyol chains can be intercalated into both of MMT and modiﬁed MMT in C20A. The dispersion of other Cloisitebranded nanoclay from Southern Clay Products, Inc., in Luprane 2090 was also investigated using the some mixing method. It was found that all of Cloisite® 25A, 10A, 20A, and 15A could be intercalated with the polyether polyol chains. Our study showed that Cloisite®C30B was the only organoclay being exfoliated in Lurpane 2090 as several types of Cloisite-branded organoclay were investigated including C25A, C10A, C20A, and C15A [33]. The organic modiﬁer for 20A and 15A is dihydrogenated tallow quaternary ammonium (2M2HT). 25A and C30B have an organic modiﬁer of dimethyl hydrogenated tallow 2-ethylhexyl quaternary ammonium (2MHTL8) and methyl tallow bis-2-hydroxyethyl quaternary ammonium (MT2EtOT), respectively. The WAXD patterns in Figure 7.20 show that the diffraction angle totally disappears in the range of 1 °–10 ° for C30B and other organoclay still can observe the diffraction angle with lower positions. It is the indication of the exfoliation of C30B and the intercalation of the other types of organoclays in the trifunctional polyol. TEM images (a and b) in Figure 7.21

191

192

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties

Figure 7.20 WAXD patterns of PU/clay nanocomposites with various clay (1) MMT; (2) C30B;

(3) 25A; (4) 10A; (5) 20A (6) 15A. Note: The mixing of the Cloisite®clay and Lurpane 2090 was carried out at 80 °C for 4 h. Reproduced from Ref. [33] with permission.

a)

b)

Figure 7.21 TEM images of PU/clay nanocomposites: (a) C30B and (b) C20A. Reproduced

from Ref. [35] with permission.

directly reveal the exfoliated C30B and intercalated C20A nanoclay in a PU matrix, respectively. Random black lines represent the silicate layers. Here, it should have more discussion on the processing details in terms of the exfoliation of C30B. The effect of mixing temperature is obviously shown in Figure 7.22, in which it is clearly seen that no exfoliation and intercalation occurs at a

7.3 PU/Clay Nanocomposites (PUCLN)

Figure 7.22 XRD patterns for PU/clay nanocomposites prepared from polyol/clay C30B

dispersions after mixing for 4 h at 20, 40, 60, 80, and 100 °C. Reproduced from Ref. [33] with permission.

Figure 7.23 XRD patterns for polyol/clay C30B liquid dispersions after mixing for 4 h at 80 °C.

Reproduced from Ref. [33] with permission.

relative low temperature of 20 °C, and the intensity of the diffraction peak at ∼5 ° gradually decreases with increasing mixing temperature and the peak ﬁnally disappears when the temperature reaches 80 °C and even higher. Therefore, the mixing temperature matters with the achievement of the exfoliation of C30B. The XRD patterns in Figure 7.23 illustrate the effect of polyol types on the exfoliation of C20B clay. A linear PCL polyester polyol with an average molecular weight of ∼2000 g mol−1 can be intercalated into C30B indicated by the downshift of the diffraction peak belonging to C30B. A very weak diffraction peak is found in the C30B

193

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties 34.0 Å 0 wt% TMP 35.0 Å Intensity (a.u.)

194

1.7 wt% TMP

37.1 Å 3.4 wt% TMP 42.8 Å

5.2 wt% TMP

7.0 wt% TMP 2

4

6 2 theta (°)

8

10

Figure 7.24 WAXD of PU/Cloisite C30B nanocomposites with varying amount of trimethylol

propane. Reproduced from Ref. [36] with permission.

mixed with a linear polyether polyol with an average molecular weight of ∼4000 g mol−1, indicating the formation of an intercalated or partially exfoliated structure. The highest exfoliation takes place in the mixture of C30B and a trihydroxylbranched polyether polyol with an average molecular weight of ∼6000 g mol−1. It can be concluded that the structure of polyol is another factor affecting the exfoliation degree of C30B. A study was also reported to disclose the relationship between the crosslinking degree of PU and the exfoliation degree of C30B [36]. Trimethylol propane (TMP) was used as branching agent to increase the crosslinking degree of PU with C30B. The WAXD patterns in Figure 7.24 indicate that the d-spacing associated with 001 basal plane of C30B increases with the increasing amount of the branching agent, and C30B can be fully exfoliated as the percent of TMP reaches 7 wt%. 7.3.1.2 Rheological Behavior of Polyol–Nanoclay Mixture In addition to the dispersion of nanoclay, another issue also needs to be considered in the fabrication of PUCLNs, which is the rheological behavior of polyol–nanoclay mixture. It has been found that the rheological behavior is related to intercalated and exfoliated types of polyol–nanoclay mixture. The effect of nanoclay types on the viscosity of a trifunctional polyol (Luprane 2090) was comprehensively investigated (R). As 5 wt%C30B was mixed in the polyol in a Haake VT500 rheometer at a shear rate of 299.7 s−1, the viscosity of the mixture was measured against mixing temperature and time. As shown in Figure 7.25, the viscosity of the mixture slightly increases from 0.8 to 1.0 Pa s after 5.5 h mixing at 40 °C. The viscosity of the mixture increases up to 1.55 Pa s after 2.5 h mixing at the temperature of 60 or 80 °C. The viscosity of the polyol/5 wt%C20A mixture at 60 or 80 °C observes a slight increase with the mixing time at a shear rate of 299.7 s−1. The

7.3 PU/Clay Nanocomposites (PUCLN) a)

b)

Figure 7.25 Viscosity of a trifunctional polyol (Luprane 2090) in the presence of 5 wt% C30B

(a) and C20A (b) as a function of temperature and time. Reproduced from Ref. [33] with permission.

maximum viscosity is much lower than that of the polyol/5 wt%C30B mixture. The relationship between the rheological behavior and dispersion type is clear, that is, exfoliated C30B generates higher viscosity than intercalated C20A because the exfoliation can produce more individual silicate layers to restrict the mobility of polyol chains. Moreover, Xia et al. analyzed this relationship in a quantitative way when ﬁtting viscosity data into the Herschel–Bulkley equation [33]. The d-spacing of nanoclay in the polyol was evaluated by XRD patterns shown in Figure 7.20. As shown Table 7.3, it is clear to see that the shear thinning parameter n decreases with increasing d-spacing of the nanoclay intercalated with polyol chains, and the lowest n of 0.77 corresponds to the fully exfoliation of C30B in the polyol.

195

196

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties Table 7.3 Relationship between rheological behavior and clay state in Luprane 2090 polyol.

Dispersion

ηL (Pa s)

ηH (Pa s)

n

Clay state (XRD evaluation)

Blank polyol Polyol/Na+ clay Polyol/15A Polyol/20A Polyol/25A Polyol/10A Polyol/30B

1.87 2.10 3.30 6.15 12.30 13.80 28.30

1.63 2.03 3.09 4.39 7.24 7.67 8.60

0.99 1.00 0.99 0.90 0.87 0.87 0.77

– Low intercalation Low intercalation High intercalation High intercalation High intercalation Exfoliation

Reproduced from Ref. [33] with permission.

Figure 7.26 dCp/dT versus temperature signals for blank polyol 6000 and polyol/clay

intercalated nanocomposites.

7.3.2 Morphology and Characterization

Nanoclay has more interesting nanoeffects on the two-phase morphology of PU in comparison with CNTs due to their layered 2D structures. It has been demonstrated that the molecular conﬁnement in the interlayer of nanoclay has a signiﬁcant effect on the glass transition of PU [37, 38]. Figures 7.26 and 7.27 show dCp/dT signals as a function of temperature for the intercalated polyol/clay and PU/clay nanocomposites, respectively. The integrated peak area of dCp/dT curves is associated with the increment of heat capacity Cp during the period of glass transition. The decrease of ΔCp should be attributed to the conﬁnement of polyol or PU chains in the interlayer spacing. A concept of characteristic length (ξa) identifying the size of cooperatively rearranging region was used to understand the glass transi-

7.3 PU/Clay Nanocomposites (PUCLN)

Figure 7.27 dCp/dT versus temperature signals for blank PU and PU/clay intercalated

nanocomposites.

tion behavior of polyol chains conﬁned in the interlayer spacing. The ξa of the glass transition for the polyols and PU was calculated to be 3.20 and 1.45 nm, respectively, in Ref. [37]. These values were found to be applicable to their nanocomposites. The original d-spacing of nanoclays for S0 (MMT), S3(triethanolamine hydrochloride-modiﬁed nanoclay), C30B, C25A, and C20A was 1.17, 1.37, 1.81, 1.86, and 2.42 nm, respectively, and correspondingly expanded to 1.59, 1.65, 2.05, 3.26, and 3.56 nm after intercalation. The thickness of silicate layer is around 1 nm. The interlayer distance (Δd) should be 0.59, 0.65, 1.05, 2.26, and 2.56 nm, respectively. As the ξa of the polyol chain (3.45 nm) is larger than the Δd of nanoclay, the conﬁnement of polyol chain takes place and decreases ΔCp crossover glass transition. PU chains hold smaller ξa (1.45 nm) due to the restriction of hard segment on the mobility of the soft segment (polyol). For Δd < 1.45 nm, PU chains are conﬁned to illustrate decreased ΔCp during glass transition. In the interlayer with Δd > 1.45 nm, PU chains have nearly the same ΔCp as the bulk PU. The nanoeffect of nanoclay can also be observed on the microphase separation of PU, which can be detected by several techniques including atomic force microscopy (AFM) and small-angle X-ray scattering (SAXS). The FTIR method proposed by Cooper and co-workers was also simply applied to determine the hydrogen index (R) and degree of phase separation (DPS) of intercalated PU/C20A nanocomposites [39]. R and DPS were found to increase with increasing C20A content. SAXS is a more precise tool to quantitatively evaluate the two-phase structure of PU by providing the data of interdomain spacing, domain size, and interfacial thickness [40]. Figure 7.28 illustrated typical SAXS intensity proﬁles (I(S)S2 vs. 2θ) for PU/C20A nanocomposites. The one-dimensional correlation function F(Z) that is related to the electron density distribution within specimens is expressed as follows: F (Z ) =

1 Q

∫

∞

0

S2I (S ) cos (SZ ) dS

197

198

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties

Figure 7.28 SAXS intensity proﬁle I(S)S2 versus scattering angle (2θ) for the polyurethane

(26 wt% hard segment)/clay nanocomposites (clay content: a: 0%, b:1%, c:3%, d:5%). Reproduced from Ref. [41] with permission.

For a two-phase system, the invariant quantity Q of overall mean-square electron density ﬂuctuation can be obtained by integrating S2I(S) over the range of scattering angle. S is the magnitude of scattering vector (S = (4π/λ)sinθ, 2θ is the scattering angle, and λ is the wavelength), and I(S) is the scattering intensity. The correlation function can be obtained through the Fourier transform of the SAXS intensity proﬁle. The correlation function curves for PU/C20A nanocomposites with different hard-segment contents are shown in Figure 7.29. The interdomain repeat distance (L), deﬁned as the average distance between two hard domains, can be obtained from SAXS intensity proﬁles according to Bragg’s equation: L = 2π/Smax, where Smax is deﬁned as the observed ﬁrst-order scattering maximum. There is a primary local maximum at a position Z in correlation function curves, which is associated with interdomain repeat distance. The domain size can be obtained by the interpretation of the 1D correlation function based on the assumption that the specimen ideally consists of an ensemble of isotropically distributed stacks of alternating crystalline and amorphous lamellae. The lamellar structure of nanoclay is neglected as a minor component in the PU matrix. The interdomain repeat distances and domain sizes evaluated from Figure 7.29 are summarized in Tables 7.4 and 7.5, respectively. Overall, the interdomain repeat distance is observed to decrease with the increasing addition of C20A. The most signiﬁcant decrease from 67 to 51 nm is found in the nanocomposites with 18 wt% hard segment. The interdomain repeat distance is slightly decreased by 7.1% from 84 to 78 nm at 26 wt% hard segment, and 3.4% from 89 to 86 nm at 36 wt% hard segment. Clearly, higher soft-segment content leads to stronger effect of C20A on the interdomain repeat distance. Table 7.5 discloses that the domain size of the hard segments is in the range of 12–32 nm, and it is nearly independent of C20A content at 18 wt% and 26 wt% hard segment. At 36 wt% hard segment, the domain size slightly decreases with increasing addition of C20A.

7.3 PU/Clay Nanocomposites (PUCLN) a)

b)

c)

Correlation function F(Z) for the PU/C20A nanocomposites with (a) 18 wt%, (b) 26 wt%, and (c) 36 wt%, respectively. Reproduced from Ref. [41] with permission.

Figure 7.29

Table 7.4 Interdomain repeat distance (L, nm) for PU/C20A

nanocomposites.. Content of hard segment

18% 26% 36%

Organoclay content 0%

1%

3%

5%

67 84 89

60 84 88

59 81 86

51 78 86

Reproduced from Ref. [41] with permission.

199

200

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties Table 7.5 Domain size (d, nm) for the PU/C30B

nanocomposites. Content of hard segment

18% 26% 36%

Organoclay content 0%

1%

3%

5%

12 16 22

12 17 20

14 15 19

12 15 18

Reproduced from Ref. [41] with permission.

AFM provides a direct approach to observe microphase-separated structure. Phase imaging using a tapping mode can be effective to map the submicron properties of multicomponent polymeric systems based on the relative elasticity of individual components. The scales of AFM phase images should be set to induce higher phase offset for hard phases, so that bright and dark regions correspond to the hard phase and soft phase, respectively. Figure 7.30 shows the tapping mode phase images of PU/C20A nanocomposites with 36 wt% hard segment. The aggregates of the hard domain in pure PU exhibit spherical structures with a size of ∼800 nm. The addition of nanoclay reduces the size of the aggregates to be ∼500 nm, whereas the SXAS results above suggest that the size of the hard domain is little affected. It should be noted that the random existence of big aggregates in the nanocomposites is attributed to the undispersed clay tactoids instead of the hard-segment aggregates. 7.3.3 Physical Properties

It should come to the beneﬁts for physical properties of PU brought from nanoclay after gaining the understanding of unique morphology of PUCLNs. Silicate sheets are featured as mechanically strong and thermally stable nanoscale block, and they can be used to enhance the mechanical properties and thermal stability of PU. Two-dimensional structures can make them play as barrier to improve the permeability of PU. Therefore, the topics in this section will mainly cover enhanced mechanical, thermal, and barrier properties of PUCLNs. 7.3.3.1 Mechanical Properties The mechanical properties of PUCLNs have been comprehensively examined. It has been reported that a series of intercalated PUCLNs presented exceptionally mechanical enhancement in comparison with pure PU. It was reported that the strength and strain at break of pure PU can be increased by >44% and 20%, respectively, with 21.5 wt% MMT [32]. Organoclay has much stronger reinforcing effects than MMT for two reasons: (i) organoclays with expanded interlayer spacing

7.3 PU/Clay Nanocomposites (PUCLN) 20 μm a)

b)

10 μm

0 μm 0 μm c)

10 μm

20 μm d)

Figure 7.30 AFM phase images of PU/C20 nanocomposites with 36 wt% hard segment as

function of clay content (a: 0%, b: 1 wt%, c: 3 wt%, d: 5 wt%). Reproduced from Ref. [41] with permission.

can be intercalated with more PU chains than MMT and (ii) the organic modiﬁer on silicate layers can improve the interfacial compatibility between silicate layers and PU matrix. Figure 7.31 illustrates the strain–stress curves of PU/C20A nanocomposites as a function of C20A contents. The tensile strength and elongation at break of the PU is substantially increased by >100% when only 1 wt% C20A is added. The tensile strength and elongation at break peaks at 3 wt% C20A, which are both increased by >150%. Hardness related to yield stress and modulus was also found to be nearly unchanged with increasing addition of C20A up to 5 wt%. This is a unique phenomenon because nanoﬁllers normally make PUs harder and stronger rather than tougher. Fatigue life associated with aging properties under dynamic external forces was also investigated. It reﬂects the ability of PUs to withstand repeated ﬂexing without development of severe cracking. Fatigue failure occurs as a result of the propagation of unstable cracks or defects under cyclic

201

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties A3

10 A5 8 Stress (MPa)

202

A7 6

A1

4

A0

2 0

0

500

1000

1500

2000

2500

Strain (%) Figure 7.31 Stress–strain curves of the intercalated PU/C20A nanocomposites as a function

of C20A contents (A0: 0%wt%, A1: 1 wt%, A3: 3 wt%, A5: 5 wt%, A7: 7 wt%). Reproduced from Ref. [34] with permission.

forces. A measurement of fatigue life was performed at a constant maximum deﬂection of 200% and a test frequency of 2 Hz under room temperature [34]. It was revealed that C20A could substantially improve the fatigue properties of PU and the addition of 3 wt% C20A lead to a maximum improvement approaching 10 times. A mechanism was speculated by Song and his co-workers to explain the signiﬁcant reinforcement in these intercalated PUCLNs, which stated that intercalated organoclay act as physical crosslink points as the result of absorbing PU chains onto the outside surface and gallery of organoclay, respectively. The surface absorption may function as mobile two-dimensional physical crosslink points. The slipping of intercalated PU chains and silicate layers in organoclays occurs when the dynamic force is applied, which could help release concentrated stress, allow more energy dissipation and avoid crack growth. Exfoliated PUCLNs with C30B exhibit different mechanical reinforcement from intercalated ones with C20A [42]. Exfoliated C30B exists as individual silicate layers without gallery space in the PU matrix. The chemical agents on the surface of the silicate layers (CH2CH2–OH) can form chemical bonding with the PU matrix. The silicate layers act as chemical crosslink points and restrict the mobility in PU chains. Thus, exfoliated nanoclay can increase the modulus and strength of PUs but decrease elongation at break in most cases. 3 wt% C30B nanocomposites [42] have the best reinforcing effect with achieving the balance between the amount of reinforcing phase and dispersion status. Higher C30B content could result in the formation of clay agglomeration in the PU matrix. The strong chemical interface effectively transfers stress from the PU matrix to the exfoliated silicate layers, resulting in nearly 100% and 54% increase in tensile strength and 100% tensile modulus, respectively. But the elongation at break seems to be nearly constant with the addition of C30B.

7.3 PU/Clay Nanocomposites (PUCLN) Table 7.6 Stress relaxation data of polyurethane/C20A nanocomposites at 26 wt% hard segment content.

Sample

Initial stress σ0 (MPa)

Equilibrium stress σe (MPa)

Relaxation ratio (σ0 − σe)/σ0

Relaxation rate at 60 s (10−4)

Relaxation time τ (s)

PU-26 1% C20A 3% C20A 5% C20A

1.44 1.52 1.63 1.87

1.17 1.20 1.23 1.39

0.188 0.210 0.241 0.257

6.80 8.04 9.67 11.50

31.9 43.1 61.8 46.3

Reproduced from Ref. [39] with permission.

Tensile properties of PUCLNs have conﬁrmed the reinforcing effects of intercalated and exfoliated nanoclay. Moreover, investigating the effect of nanoclay on the viscoelastic behavior of PUs such as stress relaxation and creeping carried by Xia et al. [39] can help us to have deeper understanding of the mechanical behavior of PUCLNs. Stress relaxation describes the stress relief of polymers under constant strain, which is a nonlinear process due to the viscoelastic characteristics of polymers. It was noted that the stress relaxation of PU should be divided into three main steps: uncoiling/disentangling of soft-segment chain network in the soft phase, breakup of interconnected hard domain, and pullout of soft-segment chains from hard domains. The ﬁrst relaxation takes place at any strain, and two latter relaxations occur at large strains. Most importantly, it was stated that the characteristic relaxation time at 5–100 s should be attributed to the uncoiling/disentangling of soft-segment chain network in the soft phase. Table 7.6 gives relaxation-related data of the PU/C20A nanocomposite with 26 wt% hard segment. The relaxation ratio and rate increase with the addition of C20A. Meanwhile, characteristic relaxation time related to uncoiling/disentangling of PU chains network in the soft phase is found to be increased with the addition of nanoclay, suggesting that characteristic relaxation time obtained from the relaxation spectrum is not always proportional to the overall relaxation rate. It was considered to be attributed to enhanced degree of microphase separation resulting from the addition of nanoclay. The soft-segment chains are more ﬂexible due to less hard-segment exists in the soft segment, thus leading to slow stress relaxation. Creep describes time-dependent permanent deformation of materials resulting from constant structural stress. The creep of polymers can be divided into two main stages: primary creep and steady-state creep. The creep strain rate decreases with time in the primary creep and is constant in the steady-state creep. Strain recovery occurs with the removal of external load after a creep time. Therefore, the total strain (e) consists of three separate parts e1, e2, and e3. The e1 and e2 are the immediate elastic deformation and delayed elastic deformation, respectively. The e3 is the Newtonian ﬂow. It was found that the e1 and e2 decreased with increasing clay content, indicating lower creep recovery with the addition of C20A. The creep compliance J, the ratio of strain and applied load, can be expressed as

203

204

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties Table 7.7 Creep data of polyurethane/C20A nanocomposites at 26 wt% hard-segment

content. Sample

Instantaneous compliance J0 (105 μm2 N−1)

Equilibrium compliance Je(105 μm2 N−1)

Creep ratio ( Je − J0)/J0

Creep rate at 60 s (10−2)

Retardant time λ(s)

PU-26 1% C20A 3% C20A 5% C20A

3.438 3.325 2.920 2.840

4.350 4.384 4.329 4.031

0.265 0.318 0.483 0.419

6.74 6.90 7.17 7.02

11.5 21.4 24.0 23.8

Reproduced from Ref. [39] with permission.

J = J1 + J2 + J3. J3 can be considered as zero for a crosslinked or highly crystalline polymers. Table 7.7 lists creep-related data of PU/C20A nanocomposites. The instantaneous compliance decreases with the addition of clay, and the nanocomposites observes the nearly same equilibrium compliance except 5 wt% C20A, which is similar to the equilibrium stress during stress relaxation. It is also similarly found that both the creep rate and retardant time increases with the addition of clay, which should also be the result of enhanced phase microseparation. 7.3.3.2 Scratch Resistance and Barrier Performance PU has been widely used as protective coatings for various surfaces in industry. This important application relies on the key properties supplied by PU materials including scratch resistance and barrier properties. Developing high-performance PU coatings is highly expected to help metal materials tackle increasingly harsh environments. In this section, we will discuss that the role of nanoclay in making PU coatings meet increasing demands for scratch resistance and barrier properties. The reinforcing effects of nanoclay in PU have been discussed above. Here, our focusing is placed on the surface scratch resistance of PUCLNs. Nanoindentation refers to use loaded indenter to penetrate the surface of materials, and the penetration depth recorded in a loading and unloading circle is informative for understanding elastic, viscoelastic, and plastic deformation of materials. Kamal et al. [43] reported the study on the subsurface mechanical properties of intercalated PU/ C20A and exfoliated PU/C30B nanocomposites using the nanoindentation technique. Figure 7.32a and b shows the typical indentation loading–unloading curves for PU/C20A and PU/C30B nanocomposites with 26% hard segment, respectively. The reduction in the penetration depth is achieved with the addition of nanoclay, and exfoliated C30B is more effective than intercalated C20A to make PU stronger to resist the penetration of the indenter. The movement of the loaded indenter over the surface of PUCLNs was conducted to evaluate the scratch resistance. Figure 7.33a and b shows the scratch depth proﬁles of the intercalated and exfoliated PUCLNs at the scratching rate of 3 μm s−1 and 5 μm s−1, respectively. At a

7.3 PU/Clay Nanocomposites (PUCLN) a)

b)

Figure 7.32 Typical indentation load–displacement curves for pristine PU and its two types of

PU nanocomposite thin ﬁlms with different clay concentrations. Reproduced from Ref. [43] with permission.

scratch length of 20 μm, the scratch depth on the PU is reduced from ∼957 to ∼448 nm with 3 wt% intercalated C20A. At a scratch rate of 5 μm s−1, the nanoclay makes better contribution to the reduction in the scratch depth. Exfoliated C30B leads to the better improvement in the scratch resistance than intercalated C20A due to the distribution of more single silicate layers into the PU matrix. Barrier property to moisture and gas is another key demand for PU coatings. It is reported that impermeable nanoclays can effectively reduce the permeability of polymers by up to hundred times. The mechanism of permeability has been widely discussed from experimental and theoretical aspects. Nielsen [44] proposed a permeability model for a regular arrangement of platelets based on the assumption, that is, rectangular platelets are uniformly dispersed and perpendicularly

205

206

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties a)

b)

Figure 7.33 Nanoscratch depth proﬁles for PU and its nanocomposite thin ﬁlms (a) 3 m s−1

and (b) 5 m s−1. Reproduced from Ref. [43] with permission.

oriented to diffusion direction. Diffusing molecules are forced to travel through tortuous paths in the composite matrix with longer time due to the existence of impermeable platelets. The tortuosity that decides diffusion coefﬁcient relates to the aspect ratio, shape, and orientation of the platelets. It is also relates to interfacial regions and dispersion types. In terms of PUCLNs, exfoliated types present lower permeability than intercalated types. Better dispersion under sonication resulted in better barrier properties to O2 and CH4 gas [45]. Osman et al. [46] investigated the effects of three intercalated organoclays with different chemical modiﬁers (Nanoﬁl 804, 32, and 15) on the permeability of pure PU. As shown in Figure 7.34, both of Nanoﬁl 32 and 804 could lead to 30% reduction in O2 permeability at 3 vol% organoclays. Surprisingly, Nanoﬁl 15 took negative inﬂuence on the barrier property of PU for O2 gas. It

7.3 PU/Clay Nanocomposites (PUCLN)

Figure 7.34 Dependence of the oxygen transmission and water vapor transmission rates

through the PU-nanocomposites. Reproduced from Ref. [46] with permission.

might result from the weak interaction between Nanoﬁl 15 and the PU matrix due to the use of nonpolar chemical modiﬁer. Osman et al. [46] also demonstrates the decreasing water vapor transmission rates (WVTR) as a function of nanoclay volume fraction. The chemical modiﬁer of Nanoﬁl 804 and 32 contains hydroxyl groups and aromatic moieties, respectively. The stronger the hydrophilicity of chemical modiﬁer, the lesser is the decrease in WVTR. The effect of Nanoﬁl 15 on WVTR was different from that on O2 permeability. It was considered by authors that the clusters of water have larger size than oxygen molecules, and probably cannot travel through the thin gaps between Nano15 and the PU matrix. Thus, the reality is very complicated, which needs to consider more than theoretical models to understand the barrier properties of PUCLNs. 7.3.3.3 Thermal Stability and Flame Retardancy It has been demonstrated that nanoclay can act as promising ﬂame retardants to reduce the heat release rate, change char structure, and decrease mass loss rate of polymers during combustion. It has been commonly suggested that the ﬂame retardant mechanism refers to the formation of carbonaceous-silicate char during burning which can act as barriers to slow heat release and mass loss rate of the gaseous and condensed phases. TGA and cone calorimeter experiments are usually carried out to measure the thermal stability and ﬂame retardancy. Song et al. [47] carried out a study on the thermal stability and ﬂame retardancy of an intercalated PUCLN. The chemical modiﬁers (hexadecyl trimethyl ammonium chloride) on the surface of the organoclays started to degrade about 200 °C, by which acid

207

208

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties

protons were left on silicate layers after the modiﬁer lost oleﬁn and amine groups. The acid site was found to act as catalyst to accelerate the decomposition of the PUCLN at a lower temperature, in which the biggest weight loss at 200–380 °C was attributed to the process of depolycondensation. The degradation of pure PU was completed at 630 °C without any char residue. The char residue of 5.23% for the PUCLNs with 5 wt% organoclay was higher than the solid residue of inorganic phase that was 74% of organoclay concentration (26% was the modiﬁer). It indicated that carbonaceous char formation in pure PU was enhanced by the organoclay, resulting from the catalytic effect of the acid protons in the oxidative depolymerization–crosslinking–charring process. The heat release and mass loss rate were measured by a cone calorimeter. The addition of 5 wt% organoclay led to signiﬁcant decreases in both of heat relate and mass loss rate. A comparative study was also performed to evaluate the advantages of organoclays over an organic ﬂame retardant (melamine polyphosphate (MPP)). It was found that the heat release and mass loss rate of the PUCLNs was lower than that of the PU with MPP, and organoclay produced less-toxic CO and CO2 gas released from combustion. The co-use of organoclay and MPP demonstrated the synergistic effect on the ﬂame retardancy of PU.

7.4 PU/Functionalized Graphene Nanocomposites (PUFGNs)

In addition to CNTs and nanoclays, the incorporations of other nanomaterials into the PU matrix via the in-situ polymerization method are also covered in many literatures. The fabricating routes are similar to those introduced above. It is highlighted again that the good dispersion and strong interface are the key points in optimizing the contributions of nanomaterials to the improved physical properties of PUs. Here, it is our great interest to introduce a rising star in the family of nanomaterials, graphene, which is a two-dimensional and 1-nm-thick carbon sheet. Single-layer graphene was ﬁrst discovered by Geim and Novoselov in 2004 using a “scotch-tape” method [48]. It has been reported to have super mechanical, electrical, and thermal properties. Scientists have found that graphene is the strongest material ever measured. Graphite has been used as an additive in polymeric materials for a long term of period. Expansion of chemically intercalated graphite (GIC) at very high temperature is a traditional route to exfoliate graphite into a number of micro-sized ﬂakes with hundreds of carbon layers. It is very far away from the single-layer graphene sheet. Recently, a chemical route has been developed to fabricate graphene, in which oxidization is ﬁrst applied to introduce the oxygen groups to the interlayers of graphite to form graphite oxide (GO) with reduced van der Waals forces among graphene layers, then GO can be exfoliated to oxidized graphene layers in water by sonication, and ﬁnally the damaged aromatic structure in oxidized graphene can be chemically restored as the oxygen functionalities are chemically removed by hydrazine. Until now, researchers have explored a variety of chemical and nonchemical routes to restore oxidized graph-

7.4 PU/Functionalized Graphene Nanocomposites (PUFGNs)

ene. However, the complete restoration is very hard to be done; thus, chemically restored graphene still present less attractive properties than pristine graphene. When more efforts are being added to polish these methods, the researchers in the ﬁeld of composites have caught the bonus of this method, which is the application of oxidized graphene also being called functionalized graphene (FG) in PNCs. FG has its natural advantages as nanoﬁllers for PNCs: (i) FG can be easily achieved in liquid media, which can be incorporated into polymeric matrices in the presence of water or organic solvents and (ii) the oxygen functionalities on the surface of graphene can enhance the interfacial interaction between polymeric matrices and the graphene oxide. It was also found that the FG can be chemically restored in the presence polymers to form conductive polymer/FG nanocomposites. 7.4.1 Fabrication

The in-situ polymerization of PUFGNs can be carried out in the presence of organic solvents. The key step is the exfoliation of FG in organic solvents. It is difﬁcult to exfoliate FG in polyols directly, which limits the application of FG in the solvent-free system. For the exfoliating mechanism of the GO in water, the oxygen groups introduced into graphite can reduce the van de Waals force and expand the space between neighboring carbon sheets. This allows water molecules to penetrate into the interlayer space of GO. The mechanical force supplied by ultrasonication can destroy the reduced van de Waals interaction and cleave the oxygenated carbon sheets from GO into water. The oxygen functionalities can create eletro static repulsion and steric hindrance to prevent the reaggregation of exfoliated graphene oxide in water. The exfoliation of the GO in organic solvents seems more complicated. Ruoff and his co-workers [49] found that the GO made by the Hummers method cannot be directly exfoliated into dimethylformamide (DMF) that is a strong polar organic solvent. It was considered that this phenomenon might be attributed to disability of organic solvents to disrupt the strong hydrogen bonds between the oxygen groups. They used phenyl isocyanate to terminate the hydrophilic oxygen groups and achieved the exfoliation of isocyanatetreated graphene in organic solvents. We modiﬁed the Hummers method and used expandable graphite (EG) to replace natural graphite as a starting material in Hummer’s method [50]. The resulting GO can be exfoliated into DMF directly without any chemical treatment with organic molecules. In this method, it could be possibly assumed that inner carbon layers of EG have more chances to be exposed to oxidization than that of natural graphite due to the sulfuric acid originally intercalated into the interlayer of EG. Paredes et al. [51] took an effect to understand the dispersion behavior of graphene oxide in different organic solvents and identiﬁed suitable organic solvents for successful dispersion of graphene oxide including N,N-dimethylformamide, N-methyl-2-pyrrolidone, tetrahydrofuran, and ethylene glycol. In contrast, some solvents such as dichloromethane, n-hexane, methanol, and o-xylene cannot accommodate graphene oxide at all. Other solvents including acetone, ethanol, 1-propanol, dimethyl sulfoxide, and pyridine can

209

210

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties

a)

b)

Figure 7.35 Board scan XPS spectra of expandable graphite (a) and expandable graphite oxide

(b). Reproduced from Ref. [50] with permission.

stabilize graphene oxide for very short term from hours to a few days. In the group of so-called suitable solvents, larger amounts of precipitate was observed in ethylene glycol and tetrahydrofuran than water, N,N-dimethylformamdie and N-methyl2-pyrrolidone. All these experimental observations suggest that the organic solvents with higher polarity generally have better dispersing ability. 7.4.2 Morphology and Characterization

The oxidation of graphite is usually conﬁrmed by X-ray photoelectron spectroscopy (XPS) which can give the surface chemical state of graphite and GO. Figure 7.35 demonstrates the board scan XPS spectra of EG and oxidized expendable graphite (EGO). The elemental analysis illustrates that the C/O atomic ratio (12.9) of EG is higher than that (2.7) of EGO, which conﬁrms that oxidation of EG is successfully conducted by Hummer’s method. The C 1s XPS spectra of EG and EGO in Figure 7.36 show that the C 1s XPS spectrum of EGO is split into three peaks (285 eV, 287 eV, and 288 eV) in comparison with two peaks split at 285 and 287 eV in that of EG, respectively. The peaks at 285, 287, and 288 eV are attributed to carbon atoms in different functional groups including the nonoxygenated ring C, the C in C–O bond, and carbonyl C in C=O bond, correspondingly. Mixture of 100 mg EGO and 10 g DMF was subjected to strong ultrasonic treatment for 1 h at room temperature. The EG(100 mg)/DMF(10 g) dispersion was also prepared by ultrasonic treatment under some conditions. Figure 7.37 shows digital pictures of the EG/DMF and EGO/DMF dispersions taken after 2 months. In the EG/DMF dispersion, all EG powder is clearly observed to precipitate at the bottom of the glass container after ultrasonic treatment. It was found that the EGO/DMF dispersion can be long-term stabilized, and even there was no precipitates appeared at the bottom of the glass container. The digital pictures are the hand-touched

7.4 PU/Functionalized Graphene Nanocomposites (PUFGNs) a)

b)

Figure 7.36 C 1s XPS spectra of expandable graphite (a) and expandable graphite oxide (b).

Reproduced from Ref. [50] with permission.

a)

b)

Figure 7.37 Digital pictures of expandable graphite/DMF (a) and expandable graphite oxide/

DMF dispersions (b). Reproduced from Ref. [50] with permission.

evidence to prove the possibility that the EGO is able to be fully exfoliated in the organic solvent. Rouff et al. used AFM to identify the thickness of single-layer FG in organic solvent as ∼1 nm. TEM images in Figure 7.38 shows exfoliated graphene nanosheets in DMF. The edge of FG layers is observed to be strongly folded indicating the high toughness of FG. The dispersion of FG in the PU matrix does not present difﬁculties since the full exfoliation of FG has been successfully achieved in DMF. The SEM images of the fracture surface in Figure 7.39 show that the layer-structured FG is dispersed in the PU matrix with nanoscale thickness. As FG joins in the polymerization of PUs, the molecular weight was found to vary from 28 000 to 155 000 g mol−1. It may result from the isocyanate consumption by oxygen groups on FG. It is also the advantage of FG to form covalent bonding (–NH–CO–) with the PU matrix. It

211

212

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties

Figure 7.38 TEM images of exfoliated graphite oxide nanoplatelets in DMF. Reproduced from

Ref. [50] with permission.

Figure 7.39 SEM image of a PU/4.4 wt%FG nanocomposite. Reproduced from Ref. [53] with

permission.

is difﬁcult to conﬁrm the covalent bonding directly since it also exists in PUs. Kim et al. placed solvent blended and in-situ polymerized PUFGN in THF and found that the solvent-blended PUFGN was dissolved in THF in 5 min and the in-situ polymerized PUFGN still could keep its original for 24 h, even with intensive mechanical perturbation. This observation could conﬁrm the role of FG as chemical crosslinkers in PUs. DMA was used to investigate the interaction between FG and the soft segment. Figure 7.40a shows that the storage modulus of the PUFGN

7.4 PU/Functionalized Graphene Nanocomposites (PUFGNs) a)

b)

0.35

a

a: 0% GONPs b: 4.4 wt% GONPs

0.30 b

0.25

Tanδ

0.20 0.15 0.10 0.05 0.00 –100

–80

–60

–40

–20

0

20

40

Temperature (°C)

Figure 7.40 Storage modulus (a) and damping factor tan δ (b) of the PU (a) and the

PU/4.4 wt% GONP nanocomposite (b) as a function of temperature. Reproduced from Ref. [53] with permission.

with 4.4 wt% FG is about 30% higher than that of pure PU at −60 °C. Figure 7.40b shows no shift in the peak of damping factor (tanδ) associated with the glass transition temperature of the soft segment, which demonstrates the interaction between FG and the soft segment is very weak. However, the huge decrease in damping capacity (amplitude of tan δ) indicates the greatly restricted motion of PU chains, which might resulting from the crosslinking of isocyanate-terminated PU chains by FG. Figure 7.41 shows the MDSC heat ﬂow signals against temperature. Line (a) shows that the melting-like transition temperature relating to the hard segment of the PU appears around 200 °C. As the incorporation of the FG increases to 4.4 wt%, it can be seen in line (b) that the melting-like transition completely disappears and is replaced by recrystallization transition. This ﬁnding indicates that a sufﬁcient amount of the FG can even result in damage to the

213

214

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties

Figure 7.41 MDSC curves of heat ﬂow against temperature: (a) the PU and (b) the

PU/4.4 wt% GONP nanocomposite. Reproduced from Ref. [53] with permission.

crystalline structure of the hard segment, which may be due to the fact that the two-dimensional structure of large graphene sheets prevents the formation of the lamellar structure in the hard segment. The FTIR technique provided by Cooper et al. was used to estimate the phase separation using the FTIR technique. The incorporation of 4.4 wt% FG resulted in the decrease in DPS of pure PU from 0.63 to 0.57, indicating the reduction of the phase separation. Kim et al. [52] also found a similar effect of FG on the hard segment. In the FTIR spectra in Figure 7.42, it can be seen that the peak at 1715 cm−1 belonging to hydrogen bonded C=O diminishes in in-situ polymerized PUFGN, which does not alter in melt and solventblended PUFGN. The PU matrix extracted from the nanocomposites also displays the reduction in the hydrogen bonding index. It seems that FG can impose the permanent change to the arrangement of the hard segment after forming covalent bonding with the PU matrix. 7.4.3 Physical Properties

Very limited publications were found in terms of PUFGNs with a short history. However, we have enough reasons to be excited to see the beneﬁts brought from FGs. Researchers from Columbia University discovered that graphene is one of the strongest materials ever measured. The breaking strength and Young’s modulus of graphene reaches 42 N m−1 and 1 TPa, respectively. However, the strength of the interface is central to the mechanical enhancement of PNCs rather than of the intrinsic strength of nanoﬁllers themselves. Therefore, it is believed that the FG can do a better job than pristine graphene to mechanically reinforce FPNs. The strong interaction between the FG and polymeric matrices could result from (i) the wrinkled surface of extremely thin graphene sheets that is capable of

7.4 PU/Functionalized Graphene Nanocomposites (PUFGNs) a)

b)

c)

d)

Figure 7.42 ATRFTIR spectra of (a) melt-

blended, (b) solvent-blended, (c) in-situ polymerized TRG, and (d) GO/TPU composites. For in-situ polymerized samples, spectra

of TPU recovered from the composites by Soxhlet extraction are also shown. Reproduced from Ref. [52] with permission.

mechanically interlocking with polymer chains; and (ii) the hydrogen bonding formed between the oxygen functionalities of the FG and polymeric matrices. It has been demonstrated above that hydrogen and covalent bonding can be formed between FG and PUs to act as a strong interface. Researchers have found signiﬁcant reinforcing effects of FG in PUs. Cai et al. found that Young’s modulus of a PUFGN with 4.4 wt% FG was nearly 10 times higher than that of pure PU with the help of the covalent interface. Figure illustrates the typical stress–strain curves of pure PU and this PUFGN. However, there was 60% decrease in elongation at break, which meant crosslinking PU chains by FG resulted in the decreased ductility of PUs. Nanoindentation tests showed that the hardness and elastic modulus increases ∼327% and ∼182%, respectively, with the incorporation of 4.4 wt% FG.

215

216

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties

a)

b) 3000

a:0% GONPs b:4.4wt% GONPs

a

a:0% GONPs b:4.4wt% GONPs

a

1500

2000

Scratch depth (nm)

Scratch depth (nm)

2500

2000

1000

1500 1000

b

500

500

b

0 Scratch direction

0

Scratch direction –500

0

20

40

60

Scratch length (μm)

80

100

0

20

40

60

80

100

Scratch length (μm)

Figure 7.43 Nanoscratch depth proﬁles for the PU and the PU/4.4 wt% GONP nanocompos-

ite at scratch rates of 3 μm s−1 (a) and 5 μm s−1 (b), respectively. Reproduced from Ref. [53] with permission.

In nanoscratch test as shown in Figure 7.43, the scratch depth of the indenter in the sample was recorded along with the scratch length at a certain scratch rate, which reﬂects the protective ability of the surface coatings for the substrates. It was revealed that the scratch depth was reduced by ∼80%, which pointed out that PUFGNs is a good candidate for organic protective coatings. Generally, FG become electrically insulating as the conductive structure of graphene is damaged. Achievement of conductive FG/PNCs needs the restoration of the conductive structure of graphene in polymeric matrices. Chemical and thermal methods have been reported to eliminate the oxygen groups on the FG and recover the conductivity of graphene. This type of graphene strictly should be called reduced graphene although often it is referred to as “graphene” in some articles. Ruoff and his co-workers [54] ﬁrst used chemically reduced graphene to electrically enhance polystyrene (PS), and found that the percolation threshold and the maximum electrical conductivity of the PS nanocomposites was 0.1 vol% and 0.1 S m−1 at 1 vol% chemically reduced graphene, respectively. The chemical reduction of FG using hydrazine was carried out in the presence of the PS to avoid reaggregation of chemically reduced graphene with much less functionalities than the FG. However, chemically reduced graphene is not suitable for PUs because the reducing agents (hydrazine) can react with the electrophilic atoms in PUs. Rapid thermal expansion was another method to reduce FG. Macosko and his co-workers [52] investigated three ways to incorporate thermally reduced graphene (TRG) into PUs including melt compounding, solution blending, and in-situ polymerization. Figure 7.44 shows DC surface resistance of those prepared PUFGNs. It can be seen that percolation threshold is >2.7 vol% and <0.5 vol% for untreated graphite and TRG, respectively. Also, the conductivity of in-situ polymerized and solvent-blended PUFGNs is higher than melt-blended ones at the same ﬁller volume fraction, which is due to better dispersion for in-situ polymerized and

7.5 Prospective of PUNs

Figure 7.44 DC surface resistance of melt-blended graphite/TPUcomposites (closed symbols,

also in inset) and melt-blended, solution-mixed, and in-situ polymerized TRG/TPU composites (open symbols). Reproduced from Ref. [52] with permission.

solvent-blended PUFGNs. In their study, those TRG-based PUFGNs exhibited exceptional N2 permeability. The barrier mechanism should be similar to that proposed in PUCLNs. Remark reduction in N2 permeation demonstrates that exfoliated graphene sheets can play as diffusion barriers in the PU matrix. Better dispersion obviously leads to higher degree of reduction.

7.5 Prospective of PUNs

The commercial values of PUs have been substantially recognized since their discovery in 1950s. The nanotechnology has provided the opportunities to create next generation of PU materials with much improved mechanical properties and other functional properties. PUs can exist in various forms for speciﬁc applications such as elastomers, solvent-based coatings, and porous foams. For in-situ polymerization of PUNs, the addition of nanoﬁllers generally takes place in organic solvents and liquid polyols with the help of mechanical forces which is not highcost process. Surface treatment of naoﬁllers paves the way to good dispersion of the inorganic phases in PU matrices. The functionalities located on the surface of naoﬁllers such as hydroxyl, carboxyl, or amine groups can potentially form chemical bonding with PU matrices acting as a strong interface for stress transfer. In addition, nanoﬁllers can bring more than mechanical reinforcement to PUs such as enhanced electrical, thermal, and barrier properties. All these advantages trigger the intention and possibility to commercialize PUNs. Here, some suggestions were given on future work for both engineering and scientiﬁc concerns.

217

218

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties

Cost is the key point to fabricate PUN products for engineering concerns. The addition of nanoﬁllers in PUs can be achieved with slight modiﬁcation to the current commercial process. High cost of nanoﬁller is still the barrier to develop PUN products. The nanoﬁllers from natural sources such as nanoclay, nanographite, and natural nanotubes are cheaper than CNTs and POSS, which should have more chance to be used in PUNs. The effort should also be placed on developing the simple and effective methods for surface treatment with scaling-up potential. Understanding of the effects of nanoﬁllers on the chemistry and physics of PUs is supposed to move on for scientiﬁc concerns. It decides how much super properties we can get out from nanoﬁllers. Regarding the polymerization kinetics of PUNs, some studies initially showed that the nanoﬁllers could act as a catalyst or consume excess NCO group in the polymerization of PUs. However, the nanoeffects on the kinetics of polycondensation and polymeric network formation are not very completely clear yet now, which should relate to dimension and surface reactivity of nanoﬁllers. For the morphology of PUNs, the random dispersion of nanoﬁllers is achieved without precise control. It has been demonstrated that nanoparticles could be smartly manipulated to be located in the hard segment of PUs, by which substantial reinforcement was achieved without losing ductility. Thus, it is believed that the precise manipulation is the way out to next generation of PUNs such as self-assembling nanoﬁllers into ordered nanophases in PU matrices. This is a big challenge for nanofabrication techniques. Increasingly harsh environment is challenging the physical properties of materials that is, ballistic impact. The ballistic impact can result in high strain rate deformation of materials, which has been demonstrated to generate new physical phenomena of polymers. It was found that the glassy transition of polyurea occurs at room temperature at a high strain rate of 105 (calorimetric Tg of polyurea is −60 °C). The physical response of PUNs to high strain rate is ambiguous. Due to the excellent performance of PUNs under low strain rate deformation, it is our expectation to see their contribution serving as protective materials for the personnel threaten by ballistic impact. In-situ polymerization has become a ﬂexible method for fabrication of PNCs. This chapter introduces in-situ polymerization of PUNs involving three typical nanoﬁllers: CNTs, nanoclays, and FGSs. It provides readers with a brief knowledge in terms of fabrication, morphology, and properties of PUNs. We have seen the powder of nanoﬁllers in PUs. However, more efforts should be encouraged to dig out more scientiﬁc and commercial values of these novel materials.

References 1 Coleman, J.N., Khan, U., and Gun’ko,

3 Haggenmueller, R., Gommans, H.H.,

Y.K. (2006) Adv. Mater., 18, 689. 2 Li, B., Sundararaj, U., and Pötschke, P. (2006) Macromol. Mater. Eng., 291, 227.

Rinzler, A.G., Fisher, J.E., and Winey, K.I. (2000) Chem. Phys. Lett., 330, 219.

References 4 Qian, D., Dickey, E.C., Andrews, R., and

5

6 7 8 9 10 11 12

13 14

15 16 17 18 19

20 21

22 23

24 25

26

Rantell, T. (2000) Appl. Phys. Lett., 76, 2868. Jia, Z., Wang, Z., Xu, C., Liang, J., Wei, B., Wu, D., and Zhu, S. (1999) Mater. Sci. Eng. A, 271, 395. Shim, B. and Kotov, N.A. (2005) Langmuir, 21, 9381. Regev, O., ElKati, P.N.B., Loos, J., and Koning, C.E. (2004) Adv. Mater., 16, 248. Cai, D. and Song, M. (2007) Macromol. Chem. Phys., 208, 1183. Iijima, S. (2002) Physica B, 323, 1. Treacy, M.M.J., Ebbesen, T.W., and Gibson, J.M. (1996) Nature, 381, 678. Berber, S., Kwon, Y.K., and Tománek, D. (2000) Phys. Rev. Lett., 84, 4613. Kim, P., Shi, L., Majumdar, A., and McEuen, P.L. (2001) Phys. Rev. B, 87, 215502. Xia, H. and Song, M. (2005) Soft Matter, 1, 386. Chen, X., Chen, X., Lin, M., Zhong, W., Chen, X., and Chen, Z. (2007) Macromol. Chem. Phys., 208, 964. Xia, H., Song, M., Jin, J., and Chen, L. (2006) Macromol. Chem. Phys., 207, 1945. Kwon, J. and Kim, H.J. (2005) J. Polym. Sci.,Part A: Polym. Chem., 43, 3973. Cai, D. and Song, M. (2008) Carbon, 46, 2107. Xia, H., and Song, M. (2006) J. Mater. Chem., 16, 1843. Jung, Y.C., Sahoo, N.G., and Cho, J.W. (2006) Macromol. Rapid Commun., 27, 126. Coleman, J.N. and Ferreira, M.S. (2004) Appl. Phys. Lett., 84, 798. Coleman, J.N., Cadek, M., Blake, R., Nicolosi, V., Ryan, K.P., Belton, C., Fonseca, A., Nagy, J.B., Gun’ko, Y.K., and Blau, W.J. (2004) Adv. Mater., 14, 791. Xiong, J., Liu, Y., and Wang, X. (2006) Carbon, 44, 2701. Chen, X., Wang, J., Zou, J., Wu, X., Chen, X., and Xue, F. (2009) J. Appl. Polym. Sci, 114, 3407. Seymour, R., Estes, G.M., and Cooper, S.L. (1970) Macromolecules, 3, 579. Grunlan, J.C., Mehrabi, A.R., Bannon, M.V., and Bahr, J.L. (2004) Adv. Mater., 16, 150. Xu, X.B., Li, Z.M., Shi, L., Bian, X.C., and Xiang, Z.D. (2007) Small, 3, 408.

27 Haggenmueller, R., Guthy, C., Lukes,

28

29 30 31

32 33 34

35 36 37 38 39

40

41

42 43 44 45

46

47

48

J.R., Fischer, J.E., and Winey, K.I. (2007) Macromolecules, 40, 2417. Du, F., Guthy, C., Kashiwagi, T., Fischer, J.E., and Winey, K.I. (2006) J. Polym. Sci., Part B: Polym. Phys., 44, 1513. Huang, H., Liu, C., Wu, Y., and Fan, S. (2005) Adv. Mater., 17, 1652. Sinha Ray, S. and Okamoto, M. (2003) Prog. Polym. Sci., 28, 1539. Sinha Ray, S., Okamoto, K., and Okamoto, M. (2003) Macromolecules, 36, 2355. Yao, K.J., Song, M., Hourston, D.J., and Luo, D.Z. (2002) Polymer, 43, 1017. Xia, H. and Song, M. (2006) Polym. Int., 55, 229. Song, M., Hourston, D.J., Yao, K., Tay, J.K.H., and Ansarifar, A. (2003) J. Appl. Polym. Sci, 90, 3239. Xia, H., Shaw, S.J., and Song, M. (2005) Polym. Int., 54, 1392. Rama, M.S. and Swaminathan, S. (2010) J. Appl. Polym. Sci, 118, 1774. Xia, H. and Song, M. (2005) Thermochim. Acta, 429, 1. Corcione, C.E. and Maffezzoli, A. (2009) Thermochim. Acta, 485, 43. Xia, H., Song, M., Zhang, Z., and Richardson, M. (2007) J. Appl. Polym. Sci., 103, 2992. Chang, S.L., Yu, T.L., Huang, C.C., Chen, W.C., Linliu, K., and Lin, T. (1998) Polymer, 39, 3479. Song, M., Xia, H.S., Yao, K.J., and Hourston, D.J. (2005) Euro. Polym. J., 41, 259. Kim, W., Chung, D., and Kim, J.H. (2008) J. Appl. Polym. Sci, 110, 3209. Yusoh, K., Jin, J., and Song, M. (2010) Prog. Org. Coating., 67, 220. Nielsen, L.E. (1967) J. Macromol. Sci. A, 1, 929. Herrera-Alonso, J.M., Marand, E., Little, J., and Cox, S.S. (2009) J. Membr. Sci., 337, 208. Osman, M.A., Mittal, V., Morbidelli, M., and Suter, U.W. (2003) Macromolecules, 36, 9851. Song, L., Hu, Y., Tang, Y., Zhang, R., Chen, Z., and Fan, W. (2005) Polym. Degrad. Stab., 87, 111. Novoselov, K.S., Geim, A.K., Morozov, S.V., Zhang, J.Y., Dubonos, S.V.,

219

220

Polyurethane Nanocomposites by In-situ Polymerization Approach and Their Properties Grigorieva, I.V., and Firsov, A.A. (2004) Science, 306, 666. 49 Stankovich, S., Dikin, D.A., Piner, R.D., Kohlhaas, K.A., Kleinhammes, A., Jia, Y., Wu, Y., Nguyen, S.T., and Ruoff, R.S. (2007) Carbon, 45, 1558. 50 Cai, D. and Song, M. (2007) J. Mater. Chem., 17, 3678. 51 Paredes, J.I., Rodil-Villar, S., MartínezAlonso, A., and Tascón, J.M.D. (2008) Langmuir, 24, 10560.

52 Kim, H. and Miura, Y.M. (2010) Chem.

Mater., 22, 3441.

53 Cai, D., Yusoh, K. and Song, M. (2009)

Nanotechnology, 20, 85712. 54 Stankovich, S., Dikin, D.A., Dommett,

G.H.B., Kohlhaas, K.M., Zimney, E.J., Stach, E.A., Piner, R.D., Nguyen, S.T., and Ruoff, R.S. (2006) Nature, 442, 282.

221

8 In-situ Synthesis and Properties of Epoxy Nanocomposites Vikas Mittal

8.1 Introduction

By the beginning of the 1980s world capacity for epoxide resins reached about 6 × 105 tonnes per annum (t.p.a) [1–4]. However, with a global consumption of about 10 million tonnes per annum for thermosetting plastics, epoxide resins had a share of about 3%. By the late 1990s, however, the world market for epoxide resins had risen to about 7.5 × 105 t.p.a. The epoxide resins of the glycidyl ether type are usually characterized by parameters such as resin viscosity (of liquid resin), epoxide equivalent, hydroxyl equivalent, average molecular weight (and molecular weight distribution), melting point (of solid resin), and heat distortion temperature (deﬂection temperature under load) of cured resin [1–4]. Amine hardeners crosslink epoxide resins either by a catalytic mechanism or by bridging across epoxy molecules. In general, the primary and secondary amines act as reactive hardeners while the tertiary amines are catalytic. Being crosslinked, the epoxy resins do not dissolve without decomposition but can be swollen by liquids of similar solubility parameter to the cured resin. The chemical resistance is as much dependent on the hardener as on the resin since these two determine the nature of the linkages formed. In the early stage of their development, the epoxy resins were used almost entirely for surface coating. With further expansion of the spectrum of applications, about half of epoxide resin production is used as surface coating applications, with the rest divided approximately equally among electronic applications (particularly for printed circuit boards and encapsulation), the building sector, in adhesives, and miscellaneous uses. The generation of composites by the incorporation of inorganic ﬁllers of various geometries helps to further enhance the performance of epoxy matrix in these applications. The generation of nanocomposites with epoxy matrix has further advantages as it helps to retain the transparency of the matrix as well as low density because the amount of ﬁller required to achieve signiﬁcant enhancements in properties is low, if it can be uniformly delaminated to the nanoscale in the polymer matrix. For example, as mentioned above, epoxies are used as adhesives to produce polymer-packaging laminates, where laminates In-situ Synthesis of Polymer Nanocomposites, First Edition. Edited by Vikas Mittal. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA.

222

8 In-situ Synthesis and Properties of Epoxy Nanocomposites

have the function of providing barrier against water vapor, oxygen, carbon dioxide, etc. In these laminates, generally polypropylene and polyethylene terephthalate ﬁlms are adhered together to bring about effective barrier to water vapor and oxygen, respectively. The use of epoxies in such laminates is limited only to hold the polymer foils together and their contribution to the overall barrier performance of the laminate is practically meager. It would be an advantage if an additional function can be added to the epoxy matrix such as enabling the adhesive to improve the barrier and mechanical performance of the laminate (by the addition of high aspect ratio platy layered silicate clay particles). This can lead to a reduction in the laminate thickness by completely avoiding the use of polypropylene or PET ﬁlms. Similarly, other speciﬁc property enhancements can be achieved for other applications by adding ﬁllers such as inorganic spherical particles, nanotubes, etc. Owing to its thermosetting nature, the synthesis of epoxy composites is achieved by following the in-situ preparation approach. The ﬁller is ﬁrst dispersed in epoxy prepolymer (with or without adding a solvent) followed by polymerization and crosslinking. This chapter presents the in-situ synthesis of epoxy nanocomposites focusing on their permeation properties for application in packaging laminates. Permeation barrier toward oxygen and water vapor form the most important property needed in the packaging materials. This can be achieved by altering the polymer network structure obtained by crosslinking of the epoxide groups with amines or other crosslinking agents [5, 6]. The use of epoxy polymer with stiff rod like units in the backbone can help to enhance the required properties. The other alternative includes the incorporation of inorganic ﬁllers in the polymer matrix, this approach being easier to monitor and control. As the ﬁller’s shape, size, and interfacial interactions affect the polymer properties greatly, organically treated plate-like inorganic aluminosilicates particles can be incorporated in the polymer matrix for improvement in barrier performance. By incorporating impermeable transparent plate-like nanoparticles in the polymer matrix, the permeating molecules are forced to wiggle around them in a random walk, hence diffusing through a tortuous pathway [7–9]. Apart from that, the decrease in the transmission rate of the permeant is a function of the aspect ratio of the inclusions, their volume fraction, and orientation.

8.2 Optimization of the Curing Conditions

The curing parameters such as time, temperature, and mole ratio of the curing agent to epoxy were required to be optimized in order to have the curing process which provides time for the intercalation of epoxy and amine in the clay interlayers, but is also quick enough to avoid excessive extragallery polymerization [10]. Figure 8.1 shows that the DSC plots of epoxy curing runs at different temperatures of 40 °C, 55 °C, 60 °C, 70 °C, and 100 °C, while keeping the amine to epoxy mole ratio constant at 1 : 1. The trials run at different amine to epoxy mole ratios, while keeping the temperature constant at 70 °C are shown in Figure 8.2.

8.2 Optimization of the Curing Conditions

Figure 8.1 DSC thermograms showing the curing of epoxy at different temperatures at a ﬁxed

amine to epoxy mole ratio of 1 : 1; I: 40 °C, II: 55 °C, III: 60 °C, IV: 70 °C, and V: 100 °C.

Figure 8.2 Curing of epoxy at 70 °C using different amine to epoxy mole ratios; I: 1 : 1, II:

0.6 : 1, III: 0.4 : 1, and IV: 0.3 : 1.

From these results, curing temperature of 70 °C and epoxy to amine ratio of 0.3 : 1 were chosen as optimal values as it provided enough curing time necessary for exfoliation, along with maintaining thermal stability of the substrate foils at the curing temperature. Approximately 45 min were required to achieve maximum curing in this chosen system. Effect of the presence of ﬁller on the overall curing

223

224

8 In-situ Synthesis and Properties of Epoxy Nanocomposites

Figure 8.3 DSC thermograms indicating the effect of ﬁller on curing at 70 °C; I: no ﬁller,

amine to epoxy mole ratio 0.3 : 1, II: 3.5 vol% 2C18-OM, amine to epoxy mole ratio 0.3 : 1, and III: 3.5 vol% 2C18-OM, amine to epoxy mole ratio 0.2 : 1.

process has also been described in Figure 8.3. The presence of dioctadecyldimethylammonium-modiﬁed clay led to the advancement of the curing signiﬁcantly as shown by curve II as compared to curve I without the presence of ﬁller. It is because of the acidic nature of the ammonium ions that have been reported to catalyze the epoxy-amine polymerization reaction [2, 10–12]. Although with quaternary ammonium ions the extent of intergallery catalysis is reported to be lower as compared to primary, secondary, and tertiary counterparts, but still it can be signiﬁcant. Curve III indicates the curing process carried out at 70 °C at a reduced amine to epoxy mole ratio of 0.2 : 1. The slowing down of the curing process is clearly visible. The inﬂuence of curing temperature on the ﬁller exfoliation was also analyzed through x-ray as shown in Figure 8.4. Using an amine to epoxy mole ratio of 0.3 : 1, the curing at 70 °C led to decrease in the peak intensity as compared to the same material cured at 60 °C, indicating optimal curing rate and time are the most important criterion.

8.3 Fillers, Surface Modiﬁcations, and Ion Exchange

Sodium montmorillonite (Cloisite Na+) and sodium vermiculite were used as ﬁller substrates. The cation exchange capacity of bentonite was 880 and 1440 μeq g−1. Higher cation exchange capacity allows for dense packing of surface modiﬁcation molecules on the surface during ion exchange reactions, thus leading to higher

8.3 Fillers, Surface Modiﬁcations, and Ion Exchange

Figure 8.4 X-ray diffractogram showing the effect of temperature on composite

microstructure (3.5 vol% 2C18-OM ﬁller).

Figure 8.5 TEM micrograph of unmodiﬁed montmorillonite dispersed in water.

basal plane spacing in the ﬁller. This can have signiﬁcant effect on the morphology if the ﬁller in the composites and hence composite properties. Figure 8.5 shows the transmission electron microscopy (TEM) micrograph of the unmodiﬁed montmorillonite ﬁller indicating the delaminated single layers. It

225

226

8 In-situ Synthesis and Properties of Epoxy Nanocomposites

indicates that the ﬁller could be delaminated by the process of sonication and shear mixing. To reduce the surface energy and hydrophilicity of the clay, hence rendering it to be compatible with the epoxy matrix, its inorganic cations were exchanged with organic ammonium ions of different chemical structures. Various ammonium salts chosen for the cation exchange reactions on the ﬁller surface are presented in Table 8.1. The chemical structures of the modiﬁcations vary owing to the presence of different number of OH groups as well as number of long alkyl chains. The presence of polar OH groups was expected to help to match the polarity with the epoxy matrix and eventual chemical reaction with the polymer during

Table 8.1 Details of the surface modiﬁcations exchanged on the ﬁller surface [13, 14].

Modiﬁcation

Abbreviation

Filler substrate

Dioctadecyldimethylammonium

2C18

Montmorillonite

Benzyldimethylhexadecylammonium

BzC16

Montmorillonite, vermiculite

Benzyldibutyl(2-hydroxyethyl) ammonium

Bz1OH

Montmorillonite

Benzyl-bis(2-hydroxyethyl) butylammonium

Bz2OH

Montmorillonite

Benzyltriethanolammonium

Bz3OH

Montmorillonite

Benzyl(2-hydroxyethyl) methyloctadecylammonium

BzC18OH

Montmorillonite, vermiculite

Chemical structure

8.3 Fillers, Surface Modiﬁcations, and Ion Exchange

curing. This way, the grafting of the long polymer chains to the ﬁller surface can be achieved. The presence of benzyl groups can also lead to van der Waals interactions with the epoxy matrix. The presence of long alkyl chains is helpful in enhancing the basal plane spacing in the modiﬁed ﬁller, thus, reducing electrostatic forces of attraction between the ﬁller platelets, thus, enhancing the potential of polymer intercalation in the ﬁller interlayers. The surface modiﬁcation of the inorganic ﬁller to render it organophillic is one of the most important steps in nanocomposite as it helps to achieve the nanoscale dispersion of ﬁller in the polymer matrix. To achieve surface modiﬁcation of both montmorillonite and vermiculite ﬁllers employed in the present study, ﬁller was dispersed in the water ethanol mixture at 70 °C for a few hours. In between, the dispersion was also sonicated as well as shear mixed using sonicator and shear mixer, respectively. To this dispersion, the ethanol solution of ammonium salt to be exchanged on the ﬁller surface was added dropwise. The amount of ammonium salt used for the purpose was generally higher than the cation exchange capacity of the ﬁller (120–150% of cation exchange capacity). The dispersion was maintained at 70 °C overnight followed by ﬁltration and washing with the water ethanol mixture. As the amount of the ammonium salt used for the exchange process was higher than the amount possible to exchange on the ﬁller surface, it was important to remove all the excess surface modiﬁcation molecules from the modiﬁed ﬁller. For this, the ﬁller was resuspended in the solvent at 70 °C followed by sonication and shear mixing. The dispersion was again stirred overnight followed by ﬁltration and washing step. The ﬁller was then dried at 70 °C in an oven under reduced pressure. It is also important to note that the excess of the surface modiﬁcation molecules may not be washed in a single washing step and more number of steps may be required as indicated in Figure 8.6 [15] for the modiﬁcation of montmorillonite ﬁller with 2C18 modiﬁcation. Four washings cycles were required to completely wash the excess surface modiﬁcation which otherwise forms a local bilayer in which the modiﬁcation molecules are intercalated in the ﬁller interlayers but are not ionically bound to the surface of the ﬁller and are less thermally stable than the bound molecules. Hi-res TGA is the most important tool to evaluate the purity of the modiﬁed ﬁller as shown by the TGA thermograms of Figure 8.6. The molecules which are not bound to the surface of the ﬁller have lower thermal degradation temperatures around 250 °C and hence their presence is easily detected. With subsequent washings, the low thermal degradation peak was eliminated indicating the absence of any excess surface modiﬁcation molecules. Hi-res TGA is also used to quantify the degree of exchange on the ﬁller surface from the total organic loss recorded in the TGA measurement. Therefore, if the full exchange of surface (>95%) with certain surface modiﬁcations was not achieved in the ﬁrst exchange reaction, the same ﬁller was re-exchanged. Table 8.1 details the surface modiﬁcations exchanged on the montmorillonite and vermiculite substrates. Figure 8.7 also shows the TGA thermograms of the montmorillonite ﬁller modiﬁed with various surface modiﬁcations. The absence of any thermal degradation peak prior to the main degradation peak ensured the cleanliness of the ﬁllers

227

228

8 In-situ Synthesis and Properties of Epoxy Nanocomposites

Figure 8.6 Stages of washing out of the excess surface modiﬁcation molecules for the

exchange of 2C18 molecules on the montmorillonite surface. Reproduced from Ref. [15] with permission from Wiley.

Figure 8.7 TGA plots of montmorillonite ﬁller modiﬁed with various surface modiﬁcations.

8.4 Nanocomposite Synthesis

except 2C18 modiﬁed ﬁller, where there was still a slight presence of excess surface modiﬁcation molecules.

8.4 Nanocomposite Synthesis

The required amounts of modiﬁed montmorillonite or vermiculite and epoxy resin were calculated by following these relations [13, 14]: MOM = MM + (MM × CEC × MOC ) MER =

MM × VER × ρER − (MM × CEC × MOC ) VM ⋅ ρM

MOV = M V + (M V × CEC × MOC ) ⎡ MM × VER × ρER ⎤ MER = ⎢ ⎥ − (M V × CEC × MOC ) VV × ρV ⎣ ⎦ where MOM is the mass of the organically modiﬁed montmorillonite, MM is the mass of the inorganic montmorillonite, MOC is the molar mass of the ammonium ion used to modify the montmorillonite surface, MEP is the mass of the epoxy resin, VM is the inorganic montmorillonite volume fraction, ρM is the density of sodium montmorillonite (2.6 g cm−3), and VEP is the epoxy volume fraction and ρEP is its density (1.18 g cm−3). MOV is the mass of the modiﬁed vermiculite, MV is the mass of the inorganic vermiculite, VV is the inorganic vermiculite volume fraction, and ρV is the density of sodium vermiculite (2.6 g cm−3). Figure 8.8 shows the steps involved in the in-situ synthesis of the epoxy nanocomposites. As the nanocomposites were developed for the packaging adhesives applications, therefore, they were supported on the polypropylene and polyamide ﬁlms. The ﬁlm substrates were chosen as they are commonly used substrates in the packaging laminates. The chosen ﬁlms also did not disturb the measurement of the barrier properties of the nanocomposite as the permeation of oxygen and water vapor was very high through polypropylene and polyamide ﬁlms, respectively. The ﬁlms were corona treated for better adhesion with the polymer. In certain cases, especially when drawing ﬁlms of pure epoxy, additional surfactant was added to the formulation in order to improve the adhesion. The ﬁrst step in the synthesis involved the swelling of the organically modiﬁed ﬁller in a suitable solvent (e.g., THF). The dispersion was then sonicated while keeping the contents ice cooled. The dispersion was then added with required amount of epoxy prepolymer and was further sonicated to allow the epoxy prepolymer chains to intercalate the ﬁller interlayers. This was followed by the addition of curing agent and light sonication to mix the contents well. It is important to achieve a balance between the extragallery and intragallery polymerization so as to allow ample time to prepolymer inside the ﬁller interlayers to cure and push apart the ﬁller platelets thereby delaminating them. For this, it is important to control the ratio of curing

229

230

8 In-situ Synthesis and Properties of Epoxy Nanocomposites Suspension and swelling of the required amount of modified filler in the solvent Cooling in ice bath sonication

Addition of epoxy resin solution (DGEBA) to the filler suspension

Sonication Addition of curing agent, tetraethylenepentamine TEPA to epoxy and filler dispersion Sonication Drawing of films with bar coater on corona treated surfaces of PP and polyamide films

Drying of films at ambient conditions for 15 min and under reduced pressure, RT, 15 min

Curing at reduced pressure at 70 (°C) overnight

Postcuring at reduced pressure at 90 (°C) for 4 h

Figure 8.8 Synthesis strategy of epoxy nanocomposites supported on the polypropylene and

polyamide ﬁlms.

8.5 Morphology

agent to the prepolymer. As mentioned above, the amine to epoxy mole ratio of 0.3 : 1 was used for this purpose. If the nanocomposites were to be synthesized for general applications as high strength materials, the obtained dispersion could be placed in a mold followed by simultaneous removal of solvent and curing. But as mentioned above, the goal of the current work was to generate nanocomposite for packaging applications; therefore, the formulation was applied as ﬁlms on the substrate ﬁlms. A bar coater with adjustable wet thickness of the ﬁlms was used. Drying of the ﬁlms was carried out ﬁrst at atmospheric pressure and room temperature for 15 min followed by another 15 min of drying at room temperature but at reduced pressure. This was followed by curing at 70 °C overnight. The postcuring was achieved by increasing the temperature to 90 °C and maintaining the ﬁlms at this temperature for 4 h. Low temperatures were chosen for curing for the same reasons directing the amount of curing agent. Apart from that, as the nanocomposite ﬁlms are drawn on substrate ﬁlms, care is necessary to maintain the dimensional stability of these substrate ﬁlms. After drying, nanocomposite ﬁlms were ca. 10 μm thick. The exact thickness of the ﬁlms was measured analytically.

8.5 Morphology

Tables 8.2 and 8.3 show the densities of the ﬁllers after treatment with various surface modiﬁcations. The ﬁller density decreased signiﬁcantly from 2.6 g cm−3 value for the inorganic ﬁller after surface modiﬁcation. Weight fraction of the ﬁller corresponding to 3.5 vol% inorganic ﬁller fraction is also reported. Densities of the nanocomposites are also reported in these tables, which also indicate that the incorporation of ﬁller did not increase their bulkiness. The composite densities lie in the range of 1.20 to 1.23 g cm−3 as compared to 1.18 g cm−3 for pure epoxy. Table 8.4 details the basal plane spacing values of the ﬁller before and after surface modiﬁcation, the suspensions of modiﬁed ﬁllers in a solvent and the epoxy

Table 8.2 Density of modiﬁed montmorillonite as well as composites containing 3.5 vol% inorganic ﬁller fraction.

OM

Modiﬁed ﬁller density (g cm−3)

Filler weight fraction (%)

Composite density (g cm−3)

Bz1OH Bz2OH Bz3OH BzC18OH BzC16 2C18

1.93 2.01 2.13 1.63 1.69 1.53

8.96 8.89 8.82 9.85 9.5l 10.60

1.22 1.23 1.23 1.21 1.21 1.21

Reproduced from Ref. [13] with permission from American Chemical Society.

231

232

8 In-situ Synthesis and Properties of Epoxy Nanocomposites Table 8.3 Density of modiﬁed vermiculite as well as composites containing 3.5 vol%

inorganic ﬁller fraction. Modiﬁcation

Modiﬁed ﬁller density (g cm−3)

Filler weight fraction (%)

Composite density (g cm−3)

BzC18OH BzC16

1.43 1.53

11.33 10.80

1.20 1.21

Reproduced from Ref. [14] with permission from Sage Publishers.

Table 8.4 Basal plane spacing of the ﬁllers, their suspensions and epoxy composites [13, 14].

Filler

d-spacing of ﬁller (nm)

d-spacing of ﬁller in solvent (nm)

d-spacing of ﬁller suspended in solvent and epoxy (nm)

d-spacing of ﬁller in composite (nm)

Cloisite Na (3.5 vol%) Bz1OH.Cloi (1 vol%) Bz1OH.Cloi (2 vol%) Bz1OH.Cloi (3.5 vol%) Bz1OH.Cloi (5 vol%) Bz2OH.Cloi (3.5 vol%) Bz3OH.Cloi (3.5 vol%) BzC18OH.Cloi (3.5 vol%) BzC16.Cloi (1 vol%) BzC16.Cloi (2 vol%) BzC16.Cloi (3.5 vol%) BzC16.Cloi (5 vol%) 2C18.Cloi (3.5 vol%) Vermiculite Na (3.5 vol%) BzC18OH.Vermi (3.5 vol%) BzC16.Vermi (3.5 vol%)

1.25 1.52 1.52 1.52 1.52 1.50 1.52 2.06 1.87 1.87 1.87 1.87 2.51 1.22 3.40 3.25

1.98 – – 1.92 – 1.88 2.41 4.12 – – No reﬂection – 3.69 1.42 3.80 3.34

1.99 – – 1.90 – 1.90 2.25 3.89 – – 3.55 – 3.54 1.42 3.80 3.53

1.35 1.89 1.88 1.87 1.87 1.62 1.98 3.45 3.07 2.99 2.93 2.90 3.46 1.29 3.96 3.68

Reproduced from Ref. [14] with permission from Sage Publishers.

composites. It was evident from the table that the basal plane spacing of the ﬁller increased according to the chemical architecture of the surface modiﬁcation. The basal plane spacing increased only marginally as compared to the unmodiﬁed montmorillonite when Bz1OH, Bz2OH, and Bz3OH surface modiﬁcations were used. On the other hand, the basal plane spacing was signiﬁcantly increased when the montmorillonite surface was modiﬁed with BzC16, BzC18OH, and 2C18 modiﬁcations owing to the presence of long alkyl chains. In the case of vermiculite, the modiﬁcations containing long alky chains also led to signiﬁcantly higher basal plane spacing values. These values were also higher than the values obtained for

8.5 Morphology

the montmorillonite for the same surface modiﬁcation. This indicated that owing to higher cation exchange capacity of vermiculite, the platelets were pushed to higher distances owing to larger amount of organic matter exchanged on the surface. The suspensions of the ﬁllers in solvent were also measured for basal plane spacing in order to analyze the interactions of the ﬁller surface modiﬁcation with the solvent. Dimethylformamide (DMF) was chosen as the solvent because of its low volatility for the time scale of X-ray experiments. All the ﬁllers were observed to be by the solvent to varying extent. The ﬁllers containing long alkyl chains were observed to swell to a higher degree. Only the diffractogram of BzC16. Cloi ﬁller did not show any diffraction peak indicating complete delamination of the ﬁller in the solvent. For the other ﬁllers, the presence of a diffraction peak indicated the presence of residual electrostatic forces. Addition of epoxy prepolymer to the ﬁller suspension showed interesting observations. Basal plane spacing in the case of Bz1OH.Cloi and Bz2OH.Cloi was not affected whereas a slight decrease in the basal plane spacing was observed for Bz3OH.Cloi system. The basal plane spacing in the case of other ﬁllers BzC18OH.Cloi, 2C18.Cloi and BzC16.Cloi also decreased. It indicated that the interaction of the epoxy prepolymer with the ﬁller surface modiﬁcation played a very important role. In the case of Bz1OH and Bz2OH ﬁllers, the polarity of the ﬁller surface matched well with the prepolymer, but in the other ﬁllers partial deswelling of the ﬁller took place owing to the mismatch between the polarities. In the case of vermiculite, no decrease in the basal planes spacing of the ﬁller was observed on the addition of epoxy prepolymer. This indicated that no or very weak intercalation of prepolymer took place in the ﬁller interlayers. Curing the polymer and subsequent evaporation of solvent did not affect the basal plane spacing in the Bz1OH.Cloi system. Even varying amounts of ﬁller in composite also did not affect the basal plane spacing indicting that the intercalated polymer was not squeezed out even at higher volume fractions of the ﬁller. This further indicated better compatibility of this system with the epoxy. In the case of Bz2OH.Cloi and Bz3OH.Cloi ﬁllers, a decrease in the basal plane spacing was observed indicating partial deswelling of the ﬁller. However, the ﬁnal basal plane spacing was still higher than the modiﬁed ﬁller powder. Similar decrease in the basal plane spacing was also observed for other ﬁllers BzC18OH.Cloi, 2C18.Cloi, and BzC16.Cloi, but in all the cases, the ﬁnal basal plane spacing was signiﬁcantly higher than the corresponding modiﬁed ﬁller powders. In the case of the vermiculite system, the basal plane spacing was observed to enhance to some extent after the composite formation indicating intercalation of the polymer occurred during the nanocomposite synthesis that pushed the ﬁller platelets further apart. Figure 8.9 shows the X-ray diffractograms of the montmorillonite modiﬁed with different modiﬁcations and their 3.5 vol% composites. Figure 8.10 shows the diffractograms of the modiﬁed vermiculite and 3.5 vol% ﬁller composites. As the X-ray analysis provided interesting insights into the interfacial interactions between the ﬁller surface and the epoxy prepolymer, further analysis of the morphology was carried out by TEM. Microstructure evaluation using TEM is also necessary, as the X-ray diffraction does not provide any information on the

233

234

8 In-situ Synthesis and Properties of Epoxy Nanocomposites a)

b)

Figure 8.9 X-ray diffractograms of (a) modiﬁed montmorillonite with various modiﬁcations

and (b) their 3.5 vol% epoxy composites.

exfoliated part of the clay. The intensity of the diffraction peaks is sometimes considered to be a measure of the content of the exfoliated and intercalated part of the ﬁller, but the intensity of the diffraction peak is itself inﬂuenced by sample preparation and mineral defects. Figures 8.11 and 8.12 show the TEM micrographs of Bz1OH.Cloi and BzC16.Cloi containing composites (3.5 vol% ﬁller

8.5 Morphology a)

b)

Figure 8.10 X-ray diffractograms of (a): (I) BzC16 modiﬁed vermiculite and (II) the ﬁller in

epoxy composite and (b): (I) BzC18OH modiﬁed vermiculite and (II) the ﬁller in epoxy composite. Reproduced from Ref. [14] with permission from Sage Publishers.

fraction). The dark lines in the micrographs indicate the ﬁller cross section. Figure 8.13 shows the TEM micrograph of BzC18OH.Vermi containing composite (3.5 vol% ﬁller fraction). Bz1OH composites were observed to have extensive ﬁller exfoliation indicated by the presence of large number of single platelets. The tactoid size was also very thin with very few layers attached to each other. On the other hand, BZC16 composite consisted mainly of intercalated tactoids of varying

235

236

8 In-situ Synthesis and Properties of Epoxy Nanocomposites

Figure 8.11 TEM micrographs of the Bz1OH.Cloi composite (with 3.5 vol% ﬁller fraction).

Reproduced from Ref. [13] with permission from American Chemical Society.

thickness. Composites with ﬁllers like BzC18OH and 2C18 had also similar morphology as BzC16 ﬁller. It conﬁrmed the ﬁndings of the X-ray diffraction that the interfacial interaction was most important factor affecting the generation of exfoliated nanocomposites. The Bz1OH though did not have any signiﬁcant increase in the basal plane spacing during the composite synthesis, but owing to the better

8.5 Morphology

Figure 8.12 TEM micrographs of the BzC16.Cloi composite (with 3.5 vol% ﬁller fraction).

Reproduced from Ref. [13] with permission from American Chemical Society.

compatibility with the polymer, majority of the ﬁller was still exfoliated. On the other hand, the ﬁllers with long alkyl chains though had much higher basal plane spacing, but their composites still had intercalated morphology owing to the mismatch between the polarity of modiﬁed ﬁller surface and polymer. The BzC18OH. Vermi composite also showed a mixed morphology with tactoids of varying

237

238

8 In-situ Synthesis and Properties of Epoxy Nanocomposites

Figure 8.13 TEM micrograph of the BzC18OH.Vermi composite (with 3.5 vol% ﬁller fraction).

Reproduced from Ref. [14] with permission from Sage Publishers.

thicknesses owing to the similar reasons. Apart from that, the ﬁller platelets were observed to be bent and folded and had no speciﬁc alignment at any magniﬁcation. Platelet alignment is a critical requirement especially when designing the composites for gas barrier applications as the misaligned ﬁller platelets are only onethird effective as compared to the aligned platelets because they provide less tortuous paths to the permeant molecules.

8.6 Barrier Properties

Oxygen permeation and water vapor transmission through the composites containing 3.5 vol% of the ﬁller were measured and are plotted in Figures 8.14 and 8.15. Oxygen permeation through pure epoxy polymer was observed to be 2.0 cm3 μm (m2 d mmHg) −1. It decreased to 1.6 and 1.7 in composites containing unmodiﬁed montmorillonite and vermiculite, respectively. As indicated by the X-ray diffraction and TEM studies, composites with Bz1OH.Cloi as ﬁller showed the best performance. The oxygen permeation value for these composites was reduced to 0.77 cm3 μm (m2 d mmHg) −1. Bz2OH.Cloi and Bz3OH.Cloi were also effective in reducing the oxygen permeation through the polymer, though with increasing number of OH groups in the surface modiﬁcation also deteriorated the reduction in oxygen permeation. It indicated that increasing the number of OH groups in the modiﬁcations may have increased the polarity beyond the required

8.6 Barrier Properties

Figure 8.14 Oxygen permeation (cm3 μm (m2 d mmHg) −1) through the nanocomposites

[3, 14].

Figure 8.15 Water vapor transmission (g μm (m2 d mmHg) −1) through the nanocomposites

[3, 14].

239

240

8 In-situ Synthesis and Properties of Epoxy Nanocomposites

extent. It conﬁrms again the interplay of interfacial interactions between the ﬁller surface modiﬁcation and the polymer for the ﬁller exfoliation and property enhancement. For the same reasons, the oxygen permeation through the composites with BzC16.Cloi as ﬁller was observed to be less reduced. A reduction of only 20% as compared to the pure polymer was observed. The permeation through BzC18OH.Cloi and 2C18.Cloi containing composites was observed to be higher than the pure polymer itself, indicating a possible phase separation between the long alkyl chains and the epoxy matrix. The oxygen permeation through the BzC16.Vermi and BzC18OH.Vermi were observed to be 1.4 and 1.5 cm3 μm (m2 d mmHg)−1. The BzC18OH ﬁller modiﬁcation in the case of the vermiculite system was observed to be more effective as compared to the montmorillonite system. It should also be noted that pure polymer itself has very low oxygen permeation, therefore, decreasing this low value of permeation further by the incorporation of modiﬁed ﬁllers is signiﬁcant. The water vapor transmission through the composites was more affected by the hydrophobicity of the ﬁllers. The unmodiﬁed ﬁllers owing to their hydrophilic nature registered extensive increases in the water vapor transmission as compared to pure polymer. The values were observed to be 23 and 37 g μm (m2 d mmHg) −1 for montmorillonite and vermiculite, respectively, as compared to 10 g μm (m2 d mmHg) −1 for the pure polymer. Reduction of the water vapor transmission in the composites indicated that the hydrophilic ﬁllers could be signiﬁcantly hydrophobized. The nature of the chemical structure of the modiﬁcation did not affect the water transmission rate much. The barrier properties were also measured as a function of ﬁller volume fraction [13] as indicated in Figure 8.16. Nanocomposites containing ﬁllers Bz1OH.Cloi and BzC16.Cloi were analyzed. The oxygen permeation as well as water vapor permeation decreased through the composites as a function of the increasing ﬁller volume fraction. The oxygen permeation decreased much more signiﬁcantly in the case of Bz1OH.Cloi containing composites as compared to BzC16.Cloi composites at all volume fractions owing to the reason indicated earlier. The water vapor transmission was reduced more in the case of the BzC16.Cloi system owing to the higher hydrophobicity of the ﬁller in this case.

8.7 Effect of Excess Surface Modiﬁcation Molecules

As mentioned above that high resolution TGA was used to check the purity of the ﬁller so that no local bilayer of the excess surface modiﬁcation molecules is present in the ﬁller. The commercially treated ﬁllers, however, are often observed to contain an excess of surface modiﬁcation molecules [16]. This excess can lead to unwanted interactions with the epoxy prepolymer or can thermally degrade at lower temperatures when composites are subjected to higher temperatures; thus, the presence of such excess amount is not required. In order to underline the effect of the excess surface modiﬁcation molecules on the ﬁller surface on the composite properties, epoxy nanocomposites with a number of commercially pro-

8.7 Effect of Excess Surface Modiﬁcation Molecules

Figure 8.16 Relative oxygen permeation and water vapor transmission through the epoxy

nanocomposites (montmorillonite ﬁller) as a function of ﬁller volume fraction [13].

Table 8.5 Abbreviations of the organic cations used for ion exchange on ﬁller surface and modiﬁed montmorillonites.

Cation Dioctadecyldimethylammonium Benzylhexadecyldimethylammonium Bis(2-hydroxyethyl) methylhydrogenatedtallowammonium

M680a) 2C18 · M680 BzC16 · M680 C182OH · M680

M880a) 2C18 · M880 BzC16 · M880 C182OH · M880

Reproduced from Ref. [16] with permission from Springer. a) M680 and M800 represent the montmorillonites with CEC values of 680 and 880 μeq g−1, respectively.

cured modiﬁed montmorillonites were prepared. For comparison, similar modiﬁcations were also exchanged in the lab on the ﬁller surface and the composites with these modiﬁed ﬁllers were also generated. The ﬁller fraction of the composites was always kept the same in these composites so as to compare the performance. Table 8.5 details the cations exchanged on the ﬁller surface. Montmorillonites with two different cation exchange capacity values were also used (M in the table signiﬁes montmorillonite). The oxygen permeation through the epoxy composites containing commercially procured and lab treated montmorillonites was measured and the results are

241

242

8 In-situ Synthesis and Properties of Epoxy Nanocomposites

Figure 8.17 Oxygen permeation through the epoxy nanocomposites with both commercially

procured and lab treated montmorillonites [16].

described in Figure 8.17. All the composites containing the commercially modiﬁed montmorillonites had the oxygen permeation much higher than the pure polymer itself. On the other hand, when the same modiﬁcation was exchanged on the surface of the ﬁller in the lab and the excess modiﬁcation was carefully washed, the oxygen permeation through the composites was much more representative of the morphology of the composites. The permeation through the composites also decreased as compared to the pure polymer and was also dependent on the chemical architecture of the modiﬁcation as well as the cation exchange capacity of the ﬁller. In the case of commercially procured montmorillonite systems, the excess modiﬁcation may have interacted negatively with the polymer thus negating the effect of ﬁller. It was also of interest to analyze the microstructure of the composites with commercially modiﬁed as well as lab-modiﬁed montmorillonites. The TEM images of commercially modiﬁed and self-modiﬁed BzC16.M680 are shown in Figure 8.18. Both the micrographs pointed toward the presence of a mixed morphology in the composites and there were no signiﬁcant differences when compared with each other. It thus indicated that the intercalation and subsequent exfoliation of the clay layers may not be generally affected by the presence of excess surface ions; however, the speciﬁc interactions of these ions with the polymer chains in the interlayer do change the polymer characteristics and the interfacial interactions to generate possible higher free volumes signiﬁcant enough to impact the sensitive permeation properties [16].

8.7 Effect of Excess Surface Modiﬁcation Molecules a)

b)

Figure 8.18 TEM micrographs of epoxy nanocomposites with (a) commercially modiﬁed

BzC16.M680 and (b) self-modiﬁed BzC16.M680 (reproduced from Ref. [16] with permission from Springer).

243

244

8 In-situ Synthesis and Properties of Epoxy Nanocomposites

References 1 Brydson, J.A. (1975) Plastic Materials, 2

3

4

5 6

7

Newnes-Butterworths, London. May, C.A. (1988) Epoxy Resins Chemistry and Technology, Dekker, New York. Lee, H. and Neville, K. (1967) Handbook of Epoxy Resins, McGraw-Hill, New York. Ellis, B. (1993) Chemistry and Technology of Epoxy Resins, Blackie Academic & Professional, London. Silvis, H.C. (1997) Trends Polym. Sci., 5, 75. Brennan, D.J., Haag, A.P., White, J.E., and Brown, C.N. (1998) Macromolecules, 31, 2622. Gusev, A.A. and Lusti, H.R. (2001) Adv. Mater., 13, 1641.

8 Eitzman, D.M., Melkote, R.R., and

Cussler, E.L. (1996) AIChE J., 42, 2. 9 Fredrickson, G.H. and Bicerano, J. (1999)

J. Chem. Phys., 110, 2181. 10 Lan, T., Kaviratna, P.D., and Pinnavaia,

T.J. (1995) Chem. Mater., 7, 2144. 11 Kamon, T. and Furakaw, H. (1986) Adv.

Polym. Sci., 80, 177.

12 Barton, J.M. (1985) Adv. Polym. Sci., 72,

120. 13 Osman, M.A., Mittal, V., Morbidelli, M.,

and Suter, U.W. (2004) Macromolecules, 37, 7250. 14 Mittal, V. (2008) J. Compos. Mater., 42, 2829. 15 Osman, M.A., Mittal, V., and Suter, U.W. (2007) Macromol. Chem. Phys., 208, 68. 16 Mittal, V. (2008) J. Mater. Sci., 43, 4972.

245

9 Unsaturated Polyester–Montmorillonite Nanocomposites by In-situ Polymerization Michal Kedzierski

9.1 Introduction

Two types of polymerization reactions are usually used to synthesize polyester resins containing unsaturated bonds: (i) polycondensation of glycols with anhydrides of unsaturated dicarboxylic acids, and (ii) polyaddition of epoxy compounds to unsaturated dicarboxylic acids (commonly acid anhydrides activated by glycols). In both the cases, the formed prepolymers are subsequently dissolved in an unsaturated crosslinking agent, most commonly styrene, and the resulted resin is ﬁnally cured by free radical copolymerization to yield the insoluble and infusible crosslinked network. Due to low-cost raw materials, simple preparation procedures, and a variety of glycols and acid monomers, which allow us to tailor the properties of the cured polymer, unsaturated polyester (UP) resins are widely applied as binders in ﬁber-reinforced laminates and composites for construction, transportation, and building industry. Since the discovery and commercialization of nylon–montmorillonite nanocomposite by Toyota researchers, the increasing attention has been paid to the modiﬁcation of polymer properties by the introduction of nanosized clay ﬁllers. Consequently, a number of studies have been conducted to prepare nanocomposites of UP resins with montmorillonite-type clays and possibly to improve mechanical, thermal, and other material properties at a much lower loading of a reinforcing additive than using conventional reinforcements like glass ﬁber. Since the process of synthesis and curing of UP resin comprises three subsequent stages (Scheme 9.1), clay nanoﬁller can be introduced in situ through one of three ways: (i) during the synthesis of prepolymer, (ii) by mixing with prepolymer before dissolution in crosslinking monomer, and (iii) by mixing with UP resin before ﬁnal copolymerization and crosslinking. Most of the literature reports concern methods (ii) and (iii), which are more convenient from the practical point of view.

In-situ Synthesis of Polymer Nanocomposites, First Edition. Edited by Vikas Mittal. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA.

246

9 Unsaturated Polyester–Montmorillonite Nanocomposites by In-situ Polymerization O O

O

+

O

O

2 HO-R-OH

O - H2O

1. polycondensation

O

+

O

O R

O

R O

n

UP pre-polymer (alkyd)

2. dissolution in crosslinking monomer

UP resin

3. crosslinking

peroxide initiator accelerator

cured UP

Scheme 9.1

Synthesis and curing of unsaturated polyester.

9.2 Nanocomposites with MMT Introduced into UP Prepolymer or Resin 9.2.1 Synthesis, Morphology, and Mechanical Properties

The preparation of UP resin–clay nanocomposite was ﬁrstly reported by Lee and Giannelis in a short article [1]. The mixing of preaccelerated resin with pristine montmorillonite for 3 h at 80 °C, followed by peroxide-initiated curing gave nanocomposite of the delaminated structure. Using montmorillonite (MMT) modiﬁed with alkylammonium cations, containing 12 to 18 carbon atoms in the chain, intercalated products were obtained. The increase in Young’s modulus, thermal stability, and ﬂame retardancy in comparison with neat polymer were stated, although no detailed information was given.

9.2 Nanocomposites with MMT Introduced into UP Prepolymer or Resin

A study on the preparation and properties of UP/MMT system was undertaken by Kornmann and coworkers [2]. They used low viscosity UP resin with high styrene content (42 wt%). With the aim to increase the compatibility of MMT with organic phase, the authors used two types of silane coupling agents, both containing unsaturated functions, that is, vinylbenzylamine and methacrylate. UP resin mixed with Co accelerator and silane-modiﬁed MMT was stirred for 4 h at 60 °C, then cured with peroxy initiator for 3 h at room temperature and postcured for 3 h at 70 °C. X-ray diffraction (XRD) and transmission electron microscopy (TEM) study showed that partially delaminated nanocomposites were formed at 5 and 10 wt% loading of the MMT (corresponding to 2.5% and 5% by volume, respectively). Tensile tests showed 32% increase in Young’s modulus at 5 vol % content of MMT in cured UP, while tensile strength was almost unchanged up to 4 vol% clay content and decreased by 32% at 5% MMT volume fraction. The signiﬁcant improvement in fracture toughness was observed for the cured UP/MMT system. The fracture energy of the nanocomposite containing 5 vol% of clay was three times as high as its value for the pure UP. The mechanism of UP-layered aluminosilicate nanocomposite formation was discussed in the paper by Park et al. [3] The authors used two methods of introducing the clay into UP resin. The ﬁrst was simultaneous mixing of UP prepolymer with styrene monomer and alkylammonium-modiﬁed MMT for 3 h at 60 °C. The second, named as sequential mixing, consisted of two stages: (i) preintercalation of organoclay with UP prepolymer and (ii) mixing of the resulting preintercalate with styrene at 60 °C. In both cases, XRD patterns of the cured products indicated that the intercalation took place, resulting in a shift of (0 0 1) basal reﬂection of MMT to lower angles. Further evidence for the formation of a nanocomposite was obtained from TEM micrographs, showing partial exfoliation of MMT layers, along with areas containing ordered clay sheets. Dynamic mechanical thermal analysis (DMTA) experiments indicated the decrease in Tg of cured UP–alkylammonium MMT nanocomposite synthesized by the simultaneous method as compared to the unmodiﬁed resin. This can be explained by lower crosslink density of UP nanocomposite, due to the fact that styrene molecules diffuse faster into the gallery of clay than UP chains. Consequently, a part of crosslinking monomer is consumed in the homopolymerization process. In a sequential process, polyester chains are preintercalated in the MMT gallery and crosslinking of UP takes place homogeneously inside and outside of the clay layers. In effect, Tg of the cured nanocomposite increases with mixing time, reaching the value close to the unﬁlled UP resin. The reinforcing effect of MMT platelets was shown by the increase of storage modulus of cured UP/MMT nanocomposite as compared with pure polyester. The relationships between morphology and properties of UP resin nanocomposites containing from 1 to 10 wt% of MMT modiﬁed with methyl-tallow-bis (2-hydroxyethyl)-quaternary ammonium (MTHEA) cations were investigated by Bharadwaj and coworkers [4]. The morphological characterization was based on TEM micrographs taken at different magniﬁcations, with an aim to study both the short- and long-range order and dispersion of aluminosilicate in the polyester matrix (Figure 9.1). The presence of fully exfoliated clay sheets in some regions

247

248

9 Unsaturated Polyester–Montmorillonite Nanocomposites by In-situ Polymerization a)

b)

c)

Figure 9.1 TEM micrographs of crosslinked

polyester nanocomposite containing 2.5 wt% clay showing: (a) microstructure at low magniﬁcation, (b) intercalated and exfoliated

sheets at high magniﬁcation of the aggregate region shown in (a), and (c) fully exfoliated sheets [4]. Reproduced with permission from Elsevier.

9.2 Nanocomposites with MMT Introduced into UP Prepolymer or Resin

Figure 9.2 Illustration of the morphological hierarchy at different length scales in the

polyester–clay nanocomposites [4]. Reproduced with permission from Elsevier.

as well as intercalated aggregates in others was revealed (Figure 9.2). However, contrary to expectations, the mechanical properties expressed by tensile modulus and dynamic storage and loss moduli were progressively decreased with the increase in clay concentration. The most signiﬁcant reduction in the tensile modulus was observed for the nanocomposite with 2.5 wt% clay, which displayed a relatively high degree of exfoliation. The authors explained this decrease in mechanical properties as partially due to the drop in the degree of polyester crosslinking along with the intercalation and exfoliation of clay sheets in the resin matrix. Another unexpected result was that the rate of thermal degradation of cured UP was slightly accelerated upon the formation of nanocomposite with MMT, which was ascribed to the inﬂuence of oxygen containing MTHEA cations. The crosslinked polyester clay nanocomposites were optically clear even at 10 wt% clay content and had improved barrier properties in comparison with unmodiﬁed UP. The greatest decrease in oxygen permeability was observed for the nanocomposite containing 2.5 wt% clay, which well correlates with its morphology showing the exfoliation occurring at a more global scale.

249

250

9 Unsaturated Polyester–Montmorillonite Nanocomposites by In-situ Polymerization

The same hydroxy functional organoclay, that is, MTHEA-treated MMT (known under the trade name Cloisite 30B®) was used for the modiﬁcation of UP resin by Inceoglu and Yilmazer [5]. They studied morphology and mechanical properties of the nanocomposites prepared from low viscosity (39 wt% styrene) UP resin, mixed for 3 h at 50 °C and sonicated for 30 min with two montmorillonites: containing sodium and MTHEA cations. XRD analysis of the cured compositions show that the intercalated nanocomposite was formed up to 3 wt% loading of organoclay (with increase in the basal spacing of MMT to d001 = 45 Å). For the compositions with organoclay content in the range 5–10 wt% the main XRD peak was corresponding to d001 = 1.9 nm (the initial basal spacing of MTHEA treated MMT) suggesting that part of the organoclay was not intercalated by UP molecules. An increase in Tg was observed, from 72 °C in the unﬁlled polyester to 86 °C in the composite with 10% organically modiﬁed MMT. This was explained by restricted segmental motion near the organic–inorganic interface. The dispersed nanoclay particles signiﬁcantly improved the mechanical properties of the cured polyester: the ﬂexural modulus reached the maximum value at 3 wt% of clay loading (increase by 35% with respect to unmodiﬁed UP) and the maximum in tensile modulus was observed with an addition of 5 wt% clay (increase by 17%). The impact strength decreased at low organoclay content, possibly due to the action of clay particles as crack initiators and then started to increase with the increasing content of nanoclay ﬁller (acting also as crack stoppers and making the tortuous crack propagation path resulting in higher impact energy). The use of ultrasonic mixing after the mechanical one had a positive effect on the mechanical properties of UP nanocomposites. For the composites with sodium MMT, only a slight increase in Young’s modulus was observed along with no signiﬁcant change in the tensile properties and a continual decrease in the impact strength with respect to the clay content. The low effect of MMT without organic treatment on the mechanical properties of UP can be explained by a lower degree of exfoliation, indicated also by SEM analysis. Xu and coworkers studied the correlation between the mechanical properties of UP–MMT nanocomposites and the interlayer spacing of organically modiﬁed clay [6]. By changing the concentration of organic modiﬁer (octadecyl trimethyl ammonium chloride), they prepared organoclays with the basal spacing varying from 2.4 to 4.0 nm, and then they used them to synthesize nanocomposites with UP resin. An increase in bending strength from 16.6 to 75.5 MPa as well as an improvement in impact strength from 0.6 to 1.9 kJ m−2 was reported for the cured UP organoclay composition containing the MMT of the largest interlayer spacing. The effect of curing monomer polarity on the properties of UP/MMT nanocomposite was investigated by Min and colleagues [7]. They used the UP resin containing hydroxypropylacrylate (HPA) monomer instead of less polar styrene and two alkylammonium modiﬁed clays of different polarity as nanoﬁllers. The clays (in loading from 1 to 5 wt%) were mixed with the molten UP prepolymer at 180 °C for 10 min, than HPA stabilized with hydroquinone was added and the mixing was continued at 50 °C for 30–120 min, followed by curing using conventional

9.2 Nanocomposites with MMT Introduced into UP Prepolymer or Resin

80

10

70

9

60

8

50

7

Impact strength (KJ/m2)

Tensile strength (MPa)

procedure. No signiﬁcant inﬂuence of the mixing time on UP conversion neither the morphology of the cured nanocomposite was observed. However, with increase in the mixing time, mechanical properties of the cured polyesters were improved. This effect was more signiﬁcant in the case of polyester modiﬁed with less polar alkylammonium MMT. The authors compared the ﬂexural properties of cured UP–MMT nanocomposites with styrene and HPA monomers. In both cases, the ﬂexural modulus increased with clay content. The more marked enhancement of the modulus was observed in the case of styrene-based nanocomposites, which was interpreted as the result of more favorable diffusion of nonpolar styrene into the galleries of alkylammonium MMT than in the case of polar HPA. However, only a slight improvement in the tensile properties was observed for nanocomposites containing 1 wt% clay obtained at long mixing times. The addition of more organoclay resulted in no further increase or even a decrease in tensile modulus. Zhang and coworkers used for the modiﬁcation of UP resin three MMT clays containing various cations, that is, sodium, alkylammonium, and methacryloxyalkylammonium [8]. The thorough mixing of UP resin with clay ﬁllers was applied, for over 24 h at 50 °C. XRD patterns of the cured UP–clay hybrids indicated only partial intercalation of sodium–MMT by polyester, while the interlayer distances of organically modiﬁed clays expanded beyond the values detected by XRD (8.8 nm). TEM analysis showed that most of MMT was dispersed in the UP matrix as small aggregates; part of them were dispersed into layers. For both organically modiﬁed clays, a distinct increase in tensile strength, impact strength and heat distortion temperature of cured UP was observed at MMT content of 4 wt% (see Figure 9.3 and Table 9.1). The hydrophilic sodium

Impact strength Tensile strength 40

6 –1

0

1

2

3

4

5

6

7

Content of MBDAC-MMT (%) Figure 9.3 Effect of the organoclay content of the mechanical properties of UP nanocompos-

ites [8]. Reproduced with permission from Wiley Interscience.

251

252

9 Unsaturated Polyester–Montmorillonite Nanocomposites by In-situ Polymerization Table 9.1 Properties of the UP organoclay nanocomposites [8].

MT type

MMT content (wt%)

Tensile strength (MPa)

Impact strength (KJ rn−2)

Hardness (Barcol)

Heat distortion temperature (°C)

Pristine UP MMT–Na MMT–CTAB MMT–MBDAC

0 4.0 4.0 4.0

44.1 47.8 56.4 71.2

6.32 4.35 8.44 9.63

12.4 13.3 13.1 13.8

86.2 89.8 103.1 110.4

Reproduced with permission from Wiley Interscience.

MMT exerted only a minimal effect on heat distortion temperature (HDT) and tensile properties, while decreasing impact resistance. The best properties were obtained for the UP modiﬁed with organoclay bearing polymerizable methacryloxy group, that is, 61% increase in tensile and 51% in impact strength as well as 24 °C increase in HDT. The simultaneous reinforcing and toughening effect of organoclays is noteworthy, because the presence of conventional toughening agents in thermosets (like thermoplastic rubbers) usually involves a decrease in the strength of the composites. However, even the improved properties of UP/clay nanocomposites still rank them among the brittle plastics. Differential scanning calorimetry (DSC) measurements showed that after the second heating glass transition peaks were present in the thermograms of unmodiﬁed UP and its composite with sodium MMT, but it did not appear on DSC curves of UP modiﬁed with organoclays. This indicates strong intermolecular or covalent bonding between clay ﬁller and polyester. Alkylammonium MMT can serve as a physical crosslinking agent, whereas MMT containing polymerizable methacryloxy group can add additional covalent crosslinks into the cured polyester. This is also reﬂected in enhanced thermal stability as shown by TGA measurements. The temperature of maximum weight loss for UP modiﬁed with organoclays was 24–30 °C higher than that of the neat UP polymer. Another property of cured polyester which was signiﬁcantly improved with the increase in the organoclay content was swelling resistance. As is known from many other studies on polymer–MMT nanocomposites, the aluminosilicate layers dispersed in the polyester matrix can act as physical barriers limiting the diffusion and swelling processes. The use of MMT intercalated with polymerizable organocations as a reactive nanoﬁller of UP resin was also investigated by Fu and Qutubuddin [9]. Two unsaturated alkylammonium salts, that is, vinylbenzyl n-alkyldimethyl (n = 12 or 18) were used for the functionalization of MMT by ion exchange in an aqueous medium. The resulting organoclays were dispersed in the UP resin by mechanical stirring and 4 h sonication. After curing, an intercalated nanocomposite was obtained with MMT containing dodecyl alkyl chain, while the organoclay with longer octadecyl chain was partially exfoliated in the cured UP matrix. For both

9.2 Nanocomposites with MMT Introduced into UP Prepolymer or Resin

nanocomposites an increase in the dynamic storage modulus as compared with the pristine UP was observed. Nanoindentation tests were used by Dhaka and collaborators to study the effect of various organoclay concentrations on the nanomechanical properties of UP nanocomposites [10]. A strong correlation was found between the mechanical properties and interlayer d-spacing of clay particles in the nanocomposite system. Cured UP resins with incorporated 1%, 3%, and 5 wt% organoclay exhibited hardness increased by 29%, 24%, and 14%, respectively. Also the elastic modulus was increased by 23% with the introduction of 5 wt% organoclay. Flexural and tribological properties of polyester resin–clay nanocomposites were examined by Balasubramanian et al. [11]. Using isophthalic UP resin and dodecylamine modiﬁed clay in amount of 1 wt%, they prepared exfoliated nanocomposites, as indicated by XRD patterns. They showed 20% increase in ﬂexural modulus and 85% decrease in speciﬁc wear rate as compared with pristine UP. Also the coefﬁcient of friction was signiﬁcantly lower for nanocomposite than in the case of unmodiﬁed polyester. In another work of the same authors, dodecylamine modiﬁed clay was applied for the modiﬁcation of the UP-based gel coat system [12]. The best mechanical properties were observed for nanocomposite gel coat with 2 wt% clay, including an increase in tensile and impact strength, respectively, by 21 and 33% compared to the conventional gel coat. Further increase in organically modiﬁed montmorillonite (OMMT) content leads to the decrease of the strength values. The nanocomposite gel coat system showed better water absorption resistance than the conventional UP gel coats. In a recent study, ¸ Sen introduced into UP resin two MMT organically modiﬁed in different ways: by ion exchange with cetyl trimethyl ammonium salt (CTA) and in the reaction with trimethoxy vinyl silane (TMVS) [13]. The compositions were prepared by mixing UP resin with the modiﬁed clays (used in amount 3% of the weight of the resin) for 24 h at room temperature, followed by the curing at room temperature for 24 h and at 80 °C for 5 h. XRD studies showed that silane treated MMT, as well as its composition with UP, did not indicate an increase in the basal spacing, possibly because the silanol groups interacted only with the edges of the clay. In the case of organoclay with CTA a successful intercalation of UP resin was conﬁrmed by an expansion of the interlayer spacing to 3.4 nm. Using the MMT modiﬁed both with TMVS and CTA, exfoliated nanocomposites were obtained, as indicated XRD and AFM analyses. This twofold MMT modiﬁcation resulted also in better thermal and dynamic mechanical properties when compared with pure UP or polyesters ﬁlled with only silanized or ion-exchanged MMT. 9.2.2 Rheology and Cure Properties

Effect of the processing parameters, including mixing mode, shear level, organoclay content, and temperature, on the morphology of UP–MMT hybrids were investigated by Narkis and coworkers [14]. They studied two nanocomposite

253

9 Unsaturated Polyester–Montmorillonite Nanocomposites by In-situ Polymerization

systems: UP resin–organoclay and UP alkyd-organoclay using 5 or 20 parts of various organically modiﬁed clays per hundred parts of the ﬁnal composite. The mixing of UP resin and organoclay was carried out for 1 h at room temperature using mechanical stirring or ultrasonication. In the second procedure, styrene-free alkyd and clay powder were hand mixed at 80 °C until a homogeneous mixture was obtained. The blends were further processed by three methods. The ﬁrst one was mechanical mixing at low (400 rpm) or high (1800 rpm) shear level and at the temperature of 80 °C or 130 °C, for several periods of time up to 24 h. Alternatively ultrasonication or static heating were applied at 80 °C for 2 or 24 h. Both procedures gave nanocomposites of intercalated structure, the extent of which depended on the type of clay treatment. For the UP resin–organoclay system, the maximum increase in basal spacing of clay (up to 3.8 nm as indicated by XRD analysis) was observed in the case of nanocomposite containing octadecylamine-treated MMT. The higher extent of intercalation with interlayer spacing above 5.5 nm was obtained for some UP alkyd–organoclay systems. Upon the addition of clays to alkyds, an increase of viscosity was observed, reaching the maximum for 24 h of mixing. Rheological studies show a correlation between the blend viscosities and the interlayer spacing of nanocomposites. The higher mixing temperature did not increase the intercalation extent, probably due to the reduced shear level. In conclusion, the authors stated that application of high shear levels (by vigorous mechanical stirring or increasing the clay content in the blend) for prolonged times promotes the intercalation and exfoliation and results in a better dispersion of clay particles in the resin matrix (Figure 9.4). The rheology and ﬂow properties of the organoclay suspension in UP resin were investigated by Rajabian and Beheshty [15]. MMT with dimethyl dehydrogenated

5.5 5.0 4.5 Normalized Intensity

254

1800 rpm, 24 h

4.0 3.5

1800 rpm, 6 h

3.0

1800 rpm, 0.5 h

2.5 2.0

400 rpm, 24 h

1.5

400 rpm, 2 h

1.0 0.5

I.28 MC Organo-Clay

0.0 1

2

3

4

5 6 2 Theta

Figure 9.4 XRD patterns of neat organoclay

(I.28MC) and alkyd/7.3 wt% organoclay nanocomposite, prepared by mechanical mixing at low- and high-shear levels/low

7

8

9

10

temperature for several durations. Intensity values were normalized and diffraction patterns shifted for clarity. [14]. Reproduced with permission from Wiley Interscience.

9.2 Nanocomposites with MMT Introduced into UP Prepolymer or Resin

tallow ammonium cations was suspended in UP resin using several steps of the mixing process with the increasing stirring rate. XRD spectra indicated the formation of nanocomposite with a mixed exfoliated /intercalated morphology. The effects of OMMT content and mixing mode on rheology behavior at different shear rates were examined and results were compared with a developed rheological model. The suspensions showed strong shear thinning behavior, caused by the orientation of MMT platelets induced by the shear ﬁeld. The steady state and transient viscosity values increased with a higher OMMT content and a longer shearing time. This work can be of practical interest in the processing of nanocomposites with the methods such as resin transfer molding and reaction injection molding, where knowledge of rheology is needed to appropriately design and control the molding stages. Research on the crosslinking of UP resins in the presence of organically modiﬁed clays was conducted by Xu and Lee. They have found that by adding 1–3 wt% organoclay to UP–styrene-poly(vinyl acetate) system cured at room temperature, the volume shrinkage of the crosslinked polymer could be eliminated [16]. In another study Xu and Lee investigated the effect of nanoclay on the cure kinetics of UP resin and the mechanical properties of the resulting UP nanocomposites [17]. Based on the DSC rate proﬁles and the evaluation of polymerization progress by Fourier transform infrared spectroscopy (FTIR) spectroscopy, they stated that the addition of nanoclay neither changes the curing reaction mechanism nor the degree of ﬁnal conversion, but it affects the initiation stage shortening the induction time. It was explained by the action of alkylammonium cations intercalated between the clay layers or the negative charged clay surface as additional promoters in the redox curing process. XRD spectra of the cured UP nanocomposites containing 3 to 7 wt% alkylammonium modiﬁed clay showed the highly disordered intercalated structure. The MMT gallery spacing increased to 3.5 nm, independently of the clay loading. The increases in tensile modulus and fracture toughness parameter KIC (respectively by 16% and 30% at 5 wt% clay content) accompanied by the reduction in tensile strength and strain has been observed. In another study on the cure behavior of UP/modiﬁed MMT nanocomposites, Zhou and collaborators applied dynamic rheology to determine the gel times of nanocomposite UP resins [18]. Two organophilic MMTs were used: nonreactive with long alkyl chain, and reactive with double bonds introduced on the interlayer surface. When using the reactive clay, the gel rate at 35 °C was decreased markedly with increasing content of clay in the range 1–3 phr. This phenomenon was explained by the decomposition of peroxide initiator by acidic sites on the MMT surface. At 5 phr clay loading the gel rate was increased to some extent, which was ascribed to the increased participation of double bonds from the layers of MMT in the crosslinking process. The gel time–temperature relationship and DSC analysis indicated that the activation energy increased and the total exotherm of cure reaction decreased with the addition of clay. From in-situ FTIR measurements it was concluded that the cure mechanism of UP-reactive clay nanocomposites was different from that of pristine UP. In contrast to the latter, the conversion of

255

256

9 Unsaturated Polyester–Montmorillonite Nanocomposites by In-situ Polymerization

styrene was lower than that of UP double bonds at room temperature curing stage and it was increased only at the postcuring stage at 80 °C. Moreover, the degree of conversion of unsaturated bonds in the UP chains was higher for the nanocomposite with reactive clay than for the pristine UP. The different behavior was observed for the nanocomposites prepared with nonreactive clay, in this case the course of the curing reaction was similar to that of pure UP. Narkis and coworkers investigated the curing of styrene free UP prepolymer (alkyd) in the presence of organoclay [19]. Depending on the peroxide initiator content either an exfoliated or a combined intercalated/exfoliated structure was obtained. The inﬂuence of the organically modiﬁed clays on the storage stability (shelf life) of UP–clay compositions was investigated by Oleksy and coworkers [20]. They employed an original method of the modiﬁcation of smectite clays (bentonites) with quaternary ammonium salts in an alkaline aqueous suspension, providing up to 93% of cation exchange [21]. This method allowed an easy processing of bentonite suspensions, without ﬁltering and rinsing problems, which is important from the practical viewpoint. The modiﬁed clays were processed into UP resins in two stages: ﬁrst a concentrate containing 10 wt% of bentonite in UP was prepared and grinded in laboratory homogenizer at 750 rpm, next the concentrate was diluted with more resin to prepare ﬁnal UP/clay compositions. The time to spontaneous gelation of the modiﬁed UP resins was measured at 70 °C. The addition to 2 wt% organoclay increased the stability time by the factor up to 16, with only a slight reduction of the resin reactivity. The most effective in stabilizing UP were bentonites modiﬁed with long aliphatic ammonium salts bearing a benzyl substituent. The proposed mechanism of stabilization involved the reaction of free radicals with the alkylammonium groups with subsequent formation of more stable radicals (e.g., mesomerically stabilized benzyl) (Scheme 9.2). The presence of modiﬁed bentonites in a cured UP resin resulted in signiﬁcant improvement of the mechanical strength, which was depended on the clay ion-exchange capacity and the type of alkylammonium salt. Increases in the tensile strength (of up to 62%), elongation at break (58%) unnotched impact strength (100%), and Brinell hardness (70%) were achieved at 4 wt% clay loading. UP–MMT nanocomposites cured under UV radiation were investigated by Kim and coworkers [22]. Two radical type photoinitiators were used: benzyl dimethyl ketal (BDK) and 1-hydroxy cyclohexyl phenyl ketone (HCPK) and hydroxyfunctional MTHEA treated MMT as a nanoﬁller. XRD analysis of the cured hybrids showed the disappearance of MMT basal spacing peak at 1 wt% clay loading due to the delamination of aluminosilicate. The patterns of nanocomposites with higher MMT loading contained a small MMT peak corresponding to interlayer spacing near 2 nm, indicating that the clay tactoids were not exfoliated completely and some of them remained as aggregates. An increase in MMT content resulted in the enhanced rubbery plateau modulus and improved tensile properties in comparison with the unﬁlled UP. The type and amount of photoinitiator affected the level of improvement in mechanical properties and HCPK seems to be more efﬁcient due to the beneﬁcial inﬂuence of its hydroxyl group on the miscibility

9.2 Nanocomposites with MMT Introduced into UP Prepolymer or Resin

+

R2 N

X

R3

_

C O

+

C• 18

R4

+

R2

C O N

X

R3

_

R4 or

N

+

+

R2

C

free radicals formed by photooxidation

or

R3

X

• CH2

mesomerically stabilized radical

_

R4 Scheme 9.2 Proposed mechanism of increase in UP resin shelf stability by quaternary ammonium salts [20]. Reproduced with permission from Wiley Interscience.

with UP resin and the interaction with organoclay. Thermal and dielectric properties of nanocomposites were also examined showing the increase in the initial temperature of the thermal degradation as well as the relative permittivity and dissipation factor proportionally to the MMT concentration. Rudd and collaborators investigated the potential use of nanoclays along with conventional low proﬁle additive for reducing volumetric shrinkage in UP resin [23]. MMT modiﬁed with dimethyl dihydrogenated tallow quaternary ammonium (DMDTA) was used as a nanoﬁller. The exfoliated nanocomposite, as evidenced by TEM, was obtained at 4 wt% clay loading. The morphology of cured hybrids changed with an increase of shear mixing speed up to 500 rpm, the higher shear rates had no signiﬁcant inﬂuence on the level of exfoliation. The resins containing organoclay exhibited increased viscosity and shear-thinning behavior. Above 10 wt% clay loading the problems with injecting the composition into the mold were experienced. The hybrid matrices consisting of clay and poly(vinylacetate) were used to impregnate random E-glass preforms via RTM. The study indicated that the exfoliated clay systems work synergistically with conventional additives in the reduction of shrinkage possibly due to the immobilization of resin particles at the clay–matrix interface. Consequently, less low proﬁle additive (LPA) level is required, which, in turn, reduces the VOC emission; moreover the presence of

257

258

9 Unsaturated Polyester–Montmorillonite Nanocomposites by In-situ Polymerization

nanoclay improves tensile strength and modulus by 53% and 108%, respectively, over that of a conventional low proﬁle laminate. Another study on UP–LPA–OMMT system was carried out by Beheshty and colleagues [24]. Using simultaneous mixing of UP prepolymer, styrene, and MMT modiﬁed with DMDTA they obtained nanocomposites with a mixed intercalatedexfoliated morphology. By several steps of 24 h mixing with stirring rate increasing from 85 to 400 rpm, UP–OMMT preblend was obtained, subsequently diluted with UP resin and ﬁnally sheared at 2000 rpm for 8 h at 40 °C. The ﬂexural modulus of the cured UP modiﬁed with 3 wt% organoclay was increased by 31.5%, while the Izod impact strength was improved by 51.7% at only 1 wt% of OMMT loading. The addition of OMMT to the low-proﬁle UP/St systems did not noticeably change the shrinkage control due to very high compatibility of the UP resin with LPA (poly(vinyl acetate)) preventing the precipitation of the LPA phase. However, the addition of organoclay substantially compensates the loss of ﬂexural and impact strength resulting from the addition of LPA. 9.2.3 Flammability

Among the most promising applications for polymer–MMT nanocomposites are the materials with increased ﬁre resistance. The suggested mechanisms of ﬁre retardancy when nanoclay platelets are dispersed in the polymer matrix are the formation of barrier to heat and mass transport as well as an increase in melt viscosity which prevents the dripping and promotes char formation. Nanoclays often act as synergists for other FR additives. The effect of MMT clays modiﬁed by various organic treatments on the burning behavior of the cured UP compositions was investigated by Nazare and coworkers [25]. Organoclays were mixed mechanically with UP resin under high shear (900 rpm) for 60 min at room temperature before polymerization. Some compositions contain additionally conventional ﬁre retardants (FR) such as ammonium polyphosphate, melamine phosphate, and alumina trihydrate, in some cases UP resin was diluted with methyl methacrylate. The compositions with 20 wt% additional FR and clay concentration higher than 5 wt% had a noticeably reduced curing rate. XRD patterns of resin nanoclay hybrids indicated the possible exfoliated structure for MMT functionalized with dimethyl-hydrogenated tallow-2-ethylhexyl quaternary ammonium salt (Cloisite 25A®). Other organoclays gave intercalated nanocomposites and Cloisite 10A® containing benzyl substituent formed only a microcomposite, possibly due to steric hindrance of the aromatic group. Flammability studies using cone calorimetry at 50 kW m−2 heat ﬂux showed that incorporation of 5 wt% organoclay reduced peak heat release rate (PHRR) by 23– 27% and total heat release (THR) values by 4–11%, depending on the clay modiﬁcation. However, no simple correlation was observed between the FR efﬁciency and the degree of clay exfoliation. The synergistic effect was observed for the combination of ammonium polyphosphate and 5 wt% amount of nanoclay, which resulted in the total reduction of the PHRR of polyester resin in the range 60–70%.

9.2 Nanocomposites with MMT Introduced into UP Prepolymer or Resin

The addition of organoclays also enhanced ﬂexural properties of the cured UP compositions. 9.2.4 Mixed-Resin and Filler Systems

Park and coworkers applied UP–MMT nanocomposites in polymer concrete with the result of improved compressive strength, elastic modulus, and splitting tensile strength along with the better thermal performance [26]. In a recent paper of Rozman and coworkers, the effect of clay addition on the tensile properties of UP resin composite with a lignocellulosic ﬁller, Kenaf, was examined [27]. Inorganic sodium MMT was used as a clay ﬁller. In the samples without MMT, with an increase in Kenaf content from 40 to 60 wt%, the tensile strength, modulus, and toughness were noticeably reduced. The same composites containing from 1 to 5 wt% clay, showed a signiﬁcant increase in all tensile properties, eliminating the negative inﬂuence of lignocellulosic ﬁbers. SEM analysis indicated good compatibility between the inorganic clay and matrix; however, MMT particles were observed in an agglomerated state. Mixed UP resin – acrylate terminated polyurethane (ATPU) – organically modiﬁed MMT system prepared by in-situ intercalative polymerization was studied by You and collaborators [28]. In the ﬁrst stage, a mixture of the ATPU, styrene, and organically modiﬁed MMT was prepared, and then the resulting preintercalate was mixed with UP resin and cured. The compositions with ATPU showed the increased impact strength; however, the tensile strength, ﬂexural strength, and heat resistance of the materials are obviously decreased. At the weight ratio between UPR, ATPU, and OMMT equal to 82 : 15 : 3, the impact strength and heat distortion temperature of the nanocomposite were greatly improved, while other properties of the nanocomposites changed only slightly. A synergistic effect of polyester urethane (PU) and organically modiﬁed MMT on the toughness properties of UP resin was described by Pan and collaborators [29]. For the cured UP compositions containing 5 wt% PU and 1 wt% OMMT (with hexadecyltrimethylammonium cations) an 80% increase in impact strength was noted, accompanied by less than 20% reduction in ﬂexural modulus and the decrease in curing shrinkage rate to 3.5%. XRD analysis conﬁrmed the good dispersion of MMT in the polymer matrix at low clay loadings. The synergistically modiﬁed UP resin also showed an improved adhesion to natural sisal ﬁbers. Chozhan and collaborators examined mechanical and thermal properties of the organoclay-modiﬁed hybrid UP–epoxy matrices [30]. The epoxy resin (diglycidyl ether of bisphenol A) was ﬁrstly stirred with 1–5 wt% cetyltrimethylammonium ion exchanged MMT at 70 °C for 24 h and subsequently was mixed with UP resin and peroxide initiator, then 4,4-diaminodiphenylmethane as epoxy curing agent was added, and ﬁnally the composition was cured at 140 °C for 3 h and postcured at 200 °C for 2 h. XRD analysis showed the formation of intercalated nanocomposites. The incorporation of organoclay into UP–epoxy matrix resulted in an increase of Tg and HDT values, compensating to some extent the loss of rigidity caused by

259

260

9 Unsaturated Polyester–Montmorillonite Nanocomposites by In-situ Polymerization Cl

O O

CH2

C CH

O

HC CH

CH2

C

Cl

O

O

CH2

O

O

CH

C

C

CH2

O

n

Figure 9.5 Structure of HUP resin.

polyester component. Also, a signiﬁcant increase in hardness, thermal stability as well as tensile, ﬂexural, and impact properties of UP–epoxy hybrids were observed with an increase in the organoclay content.

9.3 Nanocomposites with MMT Introduced during the Synthesis of Prepolymer

In this approach, organically modiﬁed clay is added in situ during the synthesis of UP prepolymer (alkyd). K˛edzierski and Penczek used this concept to prepare halogen containing unsaturated polyester (HUP)–MMT hybrids by the copolyaddition of epichlorohydrin with maleic and phthalic anhydrides activated with propylene glycol carried out in the presence of various MMT types [31]. The structure of HUP prepolymer is presented in Figure 9.5. The advantage of polyaddition process is a lower reaction temperature (synthesis of prepolymer at 130 °C followed with cis–trans isomerization to form more reactive fumarate double bonds at 170 °C) than in the case of polycondensation process (above 200 °C). This is of importance since the alkylammonium compounds commonly used for the hydrophobization of clay undergo thermal decomposition by Hoffmann elimination at temperatures higher than 180 °C. MMT with four types of cations were used: sodium form of montmorillonite (MMT-Na), DMDTA, MTHEA, and protonated aminododecanoic acid (ADA). The progress of reaction was followed by the determination of the acid number. For the reactions carried out in the presence of organoclays with quaternary ammonium cations DMDTA and MTHEA, much faster decrease in acid number was observed than in the case of neat polyester, which can be explained by the catalytic effect of OMMT on the epoxide–anhydride addition. Otherwise, no acceleration took place when sodium MMT as well as MMT containing primary ammonium ions ADA, were used. Most of the obtained clay-modiﬁed prepolymers show an increase in softening temperature by 5–10 °C as compared with unmodiﬁed HUP synthesized under the same reaction conditions (Table 9.2). With the introduction of clay, the melt viscosities of HUP prepolymers increased in the order: MMT–Na < MMT– DMDTA < MMT–MTHEA. The alkyd modiﬁed with MMT–MTHEA showed signiﬁcantly different rheological properties than other prepolymers modiﬁed with the same amount of other clays; at low shear rates it displayed non-Newtonian behavior (Figure 9.6). This may result from the interaction of the polyester matrix and hydroxyl groups of MTHEA with the possible formation of covalent bonds.

9.3 Nanocomposites with MMT Introduced during the Synthesis of Prepolymer Table 9.2

Properties of the clay modiﬁed HUP alkyds and resins [31].

MMT type

MMT content (wt%)

Pristine HUP MMT/Na MMT/DMDTA MMT/MTHEA MMT/ADA

0 2 2 2 3 2

HUP alkyd softening temperature (°C)

HUP resin Styrene content (wt%)

Viscosity (21 °C) (mPa s)

78 78 83 87 90 88

35 36 35 36 33 34 42

919 1107 1181 1993 2583 6715 1697

Reproduced with permission from Polimery (ICRI).

Viscosity (Pa s)

10000

1000

100

10 0

300

600

900

Shear rate (1/s) Figure 9.6 Melt viscosity (at 120 °C) versus shear rate for HUP alkyds modiﬁed with various

clay additives (䊏 MMT–MTHEA, 䉭 MMT–DMDTA, 䊐 MMT–Na, • unmodiﬁed) [31]. Reproduced with permission from Polimery, ICRI.

The clay-modiﬁed prepolymers were dissolved in styrene to obtain HUP resins. The curing was performed at ambient temperature using conventional redox system, followed with postcure at 80 °C. The presence of organically modiﬁed clay shows a noticeable effect on the properties of cured polyester compositions, leading to an increase in hardness and heat deﬂection temperature HDT as compared with unmodiﬁed polyester (Table 9.3). The most signiﬁcant changes in HUP properties were observed in the case of organoclay with MTHEA containing

261

262

9 Unsaturated Polyester–Montmorillonite Nanocomposites by In-situ Polymerization Table 9.3 Properties of the clay modiﬁed cured HUP resins [31].

MMT type

MMT content (wt%)

Styrene content (wt%)

Hardness (N mm−2)

Heat deﬂection temperature HDT (°C)

Limiting oxygen index LOI (%)

Pristine HUP MMT–Na MMT–DMDTA MMT–MTHEA

0 2 2 2 3 2

35 36 35 36 33 42

116 119.6 123.5 127.6 139.7 134.1

63.5 60.3 64.7 65.6 67.6 68.7

24.1 22.7 22.3 27.4 28.7 21.0

MMT–ADA

Reproduced with permission from Polimery (ICRI).

hydroxyl groups that are capable of forming covalent bonds between clay and UP. No signiﬁcant change in hardness and slight decrease in heat stability were found for the polyester containing inorganic MMT–Na. The ability of UP to bond covalently with the surface of organoclay was supported by the experiment, when the copolyaddition of maleic anhydride, phthalic anhydride, and epichlorohydrin was carried out in toluene suspension of the MMT modiﬁed with amino acid ADA. After the reaction, the clay was separated and washed out from acetone-soluble substances using Soxhlet extraction. The elemental analysis of such modiﬁed clay shows an increase in the carbon content (from initial 19.3 to 33.6 wt%) accompanied with the decrease in nitrogen the content (from initial 1.8 down to 0.7 wt%). The product also contained 5.9 wt% of chlorine. FT-IR spectrum of the modiﬁed clay shows the presence of absorption peaks at 1283 and 1727 cm−1, which may be assigned respectively to asymmetric C–O and C=O stretching vibrations of the ester bond (Figure 9.7). The bands at 1640 and 1587 cm−1 correspond to C=C stretching vibrations of maleic and phthalic units, while the peak at 1702 cm−1 may be assigned to the free carboxyl group. From these analytical data, it may be concluded that the organic fraction being strongly bound to the clay surface contains the oligomeric ester compounds formed in epoxide-anhydride reaction. The interlayer spacing of polyester-modiﬁed clay, calculated from (0 0 1) reﬂection at XRD spectrum as 1.96 nm, is slightly smaller than that of MMT–ADA (2.22 nm). This indicates that the clay preserves its stack-like organization and the polyester is intercalated into the interlayer space. The preliminary study on the ﬂammability of HUP–MMT compositions was performed by the determination of limiting oxygen indexes (LOI). The ﬂame retardance was affected by styrene content in the resin and the type of clay used (Table 9.3). An additional amount of styrene has been added to some compositions in aim to improve the casting properties; on the other hand the excess styrene makes the cured compositions more ﬂammable due to the increased content of polystyrene (LOI = 18) and decreased of HUP (LOI = 24). The cured compositions containing MMT–Na as well as organoclays with DMDTA and ADA, showed

9.4 Conclusions

Figure 9.7 FTIR spectrum of MMT/ADA modiﬁed by the reaction with epichlorohydrine,

maleic anhydride, and phthalic anhydride. Reproduced with permission from Polimery, ICRI.

decreased LOI values in comparison with neat HUP. The highest LOI values, in the range of 27–29%, were determined for the polyesters modiﬁed with hydroxylterminated organoclay with MTHEA. The observed increase in ﬂame retardance is comparable to the effect of antimony trioxide, commonly used as a ﬁre-retardant synergist in the commercial epichlorohydrin-based HUP resins. The cured polyester prepared in the presence of 3 wt% of MMT modiﬁed with MTHEA was analyzed using the XRD method. For comparison, a diffractogram for physical blend of the same organoclay and unmodiﬁed polyester was recorded. The XRD pattern of the blended composition shows the presence of the intense diffraction peak at the low-angle region, corresponding to basal spacing of 1.85 nm. No low-angle reﬂections were observed in the diffractogram of polyester with organoclay, indicating the possible delamination of MMT during the alkyd synthesis, which corresponds well to the observed changes in polyester properties.

9.4 Conclusions

After more than a dozen years of research in the ﬁeld of UP–montmorillonite nanocomposites synthesized by various variants of in-situ polymerization, a lot of experimental data has been gathered and reported in numerous publications. In aim to prepare UP–MMT nanocomposites several approaches were employed,

263

264

9 Unsaturated Polyester–Montmorillonite Nanocomposites by In-situ Polymerization

comprising an introduction of clay particles during the synthesis of UP prepolymer (alkyd), melt mixing of clay with alkyd or a mixing of clay with the solution of alkyd in the crosslinking monomer (UP resin). For an effective dispersion of MMT layers into the polyester phase, a hydrophobization of the clay was needed, which was usually accomplished by ion exchange of the inorganic cations with organic cations (commonly long-chain alkyl ammonium) or by the organic silane treatment. The morphology and properties of the cured UP/MMT hybrids depend on a range of factors, including the mixing mode, time and temperature, shear rate, UP–MMT and organic modiﬁer-MMT ratio, as well as the structure of organic cation used for modiﬁcation. The presence of functional and polymerizable groups in an organic modiﬁer may signiﬁcantly affect the nanocomposite formation and the properties of the ﬁnal product. Also the curing process of UP resin, for example the gel time and the degree of crosslinking, may be inﬂuenced by an interaction between free radical initiator and organoclay as well as the diffusion of styrene monomer and UP chains into MMT interlayer space. The exfoliation of MMT platelets was favored at low organoclay loadings. On further increase in the clay concentration, mainly intercalated and aggregated structures have been formed. The correlations between the mechanical properties and nanocomposite morphology were observed in some cases, on this basis optimal clay concentrations (usually several percent) leading to the maximal values of the selected properties can be determined. The increases in Young’s modulus, tensile and ﬂexural strength, fracture toughness, and hardness of the cured UP due to the action of MMT nanoﬁller were reported in many publications, although various authors obtained different results depending on the method used for the preparation of nanocomposite. Other examples of improved properties were increased heat resistance (HDT), thermal stability, decreased gas permeability, and reduction in volume shrinkage. The presence of MMT layers may have diverse effects on the processability of UP resin; on one hand the formation of nanocomposite is accompanied with the signiﬁcant increase in viscosity, on the other hand the clay can increase the resin stability (shelf-life) and impart the thixotropic properties, useful in some applications. The barrier properties of clay platelets make them interesting from the viewpoint of ﬂame retardance of UP resins. The synergistic action of MMT and conventional FR additives was reported creating the possibility to obtain new ﬁre resistant UP resin compositions.

Abbreviations

ADA AFM ATPU BDK CTA

aminododecanoic acid atomic force microscopy acrylate-terminated polyurethane benzylmethyl diketal cetyl trimethyl ammonium

References

DMDTA DMTA DSC FR FTIR HCPK HDT HPA HUP LOI LPA MMT MMT-Na MTHEA OMMT PHRR SEM TEM Tg TGA THR TMVS UP XRD

dimethyl dihydrogenated tallow ammonium dynamic mechanical thermal analysis differential scanning calorimetry ﬁre retardant Fourier transform infrared spectroscopy 1-hydroxycyclohexyl ketone heat distortion temperature hydroxypropylacrylate halogen containing unsaturated polyester limiting oxygen index low proﬁle additive montmorillonite sodium form of montmorillonite methyl-tallow-bis(2-hydroxyethyl) ammonium organically modiﬁed montmorillonite peak heat release rate scanning electron microscopy transmission electron microscopy glass transition temperature thermogravimetry analysis total heat release trimethoxy vinyl silane unsaturated polyester X-ray diffraction

References 1 Lee, J. and Giannelis, E.P. (1997) Polym.

8 Zhang, Y., Cai, Q., Jiang, Z., and Gong,

2

9 Fu, X. and Qutubuddin, S. (2004) Polym.

3 4

5 6

7

Preprints, 38, 688. Kornmann, X., Berglund, L.A., Skerte, J., and Giannelis, E.P. (1998) Polym. Eng. Sci., 38, 1351. Suh, D.J., Lim, Y.T., and Park, O.O. (2000) Polymer, 41, 8557. Bharadwaj, R.K., Mehrabi, A.R., Hamilton, C., Tujillo, C., Murga, M., Fan, R., Chavira, A., and Thompson, A.K. (2002) Polymer, 43, 3699. Inceoglu, A.B. and Yilmazer, U. (2003) Polym. Eng. Sci., 43, 661. Xu, F., Shen, S.-Y., Zhang, S.-X., and Xie, J. (2005) J. Wuhan Univ. Tech. – Mater. Sci. Ed., 20, 107. Kim, H.-G., Oh, D.-H., Lee, H.-B., and Min, K.-E. (2004) J. Appl. Polym. Sci., 92, 238.

K. (2004) J. Appl. Polym. Sci., 92, 2038. Eng. Sci., 44, 345.

10 Dhakal, H.N., Zhang, Z.Y., and

11

12 13 14

15 16

Richardson, M.O.W. (2006) Polym. Test., 25, 846. Jawahar, P., Gnanamoorthy, R., and Balasubramanian, M. (2005) J. Mater. Sci., 40, 4391. Jawahar, P. and Balasubramanian, M. (2006) J. Nanomater., 1, 1. ¸ Sen, S. (2010) Polym. Compos., 31, 482. Mironi-Harpaz, I.N., Narkis, M., and Siegmann, A. (2005) Polym. Eng. Sci., 45, 174. Rajabian, M., and Beheshty, M.H. (2009) Polym. Compos., 30, 399. Xu, L. and Lee, J.L. (2004) Polymer, 45, 7325.

265

266

9 Unsaturated Polyester–Montmorillonite Nanocomposites by In-situ Polymerization 17 Xu, L. and Lee, L.J. (2005) Polym. Eng. 18 19

20

21

22

23

Sci., 45, 496. Zhou, Y., Yang, X., and Jia, D. (2007) Polym. Int., 56, 267. Mironi-Harpaz, I., Narkis, M., and Siegmann, A. (2006) Macromol. Symp., 242, 201. Oleksy, M., Heneczkowski, M., and Galina, H. (2005) J. Appl. Polym. Sci., 96, 793. Heneczkowski, M., Groszek, G., Oleksy, M., Galina, H., Wójcik, Z., Mental, Z., Krawczyk, B., and Szczepanowska, M. (2000) Thixotropic unsaturated polyester resin composition of improved stability and method of obtaining same. Polish Patent 178 900. Kim, H.G., Han, D.H., Lim, J.C., Oh, D.H., and Min, K.E. (2006) J. Appl. Polym. Sci., 101, 3609. Schubel, P.J., Johnson, M.S., Warrior, N.A., and Rudd, C.D. (2006) Compos, Part A, 37, 1757.

24 Beheshty, M.H., Vafayan, M., and

25

26

27

28

29

30

31

Poorabdollah, M. (2009) Polym. Compos., 30, 629. Nazare, S., Kandola, B.K., and Horrocks, A.R. (2006) Polym. Adv. Technol., 17, 294. Jo, B.W., Park, S.K., and Kim, D.K. (2008) Constr. Build. Mater., 22, 14. Rozman, H.D., Musa, L., Azniwati, A.A., and Rozyanty, A.R. (2011) J. Appl. Polym. Sci., 119, 2549. You, C.-J., Xu, J.-G., Xi, S., Duan, X.X., Shen, J., and Jia, D.-M. (2005) Chinese J. Polym. Sci., 23, 471. Wang, C., Pan, Z., Zhang, L., Yang, D., and Li, W. (2006) Polym. Adv. Technol., 17, 528. Chozhan, C.K., Alagar, M., Sharmila, R.J., and Gnanasundaram, P. (2007) J. Polym. Res., 14, 319. Ke˛dzierski, M. and Penczek, P. (2004) Polimery, 49, 801.

267

10 Polymer Clay Nanocomposites by In-situ Atom Transfer Radical Polymerization Hanying Zhao

Since the publication of Toyota’s pioneering work on polymer–silicate (or clay) nanocomposites, the preparation of polymer clay nanocomposites has attracted intense interest in the materials science community [1–6]. By introducing a few weight percent of clay into the polymer matrix, many properties of these nanocomposites will be improved, such as mechanical properties [7], thermal stability [8], and ﬂam retardance [9]. Most commonly used layered silicate is montmorillonite clay, which is composed of micron-sized particles. The particles are constructed of platelets with thickness of ∼1 nm and width of 100–200 nm. Platelets have permanent negative charge and they are held together by charge balancing cations such as Na+ or Ca [2+] ions. The signiﬁcant disruption of individual silicate layers in polymer matrix with nanoscopic dimensions (exfoliated structure) leads to improvements of the nanocomposite properties. However, in many cases, the isolated silicate layers are not completely dispersed throughout the polymer matrix, instead, the clay particles in polymer matrix maintain the hierarchical architecture, and an interlayer expansion occurs (intercalated structure). Polymer clay nanocomposites are prepared by three methods: solution blending, melting blending, and in-situ polymerization [10–20]. Solution blending is the process of mixing of clay (or modiﬁed clay) with a polymer solution. The structure of the resulting nanocomposite is related to the interaction between polymer chains and the clay surface, the solution concentration, and the nature of solvent [21–25]. In a melting blending process, a mixture of polymer and clay is heated above the glass-transition temperature (Tg) or melting temperature of the polymer, and mixed in a extruder or a blender [18, 26]. In both solution blending and melting blending processes, polymer chains penetrate into galleries of clay layers, and an exfoliated or intercalated structure is produced. In the in-situ polymerization technique, the cations in the intergallery space of the clay are exchanged to organic ammonium salts (monomers or initiators), which leads to the expansion of the gap between the clay platelets. The in-situ polymerizations are conducted between the intercalated layers. Various different polymerization technologies were employed for preparing polymer clay In-situ Synthesis of Polymer Nanocomposites, First Edition. Edited by Vikas Mittal. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA.

268

10 Polymer Clay Nanocomposites by In-situ Atom Transfer Radical Polymerization

nanocomposites. Surfactants with double bonds or free radical initiator were tethered to the clay layers by ion exchange and polymer clay nanocomposites were synthesized by in-situ free radical polymerization [27–33]. Polylactide, poly(ε-caprolactone) and epoxy/clay nanocomposites were successfully synthesized via ring-opening polymerization (ROP) [14, 34–37]. From a surface-bound 1,1-diphenylethylene initiator, PS/clay composites were synthesized by living anionic polymerization [38, 39]. Because living/controlled radical polymerization (LRP) allows better control over architectures and compositions of the targeted polymers, a variety of LRP methods, including atom transfer radical polymerization (ATRP) [20, 34, 40–45], nitroxide-mediated polymerization (NMP) [46, 47] and the reversible addition-fragmentation chain transfer (RAFT) polymerization method [48, 49] were used in the preparation of polymer clay nanocomposites. In this chapter, we focus our topic on preparation of polymer clay nanocomposites by ATRP. ATRP is based on a fast dynamic equilibrium established between the dormant species (alkyl halides or polymer chains with halogen atoms at their chain ends) and active radicals. The transition metal complexes act as reversible halogen transfer agents that keep the active radical concentration very low so as to avoid radical termination. Because of the low radical concentration, the polymerization rate of ATRP is much slower than the conventional radical polymerization. Datta and coworkers found a remarkable enhancement in the rate of ATRP of ethyl acrylate in the presence of organically modiﬁed clay [50]. Time of dispersion of clay in monomer prior to polymerization and the extent of clay loading were found to have positive effects on the polymerization rate. The polymerization proceeded through ﬁrst-order kinetics and molecular weights increased linearly with conversion, both of which were close to the theoretical values. They proved that there is a deﬁnite interaction between the hydroxyl groups of the clay with that of the carbonyl moiety of the dormant species, thereby activating the C–Br bond next to the ester carbonyl bond generating a higher concentration of active radicals. ATRP is a powerful tool in the preparation of clay polymer nanocomposites due to the control of molecular weights and molecular architectures. The ﬁrst step in the preparation of polymer clay nanocomposites by ATRP is grafting of ATRP initiator to the platelets by ion exchange. The modiﬁcation with ATRP initiator typically turns the hydrophilic clay layers more hydrophobic, and makes them dispersed well in organic solvents. Böttcher and coworkers were the ﬁrst to report the preparation of polymer clay nanocomposites by ATRP [20]. ATRP initiator-modiﬁed layered silicate was prepared by ion exchange between clay and 1,1′-(N,N,N-trimethylammonium bromide)undecanyl-2-bromoisobutyrate; in-situ ATRP was carried out to grow polymer chains from the clay surface. In-situ ATRP between individual silicate layers leads to the direct synthesis of dispersed silicate nanocomposites. Their kinetics study and molecular weight analysis demonstrated that the polymerization followed the ATRP mechanism. Zhao et al. also synthesized polymer clay nanocomposites by ATRP [41]. An ATRP initiator, consisting of a quaternary ammonium salt moiety and a

10 Polymer Clay Nanocomposites by In-situ Atom Transfer Radical Polymerization Polymer-layered silicate nanocomposite

MMT-2 x

x x

x

x x

x

x x x

x x

x x

x

x

x

x

x

x

x

x

ATRP

x

x x

⊕ N

O 11 O

Br

Figure 10.1 Schematic for the preparation of polymer-layered silicate by in-situ ATRP

(reproduced with permission from J Polym Sci Part A: Polym Chem 2004, 42, 916).

2-bromo-2-methyl propionate moiety, was intercalated into the interlayer spacing of clay. ATRP of styrene (St), methyl methacrylate (MMA), or n-butyl acrylate (nBA) with Cu(I)X/N,N-bis(2-pyridiylmethyl) octadecylamine (BPMODA), Cu(I)X/N,N,N′,N′,N′′-pentamethyldiethylenetriamine, or Cu(I)X/1,1,4,7,10,10hexamethyltriethylenetetramine (X = Br or Cl) different catalyst systems were conducted in the presence of initiator-modiﬁed clay (Figure 10.1). Homopolymers with low polydispersities were obtained. It was noted that due to the effect of the ammonium end groups, GPC results underestimated the molecular weights of PS at low monomer conversions. Their research indicated that the PS nanocomposites contained both intercalated and exfoliated structures, whereas the PMMA nanocomposites were signiﬁcantly exfoliated. The inﬂuence of graft density on kinetics of surface-initiated ATRP of St from clay was studied by Behling and coworkers [51]. 11′-(N,N,N-trimethylammonium bromide)-undecyl-2-bromo-2-methyl propionate (BMP) and 11′-(N,N,Ntrimethylammonium bromide)-undecyl-2,2-dimethyl propionate were mixed and ion exchanged to produce modiﬁed clay with different graft densities of the ATRP initiator. The polymerization rate for PS grafted from 100%-MMT (about 1 ATRP initiator/nm2) is nearly an order of magnitude greater than unbound PS produced under analogous conditions. The polymerization rate for PS grafted from 33%MMT and 67%-MMT are accelerated relative to the polymerization initiated by the free ATRP initiator. They believed that local concentration heterogeneities shift the ATRP equilibrium in favor of the active state. Zhao and Shipp synthesized poly(styrene-block-butyl acrylate) (PSBA) by sequential ATRP. BMP was intercalated into the individual layers of clay by ion exchange, and Cu(I)Br complexed by BPMODA was used as the transition metal catalyst. After two-step ATRP, PSBA polymer brushes on the clay surface were prepared (Figure 10.2). Their transmission electron microscopy (TEM) result indicated that a mixture of exfoliated and intercalated structures was produced in the nanocomposite. On the surface of clay layers, the block copolymer chains can form nanosized domains (2 to 5 nm), which is much smaller than the size of self-assembly

269

270

10 Polymer Clay Nanocomposites by In-situ Atom Transfer Radical Polymerization ⊕ Na

⊕ Na

⊕ Na

⊕ Na

d = 1.23 nm (Na-MMT)

⊕ Na

⊕ Na

O ⊕ N NaBr ⊕ N

Br

⊕ N

Br 11

O

O N 11

Br

Br

⊕

d = 1.96 nm (MMT-A)

Br

Br 11

O

O

11

O

O

O

⊕ N

O

11

Br

O O

Br

O

O 11N

⊕

O 11N

⊕

1. Styrene + CuBr + BPMODA ATRP

2. Butyl acrylate PBA block

PS block

Figure 10.2 Schematic for the preparation of PSBA block copolymer brushes on the surface of

clay platelets by two-step ATRP (reproduced with permission from Chem Mater 2003, 15, 2693).

of pure block copolymer. The reduced size is attributed to the immobilization of block copolymer chains on the surface of clay platelets. Just like block copolymer brushes on the surface of the silicon wafer [52–54], the block copolymer brushes on the surface of exfoliated or intercalated clay layers were also found to create nanopatterns after treatment in different solvents [42]. For the block copolymner brushes after treatment in tetrahydrofuran (THF), a good solvent for the two blocks, uniform collapsed brush layers were observed (Figure 10.3a). After treatment in acetone, a selective solvent for PBA and a precipitant for PS, wormlike surface aggregates were observed (Figure 10.3b). After treatment in methanol, a precipitant for both of the blocks, micelles as well as wormlike aggregates were observed (Figure 10.4). Following the similar method, Yang and coworkers synthesized amphiphilic poly(methyl methacrylate-block-(dimethylamino)ethyl methacrylate) (PMMA-b-

10 Polymer Clay Nanocomposites by In-situ Atom Transfer Radical Polymerization a)

b)

50 nm

50 nm

Figure 10.3 TEM images of poly(styrene-block-butyl acrylate) (PSBA) block copolymer brushes

on the surface of intercalated clay layers prepared from (a) THF solution and (b) acetone solution (reproduced with permission from Polymer 2004, 45, 4473).

a)

b)

50 nm

50 nm

Figure 10.4 TEM images of nanopatterns of PSBA brushes prepared from methanol. PSBA

brushes on the surface of (a) exfoliated and (b) intercalated clay layers (reproduced with permission from Polymer 2004, 45, 4473).

PDMAEMA) block copolymer brushes on the surface of clay layers [43]. Both exfoliated and intercalated structure were found in the nanocomposites. They also found that the block copolymer brushes form the lamella structure on the clay surface after treatment in THF, and after being treated in water surface micelles and wormlike structures were observed. Because of the amphiphilicity of the polymer brushes, the clay layers with polymer brushes could be used as stabilizers

271

272

10 Polymer Clay Nanocomposites by In-situ Atom Transfer Radical Polymerization water

PDMAEMA

PMMA

MMA

polymerization MMA droplets

water

PMMA

Figure 10.5 Schematic representation for the stabilization of MMA droplets and PMMA

colloidal particles by clay layers (reproduced with permission from Langmuir 2007, 23, 2867).

b) ATRP Ethyl Acrylate, CuBr, PMDETA, MBrP 90°C

a) Esterification Et3N, dry THF

Surface initiated polyacrylate chains

Free polymer

Figure 10.6 Schematic representation for the attachment of ATRP initiator onto clay surface

via esteriﬁcation and the subsequent polymerization of ethyl acrylate by using surface-initiated ATRP (reproduced with permission from J Polym Sci Part A: Polym Chem 2008, 46, 5014).

in suspension polymerization of MMA, and PMMA colloidal particles with clay layers on the surface were prepared (Figure 10.5). Datta and coworkers prepared polymer clay nanocomposites based on cloisite 30 B, an organically modiﬁed clay with pendant bis-2-hydroxyethyl group on the modiﬁer (Figure 10.6) [44]. The ATRP initiator was grafted covalently on to the surface of clay platelets as well as at the hydroxyl-terminated edges of the modiﬁer

10 Polymer Clay Nanocomposites by In-situ Atom Transfer Radical Polymerization ⊕

⊕

Na

⊕

Na

⊕

Na

⊕

Na

⊕

Na

Na O

⊕

NaBr

N

CH2

CH2 10

(R)

O

Br

Br ⊕

⊕

R

R ⊕ R

⊕

⊕

⊕

R

⊕

R

⊕

R

R ⊕ R

R

H3C O

OH

ATRP of HEMA

O PHEMA O

O

CH3

H3C

O

O

Ring-opening polymerization of LLA PHEMA backbone

PLLA comb

Figure 10.7 Schematic representation for the preparation of PLLA comb polymer brushes on

the surface of clay layers (reproduced with permission from Polymer 2006, 47, 7374).

and used to initiate the polymerization of ethyl acrylate. The polymerization took place within the clay gallery as well as on the outer surface of the clay platelets. This method allows for a relatively high surface density of the ATRP initiator and polymer chains. Extensive exfoliation was achieved by using this method. ATRP was also combined with other polymerization methods in the preparation of polymer clay nanocomposites. Yang and coworkers reported the preparation of poly(L-lactide) (PLLA) comb polymer brushes on the surface of silicate layers by using a combination of in-situ ATRP and ROP (Figure 10.7) [34]. An ATRP initiator with a quaternary ammonium salt end group was intercalated into the interlayer spacing of clay layers via ion exchange. In-situ ATRP of hydroxyethyl methacrylate (HEMA) in the presence of CuBr and bipyridine afforded the PHEMA backbone. PLLA comb polymers were grown from the PHEMA backbone by ROP. Because

273

274

10 Polymer Clay Nanocomposites by In-situ Atom Transfer Radical Polymerization O Br

+

O

+

6

N

Cl

Na +

+

Na

Na

+

Br

O O

– NaCl

6

1.62 nm

N 6

Br

O

O

O

O Br

N+

+

N

6

(Na-MMT) (O-MMT) ATRP

2.27 nm

St

Br Br

Br Br

Br Br Br

(CROP) AgPF6 THF

Br Br

Br

Br Br Br Br Br

(PSt-b-PTHF/MMT)

(PSt/MMT)

Figure 10.8 Schematic representation for synthesis of PS-b-PTHF clay nanocomposites by

combination of ATRP and cationic ROP (reproduced with permission from J Polym Sci Part A: Polym Chem 2009, 47, 2190).

of the special structure of PLLA comb polymers on the surface of clay, the polymer brushes have strong capability to create the exfoliated structure. This method provides a straightforward approach to the preparation of clay/biodegradable polymer nanocomposites for biological purposes. Yagci and coworkers synthesized poly(styrene-block-tetrahydrofuran) (PS-bPTHF) block copolymer brushes on the surface of intercalated and exfoliated clay layers by mechanistic transformation [55]. First, the PS block was synthesized by ATRP initiated by initiator moieties immobilized within the silicate galleries. PS blocks synthesized by ATRP contain terminal halide groups, which can be converted to the initiating carbocations when reacted with silver salts containing non-nucleophilic counter anions. In the second step PS-b-PTHF block copolymer was prepared by mechanistic transformation from ATRP to cationic ROP (Figure 10.8). This research provides a new method to synthesize polymer clay nanocomposite containing soft and hard polymer segments. Besides ion exchanging, another method to introduce the ATRP initiator to the platelets is silylation reaction of the silanol groups, which are assumed to be on the edge of the clay sheets [56], followed by further chemical reactions. Organochlorosilanes can be reacted with hydroxyl groups on the external and internal surfaces of clay. By reactions of silanol groups with coupling agents, a variety of

10 Polymer Clay Nanocomposites by In-situ Atom Transfer Radical Polymerization

functional groups could be attached. Mathias and coworkers reported synthesis of covalently functionalized laponite clay synthesized through a condensation reaction of the silanol groups with mono- and trifunctional alkoxy silanes [57]. In their research aminopropyltrimethoxy silane, aminopropyldimethylethoxysilane and trimethoxypropylsilane were used in the condensation reaction with clay. Their experimental results indicated that trialkoxysilanes can link clay sheets together, reducing subsequent surface ion exchange capacity. Monoalkoxysilanes do not exhibit this behavior. In another report, Mathias and coworkers described modiﬁcation of clay with combinations of organic ammonium surfactant and/or covalently bound PMMA [58]. A compound terminated on one side with an ethoxysilane and the other side by a methacrylate group was synthesized. After silylation reaction, methacrylate group was introduced to the clay layers. A chlorosilane-terminated ATRP initiator was synthesized and linked to the clay via a reaction. Two polymer attachment methods were used in their experiments, one through reaction of a methacrylate compound with the clay’s silanol group followed by in-situ free radical polymerization of MMA, and the other through attachment of an ATRP initiator followed by in-situ ATRP. The overall scheme was shown in Figure 10.9. The free radical method yielded clays with about 75 wt% of polymer bound through multiple attachment sites to the clay, whereas the ATRP method yielded about 68 wt% of bound polymer attached only at the chain end. The polymer-modiﬁed clay can be dispersed well in organic solvents and be blended with commercial PMMA to prepare polymer blends. A mixture of intercalated and exfoliated dispersions was observed in the nanocomposites.

EtO

OH

H N

Si

O

O

O or Cl

O O

Si +

Br O

Sodium Laponite

CTAB (lon Exchange) +

Na

+

Na+ Na Na+

+

Na

Na+ Na+

MMA Polymerize

R

e riz

me

ly Po

Na+

R

A

MM

Na+ Na+ Na+

R R

R

R R = methacrylate or tertiary bromine

Figure 10.9 Schematic showing the process of covalently attaching polymer to clay by

free radical polymerization and ATRP (reproduced with permission from Chem Mater 2006, 18, 3937).

275

276

10 Polymer Clay Nanocomposites by In-situ Atom Transfer Radical Polymerization a)

b)

0.2 μm

0.2 μm c)

Figure 10.10 TEM images of clay–PS dispersed in water (a) and toluene (b). (c) A schematic representation for the structure of clay–PS particles dispersed in toluene and water (reproduced with permission from J Polym Sci Part A: Polym Chem 2009, 47, 1535).

Wu and coworkers prepared hydrophobic PS brushes on the edges of clay sheets by using edge modiﬁcation and in-situ ATRP [59]. The amino groups were attached to the edges of clay minerals by condensation reaction of silanol groups with trifunctional 3-(triethoxysilyl)-propylamine, and the reaction leads to the formation of polycondensates composed of Si–O–Si covalent bonds in the interlaminate space. ATRP initiators on clay edges were synthesized via a reaction between amino groups and 2-bromo-2-methyl propionyl bromide. PS brushes were prepared by in-situ ATRP. Owing to the hydrophilic clay surface and hydrophobic PS brushes, the clay–PS particles can self-assemble in water and toluene. Figure 10.10a is a TEM image of clay–PS dispersed in water, where the aggregation of clay particles could be observed. In the system, there exist repulsive interaction between PS and water and hydration of interlater cations on the surface of clay layers. To minimize exposure to water, PS chains on clay–PS particles collapse

10 Polymer Clay Nanocomposites by In-situ Atom Transfer Radical Polymerization a)

0.5 μm

b)

0.2 μm

Figure 10.11 (a) TEM image of PS colloidal particles stabilized by clay-PS and (b) a magniﬁed TEM image of PS colloidal particles (reproduced with permission from J Polym Sci Part A: Polym Chem 2009, 47, 1535).

forming nanometer-sized domains and the clay layers stay in the aqueous phase (Figure 10.10c). Figure 10.10b is a TEM image of clay–PS dispersed in toluene. The stacked clay particles can be observed in the image. Toluene is a good solvent for PS but it has unfavorable interaction with the hydrophilic clay surface, so the clay–PS particles tend to pile up together forming the face-to-face structure to reduce the unfavorable interaction, and PS chains stretch out into the toluene phase (Figure 10.10c). They also prepared PS colloidal particles by using clay–PS as the stabilizer [59]. Narrowly dispersed PS colloidal particles were obtained (Figure 10.11). The PS brushes on the edge of clay particles penetrate into colloidal particles and negatively charged clay face stay in the aqueous phase so that the interfacial tension between colloidal particles and water could be reduced, and colloidal particles were stabilized. Karesoja and coworkers reported grafting of clay with butyl acrylate and MMA by ATRP [60]. An ATRP initiator with trichlorosilane functionality was synthesized reacted with the hydroxyl groups on clay surfaces and initiated the copolymerization of BA and MMA. The three-step reaction was schemed in Figure 10.12. The 11-(2-bromo-2-methyl)propionyl-oxy-undecyl trichlorosilane ATRP initiator was covalently attached to clay platelets via silylation reactions. The initiator clay was used to polymerize BuA and MMA in the bulk monomer solution or dimethyl sulfoxide (DMSO). They found in the nanocomposites prepared from DMSO clay was well dispersed as individual platelets; however, in the nanocomposites prepared from bulk solution big aggregates of clay platelets were observed. This indicates that DMSO is able to expand the distance between the clay layers and make the clay gallery more accessible to chemicals. Li and coworkers chose hydrothermally synthesized Na magadiite (Na2Si14O29. nH2O) as the source of layered silicate [61]. Upon acidiﬁcation, Na-magadiite is

277

278

10 Polymer Clay Nanocomposites by In-situ Atom Transfer Radical Polymerization

Step 1. (OH) + (CH2)9

CH2

O

CH3 Br

Br

CH2

CH3

O

O

Pyridine, THF

Br

(CH2)9

CH3

CH3

Step 2. O CH2

O

Cl

CH3 Br

(CH2)9

Karstecit Catalyst HSiCl3

CH3

O

Si

Cl

O

CH3 Br

(CH2)11

CH3

Cl

Step 3. Cl Cl

O

Si

O

CH3 Br

(CH2)11

CH3

Cl

Cl

O

Triethyl amine

Si

Montmorillonite particles

Cl

H2C CH C O O CH3 H2C C CH2 C O CH2 O CH2 CH3 CH3 CH2 Br CuBr, 2,2-Bipyridyl CH3

O

O

(CH2)11

CH3 Br CH3

Polymerization of BuA and MMA

Cl

O Si Cl

O

(CH2)11

O

CH3 R CH2 C n O

O R=

Si

CH2 CH O O

m

Br O

O CH3

(CH2)11 CH3

Figure 10.12 Scheme for preparation of polymer modiﬁed clay by ATRP (reproduced with permission from J Polym Sci Part A: Polym Chem 2009, 47, 3086).

converted into H-magadiite and the dangling surface OH groups on the surface of H-magadiite are able to form covalent bonds with surfactants containing amine groups. 2-Bromopropionyl bromide was added to the mixture to react with the NH2 groups of the surfactant. PS brushes were prepared on the surface of clay layers via in-situ ATRP (Figure 10.13). The resulting nanocomposites had an intercalated and partially exfoliated structure. Synthesis of polymer-layered silicate nanocomposites by ATRP has attracted much attention due to the improved properties of the nanocomposites and many

References OEt

Me

Me Me

Si δ+ H 2N

δ+ NH2

Si

H

H δ–O

Br Br

Br HN

O

Br

H2N

Me OEt

– Oδ

OH

O

279

δ+ H2N

O δ–

H2N

R Me-Si-Me O

Me-Si-Me O

O

HN ATRP

Me-Si-Me O

Me-Si-Me O

Figure 10.13 Synthetic procedure for Br-magadiite/polystyrene nanocomposites from initiator-modiﬁed magadiite via ATRP (reproduced with permission from J Polym Sci Part A: Polym Chem 2005, 43, 534).

possible applications. In most cases, layered silicates ﬁrst need to be modiﬁed with the ATRP initiator via ion-exchange or chemical reactions, and ATRP initiator is introduced to the surfaces or edges of clay layers. Polymer chains on clay layers are able to be prepared via in-situ ATRP. Similar to block copolymer brushes on silicon wafer, the block copolymer brushes on the surface of clay layers can make different nanopatterns after treatment in different solvents. Polymer–clay nanocomposites can ﬁnd applications in polymer blends and Pickering emulsions, and new materials with different structures and properties can be prepared.

References 1 Usuki, A., Kawasumi, M., Kojima, Y.,

4 Kojima, Y., Usuki, A., Kawasumi, M.,

Okada, A., Kurauchi, T., and Kamigaito, O. (1993) J. Mater. Res., 8, 1174. 2 Usuki, A., Kojima, Y., Kawasumi, M., Okada, A., Fukushima, Y., Kurauchi, T., and Kamigaito, O. (1993) J. Mater. Res., 8, 1179. 3 Kojima, Y., Usuki, A., Kawasumi, M., Okada, A., Fukushima, Y., Kurauchi, T., and Kamigaito, O. (1993) J. Mater. Res., 8, 1185.

Okada, A., Karauchi, T., and Kamigaito, O. (1993) J. Polym. Sci., Part A: Polym. Chem., 31, 983. 5 Alexandre, M. and Dubois, P. (2000) Mater. Sci. Eng., R28, 1. 6 Zanetti, M., Lomakin, S., and Camino, G. (2000) Macromol. Mater. Eng., 279, 1. 7 Lan, T. and Pinnavaia, T.J. (1994) Chem. Mater., 6, 2216.

280

10 Polymer Clay Nanocomposites by In-situ Atom Transfer Radical Polymerization 8 Messersmith, P. and Giannelis, E.P. 9

10 11 12 13 14

15 16 17

18

19

20

21 22 23 24 25 26

27

28 29 30

(1994) Chem. Mater., 6, 1719. Zhu, J., Uhl, F.M., Morgan, A.B., and Wilkie, C.A. (2001) Chem. Mater., 13, 4649. Wang, D.Y., Zhu, J., Yao, Q., and Wilkie, C.A. (2002) Chem. Mater., 14, 3837. Pinnavaia, T.J. (1983) Science, 220, 365. Carrado, K.A. and Xu, L. (1998) Chem. Mater., 10, 1440. Giannelis, E.P. (1996) Adv. Mater., 8, 29. Messersmith, P. and Giannelis, E.P. (1995) J. Polym. Sci., Part A: Polym. Chem., 33, 1047. Rossi, G.B., Beaucage, G., Dang, T.D., and Vaia, R.A. (2002) Nano Lett., 2, 319. Huang, X. and Brittain, W.J. (2001) Macromolecules, 34, 3255. Weimer, M.W., Chen, H., Giannelis, E.P., and Sogah, D.Y. (1999) J. Am. Chem. Soc., 121, 1615. Tong, X., Zhao, H., Tang, T., Feng, Z., and Huang, B. (2002) J. Polym. Sci., Part A: Polym. Chem., 40, 1706. Zhu, J., Start, P., Mauritz, K.A., and Wilkie, C.A. (2002) J. Polym. Sci., Part A: Polym. Chem., 40, 1498. Böttcher, H., Hanllensleben, M.L., Nuss, S., Wurm, H., Bauer, J., and Behrens, P. (2002) J. Mater. Chem., 12, 1351. Vaia, R.A. and Giannelis, E.P. (1999) Macromolecules, 30, 7990. Vaia, R.A. and Giannelis, E.P. (1997) Macromolecules, 30, 8000. Balazs, A.C., Singh, C., and Zhulina, E. (1998) Macromolecules, 31, 8370. Ginzburg, V.V. and Balazs, A.C. (1999) Macromolecules, 32, 5681. Ginzburg, V.V., Singh, C., and Balazs, A.C. (2000) Macromolecules, 33, 1089. Furuichi, N., Kurokawa, Y., Fujita, K., and Oya, A. (1996) J. Mater. Sci., 31, 4307. Zhu, J., Morgan, A.B., Lamelas, F.J., and Wilkie, C.A. (2001) Chem. Mater., 13, 3774. Zeng, C. and Lee, L.J. (2001) Macromolecules, 34, 4098. Fu, X. and Qutubuddin, S. (2001) Polymer, 42, 807. Fan, X., Xia, C., Fulghum, T., Park, M., Locklin, J., and Advincula, R.C. (2003) Langmuir, 19, 916.

31 Zhang, J., Yang, Y., Zhao, C., and Zhao,

32

33

34 35

36 37 38

39

40 41

42 43 44

45 46

47 48

49 50 51

H. (2007) J. Polym. Sci., Part A: Polym. Chem., 45, 5329. Yang, Y., Zhang, J., Liu, L., Li, C., and Zhao, H. (2007) J. Polym. Sci., Part A: Polym. Chem., 45, 5759. Zhang, J., Chen, K., and Zhao, H. (2008) J. Polym. Sci., Part A: Polym. Chem., 46, 2632. Yang, Y., Wu, D., Li, C., Liu, L., Chen, X., and Zhao, H. (2006) Polymer, 47, 7374. Kubies, D., Pantoustier, N., Dubois, P., Rulmont, A., and Jerome, R. (2002) Macromolecules, 35, 3318. Lan, T., Kaviratna, P.D., and Pinnavaia, T.J. (1995) Chem. Mater., 7, 2144. Brown, J.M., Culiss, D., and Vaia, R.A. (2000) Chem. Mater., 12, 3376. Fan, X., Zhou, Q., Xia, C., Cristofoli, W., Mays, J., and Advincula, R. (2002) Langmuir, 18, 4511. Zhou, Q., Fan, X., Xia, C., Mays, J., and Advincula, R. (2001) Chem. Mater., 13, 2465. Zhao, H. and Shipp, D.A. (2003) Chem. Mater., 15, 2693. Zhao, H., Argoti, S.D., Farrell, B.P., and Shipp, D.A. (2004) J. Polym. Sci., Part A: Polym. Chem., 42, 916. Zhao, H., Farrell, B.P., and Shipp, D.A. (2004) Polymer, 45, 4473. Yang, Y., Liu, L., Zhang, J., Li, C., and Zhao, H. (2007) Langmuir, 23, 2867. Datta, H., Bhowmick, A.K., and Singha, N.K. (2008) J. Polym. Sci., Part A: Polym. Chem., 46, 5014. Datta, H., Singha, N.K., and Bhowmick, A.K. (2008) J. Appl. Polym. Sci., 108, 2398. Weimer, M.W., Chen, H., Glannelis, E.P., and Sogah, D.Y. (1999) J. Am. Chem. Soc., 121, 1615. Di, J. and Sogah, D.Y. (2006) Macromolecules, 39, 5052. Zhang, B., Pan, C., Hong, C., Luan, B., and Shi, P. (2006) Macromol. Rapid Commun., 27, 97. Salem, N. and Shipp, D.A. (2005) Polymer, 46, 8573. Datta, H., Singha, N.K., and Bhowmick, A.K. (2008) Macromolecules, 41, 50. Behling, R.E., Williams, B.A., Staade, B.L., Wolf, L.M., and Cochran, E.W. (2009) Macromolecules, 42, 1867.

References 52 Zhao, B. and Brittain, W.J. (2000) Prog.

58 Wheeler, P.A., Wang, J., and Mathias, L.J.

53

59 Wu, Y., Zhang, J., and Zhao, H. (2009)

54

55

56 57

Polym. Sci., 25, 677. Zhao, B. and Brittain, W.J. (2000) Macromolecules, 33, 8813. Zhao, B., Brittain, W.J., Zhou, W., and Cheng, S.Z.D. (2000) J. Am. Chem. Soc., 122, 2407. Yenice, Z., Tasdelen, M.A., Oral, A., Guler, C., and Yagci, Y. (2009) J. Polym. Sci., Part A: Polym. Chem., 47, 2190. Song, K. and Sandi, G. (2001) Clays Clay Miner., 49, 119. Wheeler, P.A., Wang, J., Baker, J., and Mathias, L.J. (2005) Chem. Mater., 17, 3012.

(2006) Chem. Mater., 18, 3937. J. Polym. Sci., Part A: Polym. Chem., 47, 1535. 60 Karesoja, M., Jokinen, H., Karjalainen, E., Pulkkinen, P., Torkkeli, M., Soininen, A., Ruokolainen, J., and Tenhu, H. (2009) J. Polym. Sci., Part A: Polym. Chem., 47, 3086. 61 Li, C.P., Huang, C.M., Hsieh, M.T., and Wei, K.W. (2005) J. Polym. Sci., Part A: Polym. Chem., 43, 534.

281

283

11 Polybutadiene Clay Nanocomposites by In-situ Polymerization Giuseppe Leone and Giovanni Ricci

11.1 Introduction

The second-half of the twentieth century, with the development of polymer composites based on micrometer-sized reinforcing particles, witnessed an enormous transformation in the chemical design, engineering, and performance of structural materials. Actually, the emergence of nanometer-sized particles (such as platelets, ﬁbers, and tubes) is leading to a second revolution in composite materials. Polymer nanocomposites (PNs) are currently the subject of extensive worldwide research. PNs are a class of composite materials that are two-phase hybrid systems consisting of polymers ﬁlled with high-surface-area reinforcing nanoparticles, for which at least one dimension of the dispersed particles is in the nanometer range. Three types of nanocomposites can be distinguished, depending on how many dimensions of the dispersed particles are in the order of nanometers. Thus, we deal with isodimensional nanoparticles such as spherical silica nanoparticles and semiconductor nanoclusters, when all the three dimensions are in the order of nanometers; with nanotubes or whiskers such as carbon nanotubes or cellulose whiskers if two dimensions are in the nanometer scale; with layered silicate (or clays), when only one dimension is in the nanometer range. In the last case, the ﬁller is present in the form of sheets with a thickness from one to a few nanometers, and a length from hundreds to thousands nanometers. Nowadays, ordered inorganic/organic PNs with a ﬁnely tuned structure have displaced a lot of traditional composite materials in a variety of applications because the intimate interactions between components can provide enhancement of the bulk polymer properties (i.e., mechanical and barrier properties, thermal stability, ﬂame retardancy, and abrasion resistance). The reinforcing nanoparticle/ polymer adhesion is of primarily importance, as it tunes the ﬁnal properties of the nanocomposite. Polymer/clay nanocomposites (PCNs) meet this demand due to the platelet-type dispersion of the clay ﬁller in the organic matrix [1]. Since PCNs are used as structural materials, layered silicates (e.g., dioctahedral 2 : 1 phyllosilicate clays) have attracted academic and industrial attention. The

In-situ Synthesis of Polymer Nanocomposites, First Edition. Edited by Vikas Mittal. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA.

284

11 Polybutadiene Clay Nanocomposites by In-situ Polymerization

interest is also due to their abundance, relatively cheapness, and because they exhibit a large contact area. Research involving clays, as reinforcing agents for polymers, has mostly been devoted to polar polymer (i.e., polyamide and polyester) based nanocomposites, in which silicates were easily exfoliated by water-assisted extrusion or by simple chemistry processes [2]. Hydrophobic polymeric matrices require a much more complex technique to afford a high clay dispersion degree, and their nanocomposites can be prepared by incorporation of the silicate into a solvent-swollen polymer (solution blending), latex compounding, and direct intercalation of the molten polymer (melt blending). Among the various approaches for the preparation of hydrophobic PCNs, new in-situ techniques involve the direct addition of clay intercalated by a catalyst or an initiator to the monomer mixture [3, 4]. Researches dealing with the in-situ synthesis of PCNs mostly focused on α-oleﬁns- [5–8], styrene- [9], and methylmethacrylate-based polymers [10]. In contrast, only few papers on the in-situ preparation of butadiene-based PNs have been reported in the literature, being melt and latex compounding the most investigated methods [11]. This chapter aims to give an overview on the recent advances in the synthesis of PCNs through the in-situ 1,3-butadiene homo- and copolymerization technique. To this purpose, we distinguished in-situ polymerization approaches on the basis of the polymerization method: anionic or insertion/coordinative. However, before discussing on the topic of this chapter, we wish to brieﬂy recall some peculiar aspects of clay minerals, PNs, and their methods of preparation.

11.2 Generalities 11.2.1 Clays

The layered inorganic particles of interest for PN technology have platelets from ca. 0.7–2.5 nm thick [12]. A partial and nonexhaustive list of candidates is given in Table 11.1. The silicates commonly used in nanocomposites are essentially aluminosilicate (2 : 1 phyllosilicate clays), constituted of stacks of hydrated layers whose crystal structure is composed of an octahedral alumina Al(O,OH)6 sheet (O-network) sandwiched between two silicon–oxygen tetrahedral sheets (T-network). (Scheme 11.1) The layers are separated by galleries where cations (e.g., Na, K) balance the negative charge of the aluminosilicate sheets arising from the isomorphic substitution of Al or Si with other metals [13]. The layer thickness is around 1 nm and the lateral dimensions of these layers may vary from 30 nm to several microns. These layers organize themselves to form stacks with a regular van der Walls gap in between them, named the interlayer region. Isomorphic substitution within the layers (e.g., Al3+ replaced by Mg2+ or by Fe2+; Mg2+ replaced by Li+) generates negative charges that are counterbalanced by alkali

11.2 Generalities Table 11.1

Layered nanoparticles for the potential use in polymer nanocomposites.

Chemical nature

Examples

Smectite clays

Montmorillonite (MMT), bentonite (BT), nontronite, beidellite, volkonskoite, hectorite (HT), saponite, sepiolite, stevensite, etc

Synthetic clays

Hectorite, MgO(SiO2)s(Al2O3)a(AB)b(H2O)x, (where AB is a ion pair, namely NaF)

Layered silicic acids

Kanemite, makatite, octosilicate, magadiite, kenyaite, and layered organo-silicates

Other clays

Micas, vermiculite, illite, ledikite, and tubular attapulgite

Metal chalcogenides

TiS2, MoS2, MoS3, (PbS)1.18(TiS2)2

Layered double hydroxides

M6Al2(OH)16CO3·nH2O; M = Mg, Zn

Others

Graphite, graphite oxide, etc.

Scheme 11.1

Schematic structure of a sodium-exchanged clay mineral.

or alkaline earth cations situated in between the lamellae. Based on the extent of the substitutions, a term called layer charge density is deﬁned. The layer surface has 0.25 to 1.2 negative charges per unit cell and a commensurate number of exchangeable cations in the interlamellar galleries. For the anionic clays (e.g., smectites) the ion concentration is usually expressed as the cation exchange capacity (CEC), which ranges from about 0.5 to 2 meq g−1. The distance from the Tnetwork to its analog in one of the neighboring layers is deﬁned as interlayer d-spacing. This spacing mainly depends on the size of the exchangeable cations and the amount of interlayer water. The intercalation of small molecules in between the layers is easy, being the forces holding the stacks together relatively weak [14]. In order to make these hydrophilic phyllosilicates more organophilic, the hydrated cations of the interlayer can be exchanged with cationic surfactants, such as alkylammonium or alkylphosphonium salts, to give organo-modiﬁed clay organoclay (OC). When the alkali cations are exchanged with organic cations, a

285

286

11 Polybutadiene Clay Nanocomposites by In-situ Polymerization

larger d-spacing usually results. Detailed studies on this topic have been extensively reviewed elsewhere [15, 16]. 11.2.2 Polymer Nanocomposite Structures

Depending on the nature of the components used (layered silicate, organic cation, and polymer matrix) and the preparation method, three types of hybrid PCNs can be obtained [17]. Phase-separated microcomposites (conventional composites) are obtained when the polymer chains are unable to intercalate within the inorganic sheets; clay lamellae remain stacked in structures marked as tactoids as in the pristine mineral. Otherwise, when the polymer chains penetrate in between the clay galleries, an intercalative system is obtained. In this case, the nanocomposite shows, at least in principle, a well-ordered multilayer morphology built up with alternating polymeric and clay layers. When clay platelets are randomly dispersed in the polymer matrix and the lamellae are far apart from each other, so that the periodicity of this platelet arrangement is totally lost, an exfoliated structure is achieved. PCNs are usually characterized by means of X-ray diffraction (XRD), transmission electron microscopy (TEM), and scanning electron microscopy (SEM). These tools indeed provide information about the nanocomposite structures and morphologies. Speciﬁcally, XRD is used to identify intercalated structures; in such nanocomposites, the repetitive multilayer structure is well preserved, allowing the interlayer spacing to be determined. The intercalation of the polymer chains increases the interlayer spacing, in comparison with the spacing of the pristine clay used, leading to a shift of the diffraction peak toward lower angle values (angle and layer spacing values being related through Bragg’s equation 2dhkl sin θ = λ, where λ corresponds to the wavelength of the X-ray radiation used in the diffraction experiment, d the spacing between diffractional lattice planes, and θ is the measured diffraction angle). In the case of the exfoliated structure, no more (001) diffraction peaks are detectable by XRD, either because of a too much large spacing between the layers (i.e., the polymer separates the clay platelets by 8–10 nm or more) or because the nanocomposites do not present ordering anymore. Moreover, it must be pointed out that very often the lost of any XRD signal of the inorganic component may be due to the lower clay amount (≤1 wt%) in the nanocomposite. Hence, for a better check of the nanocomposite structure, TEM investigation is the powerful analytical method allowing a qualitative understanding of the internal structure and directly providing information on morphology and defect structures [18]. However, the classiﬁcation of the composite structures as exfoliated or intercalated is not very realistic, since a mixture of different morphologies can exist presenting both intercalation and exfoliation features. In this case, a broadening of the XRD diffraction peak is observed, and the TEM observation too does not allow to assign the exact structure to the investigated nanocomposites; only a qualitative classiﬁcation of the morphology as more or less intercalated or exfoliated can be made.

11.3 Polybutadiene Nanocomposites

11.2.3 Methods of Preparation of Polymer Nanocomposites

PCNs are currently prepared in different ways: (i) clay/polymer matrix mixing in the molten state (melt intercalation), (ii) intercalation of polymer or prepolymer (in case of insoluble polymers) from solution, (iii) latex compounding, (iv) layer-by-layer assembly, and (v) in-situ intercalative polymerization [19–21]. Among them, one of the most intriguing approaches is the direct formation by in-situ polymerization. Unlike the latex and melt compounding, which are conducted directly with the polymer, the in-situ method involves the direct addition of the clay, intercalated by a catalyst or an initiator, to the polymerization reaction. According to this method, the intimate mixing of the “soft” polymer matrix with the “hard” ﬁller is promoted by the clay-mediated monomer polymerization. This technique is of special interest for the preparation of exfoliated α-oleﬁn (i.e., ethylene, propylene) and conjugated dienes (i.e., butadiene, isoprene) PNs, as the direct dispersion of silicate layers in the molten state of polyoleﬁns, for example, is difﬁcult to achieve because polyoleﬁns do not contain any polar group. Nevertheless, as most of butadiene-based polymers, which are the subject of this chapter, are available in solid and latex forms, latex and melt blending are believed to be the most feasible industrial methods for preparing polybutadiene (PB) nanocomposites. Only recently anionic in-situ polymerization using butyllithium initiator has been developed to achieve a homogeneous clay dispersion in the butadiene-based rubbers [22–27]. Further in-situ butadiene polymerization methods for the preparation of PB nanocomposites have been recently practiced by our research group [28, 29]. The synthesis has been achieved by immobilizing a butadiene polymerization catalyst onto the silicate particle, followed by the incorporation of monomer molecules into the galleries of these host materials, and successive polymerization by Ziegler–Natta coordination catalysis.

11.3 Polybutadiene Nanocomposites 11.3.1 1,3-Butadiene Polymerization Methods

The polymerization of 1,3-dienes (e.g., 1,3-butadiene and isoprene) with Ziegler– Natta catalysts began in 1954, soon after the ﬁrst results obtained in α-oleﬁn polymerization; since then many transition metal and lanthanide catalysts have been examined and several stereoregular diene polymers have been obtained [30, 31]. 1,3-Dienes can generate several types of polymers having different structures; trans-1,4; cis-1,4; 1,2 and, in the case of asymmetric monomers (e.g., isoprene), 3,4. Stereoregular 1,2- or 3,4-polydienes may also exhibit iso- or syndiotacticity. (Figure 11.1).

287

288

11 Polybutadiene Clay Nanocomposites by In-situ Polymerization

Figure 11.1 Polymer microstructures of cis-1,4 polybutadiene, trans-1,4 polybutadiene, 3,4

polyisoprene, 1,2-polybutadiene (syndio- and isotactic).

1,3-Dienes can be polymerized in different ways, by radical, cationic, anionic, and stereospeciﬁc polymerization. Radical initiators give polymers that are amorphous by X-ray at room temperature due to the different types of monomeric units along the polymer chain. Cationic catalysts give irregular crosslinked polymers. To date, the manufacture of butadiene rubber (BR) is carried out mainly in the solution process by anionic polymerization or Ziegler–Natta coordinative polymerization. Solution polymerization using the anionic initiator is usually based on lithium alkyls. Butyllithium gives a PB consisting of more than 90% 1,2 units in ethers, and about 90–93% 1,4 in hydrocarbon solvent [32]. Both these polymers are amorphous due to the fact that the latter one is constituted by a mixture of cis/trans units, while the former one lacks conﬁgurational order at the tertiary carbon atom. The stereospeciﬁc polymerization, differently from the anionic methods, is characterized by (i) high chemoselectivity, that is, it can generate polymers having only one type of monomeric unit (1,4; 1,2; or 3,4), and (ii) high stereoselectivity, that is, it can generate polymers with a very high conﬁgurational order when steric isomerism sites are present on the monomeric unit (e.g., an internal double bond, an asymmetric carbon). Beside styrene butadiene rubber (SBR), BR is the most important synthetic rubber. BR accounts for an annual consumption of ca. 2.8 million metric tons. In terms of annual production SBR and BR are only outnumbered by natural rubber (NR) with a production of ca. 6.7 million metric tons a year [33]. BR is used in four major areas. By far the largest portion is applied in tires (∼70%), especially tire treads and side walls. The second biggest use of BR is for thermoplastic polymers modiﬁcation (∼25%): the two main products are high impact polystyrene (HIPS) and acrylonitrile-butadiene-styrene (ABS) terpolymer. BR is also used to a much smaller extent in technical rubber goods (∼4%), such as conveyor belts, hose roll covers, shoe soles, and seals. The smallest application area of BR is in golf ball cores (∼1%). Since the start-up of industrial Ziegler–Natta-BR production in the 1960s, BR has continuously grown, mainly due to the general expansion of tire, HIPS, and ABS production. Commercial BR is comprised of a broad range of

11.3 Polybutadiene Nanocomposites

different BR grades [34]. These grades differ in microstructure, as shown in Figure 11.1, molecular weight, molecular weight distribution, and branching degree. Commercially available BR grades can be classiﬁed according to the type of polymerization technology and initiators/catalysts used. Some examples are reported below:

• • •

emulsion-BR, radical polymerization in aqueous emulsion; lithium-BR, anionic polymerization in solution; cobalt-, titanium-, nickel-, neodymium-BR, stereospeciﬁc polymerization in the solution.

Vulcanized rubber compounds are usually reinforced by inorganic ﬁller and/or carbon black to improve the mechanical properties, the thermal stability and the gas barrier properties of the bulky polymer. Carbon black is the most suitable reinforcing agent in the rubber industry thanks to the strong interaction with the polymer matrix. However, due to its polluting nature, the ubiquitous black color of the compounded material and its dependence on petroleum stock pushed the academic and industrially research to look out for other so-called white ﬁller. In this sense, clays seem to be a good choice to replace conventional ﬁller increasing the bulky PB performances. Despite the importance of the cis-1,4 PB, 1,2 PB is also an industrially produced polymer which is commonly used in the production of tires, ﬁlms (packaging breathing items for fruits, vegetables, and seafood), footwear soles, tubes, and hoses. The following paragraphs highlight the work concerning the synthesis of 1,3-butadiene based PCNs by in-situ, anionic and spereospeciﬁc, polymerization. 11.3.2 In-situ Anionic Polymerization

Enriched cis-1,4 PB was ﬁrst prepared by Firestone Tire and Rubber Company in 1955 [35] through anionic polymerization based on butyllithium initiator. The anionic polymerization reaction runs without polymer chain-termination, allowing the synthesis of “living polymers.” By using clay as carrier for the anionic initiator, some difﬁculties may arise since the interlayer clay water can terminate the living polymer chain growth. To overcome this disadvantage an excess of nBuLi is usually used as impurities scavenger. Zhang and coworkers [27] reported the synthesis of SBR/clay nanocomposites by anionic polymerization, as shown in Scheme 11.2. The clay used is an OC modiﬁed by the intercalation of a quaternary long ammonium salt. The ﬁller is stirred into a 5 l polymerization kettle ﬁlled with both styrene and butadiene monomer, THF, and cyclohexane. After stirring for 3 h, n-BuLi (THF/n-BuLi = 25) was added and the polymerization was carried out at 50 °C for 3 h. The authors found that the addition of the inorganic ﬁller to the reaction mixture did not change the total conversion of both monomers, leading to the synthesis of SBR nanocomposites with almost the same polymer composition as that of the materials

289

290

11 Polybutadiene Clay Nanocomposites by In-situ Polymerization

Scheme 11.2 Schematic in-situ formation process of the SBR/clay nanocomposite.

Reproduced from Ref. [27] with permission.

Figure 11.2 TEM micrograph of SBR nanocomposites at different styrene (St)/butadiene

(b) ratio (clay content = 3 wt%). Reprinted with permission from Elsevier. Reproduced from reference [27] with permission.

prepared without clay. Farther, the addition of OC did not affect: (i) the living fashion of the copolymerization until to the inorganic content of 4 wt % and (ii) the randomness of the insertion of the monomers into the growing polymer chain. In contrast, it seemed to affect the microstructure of the butadiene units increasing the cis-1,4 content up to ca. 55 wt% with respect to ca. 51 wt% of pristine SBR. More interestingly, they observed as the structure and the morphology of the SBR nanocomposite was strongly affected by the styrene content in the elastomeric matrix. The TEM image (Figure 11.2) showed some clay tactoids for the SBR/clay sample (St/B = 10 : 90), while the copolymer at the 25 wt% of styrene exhibited an homogeneous distribution of the ﬁller and exfoliated silicate lamellae in the polymer matrix. As a result of the electron withdrawing effect of benzene ring, the styrene molecules have stronger polarity than the butadiene monomer and

11.3 Polybutadiene Nanocomposites a)

b)

Figure 11.3 TEM micrograph of SIBR nanocomposites prepared by using an organoclay-

toluene suspension (a) and a cyclohexane one (b) reprinted with permission from WILEY-VCH. Reproduced from Ref. [25] with permission.

cyclohexane solvent. Hence, the authors supposed that styrene monomer ﬁrst enters within the interlayer clay spacing increasing the d-spacing (Scheme 11.2). As a result, the butadiene monomer diffuses more easily from the bulk solution to the enlarged clay interlayer, promoting the exfoliation of the ﬁller. The welldispersed organoclay lamellae have strong interaction with the SBR chains, leading to the increasing of the glass transition temperature (Tg), the tensile strength (σb), the elongation at break (εb) of the nanocomposites. Next, Zhang [25] observed that by using an OC-toluene suspension, demixing phenomenon was not detected; toluene has a stronger polarity than cyclohexane, hence, as the authors observed in the case of styrene, toluene ﬁrst easily enters within the clay sheets affording an increase of the basal spacing. Hence, exfoliated styrene-isoprene-butadiene (SIBR) nanocomposites were synthesized by living anionic method with an OC– toluene mixture (Figure 11.3a). By using an OC-cyclohexane, the TEM investigation showed the existence of tactoids structures as for pristine organoclay (Figure 11.3b). The terpolymer/clay nanocomposites exhibited a higher thermal stability, and improved mechanical properties than pristine SIBR and SIBR/clay nanocomposites prepared with OC-cyclohexane suspension. Butadiene-isoprene copolymer (BIR)/clay nanocomposites too have been prepared through anionic in-situ polymerization by Zhang et al. in the presence of polar additives such as N,N,N′,N′-tetramethylethanediamine (TMEDA) [24]. The composites were prepared as follows: a certain amount of butadiene, isoprene (butadiene/isoprene = 1/1 wt/wt), TMEDA, OC–toluene mixture was introduced into a 250 ml reactor and stirred for 3 h. Then, a small amount of n-BuLi was added as impurities scavenger, before introducing the stoichiometric amount of initiator (TMEDA/n-BuLi = 0.2) to start the polymerization. The reaction was performed at two different polymerization temperature (30 °C and 60 °C) for 6 h. TMEDA polar molecules were found to have a rather strong effect on the copolymer

291

292

11 Polybutadiene Clay Nanocomposites by In-situ Polymerization

microstructure decreasing the content of 1,2 and 3,4 units in PB and PI, respectively. The degradation pathway for all nanocomposites resulted to be shifted toward higher temperatures compared to neat BIR samples. The maximum values were registered when the content of organoclay ranging from 2 to 3 wt%. The remarkable improvement in the thermal stability was likely due to the well-known physical barrier effect of the nano-dispersed clay layers that delays the diffusion of the oxygen from the gas phase to the polymer matrix and, at the same time, the out-diffusion of the volatile decomposition products. Liao and coworkers [23] prepared PB, PI, and SBR/clay nanocomposites. They carried out a systematic study focused on the in-situ anionic polymerization method by using OCs modiﬁed with quaternary ammonium salts at different ammonium chain length (i.e., Nannolin DK1, Nannolin DK1B, and Nannolin DK4 supplied by Fenghong Clay Chemical Corp.). A suitable amount of organoclay was dispersed into a mixture of toluene and butadiene (or isoprene) monomer (toluene/ monomer = 8/1) kept under constant and vigorous magnetic stirring for 3 h and then preheated at 50 °C for 10 min before using it in the polymerization. The polymerization was started by adding the n-BuLi initiator. All experiments were performed at the ﬁxed polymerization temperature of 50 °C for a time ranging from 3 to 6 h. The method was ﬂexible and allowed to prepare a range of polymers and SBR copolymers exhibiting different microstructure and nanocomposites having different morphologies. The XRD and TEM analyses revealed that a discrete extent of exfoliation was obtained; the best ﬁller dispersion was displayed in SBR and PB nanocomposites rather than in the PI matrix. The authors attributed the worst clay dispersion in the PI matrix to a steric hindrance due to the methyl groups, likely inhibiting the monomer diffusion toward the interlayer spacing, and decreasing indeed the intercalation/exfoliation degree. The best results were obtained for SBR nanocomposites (clay content = 2.5 wt%); the much stronger interaction between the SBR matrix and the organoclay was attributed to the higher polarity of the benzene ring that easily interacted with the intercalated ammonium surfactant. The nanocomposites showed an increased thermal stability; the addition of the organoclay increased the initial decomposition temperature as well as the maximum degradation one in comparison to the pure polymers. In another paper, as already observed by Zhang et al. [25], Liao demonstrated that the experiments carried out with a toluene- or xylene–OC suspension instead of a cyclohexane one, resulted in the formation of more stable emulsions, so that aromatic solvents could be successfully used to synthesize PB/clay nanocomposites by in-situ polymerization through the anionic route [22]. Nevertheless, to date, the nonpolar cyclohexane solvent is widely used in industry for producing butadiene-based polymers. Hence, to overcome the poor stability of the clay–cyclohexane suspension, Liao and coworkers investigated a new strategy to stabilize it for a long time [26]. They modiﬁed the pristine clay by ion exchanged with a quaternary ammonium salt and with a titanate coupling agent, in order to modify both the interlayer clay spacing and the edge surface. The resulting organoclays were then mixed with cyclohexane (4 wt%) and a mixture of cyclohexane and butadiene (8/1 v/v) was added to the clay suspension. n-BuLi was

11.3 Polybutadiene Nanocomposites

added to the clay-butadiene-solvent mixture and the polymerization was started and performed at 50 °C for 4 h. First, the authors investigated the effect of ammonium surfactants with different alkyl chain length. As expected and reported by several authors [16], the clay d-spacing determined by XRD analysis increased with increasing the alkyl chain length. For instance, the diffraction peak (d001) moved to 4.36 ° 2θ for a C12 alkyl chain and to 1.80 ° 2θ for a C22 one, with respect to 5.76 ° 2θ for pristine clay. The organoclay was then treated with the coupling agent and the XRD investigation revealed that, upon the edge-clay modiﬁcation, no signiﬁcant change in the interlayer spacing was observed. The Fourier transform infrared spectroscopy (FTIR) analysis conﬁrmed that the treatment with the coupling agent restricted the clay water adsorption increasing its hydrophobicity; sedimentation experiment showed that the cyclohexane solution of the twice modiﬁed organoclay remained stable until to 72 h. The nanocomposites obtained by polymerizing butadiene in the presence of such clay–cyclohexane solution had a 1,4 content in the range of 88–91% as a function of the inorganic amount. The XRD investigation displayed no diffraction peak, although the TEM micrograph showed the coexistence of disordered intercalated structure and some exfoliated lamellae. The glass transition temperature (Tg), determined by differential scanning calorimetry (DSC), was found to increase due to the strong interfacial forces between the inorganic lamellae and the conﬁned PB chain segments. 11.3.3 In-situ Stereospeciﬁc Polymerization

In the last years much research in academic and industrial laboratories has focused on the ﬁeld of PCNs. Particularly, most effort was focused on the synthesis of polyoleﬁn nanocomposites by in-situ transition metal-catalyzed polymerization, as witnessed by several reviews recently published on this topic [36–38]. The polyoleﬁn industry plays an important role in the international economy system since polyoleﬁns are one of the biggest families of commodity polymers. The worldwide growth of polyoleﬁns has been accompanied by signiﬁcant and continuous research on new catalysts capable of synthesizing polyoleﬁns with new and improved physical and mechanical properties and new functionalizations. The potential applications of the polymers are determined by their physical and mechanical properties, which in turn are deﬁned by the morphology of the polymer. Polymer morphology largely depends on the composition and architecture of the polymer. Therefore, the development of synthetic methods for the polymerization of a wide range of monomers to polymers with controlled stereochemistry and molecular weight is a long-standing scientiﬁc challenge. To this purpose, the development of novel heterogeneous catalysts supported on inorganic clay minerals had a huge success by the research groups interested in the synthesis of polymers via Ziegler–Natta catalysis. The use of such inorganic ﬁller as metal “nano” carriers may assist in the design of nano-reinforced polymer as well as in improving the catalyst performances (i.e., catalyst life-time, polymerization activities) [5, 39]. A lot of work has been carried out in order to use layered silicates

293

294

11 Polybutadiene Clay Nanocomposites by In-situ Polymerization

as support for Ziegler–Natta catalysts for the preparation of polyethylene (PE) [5, 40], polypropylene (PP) [41], isotactic polypropylene (i-PP) [42], polystyrene (PS) [9], and polymethylmethacrylate (PMMA) clay nanocomposites [43]. So far, the in-situ approach has been little explored for the 1,3-dienes polymerization. Only very recently our research group ﬁrst reported on the synthesis of 1,2and cis-1,4 PB/clay nanocomposites by in-situ polymerization [28, 29]. In the following paragraph, we will summarize the most signiﬁcant results obtained, and will focus on identifying the remaining challenges as well as the future prospects of this technology applied to the ﬁeld of the 1,3-dienes stereospeciﬁc polymerization. Trying to summarize our experience about the in-situ synthesis of PB/clay nanocomposites, we can say that this technology involves the following three main steps: 1)

designing of modiﬁcation;

organoclay

by

interlayer

and

edge-surface

hydroxyls

2)

synthesis of a clay-supported catalyst by successive intercalation of an aluminum alkyl cocatalyst (e.g., methylaluminoxane (MAO)) and a transition metal complex;

3)

in-situ polymerization by using the clay–intercalated catalyst in a solvent– monomer mixture.

To achieve PNs via Ziegler–Natta catalysis, the catalyst has to be immobilized into the clay layers and the subsequent intercalation/exfoliation degree depends on the concentration of the reactive sites formed in between the silicate galleries. Most preparations using clay as catalyst-support involve a thermal or chemical pretreatment to remove residual water or edge hydroxyls that can lead to a low-active Ziegler–Natta catalyst or even turn it inactive in the polymerization. This approach requires an absorption of MAO onto the inorganic particle and the successive addition of the metal catalyst precursor. The metallorganic complex is activated by MAO. The activation reaction is comprised of the methylation of the metal center in the ﬁrst step and of the carbanion abstraction, to give a highly Lewis acidic monomethyl cation, stabilized by the MAO anion, in the second step [44]. The ion pair supported in such a way on the clay carrier is the actual catalyst that allows for the growth of the polymer chains. It is worth pointing out that MAO is commonly referred to as linear chain or cyclic rings [–Al(Me)–O–]n with n ≈ 5–20 and containing three-coordinate aluminum centers, meaning that the true structure of MAO is still a matter of debate [45]. In view of these ﬁndings, it is easy to imagine how the key stage of this in-situ approach is the immobilization of the aluminoxane cocatalyst within the clay carrier sheets. Therefore, a full complement of structural X-ray investigation and TEM analysis of the clay–MAO intermediate before the polymerization becomes crucial to have a right check of the ﬁller intercalation extent in the polymer matrix. In our ﬁrst study, we reported on a method of clay intercalating cobalt phosphine complexes (e.g., CoCl2(PMePh2)2 and CoCl2(PiPrPh2)2) which, in combina-

11.3 Polybutadiene Nanocomposites

tion with MAO, were found to be extremely active and stereospeciﬁc for the 1,2 polymerization of 1,3-butadiene [46]. A commercial organo-modiﬁed bentonite {Dellite®72T, modiﬁed with modiﬁed with a quaternary ammonium salt having two methyl groups and two alkyl tails (HT) with various lengths (dimethyldi(hydrogenated tallow) (DMDHT) quaternary ammonium chloride, HT ≈ 65% C18, 30% C16 and 5% C14)} has been used in this study. The clay was ﬁrst extracted by Soxhlet with ethanol to remove the excess of the DMDHT surfactant which, as observed in our previous work, had some negative effect on the direct formation of PCN by poisoning the cationic metal polymerization centers [5]. The extracted clay was then placed in a 50 ml Schlenk and a toluene/ MAO solution was added. The mixture was stirred for 90 min at room temperature. It has been shown that the MAO-treated D72T displayed a broad d001 peak shifted to the lower value of °2θ respect to pristine organoclay, suggesting the intercalation of the aluminum alkyl in between the clay galleries. The peak diffraction intensity resulted strongly decreased as a consequence of the interaction between the aluminum alkyl and the organically modiﬁed silicate that produced the ammonium salt displacement. Upon the addition of the cobalt complex, the polymerization occurs (catalyzed by the cationic cobalt species) giving 1,2 polymer (1,2 content in the range 73–84%) with a predominantly syndiotactic or atactic structure, depending on the catalyst used (i.e., type of phosphine bonded to the cobalt atom). Speciﬁcally, the system clay/MAO/CoCl2(PiPrPh2)2 gave a predominantly syndiotactic polymer (percentage of syndiotactic triads, rr%, about 70), while the system clay/MAO/CoCl2(PMePh2)2 an atactic polymer. While the clay had almost no effect on the polymer microstructure (i.e., 1,2 content) and tacticity (i.e., percentage of rr, mr, and mm triads), the polymerization rate was strongly decreased by the presence of the clay carrier. Such a decreasing may be considered a proof of the nanoconﬁnement of the active centers that resulted less mobile and therefore more difﬁcult to approach by the butadiene monomer. The wide angle X-ray diffraction (WAXD) investigation of the nanocomposites showed the presence of broad shoulder shifted to a lower diffraction angle when compared to the characteristic (0 0 1) peak of pristine clay and to the maximum of the MAO-treated clay. The enlargement of the nanoclay galleries in the nanocomposites was measured by small angle X-ray scattering (SAXS). The periodicity due to the layer spacing was found 3.7 nm for the nanocomposite prepared by clay/ MAO/CoCl2(PiPrPh2) (Figure 11.4). This value was signiﬁcantly higher than the d-spacing of D72T (2.3 nm), conﬁrming the formation of intercalated structures. The results of the morphological investigation were in good agreement with the structural characterization performed by X-ray techniques. Indeed, the TEM micrograph at high magniﬁcation conﬁrmed the presence of intercalated structures (Figure 11.4). The nanocomposites exhibited an enhanced thermal stability: the maximum decomposition temperature moved to 483 °C and 480 °C for the sample prepared with clay/MAO/CoCl2(PMePh2)2 and clay/MAO/CoCl2(PiPrPh2), respectively, with respect to 479 °C and 472 °C for the neat 1,2-PB prepared with the same catalysts in homogeneous conditions.

295

296

11 Polybutadiene Clay Nanocomposites by In-situ Polymerization a)

b)

Figure 11.4 SAXS Lorentz-corrected plot (a) and TEM micrograph (b) of PB/D72T

nanocomposite prepared with clay/MAO/CoCl2(PiPrPh2) heterogeneous catalyst.

In a more recent work [29], we reported on the synthesis of cis-1,4 PB/clay nanocomposites. We synthesized and characterized two organoclays (DVA-e and DS2) to be used as nanoreactors for the intercalation of CoCl2(PtBu2Me)2-MAO catalyst, and thus for the stereospeciﬁc polymerization of 1,3-butadiene. Speciﬁcally, starting from a natural sodium-exchanged bentonite (Dellite®HPS), DVA-e was prepared by ion exchange with (ar-vinyl-benzyl)trimethyl ammonium (VBTA) chloride surfactant and DS2 by further DVA-e modiﬁcation through the silylation of the clay edge-OHs by trimethylchlorosilane (TMSCl) coupling agent. Quite interestingly, we observed that the silylation of the edge-OHs occurred exclusively by using the VBTA-modiﬁed clay. In the presence of the sodium-clay, the organosilane was mainly intercalated within the interlayer spacing. Hence, we attributed this result to the fact that the ammonium surfactant not only increased the d-spacing, as expected, but also, due to the strong van der Walls interaction between the VBTA aromatic rings, forced the silane to react with the edge-surface of the silicate. Next, upon the MAO reaction, and by integrating the TGA, FTIR, and XRD investigations, we found that by using DVA-e clay without any edge modiﬁcation, both the presence of layers stacked in small pile (as for pristine DVA-e due to unexchanged ammonium salt) and stacks spaced by around 5.6 nm with penetration of MAO were detected. Vice versa, by using the twice-modiﬁed organoclay DS2, MAO displaced almost the entire amount of the intercalated surfactant and, as a consequence, it expanded the basal spacing to a larger extent of about 5.6 nm. Even though Morris et al. [47] determined that only 3.3% of all Al atoms are clay edge-Al-OH Brönsted acid groups, these features made it possible to design clay carriers with a tunable physical allocation of the active polymerization centers. Indeed, in the presence of the clay without any modiﬁcation of the edge-surface, MAO, due to its strong Lewis acid character, interacted also with the edge-OHs, leading to the formation of external active polymerization centers. To clear up the role of the reactive edge-surface hydroxyls, and thus of the physical allocation of the active metal centers on the nanocomposites synthesis, we per-

11.3 Polybutadiene Nanocomposites Polymerization of 1,3-butadiene catalyzed by clay/MAO/CoCl2(PtBu2Me) heterogeneous catalyst.a)

Table 11.2

Protocol

in-situ Tunnel a)

b) c) d) e) f)

Conv

Clayb)

cis-1,4c)

1,2c)

(%)

(wt%)

(%)

(%)

80 90

2.98 2.63

75 72

25 28

10−3Mwd)

800 700

Mw/Mnd)

4.8 6.4

Tge)

Tdegf )

(°C)

(°C)

−102 −104

481 487

Polymerization conditions: butadiene, 3 ml; toluene total volume, 18 ml; DVA-eM (in-situ protocol) or DS2-M (tunnel protocol), 50 mg; CoCl2(PtBu2Me)2, 3 μmol; polymerization temperature, 20 °C; polymerization time, 4 days. Calculated by the ratio of amount of the clay charged and PB/clay yield. Determined by NMR analysis. Determined by SEC relative to polystyrene standards. Tg, glass transition temperature determined by DSC on second heating scan. Tdeg, maximum decomposition temperature evaluated by TGA.

Figure 11.5 Plot of monomer conversion versus polymerization time for 1,3-butadiene polymerization by in-situ and tunnel polymerization.

formed the polymerization of 1,3-butadiene in the presence of the organoclay with a single ammonium modiﬁcation (named in-situ polymerization) and with that with a twice modiﬁcation (tunnel polymerization). We carried out a series of polymerization by changing the polymerization time over the range from 150 min to 4 days. The results are summarized in Table 11.2, while in Figure 11.5 the monomer conversion as a function of the polymerization time, for both tunnel and in-situ polymerizations, is reported. The in-situ polymerization clearly exhibited a faster initial reaction rate with respect to the tunnel one, and this behavior was attributed to the easier accessibility of the clay-edge-surface active metal centers (Scheme 11.3). Moreover, the sharpest

297

298

11 Polybutadiene Clay Nanocomposites by In-situ Polymerization

Scheme 11.3 Schematic representation of the polymerization behavior with in-situ protocol,

(upper) and tunnel protocol (bottom) clay-based catalysts.

decreasing was exhibited by in-situ protocol likely because of the clogging by polymer chain growth that leads to a rather poor accessibility of the incoming monomer toward the clay galleries. The resulting nanocomposites exhibited the same layer spacing of the clay– MAO activated parents; the growing of the PB chains ﬁlls the interlayer clay region giving an intercalated nanocomposite. However, it is noteworthy that the WAXD investigation of the sample prepared by in-situ route also showed the presence of stacks spaced by around 1.4 nm as those of the pristine DVA-e organoclay already observed for the MAO-treated clay derivates. At the same ﬁller loading, the degradation pathway for the nanocomposites synthesized by tunnel polymerization resulted to shift toward higher temperatures compared with samples prepared by in-situ polymerization (487 °C vs. 481 °C, respectively). We supposed that the lower thermal stability exhibited by the samples synthesized by in-situ route might be due to: (i) the lower ﬁller intercalation extent and (ii) the presence of layers still organized in tactoids as for pristine organoclay (Scheme 11.3). This structural dishomogeneity hinders the typical physical barrier effect of the inorganic component, and therefore the decomposition process occurred at lower temperatures. As far as concerned the molecular weight of the resulting PB nanocomposites, it is worthy to note that, in our experience, the 1,2-PB as well as the cis-1,4-PB prepared by direct polymerization showed higher molecular weights than those reported for polymers synthesized in homogeneous conditions. This fact could be due to the strong reduction of the rate of β-H transfer reactions, likely attributable to the lower amount of Al–Me moieties onto the clay surface. In fact, during the clay treatment with MAO, the free-Al(CH3)3 was removed while the PB homogeneous references were synthesized by using a commercial MAO toluene solution.

Abbreviations

11.4 Conclusions and Perspectives

In spite of the considerable number of thermoplastic PCNs synthesized by in-situ polymerization in the last 10 years, the area of rubber nanocomposites is not exhausted by far. To date, only a few number of reports have been published on the synthesis of 1,3-butadiene-based PNs. Speciﬁcally, as we have pointed out in this chapter, most of the work on the in-situ polymerization concerned with the anionic polymerization or copolymerization of 1,3-dienes by using the n-BuLi initiator. According to this technique, the monomer is intercalated in between the clay galleries and converts into polymer by in-situ addition of the anionic initiator. The heat released during the growing of the polymer chains allows to weaken the interaction between the inorganic layers (Coulomb force) increasing the basal spacing until to reach the exfoliation of the ﬁller platelets. On the other hand, the development of polybutadiene/clay nanocomposites prepared by in-situ metal-catalyzed polymerization is still in its embryonic stage. At least in principle it might assist in the design of PB-based PNs in a nanocontrolled fashion over their structural features and polymer architecture, such as composition and molecular weight. However, as we have brieﬂy showed herein, the nature of the organo-modiﬁcation of the clay, and therefore their surface character, is the key point to tune the physical allocation of the polymerization active centers and to synthesize PB nanocomposites with improved ﬁller dispersion. In conclusion, from our point of view, the growing interest in the synthesis of novel metal transition or lanthanide based catalyst systems could offer novel opportunities for further improvements in the nanocompounding process by in-situ polymerization via Ziegler–Natta coordination catalysis.

Abbreviations

ABS acrylonitrile-butadiene-styrene BIR butadiene-isoprene copolymer BR butadiene rubber CEC cation exchange capacity DSC differential scanning calorimetry DMDHT dimethyldi(hydrogenated tallow) FTIR Fourier transform infrared spectroscopy HIPS high impact polystyrene Tg glass transition temperature i-PP isotactic polypropylene MAO methylaluminoxane Mw molecular weight Mw/Mn molecular weight distribution NR natural rubber

299

300

11 Polybutadiene Clay Nanocomposites by In-situ Polymerization

OC PB PE PP PCN PN PPMA SEM SAXS SBR SIBR s-PP TGA TEM TMSCL TMEDA VBTA wt% WAXD XRD σb εb

organoclay polybutadiene polyethylene polypropylene polymer/clay nanocomposite polymer nanocomposite polymethylmethacrylate scanning electron microscopy small angle X-ray scattering styrene butadiene rubber styrene-isoprene-butadiene syndiotactic polystyrene thermogravimetric analysis transmission electron microscopy trimethylchlorosilane N,N,N′,N′-tetramethylethanediamine (ar-vinyl-benzyl)trimethyl ammonium chloride weight percentage wide angle X-ray diffraction X-ray diffraction tensile strength elongation at break

References 1 Pinnavaia, T.J. and Beale, G.W. (2000)

2 3

4

5

6

7

8

Polymer-Clay Nanocomposites, John Wiley & Sons, Inc., New York. Carrado, K.A. and Xu, L. (1998) Chem. Mater., 10, 1440. Weimer, M.W., Chen, H., Giannelis, E.P., and Sogah, D.Y. (1999) J. Am. Chem. Soc., 121, 1615. Heinemann, J., Reichert, P., Thomann, R., and Mülhaupt, R. (1999) Macromol. Rapid Commun., 20, 423. Leone, G., Bertini, F., Canetti, M., Boggioni, L., Conzatti, L., and Tritto, I. (2009) J. Polym. Sci., Part. A: Polym. Chem., 47, 548. Alexandre, M., Dubois, P., Sun, T., Garces, J.M., and Jerome, R. (2002) Polymer, 43, 2123. Halbach, T.S., Thomann, Y., and Mülhaupt, R. (2008) J. Polym. Sci., Part A: Polym. Chem., 46, 2755. Halbach, T.S. and Mülhaupt, R. (2008) Polymer, 49, 867.

9 Bruzaud, S., Grohens, Y., Ilinca, S., and

10 11

12

13

14

15 16

Carpentier, J.F. (2005) Macromol. Mater. Eng., 290, 1106. Cui, L., Tarte, N.H., and Woo, S.I. (2008) Macromolecules, 41, 4268. Sengupta, R., Chakraborty, S., Bandyopadhyay, S., Dasgupta, S., Mukhopadhyay, R., Auddy, K., and Deuri, A.S. (2007) Polym. Eng. Sci., 47, 1956. Utracki, L.A., Sepehr, M.E., and Boccaleri, E. (2007) Polym. Adv. Technol., 18, 1. Brindley, G.W. and Brown, G. (1980) Crystal Structure of Clay Minerals and Their X-Ray Identiﬁcation, Mineralogical Society, London. Theng, B.K.G. (1974) The Chemistry of Clay-Organic Reactions, John Wiley & Sons, Inc., New York. Pinnavaia, T.J. (1983) Science, 220, 365. Vaia, R.A., Teukolsky, R.K., and Giannelis, E.P. (1994) Chem. Mater., 6, 1017.

References 17 Alexandre, M. and Dubois, P. (2000) 18 19 20 21

22

23

24 25 26 27 28

29

30

31 32

Mater. Sci. Eng., 28, 1. Morgan, A.B. and Gilman, J.W. (2003) J. Appl. Polym. Sci., 87, 1329. Pavlidou, S. and Papaspyrides, C.D. (2008) Prog. Polym. Sci., 33, 1119. Ray, S.S. and Okamoto, M. (2003) Prog. Polym. Sci., 28, 1539. Podsiadlo, P., Kaushik, A.K., Arruda, E.M., Wass, A.M., Shim, B.S., Xu, J., Nandivada, H., Pumplin, B.G., Lahann, J., Ramamoorthy, A., and Kotov, N.A. (2007) Science, 318, 80. Liao, M., Zhang, W., Shan, W., and Zhang, Y. (2006) J. Appl. Polym. Sci., 99, 3615. Liao, M., Shan, W., Zhu, J., Li, Y., and Xu, H. (2005) J. Polym. Sci., Part B: Polym. Phys., 43, 1344. Zhang, Z., Zhang, L., and Li, Y. (2006) J. Appl. Polym. Sci., 102, 1167. Zhang, Z., Zhang, L., and Li, Y. (2005) J. Macromol. Mater. Eng., 290, 430. Quian, X., Liao, M., and Zhang, W. (2007) Polym. Int., 56, 399. Zhang, Z., Zhang, L., Li, Y., and Xu, H. (2005) Polymer, 46, 129. Leone, G., Boglia, A., Bertini, F., Canetti, M., and Ricci, G. (2009) Macromol. Chem. Phys., 210, 279. Leone, G., Boglia, A., Bertini, F., Canetti, M., and Ricci, G. (2010) J. Polym. Sci., Part A: Polym. Chem., 48, 4473. Porri, L., and Giarrusso, A. (1989). Comprehensive Polymer Science vol. 4, (eds G.C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt), Pergamon Press Ltd., Oxford. Thiele, S.K.H. and Wilson, D.R. (2003) J. Macromol. Sci. C: Polym. Rev., 43, 581. Forman, L.E. (1969). Polymer Chemistry of Synthetic Elastomer (eds J.P. Kennedy and

33 34 35 36 37 38 39

40

41

42

43 44

45

46

47

E. Torqvist), John Wiley & Sons, Inc., New York. IISRP (2002) Worldwide Rubber Statics. iisrp.com (last accessed 17 January 2006). Friebe, L., Nuyken, O., and Obrecht, W. (2006) Adv. Polym. Sci., 204, 1. Foster, F.C. (1967) Firestone Tire and Rubber Co. US 3.317.918. Qin, Y.-W. and Dong, J.-Y. (2009) Chin. Sci. Bull., 45, 38. Kaminsky, W., Funck, A., and Wiemann, K. (2006) Macromol. Symp., 239, 1. Dubois, P., Alexandre, M., and Jéróme, R. (2003) Macromol. Symp., 194, 13. Weiss, K., Wirth-Pfeifer, C., Hofmann, M., Botzenhardt, S., Lang, H., Brüning, K., and Meichel, E. (2002) J. Mol. Catal. A, 182–183, 143. Leone, G., Bertini, F., Canetti, M., Boggioni, L., Stagnaro, P., and Tritto, I. (2008) J. Polym. Sci., Part. A: Polym. Chem., 46, 5390. Tudor, J., Willington, L., O’Hare, D., and Royan, B. (1996) Chem. Commun., 2031. Du, K., He, A.H., Liu, X., and Han, C.C. (2007) Macromol. Rapid Commun., 28, 2294. Mariott, W.R. and Chen, E.Y.X. (2003) J. Am. Chem. Soc., 125, 15726. Bochmann, M. (1995) Comprehensive Organometallic Chemistry II (eds E.W. Abel, F.G.A. Stone, and G. Wilkinson), Pergamon, New York. Sinn, H., Kaminsky, W., Vollmer, H.J., and Woldt, R. (1980) Angew. Chem., 92, 396. Ricci, G., Sommazzi, A., Masi, F., Ricci, M., Boglia, A., and Leone, G. (2010) Coord. Chem. Rev., 254, 661. Morris, H.D., Bank, S., and Ellis, P.D. (1990) J. Phys. Chem., 94, 3121.

301

303

12 P3HT–MWNT Nanocomposites by In-situ Polymerization and Their Properties Zhongrui Li and Liqiu Zheng

12.1 Introduction

The discovery of carbon nanotubes (CNTs) has excited a great and sustained interest in 1D nanotechnologies due to their unique mechanical, thermal, electronic, optical, and chemical properties. CNTs exhibit low mass density [1], large aspect ratio (typically ∼1000), and high ﬂexibility [2]. They can be lighter than aluminum, stronger than steel, and more conductive than copper. Metallic CNTs are able to transport electrons over long tube lengths without signiﬁcant scattering due to the nearly 1D electronic structures [3]. Similarly, CNTs exhibit large phonon mean free path lengths and have a high thermal conductivity [4]. CNTs have a unique combination of these properties that make nanotubes excellent candidates to substitute or complement the conventional nanoﬁllers in the fabrication of multifunctional polymer nanocomposites. In particular, the novel structure, high aspect ratio, large surface area, high conductivity, and stability [5] make CNTs preferred for use as ﬁller materials in a polymer matrix. However, many practical applications of CNTs have been largely limited by their poor processability since they are practically insoluble and infusible. Among the various conducting polymers, 3-alkylthiophenes are the best candidates owing to their high processibility. The combination of CNTs with inherent conducting polymers (ICPs) offers an attractive route to reinforce the polymer as well as to introduce electronic properties based on a morphological modiﬁcation or an electronic interaction between the two components. In ICP–CNT composites, either the polymer functionalizes the CNTs or the ICPs are doped with CNTs, that is, a charge transfer occurs between the two constituents. Conducting polymer–fullerene (like ICP–CNT) composites generated numerous potential applications based on their unique chemical, mechanical, optical, and conducting properties [6–8]. ICP–C60 composites have shown promise for applications in photovoltaic devices, owing to efﬁcient

In-situ Synthesis of Polymer Nanocomposites, First Edition. Edited by Vikas Mittal. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA.

304

12 P3HT–MWNT Nanocomposites by In-situ Polymerization and Their Properties

electron transfer from the conjugated polymer to C60 under visible light [9]. In this respect, ICP–CNT composites are of great interest concerning the novel electronic interaction between these two elements and the mechanical reinforcement of the polymeric materials. An increased short-circuit current density has been found in the dye, N-(1-pyrenyl)maleimide (PM), functionalized single-wall carbon nanotubes (SWNTs) blended with poly(3-octylthiophene) (P3OT) photovoltaic cell, compared with the SWNT–polymer photovoltaic cell without the dye. The increased photocurrent may be due to the dye acting as an interface for better transfer of charge carriers between SWNTs and P3OT mainly because of ground-state interactions between SWNT, dye, and P3OT [10]. Poly(alkylthiophene)s such as poly(3-hexylthiophene) (P3HT) have attracted great attention due to a combination of their relatively high chemical stability in ambient conditions, their high conductivity, and the fact that the electronic band gap of these materials generally falls within the visible region of the electromagnetic spectrum. Recently, interest has been focused on the preparation of solution processed thin ﬁlms of regioregular poly(3-hexylthiophene) (rrP3HT) [11, 12], due to its great potential for a variety of applications as, for instance, ﬁeld-effect transistors [13] and photovoltaic devices [14]. P3HT also has good solubility in various solvents and high electrical conductivity when electrons are added or removed from the conjugated π-orbitals through doping [15, 16]. The alkyl group is incorporated in the thiophene ring with two different regioregularities: head-to-head (HH) and head-to-tail (HT). The HT regioregularity is preferred over HH regioregularity since it improves electroconductivity, optical nonlinearity, and magnetic properties [17]. P3HT also exhibits photoluminescence properties based on the tunability of procedures used during synthesis [18, 19]. The P3HT–CNT composites exhibit interesting physical, optical, and conductivity properties. But for a fruitful use of these properties, a clear understanding of the behavior of the P3HT– CNT composites prepared from different techniques is necessary. CNTs are difﬁcult to process due to their insolubility in commonly used solvents [20]. Wrapping appropriate polymer onto the walls of CNTs through noncovalent interactions, such as π–π interaction and/or CH–π interaction, can improve the dispersity of CNTs in polymeric materials [21]. Moreover, the polymerization of monomer, can be carried out in the presence of CNTs functionalized by nonpolar groups, to prepare polymer wrapped CNTs [22]. In this chapter, we will discuss in-situ polymerization of a soluble conductive polymer P3HT in the presence of different doping levels of CNTs in order to uniformly disperse multiwall carbon nanotubes (MWNTs) in P3HT for applications in optoelectronic devices. For the application in organic photovoltaic devices, MWNTs can provide better carrier transport than SWNTs, since semiconducting tubes usually dominate the SWNT product. MWNTs also offer a better mechanical strength. Additionally, MWNTs can be produced at a large quantity with a much lower cost. We will also explore the morphological, mechanical, thermal, optical, and electrical properties of the in-situ polymerized P3HT–MWNT nanocomposites. The progress, remaining challenges, and future directions of the nanotubes/ polymer composite research will be discussed in this chapter.

12.2 Multiwall CNTs

12.2 Multiwall CNTs

CNTs, sources of nanotubes, and some fundamental properties of nanotubes are critical to understanding P3HT–CNT composites. CNTs come in different types and vary signiﬁcantly depending on the syntheses procedures. They are long cylinders of covalently bonded carbon atoms with or without hemifullerene caps at the ends. Based on their wall numbers, CNTs can be classiﬁed into SWNT and MWNT. SWNT can be considered as a single graphene sheet (graphene is a monolayer of sp2-bonded carbon atoms) rolled into a seamless cylinder. The carbon atoms in the cylinder have partial sp3 character that increases as the radius of curvature of the cylinder decreases. MWNT consists of many graphene cylinders coaxially arranged around a central hollow core with interlayer separations of 0.34 nm, indicative of the interplane spacing of graphite [23]. When functionalization is required (this means grafting of chemical functions at the surface of the nanotubes) to add new properties to the CNT, covalent functionalization of SWNT will break some C=C double bonds, leaving “holes” in the structure on the nanotube and thus modifying both its mechanical and electrical properties. In the case of MWNT, only the outer wall is modiﬁed. The nanotubes can be ﬁlled with foreign elements or compounds, for example, with C60 molecules, to produce hybrid nanomaterials which possess unique intrinsic transport properties [24]. These hybrid nanomaterials might be a new opportunity for polymer nanocomposites. The many properties of SWNTs are determined by the tube chirality (or helicity) as deﬁned by the circumferential vector, Ch = na1 + ma2 (Figure 12.1), where the integers (n, m) are the number of steps along the unit vectors (a1 and a2) of the

Figure 12.1 Schematic diagram (a) showing how a hexagonal sheet of graphene is “rolled” to

form a carbon nanotube, and the chiral map (b) displaying metallic and semiconducting tubes.

305

306

12 P3HT–MWNT Nanocomposites by In-situ Polymerization and Their Properties

hexagonal lattice [25]. Using this (n, m) naming scheme, the three orientation types of the carbon atoms around the nanotube circumference are speciﬁed as arm chair (n = m), zigzag (n = 0 or m = 0), or chiral (all others). All armchair SWNTs are metallic with a band gap of 0 eV. SWNTs with n − m = 3i (i is a nonzero integer) are semimetallic with a band gap of the order of a few meV, while SWNTs with n − m ≠ 3i are semiconductors with a band gap inversely proportional to its diameter (usually in the range of 0.2–3 eV) [26]. Each MWNT contains a variety of tube chiralities, so its physical properties are more complicated to predict. The manifestation of mesoscopic transport properties in a MWNT is illustrated through the Aharonov–Bohm effect, universal conductance ﬂuctuations, the weak localization effect, and its power-law temperature/ﬁeld dependences [27]. Measurements of Young’s modulus of individual MWNTs show the high strength of tubes having well-graphitized walls. Electron spin resonance (ESR) measurements indicate the low-dimensional character of the electronic states even for relatively large diameter tubes. The conducting nature of the tubes, together with their large curvature tip structure, make them excellent electron and light emitters suitable for applications. Over the last two decades, different production techniques have been developed to make CNTs, and the most commonly used approaches are arc discharge [28, 29], laser ablation [30, 31], and chemical vapor deposition (CVD) [32, 33]. Arc discharge and laser ablation methods involve the condensation of hot gaseous carbon atoms generated from the evaporation of solid carbon. Compared with arc and laser methods, the catalytic CVD method attracts a lot of interest by making possible the large-scale and high-quality production of CNTs at a relatively low cost. Additionally, with the catalytic CVD method, the growth (including diameter, length, orientation, position, etc.) of CNTs can be controlled by adjusting the reaction conditions and choosing proper catalysts. In the CVD method, CNTs are produced from the thermal decomposition of the carbon-containing molecules on desirable metal catalysts (commonly Fe, Co, and Ni). Many efforts have been put to optimize catalyst formulations and operating conditions [34, 35]. The catalyst composition controlled by the preparation method affects the efﬁciency and selectivity of the catalytic reaction toward synthesis of desired CNTs. Considering the strong correlation between the diameter of the nanotubes and the size of catalytic metal particles, the high dispersity of the catalytically active metal ingredient on the support seems to be a critical point in the synthesis of CNTs by the catalytic CVD method [36]. Currently all known preparations of CNTs give mixtures of nanotube chiralities, diameters, and lengths along with different amount and type of impurities. MWNTs can have diameters from several nanometers to several hundred nanometers. The reported lengths of CNTs range from several tens of nanometers to several centimeters [37]. The properties of the ICP–CNT composites will vary signiﬁcantly depending on the distribution of the type, diameter, and length of the nanotubes. The high aspect ratio of the nanotubes coupled with a strong intrinsic van der Waals attraction between nanotubes combine to produce ropes and bundles of CNTs, particularly in SWNTs where the attractive force is 0.5 eV per

12.3 In-situ Synthesis of P3HT–MWNT Composites

nanometer of nanotube-to-nanotube contact [26]. With the aid of ultrasonication, CNTs can be moderately dispersed in some solvents, for example, in dimethylformamide, N-methyl-2-pyrrolidone, and dichlorobenzene, to produce uniform nanotube suspensions. Preparing nanotube suspensions is vital in controlling various solvent-based processes (phase separation, chemical derivatization, etc.) associated with preparing ICP–CNT composites because the initial nanotube dispersion can impact the nanotube dispersion in the polymer matrix. Isolated tubes facilitate extensive chemical derivitatization in suspension and provide the highest interfacial area for stress transfer to the polymer matrices.

12.3 In-situ Synthesis of P3HT–MWNT Composites

When preparing an ICP–CNT composite, it is essential to qualify the type of interaction between the host matrix and the guest nanoparticles (ﬁllers). The main functionalization possibilities of CNTs reported until now are [38]: (i) endohedral functionalization with C60, (ii) covalent side wall functionalization using addition reactions and subsequent nucleophilic substitution, (iii) generation and functionalization of defect sites at the tube ends and side walls by oxidation and subsequent conversion into derivatives, and (iv) noncovalent exohedral functionalization with surfactant-type molecules. These are represented in Figure 12.2. Three routes have been used to prepare ICP–CNT composites: (i) direct mixing of the ICP with CNTs, (ii) chemical polymerization of the corresponding monomer

a)

b)

e)

CNT

c)

d)

Figure 12.2 Several possible functionalization with polymers, (c) defect-group

mechanisms for CNTs. (a) endohedral functionalization with polymer, (b) noncovalent exohedral functionalization

functionalization, (d) covalent sidewall functionalization, and (e) noncovalent exohedral functionalization with surfactants.

307

308

12 P3HT–MWNT Nanocomposites by In-situ Polymerization and Their Properties

in the presence of CNTs and (iii) electrochemical synthesis of ICP on CNTs electrode. Recently, it was found that different approach could result in different materials [39]. The chemical polymerization of monomer on SWNTs leads to the breaking of SWNTs in small fragments in the presence of an oxidizing medium such as K2Cr2O7 and H2SO4, although direct mixing of CNTs with ICP solution route does not drastically affect the CNTs [40]. The methods of solution blending, electrochemical deposition, and in-situ polymerization are widely applied to produce ICP–CNT composites and will be summarized here. In addition, latex technology [41], solid-state shear pulverization [42], and coagulation spinning [43] methods also show promise. a)

Solution Blending: In solution blending, ﬁrst, CNTs are dispersed in a suitable solvent, then mixed with the polymer (at room temperature or elevated temperature), and ﬁnally the composite is recovered by precipitating or casting a ﬁlm. Ultrasonication can be used to make metastable suspensions of nanotubes or polymer–nanotube mixtures in different solvents, but high-power ultrasonication for a long period of time shortens the nanotube length, which is detrimental to the composite properties [44]. Surfactants can improve nanotube dispersion but the surfactant remaining in the resulting nanocomposite might degrade transport properties, and can also induced crystallization in ploymer [45], which might in turn affect the transparency and mechanical properties of the composites. One alternative to surfactant-aided dispersion is nanotube functionalization to improve dispersion and interfacial adhesion to the polymer matrix. When using solution blending, nanotubes tend to agglomerate during slow solvent evaporation, leading to inhomogeneous distribution of the nanotubes in the polymer matrix. The evaporation time can be reduced by putting the ICP–CNT suspension on a rotating substrate (spin-casting [46]) or dropping the ICP–CNT suspension on a hot substrate (drop-casting [47]).

b) Electrochemical Deposition: In electrochemical deposition, a potential is applied across a solution containing polymer and an electrolyte, producing a conductive polymer ﬁlm on the anode. The conducting polymer can form a continuous, uniform coating over each individual CNT when an aligned CNT preform is used as anode. In this way, the basic structure of the CNT preform is retained, making it possible to produce composites with a very high surface area. On unaligned CNT preforms, the deposited conducting polymer might tend to block electrolyte channels into the preform, preventing uniform coating of each CNT. c)

In-situ Polymerization: This fabrication strategy starts by dispersing nanotubes in monomer followed by polymerizing the monomers. In electrochemical polymerization, the CNTs and conducting polymer are simultaneously deposited; the electrolyte used contains suspended CNTs in addition to the desired monomer and any supporting electrolyte required. Polymerization is then initiated via the application of an applied potential, resulting in the

12.3 In-situ Synthesis of P3HT–MWNT Composites

uniform deposition of the conducting polymer on each CNT. As compared with solution blending, functionalized nanotubes can improve the initial dispersion of the nanotubes in the liquid (monomer, solvent) and consequently in the composites. Furthermore, in-situ polymerization methods enable covalent bonding between functionalized nanotubes and the polymer matrix using various condensation reactions. Usually, ICP–CNT composite made with in-situ polymerization has more advantages over those made with other methods, for example, poly[3-(2-hydroxyethyl)-2,5-thienylene] (PHET) produced in situ with MWNTs displays increased conductivity because of the coating of nanotube walls with polymer in opposition to the composite prepared by sonicating a mixture of PHET and nanotubes [48]. Here, we took in-situ polymerization of thiophene to P3HT in the presence of MWNTs (0.1–10 wt%) as example. (i) The MWNTs were synthesized from catalytic decomposition of acetylene over a Fe–Co/CaCO3 catalyst system (with a weight ratio Fe : Co : CaCO3 = 2.5 : 2.5 : 95) under radio frequency heating at 720 °C for 30 min. The as-produced MWNTs were puriﬁed by reﬂuxing in HCl for 24 h. The nanotubes were further washed in de-ionized water for 24 h and dried at 100 °C overnight. The purity level of the nanotubes was above 97% after one acid wash treatment. (ii) P3HT was synthesized by oxidative polymerization of the 3-hexylthiophene (3HT) monomer in the presence of anhydrous FeCl3 at room temperature. Anhydrous FeCl3 (20 mmol) was placed in a 500 ml three-necked bottle with 100 ml CHCl3. The suspension was stirred for 15 h under nitrogen ﬂow. 3HT (5 mmol) was added to the suspension via a syringe and stirred for another 24 h under nitrogen ﬂow. The black mixture was transferred to a methanol–HCl mixture (9 : 1) and stirred for 5 min. The black precipitate was then ﬁltered with a vacuum ﬁltration system using a Teﬂon membrane of 25 nm pore size. The black solid was washed in a Soxhlet’s extraction unit with methanol for 60 h. (iii) The P3HT–MWNT composites were synthesized under the same conditions and procedure as described above except by sonicating MWNTs in CHCl3 and then adding them together with 3HT in the FeCl3 suspension and stirring for the next 24 h. The P3HT-MWNT nanocomposites (named as PM-x, x is the percentage of MWNTs by the weight of monomer) were prepared with different amount of MWNTs. Fabrication methods of ICP–CNT composites have overwhelmingly focused on improving nanotube dispersion because better nanotube dispersion in the polymer matrices has been found to improve properties. Similar to the case of nanotubesolvent suspensions, it is very difﬁcult to overcome the inherent van der Waals forces of nanotubes to make pristine nanotubes soluble in polymers. The quality of CNT dispersion in polymer matrices should be evaluated over a range of length scales and can be accomplished using a selection of these imaging methods: optical microscopy, polarized Raman imaging, scanning electron microscopy (SEM), and transmission electron microscopy (TEM). A limited number of researchers have used small-angle X-ray, neutron, or light scattering with wave

309

310

12 P3HT–MWNT Nanocomposites by In-situ Polymerization and Their Properties

vectors Q in the range 10−3–1 nm−1 (corresponding to length scales 1–1000 nm) to investigate nanotubes structures in suspension [49].

12.4 The Properties and Characterization of P3HT–MWNT Nanocomposites

The properties of polymer nanocomposites containing CNTs depend on several factors in addition to the polymer: synthetic process used to produce nanotubes; nanotube puriﬁcation process (if any); amount and type of impurities in the nanotubes; diameter, length, and aspect ratio of the nanotube objects in the composite (isolated, ropes, and/or bundles); nanotube orientation in the polymer matrix. These variations in nanotubes and ICP–CNT composites account for many of the apparent inconsistencies in the literature. Reporting the nanotube concentration (specifying whether the concentration allots for the impurities or functionalization) and the matrix polymer alone is insufﬁcient. Although the variations listed above are difﬁcult to quantify, more complete reporting will reduce the discrepancies between the published results of similar composites. Entire studies should be performed using CNT materials from the same batch of puriﬁed nanotubes, thereby reducing the variability between samples and clarifying trends. Next sections highlight the dispersion/morphology and the mechanical, electrical, optical, thermal, and dielectric properties of ICP–CNT composites. 12.4.1 The Dispersion and Morphology of the P3HT–MWNT Nanocomposites

The dispersion and morphology of the P3HT–MWNT nanocomposites can be examined through transmission electron microscopic (TEM) images (Figure 12.3) [50]. The nanotubes are in the outer diameter range of 8 to 40 nm with mostly

a)

b)

Figure 12.3 TEM images of (a) the MWNTs

synthesized by cCVD; the P3HT–MWNT nanocomposites synthesized with (b) 1 wt% and (c) 10 wt% MWNTs. (From Saini, V.,

c)

et al. [2009] J. Phys. Chem. C 113[19], 8023–8029. With permission from American Chemical Society.)

12.4 The Properties and Characterization of P3HT–MWNT Nanocomposites

closed ends. The PM-1 sample shows a uniform wrapping of polymer chains around tubes with thickness between 3 and 8 nm. In the PM-10 sample, a very thin polymer layer covers the tube walls. MWNTs are aromatic or ring-based molecules with extensive delocalized covalent bonds, thus can be easily dispersed in matrix materials with delocalized bonds. Wrapping the MWNTs with watersoluble polymers can lead to a stable dispersion. The polymer coats both nanotubes and graphitic impurities, but only the polymer-coated nanotubes form a stable dispersion, while the impurities have been found to sediment out [51]. 12.4.2 HT Regioregularity

The HT regioregularity of the P3HT can be determined by nuclear magnetic resonance (NMR) spectroscopy. 1H-NMR spectra of HT regioregularity of the P3HT–MWNT composites obtained on a Bruker 200 MHz NMR are displayed in Figure 12.4. The samples were dissolved in CDCl3 (10 mg ml−1) with tetramethylsilane as the reference, and ﬁltered using a glass wool to separate out insoluble material. The inset of the ﬁgure depicts the NMR peaks generated by the corresponding protons on a single unit of P3HT. As shown in Figure 12.4a, the NMR peak of aromatic hydrogens (δ = 6.98 ppm) becomes broader in the P3HT-MWNT composites due to the π–π interactions between the thiophene ring of P3HT and the benzene ring of MWNTs [52]. The broadening was also observed in the NMR peaks of methylene protons of the pendent hexyl group (δ = 2.81, 2.58, 1.70, 1.57, 1.37 ppm). Furthermore, the –CH3 peak (0.92 ppm) shifts to somewhat upﬁeld region (0.91 ppm) and broadens in the composites, indicating the presence of the CH–π interaction between the pendent hexyl group and graphitic walls [53]. The NMR peak at position “b” can be decovoluted into two peaks at 2.8 ppm (HT) and 2.6 ppm (HH), and the ratio of the peak area [HT/(HT+HH)] can be used to estimate the head-tail (HT) regioregularity of P3HT in the nanocomposites [54]. The HT regioregularity of the P3HT decreases with increasing MWNT content in the composites, which might be caused by the interaction between MWNT and 3HT that makes the HT orientation less preferential during the polymerization process of 3HT. When MWNT are absent from the polymerization, the steric factors are the main driving force behind the coupling of 3HT monomers. The interaction of the 3HT monomers with the walls of MWNT seems to have greater inﬂuence on the coupling between 3HT than steric factors alone. The defects on the surface of MWNT might interrupt the HT coupling during the polymerization, accordingly leading to a decrease in the regioregularity of P3HT. 12.4.3 Mechanical Properties

Polymers composed of long molecular chains have unique viscoelastic properties, which combine the characteristics of elastic solids and Newtonian ﬂuids. The viscoelastic property of a polymer is studied by dynamic mechanical analysis where

311

312

12 P3HT–MWNT Nanocomposites by In-situ Polymerization and Their Properties a) f d b

g

e d, e, f

c

S

n P3HT

CHCl3

TMS

g

a

c

b

a PM-10 PM-5 PM-1 PM-0.5 PM-0.1 P3HT 7

3

2

1

0

ppm b)

H-T

H-T

PM-0.1

PM-1 73.66%

74.05%

H-H

H-H Pristine P3HT H-T

87.1% H-H

3.0 2.9 2.8 2.7 2.6 2.5 2.4 ppm H-T

PM-5

3.0 2.9 2.8 2.7 2.6 2.5 2.4 ppm H-T

PM-10 55.18%

67.18% 2.9

2.8

2.7 ppm

2.6

H-H H-H

3.0 2.9 2.8 2.7 2.6 2.5 2.4 ppm Figure 12.4 (a) 1H-NMR spectra for the

P3HT–MWNT nanocomposites and the pristine P3HT. (b) The deconvoluted NMR peaks at 2.6 and 2.8 ppm corresponding to

3.0

2.8

2.6 2.4 ppm

2.2

HH and HT regioregularity, respectively. (From Saini, V., et al. [2009] J. Phys. Chem. C 113[19], 8023–8029. Reproduced with permission from American Chemical Society.)

12.4 The Properties and Characterization of P3HT–MWNT Nanocomposites

a sinusoidal force (stress σ) is applied to a material and the resulting displacement (strain) is measured. The storage modulus measures the stored energy, representing the elastic portion, and the loss modulus measures the energy dissipated as heat, representing the viscous portion. The mechanical properties of ICP–CNT composites can be understood by investigating the temperature dependence of storage modulus, loss modulus, and tan(δ) plot of composites (δ is phase lag between stress and strain). The storage modulus or elastic modulus (G′) relates the ability of the material to store or return energy when an oscillatory force is applied to the sample, and the loss modulus (G″) relates the ability to lose the energy. Nandi et al. found that in P3HT–MWNT nanocomposites (named as PCNC-x, the number x indicates percentage of MWNT in the composite according to the literature [55]), the storage modulus of all the samples decrease with increasing temperature, and is negligibly small at about 50 °C (Figure 12.5a). However, the decrease is not linear, and two different types of transitions are clearly manifested in the loss modulus temperature plot (Figure 12.5b): the β transition (Tβ) where relaxation of hexyl side chains of P3HT occurs at the lower temperature (∼−90 °C), and R-transition arises for segmental relaxation of the main chain at the higher temperature (∼20 °C) which may be called the glass transition temperature (Tg) of the polymer [18]. In the tan(δ) verses temperature plot (Figure 12.5c) there are a broad peak at lower temperature and a sharp peak at higher temperature regions, corresponding to the Tβ and the Tg, respectively. The large difference in the transition temperatures obtained from the loss modulus and tan(δ) plots might be due to the two different modes of measurement: the dissipation of energy as heat (loss modulus) and the reduction of vibration of the material, that is, damping (tan[δ]). Both storage modulus and loss modulus values are higher in the composites than that of pure P3HT, except for the PCNC-8 where the loss modulus has lesser value at low temperature (≤0 °C). So the origins of increased storage modulus and loss modulus in the PCNCs are almost the same. The Tg values remain the same for all the nanocomposites, but the Tβ exhibits a peculiar behavior with the MWNT composition (inset of Figure 12.5b). The higher Tβ values of PCNC-1 than that of pure P3HT indicates the relaxation process of the hexyl group occurring through bond rotation is hindered. For the PCNC-8 sample, the same Tβ value was obtained as in the pure P3HT due to the balance of attractive forces from different directions on the hexyl chains for the larger concentration of MWNTs. Additionally, the percentage increase of storage modulus increases with increasing temperature. 12.4.4 Thermal Stability

The decomposition temperature is a direct measure of the thermal stability of the nanocomposite materials. The derivative thermogravimetric analysis proﬁle (DTA) (the inset of Figure 12.6a) shows two decomposition temperatures (Td1 and Td2) for all the P3HT nanocomposites. The third decomposition temperature (Tdm) assigned to MWNTs is visible only in the PM-5 and PM-10 samples. Td1 originates

313

314

12 P3HT–MWNT Nanocomposites by In-situ Polymerization and Their Properties a)

b)

c)

Figure 12.5 Mechanical property variation of

P3HT–MWNT nanocomposites with temperature: (a) storage modulus; (b) loss modulus (inset: Tβ with MWNT concentra-

tion); (c) tan (δ ). (From Kuila, B.K., et al. [2007] Macromolecules 40, 278. Reproduced with permission from American Chemical Society.)

12.4 The Properties and Characterization of P3HT–MWNT Nanocomposites

100

MWNT

80

PM-10 PM-5

60 P3HT PM-0.1 PM-1 PM-5 PM-10 MWNT

40 20 0 200

dW/dT (g/°C)

Weight (%)

a)

PM-1 PM-0.1 P3HT 200 400 600 Temperature (°C)

400 600 Temperature (°C)

800

b)

Temperature (°C)

500 480

main chain

460

330

side chain

325 320 0

8 2 4 6 MWNTs in P3HT (wt%)

Figure 12.6 (a) TGA and DTA (inset) proﬁles

of the pristine P3HT, P3HT–MWNT nanocomposites and puriﬁed MWNTs obtained at an air ﬂow rate of 150 ml min−1 from 25 to 850 °C. (b) Decomposition temperatures (Td) of P3HT side chain

10

(squares) and main chain (circles) in the P3HT–MWNT nanocomposites. (From Saini, V., et al. [2009] J. Phys. Chem. C 113[19], 8023–8029. Reproduced with permission from American Chemical Society.)

from the decomposition of side chains, whereas Td2 is due to the decomposition of thiophene rings (main chains) in the polymeric materials [56]. At low MWNT loadings (0.1% and 1%), Td1 only slightly decreases while Td2 increases as much as about 20 °C due to the interaction between main chains of P3HT and graphitic walls of MWNTs which increases the thermal stability of nanocomposites. As the MWNT concentration increases, a weaker interaction between the nanotubes and the polymers was observed due to the poorer nanotube dispersion in the polymer matrix as indicated by the TEM analysis (Figure 12.3). The composites with low MWNT loadings have therefore a better thermal stability as a result of the more uniform dispersion of the nanotubes. High MWNT concentrations would lead to MWNT agglomeration and weak interaction with the polymers. Additionally, the

315

316

12 P3HT–MWNT Nanocomposites by In-situ Polymerization and Their Properties Table 12.1 The characteristic combustion peaks and the differential scanning calorimetric

analysis of the P3HT–MWNT composites. Sample

P3HT PM-0.1 PM-1 PM-5 PM-10

HT regioregularity (%)

Td1

87 74 73 67 55

329 324 327 330 331

Td2

Melting temperature (°C)

Crystallization temperature (°C)

473 492 486 461 459

180.2 185.4 185.1 161.7 159.7

109.4 121.1 130.7 101 100.6

TGA

From Saini, V., et al. [2009] J. Phys. Chem. C 113[19], 8023–8029. Reproduced with permission from American Chemical Society). Note: (a) Tm1 and Tm2 are the melting temperatures of side chain and main chain, respectively, of the P3HT. (b) EF1 and EF2 are the enthalpy of fusion for the melting temperatures of side chain and main chain, respectively.

more nanotubes were introduced in the composites, the larger number of the oxygen-rich groups attached to the defects of the nanotube walls would affect the thermal stability. The NMR results also conﬁrmed the presence of a strong interaction between the nanotube walls and the aromatic rings (peak a − 6.9 ppm). As seen in Figure 12.4a, the change of the intensity of peak a has a similar trend to the Td2 values (Figure 12.6b) for various nanotube concentrations. The melting temperatures and crystallization temperatures were measured by the differential scanning calorimetry (DSC) data obtained for the pristine P3HT and P3HT–MWNT nanocomposites and puriﬁed MWNTs (Table 12.1). The melting temperatures and crystallization temperatures of the composites also conﬁrm that the composites prepared with low MWNT loading are more thermally stable. The probability of the P3HT interaction with the walls of more thermally stable CNTs is more via the main chain instead of side chains. The probability of this interaction decreases as the MWNT content is further increased in the polymer matrix because now the probability of the wall–wall interaction is more than that of the wall–chain interaction. 12.4.5 Optical Properties

Within materials science, the optical properties of ICP–CNT composites refer speciﬁcally to the absorption, photoluminescence, Raman and FTIR spectroscopy of ICP–CNT composites. Spectroscopic methods offer the possibility of quick and nondestructive characterization of relatively large amounts of ICP–CNT composites. There is a strong demand for such characterization from the industrial point

12.4 The Properties and Characterization of P3HT–MWNT Nanocomposites

of view: numerous parameters of the ICP–CNT composite synthesis can be changed, intentionally or unintentionally, to alter the ICP–CNT composite quality. As shown below, these spectroscopies allow quick and reliable characterization of this “ICP–CNT composite quality” in terms of nontubular carbon content, structure of the produced ICP–CNT composites, and structural defects. Those features determine nearly any other properties such as optical, mechanical, and electrical properties. a)

Optical Absorption: The UV-vis-NIR absorption spectra of the P3HT and P3HT–MWNT composite ﬁlms are shown in Figure 12.7a. Pristine P3HT

a)

Absorption

0.8 0.6

210 200 190 180 170 160 150 140 130

FWHM (nm)

P3HT PM-0.1 PM-1 PM-5 PM-10 MWNT

0

0.4

2 4 6 8 MWNT Conc. (wt%)

10

520 510 500 490 480 470 460 450

Peak Positions (nm)

1.0

0.2 0.0 300

400

500 600 700 800 Wavelength (nm)

900

1000

P.L Intensity (arb. unit)

b)

1000 1 3 500

2 4

0 600

700 Wavelength (nm)

Figure 12.7 (a) UV–vis–NIR optical absorp-

tion of the P3HT–MWNT nanocomposites prepared by in-situ polymerization. Inset is the peak positions (open triangles) and full width half maximum (solid squares) measured for the nanocomposites. (From Saini, V., et al. [2009] J. Phys. Chem. C 113[19], 8023–8029. Reproduced with permission from American

800 Chemical Society.) (b) Photoluminescence spectra of P3HT–MWNT nanocomposites after excitation by radiation of a 500 nm wavelength (normalized to ﬁlm thickness of 1 μm): (i) P3HT; (ii) PCNC-1; (iii) PCNC-2.5; (iv) PCNC-8. (From Kuila, B.K., et al. [2007] Macromolecules 40, 278. Reproduced with permission from American Chemical Society.)

317

318

12 P3HT–MWNT Nanocomposites by In-situ Polymerization and Their Properties

has absorption band at 505 nm corresponding to the π–π* transition of its conjugated segments. Addition of MWNTs gradually shifts this band toward the higher wavelength region (the inset of Figure 12.7a, right axis), suggesting a decrease in band gap because of the interaction between P3HT and nanotubes, which uncoils the P3HT chains, thereby increasing their conjugation length [15]. The spin-coated P3HT molecules from the chloroform solution form a homogeneous extension of conjugation across the whole ﬁlm [57]. The absorption peak broadens with increasing MWNT content (the inset of Figure 12.7a, left axis), suggesting a change in the morphology of spin-coated ﬁlms. Although the ﬁlms were prepared from the same weight percentage of polymer nanocomposite in chloroform (10 mg ml−1), the level of absorption is different for all the MWNT compositions, which might be caused by a variation in the weight composition of P3HT in CHCl3. The nanocomposite solutions prepared with low MWNT concentrations (PM-1, PM-0.1, and pristine P3HT) were soluble. The PM-5 solution was partly soluble with blackish brown color, whereas the PM-10 solution was light orange in color with a lot of insoluble material. The higher absorption levels in PM-5 and PM-10 over the whole spectrum was due to the high MWNT loadings. b) Photoluminescence Spectra: P3HT has a photoluminescence property [19], and the photoluminescence spectra of P3HT and composite ﬁlms, normalized to its thickness, are presented in Figure 12.7b. For an excitation wavelength of 490 nm the ﬁlms show an emission at 653, 649, and 651 nm for P3HT, PCNC-1, and PCNC-8, respectively. Here, the PCNCs also exhibit a photoluminescence quenching. The photoluminescence quenching of composites may be related to the π–π interaction of P3HT with MWNT [52], forming additional decaying paths of the excited electrons through the MWNTs, and the greater the MWNT concentration, the larger is the photoluminescence quenching. The small blue shift in the nanocomposite emission spectra indicates that the ground state energy level is more stable in the nanocomposites than that of pure P3HT. This may be possible through the resonance stability of π clouds of P3HT and graphite units of MWNT through the π–π interaction. c)

Raman Scattering from P3HT–MWNT Composites: The Raman spectra of pure P3HT and MWNTs as well as P3HT–MWNT composites synthesized by Musumeci et al. [58] are presented in Figure 12.8. The larger nanotube diameters and the wider diameter distributions inherently present in the MWNTs prevent the observation of the nanoscale phenomena found in SWNTs. The two main features in the Raman spectra are the D and G peaks at approximately 1350 and 1581 cm−1, respectively [59]. The G-band originates from the splitting of the E2g in graphite-like materials and is a characteristic mode of the SWNTs, while the D-band corresponds to disordered carbon structures. The relative intensity of the D- and G-bands (ID/IG) decreases with increasing excitation energies. The presence of the G-band and the dispersive

12.4 The Properties and Characterization of P3HT–MWNT Nanocomposites

Figure 12.8 Raman spectra of pure the MWNT and P3HT as well as P3HT–MWNT

composites of various compositions. (From Musumeci, A.W., et al. [2007] Polymer 48, 1667–678. Reproduced with permission from Elsevier Ltd.)

behavior of the D-band, together with fairly low value of half height peak width (<40 cm−1), indicates a high degree of graphitization of the MWNTs. The native poly(3-hexylthiophene) spectrum features all vibrational frequencies expected for the conjugated polymer [60]. The band at 599 cm−1 is due to in-plane thiophene ring deformation [61]. The Raman shift observed in the frequency range of 678 and 728 cm−1 are the symmetric and antisymmetric Cα–S–Cα bending deformation vibrations in the thiophene ring, respectively [62–65]. Interestingly all composite samples show upshift in both the G- and D-band frequencies due to an increase in the carbon nanotube C–C bond strength. Wise et al. [66] reported the effect of electron donor–acceptor type charge

319

320

12 P3HT–MWNT Nanocomposites by In-situ Polymerization and Their Properties

addition to SWNTs with respect to shifts in characteristic Raman active vibrational modes. Speciﬁcally, removing charge from a SWNT leads to an upshift in the G-band. A downshift in the Raman D-band was attributed to additional electron density in the SWNT’s antibonding orbitals from charge injection causing a weakened average C–C bond length. Thus, the MWNTs might donate electronic charge to the polymer matrix. Otherwise, a similar upshift of 10–15 cm−1 was observed in the Raman spectra of poly(butadiene)–MWNT composites [67]. The CH–π interactions observed between nanotube and polymer are stronger than that of the π–π interactions observed between nanotube bundles, resulting in a restriction of the C–C bond vibrations and a corresponding upshift of the Raman signal. A 17 cm−1 upshift in G-band Raman signal of MWNTs embedded in melt-blended polyethylene– MWNT composites and the evolution of a shoulder to this peak were attributed to compressive forces exerted on the MWNTs by polyethylene chains following intercalation into MWNT bundles. So, the proposed compressioninduced effect on MWNT Raman G-band position appears to be consistent with the results obtained for rrP3HT–MWNT composites. d) FTIR Properties: Fourier transform infrared (FTIR) spectra of P3HT and P3HT–MWNT composites are presented in Figure 12.9. The average conjuga-

Figure 12.9 FTIR spectra of P3HT and

P3HT–MWNT composites. Inset: relative intensity of C=C antisymmetric to symmetric stretching indicating an increase in the conjugation length. The uneven baseline, observed for the 17 wt% MWNT composite, caused by the sample strongly absorbing

incident radiation, prevents the evaluation of the ratio of relative intensities given in the inset for other compositions. (From Musumeci, A.W., et al. [2007] Polymer 48, 1667–1678. Reproduced with permission from Elsevier Ltd.)

12.4 The Properties and Characterization of P3HT–MWNT Nanocomposites

tion length of P3HT can be probed by using the ratio between the intensity of the antisymmetric C=C stretching peak (mode at 1509 cm−1) and the intensity of the symmetric stretching peak (mode at 1456 cm−1) [68, 69]. The P3HT conjugation length was found to increase with increasing MWNT content as shown in the inset of Figure 12.9. Considering that TEM image analysis of diluted composite ﬁlms was limited to a small number of P3HT layers deposited upon the MWNT surface, the result of FTIR for the bulk composite ﬁlms, on the other hand, indicates that the P3HT interfacial arrangement on the MWNT ﬁller has consequences on the structure of the bulk material as a whole. 12.4.6 Charge Transportability

CNTs in polymer–CNT composites are efﬁciently debundled and isotropically dispersed in polymer matrices, the efﬁcient interaction between CNT and polymer provides good dispersion and a low percolation threshold, but only relatively low conductivity near and above percolation, frequently around 10−4 s cm−1 is achieved at close to 2 wt% CNT loading [70, 71]. The polymer layer in the internanotube connections is supposed to be the highest resistance section in the electrical pathway. This polymer layer is a barrier to efﬁcient carrier transport between CNTs, and models for conductivity based on ﬂuctuation-induced tunneling have been proposed [72]. A power law related to percolation theory can be used to model conductivity in the following form:

σ = σ 0 ( p − pc ) for p > pc , t

(12.1)

where pc is the critical mass or percolation threshold and the exponent t should be close to 2 for a three-dimensional percolating system [73]. The results obtained from DC conductivity of P3HT–MWNT at 90 °C are pc = 0.1 wt% and t = 1.68. The low value of t has been interpreted to be associated with a ﬂuctuation-induced tunneling model. The P3HT–MWNT composites with high MWNT concentration display a frequency-independent conductivity in the range of low frequency and a tendency to increase at higher frequencies (Figure 12.10). In contrast to a typical dielectric which should exhibit a linear increase of conductivity with frequency with a slope of unity in a log–log plot, the conductivity as a function of frequency for the diluted composites shows a modulation in the low frequency range, which can be assigned to the P3HT contribution to the total conductivity. Conjugated polymers and insulating polymers are often treated in a similar way within the ﬁeld of ICP–CNT composites by applying percolation treatments. A conjugated polymer might be expected to facilitate electron transport in a composite. Conjugated polymer-based nanocomposites have presented lower levels of conductivity than insulating polymers after percolation in some works [74]. Conjugated polymer contribution to the electrical and optical properties of a P3HT–MWNT composite has been used to prepare solar cells [75]. Therefore, the results of AC and DC conductivities

321

322

12 P3HT–MWNT Nanocomposites by In-situ Polymerization and Their Properties

Figure 12.10 Real part of AC conductivity as a function of frequency for P3HT–MWNT. Compositions are indicated in the ﬁgure. (From Musumeci, A.W., et al. [2007] Polymer 48, 1667–1678. Reproduced with permission from Elsevier Ltd.)

obtained here for P3HT–MWNT composites can be considered as a combination of CNT network conductivity and conjugated polymer charge transport playing different roles in different regions of the MWNT concentration. For the low concentration composites, before percolation, the conductivity around 10−6 s cm−1 and frequency dependence are associated with charge transfer through the semiconductor polymer. There are three different regions on the conductivity changes: before percolation, after percolation and up to 4 wt%, and higher concentrated samples. A polymer conjugation length increases steadily with the addition of 0.1 wt% MWNT and stays approximately constant for more concentrated samples. The structural studies indicate an increase of long range order in the 1 wt% composite and an

Normailzied conductivity (dT/d300K)

12.4 The Properties and Characterization of P3HT–MWNT Nanocomposites 100

MWNT PM-10 PM-1 P3HT

10–1

10–2

10–3 100

150

200 250 Temperature (K)

Figure 12.11 The normalized conductivity σ(T )/σ(300 K) of the P3HT–MWNT composites with different concentrations of MWNTs above the percolation threshold. (From Saini,

300

V., et al. [2009] J. Phys. Chem. C 113(19), 8023–8029. Reproduced with permission from American Chemical Society.)

increase of mesophase self-organization upon further nanotube addition. This trend may contribute to the enhancement of composite conductivity. The conductivity (approximately 10 s cm−1) of an MWNT mat is limited by the tube–tube contact resistance [76]. Therefore, the maximum conductivity levels expected for composites, where a polymer coating is located between MWNTs, will depend on the type of polymer and coating thickness between tubes. The interplay of CNT dispersion and contact between tubes is the central key to take full advantage of the nanometric scale of the ﬁller. The use of self-ordered rod-like P3HT connections between tubes allowed the enhancement of charge transfer in the nanocomposite by three or four orders of magnitude when compared with other conjugated ICP–CNT composites. The behavior of the DC conductivity at low temperatures is an important indicator for the conduction mechanism. The thermal effects on the conductance of the P3HT–-MWNT composite ﬁlms are shown in Figure 12.11 and the temperature dependence of conductance can be expressed in the following formula [77]:

σ (T ) = σ m eTm / T + σ t e −Tb /(Ts +T ) + σ 0e − (T0 / T )

1/2

(12.2)

where the geometrical factors σm, σt, and σ0 can be treated as constants. The ﬁrst term is for quasi-1D metallic conduction along the tube axis direction, where phonons of energy kBTm (that have wave vectors 2kF spanning the Fermi surface, and kB is the Boltzmann constant) are required to backscatter carriers [78]. The second term corresponds to ﬂuctuation-assisted tunneling [79] through barriers, while the order of magnitude of typical energies reﬂected by the value of kBTb and

323

324

12 P3HT–MWNT Nanocomposites by In-situ Polymerization and Their Properties

extent of the decrease of conductivity at low temperatures is indicated by the ratio Ts/Tb. The third term represents hopping between mesoscopic metallic islands of conducting tubes separated by insulating ones, following Mott’s variable-range hopping law for disordered semiconductors [80]. T0 is a constant depending on the localization length and density of states. These three mechanisms in parallel contribute to the conductivity-temperature dependence of our P3HT–CNT composite ﬁlms. The conductivity of the P3HT exhibits strong temperature dependence and nonmetallic behavior (Figure 12.11), suggesting the presence of tunneling barriers at the homogeneous junctions, with much stronger suppression of electron tunneling, so the metallic conductance mechanism (the ﬁrst term in Eq. (12.2)) can be neglected, and the variable-range hopping contributes more to the conductance than the thermal ﬂuctuation-assisted tunneling does. This argument is based on the fact that the polymer mat is a network of dispersed bundle–bundle junctions. These junctions between the chain bundles act as a gate for many charge carriers. Therefore, the variable-range hopping conduction takes place either within the chain bundles and/or between the chain bundles through their contacts. Since the polymer mat is a 3D array of 1D linear molecules, a 3D variable-range hopping mechanism is also expected. Solution processed ﬁlms (solution cast or spin-coating) of P3HT result in a lamellar structure of P3HT molecules with two-dimensional conjugated sheets formed by interchain stacking. Orientation of lamellar parallel or perpendicular to a substrate depends on molecular weight, regioregularity, and process conditions [81]. The 2D ordered π–π stacking results in improved mobility along the stacking direction [13, 82]. The high-molecular-weight P3HT forms small-ordered domains, which are interconnected by long P3HT backbone chains to prevent the charge carrier from being trapped by disordered boundaries. In ordered domains, the charge transport parallel to the ﬁlm surface occurs via fast carrier hopping (π–π interaction hoping), and interchain hopping through parallel-oriented long chains to interconnect these domains. A MWNT usually has a much larger diameter than a SWNT, and intershell transfer can occur between concentric shells of MWNT [83]. These lead to much weaker localization in a MWNT than in a SWNT and lessen the importance of the quasi-1D suppression of phonon backscattering in Eq. (12.2) at high temperature. The tunneling effect is so strong that the quasi-1D metallic conduction is entirely suppressed in a MWNT. Additionally, in a MWNT, the thermal expansion in the directions perpendicular to the tube axis is strictly limited by the expanded diameter of the outermost shell of the MWNT for a particular temperature. So the contribution of the ﬁrst metallic term in Eq. (12.2) to the conductivity of MWNT ﬁlms is negligible, especially at T > 40 K [84]. For the SWNT ﬁlm, the variablerange hopping mechanism still plays an important role in the conductancetemperature dependence. As the wall number further increases, the dominating conduction mechanism shifts from the variable-range hopping to the ﬂuctuationassisted tunneling which governs the weak conductivity-temperature dependence of the MWNT network.

12.5 Conclusion and Outlook

In the polymer–nanotube composites, most of the nanotubes seem to be straight and separated by a thin layer of an insulating polymer [85], which acts as a potential barrier to internanotube hopping. As a result, the polymer–nanotube composite conduction can be explained by quantum tunneling in which the barrier height decreases with increasing temperature [86]. This behavior is described by the thermal ﬂuctuation-induced tunneling model [29]. The conductivity of the P3HT polymer depends on the HT regioregularity of the chain; the higher the regioregularity, the higher the conductivity [87]. According to the NMR analysis, the addition of MWNT can alter the HT regioregularity of the chain and as a consequence the polymer conductivity changes. Additionally, as the MWNT concentration in the composite increases, the conductivity-temperature dependence of the P3HT–MWNT networks gets weaker due to the increased contribution from the ﬂuctuation-assisted tunneling.

12.5 Conclusion and Outlook

These extraordinary properties of isolated carbon nanotubes greatly excite the immense potential of these nanoﬁllers. The combination of these unique properties of CNTs with various functional polymers offers many opportunities for research in chemistry, physics, and materials science that will produce novel materials with unusual electrical, magnetic, and optical properties. P3HT-MWNT nanocomposites can be synthesized by the in-situ oxidative polymerization of 3-hexylthiophene (3HT) in a dispersion of MWNTs in CHCl3. The combination of the above studies can be summarized as follows: (i) MWNT is efﬁciently wetted by P3HT; (ii) conjugation length of P3HT increases with MWNT addition; (iii) restriction of the C–C bond vibrations was observed on the MWNT through the polymer coating; (iv) P3HT crystallinity decreases with increasing MWNT concentration after 1 wt% addition. In P3HT–MWNT nanocomposites, P3HT uniformly wraps on the graphitic walls of MWNTs because of the presence of delocalized covalent bonds in both P3HT and MWNTs. Both CH–π and π–π interactions between P3HT and MWNT decrease in HT regioregularity on addition of MWNTs. The P3HT looses its solubility due to the addition of MWNTs and subsequent reduction of the HT regioregularity of the polymer. The thickness of polymer layer on the MWNT walls decreases with the increase of MWNT amount in P3HT. The addition of MWNTs increases the thermal stability of the polymer up to certain loading levels (PM-0.1 and PM-1). Despite the decrease in HT regioregularity, there is a strong interaction between the polymer chains and graphitic walls of MWNTs, which leads to better thermal stability of the composites. The red-shift in the UV-vis optical spectrum on addition of MWNTs is probably due to the interaction between P3HT and MWNTs which increases the conjugation length of the polymeric chains. Bulk conductivity of P3HT/MWNT nanocomposites reached a percolation threshold of only 0.1 wt% of MWNTs. The optimal MWNT concentration is about 1 wt% for the best P3HT–MWNT

325

326

12 P3HT–MWNT Nanocomposites by In-situ Polymerization and Their Properties

interaction. Rather than behaving as a simple insulating matrix, the conjugated polymer was observed to inﬂuence the conductivity mechanism, allowing efﬁcient charge transfer across the internanotube connections for highly concentrated samples. The temperature-dependent conductivity revealed that increasing MWNT content in P3HT–MWNT composites can shift the conduction mechanism from variable-range hopping to ﬂuctuation-assisted tunneling. Therefore, the increased conductivity in P3HT–MWNT composites can ﬁnd applications in optoelectronics easily, such as organic photovoltaic devices. Note that all current CNT preparation approaches give mixtures of nanotube chiralities, diameters, wall numbers, and lengths along with different amount of impurities and structural defects. This CNT heterogeneity varies signiﬁcantly from different batches and laboratories, accordingly makes it very difﬁcult to reproduce control experiments with these inconsistent nanoﬁllers and virtually impossible to compare results between different groups. Another great challenge in polymernanotube composites is to efﬁciently translate nanotube properties both into the polymer matrix and between nanotubes. Although some signiﬁcant insights have been achieved, there are still many unresolved issues that need to be solved theoretically and experimentally to harness the maximum beneﬁts from carbon nanotubes in polymer composites. P3HT–MWNT composites offer both great potential and great challenges, marking it as a vibrant area of work for years to come. As we advance in CNT growth, puriﬁcation and separation, as well as novel development of ICP-CNT composite synthesis, we will be able to more effectively handle the challenges, and produce these composites with desirable properties for speciﬁc applications. The signiﬁcant progress in the understanding of these composite systems in the past points toward a bright future.

References 1 Gao, G., Cagin, T., and Goddard, W.A., 2

3

4 5

6

7

III. (1998) Nanotechnology, 9, 184. Cooper, C.A., Young, R.J., and Halsall, M. (2001) Composites, Part A, 32A, 401. McEuen, P.L., Bockrath, M., Cobden, D.H., Yoon, Y.-G., and Louie, S.G. (1999) Phys. Rev. Lett., 83, 5098. Berber, S., Kwon, Y.-K., and Tomanek, D. (2000) Phys. Rev. Lett., 84, 4613. Niu, C.M., Sichel, E.K., Hoch, R., Moy, D., and Tennent, H. (1997) Appl. Phys. Lett., 70, 1480. Yeh, J.M., Liou, S.J., Lai, C.Y., Wu, P.C., and Tsai, T.Y. (2001) Chem. Mater., 13, 1131. do Nascimento, G.M., Constantino, V.R.L., Landers, R., and Temperini,

8

9

10

11

12

M.L.A. (2004) Macromolecules, 37, 9373. Sainz, R., Benito, A.M., Martínez, M.T., Galindo, J.F., Sotres, J., Baró, A.M., Corraze, B., Chauvet, O., and Maser, W.K. (2005) Adv. Mater., 17, 278. Halls, J.J.M., Pichler, K., Friend, R.H., Moratti, S.C., and Holmes, A.B. (1996) Appl. Phys. Lett., 68, 3120. Bhattacharyya, S., Kymakis, E., and Amaratunga, G.A.J. (2004) Chem. Mater., 16, 4819. Yang, H., Shin, T.J., Yang, L., Cho, K., Ryu, C.Y., and Bao, Z. (2005) Adv. Funct. Mater., 15, 671. Surin, M., Leclere, P., Lazzaroni, R., Yuen, J.D., Wang, G., Moses, D., et al.

References

13 14

15 16

17 18 19 20

21 22 23

24 25

26 27

28 29

30

31

32

(2006) J. Appl. Phys., 100, 033712/1–033712/6. Sirringhaus, H., Tessler, N., and Friend, R.H. (1998) Science, 280, 1741. Padinger, F., Rittberger, R.S., and Sariciftci, N.S. (2003) Adv. Func. Mater., 13, 85. Roncali, J. (1992) Chem. Rev., 92, 711. McCullough, R.D., and Ewbank, P.C. (1998) Handbook of Conducting Polymer (eds T.A. Skotheim, R.L. Elsenbaumer, and J.R. Renolds), Marcel Dekker Inc, NewYork. Chen, T.-A., Wu, X., and Rieke, R.D. (1995) J. Am. Chem. Soc., 117, 233. Kuila, B.K. and Nandi, A.K. (2004) Macromolecules, 37, 8577. Kuila, B.K. and Nandi, A.K. (2006) J. Phys. Chem. B, 110, 1621. Hughes, M., Chen, G.Z., Shaffer, M.S.P., Fray, D.J., and Windle, A.H. (2002) Chem. Mater., 14, 1610. Jiang, X., Bin, Y., and Matsuo, M. (2005) Polymer, 46, 7418. Kong, H., Gao, C., and Yan, D. (2004) J. Am. Chem. Soc., 126, 412. Awasthi, K., Srivastava, A., and Srivastava, O.N. (2005) J. Nanosci. Nanotechnol., 5, 1616. Monthioux, M. (2002) Carbon, 40, 1809. Thostenson, E.T., Ren, Z.F., and Chou, T.W. (2001) Compos. Sci. Technol., 61, 1899. Dyke, C.A. and Tour, J.M. (2004) J. Phys. Chem. A, 108, 11151. Forro, L. and Schonenberger, C.H. (2001), Carbon Nanotubes (eds M.S. Dresselhaus, G. Dresselhaus, and P.H. Avouris), Springe, Berlin. Ebbesen, T.W. and Ajayan, P.M. (1992) Nature, 358, 220. Ando, Y., Zhao, X., Inoue, S., Suzuki, T., and Kadoya, T. (2005) Diamond Related Mater., 14, 729. Guo, T., Nikolaev, P., Thess, A., Colbert, D.T., and Smalley, R.E. (1995) Chem. Phys. Lett., 243, 49. Maser, W.K., Munoz, E., Benito, A.M., Martinez, M.T., de la Fuente, G.F., Maniette, Y., Anglaret, E., and Sauvajol, J.-L. (1998) Chem. Phys. Lett., 292, 587. Endo, M., Takeuchi, K., Igarashi, S., Kobori, K., Shiraishi, M., and Kroto,

33

34 35

36

37

38 39

40

41

42

43

44 45

46

47

H.W. (1993) J. Phys. Chem. Solids, 54, 1841. Hernadi, K., Konya, Z., Siska, A., Kiss, J., Oszko, A., Nagy, J.B., and Kiricsi, I. (2002) Mater. Chem. Phys., 77, 636. Kong, J.A., Cassell, A.M., and Dai, H. (1998) Chem. Phys. Lett., 292, 567. Nikolaev, P., Bronikowski, M.J., Bradley, R.K., Rohmund, F., Colbert, D.T., Smith, K.A., and Smalley, R.E. (1999) Chem. Phys. Lett., 313, 91. Dai, H., Rinzler, A.G., Nikolaev, P., Thess, A., Colbert, D.T., and Smalley, R.E. (1996) Chem. Phys. Lett., 260, 471. Zheng, L.X., O’Connell, M.J., Doorn, S.K., Liao, X.Z., Zhao, Y.H., Akhadov, E.A., Hoffbauer, M.A., Roop, B.J., Jia, Q.X., Dye, R.C., Peterson, D.E., Huang, S.M., Liu, J., and Zhu, Y.T. (2004) Nat. Mater., 3, 673. Hirsch, A. (2002) Angew. Chem. Ind. Ed., 41, 1853. Baibarac, M., Baltog, I., Lefrant, S., Mevellec, J.Y., and Chauvet, O. (2003) Chem. Mater., 15, 4149. Lefrant, S., Baltog, I., Baibarac, M., Mevellec, J.Y., and Chauvet, O. (2002) Carbon, 40, 2201. Regev, O., ElKati, P.N.B., Loos, J., and Koning, C.E. (2004) Adv. Mater., 16, 248. Dufresne, A., Paillet, M., Putaux, J.L., Canet, R., Carmona, F., Delhaes, P., and Cui, S. (2002) J. Mater. Sci., 37, 3915. Vigolo, B., Penicaud, A., Coulon, C., Sauder, C., Pailler, R., Journet, C., Bernier, P., and Poulin, P. (2000) Science, 290, 1331. Badaire, S., Poulin, P., Maugey, M., and Zakri, C. (2004) Langmuir, 20, 10367. Sundararajan, P.R., Singh, S., and Moniruzzaman, M. (2004) Macromolecules, 37, 10208. De la Chapelle, M.L., Stephan, C., Nguyen, T.P., Lefrant, S., Journet, C., Bernier, P., Munoz, E., Benito, A., Maser, W.K., Martinez, M.T., De la Fuente, G.F., Guillard, T., Flamant, G., Alvarez, L., and Laplaze, D. (1999) Synth. Met., 103, 2510. Benoit, J.M., Corraze, B., Lefrant, S., Blau, W.J., Bernier, P., and Chauvet, O. (2001) Synth. Met., 121, 1215.

327

328

12 P3HT–MWNT Nanocomposites by In-situ Polymerization and Their Properties 48 Philip, B., Xie, J., Chandrasekhar, A.,

49

50

51 52

53

54

55

56

57 58

59

60

61

62

63

Abraham, J., and Varadan, V.K. (2004) Smart Mater. Struct., 13, 295. Zhou, W., Islam, M.F., Wang, H., Ho, D.L., Yodh, A.G., Winey, K.I., and Fischer, J.E. (2004) Chem. Phys. Lett., 384, 185. Saini, V., Li, Z., Bourdo, S., Dervishi, E., Xu, Y., Ma, X., Kunets, V.P., Salamo, G.J., Viswanathan, T., Biris, A.R., Saini, D., and Biris, A.S. (2009) J. Phys. Chem. C, 113, 8023. Kymakis, E. and Amaratunga, G.A.J. (2006) J. Appl. Phys., 99, 084302. Chen, J., Liu, H., Weimer, W.A., Halls, M.D., Waldeck, D.H., and Walker, G.C. (2002) J. Am. Chem. Soc., 124, 9034. Hill, D.E., Lin, Y., Rao, A.M., Allard, L.F., and Sun, Y.P. (2002) Macromolecules, 35, 9466. Amou, S., Haba, O., Shirato, K., Hayakawa, T., Ueda, M., Takeuchi, K., and Asai, M. (1999) J. Polym. Sci. A, 37, 1943. Kuila, B.K., Malik, S., Batabyal, S.K., and Nandi, A.K. (2007) Macromolecules, 40, 278. Adhikari, A.R., Huang, M., Bakhru, H., Chipara, M., Ryu, C.Y., and Ajayan, P.M. (2006) Nanotechnology, 17, 5947. Kobashi, M. and Takeuchi, H. (1998) Macromolecules, 31, 7273. Musumeci, A.W., Silva, G.G., Liu, J.-W., Martens, W.N., and Waclawik, E.R. (2007) Polymer, 48, 1667. McNally, T., Potschke, P., Halley, P., Murphy, M., Martin, D., Bell, S.E.J., et al. (2005) Polymer, 46, 8222. Osterbacka, R., An, C.P., Jiang, X.M., and Vardeny, Z.V. (2001) Synth. Met., 116, 317. Louarn, G., Trznadel, M., Buisson, J.P., Laska, J., Pron, A., Lapkowski, M., and Lefrant, S. (1996) J. Phys. Chem., 100, 12532. Baibarac, M., Lapkowski, M., Pron, A., Lefrant, S., and Baltog, I. (1998) J. Raman Spectrosc., 29, 825. Pron, A., Louarn, G., Lapkowski, M., Zagorska, M., Glowczy-Zubek, J., and Lefrant, S. (1995) Macromolecules, 28, 4644.

64 Klimov, E., Li, W., Yang, X., Hoffmann,

65

66

67

68 69

70

71

72

73

74

75

76

77 78 79 80 81

82

G.G., and Loos, J. (2006) Macromolecules, 39, 4493. Louarn, G., Buisson, J.P., Lefrant, S., and Fichou, D. (1995) J. Phys. Chem., 99, 11399. Wise, K.E., Park, C., Siochi, E.J., and Harrison, J.S. (2004) Chem. Phys. Lett., 391, 207. Baskaran, D., Mays, J.W., and Bratcher, M.S. (2005) Chem. Mater., 17, 3389. Furukawa, Y., Akimoto, M., and Harada, I. (1987) Synth. Met., 18, 151. Trznadel, M., Pron, A., Zagorska, M., Chrzaszcz, R., and Pielichowski, J. (1998) Macromolecules, 31, 5051. Du, F., Fischer, J.E., and Winey, K.I. (2005) Phys. Rev. B, 72, 121404/1–121404/4. Chatterjee, T., Yurekli, K., Hadjiev, V.G., and Krishnamoorti, R. (2005) Adv. Funct. Mater., 15, 1832. Kilbride, B.E., Coleman, J.N., Fraysse, J., Fournet, P., Cadek, M., Drury, A., et al. (2002) J. Appl. Phys., 92, 4024. Barrau, S., Demont, P., Peigney, A., Laurent, C., and Lacabanne, C. (2003) Macromolecules, 36, 5187. Kymakis, E. and Amaratunga, G.A.J. (2006) J. Appl. Phys., 99, 084302/1–084302/7. Canestraro, C.D., Schnitzler, M.C., Zarbin, A.J.G., da Luz, M.G.E., and Roman, L.S. (2006) Appl. Surf. Sci., 252, 5575. de Heer, W.A., Bacsa, W.S., Chatelain, A., Gerﬁn, T., Humphrey-Baker, R., Forro, L., et al. (1995) Science, 268, 845. Kaiser, A.B. (2001) Rep. Prog. Phys., 64, 1. Pietronero, L. (1983) Synth. Met., 8, 225. Sheng, P. (1980) Phys. Rev. B, 21, 2180. Ehinger, K. and Roth, S. (1986) Philos. Mag. B, 53, 301. Sirringhaus, H., Brown, P.J., Friend, R.H., Nielsen, M.M., Bechgaard, K., Langeveld-Voss, B.M.W., Spiering, A.J.H., Janssen, R.A.J., Meijer, E.W., Herwig, P., and de Leeuw, D.M. (1999) Nature, 401, 685. Bao, Z., Dodabalapur, A., and Lovinger, A.J. (1996) Appl. Phy. Lett., 69, 4108.

References 83 Bourlon, B., Miko, C., Forró, L., Glattli,

86 Kim, H.M., Kim, K., Lee, C.Y., Joo, J.,

D.C., and Bachtold, A. (2004) Phys. Rev. Lett., 93, 176806. 84 Skakalova, V., Kaiser, A.B., Woo, Y.S., and Roth, S. (2006) Phys. Rev. B, 74, 085403. 85 Kymakis, E., Alexandrou, I., and Amaratunga, G.A.J. (2003) J. Appl. Phys., 93, 1764.

Cho, S.J., Yoon, H.S., Pejakovi, D.A., Yoo, J.W., and Epstein, A. (2004) Appl. Phys. Lett., 84, 589. 87 Pal, S., Roy, S., and Nandi, A.K. (2005) J. Phys. Chem. B, 109, 18332.

329

331

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties Ranya Simons, Greg G. Qiao, and Stuart A. Bateman

13.1 Introduction

Polymer–montmorillonite (MMT) nanocomposites have been widely investigated over the last two decades to understand and control the signiﬁcant improvements in properties that can be achieved at relatively low concentrations of ﬁller [1, 2]. Polystyrene (PS) is an excellent model for free radical polymerized thermoplastic polymers which has lead to its use in numerous morphological, rheological, and modeling studies [3–8]. Although PS–MMT nanocomposites have been prepared through melt intercalation techniques and associated mixing methodologies, the best dispersed systems with the largest improvements in properties have been achieved using in-situ polymerization techniques [8]. The choice of modifying surfactant, experimental conditions, and polymerization strategy has been found to be crucial in determining the ﬁnal morphology of the nanocomposite and hence its properties. This chapter will review the studies performed for in-situ polymerized PS–MMT nanocomposite systems.

13.2 Morphology of Polymer–Clay Nanocomposites

The improvement in properties of polymer–clay nanocomposites is known to rely on the morphology or dispersion of the clay platelets in the polymer. Three morphologies are possible depending on the level of interaction between the clay and the polymer, and examples of these are shown in Figure 13.1 [2]. If there is very little interaction and hence compatibility between the clay and the polymer, a conventional microcomposite is formed, and the improvement in properties in this case is typically low (Figure 13.1a). If polymer chains can penetrate between the clay platelets and increase their interlayer distance an intercalated morphology forms and the improvements in properties are typically moderate (Figure 13.1b). If the interaction between the clay and the polymer are very good, the clay platelets can become completely separated leading to exfoliated structures (Figure 13.1c). In-situ Synthesis of Polymer Nanocomposites, First Edition. Edited by Vikas Mittal. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA.

332

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties a)

b)

c)

Figure 13.1 TEM images of (a) microcomposite, (b) intercalated nanocomposite, and

(c) exfoliated nanocomposite. Reproduced from Ref. [2] with permission.

Exfoliated structures tend to exhibit the largest improvements in properties due to the dimensions of individual clay platelets (1 nm thick and 100 nm in length) and thus the high surface area in which the ﬁller can interact with the polymer. Much of the work in the ﬁeld of polymer clay nanocomposites has focused on synthesizing exfoliated structures, and this can be affected through the modiﬁcation of the clay or polymer, the processing conditions, and the dispersion methods employed.

13.3 Modiﬁcation of MMT

MMT has been established as a successful ﬁller for polymers because it is cheap, plentiful in nature, and if the chemistry of the surface of the MMT and the polymer matrix is favorable, the micron-sized clay tactoids can be dispersed into the nanometer-sized clay platelets (exfoliation), effectively forming a nanocomposite. It is often difﬁcult to achieve complete exfoliation of a layered silicate in a continuous polymer matrix, due to the strong electrostatic attractions between the silicate layers and the intergallery cations. Modifying the clay with organic molecules helps to change the surface polarity of the clay and if designed correctly, improves the interaction of the polymer with the clay and allows good clay dispersion in the polymer. Every polymer matrix has a different organic modiﬁcation and dispersal method which is most appropriate for exfoliation. Liu summarized the different polymer clay modiﬁcations possible in his review [9]. In general, organic modiﬁcation used for clay can be classiﬁed into organics that are (i) ionically bonded by the cation exchange process, (ii) physically adsorbed, and (iii) covalently bonded by grafting [10]. The inherently hydrophilic MMT is not compatible with the hydrophobic styrene, and so for PS–MMT nanocomposites, clay modiﬁcation is essential. A large number of different modiﬁcations have been investigated for PS–MMT

13.3 Modiﬁcation of MMT

nanocomposites, ranging from nonreactive cations, reactive cations, cations incorporating initiator or catalyst functionality as well as covalently bonded silane modiﬁcations. The surfactants and resultant morphologies of the differently modiﬁed clays investigated for PS–MMT nanocomposites are listed in Tables 13.1–13.3. The choice of the organic modiﬁcation is probably the most important factor in order to achieve good dispersion in PS nanocomposites prepared through in-situ polymerization [8]. 13.3.1 NonReactive Modiﬁcations

The most common method to chemically modify clay is to ion-exchange the clay with nonreactive surfactants which cannot react with a styrene monomer (Figure 13.3), such as alkyl ammonium ions [4], and the nonreactive surfactants that have been used to form PS–MMT nanocomposites are listed in Table 13.1. Modiﬁed clays based on alkyl ammonium modiﬁcations are available commercially from Nanocor and Southern Clay Products. The alkyl ammonium ions used to modify MMT for use in PS nanocomposites normally contain one or two long alkyl chains attached to the ammonium ion (structures 1, 3, 4, 15, 21 in Table 13.1) [6, 11, 12, 14–19, 63, 64]. The alkyl chain portion increases the intergallery distance of the clay, which allows the styrene to more easily swell into the clay layers. The polarity of the clay surface is also modiﬁed by the addition of the alkyl groups, making the clay more hydrophobic and hence improving the interaction with the styrene monomer. Other groups can also be attached to the alkyl ammonium ions such as aromatic groups (structures 6–10, 16–20, 22, 24, 26), which allow greater interaction with the styrene monomer due to the similar nature of the organic groups. A number of research groups have also investigated nonreactive modiﬁcations that have higher thermal stability than the standard alkyl ammonium ions for use in polymer matrixes such as Syndiotactic PS. Syndiotactic PS is a special grade of PS in which the phenyl groups are oriented in an alternative manner along the backbone, leading to higher crystallinity and higher levels of packing of the chains. Syndiotactic PS is processed at much higher temperatures (above 260 °C) than atactic PS (around 200 °C), and these temperatures would lead to the decomposition of alkyl ammonium ions. It is however known that alkyl ammonium ions decompose above temperatures of 200 °C by a Hoffman elimination reaction [65, 66]. The hydrocarbon tail is ﬁrst lost and the amine may then be eliminated leaving a proton as the cation in the clay. To overcome this, phosphonium-based surfactants have been investigated (7, 24, 25) [20, 31], which have a higher thermal degradation temperature than ammonium-based surfactants. Other nonreactive cations have also been investigated which have even higher thermal stabilities [3, 17, 26, 37, 67]. These include pyridium (6, 8, 10) [17, 25, 26] and quinolium (9) [25, 26] based ions. Taking a different tact, a zwitterionic surfactant (4) was used to modify MMT for PS nanocomposites [7]. Zwitterionic surfactants contain both a cation and an anion. The claimed advantage of using this surfactant was that in certain solvents

333

9

8

7

6

5

4

3

2

1

Number

Br

NH2

N

Br

P

Br

N

13

O

N

Br

N

(CH 2) 16 CH 3

13

O

OH

HT= hydrogenated tallow 65% C18 30% C16 5% C14

HT= hydrogenated tallow 65% C18 30% C16 5% C14

N (CH2 )15 CH 3

N HT

HT

N HT

HO O

f

Surfactant-nonreactive

N+ Cl-

Quinolium salt (QC16)

Phenylacetophenone dimethylhexadecyl ammonium (BPNC16)

[2(Dimethylamino)ethyl] triphenylphosphonium bromide

Cetyl pyridium chloride

Aminoundecanoic acid

Octadecyldimethyl betaine

Cetyl trimethlammonium bromide (CTAB)

Dimethylbenzyl hydrogenated tallow ammonium (Cloisite 10A)

Dimethyldehydrogenated tallow ammonium (Cloisite 6A, 15A, 20A)

Name

Table 13.1 Nonreactive cationic surfactants used to modify montmorillonite.a)

Bulk

Bulk

Solution

Bulk

Emulsion

Bulk

Bulk, Emulsion, Suspension, Solution Bulk

Bulk syndiotactic

Polymerization technique

I

I

I

I/E mixed

E

I

I/E mixed

I

I

Morphology

[26]

[25]

[20]

[17, 24]

[23]

[7, 17]

[5, 16–22]

[14, 15]

[6, 11–13]

References 334

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties

20

19

18

17

16

15

14

13

12

11

10

Number

NH

13

Si

O

N

CH 3 (CH 2 )17

Cl

(CH2 )9 CH3

(CH2 )15 CH 3

(CH2 )15 CH 3

(CH2 )9 CH 3

N (CH2 )10 N

N (CH2 )10 N

N (CH2 )5 N

N (CH 2) 17 CH2 POSS

CH 3 (CH2 )17

Si

Si

Br

O

O

N

Surfactant-nonreactive

Si O

Benzalkonium chloride

Decylcarbazole methyldidecyllammonium (10ACDD)

Decylcarbazole dimethylhexadecylamonium (10AC)

Bulk

Bulk

Bulk

Bulk

Emulsion

Trisilanolisobutyl polyhedral oligomeric silsesquioxane Pentylcarbazole dimethylhexadecylammonium (5AC)

Emulsion

Emulsion

Emulsion

Emulsion

Octadecyldimethylammonium

Octadecyldimethyl methoxy silane

Octyldimethyl methoxysilane

Vinyl dimethylethoxy silane

Emulsion

Bulk

Pyridium salt (PyC16)

Trimethylethoxy silane

Polymerization technique

Name

I

I/E

I/E

I/E

E

I

I

I

I

I

I

Morphology

(Continued)

[5]

[29, 30]

[29]

[29]

[28]

[28]

[27]

[27]

[27]

[27]

[26]

References

13.3 Modiﬁcation of MMT 335

a)

27

26

25

24

23

22

21

y

z

H2 C R

N

R = ammoniumion

(CH 2 )11 CH3

n

H 2N (CH 2) 3 Si O Si O Si (CH 2) 3 NH 2

x

P

Cl

(CH 2) 11 CH3

P Cl-

+

N

Br

Br 2

H2 N

Cl

Surfactant-nonreactive

(Continued)

Aminopropyl terminated polydimethylsiloxane

Ammonium modiﬁed styrene/bromine copolymers

Hexadecyltributylphosphonium bromide

Triphenylphosphonium chloride (P16)

Octadecylamine (ODA)

Solution (CO2)

Bulk

Syndiotactice/ bulk

Bulk

Bulk

Bulk

Bulk

Dodecylammonium chloride

Ethylbenzyl dodecylammonium chloride

Polymerization technique

Name

I = intercalated; E= exfoliated; I/E = mixed intercalated/exfoliated morphology.

Number

Table 13.1

I/E

I

I/E

I

I

I

I

Morphology

[34]

[33]

[13]

[31, 32]

[5]

[2]

[2]

References 336

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties

33

32

31

30

29

28

Number

R

N

N

Br

Br

C18

C16

R= C12

(CH 2) 15 H 3C

N

N

N

Surfactant-reactive

octadecyl

hexadecyl

dodecyl

N

N

Bulk Bulk

N,N-trimethyl-6-(4-vinylphenyl)hexan-1aminium (6T) N,N-dimethyl-N-(4-vinylbenzyl)dodecan-1amminium (12H or VDAC)

Vinyl benzyl hexadecylammonium bromide (VB16)

Imadazolium salts

N,N-trimethyl-6-(4-vinylphenyl)hexan-1aminium (12T)

Bulk

N,N-dimethyl-N-(4-vinylbenzyl)hexan-1amminium (6H)

I E E E E

Suspension Emulsion RAFT NMP

E

I, E

I

E

I

I

Morphology

Solution

Bulk

Bulk

Bulk

Emulsion

Polymerization technique

Name

Table 13.2 Reactive surfactant cations used to modify montmorillonite in PS–MMT nanocomposites.a)

[40] (Continued)

[39]

[14, 15, 31, 32, 39]

[37, 38]

[2]

[2, 5, 7, 28, 35, 36]

[2]

[2]

Reference

13.3 Modiﬁcation of MMT 337

38

37

36

35

34

Number

Table 13.2

O

R P R

R

H 2C

O

O

CH3

N

Br

Br

N

M=

N M

(CH 2) 17 H 3C

O

Surfactant-reactive

(Continued)

BrN+

(CH 2) 15 CH 3

(CH 2) 7 CH 3

(11-Acryloyloxyundecyl)dimethylethylammonium bromide

Tributyl (4-vinyl benzyl) phosphonium

Triphenyl phosphonium

Vinyl benzyl trimethyl ammonium

Vinyl benzyl octadecylammonium bromide (C20-4VB)

2-Methacryloyloxy (MHAB)

Bulk

Solution

Bulk

Emulsion

Suspension (gamma irradiation)

Bromobenzyl (MSAB) 2-Methacrylalyoyloxy ethylhexadecyladimethyl ammonium bromide (MOAB)

Polymerization technique

Name

E

E

E

I

E

E

[46, 47]

[41]

[43–45]

[28, 41, 42]

[19]

E E

Reference

Morphology

338

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties

a)

45

44

43

42

41

40

39

O

O

O

O

Br

O

C C

HS

O

O

O

N

N+ Br -

N

N+ Cl-

N+ Br -

Cl

Surfactant-reactive

OH

O

N+ Cl-

BrN+

OH

OH

Vinyl benzyl dimethylethanolammonium chloride

2-Methacryloyloxyethyl-hexadecyldimethyl ammonium bromide (MA16)

N,N-Dimethyl-nhexadecyl-(4hydroxymethylbenzyl) ammonium chloride

N,N-dimethyl-N-octyl-N-ethylmercapatan ammonium bromide

[2-(acryloyloxyethyl)] trimethylammonium chloride

2-[(Acryloyloxy)ethyl](4-benzoylbenzyl) dimethylammonium bromide

(11-Acryloyloxyundecyl)-dimethyl (2-ydroxyethyl) ammonium bromide

Name

I = intercalated; E= exfoliated; I/E = mixed intercalated/exfoliated morphology.

Number

Bulk

RAFT

Bulk

Bulk

Emulsion

Bulk

Bulk

Polymerization technique

E

I

I/E

I

I/E

E

E

Morphology

[24]

[39]

[32]

[50]

[49]

[48]

[46, 47]

Reference

13.3 Modiﬁcation of MMT 339

49

48

47

46

Number

O

O

N H

O

N N O

H N O

O

N

CN

Br

N N

O

O

CN

O

CN

N N

O

CN

Br N

O

O

H C (CH2 )2 CN NC (CH 2) 2 C N (CH2 )2 Si O Si CN CN O x O

N

Br

Surfactant–polymeric initiators

Br N

H (CH 2) 2 N

N Br

n

Difunctional free radical initiator

Monofunctional free radical initiator

Bulk

Bulk

Bulk

Bulk

2,2-Azobis{2-methyl-N-[2-acetoxy-(2N,N,N-tributylammonium bromide) ethyl] propionamide} (ABTBA)

Macroazoinitiator

Polymerization technique

Name

Table 13.3 Polymeric initiator-based surfactants used to modify montmorillonite in PS–MMT nanocomposites.a)

I

E

I/E

E

Morphology

[56–58]

[56–58]

[55]

[51–54]

Reference

340

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties

a)

53

52

51

50

Br

HO

O

O

S

S

O

O

N

N

N

Cl

O N

COOH

(CH2 )6

N N

N

Br

N

N

OH 2HCl

Surfactant–polymeric initiators

Br N

1-(4-(2-(Benzoyloxy)-1-((2,2,6,6tetramethylpiperidin-1-yl)oxy)ethyl) phenyl) – trimethyl ammonium chloride

11-Trimethylammonium bromideundecyl-2-bromo-2-methyl propionate

10-carboxylic acide-10-dithiobenzoatedecyltrimethylammonium bromide (CDDA)

2,20-Azobis(2-(1-(2-hydroxyethyl)-2-imidazolin2-yl)propane) dihydrochloride monohydrate (VA060)

Name

I = intercalated; E = exfoliated; I/E = mixed intercalated/exfoliated morphology.

Number

NMP

Bulk, ATRP

E

I,E

E

I

Bulk, RAFT

Bulk, RAFT

Morphology

Polymerization technique

[62]

[61]

[60]

[59]

Reference

13.3 Modiﬁcation of MMT 341

342

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties

the anions at the end of the surfactants could repel each other and hence improve the clay platelet separation even further. It was found however that only intercalated structures were formed. Li et al. [23] reported the use of MMT modiﬁed with zwitterion (aminoundecanoic acid, structure 5) to synthesize PS–clay nanocomposites by emulsion polymerization. Protonation of the zwitterionic acid molecules led to the exchange of the cations in the interlayers of MMT and the carboxyl groups were then ionized in the alkaline aqueous medium which allowed exfoliation of the clay. The resulting PS–MMT nanocomposites were observed to have an exfoliated microstructure. Modiﬁcation of nanoclays and nanoparticles to make them organophilic can also be undertaken using silane grafting [68–75] to the edge hydroxyl groups on the clay or between the clay layers on the clay surface [69]. Polyfunctional silanes have been observed to graft onto the clay edges in the form of oligomers and can effectively lead to pillaring of the clay stacks [76]. Monofunctional silanes on the other hand can selectively attach to the individual clay sheets, as shown in Figure 13.2 [76]. Monofunctional silane modiﬁed clays were used to form PS–MMT nanocomposites using emulsion polymerization without the need for a surfactant (structures 11–14) [27]. It was observed that clay platelets were located inside the polymer particles and also outside of the particles and that intercalated morphologies were observed.

a)

R CH3-Si-CH3

CH3

CH3

O

R Si-O

O - Si(CH2)3O(CO)C(CH3)=CH2

CH3

CH3 O CH3-Si-CH3 R

R=

O

b)

R

O

O

Si O

R

O

O Si

O R

Si

O

R

OH

O Si

R

O

O

O

Si

O OR(H) Si

O Si

R

OR(H)

R Figure 13.2 The coupling reaction of (a) monofunctional and (b) trifunctional silane

molecules on clay edges. Reproduced from Ref. [76] with permission.

13.3 Modiﬁcation of MMT

Polyhedral oligomeric silsesquioxanes (POSS) were attached to MMT through ion-exchange reactions (16) and dispersed in styrene nanocomposites through emulsion polymerization [28]. The addition of POSS into the clay layers helped to improve the thermal stability of the resulting nanocomposite signiﬁcantly, while the glass transition temperature was only slightly increased. The commonality between the surfactants described in this section is that they do not contain a reactive group which is capable of reacting with styrene. It has been found that for these types of nonreactive surfactants, although styrene monomer can intercalate between the clay layers, exfoliated morphologies have not been achieved for in-situ polymerized PS nanocomposites. 13.3.2 Reactive Modiﬁcations

Modiﬁcations that contain reactive or polymerizable functionality have been used to improve the clay dispersion in in-situ polymerized styrene-based nanocomposites [2, 5, 15, 19, 28, 35, 36, 41, 43–47, 77], and these surfactants are listed in Table 13.2. The notion is that the growing PS chain may polymerize from or through the reactive group on the surfactant, aiding to push the clay platelets apart and hence improve the prospect of exfoliation. In most cases, the reactive group has been styrene-based although acrylic-based surfactants have also been used (structures 38–41) [46–49]. Figure 13.3 describes the difference between reactive and nonreactive modiﬁcations. Some of the earlier work in PS–MMT nanocomposites investigated the modiﬁcation of MMT with reactive cations consisting of a styrene group attached to the ammonium ion (structure 36, Table 13.2) [41, 43–45, 77]. Khalil modiﬁed sodium MMT with 4-vinyl benzyl chloride and used this to form styrene and acrylonitrile– styrene copolymers [43, 44]. Akelah et al. [45, 77] also investigated vinylbenzyl

Figure 13.3 Nonreactive and reactive ion-exchange reactions of montmorillonite.

343

344

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties

trimethylammonium (35) as a clay modiﬁer for PS–MMT nanocomposites. The nanocomposite structure in all these cases was primarily intercalated. Fu et al. further investigated surfactant containing styrenic groups (30) [35, 36]. The surfactants investigated in these studies contained a long alkyl chain with a styrene group located adjacent to the ammonium ion (head-type surfactants). This work produced exfoliated morphologies, whereas the shorter reactive cations in the earlier work did not, due primarily to the long alkyl chain which helped to increase the initial intergallery spacing and allow more styrene monomer to swell the clay in the initial stages of the reaction. Abate et al. [37] and Bottino et al. [38] investigated reactive imadazolium salts (Structure 32, Table 13.2), in an attempt to produce nanocomposites with higher thermal stabilities, while also introducing a functional group capable of reacting with the styrene. Using a slightly different tact, Akat et al. investigated exchanging a chain transfer agent (diethyl octyl ammonium ethylmercaptan bromide, 42) [50] onto MMT prior to the free radical polymerization of PS. They found that while PMMA nanocomposites formed with this modiﬁed clay led to exfoliated structures, PS nanocomposites led to intercalated morphologies. It has been found that while both reactive and nonreactive ion-exchanged surfactants can increase the d-spacing of the clay and allow intercalation of the styrene monomer, it is only the reactive modiﬁcations that have led to exfoliated nanocomposites [8]. A series of surfactant architectures were investigated (structures 22, 28–31) and the effect of the architecture of the modifying surfactant on the ﬁnal nanocomposite morphology was probed in detail [2]. It was determined that to achieve well-dispersed PS–MMT nanocomposites, a number of factors were important, and these are summarized in Figure 13.4. The propensity of the styrene monomer to swell the modiﬁed clays was observed to be a useful predictor of compatibility of the monomer and the clay and this was directly observed through transmission electron microscopy and X-ray techniques. It was found that the presence of a polymerizable group was necessary as was the position of the polymerizable group in a head position (located adjacent to the ammonium ion rather than at the end of the alkyl chain). The length of the alkyl chain was also important, with shorter alkyl chains (27, 28) not leading to exfoliated structures, whereas alkyl chains of 12 or higher (29) led to exfoliated structures. This is presumably because the initial clay d-spacing needs to be high enough to allow sufﬁcient styrene monomer to swell the clay. Finally, the solubility of the surfactant in styrene was length of akyl chain Polymerizable group N position of ammonium group Figure 13.4 Essential aspects of surfactant design for polystyrene–montmorillonite

nanocomposites. Reproduced from Ref. [78] with permission.

13.3 Modiﬁcation of MMT

necessary to ensure complete swelling of the modiﬁed clay by the styrene in the initial stages of the polymerization and lead to complete dispersion. 13.3.3 Polymeric Initiator-Based Modiﬁcations

Modifying the platelet surfaces with organic cations that improve the compatibility of MMT with the monomer as well as initiating or catalyzing the polymerization from within the clay layers offers an advantage in achieving exfoliated nanocomposites. This technique has been used to produce exfoliated nanocomposites in acid-catalyzed nylon and epoxy systems [79–81], as well as in metal-catalyzed polyoleﬁn systems [82]. The reason for the improvement in exfoliation is due to shifting the location of initiation from predominantly extra-gallery (outside of the tactoids) to intragallery (within the tactoids) [83]. The polymer initiator-based modiﬁcations used for PS–MMT nanocomposites are listed in Table 13.3. A series of papers by Uthirakumar et al. [51–54, 84] investigated the ion-exchange of macroazoinitiators onto MMT. A range of macroinititaors were exchanged onto MMT but it was found that 2,2′-azobis 2-methyl-N-(2-acetoxy-(2-N,N,Ntributylammonium bromide) ethyl) propionamide led to the highest intergallery spacing and the best interaction with styrene (46, Table 13.3) [51, 53]. Jeong et al. investigated a similar method, using a macroazoinitiator composed of repeat sequences of poly(dimethylsiloxane) (PDMS) and azo groups, which was intercalated between MMT but not ion exchanged (47) [55]. Fan et al. [56] attached an initiator to MMT via cation-exchange suitable for free radical initiation, investigating monofunctional and bifunctional (48, 49). The initiator-clay was thermally active and the clay rendered “adsorptive” rather than “organophilic” as no compatible groups with the matrix had been added to the clay. A comparative study of the monofunctional and bifunctional-modiﬁed clays used to polymerize styrene found that the monofunctional modiﬁed clays led to better exfoliated nanocomposites [57, 58]. Research has also focused on using controlled polymerization techniques to produce PS–MMT nanocomposites, and in most cases the approach has been to cationically exchange the relevant catalyst or initiator to the MMT to allow polymerization to grow from the clay surface. The controlled polymerization techniques utilized have included reverse addition fragmentation transfer (RAFT, 50, 51) [39, 59, 60], atom transfer radical polymerization (ATRP) [61, 85], and nitroxidemediated polymerization (NMP) [40, 62]. Weimer et al. investigated PS nanocomposites using MMT modiﬁed with a nitroxyl-mediated living free radical initiator (53) [40, 62]. They found that they could control the molecular weight of the PS and that exfoliated nanocomposites were formed. Shipp et al. investigated intercalating controlled radical polymerization initiators onto clay layers, including RAFT and ATRP initiators [39, 61, 85]. For the ATRP investigations [61, 85], an ATRP initiator was modiﬁed with a quaternary ammonium ion (52), and intercalated into the interlayer spacing of the clay (Figure 13.5). ATRP polymerization was then conducted for PS, polymethylmethacrylate, and poly(n-butylacrylate), with

345

346

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties Polymer-layered silicate nanocomposite

MMT-2 x

x x

x

x

x

x x

x x

x

x

x

x x

x x

ATRP

x x

x

x

x

x

x x=

N

((

11 O

O Br

Figure 13.5 ATRP of polymer from initiator modiﬁed clay (MMT-2). Reproduced from Ref. [61]

with permission.

polymethylmethacrylate and poly(n-butylacrylate) forming exfoliated nanocomposites. Low polydispersities and predictable molecular weights were obtained. The technique was also extended to poly(styrene-block-butyl acrylate) (PSBA) block copolymer brushes on the surfaces of intercalated and exfoliated silicate layers using ATRP. Block copolymer brushes on the surface of exfoliated or intercalated clay layers were found to create nanopatterns after treatment in different solvents. Taking a slightly different tact, Samakande et al. [59] synthesized a cationic RAFT initiator that was exchanged onto the surface of the clay (50). The clay was then used as a macroinititaor in the preparation of PS nanocomposites and formed intercalated structures. The beneﬁt of exchanging polymeric initiators onto the clay surface is that the growing chains are anchored to the clay surface, in a “surface initiated” polymerization and the growing PS chains may help to push the clay platelets further apart. The clay modiﬁcation itself however still needs to be compatible with the styrene, in order to allow monomer to swell the platelets. This approach is further discussed in Section 13.4, Figure 13.9.

13.4 In-situ Polymerization Methods

Several techniques may be used to produce polymer–MMT nanocomposites, including in-situ polymerization, melt intercalation, and solution casting; however, to date in-situ polymerization has produced the best dispersed systems for PS–MMT nanocomposites [8]. This technique offers the advantage of allowing the monomer to penetrate the intergallery spaces ﬁrst, which aids the exfoliation process. If the polymerization reaction can occur between the clay layers, the delamination process is further enhanced as growing polymer chains can push the clay layers apart, which is the main reason for the superior dispersed systems observed [8]. There are several in-situ polymerization techniques which may be used to form

13.4 In-situ Polymerization Methods

PS nanocomposites, including bulk free radical plymerization, solution and emulsion polymerizations, as well as controlled polymerization techniques such as RAFT, NMP, and ATRP. 13.4.1 Free Radical Polymerization Techniques 13.4.1.1 Bulk Polymerization Although many studies have investigated PS–MMT nanocomposites synthesized using different techniques, the best exfoliated structures have generally been formed through bulk polymerization methodologies. Bulk polymerization involves the free radical polymerization of the styrene monomer, clay, and an initiator without the presence of solvent. The beneﬁt of this technique is that it allows the monomer to swell the clay layers and become intercalated before polymerization commences, which aids in the dispersion of the clay. Fu et al. [35, 36] prepared the ﬁrst exfoliated PS–MMT nanocomposites by bulk polymerization, using a reactive surfactant (30). Several investigations since have found that when suitable surfactants were used, exfoliated morphologies could be formed through bulk methodologies [2, 5, 7, 14, 15, 31, 35–37, 42]. It was consistently found however that the structure of the surfactant was critical in determining the ﬁnal morphology as discussed in Section 13.3. Simons et al. [78] investigated the mechanism of intergallery expansion for PS–MMT nanocomposites formed through in-situ bulk polymerization using small angle x-ray scattering (SAXS, Figure 13.6) and kinetic studies, and an intergallery expansion mechanism postulated. For exfoliated morphologies, three stages of intergallery expansion were identiﬁed. The ﬁrst stage involved the swelling of the modiﬁed clay with styrene monomer, and the amount of swelling was

Figure 13.6 Intergallery expansion over reaction time for a modiﬁed clay–PS reaction mixture

(from small angle X-ray scattering). Reproduced from Ref. [78] with permission.

347

348

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties

limited by the solubility of the surfactant in styrene. The second stage occurred during the initial 60 min of polymerization and was characterized by fast intergallery expansion as observed via SAXS, as well as low conversions and higher molecular weights. During this stage, styrene monomer began to polymerize and because the surfactant contained double bonds, growing PS chains could polymerize through or from the surfactant on the clay surface, contributing to platelet expansion. The third stage involved a cessation of intergallery expansion as observed by SAXS, but a decrease in the intensity of the diffractable peak, as the platelets disordered rather than continued to expand, leading to exfoliated morphologies. These studies indicate that the bulk of the intergallery expansion occurs in the early stages of reaction, which is why it is important for the styrene to swell the intergallery spacing as much as possible in the ﬁrst stage. Stereospeciﬁc PS have also been formed in the presence of MMT using bulk polymerization. Syndiotactic PS is synthesized through a metallocene-catalyzed polymerization which can produce ordered PS with the phenyl groups on alternating sides. This leads to a polymer with high crystallinity and a high melting point about 270 °C. Shen et al. prepared exfoliated syndiotactic PS–MMT nanocomposites through in-situ bulk polymerization using hexadecyl treimthylammoniummodiﬁed MMT (structure 3) and a monotitanocene catalyst and a partially exfoliated morphology was formed [22]. Bruzaud et al. utilized alkylammonium (3) and alkylphosphonium (25) modiﬁed clay to form syndiotactic PS using a hemimetallocene catalyst, and a partially exfoliated morphology was formed [13, 22]. High impact polystyrene (HIPS) is a blend of PS that has been polymerized in the presence of polybutadiene. This leads to PS chains with grafted polybutadiene, as well as some free PS and PB, and the resultant polymer blend has improved impact strength. HIPS/MMT nanocomposites were formed though in-situ bulk polymerization in the presence of polybutadiene [86]. Intercalated nanocomposites with improved thermal stability were formed although the dispersion was different in the PS matrix phase compared to the rubber phase. Bulk polymerization initiated via more novel methods have also been used to form bulk PS–MMT nanocomposites. Zhang et al. [19] used gamma irradiation to initiate the polymerization of PS–MMT nanocomposites with different surface modiﬁcations (3, 34) and successfully prepared exfoliated morphologies when reactive clay modiﬁcations were used. Uthirakumar et al. [51–54] modiﬁed MMT with a cationic radical initiator which was used to initiate the bulk polymerization of styrene. Because the polymerization was initiated from the clay surface and the monomer and the clay were suitably compatible, exfoliated morphologies were formed. 13.4.1.2 Emulsion Polymerization Emulsion polymerization is an important type of free radical polymerization which is performed in aqueous systems. The most common type of emulsion polymerization is an oil-in-water emulsion, in which droplets of monomer (the oil) are emulsiﬁed (with surfactants) in a continuous phase of water. The surfactants form micelles, and when monomer is added to the system they tend to penetrate and

13.4 In-situ Polymerization Methods

accumulate within the micelle as the monomer is typically insoluble in water. The initiator is added to the aqueous phase where it decomposes and then diffuses into the hydrophobic micelle which contains the monomer. When the initiator enters the micelle, the monomer is initiated and polymer begins to grow, as shown in Figure 13.4. The advantages of emulsion polymerization are that high-molecularweight polymers can be produced without the viscosity and exothermic heat control issues that can occur in bulk and solvent polymerizations, and that the molecular weight is easily controlled by controlling the concentration of initiator and surfactant [42]. Emulsion polymerization has been employed to synthesize nanocomposites with both intercalated and exfoliated morphologies, depending on the surface modiﬁcation on the clay [27, 42, 49, 87–90]. The addition of MMT to the emulsion polymerization adds another variable to the process. If the clay has more afﬁnity for the monomer rather than the water, the MMT can diffuse into the monomer droplet. This is the desired situation, as it leads to the best dispersed systems, but this also makes the surfactant choice critical for these systems. Often the monomer and clay are premixed together to allow the platelets to disperse within the monomer, before being added to the water/surfactant/initiator solution. The nanocomposites formed often contain some silicate embedded inside the polymer particles and some adsorbed on the surface of the particles, as shown in Figure 13.7. If the clay particles are too large, the clay may form new micelles, which leads to a reduction in the ﬁnal polymer weight, as observed by Wang et al. [14] 13.4.1.3 Solution Polymerization Solution polymerization has been reported as the polymerization technique for preparing PS–MMT nanocomposites [14, 20, 41, 45]. Solution polymerization

Figure 13.7 Emulsion polymerization process.

349

350

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties

offers advantages industrially including heat transfer and hence less issues associated with the exothermic free radical reaction, as well as reduced viscosity compared to bulk polymerizations although the main disadvantage of the technique is associated with the removal of the solvent. Intercalated morphologies have been predominantly reported for PS–MMT using this polymerization method [14, 20, 45]. Wang et al. [14] investigated PS–MMT nanocomposites prepared by bulk, solution, suspension, and emulsion polymerization as well as melt blending. They found that solution polymerization always formed intercalated morphologies, whereas bulk, emulsion, and suspension polymerization could form exfoliated structures if the surfactant was suitable. The choice of solvent and the concentration was critical – the polymer as well as the monomer needs to be soluble in the solvent and the clay needs to be swollen sufﬁciently in the solvent to allow monomer/polymer insertion. The solvent however can also act to compete with the monomer/polymer in the intergallery spaces if it is more compatible with the clay than the monomer is. More recently, Akelah et al. [41] investigated in-situ PS–MMT nanocomposites polymerized in toluene with reactive surfactants for the clay modiﬁcation (35, 37) and found that exfoliated morphologies could be obtained. Zhao et al. investigated the use of supercritical carbon dioxide as the solvent for in-situ PS–MMT polymerizations [34]. Supercritical carbon dioxide has attracted interest in recent years as an alternative to conventional solvents because it is more environmentally benign and economically viable. PDMS surfactant-modiﬁed clay (27) was used and exfoliated morphologies were achieved. 13.4.2 Controlled Polymerization Techniques

Polystyrene may be synthesized using controlled and living polymerizations such as ATRP, RAFT, and NMP. These methodologies have been used to form PS–MMT nanocomposites. These polymerizations have generally been performed using two different approaches. The ﬁrst approach involves synthesizing a cationic version of the initiator and exchanging it onto the MMT. In this case, the polymerization is surface initiated from the surface of the clay, as shown in Figure 13.8, offering the advantage that the polymer chain grows from the clay surface and so can

N

Ph N Ph I I I Ph Ph I Ph Ph I

I

N N

I

N I N

I

N I N

Sty I

N N N I I Sty Sty Sty Sty I Sty I Sty I N N N Sty I

I

Ph Ph Ph Ph N

N

N

N

Figure 13.8 Surface initiated polymerization. I = initiator, Sty = styrene, Ph = phenyl group.

13.4 In-situ Polymerization Methods

disorder the platelets as the PS chains grow, improving dispersion. In the second approach, conventionally modiﬁed clays are mixed with styrene and a living or controlled polymerization initiator/catalyst system is added to the entire system. Exfoliated and intercalated morphologies have been obtained using both these approaches. 13.4.2.1 Atom Transfer Radical Polymerization ATRP [91–94] was used as the polymerization technique to form PS–MMT nanocomposites in a series of papers by Shipp et al. [61, 85]. An ATRP initiator consisting of a quaternary ammonium salt moiety and an ATRP initiator moiety (2-bromo-2-methyl proponiate, 52) was exchanged onto MMT. ATRP polymerization was surface initiated as described in Figure 13.8 in the presence of styrene and a Cu catalyst, and exfoliated morphologies were obtained containing homopolymers with predictable molecular weights and low polydispersities, characteristics of living radical polymerization. PSBA block copolymer brushes on the surfaces of exfoliated and intercalated clay layers were also formed using a similar method [85]. 13.4.2.2 Reverse Addition-Fragmentation Transfer Reverse addition-fragmentation chain transfer polymerizations (RAFTs) [95–97] have been used as the polymerization technique to form PS–MMT nanocomposites in several reports [39, 59, 60]. RAFT polymerizations are typically carried out in the presence of a RAFT agent and a polymerization initiator. Salem and Shipp [39] combined a styrenically modiﬁed clay in styrene with a RAFT agent and an initiator and found that the polymerized nanocomposite led to exfoliated morphologies with predictable molecular weights and low polydispersities, characteristics of living radical polymerization. Samakande et al. [59] modiﬁed MMT with an azo initiator (50) and then in combination with styrene and a RAFT agent, intercalated PS–MMT nanocomposites were formed, with typical RAFT characteristics. The azo intitiator was found to be less effective in initiating between the clay layers because of radical–radical coupling (the cage effect), but it was still sufﬁcient to polymerize the nanocomposite. Zhang et al. [60] took a different approach and investigated surface initiated polymerization, by synthesizing a RAFT agent (51) that was intercalated onto MMT. In-situ RAFT polymerization led to exfoliated structures with characteristic RAFT polymer properties and improved thermal stability. These studies show that while surface initiated polymerization can be beneﬁcial to the dispersion of the clay in the PS, other factors such as the compatibility of the styrene and the clay modiﬁcation, and the interaction of the clay modiﬁcation with the crystal structure of the clay itself can affect the ﬁnal morphology. 13.4.2.3 Nitroxide-Mediated Polymerization NMP is another controlled free radical polymerization technique that can be used to polymerize styrene [98–100]. Weimer et al. investigated surface-initiated PS–MMT nanocomposites using MMT modiﬁed with a nitroxyl-mediated living

351

352

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties

free radical initiator (53) [62]. They found that the molecular weight of the PS could be controlled and exfoliated nanocomposites were formed. Xu et al. [40] investigated the effect of reactively modiﬁed MMT containing a vinyl moiety (33) in combination with styrene monomer and NMP initiators, and formed exfoliated structure. In this case, the growing PS chains attached to the reactive group of the modiﬁed clay, leading to grafted PS. The living/controlled techniques used to form PS–MMT nanocomposites have led to both exfoliated and intercalated structures. Exfoliated morphologies may be obtained by either surface initiated polymerization or through modiﬁed clays that contain a reactive group such as styrene. Both scenarios allow growing PS chains to grow from or through the clay modiﬁcation, increasing the disorder of the platelets. 13.4.3 Dispersion of MMT in Styrene

The method used to disperse clay into the monomer is important as good dispersion in the monomer is vital for the ﬁnal nanocomposite morphology. Although the surface modiﬁcation of the clay is a critical aspect in determining the compatibility between the styrene and the clay, the method used to disperse the clay is also important, and in most cases some form of mechanical stirring is used as a minimum. Other methods can improve the ability to exfoliate including ultrasonic [87, 101] and direct electric current [102]. Qi et al. used a direct electrical current [102] to help to disperse the MMT in a styrene monomer. They formed intercalated nanocomposites that were highly oriented and observed improved thermal properties. Wang et al. [87] prepared intercalated unmodiﬁed MMT in PS though in-situ emulsion polymerization, using high powered ultrasound to initiate the polymerization as well as to disperse the MMT. Wang et al. [101] followed a similar procedure utilizing high powered ultrasound with unmodiﬁed MMT and formed a mixed intercalated/exfoliated morphology. In these cases, the intensity of the ultrasound needed to be high enough for dispersion and to initiate the reaction.

13.5 Properties of PS–MMT Nanocomposites Prepared via In-situ Techniques

The motivation behind the large body of work in the ﬁeld of polymer–MMT nanocomposites is that signiﬁcant improvements in properties can be achieved with low clay loadings (1–10 wt%). Different polymer matrixes have exhibited different levels of improvement, with the largest improvements being observed for nylon– MMT nanocomposites [80, 103] and epoxy–MMT nanocomposites [104–106]. PS is one of the most widely used commodity polymers and is used in many applications. Improving the properties of PS cheaply and easily would be of great advantage. The properties that are typically improved through the reinforcement with

13.5 Properties of PS–MMT Nanocomposites Prepared via In-situ Techniques

clay include the mechanical, strength, modulus, thermo-mechanical, barrier properties (gas, solvent and ﬂame resistance) and for some polymers such as PEO, the electrical conductivity [1]. A further advantage is that if the MMT is well exfoliated into the polymer, the optical clarity of the nanocomposite is not changed, due to the nanosize of the platelets being smaller than the wavelength of visible light. The weight of polymer clay nanocomposites is also lower than conventionally ﬁlled nanocomposites, due to the relatively low levels of clay addition needed to realize property improvements. 13.5.1 Mechanical Properties 13.5.1.1 Tensile Tensile tests are a versatile measurement of mechanical strength as Young’s modulus, tensile strength, and elongation at break can all be obtained. Conventional nanocomposites using standard ﬁller improve Young’s modulus, improve the tensile strength and decrease the elongation at break, typically with ﬁller loadings of 30–50 wt%. These improvements in tensile strength have also been observed for polymer–clay nanocomposites, and speciﬁcally for PS–MMT nanocomposites, generally at clay loadings of 1–10 wt%. The improvement in mechanical properties is usually more signiﬁcant for exfoliated morphologies; however, this depends on the polymerization technique as well as the nature of the clay modiﬁcation. Uthirakumar et al. [52] formed PS–MMT nanocomposites via bulk methodologies using a radical initiator that was exchanged onto the MMT surface. Tensile tests indicated improvements by up to 45% for 3 wt% clay, improvements in Young’s modulus by 25% and only a small decrease in the elongation at break. Zhu et al. [32] prepared PS–MMT nanocomposites using bulk polymerization containing reactively modiﬁed clay, and observed improvements in the tensile strength at break by 300% for exfoliated nanocomposites. In the comparative studies between different polymerization techniques for PS–MMT nanocomposites, Wang et al. [14] found that the mechanical properties of the nanocomposite depended on the polymerization technique as well as the morphology of the nanocomposite and the results are summarized in Table 13.4. For bulk polymerized nanocomposites, the highest improvements in properties were observed for exfoliated morphologies. For the emulsion polymerized nanocomposites; however, Young’s modulus decreased slightly while the tensile strength increased for exfoliated and intercalated morphologies. The reason for the apparent decrease in some of the tensile tests upon addition of clay is not well understood although it was suggested that the impact of the molecular weight of the PS was important, particularly for the emulsion and solution polymerized nanocomposites which exhibited decreased molecular weight when combined with clay. This indicates that the interactions between the clay, the surfactant, and the polymer are important in determining the ﬁnal physical properties of the nanocomposites.

353

354

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties

Table 13.4 Tensile test results for PS–MMT nanocomposites.

Surfactant (Tables 13.1–13.3)

Clay loading

Polymerization technique

Morphology

Young’s modulus (GPa)

Tensile strength (MPa)

Elongation (%)

– 33 (Reactive) 2 (Nonreactive) – 33 (Reactive) 2 (Nonreactive) – 33 (Reactive) 2 (Nonreactive)

– 3 3 0 3 3 0 3 3

Bulk

– E I – I I – E I

0.74 1.14 1.00 1.3 0.90 1.44 1.08 1.05 0.95

5.9 8.1 3.5 6.9 7.3 9.4 11.2 12.5 8.1

1.1 0.8 0.4 2.9 1.5 0.7 7.2 1.6 0.8

Suspension

Emulsion

Reproduced from Ref. [14] with permission.

13.5.1.2 Impact and Flexural Properties There are relatively few studies looking into the effect of PS–MMT on impact properties, and there have been conﬂicting reports. In general, the effect of MMT reinforcement on impact strength has not been consistent and is highly dependent on the polymer matrix and clay modiﬁcation. In general, nanocomposites which are reinforced with conventional ﬁllers tend to provide decreased impact strength when modulus and tensile strength are increased due to the ﬁllers acting as stress concentrators. In the early nylon–MMT investigations, the impact strength was observed to not decrease signiﬁcantly, which was considered to be advantageous compared to traditional nanocomposites [79, 80, 107]. Tseng et al. [24] synthesized reactively modiﬁed clay in bulk polymerized PS–MMT nanocomposites. Impact strength was improved signiﬁcantly over pure PS (23.4 J/M) for exfoliated morphologies (34.8 J/M) and less signiﬁcantly for intercalated morphologies (25.1 J/M). The ﬂexural strength and modulus were improved signiﬁcantly over pure PS for exfoliated morphologies and less signiﬁcantly for intercalated morphologies. This study demonstrated one of the beneﬁts of polymer–clay nanocomposites, that the modulus, strength, and impact can be improved at the same time, under the right conditions. 13.5.1.3 Dynamic Mechanical Thermal Analysis Dynamic mechanical thermal analysis (DMTA) provides information about a materials ability to respond under dynamic deformation, including the change of the mechanical properties as a function of temperature (Table 13.5). The storage modulus (E′), loss modulus (E′′), and glass transition (Tg) may be determined through this technique. Several studies have investigated the effect of MMT in PS using DMTA [2, 14, 35, 36, 52]. Fu et al. [36] observed an increase in the storage modulus upon addition of MMT for bulk polymerized nanocomposites although the Tg decreased with increasing

13.5 Properties of PS–MMT Nanocomposites Prepared via In-situ Techniques Table 13.5

Summary of DMTA results.

Sample

PS PS-1% MMT PS-2% MMT PS-3% MMT

Storage modulus (log E′ (Pa)) 60 °C

90 °C

100 °C

9.0 9.1 9.3 9.4

7.6 8.2 9.0 9.1

7.1 7.7 8.3 9.1

Tg (°C)

83 87 99 107

Reproduced from Ref. [52] with permission.

MMT loading. They suggested this was an effect of the decreasing molecular weight as MMT loading increased. Uthirakumar et al. [52] investigated the effect of increasing the modiﬁed MMT concentration on the storage moduli and Tg (Table 13.4). It was observed that while E′ increased with increasing concentration at all temperatures, the E′ at temperatures closest to the Tg showed the highest improvements in E′. This behavior was attributed to the exfoliated nature of the nanocomposites. The Tg increased with increasing MMT concentration, due to the restriction in the polymer motions in the vicinity of the clay platelets, and the exfoliated morphology meant that clay platelets were well dispersed and could affect a larger proportion of the polymer chains. This observation of Tg behavior was different to what was observed by Fu et al. [36], most likely due to the different surface modiﬁcations employed. 13.5.1.4 Rheological Properties The rheological properties of exfoliated PS–MMT nanocomposites prepared through bulk polymerization utilizing clay containing reactive surfactants (40) was investigated by Zhong et al. [48] They found that the rheological behavior of pure PS was temperature dependent, but that of the nanocomposites was not. This behavior was attributed to the formation of a tridimensional clay network due to the high anisotropy of the clay, which was not affected by temperature. This work showed that the relaxation dynamics of these nanocomposites was dominated by the clay–clay association, which led to the stress relaxation being independent of temperature. The leveling-off of the storage modulus at low frequency oscillatory shear measurements in combination with this temperature independent behavior implied that there was a formation of a percolating nanoclay network, which was the reason for the enhancement in the viscoelastic properties of the PS–MMT nanocomposites. 13.5.1.5 Barrier Properties As with other mechanical properties, barrier properties are also often signiﬁcantly improved by clay. Typically, MMT can lead to a decrease in the gas permeability

355

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties 700 600 Oxygen Flux (cc m–2 day–1)

356

500 400 300 200

PS PS/VDAC-MMT10 PS/C18DMB-MMT10 PS/Na-MMT10

100 0

0

1

2

3

4

5

6

Time (h) Figure 13.9 Representative experimental oxygen ﬂux data obtained at 23 °C at 10% clay

loadings. Reproduced from Ref. [7] with permission.

of the nanocomposite, which is a useful property in applications such as food packaging. Nazarenko et al. [7] investigated the gas barrier properties of bulk in-situ polymerized PS–MMT as a function of mineral composition and morphology and the oxygen permeability is shown in Figure 13.9. They found that as the clay concentration increased, the permeability of oxygen decreased. Intercalated morphologies led to the lowest permeability of oxygen as has been observed in the past [1] due to the tortuous path created by the intercalated morphology, and exfoliated nanocomposites exhibited lower permeability than microcomposites. A gas transport model was postulated from these results. It was found however that in general the decrease in permeability was not as great as expected for these PS–MMT systems, and this was related to a level of clay layer aggregation which was still present even in the exfoliated samples. 13.5.2 Thermal Properties

Polymer–clay nanocomposites are known to improve the thermal stability, ﬂame retardance and Tg of the nanocomposite [1, 80, 103]. Many studies have investigated the thermal properties of PS–MMT nanocomposites. 13.5.2.1 Thermal Gravimetric Analysis The thermal stability of nanocomposites is generally measured using thermal gravimetric analysis (TGA), in which the mass loss is monitored as the temperature is increased at a speciﬁc rate. It is known that the addition of MMT to polymers generally improves the thermal stability of the nanocomposites for most

13.5 Properties of PS–MMT Nanocomposites Prepared via In-situ Techniques

polymer systems [1]. In general, most research has shown that the thermal degradation of the PS–MMT nanocomposites are improved upon the addition of MMT [25, 26, 29, 41, 45, 52–54, 77]. Wilkie et al. observed improvements in the peak degradation temperature of up to 50 °C [12, 32]. Wang et al. [14] investigated PS–MMT via bulk, emulsion, and solution polymerizations, and observed increases in the peak degradation temperature by up to 50 °C. The exfoliated nanocomposites improved more than the intercalated nanocomposites for all three methods. Investigations into the thermal degradation of PS in the presence of MMT have been undertaken by Jang and Wilkie [108] who studied the degradation products of PS–MMT and found that the degradation reactions in the PS–MMT were different to PS alone. In virgin PS, the degradation pathway consisted of chain scission followed by depolymerization (β-scission) through an intrachain reaction. The PS–MMT degradation was affected by radical recombination reactions taking place in the intergallery spaces of the MMT. The clay layers act as a barrier to heat and mass transfer, thus spreading the degradation time and reducing the peak degradation value. To further investigate the inﬂuence of MMT on the thermal stability of nanocomposites, Wilkie et al. investigated the effect of iron in the clay on the thermal stability [109]. PS–MMT nanocomposites were prepared by bulk polymerization using both iron-containing and iron-depleted clays. The presence of structural iron, rather than that present as an impurity, signiﬁcantly increased the onset temperature of thermal degradation in polymer–clay nanocomposites. Intercalated nanocomposites showed an iron effect, but it was less important for exfoliated systems and the effect of the iron was more signiﬁcant at lower clay loadings. The iron however did not appear to affect the char formation. From this work, the authors postulated that there were two factors that could improve thermal stability in PS–MMT nanocomposites: the barrier effect and the presence of iron which could function as a radical trap, helping to prevent the degradation of PS. Uthirakumar et al. [53] investigated PS–MMT at 1, 3, and 5 wt% prepared via bulk polymerization utilizing a radical initiator tethered to the MMT. The thermal stability increased as the clay loading increased from 1 to 3 wt% (exfoliated morphologies) but at 5 wt% (intercalated morphology) there was no real increase in thermal stability, and the TGA curves are shown in Figure 13.10. This was attributed to the level of exfoliation in the nanocomposites, with the 5% loading leading to intercalated morphologies rather than exfoliated. It could also be an effect of the decrease in the molecular weight of the nanocomposites containing higher levels of clay, attributed to the silicate platelets hindering the growth of polymer chains. While the presence of clay generally improves the thermal stability, some research has shown that in some cases the clay may actually decrease the thermal stability. Wang et al. [14] found that nanocomposites prepared via emulsion polymerization exhibited slightly poorer thermal stability on the addition of clay, and this may be an effect of the decrease in molecular weight by up to 66% after the addition of clay. Research has also focused on the effect of more thermally stable clays on overall thermal degradation. While more thermally stable cations such as phophonium

357

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties

100

80 Weight loss (%)

358

60 a bc d

40

20

0 100

200

300 400 500 600 Temperature (°C)

700

800

Figure 13.10 TGA curves of (a) PS (b) PS/MMT-1 (c) PS/MMT-3 and (d) PS/MMT-5, where MMT is modiﬁed with a surfactant. Reproduced from Ref. [53] with permission.

ions and other aromatic based ions may improve the thermal stability of the clays themselves, particularly during melt processing at high temperatures, the effect on the overall thermal stability of the resultant nanocomposite is not as straightforward. Zhu et al. [32] found that while alkyl phosphonium-modiﬁed clays had a higher thermal stability than a reactively modiﬁed clay, the thermal stability of the resultant nanocomposite was actually higher for reactively modiﬁed alkyl ammonium clay nanocomposites. The explanation was that the ﬁnal morphology of the nanocomposite was more important than the clay modiﬁcation for thermal stability. A similar ﬁnding was observed by Wang et al. [20]. Other surfactants investigated for higher thermal stabilities include benzimidiazolium [67], pyridium [17, 25, 26], quinolium [25, 26], and imidazolium [3, 37, 110] based ions. Again in these cases, while the thermal stability of the clay itself was greatly improved, the thermal stability of the nanocomposite was not necessarily improved. Fu et al. [42] investigated POSS modiﬁed clay and found that the thermal stability was enhanced over unﬁlled PS as well as reactively modiﬁed PS nanocomposites formed though emulsion polymerization. These results further indicate that the improvement in properties is due to several factors combined, and it is difﬁcult to dissociate the effect of each factor. 13.5.2.2 Dynamic Scanning Calorimetry (DSC) DSC is mainly used to observe the thermal transitions of polymers such as the glass transition temperature (Tg) or the melting point (Tm). Several studies have measured the Tg of PS–MMT nanocomposites by DSC [17–19, 37].

13.5 Properties of PS–MMT Nanocomposites Prepared via In-situ Techniques

Endo

PS

86.5

CLPS1 Heat Flow (mW)

91.7 96.5

CLPS3 CLPS5

97.1 99.4

CLPS10

Exo 60 75 80 85 90 95 100 105 110 115 120 125 130 Temperature (°C) Figure 13.11 DSC curves of PS and a series of PS–MMT nanocomposites with increasing clay loading. Reproduced from Ref. [18] with permission.

Essawy et al. [17] could not detect a Tg from DSC measurements for PS–MMT prepared using nonreactive clay via bulk polymerization. This was thought to be due to the interaction of the polymer chains with the MMT and the reduced rotational and translational mobility. Abate et al. [37] did detect Tg from DSC measurements and observed a small increase in the Tg when modiﬁed clay–PS nanocomposites were formed through bulk polymerization, and that the Tg and dispersion level increased as the alkyl chain length of the surfactant increased. Yeh et al. [18] observed a larger increase in Tg (5–13 °C) for PS–MMT nanocomposites produced through bulk polymerization, with the Tg increasing as the concentration of the MMT increased (Figure 13.11); however, no relationship between the Tg and the morphology was observed. Zhang et al. [19] observed similar trends. The increase in Tg is generally attributed to the conﬁnement of PS chains located within or near the clay layers. 13.5.2.3 Fire Performance The ﬂame retardance of polymer nanocomposites containing MMT has been known to improve over pristine PS since early studies in the ﬁeld [111, 112]. The ﬁre properties of nanocomposites are most often measured using a cone calorimeter, a small-scale oxygen consumption calorimeter in which a radiant heat source in the form of a conically shaped radiator can impose a heat ﬂux on the specimen surface, and the ignition is promoted using a spark igniter. Combustion gases are extracted in an exhaust duct where instrumentation measures exhaust gas ﬂow, temperature, O2, CO, and CO2 concentrations and smoke optical density. From these measurements, quantities such as heat release rate (HRR) and smoke production can be calculated, while the time to ignition is determined by observation.

359

360

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties

The peak heat release rate (PHRR) in particular has been used to determine and compare the ﬁre performance of nanocomposites, as this measurement gives information about the size of the ﬁre and the approximate ﬁre hazard [25]. In the early review by Gilman [111] different polymer matrixes and the thermal information obtained were compared. Early work found that even intercalated nanocomposites showed signiﬁcantly reduced peak and average HRR. 50–70% reductions in the PHRR have been reported for several polymer clay nanocomposites such as nylon-6, PS, and polypropylene [113]. Gilman argued that there was a common ﬁre retardant mechanism for all matrixes in polymer nanocomposites and suggested that it was the formation of a carbonaceous-silicate char that controlled the ﬂammability. The layered silicate may also act as an insulator and mass transport barrier, slowing the escape of volatile products generated as the polymer decomposes. It is not clear whether intercalated or exfoliated structures give better thermal stability, but what is known is that the ﬂame retardance can be improved while improving the mechanical properties. Several groups have measured the ﬂame retardance of PS–MMT nanocomposites [12, 25, 26, 29, 32]. Zhu and Wilkie [12, 32] investigated the ﬁre properties of PS–MMT nanocomposites prepared through bulk polymerization techniques using modiﬁed MMT modiﬁed with both reactive and unreactive surfactants. Cone calorimetry measurements indicated that the ﬂame retardance was enhanced even when as little as 0.1% clay was added [12]. They did not however observe a linear relationship between the amount of clay and the rate of heat release. PS–MMT containing phosphorus surfactants as clay modiﬁers were investigated for ﬁre performance and although these surfactants are more thermally stable themselves, there was no improvement in the PHRR as measured by a cone calorimeter over alkyl ammonium surfactants. While the formation of a nanocomposites (intercalated or exfoliated) was necessary to lead to improvement in ﬁre properties, it was found that intercalated morphologies showed the largest decrease in the PHRR compared to exfoliated morphologies. The reason for this is not clear although this has been observed in the past for other polymer matrices [30, 111, 112]. Chigwada et al. [26] investigated intercalated PS–MMT nanocomposites prepared through bulk polymerization containing MMT modiﬁed with quinolium and pyridinium surfactants and the cone calorimetric is presented in Table 13.6. Cone calorimetry indicated that the clays reduced the PHRR by up to 50% at 7% clay loading; however, the total heat release (THR) was not signiﬁcantly decreased. In a further study, they investigated the effect of carbazole containing surfactants with a range of alkyl chain lengths [29]. While the longer alkyl chains led to better clay dispersion, the PHRR was not affected by the alkyl chain length. The ﬂame resistance was improved for nanocomposites made through bulk polymerizations rather than through melt blending. MMT modiﬁed with a biphenyl containing surfactant and used in bulk polymerization formed intercalated nanocomposites and was also found to decrease the PHRR [25]. While the PHRR of PS–MMT nanocomposites is usually reduced over unﬁlled PS, most studies have not shown a signiﬁcant decrease in the THR, and so at this time the modiﬁed clay on its own is not sufﬁcient to provide complete ﬂame

13.6 Summary Cone calorimetric data for PS and nanocomposites made through bulk polymerization.

Table 13.6

Surfactant (Tables 13.1–13.3)

Clay loading

Time to ignition

PHRR (kW/m2) (% reduction)

Time to PHRR

THR (MJ m−2)

– 9 Quinolium 9 10 Pyridinium 10 10

– 3 5 3 5 7

63 42 20 51 44 25

± ± ± ± ± ±

1351 ± 87 1100 ± 59 (19) 806 ± 29 (40) 782 ± 70 (42) 762 ± 8 (44) 683 ± 28 (50)

126 ± 21 131 ± 13 106 ± 9 111 ± 10 106 ± 2 106 ± 6

100 ± 1 95 ± 3 88 ± 2 90 ± 3 82 ± 2 88 ± 4

4 5 3 4 4 4

Reproduced from Ref. [26] with permission.

protection. Chigwada et al. [33] used bromine modiﬁed clays (26) in bulk polymerized nanocomposites to further improve the ﬂame resistance, and they found that the PHRR decreased from the effect of the clay, and the THR also decreased due to the presence of bromine. This multifaceted approach to improving the ﬁre properties was successful although only intercalated morphologies were formed using these modiﬁed clays.

13.6 Summary

A large body of work has focused on the formation of PS–MMT nanocomposites utilizing in-situ methodologies, due to the fact that PS is a widely used polymer in many different applications, and that PS can be considered a model free radical polymerization for mechanistic and fundamental investigations. Several different polymerization techniques have been investigated to produce PS–MMT nanocomposites, including bulk, solution, and emulsion, as well as living and controlled polymerization techniques. The best dispersed nanocomposites have been formed using bulk and emulsion methodologies although the modiﬁcation of the clay is very important to allow sufﬁcient interaction between the clay and the styrene. A large number of modiﬁcations have been investigated and the presence of a reactive group in the modiﬁcation has been found to be essential to form exfoliated morphologies. PS–MMT nanocomposites have exhibited improved mechanical properties, including tensile, impact, rheological, and barrier properties. The thermal and ﬁre properties have also been improved upon addition of MMT, but it is evident that for all physical properties the morphology of the nanocomposite ultimately determines the level of improvement in properties. These studies have helped to develop a level of understanding of the importance of the interplay between the clay modiﬁcation, the polymerization technique, the

361

362

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties

dispersal method, and the ﬁnal morphology and how these relate to the ﬁnal properties of the nanocomposite. While melt intercalation and mixing methodologies are the preferred methods for commercially produced composites, exfoliated PS–MMT composites have only been achieved through in-situ polymerization techniques and future work in the ﬁeld will most likely focus on leveraging the understanding of the in-situ systems to help lead to better dispersed melt intercalated nanocomposites. Other future work in the ﬁeld will no doubt continue the mechanistic studies and exfoliation behavior studies of model PS–MMT composites to further contribute to the understanding of oleﬁn–clay nanocomposites.

References 1 Chen, B. (2004) Br. Ceram. Trans., 103, 2

3 4 5 6 7

8 9 10 11 12 13

14 15 16

17

241. Simons, R., Qiao, G.G., Powell, C.E., and Bateman, S.A. (2010) Langmuir, 26, 9023. Zhao, J., Morgan, A.B., and Harris, J.D. (2005) Polymer, 46, 8641. Sinha Ray, S. and Okamoto, M. (2003) Prog. Polym. Sci., 28, 1539. Fu, X.A. and Qutubuddin, S. (2005) J. Colloid Interface Sci., 283, 373. Liu, G., Zhang, L., Zhao, D., and Qu, X. (2005) J. Appl. Polym. Sci., 96, 1146. Nazarenko, S., Meneghetti, P., Julmon, P., Olson, B.G., and Qutubuddin, S. (2007) J. Polym. Sci., Part B: Polym. Phys., 45, 1733. Carastan, D.J. and Demarquette, N.R. (2007) Int. Mater. Rev., 52, 345. Liu, P. (2007) Appl. Clay Sci., 38, 64. Gao, F. (2004) Mater. Today, 7, 50. Yilmazer, U. and Ozden, G. (2006) Polym. Compos., 27, 249. Zhu, J. and Wilkie, C.A. (2000) Polym. Int., 49, 1158. Bruzaud, S., Grohens, Y., Ilinca, S., and Carpentier, J.F. (2005) Macromol. Mater. Eng., 290, 1106. Wang, D., Zhu, J., Yao, Q., and Wilkie, C.A. (2002) Chem. Mater., 14, 3837. Wang, D. and Wilkie, C.A. (2003) Polym. Degrad. Stab., 80, 171. Chen, G., Liu, S., Chen, S., and Qi, Z. (2001) Macromol. Chem. Phys., 202, 1189. Essawy, H.A., Badran, A.S., Youssef, A.M., and Abd El-Hakim, A. (2004) Macromol. Chem. Phys., 205, 2366.

18 Yeh, J., Liou, S., Lin, C., Chang, Y., Yu,

19

20

21 22

23 24

25

26

27

28

29

30

Y., and Cheng, C. (2004) J. Appl. Polym. Sci., 92, 1970. Zhang, W.A., Chen, D.Z., Xu, H.Y., Shen, X.F., and Fang, Y.E. (2003) Eur. Polym. J., 39, 2323. Wang, H.-W., Chang, K.-C., Chu, H.-C., Liou, S.-J., and Yeh, J.-M. (2004) J. Appl. Polym. Sci., 92, 2402. Chen, G., Ma, Y., and Qi, Z. (2001) Scripta Mater., 44, 125. Shen, Z., Zhu, F., Liu, D., Zeng, X., and Lin, S. (2005) J. Appl. Polym. Sci., 95, 1412. Li, H., Yu, Y., and Yang, Y. (2005) Eur. Polym. J., 41, 2016. Tseng, C.R., Wu, J.Y., Lee, H.Y., and Chang, F.C. (2002) J. Appl. Polym. Sci., 85, 1370. Chigwada, G., Wang, D., Jiang, D.D., and Wilkie, C.A. (2006) Polym. Degrad. Stab., 91, 755. Chigwada, G., Wang, D., and Wilkie, C.A. (2006) Polym. Degrad. Stab., 91, 848. Ianchis, R., Donescu, D., Petcu, C., Ghiurea, M., Anghel, D.F., Stanga, G., and Marcu, A. (2009) Appl. Clay Sci., 45, 164. Fu, H.-K., Huang, C.-F., Huang, J.-M., and Chang, F.-C. (2008) Polymer, 49, 1305. Chigwada, G., Jiang, D.D., and Wilkie, C.A. (2005) Thermochim. Acta, 436, 113. Gilman, J.W., Awad, W.H., Davis, R.D., Shields, J., Harris, R.H., Davis, C., Morgan, A.B., Sutto, T.E., Callahan, J.,

References

31

32

33

34 35 36 37

38

39 40

41

42 43 44 45 46

47

48 49

Trulove, P.C., and DeLong, H.C. (2002) Chem. Mater., 14, 3776. Bourbigot, S., VanderHart, D.L., Gilman, J.W., Awad, W.H., Davis, R.D., Morgan, A.B., and Wilkie, C.A. (2003) J. Polym. Sci., Part B: Polym. Phys., 41, 3188. Zhu, J., Morgan, A.B., Lamelas, F.J., and Wilkie, C.A. (2001) Chem. Mater., 13, 3774. Chigwada, G., Jash, P., Jiang, D.D., and Wilkie, C.A. (2005) Polym. Degrad. Stab., 88, 382. Zhao, Q. and Samulski, E.T. (2006) Polymer, 47, 663. Fu, X. and Qutubuddin, S. (2000) Mater. Lett., 42, 12. Fu, X. and Qutubuddin, S. (2001) Polymer, 42, 807. Abate, L., Blanco, I., Bottino, F.A., Pasquale, G., Fabbri, E., Orestano, A., and Pollicino, A. (2008) J. Therm. Anal. Calorim., 91, 681. Bottino, F.A., Fabbri, E., Fragalà, I.L., Malandrino, G., Orestano, A., Pilati, F., and Pollicino, A. (2003) Macromol. Rapid Commun., 24, 1079. Salem, N. and Shipp, D.A. (2005) Polymer, 46, 8573. Xu, L., Reeder, S., Thopasridharan, M., Ren, J., Shipp, D.A., and Krishnamoorti, R. (2005) Nanotechnology, 16, S514. Akelah, A., Rehab, A., Agag, T., and Betiha, M. (2007) J. Appl. Polym. Sci., 103, 3739. Fu, H.K., Huang, C.F., Huang, J.M., and Chang, F.C. (2008) Polymer, 49, 1305. Khalil, H., Mahajan, D., and Rafailovich, M. (2005) Polym. Int., 54, 423. Khalil, H., Mahajan, D., and Rafailovich, M. (2005) Polym. Int., 54, 428. Akelah, A. and Moet, A. (1996) J. Mater. Sci., 31, 3589. Samakande, A., Hartmann, P.C., and Sanderson, R.D. (2006) J. Colloid Interface Sci., 296, 316. Samakande, A., Hartmann, P.C., Cloete, V., and Sanderson, R.D. (2007) Polymer, 48, 1490. Zhong, Y., Zhu, Z., and Wang, S.-Q. (2005) Polymer, 46, 3006. Sedláková, Z., Pletil, J., Baldrian, J., Louf, M., and Holub, P. (2009) Polym. Bull., 63, 365.

50 Akat, H., Tasdelen, M.A., Prez, F.D.,

51

52

53

54

55

56 57 58

59

60

61

62

63

64

65 66 67

68

and Yagci, Y. (2008) Eur. Polym. J., 44, 1949. Uthirakumar, P., Kim, C.-J., Nahm, K.S., Hahn, Y.B., and Lee, Y.-S. (2004) Colloids Surfaces A, 247, 69. Uthirakumar, P., Song, M.-K., Nah, C., and Lee, Y.-S. (2005) Eur. Polym. J., 41, 211. Uthirakumar, P., Nahm, K.S., Hahn, Y.B., and Lee, Y.S. (2004) Eur. Polym. J., 40, 2437. Uthirakumar, P., Hahn, Y.B., Nahm, K.S., and Lee, Y.S. (2005) Eur. Polym. J., 41, 1582. Jeong, H.M., Choi, J.S., Ahn, Y.T., and Kwon, K.H. (2006) J. Appl. Polym. Sci., 99, 2841. Fan, X., Xia, C., and Advincula, R.C. (2003) Colloids Surfaces A, 219, 75. Fan, X., Xia, C., and Advincula, R.C. (2003) Langmuir, 19, 4381. Fan, X., Xia, C., Fulghum, T., Park, M.K., Locklin, J., and Advincula, R.C. (2003) Langmuir, 19, 916. Samakande, A., Sanderson, R.D., and Hartmann, P.C. (2009) Eur. Polym. J., 45, 649. Zhang, B.Q., Pan, C.Y., Hong, C.Y., Luan, B., and Shi, P.J. (2006) Macromol. Rapid Commun., 27, 97. Zhao, H., Argoti, S.D., Farrell, B.P., and Shipp, D.A. (2004) J. Polym. Sci., Part A: Polym. Chem., 42, 916. Weimer, M.W., Chen, H., Giannelis, E.P., and Sogah, D.Y. (1999) J. Am. Chem. Soc., 121, 1615. Arioli, R., Gonçalves, O.H., Castellares, L.G., da Costa, J.M., Araújo, P.H., Machado, R., and Bolzan, A. (2006) Macromol. Symp., 245–246, 337. Kim, T.H., Lim, S.T., Lee, C.H., Choi, H.J., and Jhon, M.S. (2003) J. Appl. Polym. Sci., 87, 2106. Perrin, F.X., Bruzaud, S., and Grohens, Y. (2010) Appl. Clay Sci., 49, 113. Calderon, J.U., Lennox, B., and Kamal, M.R. (2008) Appl. Clay Sci., 40, 90. Costache, M.C., Heidecker, M.J., Manias, E., Gupta, R.K., and Wilkie, C.A. (2007) Polym. Degrad. Stab., 92, 1753. He, H., Duchet, J., Galy, J., and Gerard, J.F. (2005) J. Colloid Interface Sci., 288, 171.

363

364

13 Polystyrene–Montmorillonite Nanocomposites by In-situ Polymerization and Their Properties 69 Bauer, F., Ernst, H., Hirsch, D.,

70

71

72

73 74 75 76

77 78

79

80

81 82

83 84

85 86 87 88

Naumov, S., Pelzing, M., Sauerland, V., and Mehnert, R. (2004) Macromol. Chem. Phys., 205, 1587. Bauer, F., Gläsel, H.J., Decker, U., Ernst, H., Freyer, A., Hartmann, E., Sauerland, V., and Mehnert, R. (2003) Prog. Org. Coat., 47, 147. Müh, E., Frey, H., Klee, J.E., and Mülhaupt, R. (2001) Adv. Funct. Mater., 11, 425. Wheeler, P.A., Wang, J., Baker, J., and Mathias, L.J. (2005) Chem. Mater., 17, 3012. Halliwell, S.M. (1999) BRE Centre for Building Fabric, (Digest 442). Zhang, M. and Singh, R.P. (2004) Mater. Lett., 58, 408. Shi, Z., Yu, D., Wang, Y., and Xu, R. (2003) J. Appl. Polym. Sci., 88, 194. Herrera, N.N., Letoffe, J.M., Putaux, J.L., David, L., and Bourgeat-Lami, E. (2004) Langmuir, 20, 1564. Moet, A.S. and Akelah, A. (1993) Mater. Lett., 18, 97. Simons, R., Qiao, G.G., Bateman, S.A., Zhang, X., and Lynch, P.A. (2011) Chem. Mater 23, 2303. Usuki, A., Kojima, Y., Kawasumi, M., Okada, A., Fujushima, Y., Kurauchi, T., and Kamigaito, O. (1993) J. Mater. Res., 8, 1179. Kojima, Y., Usuki, A., Kawasumi, M., Okada, A., Fujushima, A., Kurauchi, T., and Kamigaito, O. (1993) J. Mater. Res., 8, 1185. Lan, T., Kaviratna, P., and Pinnavaia, T. (1995) Chem. Mater., 7, 2144. Bergman, J.S., Coates, G.W., Chen, H., Giannelis, E.P., and Thomas, M.G. (1999) Chem. Commun., 2179. Stadtmueller, L.M., Ratinac, K.R., and Ringer, S.P. (2005) Polymer, 46, 9574. Uthirakumar, P., Kim, H.J., Hong, C.-H., Suh, E.-K., and Lee, Y.-S. (2008) Polym. Compos., 29, 142. Zhao, H., Farrell, B.P., and Shipp, D.A. (2004) Polymer, 45, 4473. Hwang, S.J., Joo, Y.L., and Lee, S.J. (2008) J. Appl. Polym. Sci., 110, 1441. Wang, C., Wang, Q., and Chen, X. (2005) Macromol. Mater. Eng., 290, 920. Wu, H., Kawaguchi, S., and Ito, K. (2004) Colloid. Polym. Sci., 282, 1365.

89 Yang, J., Fan, H., Bu, Z., and Li, B.

(2009) Polym. Eng. Sci., 49, 1937. 90 Fan, X.W., Xia, C.J., and Advincula, R.C.

(2005) Langmuir, 21, 2537. 91 Wang, J.-S. and Matyjaszewski, K. (1995)

J. Am. Chem. Soc., 117, 5614. 92 Kato, M., Kamigaito, M., Sawamoto, M.,

93 94 95

96

97 98

99

100 101

102

103

104 105

106

and Higashimura, T. (1995) Macromolecules, 28, 1721. Kamigaito, M., Ando, T., and Sawamoto, M. (2001) Chem. Rev., 101, 3689. Matyjaszewski, K. and Xia, J. (2001) Chem. Rev., 101, 2921. Chiefari, J., Chong, Y.K., Ercole, F., Krstina, J., Jeffery, J., Le, T.P.T., Mayadunne, R.T.A., Meijs, G.F., Moad, C.L., Moad, G., Rizzardo, E., and Thang, S.H. (1998) Macromolecules, 31, 5559. Le Tam, P., Moad, G., Rizzardo, E., Thang, S., Le, T.P., Thang, S.H., Le, T., Le, J.P., Phuong, L.T., Graeme, M., Ezio, R., and Hoa, T.S. (1998) Australia Patent WO9801478-A. Moad, G., Rizzardo, E., and Thang, S.H. (2005) Aust. J. Chem., 58, 379. Solomon, D.H., Rizzardo, E., and Cacioli, P. (1985) Substd. alkyl-amine cpds.— useful as controlled growth free radical polymerisation initiators for unsaturated monomers. Patent Number EP135280-A2. Georges, M.K., Veregin, R.P.N., Kazmaier, P.M., and Hamer, G.K. (1993) Macromolecules, 26, 2987. Hawker, C.J., Bosman, A.W., and Harth, E. (2001) Chem. Rev., 101, 3661. Wang, J., Hu, Y., Wang, S., and Chen, Z. (2005) Ultrason. Sonochem., 12, 165. Qi, R., Jin, X., Nie, J., Yu, W., and Zhou, C. (2005) J. Appl. Polym. Sci., 97, 201. Yano, K., Usuki, A., Okada, A., Kurauchi, T., and Kamigaito, O. (1993) J. Polym. Sci., Part A: Polym. Chem., 31, 2493. Chen, C., Khobaib, M., and Curliss, D. (2003) Prog. Org. Coat., 47, 376. Chen, J.H.S., Poliks, M.D., Ober, C.K., Zhang, Y., Wiesner, U., and Giannelis, E. (2002) Polymer, 43, 4895. Chin, I.J., Thurn-Albrecht, T., Kim, H.C., Russell, T.P., and Wang, J. (2001) Polymer, 42, 5947.

References 107 Okada, A. and Usuki, A. (1995) Mater. 108 109

110 111

Sci. Eng., C, 3, 109. Jang, B.N. and Wilkie, C.A. (2005) Polymer, 46, 2933. Zhu, J., Uhl, F.M., Morgan, A.B., and Wilkie, C.A. (2001) Chem. Mater., 13, 4649. Su, S., Jiang, D.D. and Wilkie, C.A. (2004) Polym. Degrad. Stab., 84, 269. Gilman, J.W. (1999) Appl. Clay Sci., 15, 31.

112 Gilman, J.W., Jackson, C.L., Morgan,

A.B., Harris, R., Manias, E., Giannelis, E.P., Wuthenow, M., Hilton, D., and Phillips, S.H. (2000) Chem. Mater., 12, 1866. 113 Nazaré, S., Kandola, B.K., and Horrocks, A.R. (2006) Polym. Adv. Technol., 17, 294.

365

367

14 Aliphatic Polyester and Poly(ester amide) Clay Nanocomposites by In-situ Polymerization Laura Morales-Gámez, Alfonso Rodríguez-Galán, Lourdes Franco, and Jordi Puiggalí

14.1 Introduction: Biodegradable Polymers and Their Nanocomposites

In recent years, interest in biodegradable polymers has grown because of their sustainability and specialized applications mainly in the biomedical ﬁeld. These polymers are becoming crucial for different industrial sectors like agriculture, automotive industry, medicine, and packaging, which require the use of environmentally friendly materials and, in some speciﬁc cases, biocompatible polymers. Speciﬁcally, polymers susceptible to being employed in packaging seem to be receiving more attention since it is estimated that more than 40% of plastics are used in this area. Biodegradable polymers can be classiﬁed into three categories according to their origin: (i) synthetic polymers, particularly aliphatic polyesters, such as poly (L-lactide) (PLA) [1–3], poly(ε-caprolactone) (PCL) [4–6], poly(p-dioxanone) (PPDO) [7–9], and poly(butylene succinate) (PBS) [10–12]; (ii) polyesters produced by microorganisms, which mainly correspond to different poly(hydroxyalkanoate)s (e.g., poly(β-hydroxybutyrate) and poly(3-hydroxybutyrate-co-3-hydroxyvalerate)); and (iii) polymers derived from natural resources (e.g., starch, cellulose, chitin, chitosan, lignin, and proteins). Despite the large investment in biodegradable polymers, it is clear that they are still far for becoming ideal substitutes of conventional synthetic polymers applied as commodities. In general, the main limitations are their high manufacturing costs and disadvantageous physical properties (e.g., poor mechanical and thermal properties, high hydrophilicity, and poor processability). Therefore, it seems very necessary to modify their chemical nature and/or enhance their performance by nanotechnology. In the ﬁrst case, the incorporation of amide groups into a polyester chain is an interesting option as poly(ester amide)s (PEAs) should establish strong intermolecular interactions that may improve properties and maintain degradability [13, 14]. In the second case, the incorporation of nanoscale particles into a polymer matrix should provide polymer/nanoparticle composites with enhanced properties. Although several preparation methodologies have been developed, in-situ polymerization in the presence of layered silicates appears more In-situ Synthesis of Polymer Nanocomposites, First Edition. Edited by Vikas Mittal. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA.

368

14 Polyester and Poly(ester amide) Clay Nanocomposites by In-situ Polymerization

appropriate to render exfoliated structures, which should have a greater impact on properties than intercalated structures [15]. This chapter reviews the state-of-art in the preparation of nanocomposites based on biodegradable aliphatic polyesters and PEAs by in-situ polymerization.

14.2 Aliphatic Polyester Clay Nanocomposites by In-situ Polymerization 14.2.1 Poly(ε-Caprolactone)-Based Nanocomposites

The in-situ polymerization technique has received much attention since the successful preparation of exfoliated nanocomposite structures from polyamides [16– 18] by promoting polymerization from initiators located on the clay surfaces. Early works on in-situ polymerization of ε-caprolactone were carried out by Messersmith and Giannelis [19]. A composite consisting of poly(ε-caprolactone)intercalated silicate particles embedded in the same polymer matrix was synthesized by heating Cr3+-ﬂuorohectorite (a synthetic mica type of silicate) to 100 °C in the presence of an excess of ε-caprolactone. Intercalation of the monomer was revealed by powder X-ray diffraction (XRD), which showed an increase in the silicate d-spacing from 1.28 to 1.46 nm, in consistency with a perpendicular orientation of the ε-caprolactone ring to the silicate layers. XRD patterns after polymerization indicated a reduction in the silicate spacing from 1.46 to 1.37 nm, which suggested a dimensional change accompanying polymerization. Thus, the opening of the lactone ring in the monomer gave rise to a monolayer of fully collapsed poly(ε-caprolactone) chains (Figure 14.1). The polymer matrix was isolated and found to be indistinguishable (except for chain end groups) from the bulk polymer obtained by metal-alkoxide-initiated polymerization. The polymerization reaction was postulated to proceed through cleavage of the acyl–oxygen bond catalyzed by the interlayer Cr3+ ions. The intercalated polymer was strongly adsorbed onto the silicate layers and showed no melting transition. Messersmith and Giannelis [20] also reported the preparation of nanocomposites by reacting protonated 12-aminolauric acid-exchanged montmorillonite with ε-caprolactone monomer. The monomer ring was initially intercalated in the gaps

O 1.28 nm

O OCr3+

O OCr3+

O OCr3+

O O 1.46 nm

Cr3+

O

C

Cr3+ 1.37 nm

n

Figure 14.1 Changes in the silicate spacing as a consequence of monomer intercalation and

subsequent polymerization.

14.2 Aliphatic Polyester Clay Nanocomposites by In-situ Polymerization

between the aminolauric acid chains so that no gallery expansion was detected in the XRD patterns. During heating of this mixture at 170 °C, the organic acid groups initiated ring-opening polymerization of the heterocyclic monomer by nucleophilic attack on the ε-caprolactone carbonyl. Diffraction patterns of powdered composite samples showed no discernable (0 0 1) reﬂections due to the organomodiﬁed clay, suggesting that individual silicate layers were dispersed in the polymer matrix. It was also determined that water permeability through composite ﬁlms containing modest amounts of silicate was dramatically reduced (e.g., by nearly an order of magnitude at only 4.8 v %) due to dispersion of impermeable high aspect ratio of silicate layers in the polymer matrix. The high average grafting density (with the areal grafting density × unperturbed radius of gyration >> 1) of the obtained nanocomposites made these systems excellent models to understand the static conformations and dynamics of polymer brushes, which is a focus of intense theoretical, experimental, and simulation interest. Similarities in the rheological response of end-tethered nanocomposites and block copolymers and smectic liquid crystals were demonstrated. The rheology of these end-tethered silicate nanocomposites was investigated using linear viscoelastic measurements in oscillatory shear with small strain amplitudes [21]. It was found that the storage and loss moduli increased at all frequencies with increasing silicate loading. However, their power law dependence on the terminal zone was different from that observed in homopolymers and decreased with increasing silicate loading. At low frequencies, the rheological response became almost invariant with frequency, which suggested a solid-like response. Interestingly, the system exhibited a nonterminal rheological behavior and had molecular weights which suggested a marginal entanglement. This was in close agreement with claims concerning lamellar block copolymers and smectic liquid crystals according to which the presence of entanglements was not required to observe nonterminal behavior. The in-situ polymerization technique has recently been applied for the preparation of PCL/layered aluminosilicate nanocomposites. Tin(II) octoate and dibutyltin(IV) dimethoxide were reported as appropriated transesteriﬁcation catalysts to promote the polymerization of ε-caprolactone in the presence of organomodiﬁed clays [22–25]. Montmorillonites with surfaces modiﬁed by ammonium cations bearing hydroxyl groups (e.g., bis(2-hydroxyethyl)methyl (hydrogenated tallow alkyl) ammonium) were found to be highly efﬁcient in initiating the polymerization and “grafting” the PCL chains onto the layered silicate surface. Lactone polymerization was postulated to be initiated by all the hydroxyl functions available at the clay surface after activation into either tin(II) or Al(III) alkoxide active species. Hybrid nanocomposites were accordingly generated through covalent grafting of every polyester chain onto the ﬁller surface. Surface-grafted polycaprolactone (PCL) chains were untied and isolated by ionic exchange reaction with LiCl in THF solution, and molar masses were measured by size exclusion chromatography. The PCL molecular weight and the extent of clay exfoliation could be controlled and readily tuned by the content of hydroxyl groups available at the clay surface [25]. Another interesting conclusion was that the initiation reaction by

369

370

14 Polyester and Poly(ester amide) Clay Nanocomposites by In-situ Polymerization

aluminum trialkoxide active species yielded grafted PCL chains characterized by a very narrow molecular weight distribution (Mw/Mn ∼ 1.2). Transport properties of water and dichloromethane vapors and mechanical properties were investigated for PCL chains grafted by in-situ polymerization onto clay modiﬁed by ammonium salts bearing hydroxyl groups. The clay content was ﬁxed at 3 wt% in inorganics, whereas the hydroxyl functionality ranged between 25 and 100%, always leading to clay delamination [26]. The increase in polymer chain grafting density resulted in a better exfoliation of the clays, limiting, for example, the diffusion of permeant water molecules. In the same way, the diffusion parameters of dichloromethane exhibited a decreasing value on increasing the hydroxyl content in the nanocomposites. Mechanical and dynamic mechanical analyses showed improvement in the nanocomposite elastic modulus in a wide temperature range. Interestingly, for the higher hydroxyl contents (50, 75, and 100%), the modulus could be measured at up to 120 °C (i.e., a much higher temperature than the melting temperature of pure PCL). A surfactant mixture that contained varying proportions of hydroxyl-substituted alkylammonium and unsubstituted alkylammonium cations to exchange the initial Na+ counterions of the natural montmorillonite was employed to have a tunable amount of hydroxyl functions at the surface of the clays. As previously indicated, those functions were then derivatized into aluminum alkoxides in order to initiate the ring-opening polymerization of ε-caprolactone directly from the clay surface, which was swollen in an organic solvent. Atomic force microscopy measurements on the resulting polymer-grafted nanoplatelets demonstrated the strong dependence of the coating of individual clay particles on the composition of the surfactant mixture used for the cationic exchange [27]. This allowed the generation of a range of morphologies varying from polymer islands distributed over the clay surface to homogeneous polymer layers thoroughly coating the platelets (Figure 14.2). Direct visualization of the polymer grafted onto the clay platelets gave relevant information on the structure of the obtained nanohybrid materials. First, the grafting density increased drastically as the proportion of OH-substituted alkylammonium cations used to organomodify the clay was raised. Second, the polymer deposit was not simply a continuous ﬁlm growing in thickness with increased OH content. Instead, separate polymer islands formed in the low-OH-content systems, probably as a result of a phase separation process between the ammonium ions induced by the polymerization reaction. Homogeneous coverage and subsequent thickening only took place from 50% OH content. When this situation was achieved, adjacent platelets became fully independent of each other since they were fully covered by the polymer and exfoliation was greatly favored. Bulk polymerizations of ε-caprolactone were also conducted at 170 °C in the presence of catalytic traces of water and 10, 30 and 50 wt% of hydrated synthetic montmorillonite SOMASIF ME100 without additional catalysts [28]. 1H NMR and GPC analyses suggested that the montmorillonite present in the system induced both signiﬁcantly higher lactone hydrolysis and polymer chain growth rates. All systems gave rise to low molecular weights (Mw = 5360–22 432 g mol–1), which

14.2 Aliphatic Polyester Clay Nanocomposites by In-situ Polymerization

Long alkyl chain amoniun anions

OH

OH

OH

O

O

O

Polymerizaiton MMT surface

MMT surface Polymerizaiton

Polymer patch, ~ 2-3 nm high

MMT surface

O

O

O

MMT surface

low grafting density < 50% OH grups > 50% OH grups high grafting density

MMT surface Grafted polymer layer, ~ 5-8 nm thick

Figure 14.2 Scheme of the polymer surface grafting onto individual clay platelets and

concomitant phase separation, as explained in detail in Ref. [26].

seemed to indicate a hindrance effect in diffusion caused by increasing amounts of silicate in the system. Wide-angle XRD data revealed that within the interlamellar regions of silicate, ε-caprolactone was arranged in weakly ordered bimolecular pseudo-layers. It is worth noting that the interlayer spacings of montmorillonite in the nanocomposites were slightly smaller than for silicate dispersed in εcaprolactone. Such results were attributed to an unfavorable polymerization of εcaprolactone in the interlayer region of the silicate (i.e., chains remained small and unable to open the clay layers to give rise a delaminated structure). The comparison between measured values of gallery height and calculated dimensions of the poly(ε-caprolactone) chain indicated that polymer chains were ﬂatly arranged on each side of the silicate platelet, creating pseudo-bilayers inside the montmorillonite gallery.

371

372

14 Polyester and Poly(ester amide) Clay Nanocomposites by In-situ Polymerization

Comprehension of the nanocomposite structure and interactions at the polymer– nanoﬁller interface is crucial in controlling nanocomposite properties. In all cases, nanocomposites are characterized by the presence of a polymer layer at the interface with the inorganic surface whose properties clearly differ from those of the bulk polymer. Pucciariello et al. [29] conducted a detailed surface analysis of poly(εcaprolactone)–montmorillonite clay nanocomposites obtained by in-situ polymerization. The organophilic clay was prepared by treating the natural montmorillonite with a solution of protonated 12-aminolauric acid. In-situ polymerization of εcaprolactone was performed under a nitrogen atmosphere at room temperature for 2 h and then at 170 °C for 48 h. The organomodiﬁed montmorillonite concentration was close to 18% as this was previously determined as the maximum content at which the silicate was exfoliated [30, 31]. The X-ray photoelectron spectroscopy technique gave information on atom concentrations in the surface layer, the valence state of these atoms, and the bonding of their nearest neighbors. Spectroscopic data indicated a great polarization of Si–O and Al–O bonds in the organomodiﬁed montmorillonite, which was attributed to the electron-attracting and electron-donor effects of the NH3+ and COOH groups of the aminolauric ion, which efﬁciently coordinated the partially negatively charged oxygen and the partially positively charged silicon (aluminum) of the clay (Figure 14.3a). X-ray photoelectron spectroscopy data of the nanocomposite revealed lower binding energies of Si–O and Al–O bonds, and consequently a reduced ionicity of such bonds with respect to the organophilic clay. Thus, the presence of the polymer limited the polarizing effect of the aminolauric cation because of the electron-donor/electronattracting effects of the ester group oxygen (Figure 14.3b). Poly(ε-caprolactone) was studied as a polymer biodegradable matrix which adds potential to the use of derived nanocomposites as biodegradable packaging materials. The synthesis of a series of montmorillonite–PCL nanocomposites in which the content of the inorganic material was varied regularly from 0 to 44 wt% was performed to ﬁnd some basic structure–property correlations of the multiphase nanocomposites and investigate the permeability to organic (e.g., dichloromethane) and inorganic (e.g., water) solvents of the multiphase polymers [32]. In-situ polymerization was carried out at 85 °C using a montmorillonite modiﬁed with protonated 12-aminolauric acid. Permeability to water and dichloromethane was found to decrease signiﬁcantly with increasing clay content. In particular, the water permeability behavior was largely dominated by the diffusion parameter. The diffusion path of the polar molecules of water was assumed to be slowed down with respect to dichloromethane vapor because of not only the physical barrier of the clay layers but also the hydrophilic character of the platelets. Barrier properties associated with the biodegradability of poly(ε-caprolactone) play an important role in enhancing interest in these nanocomposites as biodegradable packaging materials. PCL is thermodynamically miscible with many other polymers (e.g., PVC or SAN copolymer) and can be used as an environmentally decomposable additive that facilitates plastic degradation [33]. A study was recently undertaken to dissolve PCL–clay systems prepared by in-situ polymerization in a SAN matrix and evaluate

14.2 Aliphatic Polyester Clay Nanocomposites by In-situ Polymerization a)

δ+Si O δ– δ+Si O δ– H N+ H H

O=C HO COOH H H +N H

HO O=C δ+Si O δ–

δ+Si O δ–

δ+Si O δ– N+ H H H

H H H N+

COOH

δ+Si O δ–

b)

δ+Si O δ–

δ+Si O δ– N+ H

O=C HO

C O

O

H O

H

C O

Figure 14.3 Schemes explained in detail in Ref. [30] showing: (a) Coordination of the

aminolauric cation HOCO −(CH2 )11−NH3+ to the clay. (b) Model of the interactions occurring in the polymer–alkylammonium cation–clay system.

the structure and mechanical properties of these ternary nanocomposites [34]. The in-situ polymerization was performed at 170 °C using a commercial montmorillonite modiﬁed with hexadecyltrimethylammonium bromide. Clay content was of 10, 30, or 50 wt% since nanocomposites were then used as masterbatches to obtain SAN–(PCL–clay) ternary blends containing 0.66–5.65 wt% of the organomodiﬁed clay. Blends of PCL nanocomposites with SAN copolymer gave rise to intercalated and semi-exfoliated structures, as deduced from XRD and TEM techniques. The increase of clay in the system led to higher values of Young’s modulus. Thus,

373

374

14 Polyester and Poly(ester amide) Clay Nanocomposites by In-situ Polymerization

compared to unmodiﬁed SAN, the addition of only 0.66 wt% resulted in approximately a 10% increase in stiffness modulus, whereas at 5.65 wt%, a 40% increase was obtained. Supercritical CO2 has been studied as an alternative polymerization medium (e.g., for ring-opening polymerization of lactones [35, 36]) for more conventional organic solvents due to its well-known advantages [37]. Thus, it is environmentally friendly, nontoxic, nonﬂammable, and economical since its critical parameters are relatively easily obtained. Furthermore, its ﬂuid/solvent properties can be tuned by small changes in temperature. Preparation of nanocomposites by redispersion of masterbatches of exfoliated PCL nanocomposites into polymers miscible with PCL, such as the styrene– acrylonitrile copolymer [34], showed encouraging results but also drawbacks, for example, difﬁculty in recovering the aggregated bulk masterbatch and need to purify it before use. Detrembleur et al. [38] proposed an alternative method based on the preparation of PCL–clay masterbatches by in-situ intercalative polymerization in supercritical carbon dioxide. The speciﬁc properties of this solvent allowed in-situ polymerization at high clay content without viscosity problems. This is never the case for bulk polymerization processes, in which clay loading is usually limited to 30 wt%. Furthermore, the product obtained after depressurization was an easily recoverable ﬁne powder and supercritical CO2 was able to extract the residual monomer during depressurization, directly providing a ready-to-use dry powder. The ring-opening polymerization of ε-caprolactone in supercritical carbon dioxide occurred as a dispersion polymerization since only the monomer was soluble in this medium under typical supercritical conditions [39]. Catalysts like Sn(Oct)2 were preferably used as they should have low sensitivity to the carbonation reaction and protic impurities. Polymerizations of ε-caprolactone were carried out in supercritical CO2 (85 °C, 28 MPa) using the natural montmorillonite and the clay organomodiﬁed with a nonfunctional (Cloisite 20A) or a functional (Cloisite 30B) quaternary ammonium salt. In all cases, nanocomposites were qualiﬁed as “pre-exfoliated” masterbatches since true exfoliation could not be reached at high clay contents. Redispersion of organomodiﬁed clay–polymer masterbatches into chlorinated polyethylene was proved to be more efﬁcient in terms of quality of clay delamination than direct blending of commercial clay. Starch, a natural polymer constituted by linear α-glucan amylose and highly branched amylopectin, is considered one of the most promising candidates to be used as environmentally friendly materials. It has an attractive combination of availability, price, and performance, but its mechanical properties are poorer than those of synthetic polymers mainly due to its hydrophilic nature, which makes it sensitive to moisture content. For this reason, starch has been modiﬁed by blending with synthetic polymers such as poly(ε-caprolactone) and even with layered silicates [40, 41]. Namazi et al. [42] have recently prepared starch-g-polycaprolactone by in-situ ring-opening polymerization of ε-caprolactone in the presence of starch, Sn(Oct)2, and an organomodiﬁed montmorillonite (i.e., Cloisite 15A) at reaction temperatures between 100 and 150 °C. Results suggested a slight improve-

14.2 Aliphatic Polyester Clay Nanocomposites by In-situ Polymerization

ment in thermal stability but intercalation of the copolymer into clay galleries was less effective than the solution intercalation method. 14.2.2 Polylactide-Based Nanocomposites

Polylactide (PLA) is a well-known green polymer that receives great attention from the polymer industry since it can be produced from renewable resources (e.g., corn starch and other carbohydrate-rich substances like maize, sugar, or wheat) and is biodegradable and compostable. These advantages make PLA an attractive alternative to classical commodity polymers (e.g., in the production of loose-ﬁll packaging, compost bags, food packaging, and disposable tableware). However, several properties need to be improved to widen its range of applications. For example, PLA is too brittle and permeable to gases to render optimal application performances, especially for packaging purposes. Despite this, PLA is currently one of the most widely used specialty polymers in the biomedical ﬁeld (e.g., for sutures, stents, dialysis media, drug delivery devices, and even for tissue engineering). The introduction of a few percent of nanoﬁllers, such as layered aluminosilicate clays, has been extensively considered to enhance PLA properties. However, the greatest improvement is usually achieved when nanoparticles are fully and uniformly delaminated (exfoliated) in the polymer matrix, a challenge that cannot be fully met by direct melt blending of the clay as this usually leads to intercalated nanocomposites [43, 44]. The use of in-situ polymerization techniques appears fully justiﬁed since the monomer can penetrate and then polymerize inside the clay sheets, enhancing the delamination efﬁciency. In-situ ring-opening polymerization of the lactide monomer in the presence of clay has been extensively studied [45, 46]. For instance, PLA–MMT nanocomposites were prepared in bulk with catalysts such as tin(II) octoate (Sn(Oct)2) or triethylaluminum in the presence of montmorillonite clays (e.g., Cloisites 25A and 30B) organomodiﬁed with an ammonium salt functionalized or not with hydroxyl groups. When Cloisite 30B was used, polymerization was coinitiated by a molar equivalent of AlEt3 or Sn(Oct)2 with respect to the hydroxyl groups borne by the ammonium cations of the ﬁller, which was added before the l,l-lactide monomer. The presence of hydroxyl groups was found to be critical because these led to aluminum alkoxide or tin alkoxide active species (Figure 14.4 and Eq. (14.1)), which acted as initiators and led to polylactide chains grafted onto the clay surface Sn (Oct )2 + n OH → Sn(Oct )2−n (OR)n + n OctH

(14.1)

The molar ratio between hydroxyl groups and aluminum cations was crucial in promoting a controlled polymerization. Thus, a defect of aluminum cations should lead to more efﬁcient aluminum trialkoxides, but this was not the case when polymerizations were performed from lactide rings [46]. It was claimed that the probability to form a large amount of trialkoxide species from the anchored hydroxyl groups was rather low, and most probably a mixture of aluminum mono-,

375

14 Polyester and Poly(ester amide) Clay Nanocomposites by In-situ Polymerization Et Et

AI ]

n

Et Et Et

Et

Et

Et Et

Et

O

Et

Et Et

Et

Et AI Et O]n

AI

AI O]n

n

a)

]

376

O

AI

AI

]

AI

]

CH3 O CH3 O O CH3 O CH3 HO HO O R O O R O O R O O R O R OH R OH CH3 + CH3 + CH3 + CH3 + CH3 + CH3 + N N N N N N AIEt3 L, L-LA Silicate layer Silicate layer Silicate layer nOH / nAI = 1 bulk, 120°C, 48h AI

]

]

b) HO CH3

R OH

HO

R OH

N+

CH3

N+

Silicate layer AIEt3 nOH / nAI = 3 L, L-LA bulk, 120°C, 48h

HO

O

Et

Et

Et

O

AI

AI

AI

O

O

N+ CH3CH3 N+

O

O

CH3 N+

AI Et O

O

R OH

CH3 N+

+

N

CH3

Silicate layer

Figure 14.4 (a) Scheme of the l,l-lactide

(b) Scheme of a mixture of aluminum mono-, in-situ polymerization performed from Cloisite di-, and trialkoxides produced by addition of 30B using triethylaluminum (AlEt3) as the AlEt3 onto Cloisite 30B in a nOH/nAl = 3 molar initiator (R stands for tallow alkyl chain). ratio.

di-, and trialkoxides was formed (Figure 14.4b). This distribution was responsible for the observed low monomer conversion, loss of polymerization control, and bimodality of the molecular weight distribution. The grafted chains pushed the lamellar sheets apart from each other and led to the achievement of an excellent degree of clay exfoliation, which, for example, leads to improved thermal stability compared to unﬁlled PLA and even intercalated counterparts. Moreover, while the Tg and Tm of the PLA matrix were not inﬂuenced by the nanoﬁller, the degree of crystallinity of the polyester in the exfoliated structure was signiﬁcantly higher than in the intercalated nanocomposite [45]. The grafting reaction was conﬁrmed by the impossibility of dissolving the soproduced PLA chains in good solvents like toluene, THF, or CHCl3. In fact, a speciﬁc cationic exchange reaction with LiCl was required to recover the PLA chains, suggesting that polyester chains were attached to the ammonium cations localized in close vicinity and in electrostatic interaction with the montmorillonite surface. Interestingly, polymerization conditions (e.g., bulk with 3 wt% of Cloisite) allowed the synthesis of PLA grafts with a number average molecular weight close

14.2 Aliphatic Polyester Clay Nanocomposites by In-situ Polymerization

to 14 000 g mol−1 and a relatively narrow distribution for such a heterogeneous initiation process (i.e., polydispersity index was close to 1.5). Dubois et al. [46] also used in-situ polymerization to prepare a PLA–MMT masterbatch and conﬁrmed that its dilution with the neat PLA during melt processing was an effective technique for improving clay dispersion. It is well known that brittleness is a strong limitation for the application of PLA. The use of plasticizing agents is a usual procedure that may allow PLA to fulﬁll mechanical requirements. Considerable efforts have been made to improve PLA brittleness to make it competitive with low-cost ﬂexible commodity polymers (e.g., polyethylene and polypropylene). Several types of compounds such as citrate ester, poly(ethylene glycol) (PEG), glucose monoesters, partial fatty acid esters, oligomeric lactic acid, and glycerol have been studied as plasticizers for PLA [47–49]. However, the addition of a plasticizer generally reduces strength and modulus. Moreover, despite the increase of deformation, PLA-based materials having a good balance of stiffness and high deformation are still required for wide applications. For this reason, the possibility of preparing nanocomposites based on the PLA matrix and adding a plasticizer compound has been considered, speciﬁcally, the preparation of nanocomposites by melt blending a PLLA matrix, a plasticizer like PEG, and a nanoﬁller. Nevertheless, PEG tends to diffuse out of the material and accumulates at the nanocomposite surface, leading to structural matrix changes upon aging [50]. Thus, in-situ polymerization of l,l-lactide in the presence of both dihydroxylated PEG (Mw = 1000 g mol–1) and Cloisite 30B has been studied as an interesting alternative since it leads to nanocomposites based on a triblock copolymer matrix in which the central polyethylene block affords ﬂexibility and is not susceptible to diffusing out of the material [45]. Results showed that intensive clay platelet destructuration was achieved independent of the PEG weight ratio. The plasticizing effect of the PEG sequence (entrapped in the triblock copolymer) was highlighted by the signiﬁcant Tg decrease of the nanocomposite (e.g., from 60 to 12 °C at 16.2 wt% content in PEG). Moreover, the thermal degradation of the resulting nanocomposites was dependent on the relative content in PEG blocks and decreased as the polyether level increased within the triblock copolymer. Polylactide–vermiculite nanocomposites were also prepared by in-situ intercalative polymerization of l,l-lactide in the presence of vermiculite clay particles (VMT) that were treated with an alkylammonium surfactant in order to decrease their highly hydrophilic character and favor diffusion of the cyclic monomer into the clay interlayer spaces [51]. XRD suggested that exfoliated structures were attained and TEM observations revealed that VMT layers were exfoliated and dispersed uniformly in the polylactide matrix. TGA indicated a slight improvement in thermal stability (i.e., the onset temperature increased from 279 to 314 °C when 5 wt% of organomodiﬁed clay was added). Dynamic mechanical analyses showed an increase in storage and loss moduli caused by the reinforcing effect of the nanoscale VMT layers, as well as a slight increase of Tg by increasing the clay content. Nanocomposites based on anionic clays or layered double hydroxides (LDHs) have also been widely studied. These clays are composed of positively charged

377

378

14 Polyester and Poly(ester amide) Clay Nanocomposites by In-situ Polymerization

layers with anions and water molecules in the interlayer region. Compounds were deﬁned with the general formula (Mg1−xAlx(OH)2)x+(A−)x·nH2O, in which A− represents an interlamellar anion (e.g., carbonate) [52]. Although LDHs occur naturally, they are usually synthesized under controlled conditions in order to obtain materials with a known, homogeneous composition [53]. LDH platelets have a high aspect ratio, tuneable layer charge density, and can be prepared by low-cost processes. The potential suitability of organo-LDHs for the intercalation of hydrophobic polymers, the potentially greater susceptibility to complete exfoliation than that of cationic clays, and the catalytic activity for polymerization of lactides in the interlayer are also worth mentioning [54, 55]. Leroux and coworkers [56] suggested in-situ polymerization as an appropriate method for preparing polymer–LDH compounds because of the more conﬁned interlayer spacing compared to a typical unmodiﬁed montmorillonite clay (0.78 vs 1.26 nm). The high charge density in LDHs and their dense packing favor the insertion of small molecules such as the lactide dimer over the insertion of long PLA chains. This should lead to an effective dispersion of LDH platelets in the growing polymer matrix. Efforts have also focused on the insertion of new organic anionic species in LDH interlayers to enhance or modify hydrophobicity of LDHs [56, 57]. Plackett et al. [58] studied in-situ polymerization of the lactide dimer in the presence of LDHs modiﬁed with carbonate (LDH-CO3) or laurate units LDH (LDHC12). LDH-CO3 was synthesized by a conventional coprecipitation method [59], whereas LDH-C12 was prepared using the reconstruction method. In this procedure, the previously synthesized LDH-CO3 was calcined to form a mixed metal oxide and then dispersed into an ethanol–water solution containing sodium laureate. This process involved fast rehydration of the layered structure, followed by a slower anion exchange reaction, giving rise to the so-called memory effect mechanism [60]. XRD, scanning electron microscopy and transmission electron microscopy revealed that exfoliated nanocomposites were obtained when using LDH-C12 but that LDH-CO3 gave a partly phase-separated morphology (Figure 14.5). Thermogravimetric analysis showed that PLA–LDH combinations exhibited higher degradation onset temperatures than unﬁlled PLA. Differential scanning calorimetry indicated that both crystallinity and temperature of crystallization increased on adding LDH-C12 or LDH-CO3, suggesting that these additives have a nucleating effect. Although in-situ polymerization of lactide in the presence of 1–5 wt% LDH-C12 could be a promising method for producing nanocomposites with an exfoliated structure, the molecular weight was signiﬁcantly reduced when compared with the polymer synthesized in the absence of LDHs. Since this phenomenon also occurred when Mg(OH)2 was used instead of LDH, a chaintermination mechanism via LDH surface hydroxyl groups and/or metal-catalyzed degradation was proposed. From an industrial point of view, the melt-intercalation technique is usually preferred to in-situ polymerization because it is simpler and uses already existing technologies. In this sense, methods that combine the efﬁciency of the in-situ

14.2 Aliphatic Polyester Clay Nanocomposites by In-situ Polymerization

LDH with Lauric anions

Melt-ROP Sn based catalyst

Intercalated LDH/PLA nanocomposite

LDH with carbonate anions and water Exfoliated LDH/PLA nanocomposite Carbonate anion:

Water:

Lauric acid anions:

Lactide (L-Lactic cyclic dimer):

Figure 14.5 Ring-opening polymerization of lactide rings in the presence of layered double

hydroxides modiﬁed with carbonate or laurate units.

polymerization approach and the practicability of the melt-intercalation technique are currently being applied. Thus, a highly ﬁlled polymer–clay masterbatch was ﬁrst synthesized using an appropriate solvent and then dispersed into the commercial polymer by melt blending. Detrembleur et al. [61] used supercritical carbon dioxide as a polymerization medium for in-situ polymerization of d,l-lactide in the presence of different organomodiﬁed clays (e.g., C20A and C30B). Polymerizations were performed at 85 °C and 240 bar at which the lactide monomer was partially soluble whereas the formed polymer precipitated. The polymerization kinetic rate was observed to decrease with increasing the clay amount due to signiﬁcant hindrance of the clay. A slightly higher conversion was observed in the C30B-based systems since its hydroxyl groups acted efﬁciently as polymerization initiators. Studies carried out at low clay levels (3 wt%) demonstrated that ﬁnal structures were also strongly inﬂuenced by the nature of the organomodiﬁer compound since intercalated and exfoliated nanocomposites were obtained using C20A and C30B, respectively. Nanocomposites were also successfully obtained with clay levels as high as 35– 50% and were then useful as masterbatches to be mixed with the commercial PLA matrix. In this way, well-delaminated nanocomposites with 3 wt% of Cloisite 30B were attained, as deduced from TEM and XRD analysis. The nanocomposites showed signiﬁcant improvement in both stiffness (up to 20%) and toughness (e.g., from 5.1 to 6.0 kJ m−2) compared with the unﬁlled matrix.

379

380

14 Polyester and Poly(ester amide) Clay Nanocomposites by In-situ Polymerization

14.2.3 PBS-Based Nanocomposites

PBS is currently one of the most commonly applied biodegradable polyesters of the poly(alkylene dicarboxylate) family. It is commercialized by Showa Highpolymer as BIONOLLE and is usually copolymerized and blended to improve mechanical properties and biodegradability [62–65]. However, these methods (i.e., copolymerization and blending) often cause problems that affect crystallinity and melting point since the incorporation of a second component results in imperfect packing and/or isomorphism, and adversely affects the temperature range over which the resulting materials can be used [66, 67]. Preparation of nanocomposites appears as a promising alternative method to the production of commercial PBSbased polymers. In this case, the aggregation trend of inorganic materials caused by strong hydrophilic interactions should be avoided. Kim et al. [68] prepared nanocomposites by in-situ polymerization of 1,4-butane diol and succinic acid in the presence of Closite 25A or a twice functionalized organoclay (TFC) using titanium(IV) butoxide as a catalyst. TFC was prepared by reaction of C25A with a silane coupling agent that gave rise to epoxy functional groups. These were expected to enhance the chemical bonding of TFC layers with PBS end groups, and consequently lead to better dispersion and exfoliation of the clay layers in the PBS matrix. Furthermore, the thermal stability of PBS–TFC was clearly improved, which was attributed to the increased interaction between the inorganic components and the polymer. Lower thermal stability and an intercalated structure were characteristic of PBS–C25A nanocomposites. Nanocomposites based on the PBS matrix were prepared by two-step polymerization consisting of direct esteriﬁcation and polycondensation from succinic acid, 1,4-butanediol, titanium tetrabutoxide, and fumed silica [69]. The ﬁrst step was carried out at 190 °C for 2 h and the second under vacuum at 240 °C for 3–5 h. Solid-state 29Si NMR and FTIR analyses indicated that silanol-bonded carbonyl groups were established within PBS–silica nanocomposite materials. Rheological effects inherent to the silica ﬁller were evaluated and showed that despite high shear force, PBS–silica nanocomposites maintained a relatively high melt viscosity attributable to a network structure resulting from covalent bonding between silica and the polymer chain. Greatly improved mechanical properties were attained with only addition of 3.5 wt% of silica. Thus, an increase in both tensile strength (from 26.3 to 38.6 MPa) and elongation (from 56% to 515%) at break compared with parent PBS was reported. PBS–silica nanocomposites showed composition dependency on biodegradation ascribable to reduced crystallinity and preferential microbial attack. Im et al. [70] treated Closite 30B with the coupling agent 1,6-diisocyanatohexane to form covalent bonds between silanol groups on the clay side and between a silanol group and the hydroxyl group of the organomodiﬁer in the silicate layer (Figure 14.6). In this way, it was possible to increase both basal spacing and the favorable interaction between clay and PBS. Nanocomposites were prepared by two-step in-situ polymerization, as previously indicated. It was found that, at the

14.2 Aliphatic Polyester Clay Nanocomposites by In-situ Polymerization

OH

OH

OH OH CH2CH2OH

OH

CH3 N HT OH

OH

OH

CH2CH2OH OH OH

O NH-C=O

CH2CH2O + O=C=N

OH

OH

NH-C=O O=C-HN

O=C-HN OH

O

OH

O

OH

381

( )6

N=C=O

CH3 N HT Dibutyl tin dilaurate 24 h, 35 °C O=C-H

CH2CH2O O=C-HN

N

OH

OH

NH-C

=O

O

OH

Figure 14.6 Formation of covalent bonds between 1,6-diisocyanatohexane and the silanol and

hydroxyl groups in an organomodiﬁed montmorillonite as explained in Ref. [70].

same level of clay content, these nanocomposites showed a higher degree of exfoliation and improvement in all mechanical properties including tensile strength, storage modulus, and elongation at breaks compared with PBS and C30B nanocomposites. Dubois et al. [71] demonstrated that the enhancement of in-situ transesteriﬁcation reactions was a highly effective method to improve organoclay dispersion and exfoliation in a polyester matrix. Thus, Bionolle–Cloisite 30B nanocomposites were prepared by melt intercalation with addition of dibutyltin dilaurate ((Bu)2Sn(Lau)2) as a catalyst. Polyester chains were covalently linked to the organomodiﬁer as a result of effective transesteriﬁcation reactions occurring between the polymer chains and the hydroxyl groups of the organoclay. As a consequence of this process, clay exfoliation was enhanced and stiffness of the ﬁnal compound clearly increased (i.e., Young’s modulus showed an increase of 60% compared with the neat polyester). 14.2.4 PPDO-Based Nanocomposites

PPDO is a biodegradable polymer of great interest for applications in the biomedical ﬁeld (e.g., bioabsorbable surgical sutures, bone ﬁxation devices, and drug delivery systems) due to its outstanding mechanical properties, for instance, high tensile strength and excellent ﬂexibility [72]. This polymer has also been applied as a drug delivery system and has great potential for general use in systems such as ﬁlms, molded products, laminates, foams, nonwoven materials, adhesives, and coatings [73]. However, PPDO has some unfavorable characteristics (e.g., high hydrophobicity, low crystallization rate, and low melt strength) that may limit its

OH

NH-C=O OH

382

14 Polyester and Poly(ester amide) Clay Nanocomposites by In-situ Polymerization

application and processing. Modiﬁcations based on the preparation of nanocomposites now appear as an ideal method to improve physical properties and avoid some of the above limitations. In this line, Wang et al. [74] studied the in-situ ringopening polymerization of PPDO with organomodiﬁed montmorillonites (natural sodium montmorillonite, montmorillonite modiﬁed by octadecyltrimethyl ammonium chloride, and montmorillonite modiﬁed by hydroxyethylhexadecyldimethyl ammonium bromine) in the presence of triethylaluminum. Reactions were performed for approximately 20 h at 50 °C. Interestingly, montmorillonites accelerated the polymerization reaction and led to a viscosity-average molecular weight of PPDO of 44 900 g mol−1 in only 0.5 h. Furthermore, a nucleating effect of montmorillonites was clearly observed and the crystallization temperature of PPDO increased by approximately 18 °C. All three montmorillonites improved the thermal stability of PPDO and increased its glass-transition and melting temperature. In addition, the melt strength of the PPDO–MMT nanocomposites increased dramatically when compared with that of the neat PPDO [73]. These results suggested that biodegradable PPDO–MMT thin ﬁlms could be prepared by blowing process since low crystallization rate and low melt strength limitations can be overcome.

14.3 PEAs Clay Nanocomposites by In-situ Polymerization

PEAs are a new class of polymers that combine the good degradability of polyesters with the high thermal stability, high modulus, and high tensile strength of polyamides, which makes it possible to obtain good material and processing properties while maintaining degradability. PEAs are also highly attractive since their properties can be tuned due to the large variety of monomers that can be used (e.g., αamino acids, α,ω-aminoalcohols, and carbohydrates). Thus, polymers can be synthesized with variable ester/amide ratio, variable aliphatic/aromatic ratio, variable hydrophilicity (e.g., incorporating poly(ethylene oxide) blocks or changing the length of polymethylene sequences), variable stereochemistry, and variable monomer distribution. Thermoplastic elastomers and amorphous and semicrystalline materials can be obtained from segmented, random, and ordered microstructures, respectively. PEAs are now being considered for use as biodegradable matrices in nanocomposite preparation, although few works focus on the in-situ polymerization method. Nanocomposites of organomodiﬁed montmorillonites (C20A, C25A, and C30B) and a biodegradable PEA were obtained by in-situ polycondensation of sodium chloroacetylaminohexanoate [75]. This synthesis was based on a thermal polycondensation reaction in which the formation of a metal halide salt became the driving force of the process [76, 77]. ClCH2CONH(CH2 )n −1 COO− Na + → [OCH2CONH(CH2 )n −1 CO]x− + x Na + Cl (14.2)

14.3 PEAs Clay Nanocomposites by In-situ Polymerization a)

b) –1.5

NaCl 400

38 Monomer 15

16

17 18 Q (nm–1)

7

t (min)

19

(a) WAXD proﬁles taken during in-situ polymerization of the monomer/C25A mixture at 125 °C. It can be observed that the disappearance and appearance of reﬂections associated to the monomer and polymer structures, respectively, and the evolution of the (1 0 0) reﬂection of the NaCl structure. Sodium chloride was produced during polymerization according to Eq. (14.2).

Figure 14.7

neat monomer neat monomer/C25A mixture

–2.5

y = –12.16x + 27.82

–3.5 In k

350

/(a.u.)

450

Polymer

y = –11.95x + 26.89

–4.5 –5.5 2.4

2.6 2.5 1000/T (K–1)

2.7

(b) Plot of ln k versus the reciprocal of the polymerization temperature for the neat monomer and its mixture with C25A. Clear differences were observed between the intersections with the ordinate axis, but no signiﬁcant differences were found between slopes and consequently the corresponding activation energies.

The great simplicity of this method raises interest in this family of polymers characterized by a semicrystalline character due to the regular distribution of the two units involved, which contrasts with the more amorphous nature of copolymers prepared by ring-opening polymerization. Exfoliated or intercalated structures were attained depending on the organomodiﬁer (i.e., the most dispersed structure was obtained by addition of Cloisite 25A). Polymerization kinetics was strongly inﬂuenced by the presence of organomodiﬁed montmorillonites under both nonisothermal and isothermal conditions. Both FTIR and WAXD experiments were used to study the polymerization process following the time evolution of the 1742 cm−1 absorption band of the ester group and the diffraction intensity of characteristic NaCl reﬂections (Figure 14.7a), respectively. The temperature dependence of the polymerization kinetic constant allowed inferring that kinetic differences between the polymerization of the neat monomer and its mixture with C25A could be attributed to the preexponential frequency factor (Figure 14.7b). In this way, clay particles seemed to reduce chain mobility and the frequency at which reactive groups were close enough to facilitate the condensation reaction. The thermal stability and crystallization behavior of the neat polymer and its nanocomposites were signiﬁcantly different. In general, exfoliated structures decreased both primary nucleation and crystal growth rate, whereas intercalated structures increased the density of primary nuclei. These nanocomposites were also prepared by the melt mixing technique, which rendered intercalated structures with a higher overall crystallization rate [78].

383

384

14 Polyester and Poly(ester amide) Clay Nanocomposites by In-situ Polymerization

14.4 Conclusion

The increasing demand for biodegradable materials in the world market is leading to the development of products with better properties than those of existing materials. Efforts mainly focus on the synthesis of new biodegradable polymers with potential applications as both commodity and specialty materials. In this sense, PEAs can be considered a modiﬁcation of conventional biodegradable polyesters that may provide enhanced properties. Incorporation of nanoparticles into biodegradable polymer matrices is another effective approach to improve the properties of pristine polymers remarkably. There is now a considerable body of literature concerning the preparation of nanocomposites from the more usual biodegradable polyesters (e.g., poly(ε-caprolactone), polylactide, poly(butylene succinate), and poly(p-dioxanone)) and silicate clays by in-situ polymerization. In general, this results in polymers grafted onto the silicate layers, favoring a subsequent exfoliation. Thus, in-situ polymerization seems to be more appropriate than other methodologies (i.e., melt or solution intercalation) that usually render intercalated structures. Despite the increasing interest in developing new biodegradable polymers, few works deal with the preparation of nanocomposites based on PEAs matrices. Thus, this may be a very promising topic for future research.

Acknowledgments

Authors want to indicate the support by CICYT and FEDER grants (MAT200911503) and by the Agència de Gestió d’Ajuts Universitaris i de Recerca (2009SGR-1208).

References 1 Perepelkin, K.E. (2002) Fibre Chem., 34, 2 3

4

5

6

85. Agrawal, A.K. and Bhalla, R. (2003) J. Macromol. Sci. Polym. Rev., C43, 479. Lee, J.H., Park, T.J., Park, H.S., Lee, D.S., Lee, Y.K., Yoon, S.C., and Nam, J.D. (2003) Biomaterials, 24, 2773. Kai, W.H., Hua, L., Zhao, L., and Inoue, Y. (2006) Macromol. Rapid. Commun., 27, 1702. Li, H.Y., Debuigne, A., Jerome, R., and Lecomte, P. (2006) Angew. Chem. Int. Ed., 45, 2264. Luong-Van, E., Grondahl, L., Chua, K.N., Leong, K.W., Nurcombe, V., and Cool, S.M. (2006) Biomaterials, 27, 2042.

7 Sabino, M.A., Feijoo, J.L., and

8

9

10 11

12

Muller, A.J. (2000) Macromol. Chem. Phys., 201, 2687. Pezzin, A.P.T., van Ekenstein, G.O.R., and Duek, E.A.R. (2001) Polymer, 42, 8303. Yang, K.K., Wang, X.L., and Wang, Y.Z. (2002) J. Macromol. Sci. Polym. Rev., C42, 373. Kim, J., Shin, T.K., Choi, H.J., and Jhon, M.S. (1999) Polymer, 40, 6873. Carroccio, S., Rizzarelli, P., Puglisi, C., and Montaudo, G. (2004) Macromolecules, 37, 6576. Dong, T., Shin, K.M., Zhu, B., and Inoue, Y. (2006) Macromolecules, 39, 2427.

References 13 Okada, M. (2002) Prog. Polym. Sci., 27, 14

15 16

17

18

19 20

21 22

23

24

25

26

27

28

29

87. Lips, P.A.M., and Dijkstra, P.J. (2005) Biodegradable Polymers for Industrials Applications, CRC Press, Boca Raton, FL. Ray, S.S. and Okamoto, M. (2003) Prog. Polym. Sci., 28, 1539. Usuki, A., Kojima, K., Kawasumi, M., Okada, A., Fukushima, Y., Kurauchi, T., and Kamigaito, O. (1993) J. Mater. Res., 8, 1179. Yano, K., Usuki, A., Okada, A., Kurauchi, T., and Kamigaito, O. (1993) J. Polym. Sci., Part A: Polym. Chem., 31, 2493. Kojima, Y., Kukumori, K., Usuki, A., Okada, A., and Kurauchi, T. (1993) J. Mater. Sci. Lett., 12, 889. Messersmith, P.B. and Giannelis, E.P. (1993) Chem. Mater., 5, 1064. Messersmith, P.B. and Giannelis, E.P. (1995) J. Polym. Sci, Part A: Polym. Chem., 33, 1047. Krishnamoorti, R. and Giannelis, E.P. (1997) Macromolecules, 30, 4097. Lepoittevin, B., Panstoustier, N., Alexandre, M., Calberg, C., Jérôme, R., and Dubois, P.H. (2002) Macromol. Symp., 183, 95. Kubies, D., Pantoustier, N., Dubois, P., Rulmont, A., and Jérôme, R. (2002) Macromolecules, 35, 3318. Lepoittevin, B., Pantoustier, N., Devalckenaere, M., Alexandre, M., Calberg, C., Jérôme, R., and Dubois, P.H. (2002) Macromolecules, 35, 8385. Lepoittevin, B., Panstoustier, N., Alexandre, M., Calberg, C., Jérôme, R., and Dubois, P.H. (2002) J. Mater. Chem., 12, 3528. Gorrasi, G., Tortora, M.S., Vittoria, V., Pollet, E., Alexandre, M., and Dubois, P.H. (2004) J. Polym. Sci., Part B: Polym. Phys., 42, 1466. Viville, P., Lazzaroni, R., Pollet, E., Alexandre, M., and Dubois, P.H. (2004) J. Am. Chem. Soc., 126, 9007. Kiersnowski, A., Dabrowski, P., Budde, H., Kressler, J., and Pigłowski, J. (2004) Eur. Polym. J., 40, 2591. Pucciariello, R., Villani, V., Langerame, F., Gorrasi, G., and Vittoria, V. (2004) J. Polym. Sci., Part B: Polym. Phys., 42, 3907.

30 Gorrasi, G., Tortora, M., Vittoria, V.,

31

32

33 34 35 36 37 38

39 40

41

42

43

44 45

46

47 48

Pollet, E., Lepoittevin, B., Alexandre, M., and Dubois, P. (2003) Polymer, 44, 2271. Pucciariello, R., Villani, V., Belviso, S., Gorrasi, G., Tortora, M., and Vittoria, V.J. (2004) J. Polym. Sci., Part B: Polym. Phys., 42, 1321. Tortora, M., Vittoria, V., Galli, G., Ritrovati, S., and Chiellini, E. (2002) Macromol. Mater. Eng., 287, 243. Eastmond, G.C. (1999) Adv. Polym. Sci., 149, 59. Kiersnowski, A. and Piglowski, J. (2004) Eur. Polym. J., 40, 1199. Stassin, F., Halleux, O., and Jérôme, R. (2001) Macromolecules, 34, 775. Bratton, D., Brown, M., and Howdle, S.M. (2002) Macromolecules, 36, 5908. Wells, S.L. and DeSimone, J.M. (2001) Angew. Chem. Int. Ed., 40, 518. Urbanczyk, L., Calberg, C., Stassin, F., Alexandre, M., Jérôme, R., Jérôme, C., and Detrembleur, C. (2008) Polymer, 49, 3979. Stassin, F. and Jérôme, R. (2001) Macromolecules, 34, 775. Avella, M., Vlieger, J.J.D., Errico, M.E., Fischer, S., Vacca, P., and Volpe, M.G. (2005) Food Chem., 93, 467. Pandey, J.K., Kumar, A.P., Misra, M., Mohanty, A.K., Drzal, L.T., and Singh, R.P. (2005) J. Nanosci. Nanotechnol., 5, 497. Namazi, H., Mosadegh, M., and Dadkhah, A. (2009) Carbohyd. Polym., 75, 665. Ray, S.S., Maiti, P., Okamoto, M., Yamada, K., and Ueda, K. (2002) Macromolecules, 35, 3104. Ray, S.S., Okamoto, K., Yamada, K., and Okamoto, M. (2002) Nano. Lett., 2, 423. Paul, M.A., Alexandre, M., Degée, P., Calberg, C., Jérôme, R., and Dubois, P. (2003) Macromol. Rapid. Commun., 24, 66. Paul, M.A., Delcourt, C., Alexandre, M., Degée, P., Monteverde, F., Rulmont, A., and Dubois, P. (2005) Macromol. Chem. Phys., 206, 484. Jacobsen, S. and Fritz, H.G. (1999) Polym. Eng. Sci., 39, 1303. Martin, O. and Averous, L. (2001) Polymer, 42, 6209.

385

386

14 Polyester and Poly(ester amide) Clay Nanocomposites by In-situ Polymerization 49 Baiardo, M., Frisoni, G., Scandola, M.,

50

51

52 53 54 55

56 57

58

59 60 61

62 63

Rimelen, M., Lips, D., Rufﬁeux, K., and Wintermantel, E. (2003) J. Appl. Polym. Sci., 90, 1731. Hu, Y., Rogunova, M., Topolkaraev, V., Hiltner, A., and Baer, E. (2003) Polymer, 44, 5701. Zhang, J.H., Zhuang, W., Zhang, Q., Liu, B., Wang, W., Hu, B.X., and Shen, J. (2007) Polym. Compos., 28, 545. Saber, O. and Tagaya, H. (2008) Mater. Chem. Phys., 108, 449. Pluta, M. (2004) Polymer, 45, 8239. Li, L., Ma, R.Z., Ebina, Y., Iyi, N., and Sasaki, T. (2005) Chem. Mat., 17, 4386. Zammarano, M., Bellayer, S., Gilman, J.W., Franceschi, M., Beyer, F.L., Harris, R.H., and Meriani, S. (2006) Polymer, 47, 652. Sabbar, E.M., Roy, M.E., and Leroux, F. (2006) J. Phys. Chem. Solids, 67, 2419. Jaubertie, C., Holgado, M.J., San Roman, M.S., and Rives, V. (2006) Chem. Mater, 18, 3114. Katiyar, V., Gerds, N., Koch, C.B., Risbo, J., Hansen, H.C.B., and Plackett, D. (2010) Polym. Degrad. Stabil., 95, 2563. Miyata, S. (1975) Clays Clay Miner., 23, 369. Cavani, F., Triﬁro, F., and Vaccari, A. (1991) Catal. Today, 11, 173. Urbanczyk, L., Ngoundjo, F., Alexandre, M., Jérôme, C., Detrembleur, C., and Calberg, C. (2009) Eur. Polym. J., 45, 643. Fujimaki, T. (1998) Polym. Degrad. Stabil., 59, 209. Yoo, Y.T., Lee, B.J., Han, S.I., Im, S.S., and Kim, D.K. (2003) Polym. Degrad. Stabil., 79, 257.

64 Han, S.I., Kang, S.W., Kim, B.S., and

65 66 67 68

69

70 71

72

73 74

75

76 77 78

Im, S.S. (2005) Adv. Funct. Mater., 15, 367. Han, S.I., Yoo, Y., Kim, D.K., and Im, S.S. (2004) Macromol. Biosci., 4, 199. Takiyama, E. and Fujimaki, T. (1992) Plastics, 43, 87. Abe, H., Doi, Y., Hori, Y., and Hagiwara, T. (1998) Polymer, 39, 59. Kim, H.S., Chen, G.X., Jin, H.J., and Yoon, J.S. (2008) Colloids Surfaces A, 313–314, 56. Han, S., Lim, J.S., Kim, D.K., Kim, M.N., and Im, S.S. (2008) Polym. Degrad. Stabil., 93, 889. Hwang, S., Yoo, E.S., and Im, S.S. (2009) Polym. Degrad. Stabil., 94, 2163. Pollet, E., Delcourt, C., Alexandre, M., and Dubois, P.H. (2006) Eur. Polym. J., 42, 1330. Yang, K.K., Wang, X.L., and Wang, Y.Z. (2002) J. Macromol. Sci. Part C Polym. Rev., 42, 373. Yang, K.K., Wang, X.L., and Wang, Y.Z. (2007) J. Ind. Eng. Chem., 13, 485. Huang, F.Y., Wang, Y.Z., Wang, X.L., Yang, K.K., Zhou, Q., and Huang, F.Y. (2005) J. Polym. Sci., Part A: Polym. Chem., 43, 2298. Morales, L., Franco, L., Casas, M.T., and Puiggalí, J. (2009) J. Polym. Sci., Part A: Polym. Chem., 47, 3616. Epple, M., and Kirschnick, H. (1996) Chem. Ber., 129, 1123. Herzberg, O., and Epple, M. (2001) Eur. J. Inorg. Chem., 2001, 1395. Morales, L., Franco, L., Casas, M.T., and Puiggalí, J. (2011) Polym. Eng. Sci., 51, 1650.

387

Index a AB/AABB type polyamides 27, 36 AC conductivity 323 accelerator 246 acids – amino, see amino acids – 12-aminolauric 12–14, 38–39, 372–373 – hydrolysis 143–145 – layered silicic 285 – surface acidity 79 – unsaturated dicarboxylic 245 acrylate terminated polyurethane (ATPU) 259 activation – energy 115–116 – metallocenes 75 – nickel catalysts 77 ADA (amino acid) 262–263 addition-fragmentation transfer 351 additives 71–72 adipate-modiﬁed LDH 47 adsorbed monomers 91 agglomeration tendency 127 aggregates 56 AIBN (2,2′-azobis(isobutyronitrile)) 89 alcohol, furfuryl 135–136 AlEt3 (triethylaluminum) 376 aligned ﬁller 8 aliphatic polyester–clay nanocomposites 367–386 alkoxy silanes 275 alkyds 261 alkylaluminoxanes 76 alkylaluminum compounds 66 alkylammonium cations 370 3-alkylthiophenes 303 alumina layer 132

aluminosilicates – inorganic 222 – layered 1 – nanocomposites 369 aluminum alkyl cocatalyst 294 amines 59 – amine : epoxy mole ratio 223–224 amino acids – ADA 262–263 – carbon number 37 – nanocomposites 36–41 12-aminolauric acid (ALA) 12–14, 38–39, 372–373 ammonium – chloride 207 – group 344 – quaternary cations 65, 73 – stearyl 11 – surfactants 290 amorphous lamellae 198 amphiphilicity 271 anhydrides 245 anhydrous melt polymerization 43 anhydrous prepolymerization 33 anionic polymerization 289–293 (ar-vinyl-benzyl)trimethyl ammonium (VBTA) 296 aromatic polyamides 28 “arrowhead” shape 150 aspect ratio – CNTs 303 – CWs 126 asymmetric carbon 288 atom transfer radical polymerization (ATRP) 23, 99, 267–281 – controlled 351 2,2′-azobis(isobutyronitrile) (AIBN) 89

In-situ Synthesis of Polymer Nanocomposites, First Edition. Edited by Vikas Mittal. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA.

388

Index

b barrier properties 204–207 – epoxy nanocomposites 238–240 – nanocomposites 118–121 – PS/MMT nanocomposites 355–356 basal spacing 5, 15 – epoxy nanocomposites 232 – montmorillonites 37 bentonite suspensions 256 benzimidiazolium 358 benzyldimethylhexadecylammonium (BzC16) 226, 233–237 bio-based nanocomposites 123–167 bioabsorbable surgical sutures 381 biodegradable polymers 367–368 BIR (butadiene-isoprene copolymer) 291 birefringence 141 bis(hydroxyethyl) terephthalate (BHET) 106 bisphenol A 15–17, 259 blending 267, 309 block copolymers 23 – brushes 270–271 bone ﬁxation devices 381 BR (butadiene rubber) 288–289 Bragg’s law 151 brushes – block copolymer 270–271 – comb polymer 273 “built-in” surface functionality 124 bulk polymerization 347–348 1,3-butadiene – BIR 291 – BR 288–289 – polymerization methods 287–289 BzC16 (benzyldimethylhexadecylammonium) 226, 233–237

c 2C18 (dioctadecyldimethylammonium) 226 calcined clays 75 calorimetric analysis 314, 358–359 calorimetry 360–361 caprolactam 12 – ring-opening polymerization 40 carbon, asymmetric 288 carbon dioxide (CO2) 374 carbon nanotubes (CNTs) 34 – DWNTs 305 – functionalization 308 – MWCNTs, see multiwall carbon nanotubes – polymer nanocomposites 100 – PUCNs 170–188 carbon number 37 carbonate units 379

carboxylation 171 catalysts – activators 74–76 – catalyst : clay ratio 81–82 – clay-based 298 – clay-supported 59, 294 – heterogeneous 297 – karstecit 278 – nickel 77 – Phillips 60 – stereoselectivity 84 – supports 63–64, 72–74, 77, 81, 294 – tacticity 84 – Ziegler–Natta 60, 287 cation exchange 14, 39 – CEC 58, 81 – clay reactions 58 – mechanism 45 cationic ROP 274 cationic surfactants 334–339 cations – hydroxyl-substituted alkylammonium 370 – interlayer 59 – organic 241 – quaternary ammonium 65, 73 C=C antisymmetric stretching 322 CEC (cation exchange capacity) 58, 81 CED (cohesive energy density) 177 cell strut 186 cellulose – microcrystalline 143–145 – molecular structure 128 cellulose ﬁber nanocomposites 99–100 cellulose whisker (CW) nanocomposites 128–132 – reactive molding 136, 140–149 chain length 13 chains, polyacrylate 272 char retention 161 characterization – P3HT/MWNT nanocomposites 310 – PUCLNs 196–200 – PUCNs 176–183 – PUFGNs 210–214 – techniques 138–140 charge transportability 322–326 chemical reactions, see reactions chemical vapor deposition (CVD) 306–307 chiral map 306 clarity 119 clays 284–286 – aliphatic polyester–clay nanocomposites 367–386 – calcined 75

Index – – – – – –

catalyst : clay ratio 81–82 catalyst activators 74–76 chemical reactions 58–59 clay action 42–43 clay-based catalysts 298 clay modiﬁed HUP resins/alkyds 261–262 – clay–PS particles 276–277 – clay state 196 – clay-supported catalysts 294 – epoxy/clay nanocomposites 268 – GASP 95–96 – general structure 54 – in-situ ATRP 267–281 – intercalated 107 – lamellae 285 – laponite 275 – layered 107 – morphological hierarchy 56, 249 – nanoclay-reinforced nanocomposites 34 – organo-, see organoclays – PEA/clay nanocomposites 382–383 – PET/clay nanocomposites 105–122 – polybutadiene clay nanocomposites 283–301 – polymerization additive 71–72 – polyol–nanoclay mixtures 194–196 – polyoleﬁn–clay nanocomposites 53–88 – PUCLNs 188–208 – surface modiﬁcation 64–66, 80–81 – synthetic 39, 285 – types 78–79 Cloisite 73–74, 80, 139, 375–376 CNFs (cellulose nanoﬁbers) 99–100 CNTs (carbon nanotubes) 34 – CNT-reinforced nanocomposites 34 – DWNTs 305 – functionalization 308 – MWCNTs, see multiwall carbon nanotubes – polymer nanocomposites 100 – PUCNs 170–188 CO2 (carbon dioxide), supercritical 374 coatings 91–92, 204 cocatalyst, aluminum alkyl 294 cohesive energy density (CED) 177 colloidal particles 272 – PS 277 comb polymer brushes 273 commercially important polyamides 29–34 commodity products 53 composites – ICP–CNT 307–309 – manufacture 133–136 – microstructure 225

– polymer/cellulose ﬁber 99–100 – polymer/CNT 100 – polymer/inorganic compound 96–99 – polypropylene 2 – polyurethane 6 – see also nanocomposites conducting polymers 303 conductive nanoﬁllers 185 conductivity – AC 323 – DC 324 – electrical 185 – thermal 187, 189 cone calorimetry 360–361 conjugated polymer charge transport 324 controlled polymerization techniques 350–352 coordination catalysts 59 coordination of amines 59 copolymerization 62 – random 96 copolymers – block 23, 270–271 – butadiene-isoprene 291 corona treatment 229 correlation function 197 coupling reaction 342 covalently functionalized laponite clay 275 cracks 201 creep 203–204 crosslink points 202 crosslinked polyester nanocomposites 248 crosslinking mechanism 136 crosslinking monomers 246 cryochemical solid-state synthesis 97 crystal vibrational energy 188 crystalline lamellae 198 crystallization – nanocomposites 109–112 – rate 381 curing – conditions optimization 222–224 – CW–PFA composites 137–138 – ﬁrst stage 156 – MMT–PFA composites 139–140 – UP/MMT nanocomposites 246, 253–258 curved particles, MMT 57 CVD (chemical vapor deposition) 306–307 CW (cellulose whisker) nanocomposites 128–132 – reactive molding 136–138, 140–149 cyclic forces 201–202 cyclic oligomers 18 cyclohexane 290–293

389

390

Index

d 2D/3D patterning 94 3D percolating system 323 damping factor 213 DC conductivity 324 decomposition – side chains 316 – thermal 307 defects 201 degradation 114–115 – nonoxidative 148–149, 160–161 – oxidative 162–163 degree of phase separation (DPS) 183 density – CNTs 303 – grafting 369 – MMT 231 – modiﬁed vermiculite 232 depolymerization 357 depth proﬁles, nanoscratch 206, 216 derivative mass 113–115 derivative weight 228 desulfonation 146 DGEBA (diglycidyl ether of bisphenol A) 15–17 diamine/diacid-based nanocomposites 41–48 dicarboxylic acids, unsaturated 245 dichloromethane vapor 370 1,3-dienes 287 differential scanning calorimetric (DSC) analysis 314, 358–359 diffusion barriers 217 diglycidyl ether 259 – bisphenol A (DGEBA) 15–17 dihydrogenated tallow 191 diisocyanate 134 1,6-diisocyanatohexane 381 γ-diketone 160 dimethyl dihydrogenated tallow quaternary ammonium (DMDTA) 257–258 dimethylformamide (DMF) 233 N,N-dimetylformamide (DMF) 177 dioctadecyldimethylammonium (2C18) 226 direct SSP 32, 44–46 dispersions – ﬁller phase 123–167 – MMT in styrene 352 – P3HT/MWCNT nanocomposites 311 – stable 311 “dispersions mixing” method 131 2,5-disubstituted furan rings 140 dodecylamine 78

double-layer hydroxide-based nanocomposites 10 double-wall nanotubes (DWNTs) 305 DPS (degree of phase separation) 183 driving force concept 107 drug delivery systems 381 drying, freeze 137 DWNTs (double-wall nanotubes) 305 dynamic mechanical analysis (DMA) 181 dynamic mechanical thermal analysis (DMTA) 247, 354–355

e EDA, see ethylene amine edge-surface hydroxyls modiﬁcation 294 EG, see expandable graphite elastic modulus 131 elastic solids 312 elastomers 170 electrical conductivity 185 emulsion-BR 289 emulsion polymerization 348–349 energy – activation 115–116 – crystal vibrational 188 – density 177 – surface 226 epoxide-anhydride reaction 262 epoxide resins 221 epoxy/clay nanocomposites 268 epoxy compounds, polyaddition 245 epoxy nanocomposites 221–244 – barrier properties 238–240 – in-situ synthesis 221–244 – morphology 231–238 – oxygen permeation 7 epoxy prepolymers 16 esteriﬁcation 142, 272 ethers, diglycidyl 15–17, 259 ethylene amine 174 ethylene uptake 72 excess surface modiﬁcation molecules 240–243 exchange capacity, cation 58, 81 exchangeable cations 55 exfoliated graphite oxide nanoplatelets 212 exfoliated morphology 4 exfoliated nanocomposites 332 – partially 42 exfoliated platelets 154 exfoliated sheets 248 – graphene 217

Index exfoliation 67–69, 82, 133 – MMT 156 – nanoclays 189–194 – tactoids 153 expandable graphite 209–211 expanded interlayer spacing 200

f FA, see furfuryl alcohol – resiniﬁcation 142–160 fabrication, see synthesis ﬁber, cellulose 99–100 ﬁber–matrix bonding 131 ﬁbrils 129 ﬁller – aligned 8 – epoxy nanocomposites 224–229 – phase dispersion 123–167 – platelets 8–9 – powders 233 – UP/MMT nanocomposites 259–260 ﬁller volume fraction 3, 240 ﬁller weight fraction 231 ﬁlms, thin 175 ﬁre performance 359–361 ﬁre retardants (FR) 258 ﬁrst stage curing 156 ﬂame retardancy 207–208 ﬂammability, UP/MMT nanocomposites 258–259 ﬂexural properties 119, 354 ﬂow birefringence 141 forces – cyclic 201–202 – driving force concept 107 – van der Waals/ionic 189, 208 formaldehyde 136 Fourier transform infrared spectra 321–322 free radical polymerization 275, 347–350 – styrene 20 free radicals 257 freeze drying 137 FTIR spectra, see Fourier transform infrared spectra functionality – “built-in” 124 – hydrophilic 174 functionalization – CNTs 308 – laponite clay 275 – organoclays 380 furan rings 140, 159 furfuryl alcohol 135–136 – resiniﬁcation 142–160

g gas-phase-assisted surface polymerization (GASP) 89–104 glass transition 196 grafting density 369 graphene – exfoliated sheets 217 – functionalized 208–217 – hexagonal sheet 306 graphite – exfoliated oxide nanoplatelets 212 – expandable 209–211 – graphite-like materials 320 groups – ammonium 344 – hydroxyl 54, 77 – methylene 45 – phenyl 350 – sulfonic acid 142, 144 – urea 169 growth – particle 68 – polymer 71 – spherulite 110

h β-H transfer reactions 298 halogen containing unsaturated polyester (HUP) 260–263 “head-to-tail” reactions 135 heat capacity 196 heat distortion temperature (HDT) 118 heat release 360 2nd heating thermogram 111 heptane 80 Herschel–Bulkley equation 173 heterogeneous catalysts 297 hexadecyl trimethyl ammonium chloride 207 hexagonal sheet 306 hexamethylenediamine 31 3-hexylthiophene 303–330 hierarchical supramolecular structure 129 high impact polystyrene 348 homopolymerization 135 hopping, variable-range 325 HPA, see hydroxypropylacrylate HT, see 3-hexylthiophene hybrid PCNs 286 hybrid PET/MMTs 110 hydrazine 216 hydrogen bonding, intermolecular 130 hydrogen bonding index 214 hydrogenated tallow 73

391

392

Index hydrolytic ring cleavage 158 hydrolytic ring opening 146 hydrophilicity 174, 226, 332 hydrophobicity 100, 284, 332 hydroxides – hydroxide-based nanocomposites 10 – LDHs 34, 46, 377–379 hydroxyfunctional MTHEA 256 hydroxyl groups 54, 381 – catalyst supports 77 hydroxyl-substituted alkylammonium cations 370 hydroxylated surface 130, 138 hydroxyls, edge-surface 294 hydroxypropylacrylate 250–251

i ICPs , see inherent conducting polymers imidazolium 358 immobilization 124 impact properties 354 in-situ anionic polymerization 289–293 in-situ intercalated nanocomposites 36–48 in-situ intercalative polymerization 133, 157 in-situ polymerization 27–51, 283–301 – aliphatic polyester–clay nanocomposites 367–386 – ICP–CNT composites 309 – methods 346 – oleﬁns 59 – PEA/clay nanocomposites 382–383 – P3HT/MWCNT nanocomposites 303–330 – polyoleﬁn–clay nanocomposites 53–88 – polyurethane nanocomposites 169–220 – reactive 123–167 – stereospeciﬁc 293–298 in-situ synthesis – alkylaluminoxanes 76 – epoxy nanocomposites 221–244 – polymer nanocomposites 1–25 – strategy 230 indentation load–displacement curves 205 inherent conducting polymers 303–304 – ICP–CNT composites 307–309 initiators 246, 345–346 inorganic aluminosilicates 222 inorganic PNs 283 inorganic volume fraction 6–7 insulating polymers 326 π–π interaction 304 interaggregate space 56 intercalation – clays 107 – MMT 156

– morphology 4 – nanoclays 189–194 – nanocomposites 332 – organoclays 206 – PA nanocomposites 36–48 – polymer 11–12 – polyols 153 – tactoids 235 intercalative polymerization 95, 133, 157 interdomain repeat distance 198–199 interfacial interaction 236 interfacial polycondensation 46–48 intergallery catalysis 224 intergallery expansion 347 interlayer cations 59 interlayer distance 14, 39 interlayer region 285 interlayer spacing – expanded 200 – PUCLNs 197 intermolecular hydrogen bonding 130 interphase 127 interrupted metallic model 326 intrinsic viscosity 108, 112 ion exchange 224–229 – organic cations 241 – reactions 53, 343 ionic bond 59 ionic forces 189 ionization laser 97 isocyanate 211 isophthalic UP resins 253 isoprene 287 – BIR 291 isotactic 1,2-polybutadiene 288

k kaolinite 105 karstecit catalysts 278 kinetic parameters 114 kinetic rate of polymerization 379

l “labyrinth” morphology 161 lactams 30 – lactam-based nanocomposites 36–41 lactide rings 379 lactone byproducts 136 lamellae 285 – crystalline/amorphous 198 laminates, packaging 222 laponite clay 275 laser vaporization/ionization 97 laurate units 379 lauryl methacrylate polymerization 21

Index layered aluminosilicates 1 layered clays 107 layered double hydroxides (LDHs) 34, 46, 377–379 – adipate-modiﬁed 47 layered nanoparticles 285 layered silicates 18, 126 – nanocomposites 132–133 layered silicic acids 285 layers, stacked 56 limiting oxygen index 262 linear polyamides 28 liquid-phase polymerization 90 liquidity 174 lithium-BR 289 LOI, see limiting oxygen index loss modulus 315 low-temperature processes 31–34

m MAO, see methylaluminoxane mass, derivative 113–115 masterbatches 374 mechanical properties – nanocomposites 117–118 – P3HT/MWCNT nanocomposites 312–316 – PS/MMT nanocomposites 353 – PUCLNs 200–204 – UP/MMT nanocomposites 246–253 melamine polyphosphate 208 melt blending 267, 309 – graphite/TPU composites 217 melt-extruded nanocomposites 321 melt viscosity 261 mesomerically stabilized radicals 257 metal chalcogenides 285 metal oxide-reinforced nanocomposites 34 metallocenes 60, 62–64 – activation 75 methyl methacrylate 92, 96, 269, 272, 277–278 methyl-tallow-bis(2-hydroxyethyl)-quaternary ammonium 247–250, 260–263 – hydroxyfunctional 256 methylaluminoxane (MAO) 61–62 – catalyst supports 63, 72–73 – polybutadiene clay nanocomposites 294–295 methylene groups 45 micas 79 microcomposite 332 microcrystalline cellulose 143–145 microﬁbrils 129 microphase separation 182, 197

microscopy, optical 176 microstructure – composites 225 – polymers 288 misalignment of ﬁller platelets 8–9 miscibility 372 mixed-resin systems 259–260 mixing – dispersion/suspension 131 – temperature 193 MLM, see multilayer model MMA, see methyl methacrylate modiﬁcation – MMT 12, 332–346 – nonreactive 333–343 – polymeric initiator-based 345–346 – reactive 343–345 – vermiculites 232 modiﬁed MAO (MMAO) 73–74 molding 133 molecular dynamics simulations 187 monofunctional silanes 275, 342 monomers – adsorbed 91 – conversion 297 – crosslinking 246 – intercalation 368 – polarity 250 montmorillonite (MMT) 1–2 – basal spacing 37 – commercially procured 242–243 – curved particles 57 – density 231 – dispersion in styrene 352 – exfoliated tactoids 155 – GASP 90 – intercalation 156 – lab treated 242–243 – modiﬁcation 5, 12, 247, 332–346 – morphology 150 – octadecylamine-treated 254 – organomodiﬁed 381 – PET/clay nanocomposites 105–122 – PET/MMT hybrids 110 – PS/MMT nanocomposites 331–365 – reactive molding 138–140, 149–164 – smectites 55–56 – structure 132 – unmodiﬁed 225 – UP/MMT nanocomposites 245–266 morphological hierarchy 56, 249 morphology 4 – CW nanocomposites 141–142 – epoxy nanocomposites 231–238 – “labyrinth” 161

393

394

Index – MMT clay 150 – PCNs 331–332 – PET/clay nanocomposites 108–109 – P3HT/MWCNT nanocomposites 311 – PUCLNs 196–200 – PUCNs 176–183 – PUFGNs 210–214 – UP/MMT nanocomposites 246–253 MPP, see melamine polyphosphate MTHEA, see methyl-tallow-bis(2hydroxyethyl)-quaternary ammonium – hydroxyfunctional 256 multifunctional nanocomposites 169 multilayer model 67–68 multiwall carbon nanotubes (MWCNTs) 40, 48, 170–173 – P3HT/MWCNT nanocomposites 303–330

n Na, see sodium nanoclays 126 – exfoliation/intercalation 189–194 – nanoclay-reinforced nanocomposites 34 nanocomposites – aluminosilicate 369 – barrier properties 118–121 – bio-based 123–167, 367–368 – cellulose whisker 128–132 – crosslinked polyester 248 – crystallization 109–112 – double-layer hydroxide-based 10 – epoxy 7, 221–244 – epoxy/clay 268 – exfoliated 332 – GASP 95–100 – intercalated 36–48, 332 – layered silicate 132–133 – mechanical properties 117–118 – melt-extruded 321 – morphology 108–109 – multifunctional 169 – partially exfoliated 42 – PCL-based 368–375 – PET/clay 105–122 – P3HT/MWCNT 303–330 – PLA-based 375–381 – polyamide 34–48 – polybutadiene/clay 283–301 – polyethylene 20 – polymer 1–25 – polymer/clay 267–281 – polymer matrix 125–133 – polymer–silicate 267

– – – – – – – – –

polyoleﬁn–clay 53–88 polypropylene 2 polystyrene/MMT 331–365 polyurethane 169–220 PPDO-based 381–382 preparation 89–91, 106–108 structure 286 ternary 373 thermal properties 112–117, 148–149, 161–164 – UP/MMT 245–266 – see also composites nanoﬁbers, cellulose 99–100 nanoﬁllers 185 nanoindentation tests 253 nanoparticles – agglomeration tendency 127 – layered 285 – silica 78 nanoplatelets 212 nanoreactors 296 nanoscratch depth proﬁles 206, 216 nanotubes – CNTs, see CNTs – double-wall 305 – MWCNTs, see multiwall carbon nanotubes Newtonian ﬂuids 312 nickel catalysts 77 nitroxide-mediated polymerization (NMP) 351–352 nitroxyl-based organic cation modiﬁcation 22 nonisothermal melting 179 nonoxidative degradation 148–149, 160–161 nonreactive cationic surfactants 334–336 nonreactive modiﬁcations 333–343 nucleating ability 179

o octadecylamine-treated MMT 254 octahedral coordination 54 OFW, see Ozawa–Flynn–Wall oleﬁns 59 oligomeric silsesquioxanes 343 oligomers, cyclic 18 one-dimensional correlation function 197 one-dimensional nanomaterials 170 optical absorption 318 optical microscopy (OM) 176 optical properties 318–322 ordered (in)organic PNs 283 organic cations 22, 241 organic–inorganic interface 250

Index organic surface modiﬁcation 64–66 organo-modiﬁer 106 organoclays 43–45 – intercalated 206 – PET/MMT hybrids 113 – PUCLN fabrication 190 – quaternary ammonium cations 65 – TFC 380 – UP nanocomposites 251–252 organomodiﬁed MMTs 381 oven-curing 137 oxidative degradation 162–163 oxidized graphene 209 oxygen ﬂux data 356 oxygen permeation – epoxy nanocomposites 7, 238–239, 241–242 – relative 3, 15 oxygen transmission rate (OTR) 120 Ozawa–Flynn–Wall (OFW) method 114, 116

p π–π interaction 304 PA 6, see poly(caproamide) PA 6.6 (poly(hexamethylene adipamide)) 30–31 packaging laminates 222 partially exfoliated nanocomposites 42 particle break-up 67–69, 82 particle expansion 71 particle growth 68 particles, colloidal 272, 277 path, “tortuous” 119 patterning, 2D/3D 94 PBA, see polybutylene adipate PCL, see poly(ε-caprolactone) PCNs, see polymer/clay nanocomposites PEA, see poly(ester amide)/clay nanocomposites PEG, see poly(ethylene glycol) percolating system 323 percolation threshold 325 permeability 120 peroxide initiator 246 PET, see polyethylene terephthalate PFA, see polyfurfuryl alcohol phase separation 371 – degree of 183 PHEMA backbone 273 phenyl group 350 Phillips catalysts 60 phonons 188 phosphorus surfactants 360

photo-induced controlled polymerization 93–94 photoinitiators 93, 256 photoluminescence spectra 318–320 photooxidation 257 P3HT (poly(3-hexylthiophene))/MWCNT nanocomposites 303–330 physical properties – PUCLNs 200–204 – PUCNs 183–188 – PUFGNs 214–217 physically controlled polymerization 92–93 platelets – exfoliated 154, 212 – ﬁller 8–9 PMMA, see polymethylmethacrylate polar polymers 284 polarity 237 – monomers 250 poly(ε-caprolactone) 268, 368–375 poly(p-dioxanone) 367, 381–382 poly(3-hexylthiophene) (P3HT)/MWCNT nanocomposites 303–330 polyacrylate chains 272 polyaddition 245 poly(alkylthiophene)s 304 polyamide ﬁlms 230 polyamide nanocomposites 34–48 polyamide salts 33 polyamides 29–34 polybutadiene 283–301 1,2-polybutadiene (syndio- and isotactic) 288 polybutylene adipate 181 poly(caproamide) 29–30 polycondensation – glycols 245 – interfacial 46–48 – PA 6.6 formation 31 – sodium chloroacetylaminohexanoate 382 polydispersity index 377 poly(ester amide) (PEA)/clay nanocomposites 382–383 polyesters – aliphatic 367–386 – crosslinked nanocomposites 248 – unsaturated 245–266 polyether polyol 193–194 poly(ethylene glycol) 377 polyethylene nanocomposites 20 polyethylene particles 69 polyethylene terephthalate 17–18 – PET/clay nanocomposites 105–122 – PET/MMT hybrids 110

395

396

Index polyfurfuryl alcohol 125 polyhedral oligomeric silsesquioxanes 343 poly(hexamethylene adipamide) 30–31 3,4-polyisoprene 288 polylactide 268, 367, 375–381 – PLA/vermiculite nanocomposites 377 polymer brushes 273 polymer/cellulose ﬁber nanocomposites 99–100 polymer/clay nanocomposites 283 – GASP 95–96 – hybrid 286 – in-situ ATRP 267–281 – morphology 331–332 polymer/CNT nanocomposites 100 polymer growth 71 polymer/inorganic compound nanocomposites 96–99 polymer intercalation 11–12 polymer-layered silicate 269, 346 polymer matrix nanocomposites (PNCs) 125–133 polymer nanocomposites (PNs) 283 – in-situ synthesis 1–25 – preparation 287 – structure 286 polymer–silicate nanocomposites 267 polymeric initiator-based modiﬁcations 345–346 polymeric initiator-based surfactants 340–341 polymerization 267–281 – additives 71–72 – anhydrous melt technique 43 – anionic 289–293 – ATRP 23, 99, 351 – bulk 347–348 – conditions 82–83 – controlled techniques 351 – coordination catalysts 60–62 – direct SSP 32, 44–46 – emulsion 348–349 – free radical 20, 275, 347–350 – GASP 89–104 – heterogeneous catalysts 297 – in-situ, see in-situ polymerization – intercalative 95, 133, 157 – kinetic rate 379 – lauryl methacrylate 21 – liquid phase 90 – low-temperature processes 31–34 – methods 287–289, 346 – NMP 351–352 – photo-induced controlled 93–94

– physically controlled 92–93 – ring-opening 18, 30, 40 – solution 349–350 – solution-melt technique 41–43 – SSP 31–32 – stereospeciﬁc 293–298 – surface-initiated 350 – temperature 383 – tunnel 297–298 – two-step 380 polymers – biodegradable 367–368 – conducting 303 – ICPs 303–304 – insulating 326 – microstructure 288 – molecular structure 83–84 – polar 284 – semicrystalline 109, 383 – sustainability 367 – water-soluble 311 polymethylmethacrylate 95, 100 polyol–nanoclay mixtures 194–196 polyoleﬁnes 130 – polyoleﬁn–clay nanocomposites 53–88 polyols – intercalating 153 – trifunctional 195 polypropylene 2, 230 polystyrene (PS) – HIPS 348 – PS/MMT nanocomposites 331–365 – PSBA 269 polyurethane (PU) – acrylate terminated 259 – composites 6 – nanocomposites 169–220 – water-based 174 poly(vinyl acetate) 255 pore space 56 POSS, see polyhedral oligomeric silsesquioxanes powders – ﬁller 233 – morphology 82 – XRD 108–109 PPDO, see poly(p-dioxanone) “pre-exfoliated” masterbatches 374 precursor, PNCs 135–136 preparation – nanocomposites 89–91 – PET/clay nanocomposites 106–108 – polymer nanocomposites 287 prepolymerization 33

Index prepolymers – epoxy 16 – synthesis 260–263 – UP 246–260 protective coatings 204 protonated monomers 40 PSBA (poly(styrene-block-butyl acrylate)) 269 pseudo-bilayers 371 PU/clay nanocomposites (PUCLNs) 188–208 PU/CNT nanocomposites (PUCNs) 170–188 PU/functionalized graphene nanocomposites (PUFGNs) 208–217 PU (polyurethane) – acrylate terminated 259 – composites 6 – nanocomposites 169–220 – water-based 174 pyridine 278 pyridium 358

q quaternary ammonium cations 65, 73 quinolium 358

r radicals – free, see free radicals – mesomerically stabilized 257 RAFT, see reverse addition-fragmentation transfer Raman scattering 320–321 Raman shift 177–178 random copolymerization 96 reaction injection molding 133 reactions – anhydrous prepolymerization 33 – ATRP 23, 99, 267–281, 351 – carboxylation 171 – clay 58–59 – copolymerization 62, 96 – coupling 342 – desulfonation 146 – epoxide-anhydride 262 – esteriﬁcation 142, 272 – β-H transfer 298 – “head-to-tail” 135 – homopolymerization 135 – hydrolytic ring cleavage 158 – ion-exchange 53 – ion exchange 343 – polyaddition 245

– polycondensation, see polycondensation – polymerization, see polymerization – silylation 274 reactive cationic surfactants 337–339 reactive modiﬁcations 343–345 reactive molding 133–136 – CW nanocomposites 136–138, 140–149 – MMT nanocomposites 138–140, 149–164 recrystallization transition 213 reducing agents 216 reinforced reaction injection molding 134 relative oxygen permeation 3, 15 relative tensile modulus 3 relaxation, stress 203 repeat distance 198–199 resiniﬁcation 137, 139 – FA 142–148, 150–160 resins – clay modiﬁed HUP 261–262 – epoxide 221 – isophthalic UP 253 – UP 246–260 resistance – scratch 204–207 – thermal 148–149, 161–164 reverse addition-fragmentation transfer 351 rheological behavior 194–196 – PS/MMT nanocomposites 355–356 – UP/MMT nanocomposites 253–258 RIM, see reaction injection molding ring cleavage, hydrolytic 158 ring opening, hydrolytic 146 ring-opening polymerization (ROP) 18, 268 – caprolactam 40 – cationic 274 – lactams 30 rings – furan 140, 159 – lactide 379 – stretching 159 – thiophene 316 rotational mobility 359 RRIM, see reinforced reaction injection molding rubber – BIR 291 – BR 288–289 – styrene butadiene 288–290

s salts 33 saponite 81 SBR, see styrene butadiene rubber β-scission 357

397

398

Index scratch resistance 204–207 second heating thermogram 111 semiaromatic polyamides 28 semicrystalline polymers 109, 383 semicrystalline PUCNs 179 shaped products 35 shear rate 261 shear thinning parameter 173 sheets – exfoliated 248 – graphene 217, 306 – hexagonal 306 – tetra-/octahedral 54 shelf stability 257 side chain decomposition 316 silane-modiﬁed MMT 247 silanes 275 silanol groups 381 silica layer 132 silica nanoparticles 78 silicates – layered 18, 126 – nanocomposites 132–133 – polymer intercalation 11 – polymer-layered 346 – spacing 368 silicic acids 285 silsesquioxanes 343 silylation 274 simulations, molecular dynamics 187 smectites 54–56, 285 sodium chloroacetylaminohexanoate 382 solid-state polymerization (SSP) 31–32 – direct 44–46 solid-state synthesis 97 solid-substrate surfaces 91–92 solution blending 267, 308–309 solution-melt polymerization technique 41–43 solution-mixed graphite/TPU composites 217 solution polymerization 349–350 Soxhlet extraction 172 speciﬁc surface area 127 spherulite growth 110 SRIM, see structural reaction injection molding stability – dispersions 311 – shelf 257 – thermal 42, 207–208, 316–317 stacked layers 56 stearyl ammonium 11 stereoselectivity, catalysts 84 stereospeciﬁc polymerization 293–298

stereospeciﬁc PS 348 storage modulus 213, 315, 355 stress relaxation 203 stress–strain curves 118 – PU 184, 202 stretching – (anti)symmetric 322 – ring 159 structural reaction injection molding 134 styrene 23 – free radical polymerization 20 – GASP 92 – hydrophobic 332 – MMT dispersion 352 – PSBA 269 – UP–styrene-poly(vinyl acetate) system 255 styrene butadiene rubber 288–290 sulfonic acid groups 142, 144 supercooling 112 supercritical CO2 374 supports, catalysts 63–64, 72–74, 77, 81, 294 supramolecular structure 129 surface-initiated polymerization 350 surface modiﬁcation – clays 64–66, 80–81 – epoxy nanocomposites 224–229 – excess molecules 240–243 – MMTs 5 – vermiculites 5 surfaces – acidity 79 – “built-in” functionalities 124 – edge-surface hydroxyls modiﬁcation 294 – energy 226 – GASP 89–104 – hydroxylated 130, 138 – solid-substrate 91–92 – wrinkled 214 surfactants – ammonium 290 – design 344 – nonreactive cationic 334–336 – phosphorus 360 – polymeric initiator-based 340–341 – reactive cationic 337–339 – tallow 45 – zwitterionic 333 surgical sutures 381 suspensions – bentonite 256 – mixing 131 swellability 79–80 swelling 14, 39

Index symmetric stretching 322 syndiotactic 1,2-polybutadiene 288 syndiotactic PS 348 synergistic enhancement 2 synthesis – cryochemical solid-state 97 – in situ 1–25, 221–244 – methods 9–12 – prepolymers 260–263 – PUCLNs 189–196 – PUCNs 170–188 – PUFGNs 209–210 – strategy 230 – UP/MMT nanocomposites 246–253 synthetic clays 39, 285

t tacticity 84 tactoids 150–153 – exfoliation 153 – intercalated 235 – tunnel polymerization 298 tallow – dihydrogenated 191 – DMTDTA 257–258 – hydrogenated 73 – MTHEA 247–250 – surfactant 45 tapping mode 200 temperature – heat distortion 118 – mixing 193 – polymerization 383 tensile properties 117 – PS/MMT nanocomposites 353–354 – PU 184 – PUCLNs 203 – relative tensile modulus 3 – test results 354 ternary nanocomposites 373 tetrahedral sheet 54 tetrahydrofuran (THF) 270–271, 278 N,N,N′,N′-tetramethylethanediamine 291 TFC, see twice functionalized organoclay thermal analysis 247, 354–355 thermal conductivity 187, 189 thermal decomposition 307 thermal gravimetric analysis (TGA) 356–358 thermal properties 112–117 – PS/MMT nanocomposites 356 thermal resistance – CW–FA nanocomposites 148–149 – MMT–FA nanocomposites 161–164

thermal stability 207–208 – P3HT/MWCNT nanocomposites 316–317 thermal stabilizer 42 thermal treatment 66 thermogram, 2nd heating 111 thermosetting 222 thin ﬁlms 175 thin layer coating 91–92 thiophene 308–309 – rings 316 three-dimensional percolating system 323 TMEDA, see N,N,N′,N′-tetramethylethanediamine toluene 80 toluene–OC suspension 292 tortuosity 120 “tortuous path” 119 transcrystallinity 131 transfer reactions, β-H 298 translational mobility 359 transmission rates 207 transportability 322–326 triethylaluminum (AlEt3) 376 trifunctional alkoxy silanes 275 trifunctional polyol 195 trifunctional silane 342 tunnel polymerization 297–298 twice functionalized organoclay 380 two-phase system 198 two-step polymerization 380

u ultralightweight CNT/PU foam 186 ultrasonic treatment 137, 210 unintercalated morphology 4 unmodiﬁed montmorillonite 225 unsaturated dicarboxylic acids 245 unsaturated polyester (UP) – halogen containing 260–263 – prepolymers 246–260 – resins 246–260 – UP–styrene-poly(vinyl acetate) system 255 unsaturated polyester (UP)/MMT nanocomposites 245–266 – organoclays 251–252 urea group 169

v van der Waals forces 189, 208 vapor transmission 207, 238–239, 241 vaporization laser 97 variable-range hopping 325

399

400

Index vermiculites 1 – epoxy nanocomposites 226, 229 – modiﬁed 232 – PLA/vermiculite nanocomposites 377 – surface-modiﬁed 5 (ar-vinyl-benzyl)trimethyl ammonium (VBTA) 296 vinyl monomers 93, 98 viscoelastic properties 312 viscosity – epoxide resins 221 – intrinsic 108, 112 – melt 261 – PUCNs 173 – trifunctional polyol 195 vitreous state 138 volume fraction – ﬁller 3 – inorganic 6–7

w wall–wall interaction 317 washings cycle 227–228 water, zeolitic 65

water-based PU (WPU) 174–175 water-soluble polymers 311 water vapor transmission 207, 238–239, 241 WAXD proﬁles 83 weight, derivative 228 whisker, cellulose, see CW nanocomposites wrinkled surfaces 214

x X-ray diffraction (XRD) – epoxy nanocomposites 233–235 – powder 108–109 xylene–OC suspension 292

y Young’s modulus 184, 215, 353

z zeolitic water 65 Ziegler–Natta catalysts 60, 287 zirconocenes 62, 82 zwitterionic ALA 38–39 zwitterionic surfactants 333