Polyolefin Blends

POLYOLEFIN BLENDS POLYOLEFIN BLENDS Edited by Domasius Nwabunma 3M Company Thein Kyu University of Akron WILEY-INTE...

Author: Nwabunma D. | Kyu Th.

141 downloads 1963 Views 21MB 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

POLYOLEFIN BLENDS

POLYOLEFIN BLENDS Edited by Domasius Nwabunma 3M Company

Thein Kyu University of Akron

WILEY-INTERSCIENCE A JOHN WILEY & SONS, INC., PUBLICATION

Cover credit: (Third image on top right side) Reprinted from European Polymer Journal, vol. 40, Smit, G. Radonjic and D. Hlavata. Phase morphology of iPP/aPS/SEP blends, page 1439, 2004. With permission from Elsevier. Copyright ß 2008 by John Wiley & Sons, Inc. All rights reserved Published by John Wiley & Sons, Inc., Hoboken, New Jersey Published simultaneously in Canada No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 750-4470, or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at http://www.wiley.com/go/permission. Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762-2974, outside the United States at (317) 572-3993 or fax (317) 572-4002. Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com. Wiley Bicentennial Logo: Richard J. Pacifico Library of Congress Cataloging-in-Publication Data: Nwabunma, Domasius. Polyolefin blends / edited by Domasius Nwabunma, Thein Kyu. p. cm. Includes index. ISBN 978-0-471-79058-7 (cloth) 1. Polyolefins. I. Kyu, Thein, 1948- II. Title. TP1180.P67N928 2007 668.40 234–dc22 2007021318 Printed in the United States of America 10 9 8 7 6 5 4 3 2 1

Contents

Preface

xv

Contributors

Part I

Introduction

1. Overview of Polyolefin Blends 1.1 Introduction 1.2 Olefinic Monomers 1.3 Polyolefin Homopolymers, Copolymers, and Terpolymers 1.4 Polyolefin Blends 1.5 Trends in Polyolefin Blends Nomenclature References

2. Miscibility and Characteristics of Polyolefin Blends 2.1 Introduction 2.1.1 Polyolefins 2.1.2 Blends 2.2 Polymer Blend Miscibility 2.3 Interfaces in Liquid and Polymer Mixtures 2.4 Polyolefin–Polyolefin Blends 2.4.1 Blends between Polyethylenes 2.4.2 Blends between Isotactic Polypropylene and Ethylene Propylene Copolymers 2.4.3 Blends between iPP and High Comonomer Concentration Polyethylene Copolymers 2.4.4 Blends between iPP and PB1 2.5 Binary Immiscible Blends 2.5.1 Polyolefin–Polystyrene Blends 2.5.2 Polyolefin–Polyamide Blends 2.6 Ternary Blends of Polyolefins with Other Polymers and Compatibilizing Agents 2.6.1 Surfactants and Compatibilizing Agents 2.6.2 Polyolefin–Polystyrene Blends with Compatibilizing Agents

xvii

1 3 3 4 5 7 13 16 18 27 27 27 29 30 33 36 36 38 39 40 42 43 43 44 44 45

v

vi

Contents 2.6.3 Polyolefin–Polyamide Blends with Compatibilizing Agents 2.7 Conclusions Nomenclature References

Part II

Polyolefin/Polyolefin Blends

3. Miscibility, Morphology, and Properties of Polyethylene Blends 3.1 Introduction 3.2 Structure and Properties of Polyethylenes 3.3 Applications of Polyethylene Blends 3.4 Molar Mass and Branching Distributions 3.5 Crystallization, Melting, and Branching of Polyethylenes 3.6 Miscibility and Crystallization 3.7 Theoretical Prediction of Miscibility 3.8 Rheology of Melted Polyethylene Blends 3.9 Mechanical Properties of Polyethylene Blends 3.10 Additives 3.11 Conclusions Nomenclature References

4. Miscibility and Crystallization Behavior in Binary Polyethylene Blends 4.1 4.2

Introduction Miscibility 4.2.1 Linear and Short Branched Polyethylene Blends 4.2.2 Blends of Linear and Long Branched Polyethylenes 4.2.3 Blends of Short and Long Branched Polyethylenes 4.3 Crystallization Behaviors 4.3.1 Blends of Linear and Short Branched Polyethylenes 4.3.2 Blends of Linear and Long Branched Polyethylenes 4.3.3 Blends of Short and Long Branched Polyethylenes 4.4 Conclusions Nomenclature References

5. Microscopically Viewed Structural Characteristics of Polyethylene Blends Between Deuterated and Hydrogenated Species: Cocrystallization and Phase Separation 5.1 5.2 5.3

Introduction Cocrystallization and Phase Separation of PE Blends Aggregation Structure of Chains in Lamella

48 50 50 51

57 59 59 64 65 66 68 72 74 76 77 80 80 81 82

84 84 86 86 88 89 90 90 92 93 93 94 94

97 97 98 101

Contents 5.4 Crystallization Behavior of D/H Blend Samples 5.4.1 Crystallization in the Cooling Process from the Melt 5.4.2 Isothermal Crystallization Process 5.4.3 Blending Effect on Crystallization Rate 5.5 Mixing Behavior of D and H Components 5.6 Conclusions Acknowledgments Nomenclature References

6. Thermal and Structural Characterization of Binary and Ternary Blends Based on Isotactic Polypropylene, Isotactic Poly (1-Butene) and Hydrogenated Oligo (Cyclopentadiene) 6.1 Introduction 6.2 Binary Blends 6.2.1 Blend Preparation 6.2.2 Glass Transition Temperature 6.2.3 Morphology and Spherulite Growth Rate 6.2.4 Isothermal Bulk Crystallization Kinetics 6.2.5 Temperature Dependence of the Spherulite Growth Rate and the Overall Kinetic Rate Constant 6.2.6 Melting Behavior 6.2.7 Polymorphism and Phase Transformation of Poly (1-Butene)/ Hydrogenated Oligo (Cyclopentadiene) 6.2.8 Supermolecular Structure of Isotactic Polypropylene/ Hydrogenated Oligo (Cyclopentadiene) Blends 6.3 Ternary Blends 6.3.1 Blends Preparation 6.3.2 Morphology and Spherulite Growth Rate 6.3.3 Glass Transition Temperature 6.3.4 Nonisothermal Crystallization and Melting Behavior 6.3.5 Isothermal Bulk Crystallization Kinetics of Isotactic Polypropylene Component 6.3.6 Melting Behavior of the Isotactic Polypropylene Component 6.3.7 Supermolecular Structure 6.4 Conclusions Nomenclature References

7. Morphological Phase Diagrams of Blends of Polypropylene Isomers with Poly(Ethylene–Octene) Copolymer 7.1 Introduction 7.2 Blends of sPP/POE 7.2.1 Thermal Characterization and Morphological Phase Diagrams: Undulated Lamella, Sheaf, and Spherulite 7.2.2 Growth of Single Crystals: Length, Width, and Periodicity

vii 105 106 108 113 114 117 118 118 119

121 121 123 123 123 124 125 126 131 133 136 141 141 141 143 143 145 146 147 153 154 155

157 157 159 161 165

viii

Contents 7.2.3 Phase Field Modeling for a Single-Component System: Sectorization and Ripple Formation in sPP 7.3 Blends of iPP/POE 7.3.1 Morphology Development in Relation to Phase Diagrams 7.3.2 Sectorization in iPP/POE Blends 7.3.3 Crystal Growth Dynamics in Binary Blends of iPP and aPP 7.4 Blends of ePP/POE 7.4.1 Characterization of Neat Elastomeric Polypropylene 7.4.2 Melting Transitions and Morphology Phase Diagrams of ePP/POE Blends 7.5 Conclusions Nomenclature References

8. Structure, Morphology, and Mechanical Properties of Polyolefin-Based Elastomers 8.1 Introduction 8.2 Thermoplastic Polyolefin Elastomers 8.2.1 Reactor Blends of PP, PE, and EPR: Impact Copolymer PP 8.2.2 Postreactor Blends of PP–EPR and ICP–EPR 8.3 Thermoplastic Vulcanized Elastomers 8.3.1 Dynamic Vulcanization and Morphology 8.3.2 Origin of Rubber Elasticity 8.3.3 Several Factors that Influence Mechanical Properties 8.4 Polyolefin Copolymers, Blends, and Composites 8.4.1 Polyolefin Copolymers 8.4.2 Blends and Composites 8.5 Conclusion Acknowledgment Nomenclature References

9. Morphology and Mechanical Properties in Blends of Polypropylene and Polyolefin-Based Copolymers 9.1 Introduction 9.2 Morphology and Dynamic Mechanical Properties 9.2.1 Blends with Polyethylene or Poly(butene-1) 9.2.2 Blends with Ethylene–a-Olefin Copolymers 9.2.3 Ethylene–Isotactic Propylene Copolymers 9.3 Tensile and Rheo-Optical Properties 9.3.1 Principles for Rheo-Optical Characterization 9.3.2 Blends with Ethylene–a-Olefin Copolymers 9.3.3 Blends with Novel Ethylene–Isotactic Propylene Copolymers 9.4 Solidification Process and Final Morphology 9.4.1 Morphology Formation During Crystallization 9.4.2 Structure and Properties of Injection-Molded Products

172 177 177 181 182 187 188 188 193 195 196

198 198 199 199 201 206 206 208 211 214 214 219 221 222 222 222

224 224 225 225 226 236 241 241 242 247 250 250 257

Contents 9.5 Conclusions Nomenclature References

10. Functionalization of Olefinic Polymer and Copolymer Blends in the Melt 10.1 10.2

Introduction Scope of Review 10.2.1 Free-Radical Grafting of Unsaturated Monomers to PO Chains 10.2.2 The Use of Monomers and Initiators 10.3 Functionalization of PP/PE Blends 10.3.1 Effect of Reacting Blend Formulation on Grafting Efficiency and Rheological and High Elastic Properties of Melt of Functionalized PP/PE Blends 10.3.2 Structure and Mechanical Properties of Functionalized PP/PE Blends 10.4 Functionalization of PP/EPR Blends 10.5 Functionalization Features of Blends: PE/EPR, PP(PE)/EOC, and PP(PE)/Styrene Polymer 10.6 Use of Functionalized Polyolefin Blends 10.7 Conclusion Nomenclature References

11. Deformation Behavior of b-Crystalline Phase Polypropylene and Its Rubber-Modified Blends 11.1 11.2

Introduction Deformation Characteristics 11.2.1 Static Tensile Behavior 11.2.2 Strain-Induced b ! a Phase Transition 11.2.3 Impact Behavior 11.3 Fracture Toughness 11.3.1 General Aspects 11.3.2 Mode I LEFM Approach 11.3.3 Impact Fracture Toughness 11.3.4 Essential Work of Fracture 11.4 Conclusions Nomenclature References

12. Multiphase Polypropylene Copolymer Blends 12.1

Introduction 12.1.1 Commercial Production 12.1.2 Morphology of Commercial Impact PP Copolymers

ix 264 265 265

269 269 270 270 275 284

284 291 295 297 299 300 301 302

305 305 309 310 314 323 330 330 331 339 341 347 347 348

351 351 352 352

x

Contents 12.2 12.3 12.4

Dispersive Mixing during Processing Molecular Structure of Impact PP Copolymers Coarsening in Multiphase PP Copolymer Systems 12.4.1 Background 12.4.2 Coarsening of High Impact Polypropylene 12.4.3 Coarsening of Model Blends 12.4.4 Interfacial Effects in Polypropylene Copolymer Systems 12.5 Conclusions Nomenclature References

13. Heterogeneous Materials Based on Polypropylene 13.1 13.2 13.3

Introduction The Interphase: Definition Magnitude Orders in the Interphase 13.3.1 The Dispersed Phase 13.3.2 The Matrix 13.3.3 The Interphase: Designing the Interface 13.4 Interfacial Modification of Heterogeneous Materials Based on Polypropylene 13.4.1 Composites: When the Dispersed Phase is Rigid 13.4.2 Blends: When the Dispersed Phase is Flexible 13.4.3 The Role of the Interfacial Modifiers from the Matrix Side 13.5 Interfacial Modifiers Based on Polypropylene 13.5.1 The Kinetic Approach: Basic Aspects 13.5.2 Chemical Modification of Polypropylenes by Grafting of Polar Monomers 13.6 Conclusions Acknowledgment Nomenclature References

14. Polypropylene/Ethylene–Propylene–Diene Terpolymer Blends 14.1 14.2

14.3

Introduction PP/EPDM Blends 14.2.1 Toughness and Crystallization Behaviors of PP/EPDM Blends 14.2.2 Compatibilization of PP/EPDM Blends 14.2.3 Ternary Blends and Composites from PP/EPDM Blends 14.2.4 Application of Radiation Dynamically Vulcanized PP/EPDM Blends (or Thermoplastic Vulcanizates (TPVs)) 14.3.1 Effect of Cross-linking on the Properties of PP/EPDM TPVs

357 360 360 360 365 368 370 373 376 377

379 379 380 380 382 383 383 385 387 387 388 397 397 398 407 408 408 408

411 411 412 412 414 416 417 419 420

Contents 14.3.2 Microstructure of PP/EPDM TPV 14.3.3 PP/EPDM/Ionomer TPVs 14.3.4 Mechanical and Rheological Properties 14.4 Applications of PP/EPDM Blends 14.5 Conclusions Acknowledgments Nomenclature References

15. Ethylene–Propylene–Diene Rubber/Natural Rubber Blends 15.1 15.2

Introduction Miscibility, Compatibility, and Thermodynamics of Polymer Blending 15.3 Blend Preparation 15.4 Covulcanization 15.5 Filler Distribution in NR/EPDM Blends 15.6 Morphology of NR/EPDM Blends 15.7 Compatibilization of NR/EPDM Blends 15.8 Mechanical and Viscoelastic Properties 15.8.1 Mechanical Properties 15.8.2 Dynamic Mechanical Properties 15.9 Rheological Properties 15.10 Thermal Properties 15.11 Electrical Properties 15.12 Aging Properties 15.12.1 Thermal Aging 15.12.2 Ozone Resistance 15.13 Transport Properties 15.14 Applications 15.15 Conclusions Nomenclature References

16. Phase Field Approach to Thermodynamics and Dynamics of Phase Separation and Crystallization of Polypropylene Isomers and Ethylene–Propylene–Diene Terpolymer Blends 16.1 16.2

16.3

Introduction Experimental Phase Diagrams 16.2.1 Cloud Point Phase Diagram of iPP/EPDM Blends 16.2.2 Cloud Point Phase Diagram of sPP/EPDM Blends Thermodynamic Free Energy Description of Crystalline Polymer Blends 16.3.1 Flory–Huggins Free Energy of Amorphous–Amorphous Blends

xi 423 425 428 436 437 438 438 439 441 441 442 443 444 447 448 450 452 452 456 458 460 461 462 462 463 465 466 469 469 470

473 473 475 475 476 478 478

xii

Contents 16.3.2

Extension of the FH Theory to Crystal–Amorphous Blends 16.3.3 Prediction of Phase Diagram Topologies 16.3.4 Comparison with Experimental Phase Diagrams of PP/EPDM Blends 16.4 Phase Field Modeling on Polymer Phase Transitions 16.4.1 Theory on Phase Separation Dynamics and Morphology Evolution 16.4.2 Dynamics of Crystal Growth in a Phase Separating System: iPP/EPDM Blends 16.4.3 Dynamics of Crystal Growth in a Phase Separating System: sPP/EPDM Blends 16.5 Conclusions Nomenclature References

Part III

479 481 484 486 486 488 491 496 496 497

Polyolefin/Nonpolyolefin Blends

499

17. Compatibilization and Crystallization of Blends of Polyolefins with a Semiflexible Liquid Crystalline Polymer

501

17.1

Blends of Polyolefins (High Density Polyethylene and Isotactic Polypropylene) with a Semiflexible Liquid Crystalline Polymer 17.1.1 Introduction 17.1.2 Blends of High Density Polyethylene (HDPE) with LCP 17.1.3 Blends of Isotactic Polypropylene with LCP 17.2 Crystallization Behavior of Blends of Polyolefins with a Semiflexible Liquid Crystalline Polymer 17.2.1 Isothermal and Nonisothermal Crystallization of Blends of Linear Low Density Polyethylene with a Semiflexible Liquid Crystalline Polymer 17.2.2 Crystallization Behavior of PE-g-LCP Copolymers 17.2.3 Effect of PP-g-LCP Compatibilizer on the Morphology and Crystallization of PP/LCP Blends 17.2.4 Isothermal Crystallization Kinetics of Compatibilized Blends of Polyolefins with a Semiflexible LCP 17.2.5 Crystallization and Morphology of Fibers Prepared from Compatibilized Blends of Polyethylene with a Liquid Crystalline Polymer 17.3 Conclusions Acknowledgment Nomenclature References

501 501 502 509 513

513 517 519 519

522 523 523 523 524

Contents

18. Functionalized Polyolefins and Aliphatic Polyamide Blends: Interphase Interactions, Rheology, and High Elastic Properties of Melts 18.1 18.2 18.3 18.4 18.5 18.6

Introduction Compounding and Interphase Phenomena in PA/PO Blends Rheological and High Elastic Properties of PA/PO Melts Morphology and Impact Strength of PA/g-PO Blends that Give Melts of High Viscosity and Strength New Applications of PA/g-PO Blends Conclusions

Nomenclature References 19. Plastic Deformation and Damage Mechanisms of Ternary PP/PA6/POE Polymer Blends 19.1 19.2

Introduction Materials Presentation and Experimental Methods 19.2.1 Materials Presentation 19.2.2 Morphological Study under SEM and TEM 19.2.3 Video-controlled Tensile System 19.3 Microstructure 19.4 General Mechanical Properties 19.4.1 Dynamic Mechanical Thermal Analysis 19.4.2 Toughness by Impact Loading and Yield Stress by Tension 19.5 Plastic Deformation under Uniaxial Tension 19.5.1 Definition of Volume Strain 19.5.2 True Axial Stress–Strain Relation 19.5.3 Volume Strain 19.5.4 Under Cyclic Tension 19.6 Mechanisms of Plastic Deformation and Damage 19.6.1 Damage Mechanisms in Polymer Blends 19.6.2 Influence of Damage Type 19.6.3 Damage-induced Shear Banding 19.6.4 Microscopic Observation under SEM 19.6.5 Microscopic Observation under TEM 19.6.6 Discussion 19.7 Conclusions Nomenclature References

20. Reactive Compatibilization of Binary and Ternary Blends Based on PE, PP, and PS 20.1 20.2

Introduction Friedel–Crafts Alkylation Reaction

xiii

527 527 528 534 543 550 550

551 551

556 556 558 558 559 559 561 566 566 567 569 569 571 573 575 579 579 580 581 582 589 589 596 597 597

600 600 601

xiv

Contents 20.3

Binary Blends 20.3.1 PE/PS Blends 20.3.2 PP/PS Blends 20.4 Ternary Blends: PE/PP/PS 20.5 Conclusions Nomenclature References

21. Polyolefin/Epoxy Resin Blends 21.1 21.2 21.3 21.4 21.5 21.6 21.7 21.8 21.9 21.10

Introduction Blend Preparation Miscibility Studies and Phase Diagrams Cure Kinetics Crystallization Behavior Morphology Dynamic Mechanical Properties Mechanical Properties Conductivity Studies Conclusion

Nomenclature References Index

603 603 611 615 620 620 621

623 623 624 625 629 631 637 644 647 649 656

658 659 663

Preface

Polyolefins are the most widely used commodity thermoplastics. They are of immense interest to polymer community because of their simple chemical structures and fascinating hierarchical structural organizations possible. To date, the field of polyolefins remains one of the most vibrant areas in polymer research. Polyolefin blends are a subset of polymer blends that emerged as a result of the need to meet application requirements not satisfied by synthesized neat polyolefins. In comparison to other subsets of polymer blends, polyolefin blends have distinct advantages of lower density, lower cost, processing ease, and good combination of chemical, physical, and mechanical properties. In the last several years, research and usage of polyolefin blends have increased due to new application opportunities (e.g., in medical and packaging) and the development of novel polyolefins. Although a sizable number of books on polyolefins and general polymer blends are available, only a few chapters address polyolefin blends. Currently, there is no single book that focuses exclusively on the fundamental aspects and applications of polyolefin blends. This is the primary source of motivation behind this book. The second motivation stems from the fact that new research trends in polyolefin blends such as in situ reactor blending and compatibilization/functionalization in the melt have emerged that need to be covered in a book format. This book is structured as follows: Chapter 1 serves as a guide to polyolefin blends introducing this important class of materials, why they are important, typical systems studied, issues of fundamental and applied interest, and current trends. The contributed chapters are divided into two main categories: polyolefin/polyolefin blends (Chapters 2–16) and polyolefin/nonpolyolefin blends (Chapters 17–21). Issues covered in these chapters include miscibility, phase behavior, functionalization, compatibilization, microstructure, crystallization, hierarchical morphology, and physical and mechanical properties. Most of the chapters are in the form of review articles. Some original articles are included to capture the latest development in polyolefin blends research. This book is intended to serve as a valuable reference for academic and industrial professionals performing research and development in the specific area of polyolefin blends or in the general area of polymer blends. Some review chapters include introductory materials to attract newcomers including senior undergraduate and graduate students and to serve as a reference book for professionals from other disciplines. Some knowledge of polymer chemistry, physics, and engineering, although not strictly essential, would be helpful to better appreciate the technical information of some chapters. Since this book is the first of its kind devoted solely to polyolefin blends, it is hoped that it will be sought after by a broader technical audience. xv

xvi

Preface

The chapters in this book were contributed by highly reputed professionals from academia, industry, and government laboratories spanning several countries from various continents. All manuscripts were peer reviewed in accordance with the guidelines utilized elsewhere by top-rated polymer journals. The editors would like to thank all contributors for believing in the realization of this book and taking painstaking tasks of going through the processes of manuscript preparation, submission, review, revision, and seeking supporting documents. Finally, sincere thanks are extended to all reviewers for their invaluable help, which undoubtedly improved the quality of this book. DOMASIUS NWABUNMA THEIN KYU

Contributors

Susana Areso Capdepoń, Department of Physics and Engineering of Polymers, Polymer Engineering Group, Institute of Science and Technology of Polymers, CSIC CL Juan de la Cierva 3, 28006 Madrid, Spain. [email protected]

Shu-Lin Bai, Centre for Advanced Composite Materials (CACM), Department of Mechanics and Engineering Science, School of Engineering, Peking University, 100871 Beijing, China. [email protected] Silvia E. Barbosa, Planta Piloto de Ingenierıá Quı´mica, PLAPIQUI (UNS-CONICET), Camino La Carrindanga km. 7 (8000), Bahıá Blanca, Argentina. [email protected]

Maurizio Canetti, C.N.R. Istituto per lo Studio delle Macromolecole, Via E. Bassini 15, I-20133 Milano, Italy. [email protected] Numa J. Capiati, Planta Piloto de Ingenierıá Quı´mica, PLAPIQUI (UNS-CONICET), Camino La Carrindanga km. 7 (8000), Bahıá Blanca, Argentina. [email protected]

Mońica F. Dıáz, Planta Piloto de Ingenierıá Quı´mica, PLAPIQUI (UNS-CONICET), Camino La Carrindanga km. 7 (8000), Bahıá Blanca, Argentina. [email protected]

Bejoy Francis, Department of Chemistry and Biochemistry, Laurentian University, 935 Ramsey Lake Road, Sudbury, Ontario, P3E 2C6, Canada. [email protected]

Soney C. George, Department of Basic Science, Amal Jyothi College of Engineering, Koovapally, Kottayam 686518, Kerala, India. [email protected]; [email protected] Christian G’Sell, Laboratoire de Physique des Mate´riaux, Ecole des Mines de Nancy, Parc de Saurupt, 54042 Nancy Cedex, France. [email protected]

Chang-Sik Ha, Department of Polymer Science and Engineering, Pusan National University, Busan 609-735, Korea. [email protected]

Jean-Marie Hiver, Laboratoire de Physique des Mate´riaux, Ecole des Mines de Nancy, Parc de Saurupt, 54042 Nancy Cedex, France. [email protected]

Benjamin S. Hsiao, Department of Chemistry, Stony Brook University, Stony Brook, NY 11794, USA. [email protected]

Boleslaw Jurkowski, Division of Plastic and Rubber Processing, Institute of Material Technology, Poznan University of Technology, Piotrowo 3, 60-950 Poznan, Poland. [email protected]

Wirunya Keawwattana, Department of Chemistry, Faculty of Science, Kasetsart University, Bangkok 10903, Thailand. [email protected]

xvii

xviii

Contributors

Gue-Hyun Kim, Division of Applied Engineering, Dongseo University, Busan 617-716, Korea. [email protected] Il Kim, Department of Polymer Science and Engineering, Pusan National University, Busan 609-735, Korea. [email protected]

Yuri M. Krivoguz, Laboratory of Chemical Technology of Polymeric Composite Materials, V.A. Belyi Metal-Polymer Research Institute of National Academy of Sciences of Belarus, 32a Kiroc Street, 246050 Gomel, Belarus. [email protected] Thein Kyu, Department of Polymer Engineering, University of Akron, Akron, OH 44325, USA. [email protected]

Jesu´s Marıá Garcıá Martıńez, Department of Physics and Engineering of Polymers, Polymer Engineering Group, Institute of Science and Technology of Polymers, CSIC CL Juan de la Cierva 3, 28006 Madrid, Spain. [email protected]

Rushikesh A. Matkar, Department of Polymer Engineering, University of Akron, Akron, OH 44325, USA. [email protected]

Liliya Minkova, Institute of Polymers, Bulgarian Academy of Sciences, Acad. G. Bonchev str. Bl.103A, 1113 Sofia, Bulgaria. [email protected]

Francis M. Mirabella, Lyondell Chemical Co., Equistar Technology Center, Cincinnati, OH 45249, USA. [email protected] Koh-Hei Nitta, Department of Chemical Engineering, Graduate School of Material Sciences, Kanazawa University, 920-1192, Japan. [email protected]

Domasius Nwabunma, 3M Company, Safety, Security, and Protection Business Services Laboratory, St. Paul, MN 55144, USA. [email protected] Emilia Pe´rez Collar, Department of Physics and Engineering of Polymers, Polymer Engineering Group, Institute of Science and Technology of Polymers, CSIC, CL Juan de la Cierva 3, 28006 Madrid, Spain. [email protected]

Stepan S. Pesetskii, Laboratory of Chemical Technology of Polymeric Composite Materials, V.A. Belyi Metal-Polymer Research Institute of National Academy of Sciences of Belarus, 32a Kirov Street, 246050 Gomel, Belarus. [email protected]

Subhendu Ray Chowdhury Department of Materials Science and Engineering, Pennsylvania State University, University Park, PA 16802, USA. [email protected]

Moonhor Ree, Department of Chemistry, Polymer Research Institute, Pohang Accelerator Laboratory, National Research Lab for Polymer Synthesis and Physics, and Center for Integrated Molecular Systems, Pohang University of Science & Technology (Postech), Pohang 790-784, Republic of Korea. [email protected]

Robert A. Shanks, School of Applied Sciences, RMIT University, GPO Box 2476V, Melbourne, Vic 3001, Australia. [email protected]

Jesu´s Taranco Gonza´lez, Department of Physics and Engineering of Polymers, Polymer Engineering Group, Institute of Science and Technology of Polymers, CSIC, CL Juan de la Cierva, 3, 28006 Madrid, Spain. [email protected]

Contributors

xix

Kohji Tashiro, Department of Future Industry-Oriented Basic Science and Materials, Toyota Technological Institute, Tempaku, Nagoya 468-8511, Japan. [email protected] Sabu Thomas, School of Chemical Sciences, Mahatma Gandhi University, Priyadarshini Hills, Kottayam 686560, Kerala, India. [email protected]

Sie C. Tjong, Department of Physics and Materials Science, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong. [email protected] Shigeyuki Toki, Department of Chemistry, Stony Brook University, Stony Brook, NY 11794, USA. [email protected] Gong-Tao Wang, School of Aerospace, Mechanical & Mechatronic Engineering, The University of Sydney, Sydney, NSW 2006, Australia. [email protected] Min Wang, Centre for Advanced Composite Materials (CACM), Department of Mechanics and Engineering Science, School of Engineering, Peking University, 100871 Beijing, China. [email protected]

James L. White, Department of Polymer Engineering, The University of Akron, Akron, OH, 44325, USA. [email protected] Masayuki Yamaguchi, School of Materials Science, Japan Advanced Institute of Science and Technology, Nomi, Japan. [email protected] Jinhai Yang, Department of Polymer Engineering, The University of Akron, Akron, OH, 44325, USA. [email protected]

Part I

Introduction

Chapter

1

Overview of Polyolefin Blends Domasius Nwabunma1

1.1 INTRODUCTION Polyolefins are synthetic polymers of olefinic monomers. They are the largest polymer family by volume of production and consumption. Several million metric tons of polyolefins are produced and consumed worldwide each year, and as such they are regarded as commodity polymers. Polyolefins have enjoyed great success due to many application opportunities, relatively low cost, and wide range of properties. Polyolefins are recyclable and significant improvement in properties is available via blending and composite technologies. Polyolefins may be classified based on their monomeric unit and chain structures as ethylene-based polyolefins (contain mostly ethylene units), propylene-based polyolefins (contain mostly propylene units), higher polyolefins (contain mostly higher olefin units), and polyolefin elastomers (1). Ethylene-based polyolefins are normally produced either under low pressure conditions using transition metal catalysts resulting in predominantly linear chain structure or under high pressure conditions using oxygen or peroxide initiators resulting in predominantly branched chain structures of various densities and crystallinity levels. Propylene-based polyolefins are normally produced with transition metal catalysts resulting in linear chain structures with stereospecific arrangement of the propylene units or special stereoblock structures from a single-site catalyst. Higher polyolefins are normally produced using transition metal catalysts resulting in linear and stereospecific chain structures. Polyolefin elastomers based mainly on a combination of ethylene and propylene may be produced using metal or single-site catalysts with or without the inclusion of dienes (for cross-linking) and are mostly amorphous with high molecular weights and heterogeneous in phase structures. One

1 3M Company, Safety, Security, and Protection Business Services Laboratory, St. Paul, MN 55144, USA

Polyolefin Blends, Edited by Domasius Nwabunma and Thein Kyu Copyright # 2008 John Wiley & Sons, Inc.

3

4

Polyolefin Blends

may conclude that a given polyolefin may be a homopolymer, copolymer, or terpolymer depending on the number of monomers used in making the polyolefin, crystalline or amorphous depending on their chain conformation, configuration, and processing conditions. Today, polyolefins and polyolefin-based materials are used in many applications. These applications include transportation (automotive, aerospace), packaging, medical, consumer products (toys, appliances, etc.), electronics, cable and wire coating, thermal and acoustic insulation, and building and construction. Polyolefins can be extruded as filaments (fibers), films (cast and blown), and pipes/profiles. They can be molded into parts of various shapes. They can be foamed with physical and chemical foaming/blowing or/and can be coated onto other materials.

1.2 OLEFINIC MONOMERS The alkenes having one or more unsaturated double bonds in their structures are the monomers used to synthesize polyolefins. They have the general formula Cn H2n ; n 2. Table 1.1 shows the first 10 members of the olefinic monomers with one double bond, which are often called a-olefins. The monomers in Table 1.1 form a homologous series of hydrocarbon compounds. Thus, apart from having the same general formula, all compounds in the series have the same functional groups. Each member of the group differs from the next in the series by the CH2 group equivalent to 14 relative molecular mass units. All members of the series have similar chemical properties. The physical properties of the compounds in the series show a progressive change with increasing relative molecular mass. The first three members of the alkenes homologous series are gases at room temperature. Those containing between 5 and 15 carbon atoms are colorless liquids and the higher compounds are waxy solids at room temperature. These a-olefinic monomers may be obtained as products of the cracking of the gas– oil and naphtha fractions of petroleum distillations. They can also be obtained from synthetic organic chemistry methods.

Table 1.1

Alkene Monomers with One Double Used in the Synthesis of Polyolefins.

No. of carbon atoms (n) 2 3 4 5 6 7 8 9 10

Formula (Cn H2n ; n 2)

Name (other name)

C2H4 C3H6 C4H8 C5H10 C6H12 C7H14 C8H16 C9H18 C10H20

Ethene (ethylene) Propene (propylene) Butene-1 (butylene) Pentene-1 Hexene-1 Heptene-1 Octene-1 Nonene-1 Decene-1

Chapter 1

Overview of Polyolefin Blends

5

1.3 POLYOLEFIN HOMOPOLYMERS, COPOLYMERS, AND TERPOLYMERS Polyolefin homopolymers, copolymers, and terpolymers are foundation materials for polyolefin blends. They may be obtained via radical or ionic chain growth polymerization of alkenes using conventional free radicals (e.g., from peroxides) and organometallic complexing (Ziegler–Natta and metallocenes) catalyst systems. Polyolefin polymerization technologies and novel catalyst systems have enabled the rapid development of polyolefins with a wide range of molecular chain structures, morphologies, properties, and particle size and shape. Polyolefin homopolymers include polyethylene (PE), polypropylene (PP), polybutene-1 (PB), polymethylpentene-1 (PMP), and higher polyolefins. Table 1.2 shows the structures of commercial polyolefin homopolymers. Of these, PE and PP are the largest by amount produced yearly by the global polyolefin companies (1). PE comes in various forms differing in chain structures, crystallinity, and density levels. These are high density polyethylene (HDPE), low density polyethylene (LDPE), linear low density polyethylene (LLDPE), ultralow density polyethylene (ULDPE), and ultrahigh molecular weight polyethylene (UHMWPE). PP and higher polyolefins come in three stereospecific forms of varying densities: isotactic, syndiotactic, and atactic forms. Polyolefin copolymers involve two olefinic monomers. The process of copolymerization is normally used to control the properties of the polyolefins. Some of the consequences of copolymerization are reduced crystallinity, melting point, modulus, strength, hardness, and low temperature impact. Polyolefin copolymers are either random or block copolymers of same or different monomers and may be a single phase or heterophasic depending on the amount of comonomer, the polymerization catalyst, and the process. For polyolefin copolymer of same monomers, this can be achieved by having different segments of the copolymer with different tacticities. One can have polyolefin block copolymers of same block or of varying block lengths. One can also have polyolefin copolymers consisting of both block and random segments together in the same macromolecule. Polyolefin copolymers are usually not homogeneous in composition but are actually mixtures of copolymers of varying compositions. It is also possible with polyolefins to have block copolymers with only one monomer. These are called stereoblock copolymers and can be achieved by having sections of the polyolefin copolymer possess different tacticities. Polyolefin copolymers started with LLDPE and ethylene–propylene rubber (EPR). Today, there are polyolefin copolymers of ethylene with butene-1, hexene1, octene, cyclopentene, and norbornene and copolymers of propylene with butene-1, pentene-1, and octene-1 in addition to ethylene. There are copolymers of butene-1 with pentene-1, 3-methylbutene-1, 4-methylpentene-1, and octene in addition to its copolymers with ethylene and propylene. There are copolymers of 4-methylpentene1 with pentene-1 and hexene-1 in addition to its copolymers with butene-1 and propylene. The function of the comonomers is to reduce crystallinity, as compared to the homopolymers, resulting in copolymers that are highly elastomeric with very low

6

Polyolefin Blends

Table 1.2

Structures of Commercial Polyolefin Homopolymers.

Name (other name) Polyethylene (polyethene, polymethylene)

Chemical structure (repeat unit) CH

CH 2 n

2

Polypropylene (polypropene)

CH

2

Polybutylene (polybutene-1)

CH

n

C H2 CH3

Polyisobutylene (polyisobutene-1)

Polybutadiene

CH

CH

CH

2

CH

CH

2

CH

2 n

n

C H2

Poly-4-methylpentene-1

CH3 C H3

CH3 CH 3

Polyisoprene CH 2

C

CH

CH

2 n

glass transition temperatures, high impact strength, low modulus, low density, and are often optically transparent. The most widely used multiphase polyolefin copolymer is polypropylene impact copolymer. These copolymers are typically composed of isotactic polypropylene (iPP) and EPR. Impact polypropylene copolymers are produced by various processes, but are generally characterized by the synthesis of iPP in the first reactor and EPR in the second reactor. Therefore, these systems are typically reactor blends. Postreactor blending can be done, but the starting material is most often the reactor blend polypropylene copolymer. Polyolefin copolymers are often used for film applications or as impact modifiers. Polyolefin terpolymers contain three olefinic monomers. A well-known example is ethylene propylene diene monomer (EPDM). The diene (double bond) monomer is

Chapter 1


7

usually ethylidene norbornene or 1,4-hexadiene. EPDM was introduced because of the difficulty in cross-linking saturated polyolefin homopolymers and copolymers. There are also functionalized polyolefins. These are usually copolymer or terpolymer containing functional groups like epoxide, anhydride, hydroxyl, acrylate, and carboxylic acid. These functional groups are either grafted onto the polyolefin after polymerization or added directly in situ during polymerization reactions involving olefins and functional groups bearing polar monomers such as vinyl acetate, methyl acrylate, butyl acrylate, glycidyl methacrylate, and acrylic acid. Functionalized polyolefins are useful compatibilizers and impact modifiers in blends and composites containing polyolefins and nonpolyolefins. In this sense, functionalized polyolefins may be considered as additives rather than matrix materials in formulation of polyolefin blends. Commercial polyolefins often contain additives such as colorants, flame retardants, antioxidants, light stabilizers, nucleating agents, antistatic agents, lubricants (microcrystalline waxes, hydrocarbon waxes, stearic acid, and metal stearates), and so on. These additives aid the processing and fabrication of products from polyolefins. Detailed treatments about specific polyolefins, polymerization systems/ mechanism/processes, structures, properties, processing, and applications may be found in References 2–9.

1.4 POLYOLEFIN BLENDS Polymer blends (mixtures of structurally different polymers (10–19)) are of interest because synthesized polymers have not satisfied increasing application demands. Polyolefin blends are a subset of polymer blends and may be classified into two groups. The first group contains polyolefins only, which are formulated to broaden the range of structures, properties, and applications offered by polyolefins. The second group contains polyolefins and nonpolyolefins, which are formulated to mitigate some of the property drawbacks of the polyolefin or the nonpolyolefin. For a blend to be classified as a polyolefin blend, it is presumed that the polyolefin component is of significant composition in the blend. In terms of miscibility, polyolefin blends may also be classified as miscible and immiscible blends (10, 11). Polyolefin blending requires knowledge of the miscibility and crystallinity of the blend, in addition to the contributions of the components of the blend. Miscibility depends on molecular structure, blend composition, and mixing temperature. To characterize miscibility, a phase diagram is needed. Nonolefinic thermoplastic polymers that in principle may be blended with polyolefins include polyamides (nylons) such as polyamide 6, polyamide 66, polyphenylene sulfide (PPS), polyphenylene ether (PPE), and polyphenylene oxide (PPO); polyesters such as polyethylene terephthalate (PET), polybutylene terephthalate (PBT), polyethylene naphthalate (PEN), polytrimethylene terephthalate (PTT), polycarbonates, polyethers, and polyurethanes; vinyl polymers such as polystyrene (PS), polyvinyl chloride (PVC), polymethylmethacrylate (PMMA), and ethylene

8

Polyolefin Blends

vinyl acetate copolymer (EVA); block and graft copolymers (styrene–acrylonitrile copolymer, styrene–butadiene copolymer, styrene–ethylene–butadiene–styrene terpolymer, etc.); and liquid crystalline polymers (LCPs). Thermosetting resins that may be blended with polyolefins include, but are not limited to, the following: epoxies, unsaturated polyesters, phenol formaldehyde, melamine formaldehyde, urea formaldehyde, silicones, and so on. The properties that polyolefins normally contribute in blends with other polymers include high melt strength and elasticity, toughness, low viscosity for processability, low polarity, dielectric constant, and loss, and chemical resistance and moisture absorption resistance. Nonpolyolefins contribute to high modulus, heat resistance, and oxygen or solvent barrier properties. For example, barrier properties of polyolefins can be improved by blending with polymers such as ethylene vinyl alcohol and polyvinylidene chloride. Blends of polyolefins with nylons and polycarbonate (PC) allow balanced control of permeability and water retention. Polystyrene (PS) is an interesting candidate for blending with polyolefins for mechanical reasons as well as for paintability and printability. There are two classes of polyolefin blends: elastomeric polyolefin blends also called polyolefin elastomers (POE) and nonelastomeric polyolefin blends. Elastomeric polyolefin blends are a subclass of thermoplastic elastomers (TPEs). In general, TPEs are rubbery materials that are processable as thermoplastics but exhibit properties similar to those of vulcanized rubbers at usage temperatures (19). In TPEs, the rubbery components may constitute the major phase. However, TPEs include many other base resins, which are not polyolefins, such as polyurethanes, copolyamides, copolyesters, styrenics, and so on. TPEs are now the third largest synthetic elastomer in total volume produced worldwide after styrene– butadiene rubber (SBR) and butadiene rubber (BR). Two important types of elastomeric polyolefin blends are reactor-made iPP/ EPR blends and postreactor blend iPP/EPDM. The latter is called thermoplastic vulcanizates (TPVs), produced by dynamic vulcanization of blends containing a thermoplastic and an elastomer. To make iPP/EPDM TPV, the two polymers PP and EPDM are mixed with curatives, such as peroxides, phenolic resins, or sulfur with accelerators, and dynamically cured in an extruder resulting in a blend consisting of micrometer-sized elastomer particles dispersed in the PP matrix (20–24). Paraffinic oils are added in the melt mixing process for viscosity control and cost. In iPP/ EPDM TPV, the crystalline iPP resin is normally the minor phase. Recently, polyolefin plastomers have been added to the class of elastomeric polyolefin blends. Polyolefin plastomers are ultralow molecular weight linear low density polyethylenes (ULMW-LLDPE). Nonelastomeric polyolefin blends are blends of polyolefins with mostly nonpolyolefin (other thermoplastic) matrices as mentioned earlier. Polyolefin blends of commercial importance are normally made via two methods: blending in the melt either during polymerization or mechanically after the polymerization process. The first method, called in-reactor blending, involves the blending of different polyolefins (homopolymers, random, and block copolymers) in a polymerization reactor. This is enabled by the presence of multiple catalyst species

Chapter 1


9

in the polymerization recipe. A good example is in-reactor-made EPR/iPP blend, which is normally prepared by adding ethylene monomer to propylene monomer toward the end of propylene polymerization process. The function of EPR is to improve iPP flexibility; hence, EPR/iPP blend is often called toughened or impact PP and finds wide applications in the consumer industry. Another good example of in-reactor-made polyolefin blend is linear low density polyethylene (LLDPE), which often contains several ethylene/a-olefin copolymers that differ in ethylene contents. LLDPE and reactor-made EPR/iPP are often called thermoplastic olefins (TPOs). The second method, called postreactor blending, involves mechanical blending of a premade polyolefin with other polyolefins or nonpolyolefins in compounding extruders. A practical example of polyolefin blend made using this method is a blend of isotactic PP with cured EPDM, as described earlier. Other examples of polyolefin blends made by postreactor mechanical blending include polyolefin/polyamides (nylons), polyolefin/polyesters, polyolefin/polystyrene (PS), polyolefin/polyvinyl chloride (PVC) blends, and so on. EPR/iPP blend can also be made by postreactor mechanical blending as well as by in-reactor blending. Postreactor blending via single and twin screw compounding is still the preferred method of polyolefin blending because it is quick, easy, economical, and efficient. Table 1.3 shows a summary of specific polyolefin blends that have been studied in the literature extracted from references (25–310). These blends involve the following polyolefins: PP, PE (LLDPE, LDPE, HDPE, and UHMWPE), EPR, EPDM, and PB. Some of the blends listed in Table 1.3 are of commercial importance (6). There are many publications (journal articles and patents) on polyolefin blends in the literature. Tables 1.4 contains a summary of the journal articles (25–310) and patents published each year during the 6-year period 2000–2005. The choice of date range is arbitrary. The number of journal articles for each year was obtained from a search of electronic version of English-based polymer and polymer-related journals using the keywords polyolefin and blends. Within polyolefin keyword, the subkeywords used in the search were polyethylene (PE, LLDPE, LDPE, HDPE, UHMWPE, PE, etc.), polypropylene (PP, iPP, sPP, aPP, etc.), polybutene-1, poly-4-methylpentene-1, ethylene–diene monomer, ethylene–propylene– diene terpolymer, ethylene propylene rubber, thermoplastic olefins, natural rubber (NR), polybutadiene, polyisobutylene (PIB), polyisoprene, and polyolefin elastomer. For the polyolefin blends patent search, polymer indexing codes and manual codes were used to search for the patents in Derwent World Patent Index based on the above keywords listed in the search strategy. Table 1.4 shows an increasing trend in the number of publications. It should be noted that while generating Table 1.4, some publications may have been missed in the reference search period (2000–2005). There are several issues of interest in polyolefin blends research. They may be categorized into formulation design and processing; miscibility, structural, molecular, and property characterization; end-use properties and performance; and

10

PE/silicon rubber (76) PE/starch (85, 277, 289, 295) PE/hydrolyzed collagen (98) PE/natural rubber/PP (238, 254, 302) PE/poly(3-hydroxybutyrate) (251) PE/nitrile rubber (100, 183)

PE/polydianilinephosphazene (91)

PE/PS (47–48, 67, 79, 123, 145, 198, 219, 223) PE/PP/PA66 (71)

PE/PA6 (25, 27, 71, 88, 92, 96, 130, 140, 144, 149, 168, 204, 215, 224, 226, 272, 273) PE/PP (33, 34, 36, 47, 62, 69, 83, 93, 107, 116, 182, 196, 200, 203, 236, 237, 239, 243, 248, 255, 263, 297) PE/PP/cycloolefin copolymers (307) PE/ethylene-co-propylene-co-butene-1 (99) PP/PPE (216, 217) PP/EPDM (50, 66, 113, 121, 136, 199, 202, 207, 224, 267) PP/EPR (35, 63, 77, 118, 124, 152, 163, 195, 231, 233, 271) PP/epoxy (62, 120) PP/PA6/PS (205) PP/cyclopolyolefin (162) PP/natural rubber (64, 241, 247, 287) PP/recycle rubber(241, 247) PP/natural rubber/recycle rubber powder (249)

PE/PP/PA6 (71)

LLDPE/LDPE/wax (80)

Polyolefin blends PE/PP/EPR (35, 114) PE/PC (29, 173, 213) PE/poly(silsesquioxanes) (156) PE/ethylene acrylic elastomer (281)

Polyolefin Blends Studied in the Literature.

HDPE/UHMWPE (81) LDPE/HDPE (84, 153, 160, 221, 259, 284) LLDPE/HDPE (285) LLDPE/LDPE (72, 104, 108, 119, 128, 129, 181, 211, 214, 221, 242, 258, 261, 292, 309)

Table 1.3

PP/EPDM/epoxidized natural rubber (246) PP/epoxidized natural rubber (301) Butadiene/styrene–butadiene rubber (293) EPDM/PA66 EPDM/SBR (68) EPDM/NBR (234)

PP/EPDM/natural rubber (244–246)

PP/PC (25) iPP/sPP (40); iPP/aPP (103)

PP/PB (26, 109)

PP/PA6 (28, 54, 58, 71, 111, 170, 178, 262, 283, 286, 288, 305) PP/PA6/ethylene-co-octene (28, 31, 43, 46)

PP/PET (137) PP/PE/EPDM (49) PP/EVOH (135, 298) PP/PA66 (28)

11

PE/PBT (82, 300) PE/PS/PMMA (209) PE/wax (256)

PE/PA12 (142) PE/PTT (37) PE/PVC (52, 201) PE/starch/PCL (90) PE/ethylene-co-octene (46, 278) PE/ethylene-co-butene-1 (89, 171) PE/PA6,6 (127, 169, 273) PE/perfluoropolyether (126) PE/EVA (45, 51, 56, 78, 97, 161, 270) PE/PET (166, 167, 208, 213, 220, 274)

PP/ethylene-co-methyl acrylate (164) PP/P12 (304) PP/TPU (228) PP/SBS (74) PP/SAN (184) PP/EVA (177, 189, 191, 193, 294) PP/SEBS (74, 112, 218) PP/SEP (105, 218) PP/SEPS (105, 218) PP/PS (38, 39, 117, 155, 159, 227, 212, 229, 299, 310) PP/ethylene-co-octene (59, 87) PP/ethylene-co-butene-1 (59) PP/LCP (225, 232) Natural rubber/polystyrene (290) Natural rubber/reclaimed rubber (252) Epoxidized natural rubber/natural rubber (253)

EPDM/SAN (53, 308) EPDM/PA6 (101, 279) EPDM/polyaniline (122) EPDM/TPU (95) EPDM/PP/PA6 (262) EPDM/PA12,10 (262) EPDM/PP6,10 (266) EPDM/natural rubber (44, 65, 240, 250, 269) EPR/polydimethylsiloxane (70) EPR/PA6 (57, 60)

12

Polyolefin Blends

Table 1.4 Summary of Number of Electronic articles on Polyolefin Blends Published Between 2000 and 2005 in English Language-based Polymer and Polymer-related Journals in Comparison to the Number of Patents. Journal articles (25–310) Year Number of Articles

2000 36

2001 31

2002 42

2003 42

2004 68

2005 66

Total 285

2004 322

2005 365

Total 2349

Patents Year Number of Patents

2000 517

2001 444

2002 406

2003 295

modeling and simulation. Regarding formulation design and processing, some of the issues of interest include i. Use of block, random, and graft copolymers in compatibilization. ii. Effects of molecular structure, weight, and additives. iii. Phase behavior, miscibility, and compatibility issues. iv. Batch and continuous mixing/compounding, extrusion, and molding into films, fibers, and other articles. v. Physical blending versus reactive blending. Regarding miscibility, structural, molecular, and property characterization, some of the issues of interest include i. Morphological characterization using techniques such as scanning electron microscopy (SEM), transmission electron microscopy (TEM), atomic force microscopy (AFM), and polarized light microscopy (PLM). ii. Structural characterization using radiation scattering and diffraction techniques such as X-ray scattering (XRS), X-ray diffraction (XRD), and electron diffraction (ED). iii. Structural characterization using spectroscopic techniques such as nuclear magnetic resonance (NMR), Fourier transform infrared (FTIR), and temperature rising elution fractionation (TREF). iv. Melt rheology. v. Phase separation dynamics. vi. Thermal transitions and thermal stability. vii. Isothermal and nonisothermal crystallization behavior under quiescent and nonquiescent conditions. viii. Crystallinity and crystal structure determination by various methods such as density, differential scanning calorimetry (DSC), XRD, and ED. ix. Tacticity, copolymer sequence distribution, and comonomer composition.

Chapter 1 Overview of Polyolefin Blends

13

Regarding end-use properties and performance, some of the issues of interest include i. Mechanical (static and dynamic) behavior under tensile, shear, or compressive, and impact mode. ii. Plastic deformation. iii. Thermomechanical and thermal stability behavior. iv. Adhesion, interfacial, and interphase behavior. v. Fire resistance/flammability behavior. vi. Barrier and transport properties. vii. Optical properties (gloss, haze, and transparency). viii. Surface/tribological properties. ix. Electrical/dielectric properties. x. Aging effects (time–temperature-dependent behavior). Modeling and simulation involves theoretical analysis of processing, characterization, and performance behavior using phenomenological, atomistic, molecular dynamics, and Monte Carlo methods, among others, and comparisons with experimental results. Polyolefin blends are generally immiscible due to hydrophobic or nonpolar nature of polyolefins. The immiscibility leads to phase separation, which is responsible for the poor mechanical properties of these blends. Immiscibility of polyolefin/ nonpolyolefin blends can be mitigated by adding a proper compatibilizer. The function of a compatibilizer is to reduce interface tension (strengthen the interface between the phases) and thus improve mechanical properties of the stabilized blend (reduce the size and morphology of the phase-separated phases). The compatibilizer strengthens the interface by broadening it from a sharp change in composition and properties to a broader gradual transition interface. Thus, the chief concern in compatibilization of polyolefin/nonpolyolefin blends is phase morphology. One way to achieve compatibilization involves physical processes such as shear mixing and thermal history, which modify domain size and shape. The second way is the use of physical additives to increase attraction between molecules and phases. The third method is reactive processing, which is used to change the chemical structure of one or more of the components in the blend and thus increase their attraction to each other. Table 1.5 contains a list of compatibilizers used in the formulation of polyolefin blends. As can be seen from Table 1.5, most of the compatibilizers used in the formulation of polyolefin blends contain compounds such as maleic anhydride, acrylic and methacrylic acid, glycidyl methacrylate, and diblock and triblock copolymers involving styrene, ethylene, and butadiene.

1.5 TRENDS IN POLYOLEFIN BLENDS The first trend is the growing use of in-reactor blending technique to produce new elastomeric or toughened polyolefins for demanding applications traditionally reserved

14

Polyolefin Blends

Table 1.5

Compatibilizers Used in the Formulation of Polyolefin Blends.

Compatibilizer

Polyolefin blend

PP-g-maleic anhydride Ethylene-co-acrylic acid Ethylene-co-glycidyl methacrylate Styrene-b-ethylene-co-propylene-g-maleic anhydride PE-g-maleic anhydride Styrene-co-ethylene-co-butadiene-costyrene-g-maleic anhydride Sodium-neutralized ethyleneco-methacrylic acid PE-g-glycidyl methacrylate Styrene-co-ethylene-co-butadiene-costyrene-g-glycidyl methacrylate Styrene-co-ethylene-co-butadiene-co-styrene Styrene-co-ethylene-co-butadiene-costyrene-g-maleic anhydride PE-g-maleic anhydride EPR-g-maleic anhydride PE-g-maleic anhydride PP-g-maleic anhydride Styrene-co-ethylene-co-butadiene-co-styrene Styrene-co-butadiene Styrene-co-ethylene-co-butadiene-co-styrene Styrene-co-butadiene-co-styrene Liquid polybutadiene þ diakyl peroxides Chlorinated PE Ethylene-co-acrylic acid Bismaleimide-g-PE PE-g-maleic anhydride PE-g-epoxidized natural rubber PE-g-diethyl maleate Sodium neutralized ethylene-co-acrylic acid PE-g-glycidyl methacrylate PE-g-maleic anhydride Ethylene-co-glycidyl methacrylate Styrene-co-ethylene-co-butadiene-co-styrene Styrene-co-ethylene-co-butadiene-co-styrene PP-g-3-isopropenyl-a,a-dimethylbenzene isocyanate PP-g-succinic anhydride and PP-g-fluorescein PP-g-maleic anhydride Ethylene-co-butyl acrylate-g-maleic anhydride Styrene-co-ethylene-co-butadiene-co-styreneg-glycidyl methacrylate

PE/PA6 PE/PA6 PE/PA6 PE/PA6

(27, 88) (27, 92, 140, 204, 215, 224) (27, 144) (130)

PE/PA6 (130) PE/PA6 (130, 168) PE/PA6 (92) PE/PA6 (144) PE/PA6 (168) PE/PP/PA6,6 (71) PE/PP/PA6,6 (71) PE/PA12 (142) PE/PP (47) PE/PP (71, 239) PE/PP (71, 239) PE/PS (47) PE/PS (48, 115) PE/PS (79) PE/PS (145) PE/PS (223) PE/PVC (52) PE/hydrolyzed collagen (98) PE/hydrolyzed collagen (98) PE/acrylonitrile-co-butadiene (100) PE/scrap rubber powder (165) PE/PET (166) PE/LCP (188) PE/PET (220) PE/ethylene-co-acrylic acid (281) PE/plasticized tapioca starch (289) PP/PA6,6 (28) PP/PA6 (28) PP/PA6 (54) PP/PA6 (111) PP/PA6 (58, 170, 178, 262, 283, 286, 289) PP/PA6 (170) PP/PA6 (178)

Chapter 1 Overview of Polyolefin Blends Table 1.5 (Continued) Compatibilizer

Polyolefin blend

PP-g-bismaleimide PP-g-acrylic acid EPDM-g-maleic anhydride Ethylene-co-octene-g-maleic anhydride PP-g-maleic anhydride Styrene/AlCl3-catalyzed friedel–crafts alkylation PP-g-PS Styrene-co-ethylene-co-butadiene-co-styrene Styrene-co-ethylene-co-butadiene-costyrene þ ionomer resin Ionomer Zn2þ Styrene-co-ethylene-co-propylene-g-glycidyl methacrylate Styrene-co-ethylene-co-butadiene-costyrene-g-glycidyl methacrylate PP-g-monomethyl itaconate PP-g-dimethyl itaconate PP-g-maleic anhydride PP-g-acrylic acid PP-g-monomethyl itaconate PP-g-primary amine PP-g-secondary amine N,N-m-Phenylene bismaleimide N,N-m-Phenylene bismaleimide Polyoctenamer Polystyrene-modified natural rubber PP-g-maleic anhydride Hydroxylated EVA Ethylene-co-butyl acrylate-co-glycidyl methacrylate EPDM-g-maleic anhydride EPDM-g-styrene-co-acrylonitrile Ethylene-co-octene-g-maleic anhydride Mercapto-modified EVA EPDM-g-maleic anhydride EPDM-g-maleic anhydride EPDM-g-vinyltriethoxysilane EPDM-g-vinyloxyaminosilane EPR-g-maleic anhydride EPR-g-MAH þ epoxy-modified lignosulfonate Styrene-co-isoprene

PP/PA6 (262) PP/PA6 (262) PP/PA6 (279) PP/PA6 (286) PP/epoxy (62, 120) PP/PS (21, 117) PP/PS (155, 299) PP/PS (299) PP/PS/PA6 (205) PP/EVOH (135) PP/PET (137) PP/PET (137) PP/EPDM (139) PP/EPDM (139) PP/EPDM/PA6 (262) PP/EPDM/PA6 (262) PP/EPR (163) PP/TPU (228) PP/TPU (228) PP/natural rubber/PE (238) PP/EPDM/natural rubber (244–245) PP/PE/natural rubber (254) PP/natural rubber (288) PP/EVA (294) PP/EVA (294) PTT/PE (37) EPDM/natural rubber (44, 240, 250) EPDM/SAN (53) EPDM/TPU (91) EPDM/NBR (234) EPDM/PA6,10 (266) EPDM/PA12,10 (266) EPDM/PE (275) EPDM/PE (296) EPR/PA6 (57, 60) EPR/lignin (172) Natural rubber/PS (290)

15

16

Polyolefin Blends

for more expensive engineering thermoplastics. There is also a growing interest in the use of elastomer-rich thermoplastic olefin blends that contain 60–75% rubber and 40– 25% polypropylene for automotive interior applications because they provide the required amount of lower and upper service temperature performance. There is also growing use of beta nucleator functional additive in polyolefin blends, for example, PP/ EPR blend. A beta nucleating agent leads to blend with improved impact and ductility. The second growing trend is the impact modification of polyolefin blends using styrenic block copolymers, which are known to be clear, strong, have low glass transition, compatible with PP, form interpenetrating polymer networks, and very efficient in contrast to maleic anhydride-grafted polyolefins. The third trend is the growing interest in functionalization of polyolefin blends in their melt by means of reactive extrusion. Particular attention has been paid to blended systems PP/PE, PP/EPR, PE/ethylene–octene copolymer (EOC), PP/EOC, PE/PS, and PP/PS functionalized in melt by reactive extrusion. The major field for application of functionalized polyolefin blends is compatibilization of blends of condensation polymers, where they can be used in place of homopolyolefins. The fourth trend is spurred by environmental sustainability concerns and the need for increased recyclability and reuse of polyolefin blends. In this regard, there is increasing replacement of PVC by polyolefin–polyolefin blends. There is also an increase in recyclability of EPDM rubber vulcanizates since EPDM is the fastest growing elastomer among synthetic rubber and the most used of nontire rubbers. Also, cryogenically ground rubber tires are being used as fillers for polyolefin blends such as LLDPE/HDPE. The fifth trend is the use of beta nucleator as opposed to alpha nucleator in PP containing blends such as PP/EPR. The use of beta nucleating agent results in blend with improved impact and ductility The sixth trend is the growing interest in foams, nonwoven, and elastic materials based on polyolefin blends The seventh trend is the increasing use of novel processing methods. For example, there is growing use of supercritical fluids (e.g., supercritical carbon dioxide and nitrogen gases) to foam polyolefin blends for density reduction. There is use of ultrasound to, for example, devulcanize cross-linked rubber. There is use of solid-state shear mechanical processing to break the polyolefin blend material into submicron particles to make environment friendly (water-based) polyolefin dispersions. There is use of electrospinning technique to make polyolefin fibers and in particular nanofibers. In conclusion, the importance of polyolefin blends has increased dramatically in the last couple of decades and this is sure to continue as new polyolefins are made and as new applications are sought for these materials.

NOMENCLATURE AFM BR

Atomic force microscopy Butyl rubber


DSC ED EOC EPDM EPR EVA EVOH FTIR HDPE LCP LDPE LLDPE NBR NMR NR PA PA12 PA12,10 PA6 PA6,6 PA6,10 PB PBT PC PCL PE PEN PET PIB PLM PMMA PMP PP iPP sPP aPP POE PPE PPO PPS PS PTT PVC SAN SB

Differential scanning calorimetry Electron diffraction Ethylene octene copolymer Ethylene propylene diene monomer Ethylene propylene rubber Ethylene-co-vinyl acetate Ethylene-co-vinyl alcohol Fourier transform infrared High density polyethylene Liquid crystal polymer Low density polyethylene Linear low density polyethylene Nitrile butadiene rubber Nuclear magnetic resonance Natural rubber Polyamide Polyamide 12 Polyamide 12,10 Polyamide 6 Polyamide 6,6 Polyamide 6,10 Polybutene-1 Polybutylene terephthalate Polycarbonate Polycaprolactone Polyethylene Polyethylene naphthalate Polyethylene terephthalate Polyisobutylene Polarized light microscopy Polymethylmethacrylate Polymethylpentene-1 Polypropylene Isotactic polypropylene Syndiotactic polypropylene Atactic polypropylene Polyolefin elastomer Poly(2,6-dimethyl-1,4-phenylene ether) Polyphenylene oxide Polyphenylene sulfide Polystyrene Polytrimethylene terephthalate Polyvinyl chloride Styrene-co-acrylonitrile Styrene-co-butadiene

17

18

Polyolefin Blends

SBR SBS SEM SEBS SEP SEPS TEM TPE TPO TPU TPV TREF UHMWPE ULDPE ULMW-LLDPE XRD XRS

Styrene butadiene rubber Styrene-co-butadiene-co-styrene Scanning electron microcopy Styrene-co-ethylene-co-butadiene-co-styrene Styrene-co-ethylene-co-propylene Styrene-co-ethylene-co-propylene-co-styrene Transmission electron microscopy Thermoplastic elastomer Thermoplastic olefin Thermoplastic polyurethane Thermoplastic vulcanizate Temperature rising elution fractionation Ultrahigh molecular weight polyethylene Ultralow density polyethylene Low molecular weight linear low density polyethylenes X-ray diffraction X-ray scattering

REFERENCES 1. M. Gahleitner, Prog. Polym. Sci., 26, 895 (2001). 2. J. L. White, Polyolefins, Hanser, Munich, 2005. 3. N. Pasquini (ed.), Polypropylene Handbook, Hanser, Munich, 2005. 4. K. Harutun (ed.), Handbook of Polypropylene and Polypropylene Composites, Marcel Dekker, New York, 2003. 5. C. Vasile (ed.), Handbook of Polyolefins, Marcel Dekker, New York, 2000. 6. A. J. Peacock (ed.), Handbook of Polyethylene: Structures, Properties, and Applications, Marcel Dekker, New York, 2000. 7. O. Olagoke, Handbook of Thermoplastics, Marcel Dekker, New York, 1997. 8. E. Moore (ed.), Polypropylene Handbook, Hanser, Munich, 1996. 9. J. Karger-Kocsis (ed.), Polypropylene: Structure, Blends, and Composites, Chapman and Hall, 1995. 10. L. A. Utracki (ed.), Polymer Blends Handbook, Vols. 1 and 2, Springer, New York, 2003. 11. C. Vasile and A. K. Kulshreshtha (eds.), Handbook of Polymer Blends and Composites, Rapra Technology Limited, 2003. 12. L. A. Utracki (ed.), Polymer Blends Handbook, Vols. 1 and 2, Kluwer Academic Publishers, Dordrecht, 2002. 13. D. R. Paul and C. B. Bucknall (eds.), Polymer Blends: Formulations and Performance, Vols. 1 and 2, Wiley, New York, 2000. 14. G. O. Shonaike and G. P Simon (eds.), Polymer Blends and Alloys, Marcel Dekker, New York, 1999. 15. T. Araki, Q. Tran-Cong, and M. Shibayama (eds.), Structure and Properties of Multi-Phase Polymeric Materials, Marcel Dekker, New York, 1998. 16. I. S. Miles and S. Rostani (eds.), Multi-Component Polymer Systems, Longman Scientific, Essex, 1992. 17. L. A. Utracki (ed.), Polymer Blends and Alloys: Thermodynamics and Rheology, Hanser, Munich, 1989.


19

18. D. R. Paul and S. Newman (eds.), Polymer Blends, Academic Press, New York, 1978. 19. J. Manson and L. H. Sperling (eds.), Polymer Blends and Composites, Plenum Press, New York, 1976. 20. G. Holden and N. R. Legge (eds.), Thermoplastic Elastomers, Hanser, New York, 1996. 21. W. K. Fisher, US Patent 3,835,201 (1974). 22. S. Daniel and R. S. Porter, Polymer, 19, 448 (1978). 23. A. Y. Coran and R. P Patel, Rubber Chem. Technol., 53, 141 (1980). 24. S. Abdou-Sabet and R. P. Patel, Rubber Chem. Technol., 64, 769 (1991). 25. N. Chapleau, B. D. Favis, and P.J. Carreau, Polymer, 41, 6695 (2000). 26. Y. T. Shieh, M. S. Lee, and S. A. Chen, Polymer, 42, 4439 (2001). 27. V. Chiono, S. Filippi, H. Yordanov, L. Minkova, and P. Magagnini, Polymer, 44, 2423 (2003). 28. S.-L. Bai, G.-T. Wang, J.-M. Hiver, and C. G’Sell, Polymer, 45, 3063 (2004). 29. B. Lin and U. Sundararaj, Polymer, 45, 7605 (2004). 30. K. Shimizu, H. Wang, Z. Wang, G. Matsuba, H. Kim, and C. C. Han, Polymer, 45, 7061 (2004). 31. C. G’Sell, S.-L. Bai, and J.-M. Hiver, Polymer, 45, 5785 (2004). 32. J. I. Uriguen, L. Bremer, V. Mathot, and G. Groeninckx, Polymer, 45, 5961 (2004). 33. B. Na, K. Wang, Q. Zhang, R. Du, and Q. Fu, Polymer, 46, 3190 (2005). 34. D. Dikovsky, G. Marom, C. A. Avila-Orta, R. H. Somani, and B. S. Hsiao, Polymer, 46, 3096 (2005). 35. K.-H. Nitta, Y.-W. Shin, H. Hashiguchi, S. Tanimoto, and M. Terano, Polymer, 46, 965 (2005). 36. B. Na, Q. Zhang, K. Wang, L. Li, and Q. Fu, Polymer, 46, 819 (2005). 37. S. H. Jafari, A. Yavari, A. Asadinezhad, H. A. Khonakdar, and F. Bohme, Polymer, 46, 5082 (2005). 38. Q. Xing, M. Zhu, Y. Wang, Y. Chen, Y. Zhang, J. Pionteck, and H. J. Adler, Polymer, 46, 5406 (2005). 39. M. F. Diaz, S. E. Barbosa, and N. J. Capiati, Polymer, 46, 6096 (2005). 40. X. Zhang, Y. Zhao, Z. Wang, C. Zheng, X. Dong, Z. Su, P. Sun, D. Wang, C. C. Han, and D. Xu, Polymer, 46, 5956 (2005). 41. A. V. Machado and M. van Duin, Polymer, 46, 6575 (2005). 42. W. G. F. Sengers, M. Wu¨bbenhorst, S. J. Picken, and A. D. Gotsis, Polymer, 46, 6391 (2005). 43. S.-L. Bai, C. G’Sell, J.-M. Hiver, and C. Mathieu, Polymer, 46, 6437 (2005). 44. S. H. El-Sabbagh, J. Appl. Polym. Sci., 90, 1 (2003). 45. G. Takidis, D. N. Bikiaris, G. Z. Papageorgiou, D. S. Achilias, and I. Sideridou, J. Appl. Polym. Sci., 90, 841 (2003). 46. N. Phochalam, A. Tabtiang, and R. A. Venables, J. Appl. Polym. Sci., 90, 1655 (2003). 47. T. Kallel, V. Massardier-Nageotte, Mo. Jaziri, J.-F. Ge´rard, and B. Elleuch, J. Appl. Polym. Sci., 90, 2475 (2003). 48. I. Fortelny´, J. Mikesôva´, J. Hromadkova´, V. Hasôva´, and Z. Hora´k, J. Appl. Polym. Sci., 90, 2303 (2003). 49. T. Nedkov and F. Lednicky´, J. Appl. Polym. Sci., 90, 3087 (2003). 50. Y. Chen, Y. Cao, and H. Li, J. Appl. Polym. Sci., 90, 3519 (2003). 51. T. Wu, Y. Li, D.-L. Zhang, S.-Q. Liao, and H.-M. Tan, J. Appl. Polym. Sci., 91, 905 (2004). 52. Z. Fang, G. Ma, C. Liu, and C. Xu, J. Appl. Polym. Sci., 91, 763 (2004). 53. X. Qu, S. Shang, G. Liu, S. Zhang, Y. Zhang, and L. Zhang, J. Appl. Polym. Sci., 91, 1685 (2004). 54. G.-H. Hu, H. Cartier, L.-F. Feng, and B.-G. Li, J. Appl. Polym. Sci., 91, 1498 (2004). 55. M. Canetti, A. De Chirico, and G. Audisio, J. Appl. Polym. Sci., 91, 1435 (2004). 56. H. A. Khonakdar, J. Morshedian, H. Eslami, and F. Shokrollahi, J. Appl. Polym. Sci., 91, 1389 (2004).

20

Polyolefin Blends

57. S. C. George, K. N. Ninan, G. Geuskens, and S. Thomas, J. Appl. Polym. Sci., 91, 3756 (2004). 58. D. Shi, J. Yin, Z. Ke, Y. Gao, and R. Ky Li, J. Appl. Polym. Sci., 91, 3742 (2004). 59. J. Lee, C. W. Macosko, and F. S. Bates, J. Appl. Polym. Sci., 91, 3642 (2004). 60. Z. Oommen, S. R. Zachariah, S. Thomas, G. Groeninckx, P. Moldenaers, and J. Mewis, J. Appl. Polym. Sci., 92, 252 (2004). 61. B. C. Poon, S. P. Chum, A. Hiltner, and E. Baer, J. Appl. Polym. Sci., 92, 109 (2004). 62. X. Jiang. H. Huang, Y. Zhang, and Y. Zhang, J. Appl. Polym. Sci., 92, 1437 (2004). 63. M. Pires, R. S. Mauler, and S. A. Liberman, J. Appl. Polym. Sci., 92, 2155 (2004). 64. S. Varghese, R. Alex, and B. Kuriakose, J. Appl. Polym. Sci., 92, 2063 (2004). 65. S. Kiatkamjornwong and K. Pairpisit, J. Appl. Polym. Sci., 92, 3401 (2004). 66. L. Tang, B. Qu, and X. Shen, J. Appl. Polym. Sci., 92, 3371 (2004). 67. G. Chen, S. Guo, and Y. Li, J. Appl. Polym. Sci., 92, 3153 (2004). 68. T. Muraleedharan Nair, M. G. Kumaran, and G. Unnikrishnan, J. Appl. Polym. Sci., 93, 2606 (2004). 69. G. Liu, Y. Chen, and H. Li, J. Appl. Polym. Sci., 94, 977 (2004). 70. M. Carlberg, D. Colombini, and F. H. J. Maurer, J. Appl. Polym. Sci., 94, 2240 (2004). 71. R. Krache, D. Benachour, and P. Po¨tschke, J. Appl. Polym. Sci., 94, 1976 (2004). 72. H. Wu, S. Guo, G. Chen, and H. Shi, J. Appl. Polym. Sci., 94, 1917 (2004). 73. A. Colbeaux, F. Fenouillot, J.-F. Ge´rard, M. Taha, and H. Wautier, J. Appl. Polym. Sci., 95, 320 (2005). 74. F. O. M. S. Abreu, M. M. C. Forte, and S. A. Liberman, J. Appl. Polym. Sci., 95, 263 (2005). 75. F. Chen, R. A. Shanks, and G. Amarasinghe, J. Appl. Polym. Sci., 95, 1549 (2005). 76. A. Jalali-Arani, A. A. Katbab, and H. Nazockdast, J. Appl. Polym. Sci., 96, 155 (2005). 77. B. Patham, P. Papworth, K. Jayaraman, C. Shu, and M. D. Wolkowicz, J. Appl. Polym. Sci., 96, 423 (2005). 78. T. Wu, Y. Li, Q. Wu, G. Wu, and H.-M. Tan, J. Appl. Polym. Sci., 96, 261 (2005). 79. G. J. Nam, K. Y. Kim, and J. W. Lee, J. Appl. Polym. Sci., 96, 905 (2005). 80. S. Luyt and M. J. Hato, J. Appl. Polym. Sci., 96, 1748 (2005). 81. K. L. K. Lim, Z. A. Mohd. Ishak, U. S. Ishiaku, A. M. Y. Fuad, A. H. Yusof, T. Czigany, B. Pukanszky, and D. S. Ogunniyi, J. Appl. Polym. Sci., 97, 413 (2005). 82. J. Sook. Hong, J. L. Kim, K. H. Ahn, and S. J. Lee, J. Appl. Polym. Sci., 97, 1702 (2005). 83. S. Bertin and J.-J Robin, Eur. Polym. J., 38, 2255 (2002). 84. D. Gheysari, A. Behjat, and M. Haji-Saeid, Eur. Polym. J., 37, 295 (2001). 85. B. Raj, V. Annadurai, R. Somashekar, M. Raj, and S. Siddaramaiah, Eur. Polym. J., 37, 943 (2001). 86. J. Kim, J. H. Kim, T. K. Shin, H. J. Choi, and M. S. Jhon, Eur. Polym. J., 37, 2131 (2001). 87. X. Zhang, F. Xie, Z. Pen, Y. Zhang, Y. Zhang, and W. Zhou, Eur. Polym. J., 38, 1 (2002). 88. B. Jurkowski, Y. A. Olkhov, K. Kelar, and O. M. Olkhova, Eur. Polym. J., 38, 1229 (2002). 89. S. Glowinkowski, M. Makrocka-Rydzyk, S. Wanke, and S. Jurga, Eur. Polym. J., 38, 961 (2002). 90. P. Matzinos, V. Tserki, C. Gianikouris, E. Pavlidou, and C. Panayiotou, Eur. Polym. J., 38, 1713 (2002). 91. D.-H. Chen, L. Hong, X.-W. Nie, X.-L. Wang, and X.-Z. Tang, Eur. Polym. J., 39, 871 (2003). 92. H. Yordanov and L. Minkova, Eur. Polym. J., 39, 951 (2003). 93. S. Jose, A. S. Aprem, B. Francis, M. C. Chandy, P. Werner, V. Alstaedt, and S. Thomas, Eur. Polym. J., 40, 2105 (2004). 94. A. Krumme, A. Lehtinen, and A. Viikna, Eur. Polym. J., 40, 371 (2004). 95. X. Wang and X. Luo, Eur. Polym. J., 40, 2391 (2004). 96. A. Lahor, M. Nithitanakul, and B. P. Grady, Eur. Polym. J., 40, 2409 (2004).


21

97. K. A. Moly, H. J. Radusch, R. Androsh, S. S. Bhagawan, and S. Thomas, Eur. Polym. J., 41, 1410 (2005). 98. M. C. Dasca˘lu, C. Vasile, C. Silvestre, and M. Pascu, Eur. Polym. J., 41, 1391 (2005). 99. A. C. Quental and M. I. Felisberti, Eur. Polym. J., 41, 894 (2005). 100. J. George, K. Ramamurthy, K. T. Varughese, and S. Thomas, J. Polym. Sci. B Polym. Phys., 38, 1104 (2000). 101. S. C. George, G. Groeninckx, K. N. Ninan, and S. Thomas, J. Polym. Sci. B Polym. Phys., 38, 2136 (2000). 102. M.-H. Kim, J. D. Londono, and A. Habenschuss, J. Polym. Sci. B Polym. Phys., 38, 2480 (2000). 103. Z.-G. Wang, R. A. Phillips, and B. S. Hsiao, J. Polym. Sci. B Polym. Phys., 38, 2580 (2000). 104. K. Lee, H. Kwag, B. Kim, G.-J. Kim, K.-H. Kim, and S. Choe, J. Polym. Sci. B Polym. Phys., 39, 218 (2000). 105. G. Radonjicˇ and Ivan sˇmit, J. Polym. Sci. B Polym. Phys., 39, 566 (2000). 106. H. Bashir, A. Linares, and J. L. Acosta, J. Polym. Sci. B Polym. Phys., 39, 1017 (2001). 107. C. Zhang, T. Mori, T. Mizutani, M. Ishioka, and Y. Cheng, J. Polym. Sci. B Polym. Phys., 39, 1741 (2001). 108. J. Lu and H.-J. Sue, J. Polym. Sci. B Polym. Phys., 40, 507 (2002). 109. Y.-T. Shieh, M.-S. Lee, and S.-A. Chen, J. Polym. Sci. B Polym. Phys., 40, 638 (2002). 110. J. Chatterjee and R. G. Alamo, J. Polym. Sci. B Polym. Phys., 40, 878 (2002). 111. C. Marco, E. P. Collar, S. Areso, and J. Ma Garcıá-Martıńez, J. Polym. Sci. B Polym. Phys., 40, 1307 (2002). 112. S. C. Tjong, S.-A. Xu, and Y.-W. Mai, J. Polym. Sci. B Polym. Phys., 40, 1881 (2002). 113. Y. Wang, Q. Fu, Q. Li, G. Zhang, K. Shen, and Y.-Z. Wang, J. Polym. Sci. B Polym. Phys., 40, 2086 (2002). 114. M. Moffitt, Y. Rharbi, J.-D. Tong, J. P. S. Farhina, H. Li, M. A. Winnik, and H. Zahalka, J. Polym. Sci. B Polym. Phys., 41, 637 (2002). 115. I. Fortelny´, D. Hlavata´, J. Mikesˇova´, D. Micha´lkova´, L. Potrokova´, and I. sˇloufova´, J. Polym. Sci. B Polym. Phys., 41, 609 (2002). 116. J. Finlay, M. J. Hill, P. J. Barham, K. Byrne, and A. Woogara, J. Polym. Sci. B Polym. Phys., 41, 1384 (2003). 117. M. F. Dıáz, S. E. Barbosa, and N. J. Capiati, J. Polym. Sci. B Polym. Phys., 42, 452 (2003). 118. A. Opdahl, R. A. Phillips, and G. A. Somorjai, J. Polym. Sci. B Polym. Phys., 42, 421 (2003). 119. K. Lee, S. E. Shim, B. H. Lee, S. U. Hong, and S. Choe, J. Polym. Sci. B Polym. Phys., 42, 1114 (2003). 120. X. Jiang, Y. Zhang, and Y. Zhang, J. Polym. Sci. B Polym. Phys., 42, 1181 (2003). 121. W. Jiang, D. Yu, L. An, and B. Jiang, J. Polym. Sci. B Polym. Phys., 42, 1433 (2003). 122. S. C. Domenech, J. H. Bortoluzzi, V. Soldi, C. V. Franco, W. Gronski, and H.-J. Radusch, J. Polym. Sci. B Polym. Phys., 42, 1767 (2003). 123. I. Banik, P. J. Carreau, and H. P. Schreiber, J. Polym. Sci. B Polym. Phys., 42, 2545 (2004). 124. C.-I Huang, C.-P. Chang, K. Shimizu, and C. C. Han, J. Polym. Sci. B Polym. Phys., 42, 2995 (2004). 125. C. Silvestre, S. Cimmino, and J. S. Lin, J. Polym. Sci. B Polym. Phys., 42, 3368 (2004). 126. E. Puukilainen and T. A. Pakkanen, J. Polym. Sci. B Polym. Phys., 43, 2252 (2005). 127. Z. Chen, T. Li, X. Liu, and R. Lu¨, J. Polym. Sci. B Polym. Phys., 43, 2514 (2005). 128. H. Wu, S. Guo, and Z. Li, J. Polym. Sci. B Polym. Phys., 43, 3030 (2005). 129. R. Pe´rez, E. Rojo, M. Fernańdez, V. Leal, P. Lafuente, and A. Santamarıá, Polymer, 46, 8045 (2005). 130. S. Filippi, L. Minkova, N. Dintcheva, P. Narducci, and P. Magagnini, Polymer, 46, 8054 (2005).

22

Polyolefin Blends

131. C.-T. Lo, S. Seifert, P. Thiyagarajan, and B. Narasimhan, Macromol. Rapid Commun., 26, 533 (2005). 132. H. Hu, C. Chong, A. He, C. Zhang, G. Fan, J.-Y. Dong, and C. Han, Macromol. Rapid Commun., 26, 973 (2005). 133. X. Zhang, Z. Wang, M. Muthukumar, and C. C. Han, Macromol. Rapid Commun., 26, 1285 (2005). 134. J. Luettmer-Strathmann, J. Chem. Phys., 123, 014910–12 (2005). 135. M. Montoya, M. J. Abad, L. Barral Losada, and C. Bernal, J. Appl. Polym. Sci., 98, 1271 (2005). 136. S. Datta, K. Naskar, J. Jelenic, and J. W. M. Noordermeer, J. Appl. Polym. Sci., 98, 1393 (2005). 137. M. Pracella, D. Chionna, A. Pawlak, and A. Galeski, J. Appl. Polym. Sci., 98, 2201 (2005). 138. D. Xu, J. Wang, Z. Liu, Y. Ke, and Y. Hu, Macromol. Chem. Phys., 202, 1817 (2001). 139. M. A. Lo´pez-Manchado, J. M. Kenny, R. Quijada, and M. Yazdani-Pedram, Macromol. Chem. Phys., 202, 1909 (2001). 140. S. Filippi, V. Chiono, G. Polacco, M. Paci, L. I. Minkova, and P. Magagnini, Macromol. Chem. Phys., 203, 1512 (2002). ´ . Prieto, J. M. Perenã, R. Benavente, M. L. Cerrada, and E. Pe´rez, Macromol. Chem. Phys., 203, 141. O 1844 (2002). 142. H. Sato, S. Sasao, K. Matsukawa, Y. Kita, H. Yamaguchi, H. W. Siesler, and Y. Ozaki, Macromol. Chem. Phys., 204, 1351 (2003). 143. A. J. Mu˝ller, M. L. Arnal, A. L. Spinelli, E. Can˜izales, C. C. Puig, H. Wang, and C. C. Han, Macromol. Chem. Phys., 204, 1497 (2003). 144. Q. Wei, D. Chionna, and M. Pracella, Macromol. Chem. Phys., 206, 777 (2005). 145. J. Li, C.-M. Chan, B. Gao, and J. Wu, Macromolecules, 33, 1022 (2000). 146. M. L. D. Lorenzo, S. Cimmino, and C. Silvestre, Macromolecules, 33, 3828 (2000). 147. I. G. Economou, Macromolecules, 33, 4954 (2000). 148. Y. A. Akpalu, A. Karim, S. K. Satija, and N. P. Balsara, Macromolecules, 34, 1720 (2001). 149. H.-J. Kim, K.-J. Lee, Y. Seo, S. Kwak, and S.-K. Koh, Macromolecules, 34, 2546 (2001). 150. H. Wang, K. Shimizu, E. K. Hobbie, Z.-G. Wang, J. C. Meredith, A. Karim, E. J. Amis, B. S. Hsiao, E. T. Hsieh, and C. Han, Macromolecules, 35, 1072 (2002). 151. S. Costeux, P. Wood-Adams, and D. Beigzadeh, Macromolecules, 35, 2514 (2002). 152. A. Opdahl, R. A. Phillips, and G. A. Somorjai, Macromolecules, 35, 4387 (2002). 153. I. A. Hussein, Macromolecules, 36, 2024 (2003). 154. Y. Ma, J. P. S. Farinha, M. A. Winnik, P. V. Yaneff, and R. A. Ryntz, Macromolecules, 37, 6544 (2004). 155. L. Caporaso, N. Iudici, and L. Oliva, Macromolecules, 38, 4894 (2005). 156. F. M. Capaldi, G. C. Rutledge, and M. C. Boyce, Macromolecules, 38, 6700 (2005). 157. J. E. Wolak and J. L. White, Macromolecules, 38, 10466 (2005). 158. L. Minkova, M. Velcheva, and P. Magagnini, Macromol. Mater. Eng., 280-281, 7 (2000). 159. Z. Funke, C. Schwinger, R. Adhikari, and J. Kressler, Macromol. Mater. Eng., 286, 744 (2001). 160. G. Radonjicˇ and N. Gubeljak, Macromol. Mater. Eng., 287, 122 (2002). 161. B. Borisova and J. Kressler, Macromol. Mater. Eng., 288, 509 (2003). 162. J. Kolarˇ´ık, A. Pegoretti, L. Fambri, and A. Penati, Macromol. Mater. Eng., 288, 629 (2003). 163. M. Yazdani-Pedram, R. Quijada, and M. A. Lo´pez-Manchado, Macromol. Mater. Eng., 288, 875 (2003). 164. A. Genovese and R. A. Shanks, Macromol. Mater. Eng., 289, 20 (2004). 165. B. Guo, Y. Cao, D. Jia, and Q. Qiu, Macromol. Mater. Eng., 289, 360 (2004). 166. M.-B. Coltelli, S. Savi, I. D. Maggiore, V. Liuzzo, M. Aglietto, and F. Ciardelli, Macromol. Mater. Eng., 289, 400 (2004).


23

167. Z.-M. Li, W. Yang, R. Huang, X.-P. Fang, and M.-B. Yang, Macromol. Mater. Eng., 289, 426 (2004). 168. S. Filippi, H. Yordanov, L. Minkova, G. Polacco, and M. Talarico, Macromol. Mater. Eng., 289, 512 (2004). 169. Z.-B. Chen, T.-S. Li, Y.-L. Yang, Y. Zhang, and S.-Q. Lai, Macromol. Mater. Eng., 289, 662 (2004). 170. U. Hippi, M. Korhonen, S. Paavola, and J. Seppa¨la¨, Macromol. Mater. Eng., 289, 714 (2004). 171. C. Li, Q. Kong, J. Zhao, D. Zhao, Q. Fan, and Y. Xia, Macromol. Mater. Eng., 289, 833 (2004). 172. G. Cazacu, M. Mihaies, M. C. Pascu, L. Profire, A. L. Kowarskik, and C. Vasile, Macromol. Mater. Eng., 289, 880 (2004). 173. Z.-M. Li, C.-G. Huang, W. Yang, M.-B. Yang, and R. Huang, Macromol. Mater. Eng., 289, 1004 (2004). 174. H. Kwag, D. Rana, K. Cho, J. Rhee, T. Woo, B. H. Lee, and S. Choe, Polym. Eng. Sci., 40, 1672 (2000). 175. R. Zacur, G. Goizueta, and N. Capiati, Polym. Eng. Sci., 40, 1921 (2000). 176. G. Flodberg, A. Hellman, M. S. Hedenqvist, E. R. Sadiku, and U. W. Gedde, Polym. Eng. Sci., 40, 1969 (2000). 177. E. Ramı´rez-Vargas, D. Navarro-Rodrı´guez, F. J. Medellin-Rodrı´guez, B. M. Huerta-Martıńez, and J. S. Lin, Polym. Eng. Sci., 40, 2241 (2000). 178. J. D. Tucker, S. Lee, and R. L. Einsporn, Polym. Eng. Sci., 40, 2577 (2000). 179. H. Tang, B. Foran, and D. C. Martin, Polym. Eng. Sci., 41, 440 (2001). 180. M. A. Huneault, F. Mighri, G. H. Ko, and F. Watanabe, Polym. Eng. Sci., 41, 672 (2001). 181. I. A. Hussein and M. C. Williams, Polym. Eng. Sci., 41, 696 (2001). 182. K. Wang and C. Zhou, Polym. Eng. Sci., 41, 2249 (2001). 183. D. K. Setua, C. Soman, A. K. Bhowmick, and G. N. Mathur, Polym. Eng. Sci., 42, 10 (2002). 184. J. Kolarˇ´ık, L. Fambri, A Pegoretti, A. Penati, and P. Goberti, Polym. Eng. Sci., 42, 161 (2002). 185. A. Garcia-Rejon, A. Meddad, E. Turcott, and M. Carmel, Polym. Eng. Sci., 42, 346 (2002). 186. D. Roy, G. P. Simon, and M. Forsyth, Polym. Eng. Sci., 42, 781 (2002). 187. Y. C. Liang and A. I. Isayev, Polym. Eng. Sci., 42, 994 (2002). 188. Y. Son and R. A. Weiss, Polym. Eng. Sci., 42, 1322 (2002). 189. E. Ramı´rez-Vargas, F. J. Medelln-Rodrı´guez, D. Navarro-Rodrı´guez, C. A. Avila-Orta, S. G. Solı´sRosales, and J. S. Lin, Polym. Eng. Sci., 42, 1350 (2002). 190. J.-T. Yeh, and S.-S. Chang, Polym. Eng. Sci., 42, 1558 (2002). 191. F. Marguerat, P. J. Carreau, and A. Michel, Polym. Eng. Sci., 42, 1941 (2002). 192. M. L. Arnal, E. Canízales, and A. J. Mu¨ller, Polym. Eng. Sci., 42, 2048 (2002). 193. C. Joubert, P. Cassagnau, A. Michel, and L. Choplin, Polym. Eng. Sci., 42, 2222 (2002). 194. I. Sendijarevic, A. J. McHugh, J. A. Orlicki, and J. S. Moore, Polym. Eng. Sci., 42, 2393 (2002). 195. C. Grein, C. J. G. Plummer, Y. Germain, H.-H. Kausch, and P. Be´guelin, Polym. Eng. Sci., 43, 223 (2003). 196. N. Kukaleva, G. P. Simon, and E. Kosior, Polym. Eng. Sci., 43, 431 (2003). 197. G. Flodberg, M. S. Hedenqvist, and U. W. Gedde, Polym. Eng. Sci., 43, 1044 (2003). 198. H. Padilla-Lopez, M. O. Va´zquez, R. Gonza´lez-Nuń˜ez, and D. Rodrigue, Polym. Eng. Sci., 43, 1646 (2003). 199. W. Wang, Q. Wu, and B. Qu, Polym. Eng. Sci., 43, 1798 (2003). 200. G. Liu, M. Xiang, and H. Li, Polym. Eng. Sci., 44, 197 (2004). 201. N. Sombatsompop, K. Sungsanit, and C. Thongpin, Polym. Eng. Sci., 44, 487 (2004). 202. Y. Chen and H. Li, Polym. Eng. Sci., 44, 1509 (2004). 203. P. Rachtanapun, S. E. M. Selke, and L. M. Matuana, Polym. Eng. Sci., 44, 1551 (2004). 204. L. Canfora, S. Filippi, and F. P. La Mantia, Polym. Eng. Sci., 44, 1732 (2004).

24

Polyolefin Blends

205. M. A. Debolt and R. E. Robertson, Polym. Eng. Sci., 44, 1800 (2004). 206. M. H. Ha and B. Y. Kim, Polym. Eng. Sci., 44, 1858 (2004). 207. W. Feng and A. I. Isayev, Polym. Eng. Sci., 44, 2019 (2004). 208. Z.-M. Li, B.-H. Xie, R. Huang, X.-P. Fang, and M.-B. Yang, Polym. Eng. Sci., 44, 2165 (2004). 209. K. P. Tchomakov, B. D. Favis, M. A. Huneault, F. Champagne, and F. Tofan, Polym. Eng. Sci., 44, 749 (2004). 210. J.-T. Yeh, S.-S. Huang, and H.-Y. Chen, Polym. Eng. Sci., 45, 25 (2005). 211. Y. Fang, P. J. Carreau, P. G. Lafleur, and S. Ymmel, Polym. Eng. Sci., 45, 343 (2005). 212. V. Thirtha, R. Lehman, and T. Nosker, Polym. Eng. Sci., 45, 1187 (2005). 213. H.-S. Xu, Z.-M. Li, S. Y. Yang, J.-L. Pan, W. Yang, and M.-B. Yang, Polym. Eng. Sci., 45, 1231 (2005). 214. Y. Fang, P. J. Carreau, and P. G. Lafleur, Polym. Eng. Sci., 45, 1254 (2005). 215. F. P. La Mantia, L. Canfora, and N. Tzankova Dintcheva, Polym. Eng. Sci., 45, 1297 (2005). 216. S. S. Morye, Polym. Eng. Sci., 45, 1369 (2005). 217. S. S. Morye, Polym. Eng. Sci., 45, 1377 (2005). 218. Y. Matsuda, M. Hara, T. Mano, K. Okamoto, and M. Ishikawa, Polym. Eng. Sci., 45, 1630 (2005). 219. C. Lu, X. Yu, and S. Guo, Polym. Eng. Sci., 45, 1666 (2005). 220. F. Pazzagli and M. Pracella, Macromol. Symp., 149, 225 (2000). 221. J. Murıń, J. Uhrin, and I. Choda´k, Macromol. Symp., 170, 115 (2001). 222. B. L. Lopez, C. Gartner, and M. Hess, Macromol. Symp., 174, 277 (2001). 223. D. Hlavata´, Z. Krulisˇ, Z. Hora´k, F. Lednicky´, and J. Hroma´dkova´, Macromol. Symp., 176, 93 (2001). 224. M. R. Arroyo and M. L. Bell, Macromol. Symp., 170, 181 (2001). 225. S. Bualek-Limcharoen, S. Saengsuwan, T. Amornsakchai, and B. Wanno, Macromol. Symp., 170, 189 (2001). 226. F. P. la Mantia, R. Scaffaro, A. Valenza, A. Marchetti, and S. Filippi, Macromol. Symp., 198, 173 (2003). 227. J. Pointeck, P. Po¨tschke, N. Proske, H. Zhao, H. Malz, D. Beyerlein, U. Schulze, and B. Voit, Macromol. Symp., 198, 209 (2003). 228. Q.-W. Lu, C. W. Macosko, and J. Horrion, Macromol. Symp., 198, 221 (2003). 229. Y. Roiter, V. Samaryk, S. Varvarenko, N. Nosova, I. Tarnavchyk, J. Pointeck, P. Po¨tschke, and S. Voronov, Macromol. Symp., 210, 209 (2004). 230. T. Trongsatikul, D. Aht-Ong, and W. Chinsirikul, Macromol. Symp., 216, 265 (2004). 231. H. H. Kausch, Macromol. Symp., 225, 165 (2005). 232. G. Guerrica-Echevarrıá, J. I. Eguiazaba´l, and J. Naza´bal, J. Polym. Sci. B. Polym. Phys., 38, 1090 (2000). 233. F. Mighri, M. A. Huneault, A. Ajji, G. H. Ko, and F. Watanabe, J. Appl. Polym. Sci., 82, 2113 (2001). 234. M. G. Oliveira and B. G. Soares, J. Appl. Polym. Sci., 91, 1404 (2004). 235. J. Luettmer-Strathmann, J. Chem. Phys., 123, 014910 (2005). 236. D. Heine, D. T. Wu, J. G. Curro, and G. S. Grest, J. Chem. Phys., 118, 914 (2003). 237. R. Strapasson, S. C. Amico, M. F. R. Pereira, and T. H. D. Sydenstricker, Polym. Test., 24, 468 (2005). 238. A. Hassan, M. U. Wahit, and C. Y. Chee, Polym. Test., 22, 281 (2003). 239. C. Li, Y. Zhang, and Y. Zhang, Polym. Test., 22, 191 (2003). 240. S. H. El-Sabbagh, Polym. Test., 22, 93 (2003). 241. H. Ismail and Suryadiansyah, Polym. Test., 21, 389 (2002). 242. J.-Z. Liang, Polym. Test., 21, 69 (2002). 243. C. M. Tai, R. K. Y. Li, and C. N. Ng, Polym. Test., 19, 143 (2000). 244. Halimatuddahliana, H. Ismail, and H. Md. Akil, Polym. Plast. Technol. Eng., 44, 1429 (2005).


25

245. Halimatuddahliana, H. Ismail, and H. Md. Akil, Polym. Plast. Technol. Eng., 44, 1217 (2005). 246. Halimatuddahliana and H. Ismail, Polym. Plast. Technol. Eng., 43, 357 (2004). 247. H. Ismail and Suryadiansyah, Polym. Plast. Technol. Eng., 43, 319 (2004). 248. A. P. Gupta, U. K. Saroop, and M. Verma, Polym. Plast. Technol. Eng., 42, 357 (2003). 249. H. Ismail and Suryadiansyah, Polym. Plast. Technol. Eng., 41, 833 (2002). 250. S. H. Botros, Polym. Plast. Technol. Eng., 41, 341 (2002). 251. A. A. Ol’khov, A. L. Iordanskii, G. E. Zaikov, and L. S. Shibryaeva, Polym. Plast. Technol. Eng., 39, 783 (2000). 252. T. D. Sreeja and S. K. N. Kutty, Polym. Plast. Technol. Eng., 39, 501 (2000). 253. B. T. Poh and G. H. Khok, Polym. Plast. Technol. Eng., 39, 151 (2000). 254. A. Hassan, M. U. Wahit, and C. Y. Chee, Polym. Plast. Eng., 44, 1245 (2005). 255. N. M. Abdel Monem, Z. I. Ali, H. M. Said, H. A. Youssef, and H. H. Saleh, Polym. Plast. Technol. Eng., 44, 1025 (2005). 256. I. Krupa and A. S. Luyt, Polym. Degrad. Stabil., 70, 111 (2000). 257. G. Luo, T. Suto, S. Yasu, and K. Kato, Polym. Degrad. Stabil., 70, 97 (2000). 258. I. A. Hussein, K. Ho, S. K. Goyal, E. Karbashewski, and M. C. Williams, Polym. Degrad. Stabil., 68, 381 (2000). 259. J. C. M. Suarez, E. B. Mano, and R. A. Pereira, Polym. Degrad. Stabil., 69, 217 (2000). 260. E. Ramı´rez-Vargas, D. Navarro-Rodrı´guez, A. I. Blanqueto-Menchaca, B. M. Huerta-Martıńez, and M. Palacios-Mezta, Polym. Degrad. Stabil., 86, 301 (2004). 261. A. A. Basfar, K. M. Idriss Ali, and S. M. Mofti, Polym. Degrad. Stabil., 82, 229 (2003). 262. R. N. Darie, M. Brebu, C. Vasile, and M. Kozlowski, Polym. Degrad. Stabil., 80, 551 (2003). 263. W. Camacho and S. Karlsson, Polym. Degrad. Stabil., 78, 385 (2002). 264. R. A. Assink, M. Celina, K. T. Gillen, R. L. Clough, and T. M. Alam, Polym. Degrad. Stabil., 73, 355 (2001). 265. M. P. Anachkov, S. K. Rakovski, and R. V. Stefanova, Polym. Degrad. Stabil., 67, 355 (2000). 266. I. Vieira, V. L. S. Severgnini, D. J. Mazera, M. S. Soldi, E. A. Pinheiro, A. T. N. Pires, and V. Soldi, Polym. Degrad. Stabil., 74, 151 (2001). 267. T. Zaharescu, Polym. Degrad. Stabil., 73, 113 (2001). 268. R. Xie and B. Qu, Polym. Degrad. Stabil., 71, 375 (2001). 269. T. Zaharescu, V. Meltzer, and R. Vı˘lcu, Polym. Degrad. Stabil., 70, 341 (2000). 270. H. A. Khonakdar, S. H. Jafari, A. Yavari, A. Asadinezhad, and U. Wagenknecht, Polym. Bull., 54, 75 (2005). 271. T. Zaharescu and P. Budrugeac, Polym. Bull., 49, 297 (2002). 272. R. Gonza´lez-Nun˜ez, H. Padilla, D. De Kee, and B. D. Favis, Polym. Bull., 46, 323 (2001). 273. C. Albano, J. Trujillo, A. Caballero, and O. Brito, Polym. Bull., 45, 531 (2001). 274. C. G. In˜iguez, E. Michel, V. M. Gonza´lez-Romero, and R. Gonza´lez-Nun˜ez, Polym. Bull., 45, 295 (2000). 275. M. Alagar, S. M. Abdul Majeed, and S. Nagendiran, Polym. Adv. Technol., 16, 582 (2005). 276. J. K. Mishra and C. K. Das, Polym. Compos., 24, 83 (2003). 277. A. G. Pedroso and D. S. Rosa, Polym. Adv. Technol., 16, 310 (2005). 278. R. Adhikari, R. Godehardt, W. Lebek, S. Frangov, G. H. Michler, H.-J. Radusch, and F. J. Balta´ Calleja, Polym. Adv. Technol., 16, 156 (2005). 279. M. S. Kim and B. K. Kim, Polym. Adv. Technol., 15, 419 (2004). 280. D. Tomova and H.-J. Radusch, Polym. Adv. Technol., 14, 19 (2003). 281. J. K. Mishra, S. Roychowdhury, and C. K. Das, Polym. Adv. Technol., 13, 112 (2002).

26

Polyolefin Blends

282. R. R. N. Sailaja, Polym. Int., 54, 1589 (2005). 283. N. Abacha and S. Fellahi, Polym. Int., 54, 909 (2005). 284. Y. Lin, W. Du, D. Tu, W. Zhong, and Q. Du, Polym. Int., 54, 465 (2005). 285. I. A. Hussein, Polym. Int., 53, 1327 (2004). 286. N. Zeng, S. L. Bai, C. G’Sell, J.-M. Hiver, and Y. W. Mai, Polym. Int., 51, 1439 (2002). 287. A. S. Hashim and S. K. Ong, Polym. Int., 51, 611 (2002). 288. A. Tedesco, P. F. Krey, R. V. Barbosa, and R. S. Mauler, Polym. Int., 51, 105 (2001). 289. R. R. N. Sailaja, A. P. Reddy, and M. Chanda, Polym. Int., 50, 1352 (2001). 290. S. Chattopadhyay and S. Sivaram, Polym. Int., 50, 67 (2001). 291. J. K. Mishra, S. Raychowdhury, and C. K. Das, Polym. Int., 49, 1615 (2000). 292. P. Micic and S. N. Bhattacharya, Polym. Int., 49, 1580 (2000). 293. A. J. Marzocca, S. Cerveny, and J. M. Meńdez, Polym. Int., 49, 216 (2000). 294. A. Maciel, V. Salas, and O. Manero, Adv. Polym. Technol., 24, 241 (2005). 295. F. J. Rodriguez-Gonzalez, N. Virgilio, R. A. Ramsay, and B. D. Favis, Adv. Polym. Technol., 22, 297 (2003). 296. M. Suresh Chandra Kumar and M. Alagar, Adv. Polym. Technol., 21, 201 (2002). 297. K. Wang, C. Zhou, H. Zhang, and D. Zhao, Adv. Polym. Technol., 21, 164 (2002). 298. J. H. Yeo, C. H. Lee, C.-S. Park, K.-J. Lee, J.-D. Nam, and S. W. Kim, Adv. Polym. Technol., 20, 191 (2001). 299. A. A. Adewole, A. Denicola, C. G. Gogos, and L. Mascia, Adv. Polym. Technol., 19, 180 (2000). 300. D. N. Saheb and J. P. Jog, Adv. Polym. Technol., 19, 41 (2000). 301. H. Huang, L. Yang, and X. Ni, J. Macromol. Sci. B Phys., 44, 137 (2005). 302. F. Hernańdez-Sańchez, A. Manzur, and R. Olayo, J. Macromol. Sci. B Phys., 43, 1183 (2004). 303. A. K. Akinlabi, F. E. Okieimen, and A. I. Aigbodion, Polym. Adv. Technol., 16, 318 (2005). 304. J. Varga, A. Breining, and G. W. Ehrenstein, Int. Polym. Process., 15, 53 (2000). 305. C. Colletti, S. Piccarolo, and A. Valenza, Int. Polym. Process., 15, 46 (2000). 306. Y. Yu and J. L. White, Int. Polym. Process., 18, 388 (2003). 307. S. Pimbert, Int. Polym. Process., 19, 27 (2004). 308. Z. Hrnjak-Murgic, L. Kratofil, Z. Jelcic, J. Jelencic, and Z. Janovic, Int. Polym. Process., 19, 139 (2004). 309. H. Uehara, K. Sakauchi, T. Kanai, and T. Yamada, Int. Polym. Process., 19, 163 (2004). 310. H. Chen, U. Sundararaj, K. Handakumar, and M. D. Wetzel, Int. Polym. Process., 19, 342 (2004).

Chapter

2

Miscibility and Characteristics of Polyolefin Blends James L. White1 and Jinhai Yang1

2.1 INTRODUCTION It is our purpose here to summarize the characteristics of blends of polyolefins. We consider both blends of polyolefins with other polyolefins and with nonpolyolefins including polyamides and polystyrenes. In the case of interpolyolefin blends, our primary concern is with miscibility. In the case of blends of polyolefins with nonpolyolefins, the blends are all immiscible and our concern is with phase morphology. We also consider three component blend systems where the third component is a surfactant or compatibilizing agent, which collects at the interface.

2.1.1 Polyolefins Olefins are a group of unsaturated hydrocarbons of the general formula CnH2n. The polymers of olefins are known as polyolefins. Commercial polyolefins mainly include homopolymers of the olefin monomers: ethylene, propylene, butene-1, isobutene, and 4-methylpentene-1. These homopolymers have structure units as shown in Fig. 2.1, where the asterisk indicates asymmetric carbon atoms. Thus, polypropylene (PP), poly (butene-1) (PB1), and poly(4-methylpentene-1) (P4MP1) have different tactic forms. The most important commercial polyolefins are polyethylene, polyisobutene, and the isotactic forms, that is, iPP, iPB1, and iP4MP1. Polyisobutene was first polymerized by the IG Farbenindustries (BASF) in the late 1920s. Polyethylene was first polymerized by ICI in the late 1930s in a branched form (1). Linear polyethylene

1

Department of Polymer Engineering, The University of Akron, Akron, OH 44325, USA


27

28

Polyolefin Blends

CH2 CH2

PE

n

*

CH2 CH n CH3

PP

CH3 CH2 C n CH3

PIB

*

CH2 CH n CH2 CH3 PB1

*

CH2 CH n CH2 CH CH3 CH3 P4MP1

Figure 2.1 Structure units of polyolefins.

was developed by Ziegler and coworkers (2) and by Phillips Petroleum (3) in the 1950s. The other isotactic polyolefins were polymerized by Natta et al. (4,5) using modified Ziegler catalysts in the 1950s. It is a very interesting question why only these five types of polyolefins are commercialized, considering that there are many more types of olefin monomers. This is largely associated with their crystalline melting temperatures. Polyethylene, polypropylene, and polybutene-1 have reasonable melting temperatures, in particular, 135, 165, and 120 C, respectively. However, the melting temperatures decrease greatly with increase of the pendant group length in the isotactic polymers, for example, polypropylene (165 C), polybutene-1 (120 C), polypentene-1 (70 C), polyhexylene-1 (55 C), polyheptylene-1 (40 C), and polyoctylene-1 (38 C) (6,7). Introducing methyl groups onto the side group raises the melting temperatures, for example, poly(3-methylbutene-1) (300 C), poly(4-methylpentene-1) (235 C) (7). Polyisobutylene, which crystallizes with difficulty has become a significant synthetic rubber. In addition to the homopolymers, there are various polyolefin copolymers. Linear low density polyethylene (LLDPE) is probably the largest. This involves ethylene copolymerized variously with butene-1, hexene-1, octene-1, and 4methylpentene-1, and so on. Here the second comonomer content is around 5 mol%. In the 1990s, Exxon (8) and Dow Chemical (9,10) began to make polyolefin copolymers using new generation of Ziegler–Natta catalysts called metallocene catalysts. Polymers were also made with high comonomer contents and the density was as low as 0.86 g cc1 (compared to 0.83 g cc1 for amorphous polyolefins). It should be noted that these are random copolymers. Recently, Dow Chemical (11) made polyolefin block copolymers using ‘‘chain shuttling polymerization.’’ These block copolymers display melting temperature around 40 C higher than those of random copolymers of equivalent density. Ethylene–propylene copolymers, which usually contain roughly 50–75 mol% of ethylene, are also important polyolefin copolymers. These comonomers are elastomers and are defined as ethylene– propylene copolymer (EPM) or ethylene–propylene rubber. Terpolymers are commercially more important than EPM and are produced by polymerizing ethylene and propylene together with a diolefin, which are called ethylene–propylene diene terpolymer (EPDM). Since the late 1990s, another group of polyolefins, cyclopolyolefins, were developed (12,13). They are homo- or copolymers formed by the polymerization

Chapter 2 Miscibility and Characteristics of Polyolefin Blends

29

Figure 2.2 Norbornene form (I).

of norbornene derivatives of form (I) shown in Fig. 2.2. The bridged ring structure of norbornene does not readily fit into crystalline lattices, which together with comonomers make the resins completely amorphous with high transparency and high glass transition temperature, for example, 200 C with 60 mol% of norbornene (14). They are used in competition with polymers such as polycarbonate and polymethylmethacrylate for applications such as optical discs and lenses.

2.1.2 Blends Though more and more polymers have been synthesized since 1930s, however, they could not satisfy the fast growing millions of applications especially in electric, architecture, and automobile after 1960s. Thus, both blends and composites are compounded greatly to compensate the gap. So far, compounding is the most fast, economic, and efficient material development method for new applications. At present, about 36% of the synthetic resins are used in blends and about 39% in compounds with particle and fibers (15). Polymer blends date to the nineteenth century. In 1846, Parkes patented the first polymer blends of cis-1,4 polyisoprene (natural rubber) and trans-1,4 polyisoprene (gutta percha) (16). The modern era of polymer blending is thought to begin with the development of high impact polystyrene (HIPS) (17) and poly(acrylonitrile–butadiene–styrene) terpolymer (ABS) (18–21) in the 1950s and polyphenylene ether (PPE)/ polystyrene (PS) blends by General Electric (22) in the 1960s. Polymer blends can be divided into two groups: miscible and immiscible blends. Miscible blends are homogeneous and stable. Their properties tend to be intermediate. However, they are relatively few. Most polymer blends are immiscible. Their properties are strongly affected by their phase morphologies, which are decided by their viscosity, interfacial tension, and processing methods. In this review we will describe polyolefin blends. Many of these blends involve polar polymers with polyolefins.

30

Polyolefin Blends

These blends are immiscible and their interfaces are unstable. Special interfacial treatments are required to make them suitable as materials of commerce. A second group of blends are those in which the components are all polyolefins. These blends will be miscible or nearly so. Also in this paper the general questions of blend miscibility and interfacial characteristics will be treated.

2.2 POLYMER BLEND MISCIBILITY The question of polymer blend miscibility derives out of the nineteenth century development of thermodynamics and studies in the same periodic of binary mixtures of low molecular weight liquids. From a thermodynamic viewpoint, Gibbs (23) formulated the stability of multiphase systems in terms of the quantity G defined by (in modern notation (24)) G ¼ U þ PV TS ¼ H TS

ð2:1Þ

where U is the internal energy, P is the pressure, V is the volume, T is the absolute temperature, and S is the entropy. According to Clausius (25), equilibrium was associated with the maximization of Suniv , the entropy of the universe. Gibbs showed that this was equivalent to G for a system at constant temperature and pressure to reach a minimum. Thus, the formation of a solution requires DG < 0

ð2:2Þ

The quantity, DG, known today as the Gibbs free energy may be expressed DG ¼ DH TDS

ð2:3Þ

There is an extensive literature on modeling DH, the enthalpy or heat of solution, and DS, the entropy of solution. The simplest situation involves isotropic molecules with no energetic interactions. We may consider an ensemble of systems of equal numbers of component molecules and of energy (microcanonical ensemble). The entropy of mixing involves Boltzmann’s formulation of entropy as (24,26) S ¼ kb ln V

ð2:4Þ

where kb is the Boltzmann constant and V is the number of configuration. Considering N1 the number of isotropic molecules of molecule ‘‘1’’ and N2 the number of isotropic molecules of molecule ‘‘2’’ randomly distributed on a lattice, this leads to V¼

ðN1 þ N2 Þ! N1 !N2 !

ð2:5Þ

On the basis of Stirling’s approximation, as long as N is large the log of factorial, N!, is given by ln N! ¼ N ln N N

ð2:6Þ


31

The entropy of mixing is then DSm ¼ kb ðN1 þ N2 Þ½x1 ln x1 þ x2 ln x2

ð2:7Þ

where x1 ¼

N1 N2 ; x2 ¼ N1 þ N2 N1 þ N2

ð2:8Þ

The second problem, which needed to be resolved was the heat of solution, DH. For solutions with very similar molecules such as benzene and toluene, or hexane and octane, the heat of mixing is near zero. For various pairs of molecules, for example, acetone and chloroform, significant heat is evolved (DH < 0) and the solution process is said to be exothermic. For other pairs of molecules, heat is absorbed (DH > 0) and the solution process is called endothermic. As early as 1906, van Laar (27) modeled the heat of mixing of a binary solutions using van der Waals equation of state. The views of van Laar were subsequently expanded by van Laar and Lorenz (28), Hildebrand (29,30), and Scatchard (31). The latter authors considered a mixtures of two fluids each with isotropic molecules with interaction energies, c11 and c22 . The interaction energy between the two fluid molecules is c12 . Scatchard (31) obtained the form DUm ¼ Um U1 x1 U2 x2 ¼ Vx1 x2 ðc11 þ c22 2c12 Þ where V is the mixture volume. He further presumed pffiffiffiffiffiffiffiffiffiffiffiffi c12 ¼ c11 c22

ð2:9Þ

ð2:10Þ

which leads to 1=2

1=2

DUm ¼ Vx1 x2 ðc11 c22 Þ2 Hildebrand (32) defined ‘‘solubility parameters’’ of liquids as DUi 1=2 1=2 ; i ¼ 1; 2 di ¼ cii ¼ Vi

ð2:11Þ

ð2:12Þ

Thus, Equation 2.11 is equivalent to DUm ¼ Vx1 x2 ðd1 d2 Þ2

ð2:13Þ

Hildebrand (32) also determined experimental values for various liquids. Typical values are listed in Table 2.1 taken from Hildebrand and Scott (33). A major problem with Equation 2.13 is that it leads to the heat of mixing being positive, that is, always endothermic, absorbing heat. Many solutions are exothermic, that is, heat is evolved when the solution is formed. There have been various efforts to model exothermic heats of mixing. These are generally dated to Dolezalek (34) who considered dissolution to represent processes similar to chemical reactions. The modeling of the formation of solutions that we have described assumes that there is no change in volume in mixing. Theories of liquids that consider cell models

32

Polyolefin Blends Table 2.1 Solubility Parameters of Organic Liquids at 25 C. Solubility parameter, (J cm3)1/2

Substance n-Hexane n-Octane n-Hexadecane Cyclohexane Benzene Ethyl benzene Naphthalene Fluorinated hexane (C6F14) Fluorinated octane (C8F18) Fluorinated cyclohexane (C6F12) Fluorinated benzene (C6F6) Dimethyl ether Carbon disulfide Chloroform

7.30 7.55 8.0 8.20 9.15 8.80 9.9 5.6 5.7 6.0 8.1 8.8 10.0 9.3

Adapted from Reference 33.

volume changes begin with Lennard-Jones and Devonshire (35) in 1937. This approach was extended to solutions by Prigogine and Garikian (36) and LonguetHiggins (37) in the 1950s. These developments up to the mid 1950s are described by Prigogine in his book ‘‘The Molecular Theory of Solutions’’ (38). Volume change in mixing can affect the change in DG and induce phase separations which are not predicted from Equations 2.7 and 2.13. Early investigations of the thermodynamics of polymer solutions indicated that solutions exhibited large deviations from ideality. Calorimetric measurement showed that heats of solution were small indicating that these deviations were due to the entropy of mixing. Subsequently, Meyer (39) proposed that the low entropy of mixing was associated with the reduced number of configurations available to polymer chains compared with low molecular weight molecules. More quantitative formulations were subsequently developed by Huggins (40) and notably Flory (41–43). The entropy of mixing DSm was shown to have the form DSm ¼ kb ðN1 þ N2 Þ½x1 ln f1 þ x2 ln f2

ð2:14Þ

where f1 and f2 are the volume fractions of solvent and polymer. The heat of mixing DHm in polymer solutions that has been expressed by Flory (43) is DHm ¼ xkb TN1 f2

ð2:15Þ

where x characterizes the interaction energy per solution molecule divided by kb T. According to Flory x may be positive or negative. The free energy of solution, DGm , may be expressed DGm ¼ kb TðN1 ln f1 þ N2 ln f2 þ xN1 f2 Þ

ð2:16Þ


33

This formulation was subsequently extended to polymer blends by Scott (44) in 1949 as follows: V f1 f2 ð2:17Þ ln f1 þ ln f2 þ xf1 f2 DGm ¼ kb T M2 V s M1 where M1 and M2 are the numbers of segments of the two polymers, V is the total volume of mixture, and Vs is the volume of segments, which usually are regarded as the volume of repeating units of the polymers. The formulation of Scott (44) does not present the range of phenomena occurring in polymer blends. Various binary blends exhibit lower critical solution temperatures (LCST) where phase separations occur at lower temperature. Other blends exhibit upper critical solution temperatures (UCST) where miscible blends exhibit phase separations at higher temperatures (45). It was shown by McMaster (46) that volume changes occurred in mixing.

2.3 INTERFACES IN LIQUID AND POLYMER MIXTURES The concept of surface tensions between air and liquids seems to date back to Thomas Young (47). Measurements of surface tensions of liquids (48) and its relationship to waves on liquids (49) and the breakup of capillary jets and viscous filaments (50,51) occur throughout the nineteenth century. In thermodynamics point of view, the surface tension is the increase of Gibbs free energy as a result of creating a unit surface area. The surfaces of liquids can be easily understood as an elastic film that inherently tends to shrink to decrease its surface. A summary of surface tensions of various low molecular weight liquids is contained in Table 2.2. It can be seen that the highest values are for molten metals. For organic liquids, the highest values are for polar compounds. Aromatic and aliphatic compounds are similar and the lowest values are for fluorinated compounds. It was believed that the intermolecular interactions play similar role in determining cohesive energy density and surface tension. The surface tension was suggested to be estimated by the cohesive energy density based on the following equation (53) Vð298KÞ 4 2=3 ð2:18Þ gðTÞ ¼ 0:75 ecoh VðTÞ where g is the surface tension, ecoh is the cohesive energy density, V is the molar volume, and T is the absolute temperature. However, Equation 2.18 cannot be used to evaluate the polymers containing polar units such as amide, urea, and hydroxyl groups, which form strong hydrogen bonds because the effect of hydrogen bonding on surface tension is generally smaller than its effect on cohesive energy density (54). For high molecular weight materials, the surface tensions of these materials has been found to vary with their molecular weight roughly as (55) g ¼ g1

Ke M 2=3

ð2:19Þ

34

Polyolefin Blends Table 2.2 Surface Tensions of Low Molecular Weight Compounds. Surface tension, dyn cm1

Material Iron Gold Silver Tin Lead Mercury Water (H2O) Ethylene glycol Epsilon-caprolactam Phthalic acid Lactic acid Styrene Acrylic acid Methyl methacrylate Benzene Cyclohexane Hexafluorobenzene n-Octane n-hexane Octafluoro-2-butene Decafluorobutane

Temperature, C

1962 1083 882 540 442 318 74.5 50.7 47.1 42.8 39.5 32.6 29.0 26.1 28.8 25.2 22.6 21.5 18.5 12.7 11.8

1200 1200 1200 455 455 455 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20

Adapted from Reference 52.

where g 1 is the surface tension at infinite molecular weight, Ke is a constant, and M is the molecular weight. The surface tension of polymers seems to vary with temperature approximately as (56) g ¼ g 0 ð1 T=Tc Þ11=9

ð2:20Þ

where g 0 is the surface tension at T ¼ 0K and Tc is the critical temperature whose value is around 1000 K for most polymers. The temperature coefficient of surface tension is given by differentiation of Equation 2.20 to be dg=dT ¼ ð11=9Þðg 0 =Tc Þð1 T=Tc Þ2=9

ð2:21Þ

Thus, the coefficient is practically a constant at ordinary temperature, that is, T Tc ¼ 1000 K. The surface tensions of some polymers are listed in Table 2.3. It came to be realized that interfaces between low molecular weight liquids in immiscible liquid mixtures possess an interfacial tension. Experimental measurements of interfacial tension between low molecular weight liquids were first made in the nineteenth century (59,60). Various experimental techniques have been developed to measure interfacial tensions. A popular method has been the pendant drop measurement, which was originally developed by Andreas et al. (61)


35

Table 2.3 Surface Tension (g) of Polymers at Different Temperature in dyn cm1 .

Polymer Linear polyethylene Atactic polypropylene Isotactic polypropylene Isotactic polypropylene

Molecular weight g mol1 Mw ¼ 67; 000 — — Mw ¼ 267; 000

g g (20 C) (140 C)

—

25.5

Polystyrene Polystyrene Poly(ethylene terephthalate)

Mn ¼ 44; 000 Mn ¼ 9290 Mn ¼ 25; 000

40.7 44.6

28.8 22.7 23.1 19.7 (210 C) 19.2 (180 C) 32.1 31.5 36.7

Poly(ethylene oxide) Poly(methyl methacrylate) Poly(e-caprolactam)

M ¼ 6000 Mn ¼ 3000 —

42.9 41.1

33.8 32.0

Poly(hexamethylene adipamide) Bisphenol-A-polycarbonate Poly(dimethyl siloxane) Polytetrafluoroethylene

Mn ¼ 19; 000

46.5

— Mn ¼ 75; 000 M ¼ 1088

42.9 20.9 21.5

30.2 (270 C) 35.7 14.3 13.7

Polybutene-1

35.7 29.4 30.1 30.0

g (180 C)

References

26.5 20.4 20.8 19.1 (220 C) 17.5 (220 C) 29.2 28.8 34.2

(55, 56) (55, 56) (55, 56) (57)

30.7 28.9 36.1 (265 C) 28.3 (300 C) 33.3 12.3 11.1

(55, 56) (55, 56) (55, 56)

(57) (55, 56) (58) (55, 56)

(55, 56) (55, 56) (55, 56) (55, 56)

Adapted from References 55–58.

for measuring surface tension. Another method involves the breakup of molten polymer threads. Interfacial tensions of some polymer pairs are listed in Table 2.4. Measurements of interfacial tensions of polymer melts were reviewed by Wu (55), Koberstein (65), and Demarquette (66). The measurements usually need long equilibrium time because of the high viscosities of polymer melts. The measurements can be divided into two groups: static methods in which interfacial tension is calculated based on the equilibrium profile of the drops and dynamic methods that study the evolution of fiber or drop profiles with time. Static methods include pendant drop method, sessile drop method, and rotating drop method. Dynamic methods include breaking thread method, imbedded fiber method, and deformed drop retraction method. Interfacial tension plays a very important effect in controlling the phase morphology of immiscible polymer blends. In mixing of polymer blends, the minor phase is deformed into long fibrils or thin films by shear or elongation flows, and then is broken into small particles. At the same time, the dispersed particles can collide and merge together, which is called coalescence process. The final morphology is decided by the balance between breakdown and

36

Polyolefin Blends

Table 2.4

Interfacial Tensions of Molten Polymer Blends in dyn cm1 .

Polymer A

Polymer B

Polyethylene Polyethylene Polyethylene

Polystyrene Polysulfone Poly(phenylene sulfone) Poly(ethylene terephthalate) Polycarbonate Polyamide 6 Polyethylene Polystyrene Polycarbonate Polyamide 6

Polyethylene Polyethylene Polyethylene Polypropylene Polypropylene Polypropylene Polypropylene

Interfacial tension, dyn cm1

Temperature, C

References

4.0 7.0 7.9

290 290 290

(62) (62) (62)

9.4

290

(62)

12.5 12.8 1.63 4.5 8 12.2

290 290 290 290 290 290

(62) (62) (63) (64) (64) (64)

Adapted from References 62–64.

coalescence. Large interfacial tension resists this process in the following three ways. First, larger interfacial tension indicates lower interfacial adhesion, which prevents passing shear stress from one phase to another phase. Second, the interface with large tension is like an elastic film as we mentioned earlier. Inherently, it tends to shrink the deformed phases into spherical shape. Thus, it resists both the shear or elongation deformation of the dispersed phase. Third, large interfacial tension increases the tendency to coalescence. The low interaction between the two phases leads the matrix molecules easy to be excluded when two dispersed particles collide together. Also, large interfacial tension accelerates the merging process of the collided particles.

2.4 POLYOLEFIN–POLYOLEFIN BLENDS There is a history of investigations of blends of polyolefins. Many of these blends were not produced on purpose but were the results of incompletely understood polymerization processes. Examples of these are the various studies of elastomeric polypropylenes, which are mixtures of polypropylenes of varying tacticity levels (67–71). These materials often have interesting technological properties. It is of more concern to consider the results of specially prepared blends of known characterized polymers.

2.4.1 Blends between Polyethylenes Hundreds of polyethylenes are commercially available, and their blends have been used to provide desired range of properties. The blends have high level of mis-


37

cibility. As mentioned earlier, LLDPE is a copolymer of ethylene with other olefins, for example, butene, hexene, and octene. Thus, it includes many short branches. LDPE is synthesized by free radical polymerization at high pressure. It includes both short and long chains. Thus, except molecular weight, both branch type and branch content are the two most important molecular structures of polyethylenes. Although the blends of polyethylenes are nearly miscible, their miscibility has been studied extensively. Three methods have mainly been used to study the miscibility. Hill et al. (72–75) mainly used differential scanning calorimetry(DSC) combined with transmission electron microscope (TEM). The melts of the blends are quenched into acetone at freezing point. If DSC shows only one melting peak and there is only one group of crystal morphology based on TEM, the blend is thought miscible in the melt. If DSC shows double melting peak and there are two groups of crystal structures, the blend is thought to be separated in the melt. Hussein et al. (76– 79) used rheology method. They measured both the dynamic viscosity and viscosity at zero shear rate of the blends. If the two parameters of the blends show log-additive linear relationship based on blend content, the blends are considered miscible. If they deviate from the linearity, the blends are thought immiscible. Alamo et al. (80–82) used small-angle neutron scattering (SANS). One component of the blends is deuterated. It was concluded that deuterating can increase the interaction parameter. In some cases, it is enough to lead to phase separation. Tashiro et al. (83–86) argued that they found phase-segregated structures between polyethylene and deuterated polyethylene. These studies suggest the following conclusions regarding the miscibility of polyethylene blends. (1) The branch content is the most important factor to control the miscibility. The miscibility decreases with increase of branch content. (2) The branch type also affects the miscibility. The miscibility decreases with increase of branch length. For example, LLDPE with 4.4 mol% of butene is miscible with high density polyethylene (HDPE) at all compositions (82). However, LLDPE with 4 mol% of hexene is immiscible with HDPE when the HDPE fraction is less than 60 wt% (87), and LLDPE with 2.1 mol% of octene is claimed immiscible with HDPE when HDPE fraction is less than 50 wt% (73). It is hard to deduce the effects of long-branch chains in LDPE. (3) Decreasing molecular weight increases the miscibility. For example, in a study of Hill et al. (72), HDPE with molecular weight 105 g mol1 shows phase separation with the LDPE with 5.2 mol% of branches when the HDPE fraction is lower than 50 wt%. However, HDPE with molecular weight 2553 g mol1 is miscible with LDPE at all HDPE composition. (4) Most of the blends possessed an UCST. Thus, the blends become more miscible with the increase of melt temperatures. (5) Usually, the blends become more miscible with the increase of linear polyethylene content. (6) In summary, the miscibility between polyethylenes is dependent on molecular structures (molecular weight, branch content, and type), temperature, and composition. Thus, phase diagrams based on the composition and the temperature are necessary to characterize the miscibility of specific polyethylene blends.

38

Polyolefin Blends

2.4.2 Blends between Isotactic Polypropylene and Ethylene Propylene Copolymers There have been extensive applications of isotactic polypropylene (iPP)/EPM blends. These were used to produce rubber toughened polypropylene blends and subsequently polyolefin thermoplastic elastomers (88,89). Most commercial EPMs contain more than 50 mol% of ethylene, and these are elastomers. The solubility parameter of EPM should be intermediate to those of polyethylene and polypropylene dependent on ethylene content. Thus, it is often used to compatibilize PE/PP blends (90,91). IPP–EPM blends were generally considered to be immiscible (92–95). Chen et al. (96) have reported miscibility and a LCST in blends of isotactic polypropylene and an ethylene-propylene terpolymer (EPDM). Seki et al. (97) reported that iPP was miscible with EPM with both 19 and 47 mol% of ethylene based on both transmission electron microscopy (TEM) and SANS. However, the molecular weight of the iPP was below 10; 000 g mol1. More recently both Nitta et al. (98) and Kamdar et al. (99) reported that the miscibility between iPP and EPM, whose molecular weight is higher than 100; 000 g mol1 , is strongly dependent on ethylene content in EPM. The miscibility increases with decrease of ethylene content. IPP was miscible with EPM with ethylene content lower than around 17 mol% based on both dynamic mechanical analysis (DMA) and TEM. The EPM domain sizes decrease with decrease of ethylene content as shown in Fig. 2.3.

Figure 2.3 TEM micrographs for the iPP/ethylene–propylene copolymer (80/20) blend films. The propylene contents by mol% in the copolymers are (a) 89.3; (b) 84.3; (c) 76.5; (d) 52.5. (From Reference 98 with permission from Elsevier.)


39

2.4.3 Blends between iPP and High Comonomer Concentration Polyethylene Copolymers Since about 1990, ethylene copolymers with substantially high level of olefin comonomers (e.g., butene, hexene, and octene) have become available (100–104). Compared with most commercial EPM, they are available in a pellet form that facilitates continuous processing with twin screw extruders. These copolymers are becoming widely used to substitute for EPM to toughen iPP (105–107). The miscibility between these copolymers and iPP is also strongly dependent on the comonomer content. In a 1996 paper, Yamaguchi et al. (108) of Tosoh studied the miscibility of isotactic polypropylene with ethylene–hexene-1 (EHR) and ethylene–butene-1 (EBR) copolymers. It was found that iPP was miscible with ethylene–hexene-1 copolymers with more than 50 mol% of hexene based on both TEM and DMA (as shown in Figs. 2.4 and 2.5). However, though DMA indicated miscibility between iPP and ethylene– butene-1 copolymers with 56 and 62 mol% of butene-1 (as shown in Fig. 2.4), TEM suggested that there were tiny particles in the blends (as shown in Fig. 2.5). Thomann et al. (109) showed that ethylene–butene-1 copolymers are miscible with iPP when the butene-1 content was higher than 78 mol%. Yamaguchi et al. subsequently investigated the mechanical properties of blends of polypropylene with ethylene–hexene copolymers (110,111). Interestingly

Figure 2.4 Variation of the mechanical loss modulus (E00 ) with temperature for the iPP (O), the copolymers: EHR57 (with 57 mol% of hexene-1); EBR56 (with 56 mol% of butene-1); EBR62 (with 62 mol% of butene-1) (D), their blends (.). (From Reference 108 with permission from John Wiley & Sons, Inc.)

40

Polyolefin Blends

Figure 2.5 TEM micrographs of thin sections of series of blends with 50wt% of iPP. EHR57 has 57 mol% of hexene-1; EBR56 has 56 mol% of butene-1; EBR62 has 62 mol% of butene-1. (From Reference 108 with permission from John Wiley & Sons, Inc.)

Yamaguchi and Miyata (112) found a lack of miscibility of these copolymers with syndiotactic polypropylene (as shown in Fig. 2.6).

2.4.4 Blends between iPP and PB1 Based on the research of Thomann et al. (109), iPP is miscible with ethylene–butene copolymer with 78 mol% of butene. It is an interesting question whether iPP is miscible with PB1. For iPP/PB1 (30/70 by weight) blends made by melt mixing, Cham et al. (113) reported that liquid–liquid demixing happened at 180, 220, and 250 C. The demixing


41

Figure 2.6 Transmission electron micrographs for (a) iPP/EHR and (b) sPP/EHR. Hexene-1 content is 57.1 mol%. The samples were stained with RuO4. (From Reference 112 with permission from American Chemical Society.)

process was observed by optical microscope, which indicates that the refractive index difference between iPP and PB1 is large enough for microscopy observation. Bartczak et al. (114) also reported that PB1 particles were found to disperse in the iPP spherulites when the blends were isothermally crystallized at 125 C. These suggest that the blends are immiscible at least up to 250 C. Cham et al. (113) found some evidence of partial miscibility. The demixed blends were isothermally crystallized at 145 C. It was found that the crystals were nucleated in the iPP domains, then, they gradually extended to all the domains. Surprisingly they continued to grow into the continuous PB1 phase with low density. When the crystals growing in PB1 phase reach some iPP domains, they can nucleate the crystal grow in those domain. Since at 145 C PB1 cannot crystallize, the crystals growing in PB1 phase must be PP crystals. This indicates that in the continuous PB1 phase there are iPP molecules, indicating that they are partially miscible. The authors also showed that the iPP crystals growing in PB1 phase have lower melting temperatures than those growing in iPP domains. It was reported that iPP/PB1 blends showed only one tand peak in a DMA test (113,115,116). This was usually taken as a proof for the miscibility between the amorphous iPP and PB1. However, a phase separation was reported at 250 C by Cham et al. (113). They suggested an explanation in their paper. First, the difference of Tg of iPP and PB1 is not large, around 25 C. Second, the separated phases are not pure iPP or PB1 phases, but iPP-rich and PB1-rich phases. Thus, the difference of the Tg of the two phases should be less than 25 C. The small Tg difference may make the two tand peaks merge together. Here, we also want to mention another point based on the study of Hsu and Geil (116), where blends with iPP content lower than 30 wt% show only one peak, but, those with higher iPP content show two peaks (as shown in

42

Polyolefin Blends

Figure 2.7 DMA curves of iPP/PB1 blends. (From Reference 116 with permission from John Wiley & Sons, Inc.)

Fig. 2.7). First, the tand peak of PB1 is much larger than that of iPP. Thus, the iPP peak in the blends with low iPP content is easy to hide, but the PB1 peak in the blends with low PB1 content cannot. It was reported by Hsu et al. (116) and Lee et al. (117) that iPP can nucleate the crystallization of PB1 to increase its crystallization temperature, and PB1, especially in the blends with high content of PB1 more than 70 wt%, can obviously retard the crystallization of iPP. On the basis of the publications available so far, we can make the following conclusions: (1) iPP and PB1 are partially miscible and phase separation occurs at reported temperatures up to 250 C. Melt mixed blends should have a two-phase morphology. (2) The blends made by precipitation from dilute solution show homogeneous mixing. But they are in a metastable state and tend to demix at high temperature.

2.5 BINARY IMMISCIBLE BLENDS There is a large literature involving binary blends of polyolefins, especially polyethylene and isotactic polypropylene, with other polymers. Among the polymers included in these studies are polystyrenes and polyamides. Polyolefins contribute to increase toughness, processability, chemical resistance, and moisture absorption resistance to these polymers. The other polymers contribute high modulus, heat resistance, and oxygen or solvent barrier properties to polyolefins. However, these blends are all immiscible. Useful polymer pairs generally require a compatibilizing agent as described in Section 2.6.


43

Figure 2.8 Scanning electron microscopy micrographs of PS/PP (70/30) blend prepared at 190 C.

2.5.1 Polyolefin–Polystyrene Blends Polyolefin–polystyrene blends have long been studied. They are immiscible showing two-phase morphologies. The blends showed poor mechanical properties, especially elongation at break and impact strength, much lower than those predicted based on an additive rule (118). Their fracture surfaces were observed by electron microscope (119,120). As shown in Fig. 2.8, the dispersed phase is easy to be pulled from the matrix and leaves very smooth surface, indicating low interfacial adhesion. Because of the high interfacial tension, the morphology of the blends is not stable. Coalescence readily occurs in the molten state. As suggested by Macosko et al. (121), in melt mixing of immiscible polymer blends, the disperse phase is first stretched into threads and then breaks into droplets, which can coalesce together into larger droplets. The balance of these processes determines the final dispersed particle sizes. With increase of disperse phase fraction (usually more than 5 wt%), the coalescence speed increases and the dispersed phase sizes increase (121–123).

2.5.2 Polyolefin–Polyamide Blends Polyolefin–polyamide melt blends are striking in not only the lack of miscibility, but also the large interfacial tensions between the two melt phases. Investigations of these phenomena in our laboratories (118,124–126) have made numerous studies of these polymer blend systems and found that their phase morphology are quite unstable and trend to coalesce especially under quiescent or low deformation rate conditions. Similar to polyolefin–polystyrene blends, they also show weak interfacial adhesion (118,124,127) (as shown in Fig. 2.9). The mechanical properties of the

44

Polyolefin Blends

Figure 2.9 Scanning electron microscopy micrographs of PA12/PP (70/30) blend prepared at 190 C.

blends are also much lower than those predicted based on the additive rule (125). The disperse phase particles increase in size with increase of concentration (126,128).

2.6 TERNARY BLENDS OF POLYOLEFINS WITH OTHER POLYMERS AND COMPATIBILIZING AGENTS There is a long history of the use of surfactants with a mixture of low molecular weight molecules. These were used as soaps in the preparation of emulsion and other applications. This science and technology were highly developed in the 1920s. The use of high molecular weight surfactants dates back to the 1950s and 1960s and was only studied scientifically in the 1970s.

2.6.1 Surfactants and Compatibilizing Agents The term surfactant has the same meaning with the term surface active agent. Actually it is a blend of ‘‘surface active agent.’’ It has been extensively accepted to substitute the term surface active agent since the early 1960s. Surfactants are usually organic amphipathic compounds with both hydrophobic and hydrophilic groups. They can lower the surface tension of liquids to allow them to spread easier. Their earlier name ‘‘surface active agent’’ was given because of this characteristic. Also they can lower the interfacial tension between two liquids. This characteristic leads them to be applied extensively to make emulsions. Soap is the earliest surfactant. The earliest soap was made by boiling fats together with ashes of plant, which dated as early as 2800 B.C. in ancient Babylon. In 1823, Michel Eugene Chevreul, a French chemist, worked out the structure of fats


45

as compounds of organic acids and glycerol. He showed that alkalis converted fats into soaps. He isolated stearic acid and various other organic fatty acids. He sought to commercially develop stearic acid. This was the beginning of the scientific studies of modern soaps and surfactants. From then on more and more surfactants were developed and were used in more applications, for example, detergents, fabric softeners, emulsifiers, paints, and so on. Though surfactants were mainly used to stabilize the immiscible small molecular mixtures, they were also reported in more recent to be used to make immiscible polymer blends more compatible (129–132). They could suppress the coalescence of dispersed particles to decrease their sizes, but could not increase the interfacial adhesion to increase the mechanical properties. The term ‘‘compatibilizing’’ was defined by Gaylord in 1966 (133) as ‘‘rendering a mixture of two or more polymeric materials permanently miscible so as to form a homogeneous composition that has useful plastic properties and which does not separate into its component parts.’’ Actually, in his definition the term ‘‘miscible’’ was not appropriate and another term ‘‘uniform’’ would be better. A few years later, he defined another term ‘‘compatibilization’’ as ‘‘absence of separation or stratification of the components of the polymeric alloy during the expected useful lifetime of the product’’ (134). Different from surfactants, substance considered as compatibilizing agents are today considered not only to decrease the interfacial tension, but also to enhance mechanical behavior presumably by increasing the interfacial adhesion. They must be high molecular weight compounds or form high molecular weight compounds in the blends. They stay on the interface with the part of their molecular chain in one component of the blends and the other part of chain in the other component. Though it was usually thought that the most efficient compatibilizing agents are block copolymers of the two polymers in the blends, however, there are only few block copolymers commercially available. Thus, the most applicable way is to use some commercially available polymers whose surface tensions are between those of the two blend components. For blends between polyolefins and some polar polymers, for example, polyamide, polycarbonate, polyester, reactive group grafted polyolefins are usually used, for example, maleic anhydride grafted polyethylene for polyamide/PE blends. In mixing, the maleic anhydride can react with the amine groups on polyamide to produce polyamide grafted polyolefins, which have good compatibilizing effect. Table 2.5 indicates that compatibilizing agents can decrease the interfacial tensions of some polymer melt pairs greatly.

2.6.2 Polyolefin–Polystyrene Blends with Compatibilizing Agents 2.6.2.1 Polyethylene–Polystyrene Blends with Compatibilizing Agents In case of the first studies using polymer compatibilizing agents, Anderson et al. of Dow Chemical in 1960 (135) irradiated polyethylene with g rays and

46

Polyolefin Blends

Table 2.5 Interfacial Tensions between Polymer Melt Pairs Including Compatibilizing Agents. MAH–PP: Maleic Anhydride Grafted Polypropylene; SEBS: Hydrogenated Triblock Copolymer of Styrene and Butadiene; MAH-g-SEBS: Maleic Anhydride Grafted SEBS; PEMA–Zn: Poly(ethylene-co-methacrylic Acid) Ionomer Neutralized by Zinc. Binary system PE/PA 6

PE/PS

Compatibilizing agent None MAH-g-PP MAH-g-SEBS PEMA-Zn None SEBS

Temperature, C

Interfacial tension, dyn cm

230 230 230 230 180 180

12.5 2 1.5 1.8 5.8 1.1

Adapted from Reference 118 by Chen & White.

subsequently treated it with styrene to make styrene grafted polyethylene, which increased the tensile strength of PS/LDPE blends. Locke and Paul in 1973 (136) also reported that styrene-grafted polyethylene could improve the tensile strength, modulus, and elongation of LDPE/PS blends and greatly decreased the dispersed particle sizes. Kishimoto et al. in 1970 (137) masticated HDPE/ PS blends with the presence of a free radical initiator. They suggested that the formed block graft copolymers improved the properties of the blends. Carrick in 1970 (138) prepared polyethylene grafted polystyrene copolymers based on Friedel–Crafts (F-C) alkylation reaction of polystyrene. The aluminum chloride (AlCl3) was put into the solution of PE/PS blends in cyclohexane to initiate the reaction. After a period the solution was precipitated in methyl–ethyl ketone (MEK) to extract polystyrene. In the following years, Heikens et al. (139–142) used polyethylene-graft-polystyrene prepared as suggested by Carrick to compatibilize LDPE/PS blends and found that they showed good effect to increase interfacial adhesion and impact strength. In 1980, Sjoerdsma et al. (143) reported that tapered butadiene–styrene block copolymer (PS-(PS-PB)random-PB) could increase the interfacial adhesion of LDPE/ PS blends. In the followed years, hydrogenated butadiene–styrene copolymers (SEBS or SEB) was used as compatibilizing agents for HDPE/PS blends by Lindsey et al. (144), Fayt et al. (145), and for LDPE/PS blends by Fayt et al. (146–149). Fayt et al. (147,148) also reported that tapered hydrogenated butadiene–styrene copolymers (PS-(PS-PB)random-PB) showed better compatibilizing effect than pure hydrogenated butadiene–styrene diblock copolymer in decreasing dispersed phase sizes and increasing the elongation and tensile strength of LDPE/PS blends. Today these butadiene–styrene copolymers, for example, SB, SBS, SEB, SEBS, and so on, have been the major group of compatibilizers for PE/PS blends (150–157), presumably because of their large commercial availability. With 5–10 wt% of these compatibilizers in PE/PS blends, the dispersed phase sizes are decreased, and the elongation at break and impact strength are increased.


47

Gao et al. (158) showed that LDPE-g-PS prepared based on Friedel-Crafts alkylation reaction could increase the impact strength, elongation at break and tensile strength of LDPE/PS blends. Wang et al. (159) reported that adding dicumyl peroxide (DCP) into HDPE/PS/SBS blends can obviously increase their impact and tensile strength. The mixing order affected the blend properties very much. However, DCP did not show such effect with HDPE/PS blends. 2.6.2.2 Polypropylene–Polystyrene Blends with Compatibilizing Agents The most commonly used compatibilizers for PP/PS blends are also di- or triblock copolymers of styrene and butadiene (SB and SBS) and their hydrogenated products (SEB and SEBS) (160–164). They form dispersed phases in both pure PP and PS. In PP/PS blends, they locate at the interface to connect both PP and PS phase together. Thus, the interfacial tension is decreased and the dispersed phase sizes are greatly decreased. Polystyrene-b-poly (ethylene-co-propylene) block copolymers were also used to compatibilize PP/PS blends (165,166). They showed similar effect to block copolymers of styrene and butadiene. Fig. 2.10 indicates that they locate in the blend interface. Graft and block copolymers of propylene and styrene have been developed to compatabilize PP/PS blends. Del Giudice et al. (167) and Xu and Lin (168) have synthesized PP-b-PS. Kim et al. (169) and Li et al. (170) first polymerized propylene together with some functional monomers, then polymerized styrene from these monomers units to form polystyrene branches. Diaz et al. (171,172) grafted PP chains onto PS chains based on F-C alkylation reaction when mixing PP/PS blends in the presence of AlCl3 catalyst and styrene. All these copolymers help form very

Figure 2.10 Transmission electron micrograph of iPP/aPS 70/30 blends compatibilized with poly (styrene-b-ethylene-co-propylene) (SEP) and stained with RuO4: 10 wt% of SEP. (From Reference 165 with permission from Elsevier.)

48

Polyolefin Blends

strong interfacial adhesion to decrease the dispersed phase size and increase the tensile strength.

2.6.3 Polyolefin–Polyamide Blends with Compatibilizing Agents To increase the affinity between polyolefins and polyamides, some polar groups, such as maleic acid or its anhydride, can be grafted onto polyolefin chains. In mixing the polar groups react with polyamides to form polyamide-grafted polyolefins, which are the real compatibilizing agents. The studies about polar group grafted polyolefins were stimulated by trial to increase the dyeability of polyolefin fibers or wettability of polyolefin films, which was earlier than their application to compatibilize polyolefin– polyamide blends. In the 1960s, Nowak and Jones (173–175) of Dow Chemical grafted reactive reagents such as acrylic acid onto melt polyethylene in a screw extruder. In the 1960–1970s, unsaturated acid group grafted polyolefins were seen to be used as compatibilizing agents for polyolefin/polyamide blends first in patents (176–181), then in publications by Ide and Hasegawa (182), Braun and Eisenlohr (183). Since the late 1980s, compatibilizing agent effects on polyolefin/polyamide blends were extensively studied. 2.6.3.1 Polyethylene–Polyamide Blends with Compatibilizing Agents The most important group of compatibilizers for polyethylene/polyamide blends is the free radical initiated graft copolymers of polyethylene, SEBS, EVA, EPDM, and polypropylene with unsaturated acid or its anhydride. They include graft copolymers of polyethylene with maleic acid or its anhydride, glycidyl methacrylate, and ricinoloxazoline maleinate; graft copolymers of SEBS with glycidyl methacrylate, maleic anhydride, and ricinoloxazoline maleinate; graft copolymers of EVA with mercaptoacetic acid and maleic anhydride; graft copolymer of EPDM with maleic anhydride; graft copolymer of polypropylene with maleic anhydride. The acid or anhydride groups of the grafted copolymers can react with the -NH2 groups in polyamide to form polyolefin-graft-polyamide, the real compatibilizers. These graft copolymers have been called precursor compatibilizers. An extensive study comparing various graft copolymers was given by Chen and White (118). Because the free radical initiated graft reaction can also lead to the cross-linking of polyethylene, copolymers of ethylene and with acrylic acid (184,185), glycidyl methacrylate (184,186), methacrylic acid and 10-undecenoic acid (187–189) were synthesized to compatibilize polyethylene/polyamide blends. The poly (ethyleneco-methacrylic acid) ionomers neutralized by sodium (184) and zinc (45,118,190– 192) has also used as compatibilizers. High energy irradiation, used to modify the surface of fibers or films at beginning, was also used to compatibilize the polyethylene/polyamide blends (193–196).


49

2.6.3.2 Polypropylene–Polyamide Blends with Compatibilizing Agents The most commonly used compatibilizers for the blends are graft copolymers of polypropylene with maleic anhydride (PP-g-MAH) (197–203), glycidyl methacrylate (PP-g-GMA) (204,205), and acrylic acid (PP-g-AA) (198,206). They can increase the tensile strength and impact strength of the blends. Favis et al. (207– 210) used an ionomer, a copolymer of ethylene and a mixture of methacrylic acid, zinc methacrylate, and isobutylacrylate to compatibilize PP/PA blends and found it effective to decrease the dispersed particles and improve mechanical properties. To further increase the toughness of polyamides and polyamide-polypropylene blends, elastomers grafted with maleic anhydride were used, for example, SEBS-g-MAH (203,211–214), poly(ethylene-co-octene) grafted maleic anhydride (POE-g-MAH) (202,215), EPR-g-MAH (214) and EVA-g-MAH (216). For example, Fig. 2.11 shows the toughening effect of EPR-g-MAH and SEBS-g-MAH on PP/polyamide 6 blends. For PP-rich blends, the notched impact strength of the toughened blends is usually less than 20 kJ m2 , however, for polyamide-rich blends around 100 kJ m2 notched impact strength was obtained. For polypropylene/polyamide/grafted rubber ternary blends, TEM was proved to be a good tool to study their phase morphology. Without grafting, EPR (217) or SEBS (218) formed dispersed particles together with the minor component of the blends separately; however, after grafting with maleic anhydride they could encapsulate the dispersed minor component of the blends (214,217–221).

Figure 2.11 Izod impact strength of polyamide 6/polypropylene blends modified with 20% EPR-gMA or SEBS-g-MA. The composition shown on the abscissa is the percentage of polypropylene in the blend on a rubber-free basis. The dashed lines correspond to the properties of unmodified binary blends. (From Reference 214 with permission from Elsevier.)

50

Polyolefin Blends

2.7 CONCLUSIONS The miscibility between polyolefins is dependent on molecular structure, blend composition, and mixing temperature. Phase map is needed to characterize their miscibility. The binary blends between polyolefins and polar polymers are immiscible and have large interfacial tension. They show weak interfacial adhesion, poor mechanical properties, and trend to coalesce in the melt state. Compatibilizing agents are needed in these binary blends, which can decrease interfacial tension, prevent phase coalescence, and improve properties largely. Compatibilizing agents can be grafted or biblock copolymers of the two blend components, or can be prepared by functionalizing one blend component, which can graft onto the other component to form grafted copolymers through the functional groups in mixing. To be effective, these compatibilizing agents must stay at phase interface. Both their molecular structures and melt mixing procedure can affect their diffusion behavior to the phase interface.

NOMENCLATURE G U DUm P V Vs T Tc S DS DSm Suniv H DHm kb V N N1 N2 x1 x2 c11 c22 c12 d d1

Gibbs free energy Internal energy Internal energy change in mixing Pressure Volume Volume of segments of polymers Absolute temperature Critical temperature Entropy Entropy change Entropy change of mixing Entropy of the universe Enthalpy Enthalpy change in mixing Boltzmann constant Number of configuration Number of molecule Number of molecule 1 Number of molecule 2 Mole fraction of molecule 1 Mole fraction of molecule 2 Interaction energy between molecule 1 Interaction energy between molecule 2 Interaction energy between molecules 1 and 2 Solubility parameter Solubility parameter of molecule 1


d2 f1 f2 x M1 M2 g g0 g1 ecoh M Ke

51

Solubility parameter of molecule 2 Volume fraction of molecule 1 Volume fraction of molecule 2 Interaction parameter Number of segments of molecule 1 Number of segments of molecule 2 Surface tension Surface tension at zero absolute temperature Surface tension of molecules with infinite molecular weight Cohesive energy density Molecular weight Molecular weight–surface tension constant

REFERENCES 1. E. W. Fawcett, R. O. Gibson, M. W. Perrin, J. G. Patton, and E. G. Williams, British Patent 471,590 (1937). 2. K. Ziegler, E. Holzkamp, H. Breil, and H. Martin, Angew. Chem., 67, 541 (1955). 3. J. P. Hogan and R. L. Banks, U.S. Patent 2,825,721 (1958). 4. G. Natta, P. Pino, P. Corradini, F. Danusso, E. Mantica, G. Mazzanti, and G. Moraglio, J. Am. Chem. Soc., 77, 1708 (1955). 5. G. Natta, J. Polym. Sci., 16, 143 (1955); Makromol. Chem., 16, 213 (1955). 6. G. Natta, P. Pino, G. Mazzanti, P. Corradini, and U. Giannini, Rend., Accad. Naz. Lin., 19, 397 (1955). 7. F. P. Reding, J. Polym. Sci., 21, 547 (1956). 8. A. Montagna and J. C. Floyd, MetCon’93, Houston, May 26–28, 1993. 9. J. C. Stevens, F. J. Timmers, D. R. Wilson, G. F. Schmidt, P. N. Nickias, R. K. Rosen, G. W. Knight, and S. Y. Lai, Eur. Patent 416,815-A2 (1991). 10. S. Y. Lai, D. R. Wilson, G. W. Knight, J. C. Stevens, and S. Chum, U.S. Patent 5,272,236 (1993). 11. D. J. Arriola, E. M. Carnahan, P. D. Hustad, R. L. Kuhlman, and T. T. Wenzel, Science, 312, 714 (2006). 12. W. Kaminsky and M. Arndt, Adv. Polym. Sci., 127, 143 (1997). 13. Y. Nishi, M. Ohshima, T. Kohara, T. Matsui, and T. Natsuume, U.S. Patent 5,599,882 (1997). 14. V. Dragutan and R. Streck, Cataytic Polymerization of Cycloolefins, Elsevier Science Ltd., New York, 2000. 15. L. A. Utracki, Introduction to polymer blends, in: Polymer Blends Handbook, Vol. 1, L. A. Utracki (ed.), Kluwer Academic Publishers, Boston, 2002. 16. A. Parkes, English Patent, 11,147 (1846). 17. W. H. Smyers and D. W. Young, U.S. Patent 2,610,962 (1952). 18. L. E. Daly, U.S. Patent 2,439,202 (1948). 19. J. D. Rourk, CA Patent 529, 621 (1956). 20. G. H. Fremon and W. N. Stoops, DE Patent 1,040,249 (1958). 21. T. G. Heggs and G. P. Lee, GB Patent 817,141 (1959). 22. NL Patent 6,517,240 (1966). 23. J. W. Gibbs, Trans. Conn. Acad. Sci., 3, 228 (1876). 24. K. Denbigh, The Principles of Chemical Equilibrium: With Applications in Chemistry and Chemical Engineering, 2nd edition, Cambridge University Press, Cambridge, 1966.

52

Polyolefin Blends

25. R. Clausius, Abhandlungen u¨ber die mechanische wa¨rmetheorie, Vol. 2, Vieweg, Braunschweig, 1867, p. 44. 26. T. L. Hill, An Introduction to Statistical Thermodynamics, Addison-Wesley Publishing Co., Reading, 1960. 27. J. J. Van Laar, Sechs Vortrage uber das Thermodynamische Potential, Vieweg, Braunschweig, 1906. 28. J. J. Van Laar and R. Lorenz, Z. Anorg. Allgem. Chem., 146, 42 (1925). 29. J. H. Hildebrand, J. Am. Chem. Soc., 38, 1452 (1916). 30. J. H. Hildebrand, Proc. Nat. Acad. Sci., U.S.A. 13, 267 (1927). 31. G. Scatchard, Chem. Rev., 8, 321 (1931). 32. J. H. Hildebrand, The Solubility of Non-Electrolytes, Reinhold, New York, 1936. 33. J. H. Hildebrand and R. L. Scott, The Solubility of Nonelectrolytes, 3rd edition, Reinhold, New York, 1950. 34. F. Dolezalek, Z. Phys. Chem., 64, 727 (1908). 35. J. E. Lennard-Jones and A. F. Devonshire, Proc. Roy. Soc., A163, 53 (1937). 36. I. Prigogine and G. Garikian, Physica, 16, 239 (1950). 37. H. C. Longuet-Higgins, Proc. R. Soc., A205, 247 (1951). 38. I. Prigogine, The Molecular Theory of Solution, North-Holland Publishing Co., Amsterdam, 1957. 39. K. H. Meyer, Z. Physik. Chem., B44, 383 (1939). 40. M. L. Huggins, J. Phys. Chem., 46, 151 (1942). 41. P. J. Flory, J. Chem. Phys., 10, 51 (1942). 42. P. J. Flory, J. Chem. Phys., 12, 425, 1944. 43. P. J. Flory, Principles of Polymer Chemistry, Cornell University Press, New York, 1953. 44. RJ. Scott, J. Chem. Phys., 17, 268 (1949). 45. R. Koningsveld, W. H. Stockmayer, and E. Nies, Polymer Phase Disgrams, Oxford University Press, Oxford, 2001. 46. L. P. McMaster, Macromolecules, 6, 760 (1973). 47. T. Young, Philos. Trans. R. Soc., 95, 65 (1805). 48. N. K. Adam, The Physics and Chemistry of Surfaces, Oxford University Press, London, 1941. 49. W. Thomson, Philos. Mag., 42, 368 (1871). 50. J. W. Strutt, Proc. Lond. Math. Soc., 10, 4 (1878). 51. J. W. Strutt, Proc. R. Soc., 29, 71 (1879). 52. C. L. Yaws, X. Y. Lin, L. Bu, and S. Nijhawan, Surface tension, in: Chemical Properties Handbook, C. L. Yaws (ed.), McGraw Hill, New York, 1999. 53. J. Bicerano, Prediction of Polymer Properties, 2nd edition, Marcel Dekker, New York, 2002. 54. D. W. van Krevelen, Properties of Polymers, 3rd edition, Elsevier, Amsterdam, 1990. 55. S. H. Wu, Polymer Interface and Adhesion, Marcel Dekker, New York, 1982. 56. S. H. Wu, Surface and interfacial tensions of polymers, oligomers, plasticizers, and organic pigments, in Polymer Handbook, J. Brandrup, E. H. Immergut, and E. A. Grulke (eds.), Wiley, New York, 1999. 57. T. J. Menke, Z. Funke, R. D. Maier, and J. Kressler, Macromolecules, 33, 6120 (2000). 58. G. W. Bender and G. L. Gaines, Macromolecules, 3, 128 (1970). 59. J. C. Maxwell, Surface tension, in: Encyclopedia Britannica, Samuel L. Hall, New York, 1878. 60. G. Quincke, Ann. Phys., 137, 1 (1870). 61. J. M. Andreas, E. A. Hauser, and W. B. Tucker, J. Phys. Chem., 42, 1001 (1938). 62. P. J. Yoon and J. L. White, J. Appl. Polym. Sci., 51, 1515 (1994). 63. G. Macaubas and P. H. Pierin, Polym. Eng. Sci., 43, 670 (2003). 64. G. Palmer and N. R. Demarquette, Polymer, 46, 8169 (2005).


53

65. J. T. Koberstein, Interfacial properties, in: Encyclopedia of Polymer Science and Engineering, J. I. Kroschwitz (ed.), Wiley Interscience, New York, 1987. 66. N. R. Demarquette, Int. Mater. Rev., 48, 247 (2003). 67. E. D. Carlson, M. T. Krejchi, C. D. Shah, T. Terakawa, R. M. Waymouth, and G. G. Fuller, Macromolecules, 31, 5343 (1998). 68. R. Chen, M. R. Xie, Q. Wu, and S. G. Lin, J. Polym. Sci. A Polym. Chem., 38, 411 (2000). 69. Y. Hu, M. T. Krejchi, C. D. Shah, C. L. Myers, and R. M. Waymouth, Macromolecules, 31, 6908 (1998). 70. E. D. Carlson, G. G. Fuller, and R. M. Waymouth, Macromolecules, 32, 8094 (1999). 71. W. Wiyatno, J. A. Pople, A. P. Gast, R. M. Waymouth, and G. G. Fuller, Macromolecules, 35, 8488 (2002). 72. M. J. Hill, P. J. Barham, and A. Keller, Polymer, 33, 2530 (1992). 73. M. J. Hill, P. J. Barham, and J. V. Ruiten, Polymer, 34, 2975 (1993). 74. M. J. Hill, R. L. Morgan, and P. J. Barham, Polymer, 38, 3003 (1997). 75. R. L. Morgan, M. J. Hill, and P. J. Barham, Polymer, 40, 337 (1999). 76. T. Hameed and I. A. Hussein, Polymer, 43, 6911 (2002). 77. I. A. Hussein, Macromolecules, 36, 2024 (2003). 78. T. Hameed and V Hussein, Macromol. Mater. Eng., 289, 198 (2004). 79. I. A. Hussein and M. C. Williams, Polym. Eng. Sci., 44, 660 (2004). 80. G. D. Wignall, J. D. Londono, J. S. Lin, R. G. Alamo, M. J. Galante, and L. Mandelkern, Macromolecules, 28, 3156 (1995). 81. R. G. Alamo, W. W. Graessley, R. Krishnamoorti, D. J. Lohse, J. D. Londono, L. Mandelkern, F. C. Stehling, and G. D. Wignall, Macromolecules, 30, 561 (1997). 82. G. D. Wignall, R. G. Alamo, J. D. Londono, L. Mandelkern, M. H. Kim, J. S. Lin, and G. M. Brown, Macromolecules, 33, 551 (2000). 83. K. Tashiro, R. S. Stein, and S. L. Hsu, Macromolecules, 25, 1801 (1992). 84. K. Tashiro, M. M. Satkowski, R. S. Stein, Y. Li, B. Chu, and S. L. Hsu, Macromolecules, 25, 1809 (1992). 85. K. Tashiro, M. Izuchi, M. Kobayashi, and R. S. Stein, Macromolecules, 27, 1221 (1994). 86. K. Tashiro, M. Izuchi, M. Kobayashi, and R. S. Stein, Macromolecules, 27, 1234 (1994). 87. B. S. Tanem and A. Stori, Polymer, 42, 4309 (2001). 88. A. Y. Coran and R. Patel, U.S. Patent 4,297,453 (1981). 89. A. Y. Coran and R. Patel, U.S. Patent 4,310,638 (1982). 90. E. Nolley, J. W. Barlow, and D. R. Paul, Polym. Eng. Sci., 20, 364 (1980). 91. D. W. Bartlett, J. W. Barlow, and D. R. Paul, J. Appl. Polym. Sci., 27, 2351 (1982). 92. N. K. Kalfoglou, Angewandte Makromolekulare Chemie, 129, 103 (1985). 93. D. J. Lohse, Polym. Eng. Sci., 26, 1500 (1986). 94. L. Dorazio, C. Mancarella, E. Martuscelli, and G. Sticotti, J. Mater. Sci., 26, 4033 (1991). 95. E. Martuscelli, C. Silvestre, and G. Abate, Polymer, 23, 229 (1982). 96. C. Y. Chen, W. M. Z. W. Yunus, H. W. Chiu, and T. Kyu, Polymer, 38, 4433 (1997). 97. M. Seki, H. Nakano., S. Yamauchi, J. Suzuki, and Y. Matsushita, Macromolecules, 32, 3227 (1999). 98. K. H. Nitta, Y. W. Shin, H. Hashiguchi, S. Tanimoto, and M. Terano, Polymer, 46, 965 (2005). 99. A. R. Kamdar, Y. S. Hu, P. Ansems, S. P. Chum, A. Hiltner, and E. Baer, Macromolecules, 39, 1496 (2006). 100. P. S. Chum, W. J. Kruper, and M. J. Guest, Adv. Mater., 12, 1759 (2000). 101. V. B. F. Mathot and M. J. F. Pijpers, J. Appl. Polym. Sci., 39, 979 (1990).

54

Polyolefin Blends

102. F. Defoor, G. Groeninckx, H. Reynaers, P. Schouterden, and B. Van der Heijden, Macromolecules, 26, 2575 (1993). 103. K. Sehanobish, R. M. Patel, B. A. Croft, S. P. Chum, and C. I. Kao, J. Appl. Polym. Sci., 51, 887 (1994). 104. S. Bensason, J. Minick, A. Moet, S. Chum, A. Hiltner, and E. Baer, J. Polym. Sci. B, Polym. Phys., 34, 1301 (1996). 105. A. L. N. Da Silva, M. I. B. Tavares, D. P. Politano, F. M. B. Coutinho, and M. C. G. Rocha, J. Appl. Polym. Sci., 66, 2005 (1997). 106. F. Stricker, R. D. Maier, Y. Thomann, and R. Mulhaupt, Kunststoffe-Plast Europe, 88, 527 (1998). 107. T. McNally, P. McShane, G.M Nally, W. R. Murphy, M. Cook, and A. Miller, Polymer, 43, 3785 (2002). 108. M. Yamaguchi, H. Miyata, and K. H. Nitta, J. Appl. Polym. Sci., 62, 87 (1996). 109. Y. Thomann, J. Suhm, R. Thomann, G. Bar, R. D. Maier, and R. Mu¨lhaupt, Macromolecules, 31, 5441 (1998). 110. M. Yamaguchi, K. Suzuki, and H. Miyata, J. Polym. Sci. B, Polym. Phys., 37, 701 (1999). 111. M. Yamaguchi and K. Nitta, Polym. Eng. Sci., 39, 833 (1999). 112. M. Yamaguchi and H. Miyata, Macromolecules, 32, 5911 (1999). 113. P. M. Cham, T. H. Lee, and H. Marand, Macromolecules, 27, 4263 (1994). 114. Z. Bartczak, A. Galeski, and M. Pracella, J. Appl. Polym. Sci., 54, 1513 (1994). 115. A. Siegmann, J. Appl. Polym. Sci., 27, 1053 (1982). 116. C. C. Hsu and P. H. Geil, Polym. Eng. Sci., 27, 1542 (1987). 117. M. S. Lee and S. A. Chen, J. Polym. Sci. C, Polym. Lett., 25, 37 (1987). 118. C. C. Chen and J. L. White, Polym. Eng. Sci., 33, 923 (1993). 119. K. Min, J. L. White, and J. F. Fellers, J. Appl. Polym. Sci., 29, 2117 (1984). 120. I. Banik, P. J. Carreau, and H. P. Schreiber, J. Polym. Sci. B, Polym. Phys., 42, 2545 (2004). 121. C. W. Macosko, P. Gue´gan, A. K. Khandpur, A. Nakayama, P. Marechal, and T. Inoue, Macromolecules, 29, 5590 (1996). 122. U. Sundararaj and C. W. Macosko, Macromolecules, 28, 2647 (1995). 123. J. M. Willis, B. D. Favis, and J. Lunt, Polym. Eng. Sci., 30, 1073 (1990). 124. B. R. Liang, J. L. White, J. E. Spruiell, and B. C. Goswami, J. Appl. Polym. Sci., 28, 2011 (1983). 125. K. Min, J. L. White, and J. F. Fellers, Polym. Eng. Sci., 24, 1327 (1984). 126. C. C. Chen, E. Fontan, K. Min, and J. L. White, Polym. Eng. Sci., 28, 69 (1988). 127. A. Valenza, L. Gallo, G. Spadaro, E. Calderaro, and D. Acierno, Polym. Eng. Sci., 33, 1336 (1993). 128. Y. G. Cho and M. R. Kamal, Polym. Eng. Sci., 42, 2005 (2002). 129. M. Baer and E. H. Hankey, U.S. Patent 3,085,082 (1963). 130. E. H. Hankey and A. Kianpour, U.S. Patent 3,145,187 (1964). 131. M. Baer and E. H. Hankey, U.S. Patent 3,399,155 (1968). 132. D. Q. Khang, L. Starke, and Z. Funke, J. Macromol. Sci. Pure Appl. Chem., A33, 1973 (1996). 133. N. G. Gaylord, U.S. Patent 3,485,777 (1969). 134. N. G. Gaylord, in: Copolymers, Polyblends and Composites, ACS Advances in Chemistry Series, Vol. 142, ACS, Washington, 1975, p.76. 135. L. C. Anderson, D. A. Roper, and J. K. Rieke, J. Polym. Sci., 43, 423 (1960). 136. C. E. Locke and D. R. Paul, J. Appl. Polym. Sci., 17, 2597 (1973); 17, 2791 (1973). 137. A. Kishimoto, S. Hirata, and H. Ueno, DE Patent 1,950,479 (1970). 138. W. L. Carrick, J. Polym. Sci., 8, 215 (1970). 139. D. Heikens and W. Barensten, Polymer, 18, 69 (1977). 140. W. M. Barentsen and D. Heikens, Polymer, 14, 579 (1973). 141. W. M. Barentsen, D. Heikens, and P. Piet, Polymer, 15, 119 (1974).


55

142. D. Heikens, N. Hoen, W. Barentsen, P. Piet, and H. Ladan, J. Polym. Sci. Polym. Symposia, 62, 309 (1978). 143. S. D. Sjoerdsma, J. Dalmolen, A. C. A. M. Bleijenberg, and D. Heikens, Polymer, 21, 1469 (1980). 144. C. R. Lindsey, D. R. Paul, and J. W. Barlow, J. Appl. Polym. Sci., 26, 1 (1981). 145. R. Fayt, R. Jerome, and P. Teyssie, J. Polym. Sci. B, Polym. Phys., 19, 1269 (1981). 146. R. Fayt, R. Jerome, and P. Teyssie, J. Polym. Sci. C, Polym. Lett., 19, 79 (1981). 147. R. Fayt, R. Jerome, and P. Teyssie, J. Polym. Sci. B, Polym. Phys., 20, 2209 (1982). 148. R. Fayt, R. Jerome, and P. Teyssie, Makromol. Chem., 187, 837 (1986). 149. R. Fayt, R. Jerome, and P. Teyssie, J. Polym. Sci. C Polym. Lett., 24, 25 (1986). 150. I. Fortelny, J. Mikesova, J. Hromadkova, V. Hasova, and Z. Horak, J. Appl. Polym. Sci., 90, 2303 (2003). 151. M. Tasdemir and H. Yildirim, J. Appl. Polym. Sci., 83, 2967 (2002). 152. C. Harrats, R. Fayt, and R. Jerome, Polymer, 43, 863 (2002). 153. J. S. Wu, B. H. Guo, C. M. Chan, J. X. Li, and H. S. Tang, Polymer, 42, 8857 (2001). 154. I. Luzinov, C. Pagnoulle, and R. Je´roˆme, Polymer, 41, 3381 (2000). 155. J. X. Li, C. M. Chan, B. H. Gao, and J. S. Wu, Macromolecules, 33, 1022 (2000). 156. H. F. Guo, S. Packirisamy, R. S. Mani, C. L. Aronson, N. V. Gvozdic, and D. J. Meier, Polymer, 39, 2495 (1997). 157. T. Li, V. A. Topolkaraev, A. Hiltner, E. Baer, X. Z. Ji, and R. P. Quirk, J. Polym. Sci. B, Polym. Phys., 33, 667 (1995). 158. Y. Gao, H. L. Huang, Z. H. Yao, D. Shi, Z. Ke, and J. H. Yin, J. Polym. Sci. B, Polym. Phys., 41, 1837 (2003). 159. Z. Wang, C. M. Chan, S. H. Zhu, and J. R. Shen, Polymer, 39, 6801 (1998). 160. A. Halimatudahliana, H. Ismail, and M. Nasir, Polym. Test., 21, 163 (2002). 161. D. Hlavata, Z. Horak, F. Lednicky, J. Hromadkova, A. Pleska, and Y. V. Zanevskii, J. Polym. Sci. B Polym. Phys., 39, 931 (2001). 162. P. H. P. Macaubas and N. R. Demarquette, Polymer, 42, 2543 (2001). 163. D. Hlavata, Z. Horak, J. Hromadkova, F. Lednicky, and A. Pleska, J. Polym. Sci. B, Polym. Phys., 37, 1647 (1999). 164. G. Radonjic, V. Musil, and I. Smit, J. Appl. Polym. Sci., 69, 2625 (1998). 165. I. Smit, G. Radonjic, and D. Hlavata, Eur. Polym. J., 40, 1433 (2004). 166. G. Radonjic and V. Musil, Die Angewandte Makromolekulare Chemie, 251, 141 (1997). 167. L. Del Giudice, R. E. Cohen, G. Attalla, and F. Bertinotti, J. Appl. Polym. Sci., 30, 4305 (1985). 168. G. X. Xu and S. G. Lin, Polymer, 37, 421 (1996). 169. J. Kim, J. Kwak, and D. Kim, Polym. Adv. Technol., 14, 58 (2003). 170. Z. Li, Y. C. Ke, and Y. L. Hu, J. Appl. Polym. Sci., 93, 314 (2004). 171. M. F. Diaz, S. E. Barbosa, and N. J. Capiati, Polymer, 46, 6096 (2005). 172. M. F. Diaz, S. E. Barbosa, and N. J. Capiati, J. Polym. Sci. B, Polym. Phys., 42, 452 (2004). 173. R. M. Nowak and G. D. Jones, U.S. Patent 3,177,269 (1965). 174. G. D. Jones and R. M. Nowak, U.S. Patent 3,177,270 (1965). 175. R. M. Nowak, U.S. Patent 3,270,090 (1966). 176. R. G. Armstrong, U.S. Patent 3,373,222 (1967). 177. H. Terada, T. Kono, M. Kawajima, and A.Hasegawa, Japan Patent 46,038,022 (1971). 178. T. Ozeki, T. Kono, and M. Kawajima, Japan Patent 46,038,023 (1971). 179. W. Nielinger, R. Dhein, K. Schneider, and P. Tacke, DE Patent 2,426,671 (1975). 180. J. H. Davis, GB Patent 1,403,797 (1975).

56

Polyolefin Blends

181. H. W. Starkweather, DE Patent 2,551,023 (1976). 182. F. Ide and A. Hasegawa, J. Appl. Polym. Sci., 18, 963 (1974). 183. D. Braun and U. Eisenlohr, Angewandte Makromolekulare Chemie, 58–59, 227 (1977). 184. A. Lahor, M. Nithitanakul, and B. P. Grady, Eur. Polym. J., 40, 2409 (2004). 185. A. Valenza, L. Gallo, G. Spadaro, and E. Calderaro, Polym. Eng. Sci., 33, 1336 (1993). 186. S. Filippi, V. Chiono, G. Polacco, M. Paci, L. Minkova, and P. Magagnini, Macromol. Chem. Phys., 203, 1512 (2002). 187. A. Valenza, G. Spadaro, E. Calderaro, and D. Acierno, Polym. Eng. Sci., 33, 845 (1993). 188. Y. J. Yoo, C. H. Park, S. G. Lee, K. Y. Choi, D. S. Kim, and J. H. Lee, Macromol. Chem. Phys., 206, 878 (2005). 189. M. Mehrabzadeh and M. Kamal, Polym. Eng. Sci., 44, 1152 (2004). 190. B. D. Favis, Polymer, 35, 1552 (1994). 191. P. Leewajanakul, R. Pattanaolarn, J. W. Ellis, M. Nithitanakul, and B. P. Grady, J. Appl. Polym. Sci., 89, 620 (2003). 192. G. Fairley and R. E. Prudhomme, Polym. Eng. Sci., 27, 1495 (1987). 193. A. Valenza, L. Gallo, G. Spadaro, and E. Calderaro, Polym. Eng. Sci., 33, 1336 (1993). 194. A. Valenza, G. Spadaro, E. Calderaro, and D. Acierno, Polym. Eng. Sci., 33, 845 (1993). 195. G. Spadaro, D. Acierno, C. Dispenza, E. Calderaro, and A. Valenza, Radiat. Phys. Chem., 48, 207 (1996). 196. S. S. Wu, B. O. Chi, H. Yan, and J. Shen, J. Appl. Polym. Sci., 99, 2029 (2006). 197. S. J. Park, B. K. Kim, and H. M. Jeong, Eur. Polym. J., 26, 131 (1990). 198. A. Valenza and D. Aciermo, Eur. Polym. J., 30, 1121 (1994). 199. S. Jose, B. Francis, S. Thomas, and J. Karger-Kocsis, Polymer, 47, 3874 (2006). 200. S. Jose, S. V. Nair, S. Thomas, and J. Karger-Kocsis, J. Appl. Polym. Sci., 99, 2640 (2006). 201. D. X. Li, D. M. Jia, and P. Zhou, J. Appl. Polym. Sci., 93, 420 (2004). 202. N. Zeng, S. L. Bai, C. G’Sell, J. M. Hiver, and Y. M. Mai, Polym. Int., 51, 1439 (2002). 203. J. D. Tucher, S. Lee, and R. L. Einsporn, Polym. Eng. Sci., 40, 2577 (2000). 204. X. M. Zhang, Z. H. Yin, T. H. Na, and J. H. Yin, Polymer, 38, 5905 (1997). 205. X. M. Zhang, G. Li, J. S. Li, and J. H. Yin, Angew. Makromol. Chem., 248, 189 (1997). 206. X. M. Zhang and J. H. Yin, Polym. Eng. Sci., 37, 197 (1997). 207. P. Van Gheluwe, B. D. Favis, and J. P. Chalifoux, J. Mater. Sci., 23, 3910 (1988). 208. P. Van Gheluwe and B. D. Favis, Polym. Mater. Sci. Eng., 58, 966 (1988). 209. J. M. Willis, V. Caldas, and B. D. Favis, J. Mater. Sci., 26, 4742 (1991). 210. J. M. Willis, B. D. Favis, and C. Lavallee, J. Mater. Sci., 28, 1749 (1993). 211. M. J. Modic and L. A. Pottick, Plast. Eng., 47, 37 (1991). 212. J. Roesch and R. Muelhaupt, Makromol. Chem. Rapid Commun., 14, 503 (1993). 213. M. Tasdemir, J. Appl. Polym. Sci., 89, 3485 (2003). 214. A. Gonza´lez-Montiel, H. Keskkula, and D. R. Paul, Polymer, 36, 4587 (1995). 215. S. L. Bai, G. T. Wang, J. M. Hiver, and C. G’Sell, Polymer, 45, 3063 (2004). 216. H. Z. Liu, T. X. Xie, L. L. Hou, Y. C. Ou, and G. S. Yang, J. Appl. Polym. Sci., 99, 3300 (2006). 217. J. Ro¨sch, Polym. Eng. Sci., 35, 1917 (1995). 218. A. N. Wilkinson, L. Laugel, M. L. Clemens, V. M. Harding, and M. Marin, Polymer, 40, 4971 (1999). 219. B. Lu and T. C. Chung, Macromolecules, 32, 2525 (1999). 220. G. M. Kim, G. H. Michler, J. Ro¨sch, and R. Mu¨thaupt, Acta Polym., 49, 88 (1998). 221. J. Ro¨sch and R. Mu¨lhaupt, Polym. Bull., 32, 697 (1994).

Part II

Polyolefin/Polyolefin Blends

Chapter

3

Miscibility, Morphology, and Properties of Polyethylene Blends Robert A. Shanks1

3.1 INTRODUCTION Polyethylenes (PE) are a group of polymers having the same chemical structure, but differing in molecular architecture. They have subambient glass transition temperatures so the amorphous phase is rubbery. Several additional subambient relaxation temperatures have been detected depending on the branching characteristics. Molecular flexibility contributes yield, drawing, and toughness. Strength and moderate thermal resistance are derived from crystallinity. Weak intermolecular forces means that interactions and crystallinity result from symmetry and close packing. Branching restricts symmetry and close packing. Branch points are excluded from the crystals. Polyethylenes have many applications and through their price and performance they rank as commodity polymers. They are used in films, sheets, fibers, pipes, and many types of moldings. Polyethylene products are prepared by extrusion, injection molding, thermoforming, and rotational casting. A particular polyethylene must be chosen for the molding process, product properties, and longer term product performance. Even with the large range of polyethylenes available, it is difficult to choose a polyethylene with the best properties. Blending is usually the solution to property optimization. More polyethylenes are manufactured as blends than they are as the pure polymers. Most polyethylene blends are with other polyethylenes, though some blends are formed with ethylene copolymers with polar monomers such as vinyl acetate (VAc), methyl acrylate (MA), butyl acrylate (BA), glycidyl methacrylate

1

School of Applied Sciences, RMIT University, GPO Box 2476V, Melbourne, Vic 3001, Australia


59

60

Polyolefin Blends

(GMA), acrylic acid (AA), or grafted with maleic anhydride (MAn). Hydrocarbon waxes, sometimes highly branched, are other compounds that may be blended with polyethylenes. Polyethylene in its ideal form is a linear polymer (LPE) that should correctly be called poly(methylene), Table 3.1. LPE can reach high crystallinity (Xc > 0:90) due to its compact symmetrical molecules. It has low strength dispersive molecular forces. The melting temperature (Tm ) of LPE is variable but usually about 135 C. The equilibrium melting temperature of LPE has been measured many times and is about 145 C. LPE is insoluble at ambient temperatures, but it can be dissolved in hydrocarbon, alkylaromatics, and halogenated hydrocarbons at temperatures approaching its melting temperature. Two crystalline morphologies have been observed; the chain folded form is most common and forms spontaneously from

Table 3.1

Properties of Linear Polyethylene.

Property

Value

Comment

Monomer

CH2 CH2

Polymerization

Ziegler–Natta, supported metal oxides such as Philips, Unipol; and metallocene 120 to 60 C 137 C

Symmetrical and compact for close packing, dispersive forces; no other intermolecular interactions, polarity, or bulky steric groups Stereospecific chain formation required; weak interaction/complexation with any initiators/catalysts; metal coordination complexes required

Glass transition (Tg ) Melting temperature (Tm )

Crystallinity

0.60–0.95

Molar mass

Wide range, 103–107 g mol1

Density

0.90–0.96 g cm3

Stability

No chemical groups to absorb ultraviolet light and no reactive tertiary hydrogens

An elastomer or viscous liquid at ambient folded chain crystals; strength properties derived from crystallinity; low thermal resistance. Extended chain crystals can melt at about 145 C Properties dependent on crystallinity, with amorphous phase providing flow, flexibility, and toughness. Kinetically stable folded chain crystals; thermodynamically stable extended chain crystals can be formed by gel drawing Molar mass must be high since intermolecular forces are weak; all values are available from waxes to ultrahigh molar mass. Dependant on crystallinity; often HDPE contains some butene comonomer to avoid brittleness The ideal polyethylene structure should be very stable; however, some unsaturated end groups, impurity oxygen containing groups, and catalyst residues accelerate degradation

Chapter 3 Miscibility, Morphology, and Properties of Polyethylene Blends

61

Figure 3.1 Density, crystallinity, and polyethylene type scale.

the melt, being kinetically favored. The chain extended form is thermodynamically favored, but can be formed by only flow elongation of molecules, typically by gel spinning. Significant characteristics of LPE are molar mass (M) and molar mass distribution. Molar mass and morphology are the only variables available to provide required properties with LPE. LPE is usually called high density polyethylene (HDPE) because of the high density arising from high crystallinity. Commercial HDPE often contains a small proportion of comonomer, typically 1-butene, to decrease its brittleness. The comonomer will lower the density, crystallinity, and melting temperature; Tm may be in the range 130–135 C. A schematic showing polyethylene type, density, melting temperature, and spherulitic crystal ranges is shown in Fig. 3.1. Branching of polyethylenes provides the second dimension, after molar mass, with which to control properties, Tables 3.2 and 3.3. Branching of polyethylenes is a complex topic; in this review it will be treated starting with the ideal copolymer structures formed with the new metallocene catalysts. Branched polyethylenes (BPE) provide increased toughness though decreased modulus and strength compared with LPE. Branches are obtained by copolymerization with 1-alkenes, such as Table 3.2 Properties of Typical Branched Polyethylenes. Properties

VLDPE-B VLDPE-O

Comonomer Butene Catalyst type M MFI (dg min1) 27.0 0.901 Density (g cm3) Mw 58,000 Mw/Mn 2.65 Comonomer 6.3 Content (%mol) Tm ( C) 92.7 Tc( C) 76.6

VLDPE-O(ZN)

LDPE-highM LDPE-lowM

Octene S 1.0 0.908 96,700 2.86 2.4

Octene ZN 1.0 0.912 120,000 3.80 4.2

P 7.0 0.919 474,000 23.3

P 22.0 0.918 89,000 4.4

105.4 90.3

123.0 100.5

106.5 87.8

103.9 87.1

Data were obtained from data sheets published by the manufacturer. ZN ¼ Ziegler–Natta catalyst, S ¼ Constrained geometry single-site catalyst, M ¼ Metallocene, and P ¼ Peroxide. Crystallisation (Tc) and melting (Tm) temperature were obtained using DSC at scanning rates of 10 C min1.

62

Polyolefin Blends

Table 3.3

Types of Polyethylenes and Their Properties.

Polyethylene

Structure

Comment

HDPE or linear polyethylene, poly(methylene) LDPE

Linear carbon chain

See Table 3.1

Random branching with either short (average C4) or long (many C) branches; dependant on autoclave or tubular reactor type

LLDPE

Short branches introduced by a comonomer, usually butene, hexene, or octane; with many others; comonomer amount and structure limited with ZN catalysts

mLLDPE

Metallocene-catalyzed PE with more uniform short branch distribution; increased monomer content and structure available Increased comonomer content giving lower density and lower Tm. ZN and metallocene type available Highest comonomer contents, lowest Tm ; crystallization can continue below ambient; can be dissolved at ambient

Many short branches limit crystallinity and lamella thickness (reduce Tm ); long branches provide nonNewtonian rheology and melt strength Short branch content is controlled by comonomer amount, distribution of comonomer not controlled. Slurry (or solution) and gas phase variations. Bimodal crystal distributions and possibly two liquid phases Structure and property combinations highly specialized; new applications possible with a wide range of molar mass and branching Soft PE that can be blended with stronger PE to improve toughness, heat sealing, and elasticity Elastomeric properties that require cross-linking if used alone. Often used in blends

VLDPE

ULDPE

1-butene, 1-hexene, and 1-octene. Polar polyethylenes are formed by copolymerization with VAc, MA, BA, GMA, or AA. Grafting of MAN can be performed in a solution reaction or in an extruder. The properties of BPE are determined by the proportion of branches, or more specifically the methylene sequence length (MSL) between branches, in addition to molar mass and its distribution. When the comonomers are unevenly distributed along polyethylene molecules, MSL will be a distribution or this can be expressed and a comonomer distribution. A two-dimensional contour chart must define the architecture of BPE, with molar mass distribution and comonomer distribution axes. The comonomer distribution can be intermolecular, such that some molecules have mainly linear structure while others have many branches. The comonomer distribution may be intramolecular, where the density of comonomer units changes along the length of a molecule, so that segments


63

Figure 3.2 PE branch distributions, intermolecular branch distribution (top and middle), and intramolecular branch distribution (bottom).

of a molecule may be mainly linear, while other segments have many comonomers or branches (1). Inter- and intramolecular branch distributions are illustrated in Fig. 3.2. The morphology is independent of branch length, that is, comonomer structure, since the branches are excluded from crystals. BPE polymerized by Ziegler–Natta, Philips, Unipol, or similar multisite catalysts show increased dispersion of comonomers. The distributions are typically bimodal. Gas phase and solution polymerizations differ in distribution, with solution or slurry prepared BPE generally showing broader distributions. Bimodal distributions of branches result in bimodal distributions of lamella thickness and hence melting temperatures. Bimodal BPE behave like blends though there is poor control over the structure and composition of each component. The amount of braches that can be included is limited to low amounts compared with single-site-catalyzed BPE where highly branched elastomeric BPE can be produced. There is controversy as to whether the components are miscible in the liquid phase and if they become immiscible at what difference in branch composition and at what temperature. The phase diagram is proposed to be of an upper critical solution temperature type. The BPE described are prepared by copolymerization and only short branches are present. PBE with both short and long branches are prepared from radical polymerization in the traditional high pressure process. These PE are called low density polyethylene (LDPE). A molecular model of a conformation of a typical LDPE is shown in Fig. 3.3. LDPE with different proportions of short and long branches are formed using autoclave or tubular reactors. There are many short branches with average

Figure 3.3 Molecular model of typical LDPE with 1000 carbons, 3 long branches, and 30 short branches.

64

Polyolefin Blends

length of four carbons formed by intramolecular chain transfer by the growing chain. There are relatively few long branches, about 10% of the short branches, formed by intermolecular chain transfer by a growing chain to a terminated chain. The special properties of LDPE compared with other PBE are brought about by the long branches. Long branches contribute to melt rheology during processing by providing increased shear thinning and melt strength. The branch distribution in LDPE is random because of the mechanism of formation, and the overall distribution demonstrated by melting is narrow compared with most multisite-catalyzed BPE. Techniques are now available for polymerization with single-site catalysts that allow long branches to be included by copolymerization with alkene-terminated BPE formed during the same polymerization reaction. Alternatively, the long branches can be added to any PE composition by blending with LDPE.

3.2 STRUCTURE AND PROPERTIES OF POLYETHYLENES The important factors (in order of most to least) in controlling the properties of polyethylenes are as follows: Proportion of branches (i.e., the composition of the comonomer). Each comonomer unit in the polymer is unable to crystallize, so this breaks the ability of the polyethylene chain to take part in crystallization. The distribution of the comonomer units in the chain (are they distributed evenly or are they clustered within some molecules or within parts of some molecules). This is determined by the type of catalyst—metallocenes give relatively even distribution of branches. Ziegler–Natta catalysts give broad distribution. Bimodal distribution of branches, similar to the above two, except that the melting shows two endotherms. Usually a broad lower temperature melting and a sharp higher temperature melting, the latter for crystals of molecules with few branches. This is caused by uneven reactivity giving two phases in the reactor. Type of branches—1-butene comonomer gives ethyl branches, 1-hexene comonomer gives butyl branches, and 1-octene comonomer gives hexyl branches. The length of branches has a small influence, but the number of branches is more important since the branches do not take part in crystallization. Molar mass (inversely proportional to melt flow index, MFI) has a small influence. Generally in the chain folded crystal structure, the molar mass will not be important. It will mean that a particular molecule has more folds in the crystal. Melting is determined by the crystal thickness. Crystal thickness is determined by crystallization conditions and the presence of branches that sterically prevent crystallization of some segments. Often molecules with different numbers of branches will be immiscible in the liquid state (melt). So a polymer with a broad branch distribution may have two


65

liquid phases and so crystallize with two crystal populations. This is another manifestation of the above three, within the one polymer. Blends of polyethylenes with much different branch contents will also show two immiscible liquid phases.

3.3 APPLICATIONS OF POLYETHYLENE BLENDS Rheological control is a primary reason for the preparation of PE blends. The PE chosen to provide optimum properties for a particular application may not have desired processing characteristics, Table 3.4. Ideally shear thinning is preferred so that high molar mass PE required for a product can be processed under moderate conditions. Even a small amount of BPE with significant shear thinning can impart shear-thinning properties when used in a blend. Branching increases the elasticity, elongation at break, and toughness of PE films, though modulus, yield stress, and break stress are decreased. Heat sealing of films is enhanced by BPE since branching reduces the melting temperature. PE films are usually produced by the blown film extrusion method. The alternative method is sheet extrusion followed by passing through chilled rollers. Blown film is more economic; biaxial orientation is introduced by the draw-off and blow ratios. Bubble stability is critical to the blown film process. Bubble stability is provided by the melt strength and rheological characteristics of the PE. Long

Table 3.4 Some Examples of Typical Polyethylene Blends and Their Applications. Polymer 1

Polymer 2

HDPE

LDPE

HDPE

LLDPE

LLDPE mLLDPE UHMWPE

HDPE

Application

Toughened HDPE with improved processing, especially melt strength; HDPE and LDPE are immiscible LLDPE Pipe grade HDPE required rheology for (or LDPE) extrusion; rigidity and impact resistance for pipes LDPE Typical film combination for increased melt strength, elasticity combined with LLDPE strength. These are the most widely used PE blends mLLDPE or Packaging films with improved melting and mVLDPE flow character for heat seal LDPE Introduction of long branches to assist processing into specialized metallocene PE HDPE Lower molar mass (HDPE) to moderate the properties of ultrahigh molar mass PE for processing and flexibility HDPE Bimodal molar mass distributions provide (‘‘reactor blends’’) unique properties that cannot be achieved by mixing the two PE grades

66

Polyolefin Blends

branches and high molar mass increase melt strength and so LDPE is preferred for blown film manufacture, but other properties may be required of the film after formation. The solution to optimization of production and application properties is to use blends of PE to contribute their unique characteristics. Such an important blend use for films is LLDPE–LDPE. Better-defined properties can be obtained using particular BPE formed by single-site catalysts. These BPE are more expensive, so a blend with lower cost PE will often provide the desired contribution to properties. Often combinations of properties are required and these are more applicable to a PE blend. A combination of high melt strength and shear-thinning rheology requires long branches, with high film strength from low branch content, to allow thin films to be made. Heat sealable films require a low temperature melting component, together with a strength-providing component. PEs are being increasingly used for cable coverings, either for large cables or for flexible appliance coverings. Many competing requirements are expected of these cable coverings, most specified by regulations or standards. The inherent properties of the PE are often supplemented by peroxide or grafted silane-induced cross-linking. Metallocene PE have suitable properties for cable covering, but blends are needed to satisfy all specifications. Plasticised poly(vinyl chloride) has been used so a PE covering will often need to possess similar properties to be considered as a substitute. Extruded pipes need to be rigid yet tough so a significant component is likely to be HDPE, yet toughness must be included and flexibility to enable coiling. Toughness of HDPE can be improved by inclusion of some comonomer, though a blend with a PE plastomer or elastomer will produce dispersed phase elastomer that is know to provide efficient toughness. Other extruded profiles for building components will require similar properties. PE moldings such as garbage bins need a combination of stiffness, strength, and toughness. In addition high melt flow is needed to completely fill the mold before solidification. HDPE is the base polymer, but other blended polymer such as a PE elastomer will be required to increase the toughness while retaining strength and dimensional stability, particularly when the bin is used to carry a load. PE is used for large rotationally molded tanks. The tanks must be strong, stiff, and tough as for the other molded products. Rotomolding is a special molding technique where flow of the molten polymer is required under a small centrifugal or gravitational force and the walls must be void free. The slow molding process usually does not produce stresses within the walls. Low molar mass could provide suitable melt flow, but the tank would have insufficient strength so the PE must be a blend to obtain the individual distinct properties. See Table 3.5 for a summary of processing methods, the PE and product requirements.

3.4 MOLAR MASS AND BRANCHING DISTRIBUTIONS Metallocene catalysts provide PE with control of molar mass and a narrower molar mass distribution than other catalysts. These PE will have precise control of


67

Table 3.5 The Main Processing Methods for Polyethylenes and Their Requirements. Processing method

Polyethylene

Comment

Extrusion

All types

Cast-film extrusion

LLDPE, HDPE

Laminate-film extrusion

LLDPE, LDPE

Coextrusion

VLDPE–mPE

Blown-film extrusion

VLDPE–LDPE blends

Injection molding

HDPE

Vacuum forming

HDPE, VLDPE

Rotational casting

LDPE, LLDPE

Powder coating

LDPE, LLDPE

Extrusion is the main process for compounding with additives, fillers, pigments, blended polymers, and then forming pellets for subsequent processing. Further processing may form various profiles such as pipes, beading, wire, and cable coating Film is extruded through a slit-die then passed through chilled rollers to crystallize and smooth the surface; orientation by cold drawing PE can be extruded onto other substrates, paper, and paperboard mainly; though other polymer films and metal foils are used Two or more polymer films are extruded together for combination properties not available by blending; strength and heat-seal layers The tubular extrudate in expanded into a thinner film; melt strength to resist bursting is needed Molding of articles of many sizes, from rubbish bins to small fittings requires melt high flow and lubrication Partially melted sheet is rapidly shaped with high extensional flows and critical edge strengths Melting, flow, and particle coalescence are required; particularly flow under little more than gravity, though log times are involved Powder must adhere to hot objects, then flow under adsorption and surface tension forces

properties associated with molar mass, but this may not be effective in providing a broad range of properties. A broad molar mass distribution will combine good processing with sufficient strength, but with less precise control. Multisite catalysts give a broad molar mass distribution, but the molar mass can be controlled to give a range of grades for each polymer. Radical polymerization to manufacture LDPE gives the least control of molar mass and distributions or offer very broad due to the many long branches.

68

Polyolefin Blends

Bimodal molar mass distributions have become of increasing interest due to their combination of processability and mechanical properties. Combinations of toughness, tensile strength, and ultimate strain can be found in bimodal PE. Bimodal HDPE can retain modulus and yield strength, while providing some flexibility and toughness. The alternative of blending with a BPE often decreases modulus and tensile strength too much for applications such as extruded pipe. Resistance to creep is an important property retained in bimodal HDPE. The bimodal PE are usually not blends of differing molar mass, they are created in a polymerization reactor. Bimodal LLDPE are available where the high molar mass component contributes to tensile strength and creep resistance required in films, while the low molar mass facilitates processing and flexibility of the film. Branch distribution is the orthogonal distribution to molar mass distribution. Branching determines crystallinity and melting temperature range, so the mechanical properties are more dependent on branch distribution than molar mass distribution. As described above, catalyst technology has mainly revolutionized the copolymerization reaction. The single-site or metallocene catalysts produce a more uniform structure, in both molar mass and branch distribution. These catalysts are less selective of monomer, and the branch distribution becomes more statistical compared with multisite catalysts such as Zeigler–Natta. The distribution of short branches approaches that of LDPE, except that there are normally no long branches. Long branches can be introduced through single-site copolymerization. This is done by maximizing termination by disproportionation so that unsaturation is formed at chain ends. The end unsaturated chains then become macromonomers for further copolymerization with ethylene and the respective 1-alkene (2). The long branches are desired for their contribution to shear thinning and melt strength during processing. These rheological properties assist blown film production. Many LLDPE are compositionally heterogeneous, since they contain molecules with few branches and molecules that are highly branched. The more highly branched fraction can be separated by solvent extraction. Small-angle neutron scattering (SANS) has confirmed the prediction that structurally disperse LLDPE display liquid–liquid phase separation, even if the overall branch content is low. These LLDPE behave like polyethylene blends even though they are the result of a single polymerization. PE with high levels of 1-alkenes can be produced using single-site catalysts. The near statistical highly branched PEs are elastomeric. They have similar compositions to EP and EPDM, but they are formed from comonomers such as 1-butene, and their comonomer distribution is more uniform. They have processing characteristics similar to thermoplastics, as opposed to rubbers.

3.5 CRYSTALLIZATION, MELTING, AND BRANCHING OF POLYETHYLENES The melting temperature of PE varies over a broad range with the equilibrium melting temperature as the upper limit for LPE. The melting temperature varies with lamella


69

thickness that in turn varies with crystallization conditions. Slow crystallization will give larger lamella thickness. Slow crystallization will occur at higher isothermal conditions or upon slow cooling. As the crystallization temperature (Tc ) approaches the melting temperature, the melting temperature approaches the equilibrium melting temperature (Tm ). Branching produces a new limiting temperature called the copolymer melting temperature (Tmc ). Tmc still depends on crystallisation conditions and represents the limiting case for Tc ¼ Tmc ; however, Tmc depends on the mole fraction of branches, or more specifically the noncrystallizable component. Branches suppress lamella morphology due to their causing variable crystallizable sequences (3). Branches are excluded from the lamella, thus the lamella thickness is limited by the length of ethylene segments, the methylene sequence length (MSL). Methyl and sometimes ethyl branches can be included in lamella if crystallization is fast; the resulting crystals are less well formed and the melting temperature is further reduced. Linear polyethylene (LPE or HDPE) has been blended with LLDPE with varying branch distributions. The critical branch content for phase separation was lower for mPE, where the branches were more evenly distributed, than for Ziegler– Natta LLDPE, where the branches are heterogeneously distributed. Where cocrystallization occurred, the unit cell of the crystals showed marked expansion (4). Crystallization conditions were shown to be important for the cocrystallization of blends of linear and branched polyethylenes, low isothermal temperatures promote cocrystallization as did quench cooling (5). Figure 3.4 shows the crystallization (on cooling at 10 C min1) followed by melting (on heating at 10 C min1) of ZN–octene–LLDPE. This slurry or solutionpolymerized LLDPE shows considerable molecular or branching heterogeneity according to the broad bimodal melting temperature range. There is a sharp higher

Figure 3.4 Crystallization and melting of ZN–octene–LLDPE after cooling and then heating at 10 C min1 using a Perkin-Elmer Pyris 1 DSC.

70

Polyolefin Blends

Figure 3.5 DSC melting curves for HDPE, LDPE, and their 1:1 blend.

melting double peak at 125 C, where the double peak is attributed to melting– recrystallization–melting during the DSC scan. Crystallization similarly shows a sharp higher temperature exotherm followed by a broad large exotherm (not both exotherms and endotherms are positive since the heat flow data have been converted to apparent specific heat by baseline corrections and calibration). DSC melting scans of blends of polyethylenes reveal changes in all aspects of crystallinity as shown by the melting peaks, though some peaks of the original components are retained in modified form. This indicates different crystal populations that may arise from liquid–liquid immiscibility or fractionation of a singlephase liquid during crystallization. Figure 3.5 shows melting of HDPE and LDPE and their blend. The HDPE in the blend melts similarly to the pure HDPE but at a slightly lower temperature. The melting of LDPE in the blend occurs at the same temperatures as the pure LDPE, but the main peak at 110 C is much reduced indicating that LDPE is miscible with the HDPE in the amorphous phase and that crystallinity of the blend is much reduced. Figure 3.6 shows the melting of a blend of LDPE–VLDPE and the melting of the pure components. The individual peak melting temperatures are retained in the blend, but the distribution of crystals melting under each peak has changed. This indicates miscibility of the components with similar branching distributions to provide a new branching distribution in the liquid phase, which may exist as a phase separated liquid phase. A small low temperature endotherm shifted from about 90 C in pure LDPE, to 80 C in the blend. The LDPE–ULDPE blend shown in Fig. 3.7 provides the greatest difference between the PE in these examples. ULDPE melts between 30 and 75 C because it is highly branched; such PE are called thermoplastic polyolefin elastomers (TPO). ULDPE and LDPE in the blend seem to mainly melt independently, though with overall decreased crystallinity. The main peak shift is the LDPE lower melting temperature peak shifting from 90 C to about 80 C. The melting endotherms of LDPE and


71

Figure 3.6 DSC melting curves for LDPE, VLDPE, and their 1:1 blend.

ULDPE do not overlap, so it can be assumed that the branching distributions do not overlap making it unlikely that any components from each polymer would be dissolved in the other. Within a BPE with a distribution of branches, there will be a distribution of lamella thickness. This will result in a broad melting range. BPE with a bimodal distribution of branches will have a bimodal distribution of lamella thickness and a corresponding melting temperature range. When two polyethylenes are blended, assuming they are miscible, they will cocrystallize only where they have common MSL. Some molecular segments in each BPE will crystallize independently of

Figure 3.7 DSC melting curves for LDPE, ULDPE, and their 1:1 blend.

72

Polyolefin Blends

the other BPE in the blend. Note that it is the segments that enter lamella, not whole molecules. Where there are intramolecular branch distributions, then segments of the one molecular can join different lamella as appropriate to their MSL. Thermal fractionation techniques have been used to separate crystals with differing lamella thickness formed from molecular segments with differing methylene sequence lengths. Stepwise isothermal cooling (SC) has been used to allow each fraction to crystallize according to the temperature appropriate to each methylene sequence length, where longer sequences can crystallize at higher temperatures (6–8). Another technique that provides better resolution of the fractions is successive self-nucleation and annealing (SSA) (9). SSA has been used to fractionate ethylene–hexene copolymers to distinguish the contribution of catalyst on the branch distribution and structure (10).

3.6 MISCIBILITY AND CRYSTALLIZATION Polyethylenes have a broad distribution of melting temperatures resulting from varying lamella thickness. The lamella thickness is determined by the fold length, which is limited by MSL. At each crystallization temperatures, only those molecules with sufficient MSL can crystallize. If a blend is homogeneous, then crystallization will proceed, which is dependent on MSL and temperature, irrespective of the component from where the molecules originated. The overall distribution of lamella thickness will be a combination of the lamella thicknesses if each component were crystallized separately. Difference in the distribution will depend on the conditions of crystallization, such as rate of cooling. This situation will provide a broadened and probably bimodal melting endotherm. In this case phase separation does not occur, but fractionation according to MSL occurs for the combined blend components. If the PEs in a blend are immiscible at the temperature of crystallization, then each phase must crystallize independently and there will normally be a disperse phase and a continuous phase, except for nearly equal volume fractions where the phases could be cocontinuous. Though the phases crystallize independently, they will not crystallize at the same temperatures, the crystallization temperatures will depend on branch density. This situation will provide a bimodal melting endotherm and a double crystalline morphology. Note that the two crystalline phases will not have the composition of the original PE. The composition of the phases will be according to the tie-lines of the phase diagram (Fig. 3.8). As the blend is cooled and the mutual solubility changes, the composition of the two phases will change. Descriptively, the more linear PE will dissolve the more linear molecules from the more branched PE of the blend. Conversely, the more branched PE will dissolve the more branched molecules from the less branched PE. In this case phase separation occurs before crystallization and continues during crystallization. The phase equilibria correspond to a typical U-shaped upper critical solution temperature diagram. The competition between liquid–liquid phase separation and crystallization from homogeneous melt has been observed for


73

Figure 3.8 Temperature–composition phase diagram for an immiscible polyethylene blend with differing branching.

blends of ethylene copolymers with hexene and butene, using time resolved SAXS and WAXS, complemented by optical and atomic force microscopies (11). Similar conclusions were made concerning blends of ethylene copolymers of butene, hexene, and octane (12–14). In both miscible and immiscible blends, the resulting crystalline morphology will be complex and there will be a broad melting temperature range. Properties will be significantly modified by the blend morphology and events such as impact; elongation or creep conditions will differ. Only a small amount of a modifying PE may be necessary to provide significant changes. If miscibility is required then the two polyethylenes must be similar, but different enough to change morphology in the blend. Blends of LLDPE with LDPE with only small differences in branch composition (3% and 5%, respectively) were found to be immiscible in the liquid (15). Blends of HDPE with long-chain branched polyethylenes (HBPE) prepared from metallocene catalysts have been studied by DSC and their crystal structures interpreted in terms of phase behavior. The HBPE contained long-chain branches and short branches formed form octane comonomer. HBPE with 7.5–12.0% octane exhibited phase separation, whereas HBPE with 2% octane were found to be miscible with HDPE over the whole composition range. Long branches were few and did not contribute to the immiscibility (16). Crystallization and melting of ternary blends containing mPE, LLDPE, and LDPE have been studied where the mPE and LLDPE varied in content and had the same melting temperature, Tm ¼ 122 C and a fixed content of LDPE, Tm ¼ 114 C. Crystallinity increased with mPE content, and it was proposed that mixed crystals may be formed since no separated melting peaks were observed for the component PE (17). Blends of LPE and several different poly(ethylene-cohexene)s were shown,

74

Polyolefin Blends

by DSC measurements of melting, to be immiscible even with only 1.8 mol % hexene (18). SSA has been used to determine the miscibility of linear and branched polyethylenes and the results showed that only those PE fractions that were similar in branch content and distribution of short branches were miscible in the melt. The SSA thermal fractionation helped to distinguish miscibility effects from cocrystallization effects (19). Phase diagrams have been proposed for blends of polyethylenes where the first component is linear, or less branched, and the second component is more branched. The method involves quench cooling each blend composition from the melt at various temperatures, so that there is insufficient time for liquid–liquid phase separation. Separated blends are considered to be separated at the particular temperature of the melt prior to quenching. TEM and DSC are used to characterize the quenched blends. The phase diagrams exhibit UCST behavior, depending on the difference in branching content of the component polyethylenes. Branch length has been found not to be important since it is the branch points that are excluded from the crystals (20,21).

3.7 THEORETICAL PREDICTION OF MISCIBILITY The miscibility of polyolefins in the melt is difficult to predict. Their miscibility or immiscibility is in contrast to the chemical similarity of their structural groups: methyl, methylene, and methyne. Polyolefin melts have similar structures, including density–temperature relationships and optical characteristics such as the refractive index, making most characterization techniques inappropriate for the detection of phase separation. The general rule of like dissolves like being the determinant of miscibility is inadequate for polyolefin miscibility. Some blends of polyethylenes are considered to phase-separate the above crystallization temperatures of the components according to an upper critical solution temperature spinodal phase diagram (Fig. 3.8). Small-angle X-ray scattering and hot-stage optical microscopy have not provided definitive evidence of phase separation. Crystallization of polyethylene from solution has been shown to follow selfnucleation to form gels where the network links are formed by the crystals. Crystallization occurs at a temperature-dependent critical gelation concentration. The lamella may join edge-to-edge when near coplanar (22). Crystallization from miscible polyethylene blends is expected to be consistent with solution gelation studies. Two processes are predicted to occur on solidification of polyolefin blends. Crystallization from a homogeneous melt of the blend where the composition of the crystalline phase will initially contain molecules with the longest methylene sequence length irrespective of which component from which the molecules originated. Alternatively liquid–liquid separation may occur first, followed by independent crystallization of the two liquid phases, and in this case the composition of the liquid phases will be determined by the temperature and associated tie-lines of the phase diagram for the blend. Analogous phase separation mechanisms have been observed for polyethylene–polypropylene blends (23–25).


75

Studies, where one of the blended polyethylenes was deuterated, have revealed immiscibility of polyolefins with SANS, by exploiting the difference in coherent scattering lengths between proton and deuterium nuclei to provide suitable enhanced scattering contrast. SANS techniques have been applied to polyolefin blends to investigate the thermodynamic interactions relating to phase behavior and purecomponent pressure–volume-temperature (PVT) relationships. Qualitative predictions of the miscibility of model homopolymer blends of PE and deuterated PE have been achieved by simulation. Optical microscopy, differential scanning calorimetry, rheological measurements, and scanning electron microscopy have been applied. Molecular architecture and the effects of molar mass have been studied in relation to phase separation and PVT data (26). Thermodynamic interactions in polyolefin blends have been shown to originate from induced- dipole forces but they depend upon the component structures. SANS has been used to determine interactions of blends, including solubility data, light scattering to determine phase boundaries, and PVT measurements to establish cohesive energies. Isotope and microstructure interactions have been shown to contribute to the Flory–Huggins interaction parameter in mixtures of deuterated and protonated PE. Lack of a single dominant thermodynamic parameter determining phase behavior means polyolefin blends are influenced by small variations in molecular architecture. Binary phase behavior was correlated with statistical segment length asymmetry and its dependence on the Flory–Huggins interaction parameter. Additional contributions from temperature, composition, and molar mass have been established, so Flory–Huggins theory of polymer blends and their thermodynamics has been extended to include them. Miscibility loop behavior has been predicted and described using a temperature-dependent interaction parameter, leading to an UCST phase diagram. The predictions have been attributed to nonrandom packing effects where constituents have sufficient asymmetry in the polymerization indices (27). Molecular packing may have a dominant influence on polyolefin miscibility, so the accurate prediction of polyolefin melt structures would enable understanding of reasons for the difficulty of PE miscibility prediction to be obtained. In the melt, molecules of a homopolymer polyolefin, such as LPE, generally exist in random conformations, but some order is present because of weak intramolecular and intermolecular interactions, such as dispersive forces. Molecular conformation is determined by the configuration of branches, which influence the packing of the PE chains, determining the melt structure, density, and, hence, the miscibility of PE in the melt. The physical properties of the melt are thus determined by conformation. The branch structure, length, frequency, and distribution determine the most favorable conformations. The conformations allow or limit interactions to the extent that PE with differing branch structures may be immiscible. The thermodynamic stability of a single liquid phase has been theoretically predicted by computation of pure component and blend parameters dependant upon entropic considerations (28). Microphase separation has been ascribed to a considerable increase in unfavorable noncombinatorial entropy, using a Flory–Huggins equation to predict the free energy of mixing (29).

76

Polyolefin Blends

3.8 RHEOLOGY OF MELTED POLYETHYLENE BLENDS Linear polyethylene rheology is dependant on the critical molar mass for entanglements (940 g mol1 (30), which is low due to lack of chain stiffening groups or side chains. Chain entanglements and branch entanglements provide non-Newtonian rheology for higher molar mass PE. Entanglements of branches increase the melt strength of PE with long branches, especially LDPE, and also some newer mPE that are polymerized to contain long branches in addition to the short branches from a comonomer. Short branches do not interact or entangle sufficiently to provide melt strength. PE with differing branch content have been shown to form immiscible liquid–liquid mixtures. The rheology of immiscible PE blends deviate from additivity expected for a homogeneous liquid. For example, HDPE and LLDPE are miscible when the branch content is low (8 branches/ 100 Cs, phase separation takes place. The rheology of melted polyethylene blends is of importance in processing (31). Processing is often by extrusion where the main product is film, either from the blown film process or cast film. The molecular structure and hence meltingtemperature profile determines screw design. Usually PE blends melt over a broad temperature range, so a long compression zone is required. Orientation of blown film requires high melt strength that is enhanced by long branches while subsequent film strength is dependent on crystallinity requiring long methylene sequence lengths and so few branches. The optimum structure for processing and strength of PE films is best achieved by blending different PE. Injection molding requires high melt flow and hence low molar mass, while the mechanical properties will require high molar mass; the optimum properties may require a blend of PE with differing molar mass, though bimodal molar mass distribution. Foams are prepared by gas injection molding, and successful foams require high melt strength. Rotational casting, powder coatings, heat sealing all require melt flow under relatively low shear, so non-Newtonian characteristics may not have the opportunity to become apparent. Rheology can be modified by processing aids and lubricants and some of these are low molar mass PE, such as PE waxes and microcrystalline waxes. Rheology has been used to interpret the miscibility of polyethylene blends in lieu of direct measurement of phase separation. The sensitivity of various models has been shown suitable for distinguishing homogeneous melts from liquid–liquid phase separated melts (32,33). Such phase separated melts may be of advantage in extrusion processing of polyethylene blends; particularly in blown film formation where melt strength and rheological characteristics are critical to bubble stability, film uniformity, orientation, and clarity. Metallocene poly(ethylene-cobutene)s were blended with LDPE and the viscosities were increased at low frequencies though shear thinning was more evident. The relaxations were retarded in the presence of LDPE and the effects were more noticeable for the metallocene copolymers with higher butene content (34). Metallocene HDPE and metallocene LLDPE blends were confirmed by rheological analysis to be miscible, contrary to blends of the same metallocene HDPE with LDPE (35).


77

3.9 MECHANICAL PROPERTIES OF POLYETHYLENE BLENDS PE blends are compatible; they can be miscible, cocrystallize, or crystallize with separate crystallite distributions. These morphologies effect modulus, yield and any double yield phenomena, strain hardening, elasticity and recovery, creep, creep recovery, stress relaxation, tear strength. These properties are all measured using static stress or strain. Typical tensile stress–strain curves for LLDPE, LDPE, and a blend of LDPE–EP rubber are shown in Fig. 3.9. LDPE shows a sharp yield, plastic flow with some strain hardening with increasing strain. The LDPE–EP blend is a thermoplastic vulcanizate, though it still yields, it extends with constant stress. LLDPE shows a sharp yield followed by plastic flow, until extensive strain hardening occurs over the strain range 4 to 13. Strain hardening is a problem for films where rapid stretching to high strain is required, such as pallet wrap, stretch wrap, and agricultural films. The film exhibits decreased elasticity during strain hardening, as crystallinity is oriented in the direction of flow and the crystalline morphology changes. Figure 3.10 shows atomic force microscopy of the Ziegler–Natta-catalyzed LLDPE containing 5% hexene, shown in the stress–strain curve (Fig. 3.9). The original crystals become elongated and orientated during straining with loss of film elasticity, increase in surface roughness, and loss of film transparency. Blending with LDPE or metallocene VLDPE will reduce the strain hardening. Polyethylenes undergo interlaminar deformation during tensile elongation. This can give rise to a double yield phenomenon. At the onset of the first tensile yield, chain slip and lamella rotation occur and this process is reversible. The second tensile yield is irreversible and coinsides with lamella fragmentation (36). These mechanically induced morphological changes and the observation of any double yield phenomena are dependent on several structural factors that can be controlled by

Figure 3.9 Tensile stress–strain curves for LLDPE, LDPE, and LDPE–EP blends.

78

Polyolefin Blends

Figure 3.10 Atomic force microscopy of ZN–hexene–LLDPE before strain (a) and after strains of 3 (b) and 6 (c) (10 mm 10 mm).

polyethylene blending. Homogeneous or heterogeneous slips have been identified as responsible for different macroscopic deformation regimes. Homogeneous slip, arising from dislocations within the crystals, has a higher thermal activation and predominates as temperature increases or strain rate decrease. Heterogeneous slip, operating through defective crystal block boundaries, proceeds with lower strain hardening (37). Dynamic mechanical properties exhibit side chain or branch motions; short main chain segment motions, main chain segmental motions, recrystallization, and melting. These transitions are observed as inflections in the storage modulus curve with temperature, peaks in either the loss modulus or damping factor (tan(d)) curves. Figure 3.11 shows the dynamic mechanical spectroscopy (DMS) of a ZN–VLDPE at 1 Hz in tensile mode. The glass transition temperature (maxima of the loss modulus


79

Figure 3.11 Dynamic mechanical spectroscopy of ZN–VLDPE in tensile mode using frequency of 1 Hz.

peak, or the damping factor peak, as often used) is at about 40 C. The transitions of individual components of a blend are often detected independently. The activation energy for each of the transitions can be determined using multifrequency data and correlation with the Arrhenius equation. Williams–Landel–Ferry (WLF) analysis of the multifrequency data is used to construct time (frequency)–temperature master curves to obtain plateau or terminal viscoelastic properties. These properties reveal the contributions of differing branching levels or distributions within a blend. This is illustrated in Fig. 3.12 that shows a LDPE–EP rubber blend 1, 2, 5, 10, and 20 Hz in

Figure 3.12 Dynamic mechanical spectroscopy of LDPE–EP rubber blend in tensile mode using synthetic frequency multiplexing of 1, 2, 5, 10, and 20 Hz.

80

Polyolefin Blends

tensile mode measured using synthetic frequency multiplexing. The glass transition temperature (50 C) is shifted to higher temperatures with higher frequencies. A second, lower temperature transition is shown in the range 70 to 90 C. Metallocene poly(ethylene-cooctene)s showed changes in the b-transition region that were related to motions of segments located at interfacial regions (38).

3.10 ADDITIVES Many additives are included in PE and their blends. These include colorants, stabilisers, ultraviolet protection, flame retardants such as antimony oxide and decabromodiphenyl ether, or halogen free flame retardants such as magnesium and aluminum hydroxides. Prodegradents are an alternative to stabilizers, such as prooxidants that are currently being used in some packaging films to reduce environmental impact. Surface properties such as tack, antiblocking, corona discharge treatment for printing, and its subsequent reversion over time are modified by additives. Additives for processing include internal and external lubricants. Microcrystalline waxes, hydrocarbon waxes, stearic acid, and metal stearates such as calcium, magnesium, and zinc stearates are common lubricants. Often blends with lower melting temperature polyethylenes, particularly those with long branches are preferred for processing enhancement. Elastomeric or plastomeric polyethylenes often require cross-linking, so radical initiators such as cumyl peroxide are added, or alkoxyvinylsilanes grafted for subsequent water-induced cross-linking.

3.11 CONCLUSIONS The polyethylenes are a large and diverse family of chemically similar polymers; increased understanding is required to learn how to combine them to achieve blends with a custom range of properties. Single-site (metallocene)-catalyzed PE have brought a new life cycle to PE developments and applications. These new PE have initially found applications in blends with existing PE whose characteristics are well known but require processing and performance improvement. Most PE blends are with other PE to provide combinations of molar mass, short branching, and long branching. Other blends are with PE copolymers with polar or functionalized comonomers so that increased surface polarity and functionalization can be included. Production and cost pressures increase the performance requirements for PE. PE can be enhanced to replace other polymers as well as expanding into areas occupied by other materials. The similarity of the chemistry and lack of distinguishing features for many characterization techniques make investigation of the structure–property– performance of PE blends a difficult and often controversial task. Critical requirements for PE are rapid processing, thinner yet stronger films, ease of direct printing or adherence of labels, rapid heat sealing, high strain rate at high strain for stretch wrap, plastic memory for shrink wrap, cross-linkability to resist creep and stress relaxation, cross-linking to resist elevated temperatures when the


81

low melting temperature of PE would otherwise be a constrain, increased elasticity covering the range from traditional elastomers to stiff thermoplastics. PE compositions need performance compatible with newer high speed processing equipment and high speed packaging and finishing machinery. The new PE and their blends are meeting these needs and opening new opportunities where they easily meet or exceed expectations. All of these properties and structural diversity are now being derived from the polymer with the simplest chemical structure.

NOMENCLATURE Tg Tm Tm Tmc Tc PE LPE mPE M HDPE BPE LLDPE VLDPE ULDPE LDPE TPO TPE VAc MA BA GMA AA MAn MSL SANS PVT UCST SAXS WAXS SSA SC HBPE DMS ZN

Glass transition temperature Melting temperature Equilibrium melting temperature Copolymer melting temperature Crystallization temperature Polyethylene Linear polyethylene Metallocene polyethylene Molar mass High density polyethylene Branched polyethylene Linear low density polyethylene Very low density polyethylene Ultralow density polyethylene Low density polyethylene Thermoplastic polyolefin Thermoplastic polyolefin elastomer Vinyl acetate Methyl acrylate Butyl acrylate Glycidyl methacrylate Acrylic acid Maleic anhydride Methylene sequence length Small-angle neutron scattering Pressure–volume–temperature Upper critical solution temperature Small-angle X-ray scattering Wide-angle X-ray scattering Successive self-nucleation and annealing Stepwise isothermal cooling Long-chain branched polyethylenes Dynamic mechanical sprectroscopy Ziegler–Natta catalyst

82 EP TPO

Polyolefin Blends

Ethylene–propylene copolymer elastomer Thermoplastic polyolefin elastomer

REFERENCES 1. S. P. Chum, C. I. Kao, and G. W. Knight, Structure, properties and preparation of polyolefins produced by single-site catalyst technology, in: Metallocene-Based Polyolefins: Preparation, Properties and Technology, Wiley Series in Polymer Science, Vol. 1, J. Schiers and W. Kaminsky (eds.), Wiley, New York, 2000, p. 261. 2. G. D. Wignall, R. G. Alamo, J. D. Londono, L. Mandelkern, and F. C. Stehling, Macromolecules, 29, 5332 (1996). 3. F. M. Mirabella, J. Polym Sci. B: Polym. Phys., 44, 2369 (2006). 4. Y. Zhao, S. Liu, and D. Yang, Macromol. Chem. Phys., 198, 1427 (1996). 5. M. J. Galante, L. Mandelkern, and R. G. Alamo, Polymer, 39, 5105 (1998). 6. G. Amarasingh, F. Chen, A. Genovese, and R.A. Shanks, J. Appl. Polym. Sci., 90, 681 (2003). 7. R. A. Shanks and G. Amarasinghe, Application of differential scanning calorimetry to analysis of polymer blends, in: Polymer Characterization Techniques and Their Application to Blends, G. Simon (ed.), Oxford University Press, New York, 2003, pp. 22–67. 8. P. Starck, P. Lehmus, and V. Seppala, Polym. Eng. Sci., 39, 1444 (1999). 9. A. Muller and M. L. Arnal, Prog. Polym. Sci., 30, 559 (2005). 10. M. Bialek, K. Czaja, and B. Sacher-Majewska, Thermochimica Acta, 429, 149 (2005). 11. Z. Wang, H. Wang, K. Shimizu, J. Y. Dong, B. S. Hsiao, and C. C. Han, Polymer, 46, 2675 (2005). 12. G. Matsuba, K. Shimizu, H. Wang, Z. Wang, and C. C. Han, Polymer, 44, 7459 (2003). 13. G. Matsuba, K. Shimizu, H. Wang, Z. Wang, and C. C. Han, Polymer, 45, 5137 (2004). 14. K. Shimizu, H. Wang, Z. Wang, G. Matsuba, H. Kim, and C. C. Han, Polymer, 45, 7061 (2004). 15. M. J. Hill and C. C. Puig, J. Appl. Polym. Sci., 65, 1921 (1997). 16. J. Schellenberg and B. Wagner, J. Thermal Anal., 52, 275 (1998). 17. M. Run, J. Gao, and Z. Li, Thermochim. Acta, 429, 171 (2005). 18. B. S. Tanem and A. Stori, Polymer, 42, 5389 (2001). 19. M. L. Arnal, J. J. Sanchez, and A. J. Muller, Polymer, 42, 6877 (2001). 20. M. J. Hill, P. J. Barham, A. Keller, and C. C.A. Rosney, Polymer, 32, 1384 (1991). 21. M. J. Hill and P. J. Barham, Polymer, 38, 5595 (1997). 22. J. H. Kim and R. E. Robertson, Polymer, 39, 5371 (1998). 23. M. Razavi-Nouri and J. N. Hay, Polym. Eng. Sci., 46, 889 (2006). 24. J. Li, R. A. Shanks, and Y. Long, J. Appl. Polym. Sci., 87, 1179 (2003). 25. J. Li, R. A. Shanks, and Y. Long, Chin. J. Polym. Sci., 20, 497 (2002). 26. H. M. Freischmidt, R. A. Shanks, G. Moad, and A. Uhlherr, J. Polym. Sci. B: Polym. Phys., 39, 1803 (2001). 27. H. W. Kammer, Polymer, 40, 5793 (1999). 28. G. H. Fredrickson and A. J. Liu, J. Polym Sci. B: Polym. Phys., 33, 1203 (1995). 29. H. W. Kammer, Polym. Netw. Blends, 5, 69 (1995). 30. H. G. Elias, An Introduction to Polymer Science, VCH Publishers Weinheim, New York, 1997, Chapter 7. 31. L. A. Utracki, Polymer Alloys and Blends: Thermodynamics and Rheology, Hanser Publishers, Munich, 1989, p. 201.


83

32. Y. Fang, P. J. Carreau, and P. G. Lafleur, Polym. Eng. Sci., 45, 1254 (2005). 33. F. Chen, R. A. Shanks, and G. Amarasinghe, J. Appl. Polym. Sci., 95, 1549 (2005). 34. J. Xu, X. Xu, Q. Zheng, L. Feng, and W. Chen, Eur. Polym. J., 38, 365 (2002). 35. C. Liu, J. Wang, and J. He, Polymer, 43, 3811 (2002). 36. M. F. Butler and A. M. Donald, Polymer, 38, 5521 (1997). 37. V. Gaucher-Miri and R. Seguela, Macromolecules, 30, 1158 (1997). 38. A. G. Simanke, G. B. Galland, L. Freitas, J. A. H. da Jornada, R. Quijada, and R. S. Mauler, Polymer, 40, 5489 (1999).

Chapter

4

Miscibility and Crystallization Behavior in Binary Polyethylene Blends Moonhor Ree1

4.1 INTRODUCTION Polyethylene (PE) is currently the most widely used commercial polymer throughout the world, and the industrial market for this polymer is still growing due to the variety of applications that are based around its superior properties, such as high chemical and mechanical resistances, easy processability, and low specific gravity, as well as low manufacturing costs. So far, several types of PE materials have been developed, each with different levels of mass density, branching content, and molecular weight. These PE materials include linear high density PE (HDPE), linear ultrahigh molecular weight PE (UHMWPE), linear low density PE (LLDPE), and low density PE (LDPE) (Fig. 4.1) (1–5). These polymers are usually synthesized by the conventional polymerization of the ethylene monomer, or with one or more a-olefin comonomers (1-butene, 1-hexene, 1-octene, etc.) with the aid of Ziegler–Natta catalysts (1–5). Advances in the catalysts used for the polymerization of olefins have allowed manufacturers to expand their product line to accommodate a wide range of densities and branching structures. In particular, new high performance PEs are mainly produced via the use of metallocene catalysts (i.e., single-site catalysts), which have been shown to have superior properties compared with conventional PEs (6,7). The single-site nature of the catalysts allows for the formation of polymers

1

Department of Chemistry, Polymer Research Institute, Pohang Accelerator Laboratory, National Research Lab for Polymer Synthesis and Physics, and Center for Integrated Molecular Systems, Pohang University of Science & Technology (Postech), Pohang 790-784, Republic of Korea Polyolefin Blends, Edited by Domasius Nwabunma and Thein Kyu Copyright # 2008 John Wiley & Sons, Inc.

84

Chapter 4 Miscibility and Crystallization Behavior in Binary Polyethylene Blends

85

Figure 4.1 Schematic molecular structures of LDPE, LLDPE, HDPE, and UHMWPE.

with a narrower molecular weight distribution and a more uniform distribution of comonomers, as compared with conventional PEs (6,7). HDPE gives rise to stiffer materials with good tensile properties but poor impact and tear resistance (1,5). Moreover, UHMWPE has excellent mechanical properties but very poor melt properties, causing severe processing difficulties (4,5). LLDPE, which is synthesized by copolymerization of ethylene with one or more a-olefin comonomers, has short chain branching and presents excellent mechanical properties, such as impact and tear resistance, as well as high tensile strength (1–5). However, due to a narrow molar mass distribution, LLDPE exhibits an elevation in the melt viscosity and a lowering of the melt strength, causing relatively poor processability (1–5). In general, metallocene-catalyzed LLDPE is synthesized with a much narrower molecular weight distribution, compared to that of Ziegler–Nattacatalyzed LLDPE, and exhibits very poor melt properties that cause processing difficulties (6–15). On the contrary, LDPE is a polymer composed of both long and short chain branches and is well known to have good melt properties and therefore efficient processability (16). In general, the processing behavior of PEs mainly depends on parameters such as the weight-average molecular weight and distribution, comonomer content, and the long-chain-branching content and distribution (17–22). In particular, the longchain-branching content is known to have a large impact on polymer processability; PE polymers with long-chain-branching demonstrate excellent processing properties (18,23).

86

Polyolefin Blends

These PE polymers offer a broad spectrum of structures, properties, and applications. However, the blending of different types of PEs (HDPE, LDPE, LLDPE, and UHMWPE) has attracted growing interest because of the potential for obtaining low cost materials with improved mechanical properties and better processabilities, as compared to those of the pure constituents (1–34). Nowadays, 70% of PEs in the market are blends (24). The processability and properties of PE blends are dependent on the melt miscibility. Moreover, the properties are also dependent on the morphological structure of the blend, which is basically a combination of the crystallization behavior and melt miscibility. Therefore, the miscibility and crystallization behavior of PE blends have been prevalent research topics over the last two decades.

4.2 MISCIBILITY 4.2.1 Linear and Short Branched Polyethylene Blends Ree (5) first studied quantitatively the miscibility of two HDPE/LLDPE blend systems using the small-angle neutron scattering (SANS) technique: (i) blends of a HDPE-d (237,000 weight-average molecular weight Mw , 2.06 polydispersity PDI, and 98% deuteration) and a LLDPE-B (114,000 Mw , 4.50 PDI, and 18 ethyl branches per 1000 backbone carbons); (ii) blends of a HDPE-d (109,000 Mw , 1.80 PDI, and 98% deuteration) and a LLDPE-O (96,000 Mw , 4.50 PDI, and 15 hexyl branches per 1000 backbone carbons). For these blends in the melt, the Flory–Huggins interaction parameter x was determined to be small, with positive values of the order 10 4, but did not exceed the critical x values, that is, the upper limit of the stability of a single miscible phase. Moreover, these blends revealed no SANS cloud point over the temperature range 140–300 C, indicating that the HDPE blend systems with LLDPE-B and LLDPE-O are both miscible in the melt. Nicholson et al. (35) extended the SANS analysis approach to accommodate blends of HDPEs with hydrogenated and deuterated polybutadienes (PB, a LLDPE-B and PB-d, a LLDPE-B-d) with different amounts of ethyl branches ranging from 18 to 106 per 1000 backbone carbons: HDPE (111,830 Mw and 1.23 PDI), HDPE-d (116,580 Mw and 1.27 PDI), LLDPE-B-d(1) (102,660 Mw , MeSt>>NVP > TMVS > MMA > MAc > VCH. The other approach is based on the double-bond activation in MAH by substitution of one of the MAH hydrogen atoms for a strong electron-donating or electronaccepting group. In this case, the MAH double bond undergoes polarization. It is known that the stronger the double bond is polarized, the more actively the monomer reacts with macroradicals. According to Al-Malaika (6), the substitution of one of the MAH hydrogen atoms for a halogen atom promotes the yield rise of the grafted product even more than that by CTC formation. Diels–Alder system, which includes MAH and dicyclopentadiene, showed increased grafting efficiency only in a mixer of intermittent action. No special advantages have been detected in this system in comparison with neat MAH when extruder reactors were used (24). Besides MAH, acryl monomers, for example, glycidyl methacrylate (GMA), have been widely used for polyolefins functionalization. In its chemical structure, GMA belongs to 1,1-disubstituted ethylenes and has a stronger tendency to homopolymerization. Because of this, its grafting power to polyolefin chains is still lower than that of MAH. In order to raise the reacting power of GMA, electron-donating monomers (styrene or others) are added to the reacting system. Unlike MAH, GMA cannot form CTC with electron-donating monomers. There is another mechanism of raising the grafting efficiency of GMA to PO. It has been reported (6) that styrene first interacts with macroradicals. Then grafted styrene fragments copolymerize with GMA. Here, GMA is not directly grafted to chains but grafted through styrene units. The so-called ‘‘comonomer concept’’ is effective in grafting several other monomers, for example, 3-isopropenyl-a, a-dimethylbenzene isocyanate or 1-heptadecafluorooctylethyl acrylate (25,26). Vainio et al. (27) have studied the grafting of ricinoloxazoline maleate to PP and reported that the use of styrene did not increase the reaction efficiency; in some instances it decreased the latter while the degradation of polypropylene decreased markedly. The copolymerization constant value could be the basic point in explaining the effect of styrene on the grafting of vinyl monomers to polyolefins and concurrent secondary reactions (27). This means that ‘‘comonomer concept’’ acts quite successfully if at least two basic conditions are fulfilled: (i) a high rate of interaction of a comonomer with

278

Polyolefin Blends

macroradicals, and (ii) a high rate of interaction of a comonomer macroradical with the monomer being grafted (6,23,24). Macromonomers used in RE include poly- or oligomers with a reactive double bond. Because of mostly steric factors, macromonomers usually display a weaker tendency to homopolymerization in comparison with monomers. A disadvantage of macromonomers used in RE is that owing to lower volatility the ungrafted portion cannot be removed from the melt during the stage of degassing. The tendency of a monomer to homopolymerization depends on the propagation rate of chain (kp) and on the condition of dynamic equilibrium between the break and the formation of the chains (Fig. 10.1). The equilibrium depends on the temperature at which grafting takes place. The temperature at which a dynamic equilibrium is reached between the formation and the decay of monomer macroradicals is called a ceiling temperature. For certain monomers, there are published ceiling temperatures, heats, and entropy of polymerization (28,29). Their values are, for example, 150 C for MAH, 200 C for methacrylate, 400 C for acrylate and styrene (28). It should be noted that these values are typical of reactions occurring at a constant (atmospheric) pressure and monomer concentration (usually 1 mol). The peak temperature rises with monomer concentration and pressure. That is why MAH was observed to homopolymerize at an extrusion temperature above 160 C (30). When grafting is done by RE, many factors must be accounted for that concern a specific influence of the monomer on the process kinetics as well as the yield of the grafted product (1,3,5), that is, monomer concentration, its solubility in the molten PO, thermal stability, method of introduction into the reacting system, reactivity toward the initiator and macroradicals, and tendency to homopolymerization. The role of monomer is in trapping of radicals which otherwise are spent on degradation or cross-linking of the chains. If a monomer concentration is too high, phase separation may occur and lead to a lower yield of the grafted product and a stronger probability of homopolymerization (1). In such a case, a higher level of grafting can be achieved by stepwise feeding of the monomer and the initiator (31). The target-oriented selection of an initiator should be done depending on the monomer. The most important factors are the distribution of the initiator between monomer and PO phases, and also monomer’s reactivity in comparison with PO in reactions with free radicals generated during the thermal decomposition of the initiator (1). The stage of initiation of free-radical reactions taking place during monomer grafting to PO, as shown in Fig. 10.1, is the first and the most important stage. With other equal conditions, the type of initiator determines not only the yield of grafted product and the course of secondary reactions, but also the performance qualities of final materials (presence of unreacted initiator or its products of thermal decomposition within the functionalized PO can lead to accelerated ageing of goods in the course of their service). The problems of selection of initiators, discussion of the mechanism of their action, description of free-radical reactions at RE have been treated in numerous

Chapter 10 Functionalization of Olefinic Polymer and Copolymer Blends in the Melt

279

publications (1,5–7,15–20,32,33). We shall deal with most important points concerning grafting to PO blends. Most suitable initiators are organic peroxides (Table 10.2). In certain cases, it is advisable to use azo compounds as initiators (33). It is particularly typical of grafting amine-containing monomers: peroxides cause oxidation of amino groups, which leads to a loss of commercial value of a grafted product. It should be remembered that most of the azo compounds are unsuitable as initiators not only because of half-life (t0:5 ) but also because cyanoalkyl radicals formed are inactive in abstraction reactions of hydrogen from chains. An exception is phenylazo-compounds (33) being a source of phenyl radicals that are most active in hydrogen abstraction reactions (Table 10.3). The important factors concerning initiators that are to be accounted for in grafting experiments at RE include as follows: half-life; reactivity of products of the thermal decomposition of the initiator with respect to PO, the monomer, and other components of a reacting blend; solubility and distribution factor for the initiator among the components of a reacting blend in the molten PO; thermal stability and volatility of the initiator, its physical state, method of introduction into a reacting system, concentration, and tendency to form secondary products. It is advisable that the initiator should undergo thermal breakdown entirely within the reaction zone of the extruder. As the transport time of PO melt through the cylinder of the extruder does not exceed 3–5 min as well as t 0:5 —value for peroxides (Table 10.2), it is clear that thermal breakdown of a peroxide can be controlled by varying the melt temperature. At T 200 C, most peroxides have quite low values of t 0:5 . It is reasonable to assume that they undergo thermal decomposition virtually in full. If a residence time for a reacting blend in the extruder exceeds five times the half-life, the peroxide consumption exceeds 97% (1,6). On the contrary, the use of initiators with a very short t0.5 can result in negative consequences. An initiator with a short half-life leads to an increased concentration of radicals in a polymer melt. This can increase the probability of macromolecular cross-linking through recombination reactions of macroradicals. Besides, the yield of grafted product can be reduced because of a limited diffusion rate of the monomer or macroradical into the reaction zone. The latter fact is especially important for heterogenous melts. Therefore, if peroxides with short t 0:5 are used, the factors like concentration and introduction method of initiator become more important. In order to ensure a high yield of grafted product, it is advisable to introduce the initiator repeatedly. The advantage of reactive compounding, when short half-life peroxides are used, is the possibility of functionalizing PO in the first stage of compounding and mixing it with the basic thermoplastic polymer in the second stage of mixing. The use of extruder with large screw lengths (large value of L=D) and initiators with short t 0:5 values is not reasonable because of the weak influence of the melt residence time in the extruder upon the grafting efficiency and probable thermomechanical degradation of the grafted product. It should be noted that t 0:5 values in Table 10.2 have been determined at conditions different from those typical of RE process. In case of model experiments,

280

2,5-Dimethyl-2,5di(t-butylperoxy) hexane, DTBH, L-101

a:a’-Di (t-butylperoxy) diisopropyl benzene, DIPB, Perk-14

Dicumyl peroxide, DCP

PE

PP

EPDM (70–80 wt% ethylene)

Type of polymer/ peroxide, trade name

H

CH2 CH2 CH2 CH2

C

CH 3

H

CH 3

CH 2 C CH 2

H

CH2 C CH2 CH2 CH2 CH2

CH3

Structure

H

CH2

CH3 C

CH3 H

C

t-Butoxyl alkoxy

Methyl alkyl

Methyl alkyl

Methyl

Cumyloxy

t-Butoxyl alkoxy

—

—

—

—

—

—

Liquid

Solid

Solid

Solid

Solid

Solid

11.03

9.0

5.92

—

—

—

176

192

202

—

—

—

15.5

16.4

17.4

16.1

16.6

17.2

TempSolubility Concenerature of parameter tration of complete at 25 Cc Forming radicals Physical active decomposi- (calc.), Primary Secondary state oxygen,% tion, Cb (J cm3 Þ1=2

Table 10.2 Peroxide Initiators Used in RE Technology.

11.6

12.8

14.6

—

15.1

16.6

14.0

13.0

9.2

—

—

—

0.30

0.31

0.25

—

—

—

Solubility parameter at 180 Cc t 0:5 , mina (calc.), (J cm3 Þ1=2 150 C 200 C

281

CH 3

CN

OCH3

CH CH 3 3 N C CH C CH 2 3

CH 3

OO C

C H3

From derivatographic analysis (heating rate, 5 C min).17,18

N

CH3

C

CH 3

[From References (1,17,18) with permission from Elsevier].

c

Ph

H3C

According to Moad (1).

From studies.1,17,18

b

a

2-Phenylazo-2.4dimethyl-4-methoxylvaleronitrile, V-19

Di-t-butyl peroxide, DTBP

t-Butylcumyl peroxide, TBCP

2,5-Dimethyl-2,5di(t-butylperoxy) hexyne-3, DTBHY, L-130

Phenylazo alkyl

Phenyl Liquid

Methyl Liquid

—

10.95

7.3

Methyl Liquid

Cumyloxy t-butoxyl t-Butoxyl

10.2

Methyl Liquid

t-Butoxyl alkoxy

a

—

125

—

195

—

15.3

16.2

15.0

—

—

13.8

11.5

30.0

18.0

—

45.0

1.5

0.35

—

0.74

282

Polyolefin Blends

Table 10.3 180 C.

Estimated Equilibrium Constants for Hydrogen Abstraction Reaction at

No.

Radical

Primary

1. 2. 3. 4. 5. 6.

Ph RO CH3 RCH2 (R)2CH (R)3C

6 106 1 103 8 102 1 4 102 4 104

Carbon type Secondary 2 108 7 104 2 104 3 10 1 1 102

Tertiary 1 1010 6 106 2 106 2 103 9 101 1

[From Reference (33) with permission from Plenum Press].

it is practically impossible to account for the effects of pressure, shear, and other factors on the thermal decomposition of peroxides. Turcsanyi (34–36) has suggested that in selecting an initiator one of the criteria should be the temperature (T*) at which a maximum decomposition rate of peroxide is reached, and not t0:5 value. The reactivity and specific behavior of free radicals produced during initiator’s thermal decomposition strongly depend on the type of the radicals formed, which is determined by the nature of peroxide (28,37). Table 10.2 lists primary and secondary radicals formed during the decomposition of an initiator, while Table 10.3 gives data on the activity of certain types of free radicals in abstraction reactions of hydrogen atoms from carbon (33). Primary radicals are formed directly at breakdown of an initiator molecule; secondary radicals result from transformations of primary radicals by a monomolecular mechanism. It is shown in Table 10.3 that phenyl (Ph*) and alkoxyl (RO*) radicals are most active in abstraction reactions of hydrogen from carbon atoms. Methyl radicals (CH3*) are less active in abstraction reactions of hydrogen from carbon (especially primary carbon). These are especially effective in addition reactions by double bonds (1). For higher alkyl radicals, this tendency is stronger (Table 10.3). In should also be noted that of importance is the selective interaction of radicals, generated by an initiator, with PO base. It is owing to the interaction of monomer molecules with macroradicals and chain behavior of the process that grafting occurs (Figure 10.1). Up to 20 monomer units can be grafted onto one radical generated (38,39). It has been mentioned already that the initiator’s solubility in PO–monomer systems is one of the most important factors in RE technology. When grafting a monomer onto blends of olefin polymers or copolymers, the role of initiator’s solubility grows since it influences the distribution factor of the initiator between individual phases in multiphase melts (1). The distribution factor influences the grafted product yield and grafting selectivity (12,17,22). The initiator solubility in the components of a reacting system can be predicted using calculated values of the solubility parameter (d).


283

According to other authors (40,41) vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u P DE i u u d ¼ t iP ; Na DVi

ð10:1Þ

i

Values of d calculated by the procedure of group shares (40,41) from Equation 10.1 are in good agreement (the error does not exceed 10%) with the experimental values found by measurements of evaporation heats of the substances (32). It is stated in other studies (32,40) that calculated d-values, an absolute error of determination for which is 0:1ðJ cm3 Þ1=2 , are as a rule more precise than experimental ones. Values of d depend on the temperature and can be found (32) from Equation 10.2 if the temperature varies: ln dT ¼ ln d298 bkðT 298Þ

ð10:2Þ

When determining the mutual solubility of substances by d-values one should remember that in the absence of strong specific intermolecular interactions, thermodynamic miscibility between the substances mixed (complete mutual dissolution among them) has been detected if their solubility parameters differed by no more than 2 (J cm3 Þ1=2 (42). The thermal stability of the initiator influences not only the grafting course, but also the method for introducing an initiator into molten PO and the safety requirements for RE procedure (1,5). Although for RE method there are recommended solid and thermally stable—up to PO melting temperature in the reaction zone—initiators, some authors have reported high yields of the grafted product with the use of thermally unstable peroxides that undergo decomposition near the feeding zone of the extruder (12,17). This fact supports the opinion that the major role in grafting involves the interaction of monomers with the radicals generated in the reaction of PO chains with products of thermal decay of peroxides. At RE conditions, macroradicals have, probably, quite a long lifetime since the reaction zone of their interaction with a monomer can be shifted for some distance from the introduction zone of a peroxide (10–18). There are different versions of introducing an initiator into a reacting system (1,5), for example, together with PO and a monomer into the major feeding hopper; into molten PO together with a monomer, before or after the monomer; an initiator can be preliminarily absorbed into PO; by introducing an initiator into PO as a solution in a monomer or in a solvent; by addition of the whole of the initiator at once or by portions. The introduction method of initiator, as well as the determination of its required concentration, is usually selected in view of the requirement to obtain a maximum yield of the grafted product and suppress secondary reactions. If a monomer is grafted to PO blends, an initiator can be introduced locally in metered quantities into an individual polymer phase.

284

Polyolefin Blends

10.3 FUNCTIONALIZATION OF PP/PE BLENDS The problem of functionalization of PP/PE blends is of interest for two reasons. First, such a functionalization can be treated as a means of obtaining blend compositions with better—in comparison with homopolyolefins—adhesional, compatibilizing, technological and other qualities as well as price. Second, addition of one polyolefin to another during free-radical grafting can be a means of controlling the course of concurrent secondary reactions. In PP/PE blends, the degradation of chains and their cross-linking can, in fact, be balanced at the expense of cross-reactions of PP and PE macroradicals that lead to grafted copolymers such as PP-g-PE (43).

10.3.1 Effect of Reacting Blend Formulation on Grafting Efficiency and Rheological and High Elastic Properties of Melt of Functionalized PP/PE Blends In an earlier study (44), dedicated to functionalization of PP/PE blends, there was considered the effects of MAH and DCP concentration on the viscosity of the product (PE/PP)-g-MAH and on the MAH-grafted level. It was found that on increasing DCP concentration from 0.05 to 0.3 wt%, the viscosity of (PE/PP)-g-MAH melt—polymer components ratio being 90:10—varies but negligibly (Fig. 10.2a). Obviously, with increased concentration of the initiator, both degradation and cross-linking of the chains are initiated alike, which oppositely influence the polymer melt viscosity. An increase in MAH content in the PE/PP (90:10) blend causes some drop in (PE/PP)-g-MAH viscosity (Fig. 10.2b). The main reason for this is the degradation process in PP, which proceeds faster than the PE cross-linking (44). The dependence of grafting efficiency on the monomer and initiator concentration is indicative of a strong effect of PP additions on the yield of grafted product (Table 10.4). Irrespective of DCP and MAH ratio in a reacting system, grafting efficiency of PE/PP blends exceeds that of neat PE. According to Chaoqin Li et al. (44), this can be explained by a lower viscosity of PE/PP melt against neat PE, which makes diffusion of reagents easier and raises the homogeneity of the reacting system. On the basis of the IR spectral analysis of the grafted products, Chaoqin Li et al. (44) have concluded that in the PP/PE melt, MAH gets grafted to chains of both PE and PP. This conclusion is based on the fact that the values of characteristic band frequencies of carbonyl absorption of MAH grafted to PE/PP blends (1864.1 and 1785:6 cm1 ) are between frequency values for MAH grafted to PE (1865.2 and 1784:9 cm1 ) and to PP (1862.5 and 1785:8 cm1 ). However, if Chaoqin Li et al. (44) had separated the (PE/PP)-g-MAH blend into separate PE and PP fractions by some known methods used in polymer blend fractionation (e.g., temperature rising elution fractionation, TREF, or some others), then determination of grafting location by IR spectral analysis would have been much more certain and precise. Besides the study of MAH and DCP concentration effects, Chaoqin Li et al. (44) analyzed the effect of polymer components ratio on the parameters under


285

Figure 10.2 Effect of DCP content (a) and MAH content (b) on melt viscosity of MAH grafted LDPE/PP blend. (From Reference (44) with Permission from Elsevier.)

Table 10.4 Effect of DCP and MAH on the Grafting Degree.

MAH

DCP

1.5 1.5 1.5 1.5 0.75 3.0

0.05 0.1 0.2 0.3 0.1 0.1

Grafting degree, % MAH-g-LDPE/PP MAH-g-LDPE (90/10 blend)

[From Reference (44) with permission from Elsevier].

0.42 0.69 0.71 0.68 0.51 0.90

0.47 0.71 0.74 0.79 0.54 0.98

286

Polyolefin Blends

Table 10.5 Dependence of Properties of Initial Polyolefins, Functionalized (PP/LDPE)g-IA Blends and Unmodified PP/LDPE Blends on the Ratio of Polymer Components. T, C 190 C Test material, wt%

a, %

PP PP-g-IA [99PP/1LDPE]-g-IA [95PP/5LDPE]-g-IA [75PP/25LDPE]-g-IA [50PP/50LDPE]-g-IA [25PP/75LDPE]-g-IA [5PP/95LDPE]-g-IA [1PP/99LDPE]-g-IA LDPE-g-IA LDPE 95PP/5LDPE 75PP/25LDPE 50PP/50LDPE 25PP/75LDPE 5PP/95LDPE

— 60.2 61.8 66.8 74.2 78.3 85.1 89.8 90.6 91.8 — — — — — —

230 C

Gel MFI, MFI, Ea, g/10 min g/10 min kJ mol1 content, % 4.9 15.6 16.9 17.3 16.6 5.7 5.5 0.1 0.2 0.3 7.4 5.2 6.1 6.6 6.8 7.1

11.4 16.2 17.0 17.4 16.2 17.1 16.5 0.5 0.7 1.6 18.0 13.3 15.5 15.7 16.0 17.1

40.8 1.5 0.3 0.3 7.0 53.6 53.2 77.8 62.0 81.0 43.0 45.4 45.1 42.0 45.1 42.6

— — — — — 3.4 5.7 27.4 22.1 16.7 — — — — — —

sm, kPa

Km, rel. unit

2.3 0.4 0.3 0.5 2.0 13.6 20.2 21.3 32.6 35.0 3.1 1.5 1.7 2.2 2.4 2.7

2.5 1.5 1.5 1.6 2.1 3.0 2.2 2.1 2.0 1.9 3.0 1.8 2.1 2.2 2.1 2.0

[From Reference (46) with permission from Wiley].

investigation. The PE/PP blend composition, however, was varied within a narrow range (100:0; 90:10; 80:20) with PE predomination. Therefore, the effects observed were negligible and the analysis of data from an earlier work (44) cannot allow general conclusions about mutual influence of polymer components on functionalization of these blends. The effect of PP and PE ratios has been studied earlier (45–47) when itaconic acid (IA) was grafted to their blends containing L-101 peroxide as the initiator. Grafting was done in the extruder reactor assembled on the base of Brabender plastograph (Duisburg, Germany) equipped with the dynamic mixer (48,49). The information showing the course of grafting reaction of IA onto PP, PE, and PP/PE blends is presented in Table 10.5 and Fig. 10.3. The ratios of PP and PE in the reacting mixture strongly influence the grafting efficiency (a values). The higher the PE concentration in PP/PE blends, the higher is the grafting efficiency, whereas a higher concentration of PP causes the a-value to decrease. The extent to which the addition of PE or PP influences a-values depends on their concentrations. For relatively low amounts (up to 25 wt%) of PE added, a-value increases at a higher rate and the relationship is nonlinear. For PE concentrations between 25 and 95 wt%, a values rise monotonically with the PE concentration. The increased grafting efficiency of IA, in contrast with the additive one, at PE below 25 wt% (the continuous phase in PP/PE blends is formed by the PP phase) can


287

Figure 10.3 Effect of PE concentration in PP/PE Blends on IA grafting efficiency (a) (solid line). In Figs. 10.3 and 10.6 dotted lines stand for additive values of the tested variables calculated as the sum of the values for neat PP and PE (EPR) with account for their concentrations in blends under consideration. (From Reference (46) with permission from John Wiley & Sons.)

result from PE* macroradicals formed—as being initiated by PP* macroradicals. Judging by melt flow index (MFI) values PEs participate chiefly in the grafting reaction of IA and do not undergo recombination between themselves. It is quite probable, therefore, that at PE concentrations below 25 wt%, increased a-values, in contrast with additive ones, result from prevailing grafting of the monomer onto PE chains and not onto PP ones. The MFI values determined at T ¼ 190 C and P ¼ 5 kg were found to depend on blend composition in a more complicated manner than the grafting efficiency (Table 10.5). The data in Table 10.5 permitted distinguishing three groups of blends that showed different rheological behaviors depending on the ratio of PP to PE (45,46). The (PP/PE)-g-IA systems, for example, containing PP between 99 and 75 wt%—the continuous phase in PP/PE blends is formed by PP phase—showed higher MFI values (more than three times as high as MFI of the neat PP). Maximum magnitudes of MFI are typical of (95PP/5PE)-g-IA systems. Such blends, moreover, show a sharp decrease in the apparent activation energy of viscous flow Ea, up to its negative values for the (75PP/25PE)-g-IA system (Table 10.5). For PP/PE blends with polymer component ratios of 50:50 and 25:75, both components most likely formed the continuous phase in the blend and, as a result, the apparent viscosity was observed to rise sharply (MFI decreased). These blends have MFIs close to the values for the neat PP and PE. A further increase in the PE concentration (compositions of 5:95 and 1:99 in which PE makes the continuous phase) led to still much lower MFI values. At the same time, Ea was observed to grow (Table 10.5). It is noticeable that (PP/PE)-g-IA systems containing 1 and 5 wt% of PP have lower MFIs than PE-g-IA. The a values for these compositions are lower than additive ones (Fig. 10.3). The pattern of dependence of MFI (determined at T ¼ 230 C and P ¼ 5 kg) on the composition of (PP/PE)-g-IA systems is roughly the same as at T ¼ 190 C and

288

Polyolefin Blends

Figure 10.4 Scheme of probable chemical reactions occurring at free-radical grafting of IA onto PP/PE blends. (From Reference (46) with permission from John Wiley & Sons.)

P ¼ 5 kg (Table 10.5). The only difference is that a greater number of systems differing in composition show higher MFIs (several times exceeding MFI of the PP-g-IA and PE-g-IA), while the viscosity begins to increase sharply only with Tg, The KIQ values remain nearly constant up to about 200 MPa m1/2 s1 but decrease sharply at higher rates. This implies that b-PP is ductile up to 200 MPa m1/2 s1 (0.1 m s1) at 25 C and up to 800 MPa m1/2 s1 (0.4 m s1) at 60 C. The KIQ values of b-PP are higher than those of a-PP at

Chapter 11 Deformation Behavior of b-Crystalline Phase

335

Figure 11.26 Variation of KIQ with the crack-tip loading rate, dK/dt, for a-PP and b-PP CT specimens tested at (a) 30 C, (b) 5 C, (c) 25 C, and (d) 60 C. The upper scale gives an indication of the test speed. The arrows indicate the ductile–brittle transition in b-PP. (From Reference 31 with permission from Elsevier.)

25 and 60 C. Similarly, formation of b-form PP phase also leads to substantially higher GIQ values at 25 and 60 C. The higher fracture toughness of b-form PP is related to the formation of stress-whitened zone ahead of crack tip. TEM images reveal that extensive microcavitation in the form of dense zone of craze-like structures are developed within the stress-whitened zone. As mentioned above, bPP is much more prone to crazing than a-PP due to its lower yield stress that promotes microvoiding in the amorphous layer (15). The bundled b-lamellar structure linked by the tie molecules can easy detach from one another on loading. This lamellar detachment is accompanied by massive voiding with the simultaneous onset of a craze-like microporous structure (15,18,25–27). In the case of iPP/EPR and b-PP/EPR blends containing 15 vol% impact modifier, Grein et al. indicated that the b-nucleation had little effect on the brittle– ductile transition at 30, 5, 25, and 60 C on the basis of the results of KIQ versus dK/dt plots. Thus, elastomer particles play a major role during deformation of

336

Polyolefin Blends

rubber-modified a-PP or b-PP blends. These particles act as stress concentrators, causing cavitation and crazing of the matrix ahead of the crack tip. The elastomeric particles are therefore inferred to control the initiation and propagation of the plastic zone ahead of the crack tip during fracture of the rubber-modified blends. From the compact tension results and SEM observations, they further developed deformation maps to illustrate the deformed regions of the iPP/EPR and b-PP/EPR blends at different temperatures and crack-tip loading rates (Fig. 11.27). Little difference is

Figure 11.27 Deformation maps of (a) a nonnucleated PP/EPR with 15% EPR and (b) its b-nucleated counterpart for different temperatures and crack-tip loading rates as deduced from the fracture surfaces of compact tension specimens. A rough indication of the test speed is provided by the upper scale. () shearing, (&) shearing and crazing, (D) multiple crazing, and (X) single craze. (From Reference 32 with permission from Springer Science and Business Media.)


337

observed in the deformation modes of these two rubber-modified blends. The presence of b-phase extends ductile failure of the b-PP/EPR blend slightly to higher test speeds. It is obvious that the b-phase is more effective in improving the fracture resistance of PP homopolymer than of the rubber-modified blends. As b-PP shows ductile behavior at test velocities 1 m s1, crack-tip plasticity must be taken into account in the determination of fracture toughness. LEFM concept can still be applied to b-PP under conditions of small-scale yielding ahead of crack tip. In this regard, radius of the plastic zone must be determined and incorporated into the existing crack length (a) in solving Equation 11.8. Figure 28a and b shows the typical load–displacement curves of the b-PP compact tension specimens having different crack lengths loaded under velocities of 0.001 and 3 m s1 at room temperature. The b-PP specimens exhibit ductile mode tested at 0.001 m s1 and display brittle behavior at a high velocity of 3 m s1. These figures can be expressed in terms of Fmax versus BW1/2/f(a/w) (Fig. 11.29). The regression line of brittle b-PP specimens tested at 3 m s1 passes through the origin as expected. The slope of the line yields KIC. By contrast, the ordinate of the regression line of ductile b-PP specimens tested at 0.001 m s1 has a negative value. This is because plastic yielding takes place ahead of crack tip, and a plastic zone correction is needed to obtain intrinsic K values. The radius of plastic zone, rp, can be determined by numerical iteration method in which the f(a/W) term of Equation 11.8 is replaced by f ða þ rp Þ=W) such that all data points fall on a line through the origin as shown in Fig. 11.30a. The radius of the plastic zone is determined to be about 2:22 0:45 mm. The effective toughness (Keff) can be determined from the slope of this corrected line, that is, 4.9 MPa m1/2 (Fig. 11.30b). Similar procedures are adopted for the determination of Gc in which the effective crack length is expressed in terms of a þ dp , where dp is the diameter of plastic zone (63). Figure 11.31 shows the variations of Keff and Geff values with temperature for the b-PP specimens tested at 0.001 m s1. The Geff values are very low at low temperatures (30 and 5 C) but increase significantly at room temperature and above. The Keff values of the b-PP specimens at below and above room temperature are slightly lower than that at room temperature, reflecting temperature dependence of the yield stress. The temperature and strain-rate dependence of the mechanical behavior of polymers is well known. It is considered that in many cases the effect of an increase in temperature is similar to the effect of a decrease in strain rate. For the b-nucleated iPP/EPR blend containing 15% impact modifier, Grein and coworkers determined the radius of plastic zone at 30, 5, 23, and 60 C under 0.001 m s1 using the corrected LEFM approach to be 2:07 0:28, 2:17 0:28, 2:59 0:25, and 1:78 0:45 mm, respectively (63). The corresponding fracture toughness values of the b-nucleated iPP/EPR blend at these temperatures are 7:76 0:25, 6:63 0:18, 4:75 0:30, and 2:48 0:11 MPa m1=2 , respectively. Compared to the fracture toughness of b-PP homopolymer at 30, 5, 23, and 60 C (Fig. 11.30), the beneficial effect of EPR addition to b-PP can be only found at low temperatures, that is, below 23 C. This implies that the impact modifier improves the fracture toughness of b-PP at low temperatures only.

338

Polyolefin Blends

Figure 11.28 Force–displacement curves of b-PP compact tension specimens having different crack lengths loaded under velocities of (a) 0.001 m s1 and (b) 3 m s1 at room temperature. (From Reference 63 with permission from Elsevier.)


339

Figure 11.29 Fmax versus BW1/2/f(a/W) plots for b-PP compact tension specimens loaded under velocities of 0.001 and 3 m s1 at room temperature. (From Reference 63 with permission from Elsevier.)

11.3.3

IMPACT FRACTURE TOUGHNESS

Tjong et al. have determined the impact fracture toughness of b-PP using the falling weight Charpy impact tests on SEN specimens at room temperature under a test speed of 1.2 m s1 (64). The b-PP was nucleated from an agent consisting 0.1 wt% mixture of pimelic acid and calcium stearate (K value ¼ 0:94). Figure 11.32 shows the typical impact force–time traces of a-PP and b-PP specimens. Apparently, the a-PP and b-PP specimens fracture in brittle mode as evidenced by nearly triangular shape of load–unloading trace. In this case, Equation 11.9 can be used to determine the critical strain energy release rate at high impact speed of 1.2 m s1. The GIC values for the a-PP and b-PP specimens are determined to be 5.26 and 6.71 kJ m2, respectively. The higher impact fracture value of b-PP is attributed to the formation of fibrillated zone next to the sharp notch region on the basis of SEM observation (64). Recently, Karger-Kocsis and coworkers also determined the Charpy impact fracture toughness of the a-PP and b-PP specimens at various test temperatures under an impact speed of 1.2 m s1 (47). The b-nucleant is of quinacridone type and the amount of b-PP phase is 9.49%. Figure 11.33a and 33b shows the impact force– time traces of a-PP and b-PP specimens at room temperature and 40 C, respectively. The a-PP homopolymer is brittle at both test temperatures as expected. The impact force–time curve of b-PP shows ductile yielding behavior, particularly at room temperature, as evidenced by the occurrence of crack propagation stage after reaching maximum load. The resulting Kc and Gc values are listed in Table 11.2. The results clearly indicate that the b-PP exhibits higher impact fracture toughness than their a-PP at room temperature and at T < Tg . It is noted that the Gc value of a-PP at room temperature is comparable to that of Tjong et al. However, the Gc value of b-PP at room temperature is higher than that reported by Tjong et al. This is due to higher molecular weight of b-PP employed by Karger-Kocsis and coworkers

340

Polyolefin Blends

Figure 11.30 (a) Determination of radius of plastic zone and effective toughness for the b-PP specimens tested at 0.001 m s1. (b) Geometry independence of Keff values. (From Reference 63 with permission from Elsevier.)


341

Figure 11.31 Variations of Keff and Geff values with temperature for the b-PP specimens tested at 0.001 m s1. (From Reference 63 with permission from Elsevier.)

(Mw ¼ 103 kg mol ). The larger Gc value is derived from the formation of smallscale yielding ahead of the propagating crack tip. The improved impact fracture toughness of b-PP is related to several factors such as molecular weight and tie molecules density, lamellar arrangement, and b ! a phase transition (47). However, there are no solid experimental evidences for the b ! a phase transition in b-PP subjected to impact measurements, possibly due to short testing time periods (18).

11.3.4 Essential Work of Fracture The fracture criterion of LFEM based on the Kc is invalid when extensive plastic deformation occurs ahead of the crack tip. For ductile polymers and their blends, the

Figure 11.32 Falling weight impact load–time traces of the a-PP and b-PP SEN specimens. (From Reference 64 with permission from Elsevier.)

342

Polyolefin Blends

Figure 11.33 Characteristic load–time traces due to impact of the notched Charpy specimens (a=W ¼ 0:5) for the a-PP and b-PP homopolymers at (a) room temperature and (b) 40 C. (From Reference 47 with permission from Elsevier.)

plastic zones may even extend across the whole uncracked ligament of the specimens. In this case, nonlinear elastic fracture mechanics parameters should be adopted. The J-integral concept developed by Rice (65) and the essential work of fracture (EWF) approach proposed by Broberg (66) can be used to characterize the fracture behavior of ductile materials. The determination of a critical value of J-integral (Jc) is generally carried out through the construction of the resistance


343

Table 11.2 Impact Fracture Toughness (Kc and Gc ) of a-PP and b-PP (Mw ¼ 103 kg mol1 ) at Various Temperatures. [From Reference 47 with permission from Elsevier].

40 C a-PP b-PP

Gc , kJ m2

Kc, MPa m1/2

Sample

2.4 4.2

20 C

0 C

2.6 4.2

2.5 3.8

RT

40 C 20 C

2.2 2.6

1.8 5.7

2.0 6.0

0 C

RT

2.1 5.9

6.2 9.3

curve J–Da, where Da is the advanced crack length. The Jc value is determined at the point of intersection between the crack growth resistance (J–R) curve and the blunting line (J ¼ 2s y Da, where s y is the yield stress). The process of determination of Jc for ductile polymer is rather tedious and controversial as the application of different standard practices can yield different critical values (67–69). In postyielding fracture mechanics, the EWF concept has been successfully employed to characterize the fracture toughness of ductile polymers and tough composites due to its simplicity over conventional J-integral analysis (70–75). The EWF approach involves the determination of the total fracture energy (Wf) of notched specimen. It can be divided into two components: Wf ¼ We þ Wp

ð11:11Þ

The first term in Equation 11.11 is the essential work of fracture (We), that is, the work required to create new surfaces in inner fracture surface zone and is surface related. The second term is referred to as nonessential work (Wp). It is associated with various energy dissipation mechanisms that occur in the outer plastic zone and is volume related. Thus, Wf can be written as Wf ¼ we LB þ bwp L2 B wf ¼

Wf ¼ we þ bwp L LB

ð11:12Þ ð11:13Þ

where wf is the specific total fracture work, we and wp are the specific essential fracture work and specific plastic work, respectively, L is the ligament length, B is the sample thickness, and b is a shape factor of the plastic zone. The validity of EWF approach under plane-stress condition requires the ligament to be fully yielded prior to crack propagation. The validity range of ligament L under plane-stress condition is given by ð3 5ÞB L minðW=3; 2rp Þ

ð11:14Þ

where W is the width of the specimen and 2rp is the size of plastic zone. The ESIS protocol (76) for EWF recommends the use of double edge notched tensile specimen. Apparently, EWF concept is a simple method that consists of testing specimens with different ligament lengths, recording the area under the load displacement curve

344

Polyolefin Blends

(Wf), plotting the wf versus L diagram, and evaluating the best fit linear regression line. The we value is then evaluated by extrapolation to zero ligament length. It has been demonstrated by Mai that the plane strain essential work of fracture, wIe is equivalent to JIc if the sample thickness meets the condition: B 25 (wIe =s y ) (77,78). Otherwise, wIe is considered a near-plane-strain fracture toughness and is dependent on B. Karger-Kocsis and Varga determined the plane stress we and bwp values of b-PP using SEN and DENT specimens with dimensions of 100 mm 35 mm 1 mm under tensile mode (24,25). Figure 11.34a and b shows typical load–displacement curves for the a-PP and b-PP SEN specimens with different ligament lengths. Selfsimilarity in the shape of the load–displacement curves for the a-PP is not maintained when L 15 mm. At this stage, the samples fail with fast fracture due to an incomplete development of the plastic zone. The EWF approach becomes invalid for the a-PP. By contrast, the b-PP specimens show more ductile behavior as evidenced by nonlinearity in the force–displacement curves. The specific total fracture work

Figure 11.34 Tensile load–displacement curves of the SEN specimens at various ligament lengths for (a) a-PP and (b) b-PP. (From Reference 24 with permission from John Wiley & Sons.)


345

Figure 11.35 Specific total work of fracture vs. ligament length plots for the tensile SEN specimens of a-PP and b-PP. Note: The data related to fast fracture of a-PP at L 15 mm (filled circles) are excluded. (From Reference 24 with permission from John Wiley & Sons.)

versus ligament plots for the a-PP and b-PP specimens are shown in Fig. 11.35. It is apparent that both PP modifications exhibit the same we value of 34 kJ m2. However, the plastic work dissipation per unit volume (bwp) of b-PP is three times higher than that of the a-PP. This implies that more energy is spent in the enlarged plastic zone of b-PP, leading to the toughness improvement as revealed by the infrared thermography frames taken during the tensile loading of SEN specimens of a-PP and b-PP. The EWF measurements also yield similar we value of 32 kJ m2 for both the a-PP and b-PP using DENT specimens. The bwp values for the a-PP and b-PP DENT specimens are 2.8 and 10 MJ m3, respectively (25). Compared to a-PP, the plastic work term is 3.6 times higher for b-PP. It is noted that the same we value obtained for both the a-PP and b-PP using either SEN or DENT specimens is derived from a lack of the ligament yielding in a-PP. The accuracy of we value is doubtful despite the data related to fast fracture of a-PP at L 15 mm are neglected (Fig. 11.35). The b-PP is expected to exhibit larger we and bwp values than a-PP owing to its ductile nature. Recently, Tordjeman et al. used EWF approach to determine the fracture toughness of PP having different b-phase contents via three-point bending tests on SEN specimens with dimensions of 100 mm 10 mm 3 mm at 500 mm min1. The PP homopolymer with variable a/b phase contents but with constant crystallinity and constant spherulite size were controlled with proper heat treatment (79). Figure 11.36 is a typical plot showing the total work of fracture versus ligament length plot for the PP homopolymer containing 70% b-phase. The resulting near-plane-strain fracture toughness and plastic work versus b-phase content for PP homopolymers are depicted in Fig. 11.37a and b, respectively. It is apparent that wIe increases markedly with increasing b-phase content. It increases from 1.1 kJ m2 for a-PP specimen to more than 6 kJ m2 for PP containing 80% b-phase. The wIe value obtained for b-PP is somewhat much smaller than that reported by Karger-Kocsis

346

Polyolefin Blends

Figure 11.36 Total work of fracture versus ligament length for PP containing 70% b-phase. (From Reference 79 with permission from Springer Science and Business Media.)

and Varga (24,25). This is attributed to the difference in the thicknesses of the specimens employed. As mentioned above, a near-plane-strain fracture toughness is thickness dependent, whilst the plane strain wIe is independent of the specimen thickness (71).

Figure 11.37 (a) Near-plane-strain essential work of fracture and (b) plastic work dissipation per unit volume as a function of b-phase content. (From Reference 79 with permission from Springer Science and Business Media.)


11.4

347

CONCLUSIONS

This chapter reviews the structure–property relationship, deformation and failure behavior of b-PP homopolymer and b-PP/EPR blends subjected to tensile and impact tests. The unique sheaflike lamellar morphology of the spherulites of b-PP renders it to possess superior tensile ductility and impact strength. The molecular weights of PP and dispersed elastomer phase (EPR) influence the impact strength of b-PP and its blends considerably. Compared to a-PP, the incorporation of b-nucleator brings about distinct softening, as evidenced by lower values of yield stress at various strain rates. This behavior is also observed in the b-nucleated iPP/EPR blends. The b-PP homopolymer exhibits large strain hardening and stress whitening during tensile drawing. This is considered to be resulted from the b ! a phase transformation on the basis of TEM, DSC, and XRD measurements. However, the exact mechanism by which the strain induced b ! a phase transition takes place remains unclear. Furthermore, there is no evidence for the b ! a phase transformation during impact as revealed by the DSC results. Large plastic zone is developed in b-PP and b-nucleated iPP/EPR blends during tensile test. Similarly, extended plastic zone is observed in the b-PP during impact loading; the size of plastic zone tends to increase with increasing b-phase content. The b-phase is more effective in improving the impact strength and fracture resistance of PP homopolymer than in the rubbermodified blends. In the later case, the elastomer particles mainly control the mechanical deformation behavior. The fracture toughness of b-PP homopolymer and b-nucleated iPP/EPR blends at low strain rates (mode I) can be determined by means of the corrected LEFM approach. This approach can be applied to characterize the toughness of ductile b-PP and b-nucleated iPP/EPR blends provided that the cracktip plasticity effect is taken into account. In this regard, a plastic zone correction is needed to obtain intrinsic fracture toughness values. Finally, the EWF concept is more adequate to evaluate the fracture toughness of ductile b-PP. The specific work of fracture of b-PP is found to increase markedly with increasing b-phase content.

NOMENCLATURE a B b e_ E f ða=WÞ Fmax Gc DH Ið110Þa Ið040Þa Ið130Þa

Crack length Sample thickness Shape factor of the plastic zone Strain rate Young’s modulus Geometric factor Maximum load Critical strain energy release rate Activation energy of the plastic flow Integrated area of (110) peak of a-PP Integrated area of (040) peak of a-PP Integrated area of (130) peak of a-PP

348

Polyolefin Blends

Ið300Þb Ið301Þb K KI L y Fða=WÞ 2rp R S t T V we wp Wf W

Integrated area of (300) peak of b-PP Integrated area of (301) peak of b-PP Content of b-PP phase Stress intensity under mode I Ligament length Poisson’s ratio Calibration factor depending on sample geometry Size of plastic zone Universal gas constant Order parameter Sample thickness Absolute temperature Activation volume Specific essential fracture work Specific plastic work Total fracture work Width of sample

REFERENCES 1. S. V. Meille, D. R. Ferro, S. Bruckner, A. J. Lovinger, and F. J. Padden, Macromolecules, 27, 2615 (1994). 2. J. Varga and J. Jarger-Kocsis, J. Polym. Sci. B: Polym. Phys., 34, 657 (1996). 3. H. J. Leugering, Makromol. Chem., 109, 204 (1967). 4. J. Garbarczyk and D. Paukszta, Colloid Polym. Sci., 263, 985 (1985). 5. P. Jacobi, B. H. Bersted, W. J. Kissel, and C. E. Smith, J. Polym. Sci. Polym. Phys. Edn., 24, 461 (1986). 6. G. Shi and Z. Zhang, US Patent 5231126 (1993). 7. G. Shi and X. Zhang, Thermochim. Acta, 205, 235 (1992). 8. J. Varga, I. Mudra, and G. W. Ehrenstein, J. Appl. Polym. Sci., 74, 2357 (1999). 9. F. Chu and Y. Kimura, Polymer, 37, 573 (1996). 10. M. Obadal, R. Cermak, M. Raab, V. Verney, S. Commereuc, and F. Fraisse, Polym. Degr. Stab., 91, 459 (2006). 11. J. Liu, Z. Wei, and Q. Guo, J. Appl. Polym. Sci., 41, 2829 (1990). 12. P. M. McGenity, J. J. Hooper, C. D. Paynter, A. M. Riley, C. Nutbeem, N. J. Elton, and J. M. Adams, Polymer, 33, 5215 (1992). 13. A. Turner Jones, J. M. Aizlewood, and D. R. Beckett, Makromol. Chem., 75, 134 (1964). 14. G. Zhou, Z. He, J. Yu, Z. Han, and G. Shi, Makromol. Chem., 187, 633 (1986). 15. S. C. Tjong, J. S. Shen, and R. K. Y. Li, Polym. Eng. Sci., 36, 100 (1996). 16. R. Cermak, M. Obadal, P. Ponizil, M. Polaskova, K. Stoklasa, and A. Lengalova, Eur. Polym. J., 41, 1838 (2005). 17. D. R. Norton and A. Keller, Polymer, 26, 704 (1985). 18. S. C. Tjong, J. S. Shen, and R. K. Y. Li, Polymer, 37, 2309 (1996). 19. J. Varga and G. W. Ehrenstein, Colloid Polym. Sci., 275, 511 (1997). 20. J. Varga, J. Macromol. Sci. B: Phys., 41, 1121 (2002).


349

21. A. Menyhard, J. Varga, A. Liber, and G. Belina, Eur. Polym. J., 41, 669 (2005). 22. M. C. Cramez, M. J. Oliveira, and R. J. Crawford, J. Mater. Sci., 36, 2151 (2001). 23. T. Labour, L. Ferry, C. Gauthier, P. Hajji, and G. Vigier, J. Appl. Polym. Sci., 74, 195 (1999). 24. J. Karger-Kocsis and J. Varga, J. Appl. Polym. Sci., 62, 291 (1996). 25. J. Karger-Kocsis, Polym. Eng. Sci., 36, 203 (1996). 26. J. X. Li and W. L. Cheung, Polymer, 39, 6935 (1998). 27. J. X. Li, W. L. Cheung, and C. M. Chan, Polymer, 40, 2089(1999). 28. M. Aboulfaraj, C. G’Sell, B. Ulrich, and A. Dahoun, Polymer, 36, 731 (1995). 29. W. Jiang, S. C. Tjong, and R. K. Y. Li, Polymer, 41, 3479 (2000). 30. M. Denac, I. Smit, and V. Musil, Composites Part A, 36, 1094 (2005). 31. C. Grein, C. J. G. Plummer, H. H. Kausch, Y. Germain, and Ph. Beguelin, Polymer, 43, 3279 (2002). 32. C. Grein, Adv. Polym. Sci., 188, 43 (2005). 33. S. C. Tjong and R. K. Y. Li, J. Vinyl Technol., 3, 89 (1997). 34. S. C. Tjong, R. K. Y. Li, and T. Cheung, Polym. Eng. Sci., 37, 166 (1997). 35. H. B. Chen, J. Karger-Kocsis, J. S. Wu, and J. Varga, Polymer, 43, 6505 (2002). 36. M. Raab, J. Scudla, and J. Kolarik, Eur. Polym. J., 40, 1317 (2004). 37. F. Chu and Y. Kimura, Polymer, 37, 573 (1996). 38. G. Shi, X. Zhang, J. Zheng, and G. Zhou, Intern. Polym. Process., 10, 330 (1995). 39. F. Chu, T. Yamaoka, and Y. Kimura, Polymer, 36, 2523 (1995). 40. ASTM D256-05a: Standard Test Methods for Determining the Izod Pendulum Impact Resistance of Plastics, p. 1–20, American Society for Testing and Materials, Philadelphia, 2005. 41. ASTM D 5628-96: Standard Test Method for Impact Resistance of Flat, Rigid Plastic Specimens by Means of a Falling Dart (Tup or Falling Mass), pp. 1–10, American Society for Testing and Materials, Philadelphia, 1996. 42. T. Yoshida, Y. Fujiwara, and T. Asano, Polymer, 24, 925 (1983) 43. T. Asano, Y. Fujiwara, and T. Yoshida, Polym. J., 11, 383 (1979). 44. J. Kotek, I. Kelnar, J. Baldrian, and M. Raab, Eur. Polym. J., 40, 679 (2004). 45. J. Karger-Kocsis, J. Varga, and G. W. Ehrenstein, J. Appl. Polym. Sci., 64 (1997) 2057. 46. J. Karger-Kocsis, D. E. Mouzakis, G. W. Ehrenstein, and J. Varga, J. Appl. Polym. Sci., 73, 1205 (1999). 47. H. B. Chen, J. Karger-Kocsis, J. S. Wu, and J. Varga, Polymer, 43, 6505 (2002). 48. W. Xu, D. C. Martin, and E. M. Arruda, Polymer, 46, 455 (2005). 49. G. Coulon, G. Castelein, and C. G’Sell, Polymer, 40, 95(1998). 50. K. Cho, D. N. Saheb, J. Choi, and H. Yang, Polymer, 43, 1407 (2002). 51. K. Cho, D. N. Saheb, H. Yang, B. I. Kang, J. Kim, and S. S. Lee, Polymer, 44, 4053 (2003). 52. T. Labour, G. Vigier, R. Seguela, C. Gauthier, G. Orange, and Y. Bomal, J. Polym. Sci. B: Polym. Phys., 40, 31 (2002). 53. J. C. Wittmann and B. Lotz, Polymer, 30, 27 (1989). 54. B. Lotz, C. Mathieu, A. Thierry, A. J. Lovinger, C. De Rosa, O. Ruiz de Ballesteros, and F. Auriemma, Macromolecules, 31, 9253 (1998). 55. K. S. Macturk, R. K. Eby, and B. L. Farmer, Polymer, 37, 4999 (1996). 56. C. Grein, K. Bernreitner, A. Hauer, M. Gahleitner, and W. Neibl, J. Appl. Polym. Sci., 87, 1702 (2003). 57. G. R. Irwin, Fracture, in Encyclopedia of Physics, Vol. VI, Springer, Berlin, 1958. 58. G. R. Irwin, J. Appl. Mech., 24, 361 (1957). 59. A. Pavan, Fracture Mechanics Testing Methods for Polymers, Adhesives and Composites, in: ESIS Publication 28, D. R. Moore, A. Pavan, and J. D. Williams (eds.), Elsevier, Oxford, 2001, p. 27.

350

Polyolefin Blends

60. J. G. Williams and M. J. Cawood, Polym. Test., 9, 15 (1990). 61. ASTM Annual Book of Standards, American Society of Testing Materials and Philadelphia, PA, E-399, 1987, pp. 417–427. 62. R. Gensler, C. J. G. Plummer, C. Grein, and H. H. Kausch, Polymer, 41, 3809 (2000). 63. C. Grein, H. H. Kausch, and Ph. Beguelin, Polym. Test., 22, 733 (2003). 64. S. C. Tjong, J. S. Shen, and R. K. Y. Li, Scr. Metall. Mater., 33, 503 (1995). 65. J. R. Rice, J. Appl. Mech., 35, 379 (1968) 66. K. B. Broberg, Int. J. Fract., 4, 11 (1968). 67. ASTM E813–81: Standard Method for JIC, A Measure of Fracture Toughness, American Society for Testing and Materials, Philadelphia, 1981, p. 810. 68. ASTM E813–87: Standard Method for JIC, A Measure of Fracture Toughness, American Society for Testing and Materials, Philadelphia, 1987, p. 1968. 69. ASTM E813–89: Standard Method for JIC, A Measure of Fracture Toughness, American Society for Testing and Materials, Philadelphia, 1989, p. 700. 70. C. A. Paton and S. Hashemi, J. Mater. Sci., 27, 2279 (1992). 71. S. Hashemi, J. Mater. Sci., 28, 6178 (1993). 72. D. E. Mouzakis, F. Stricker, R. Mulhaupt, and J. Karger-Kocsis, J. Mater. Sci., 33, 2552 (1998). 73. S. C. Tjong, S. A. Xu, and R. K. Y. Li, J. Appl. Polym. Sci., 77, 2074(2000). 74. S. C. Tjong and S. P. Bao, J. Polym. Sci. B: Polym. Phys., 43, 585 (2005). 75. S. C. Tjong, S. P. Bao, and G. D. Liang, J. Polym. Sci. B: Polym. Phys., 43, 3112 (2005). 76. E. Clutton, Essential work of fracture, in: Fracture Mechanics Testing Methods for Polymers, Adhesives and Composites, ESIS Publication 28, D. R. Moore, A. Pavan, and J. D. Williams (eds.), Elsevier, Oxford, 2001 77. Y. W. Mai and B. Cotterell, Int. J. Fract., 32, 105 (1986). 78. J. S. Wu and Y. W. Mai, Polym. Eng. Sci., 36, 2275 (1996). 79. Ph. Tordjeman, C. Robert, G. Marin, and P. Gerard, Eur. Phys. J., 4, 459 (2001).

Chapter

12

Multiphase Polypropylene Copolymer Blends Francis M. Mirabella1

12.1 INTRODUCTION The most widely used multiphase polymer system is polypropylene impact copolymer. These copolymers are typically composed of isotactic polypropylene (iPP) and poly(ethylene–propylene), referred to as ethylene–propylene rubber or EPR. The world demand for all types of polypropylene is about 90 billion pounds per year. About 22 billion pounds per year of that total are impact polypropylene copolymers, referred to by various names such as high impact polypropylene (hiPP) and thermoplastic olefin (TPO). Growth is very strong at about 10% per year. These PP copolymers are primarily used in injection-molded parts for automotive, appliances, and other durable goods applications, as well as for extruded sheet and thermoforming. The wide range of physical and mechanical properties, relative ease of processing, and low density constitute these polypropylene copolymers as extremely attractive materials capable of competing with more expensive plastics in many demanding applications. The automotive industry has made TPO resins the primary choice for an increasing range of interior and exterior applications. Interior applications include instrument panels, consoles, door panels, and pillars. Exterior applications include bumpers, fascia, body side cladding, rocker panels, and cowl vent grilles. Over the last several decades, these TPO resins have progressively replaced other polymeric compositions in interior and exterior applications due to their desirable balance of properties and safety attributes. This chapter presents a discussion of the multiphase polypropylene copolymers with emphasis on the commercially important blends. The discussion will be more narrowly focused on the composition, structure, and phase morphology of these

1

Lyondell Chemical Co., Equistar Technology Center, Cincinnati, OH 45249, USA


351

352

Polyolefin Blends

commercial polymer systems to the exclusion of related systems of mainly academic interest.

12.1.1 Commercial Production Impact polypropylene copolymers are produced by various processes, but are generally characterized by the synthesis of iPP in the first reactor and EPR in the second reactor (1–3). Therefore, these systems are typically reactor blends. Postreactor blending is a common practice, but the starting material is most often the reactor blend polypropylene copolymer. In the minority are blends employing polypropylene as the starting material to which the rubber phase and other components are added. The iPP is typically the majority, that is, matrix, phase, while the EPR is the minority, that is, dispersed, phase. The EPR has low Tg and increases the impact strength, but lowers the stiffness of the system. The iPP has high crystallinity and stiffness and acts as a rigid matrix. The iPP also has high Tg . The EPR particles typically average about 0.5–1.0 mm in diameter with the entire distribution of particle sizes ranging from about 0.1 to 3 mm. The reactor grades of impact polypropylene copolymers are often compounded with other components, especially other toughening agents with low Tg, such as ethylene–propylene diene monomer (EPDM), metallocene ethylene–a-olefin copolymers, styrenic block copolymers, and so on. The compounded systems containing much higher rubber content relative to the base resin are often called thermoplastic elastomers (TPEs). In the TPEs, the rubbery components may constitute the major phase. However, TPEs include many other base resins, which are not polyolefins, such as polyurethanes, copolyamides (segmented block copolymers), copolyesters (segmented block copolymers), styrenics, and so on. Polypropylene homopolymer, compounded with EPDM in a dynamic melt-mixing/curing process, is often called thermoplastic vulcanized elastomer (TPV), which offers chemical cross-linking of the rubber phase. Paraffinic oils are always included in the melt-mixing process for viscosity control and cost. The entire range of systems in these categories is difficult to specify, since compounders have the option to blend an extremely diverse range of materials. The rigid hiPP and TPO grades account for the majority of PP copolymer demand, while the elastomeric TPE grades are in the minority. Total TPE demand is about 5.0 billion pounds per year. Other common additives to polypropylene copolymer compounds are talc, nucleating agents, clarifiers, other inorganic fillers, and so on. A recent movement toward the addition of nanofillers such as organoclay (especially oganically modified montmorillonite), TiO2, SiO2, and so on has begun to gain wider acceptance and utility.

12.1.2 Morphology of Commercial Impact PP Copolymers Excellent treatments of the compatibility, thermodynamics, phase separation (including kinetics), rheology, physical and mechanical properties, fracture phenomena, and so on have been given in numerous books, for example, the two-volume

Chapter 12 Multiphase Polypropylene Copolymer Blends

353

treatise by Paul and Newman (4). Quite a bit of attention in such treatments is inclined toward miscibility and phase separation. In the case of reactor grades or compounded polypropylene impact copolymers, however, these exhibit a phaseseparated morphology in the commercial reactors and undergo modification in that morphology in the postreactor compounding extruders. These systems are never one phase, but are already multiphasic, and remain so, when transferred into the melt state. It is the nature of the production of these copolymers, which yields systems that are always multiphasic. It should be noted, however, that the equilibrium concentrations of the components are not necessarily established until the already multiphasic system is transferred into the melt state at a specified temperature. Phase changes during this process are expected to be relatively small, but significant for the establishment of the true equilibrium concentrations of the components under the specified conditions in the melt. Therefore, considerations of miscibility, conditions for and kinetics of phase separation, and the like are not typically relevant to polypropylene impact copolymers. The development of the morphology of impact polypropylene copolymers follows a pattern from the commercial reactor to finished parts, such that the system is multiphasic throughout these conversions. However, the morphology exhibits some large changes in some steps. The form of the polymer after synthesis in the commercial reactors is typically powder, or particles in some newer processes. The powder or particles are in the range of 0.5–1.0 mm (flakes) and 1–2 mm in diameter, respectively. The morphology in the case of powder is extremely heterogeneous with some regions in the powder containing no observable EPR and other regions containing large ‘‘pools’’ of EPR in the rigid iPP matrix. This heterogeneous morphology is shown in atomic force microscopy (AFM) images of several representative powder flakes from a single lot of reactor powder in Figs. 12.1–12.4. Figure 12.1 is an example of a region in the powder containing no observable EPR, while Fig. 12.4 shows a region containing ‘‘pools’’ of EPR, which are roughly 5 mm in diameter and exhibit connectivity between EPR domains. Figure 12.2 shows a region with a very sparse population of (0.1–0.3 mm diameter spherical) EPR domains, while Fig. 12.3 shows interconnected (0.1–0.7 mm wide) channels of EPR. Note that AFM is performed on the powder by imbedding it in an adhesive, microtoming a flat surface, and imaging the surface with no further treatment, such as etching or staining used in scanning electron or transmission electron microscopy, respectively (5). In the AFM micrographs, the topographic image is on the left and phase contrast image is on the right. In the topographic image, dark areas are low and bright areas are high and in the phase contrast image dark areas are soft and bright areas are hard. Therefore, the EPR particles appear dark, since these are noncrystalline, in the phase contrast images and, in fact, also typically appear dark in the topographic images due to an artifact of the microtomy (5). The iPP matrix phase appears bright in the phase contrast image, since it is crystalline and, therefore, hard. Similar results were obtained in a study of impact polypropylene copolymer particles produced in the Basell Spheripol process (6) in which EPR concentration varied widely in particles of different sizes (7). The EPR concentration was found to decrease with increasing particle size. In a recent study, the detailed physical

354

Polyolefin Blends

Figure 12.1 Polypropylene TPO reactor powder morphology. AFM image of embedded and cryofaced-off Rx powder particle. Topographic image is on left and phase contrast image is on right. In the topographic image, dark areas are low and bright areas are high and in the phase contrast image dark areas are soft and bright areas are hard.

Figure 12.2 Polypropylene TPO reactor powder morphology. AFM image of embedded and cryofaced-off reactor powder particle.


355


structure of the hiPP particle produced in a two-stage laboratory process was studied, as well as the formation and distribution of the EPR (8). The commercially produced polymer is normally formed into pellets in the commercial plant by mechanical mixing in a pelletization extruder (9) for convenient shipment by railcar, truck, and so on. In order to produce a consistent material for


356

Polyolefin Blends

Figure 12.5 Polypropylene TPO extruded pellet (made from reactor powder in Figs. 12.1–12.4) morphology. AFM image of embedded and cryo-faced-off reactor powder particle.

commercial sale, the pellets are required to exhibit a much more homogeneous dispersion of the EPR in the iPP matrix than exhibited in the commercial reactor product. This is accomplished by the input of a large amount of work in the form of shearing into the molten polymer in the extruder. The extruder design is judiciously done to maximize dispersion of the EPR phase, but minimize polymer degradation. The pellets in fact normally do exhibit homogeneous dispersion of the EPR, although particle size distribution is typically relatively broad. This is shown in Fig. 12.5 for pellets formed from the powder in Figs. 12.1–12.4. It may be observed in Fig. 12.5 that the EPR particle size distribution is broad and that particles are generally round (except for obvious coalescence of neighboring particles) indicating that EPR particles are spherical droplets in the molten state. The EPR particles in Fig. 12.5 have an included hard phase, which is due to crystalline polyethylene copolymers that segregate away from the iPP matrix due to their incompatibility (10,11). The fabrication of useful articles from the pellets is the usual next step (8). Injection molding is one of many processes that are used to fabricate articles. Injection molding is used to fabricate a variety of articles, such as automotive and appliance parts. Figure 12.6 shows AFM images of an injection-molded tensile bar used for mechanical property testing, which was molded from the commercial pellets in Fig. 12.5. The specimen was taken from the ‘‘core’’ of the bar, where residual stresses are minimal due to slow cooling (12). This core specimen of the injectionmolded bar in Fig. 12.6 exhibits very similar morphology to the pellets in Fig. 12.5, although, some evidence of EPR particle coarsening may be observed in Fig. 12.6. This coarsening is due to incorporation of material in smaller particles into larger particles, which will be discussed later. In some applications, the commercial plant


357

Figure 12.6 Polypropylene TPO injection-molded bar (made from commercial pellets in Fig. 12.5) morphology. AFM image of cryo-faced-off section of core of injection-molded bar.

pellets are often compounded with additional components, as discussed in Section 12.1.1, and are repelletized for subsequent use in the fabrication process.

12.2

DISPERSIVE MIXING DURING PROCESSING

Figures 12.1–12.6 show the radical change in EPR particle morphology from reactor powder to pellets, but the relatively static morphology from pellets to fabricated articles. This is due to the great efficiency of commercial-scale corotating twin-screw pelletization extruders (8). The EPR phase is efficiently dispersed and attains the ‘‘stationary’’ value of particle size, as described by theoretical treatments of droplet breakup and coalescence (13–15). This droplet breakup and coalescence occurs in the molten state of the viscoelastic iPP and EPR, matrix and dispersed phases, in the extruder under a complex strain field, which is a combination of nonuniform, transient shear and elongational fields. Further, a variable temperature profile is used along the barrel of the extruder causing complex variation in the viscoelastic properties of these components. A set of empirical equations was obtained by Wu to describe the dispersed phase average particle size obtained after dispersive mixing in an extruder (13). The equations were based on the case of a Newtonian drop suspended in a Newtonian matrix, that is, Taylor’s theory (16,17) with an extension to the case of a viscoelastic drop in a viscoelastic matrix. The empirical data employed were for blends containing 15 wt% dispersed phase and 85 wt% matrix phase. The particle size was found to be critically dependent on the ratio of the dispersed phase to the matrix phase melt

358

Polyolefin Blends

viscosity (p). The equation for the case of higher melt viscosity of the dispersed phase than the matrix phase is an ¼

4gp0:84 Ghm

ðfor p > 1Þ

ð12:1Þ

where p ¼ hd =hm , hd is the melt viscosity of the dispersed phase, hm is the melt viscosity of the matrix phase, G is the effective shear rate, an is the number-average particle diameter, and g is the interfacial tension. The same equation is used for p < 1, except that the exponent is 0:84. The two equations for p > 1 and p < 1 predict a V-shaped curve with particle size at a minimum at p ¼ 1. This empirical treatment might be expected to be fairly realistic, due to its recognition of the nonNewtonian behavior of the multiphase systems and the finite dispersed phase concentration. In contrast and for comparison, the theoretical equation from Taylor’s theory (16, 17) for a Newtonian drop suspended in a Newtonian matrix with the concentration of the dispersed phase particle assumed to be vanishingly small is an ¼

4gðp þ 1Þ Ghm p 19 4 þ4

ðfor p < 2:5Þ

ð12:2Þ

The calculated particle diameters from Equation 12.2 may be considered a lower limit, that is, the ‘‘Taylor limit,’’ due to the assumption of Newtonian behavior of the system and vanishingly small concentration of the dispersed phase. Polymers exhibit non-Newtonian behavior, namely, the droplets elongate elastically before breaking. This behavior corresponds to an increase in interfacial tension, and therefore, particle size increases as predicted by Equation 12.1, over that predicted from Equation 12.2. (This is discussed below and can be seen in the last two columns of Table 12.3). A study was done to determine the effect of extruder conditions on dispersed phase particle size for an hiPP. Table 12.1 presents polymer data on the hiPP (assumed to be a binary iPP/EPR blend) studied. Table 12.2 shows the Berstorff twin-screw corotating extruder conditions. Extruder runs were done under low (0-series) and high (1-series) shear rate conditions and nitrogen purge. Temperature profiles were somewhat different along the extruder for the 0- and 1-series experiments. Table 12.3 presents particle size determined by scanning electron microscopy (SEM), along with the calculated values of an from Equations 12.1 and 12.2. The dn, dw, and dz values in Table 12.3 are the number, weight and Z-average diameters determined by SEM, respectively. The molecular weight of the EPR in the impact PP Table 12.1

PP/EPR Blend Data.

Polymer iPP/EPR blend iPP EPR (33 wt% C2)

wt%

Mw ð103 Þ

— 84 16

620 510 935

Mw =Mn 10.3 7.3 20.4

h @ 230 C @100 s1 (Pa s) 19000 9400 82600

g (mN/m1) 0.2 — —

Chapter 12 Multiphase Polypropylene Copolymer Blends Table 12.2 Run ID 0-1 0-2 0-3 1-1 1-2 1-3

359

Berstorff Twin-Screw Corotating Extruder Conditions. T ( F) zones 1–7

Residence time (s)

257–422 323–454 325–416 360–419 394–447 394–411

Shear rate (s1)

Feed rate (lbh1)

115 115 115 172 172 172

250 250 250 250 250 250

75 75 75 87 87 87

is approximately two times that of the iPP. Higher molecular weight of the EPR, relative to the iPP in impact PP resins, is often the case in hiPP resins. The melt viscosities at the operating shear rate of the extruder followed these trends; therefore, the values of p were much larger than 1 in this case (see Table 12.1). The concentration of the dispersed phase (EPR) was 16 wt%. The particle size observed (dx values in Table 12.3) was two to three orders of magnitude larger than that calculated (an values in Table 12.3) from Equations 12.1 and 12.2. The much larger particle sizes observed compared to those calculated are probably due to the very high values of hm and p, and, also, due to the effects of flow-induced coalescence, as has been clearly demonstrated for similar systems (18,19). In those studies it was found that the observed particle size for systems with higher dispersed phase concentration increased exponentially at high dispersed phase concentration. However, the values of hm and p were significantly smaller (950 Pa s and 0.9, respectively) in that study than those in the present study. Therefore, it is not surprising that the observed particle sizes in this study exceeded those calculated by Wu’s empirical equation and by Taylor’s theoretical equation. This indicates the very crude nature of the predictions and the necessity of experimental determination of particle morphology in dispersively mixed systems. Although, the 0-series was run at lower shear rate than the 1-series, the experimentally observed particles sizes were larger in the higher shear rate 1-series. The cause of this observation is not known, but might relate to the differing extruder temperature profiles.

Table 12.3 Particle Size Data for Dispersed Phase in hiPP a.

Run

dn ðmmÞ

0-1 0-2 0-3 1-1 1-2 1-3

0.49 0.58 0.51 0.57 0.57 0.61

a

dw ðmmÞ 0.50 0.59 0.53 0.61 0.58 0.62

dz ðmmÞ

dw =dn

dz =dw

Wu limit, an ðmmÞ

0.63 0.73 0.67 0.76 0.72 0.82

1.02 1.02 1.04 1.07 1.02 1.02

1.26 1.24 1.26 1.25 1.24 1.32

0.0046 0.0046 0.0046 0.0030 0.0030 0.0030

From Reference 28 with permission from John Wiley & Sons, Inc.

Taylor limit, an ðmmÞ 0.00015 0.00015 0.00015 0.00010 0.00010 0.00010

360

Polyolefin Blends

12.3 MOLECULAR STRUCTURE OF IMPACT PP COPOLYMERS The nominal components of impact PP that are targeted in commercial production are iPP and EPR containing about 25–50 wt% ethylene, so that the EPR is essentially noncrystalline. However, the composition of these copolymers is more complex, due to the broad composition distribution of the ethylene/propylene copolymers, typically made in the second reactor, and to a lesser extent due to the tacticity distribution of the iPP, typically made in the first reactor. The structure of a commercial impact PP was determined by preparative temperature rising elution fractionation (P-TREF), followed by analysis of the fractions by nuclear magnetic resonance (NMR) spectroscopy in order to determine chemical composition and tacticity (10,20). It was found that the hiPP (MFR ¼ 6) was composed of about 75 wt% iPP with average tacticity of about 97% and exhibiting a broad tacticity distribution extending to very low values, 17 wt% EPR containing an average of about 48 wt% ethylene and exhibiting a broad composition distribution of about 15–80 wt% ethylene and about 8 wt% of an ethylene-rich copolymer containing 92 wt% ethylene. It should be noted that, although these commercial PP copolymers are often called ‘‘block’’ copolymers, no evidence was found in this study to indicate that there was significant block copolymer formation. The ethylene–propylene copolymers made in the second reactor are random copolymers and no detectable block copolymers are formed between first and second reactor polymers. That is, the iPP is ‘‘dead’’ polymer upon arriving in the second reactor. This result has been found for a wide range of commercial PP copolymer resins. The compositional structure is further complicated by the molecular weight heterogeneity of the various components. For the case of this MFR ¼ 6 hiPP, which had Mw ffi 380,000 and Mw =Mn ¼ 6:3, the iPP had Mw ffi 382,000 and Mw =Mn ¼ 4:3 and the EPR had Mw ffi 342,000 and Mw =Mn ¼ 14:2. Therefore, theoretical calculations for the prediction of the properties of these systems are complicated by the heterogeneity in composition and molecular weight. The composition and molecular weight of the various components of the hiPP and TPO resins may vary widely. Although, this is the case, theoretical calculations are routinely done under the assumption that these systems are simple binary blends of homogeneous (composition and molecular weight) iPP and EPR.

12.4 COARSENING IN MULTIPHASE PP COPOLYMER SYSTEMS 12.4.1 Background It was argued above that phase separation from a one-phase melt was not typically relevant to systems based on impact polypropylenes. Further, it was shown how the initial particle size distribution is created in an extruder, due to efficient dispersive mixing. Also, it was shown that this initial particle size distribution did not tend to be greatly modified in subsequent processing steps, namely part fabrication.


361

However, it was mentioned that coarsening of the dispersed phase particles is a potentially important effect in systems in which the two-phase melt experiences, at least partly, a quiescent state. In the case in which a polymer system begins as a one-phase melt, the sequence of events is as follows: (1) rapid liquid–liquid phase separation proceeding with continuous change in phase compositions and volume fractions, (2) attainment of ‘‘equilibrium,’’ that is, stationary values of phase compositions and volume fractions and particle size distribution, (3) the excess free energy of the system due to excess interfacial energy of the small dispersed phase droplets (or particles) may be reduced by the much slower process of coarsening of the droplets, during which process the phase compositions and volume fractions remain constant. The late-stage coarsening process occurs over a time scale many orders of magnitude larger than the early-stage phase separation process. The PP copolymers do not normally exist as a one-phase melt; therefore, only the coarsening process is of material interest for these systems. The demixing of the immiscible components in an initially one-phase melt proceeds rapidly due to a thermodynamic instability, this is spinodal decomposition, or a thermodynamic metastability (i.e., nucleation and growth of a new phase). However, this distinction of the character of the liquid–liquid phase transition is largely academic for the purposes of this work, because in either case the system rapidly produces droplets of a new phase (i.e., for off-critical mixtures). The demixing process itself was not studied in detail in this work. However, further details of this process may be found elsewhere (21). The production of droplets of a new phase in these immiscible systems is extremely rapid compared with the subsequent coarsening of the system to the final morphology. The coarsening of the phase-separated system can occur by two mechanisms. Particle diffusion, collision, and coalescence is one mechanism. Particle diffusion occurs in the quiescent melt by Brownian motion of the particles. Another mechanism is evaporation and condensation, called Ostwald ripening. Ostwald ripening occurs by molecular diffusion of the minor component, which primarily makes up the minor phase particles, through the matrix phase. This results in ‘‘evaporation’’ of particles smaller than a critical radius by diffusion of the minor component out of these and growth of particles larger than the critical radius by ‘‘condensation’’ of the diffusing molecules into these. A simplistic rationale for the differing timescales of phase separation, for example, by the nucleation and growth process, and the subsequent coarsening process may be developed as follows. At time zero, the molecules of a binary solution, with molecules A and B being the major and minor components, respectively, are intimately mixed at the molecular level (i.e., there is no macroscopic phase segregation). This system has an extremely high driving force for segregation of the two types of molecules, since the solution is highly supersaturated. Therefore, the molecules separate rapidly into two macroscopic phases. This results in two phases with the thermodynamic equilibrium concentrations of B in the A-rich phase and A in the B-rich phase. At this point of the very rapid process, the system still has significant excess of energy due to the large surface area of the small particles of the minor phase B. In the next step of the process, the particles of the minor phase, B,

362

Polyolefin Blends

grow relatively slowly by combining of the nuclei of B. This occurs by the diffusion of molecules of B through the dilute solution of B molecules in the A matrix. The driving force for this process is different from the first process. In this case, the B nuclei have extremely high surface area and, therefore, extremely high cumulative interfacial tension against the A matrix. The driving force then is to minimize the surface area, thereby minimizing the cumulative interfacial tension between A and B. This is done by simply increasing the particle size of B. The minimum of the interfacial tension is one particle of B in the A matrix. The slower coarsening process is of interest because it occurs on a timescale comparable to melt processing procedures, such as pelletization, injection molding, film blowing, and the like. Coarsening can occur by coalescence or Ostwald ripening, or by both simultaneously. The coarsening of a two-phase immiscible system has been kinetically modeled for the case of metal alloys. The classical derivation was done, following the theory of Ostwald ripening (22) for immiscible metals in which grains of a new phase grow from a matrix of the original supersaturated solution. Again, the first step is very rapid (i.e., nucleation and growth) and is not included in the kinetic scheme of the model. The second step treats the slower growth of the grains of the minor phase in order to reduce the interfacial tension between the grains and the matrix. The results of this derivation for metals are most compactly expressed by the Lifshitz–Slyozov equation for a binary system (23): r 3 ðtÞ ¼ r 3 ð0Þ þ Kt

ð12:3Þ

where rðtÞ and rð0Þ are the particle radii at time t and t ¼ 0 (where zero time is defined as the beginning of coarsening, or some arbitrary time during coarsening, in the longtime regime) and K is a constant, which is independent of the volume fraction of the coarsening phase f. Although the overall driving force of the coarsening process is generally accepted to be a reduction of the interfacial area (and concomitantly the system energy), the complex diffusional processes occurring during coarsening have undergone limited study. A central assumption in the theory leading to equations of the form of Equation 12.3 is that coarsening occurs for the case of infinitely separated, noninteracting spheres. Therefore, it is not certain that polymeric systems with high volume fraction of the coarsening phase will generally adhere to this theory developed for metals. An alternative mechanism for the longtime coarsening regime, which also follows a r t1=3 dependence, is diffusion and coalescence. Coalescence occurs by the movement of the dispersed phase particles through the matrix by Brownian motion (diffusion), collision, and formation of fewer, larger particles (24,25). Coalescence follows the same law as stated in Equation 12.3. The rate constant K in Equation 12.3 can be expressed for Ostwald ripening as KOR ¼

8 DVm gCea 9 RT

ð12:4Þ

where D is the molecular diffusion coefficient of the minor component in the matrix phase (usually the mutual diffusion coefficient), Vm is the molar volume of the minor


363

component, g is the interfacial tension between the matrix and droplet phase, Cea is the equilibrium concentration (mole fraction) of the minor phase in the matrix phase, R is the gas constant, and T is the absolute temperature. As noted in Section 12.1.2 the equilibrium concentrations of the components are not necessarily established until the already multiphasic system is transferred into the melt state at a specified temperature. Although phase changes during this process are expected to be relatively small, these may be significant for the establishment of the true equilibrium concentrations of the components under the specified conditions in the melt, for example, Cea . The rate constant K in Equation 12.3 can be expressed for coalescence as (26) Kc ¼

DBr ¼

2kTf ph

ð12:5Þ

kT 6ph r

ð12:6Þ

Particle diameter

Long-time regime

log dw

Short-time regime

where k is the Boltzman constant, f is the volume fraction of the minor (droplet) phase, h is the melt viscosity of the major phase, and r is the particle radius. DBr is the droplet diffusion coefficient, due to Brownian motion (26). It may be noted that the coalescence mechanism depends on the reciprocal of the melt viscosity (Eq. 12.5). The two different timescales for phase separation and coarsening are shown schematically in Fig. 12.7. As may be seen in Fig. 12.7, particle diameter increases

log time dwo

0 Time

Figure 12.7 Schematic representation of the short-time and longtime coarsening regimes. The diameter at the beginning of the longtime regime (dw0) corresponds to zero time in Equation 12.3. Note that the lower time axis represents an unspecified time dependence for the two regimes. However, in the coarsening regime the dependence is specified as log time. (From Reference 28 with permission from John Wiley & Sons, Inc.)

364

Polyolefin Blends

on a steeply ascending line in the short-time demixing regime, while it increases on a much less steeply ascending line in the longtime coarsening regime. The changeover from the short-time phase demixing regime to the longtime coarsening regime has been interpreted as ‘‘pinning,’’ that is, cessation of morphology modification (27). However, the morphology of the system continues to change, that is, coarsens, in the longtime regime, but at a much slower rate of change as compared to the short-time regime (28). A qualitative consideration of the two mechanisms, according to Equations 12.4–12.6, logically results in the assumption that coalescence is only favored in systems with lower melt viscosity matrices, while Ostwald ripening is favored in higher melt viscosity matrices. Additionally, Ostwald ripening is only favored if the minor phase has finite solubility in the matrix phase. In fact, a quantitative comparison of theory to experiment was made in which binary blends of essentially monodisperse components were made by Crist and Nesarikar (29). That study showed that coalescence could be effectively suppressed by increasing the melt viscosity h of the matrix, thereby decreasing the diffusion DBr due to Brownian motion and, thereby, inhibiting collision of the droplet phase. Conversely, Ostwald ripening could be suppressed by decreasing either the rate of molecular diffusion D of the minor phase or the solubility Cea of the minor phase in the matrix phase. By the manipulation of these parameters, the coarsening mechanism was modified from nearly pure coalescence to equally coalescence and Ostwald ripening to nearly purely Ostwald ripening. Previous studies of the coarsening of PP copolymer systems were conducted in a way that the one-phase melt was artificially produced (for these normally multiphase systems) to establish a true ‘‘zero time’’ for the onset of late-stage coarsening (30). To ensure the homogeneity of the initial system the ex-reactor hiPP powder was dissolved in a solvent and precipitated by a nonsolvent to form a one-phase ‘‘molecular dispersion.’’ This artificially produced system was then stored in the quiescent, isothermal melt state for various times (5 s to 1 h) to observe the phase separation and subsequent coarsening processes. The system was subsequently quenched in ice water to produce the ‘‘frozen’’ liquid, which was assumed to be an unaltered record of the system in the melt state. The coarsening behavior of such systems would find application for the case of thick moldings of these multiphase polymers or for other unmixed melts for which coarsening of the EPR phase in the quiescent melt would play a significant role in the formation of the final two-phase morphology. Therefore, the study was done on the quiescent melt, although it was recognized that subsequent mechanical mixing of the system may have a profound effect on the phase morphology. For the typical commercial polypropylene copolymer systems the viscosity of the matrix phase is quite high, and the molecular diffusion and solubility of the minor phase component in the matrix phase are relatively high. These factors tend to favor the evaporation/condensation, that is, Ostwald ripening, mechanism and suppress the coalescence mechanism in these systems. Mirabella and coworkers studied a series of multiphase systems, including a hiPP (30), a high density polyethylene (HDPE)/ hydrogenated polybutadiene (HPB) blend, (31) and an unbranched PE molecular


365

weight fraction/HPB blend (32), in order to compare the theoretical predictions of coarsening to the experimental observations.

12.4.2 Coarsening of High Impact Polypropylene The coarsening of the EPR-rich phase in the hiPP is shown in scanning electron micrographs in Fig. 12.8. Coarsening was monitored for the ‘‘molecular disper-

Figure 12.8 Scanning electron microscopy photomicrographs of quiescently coarsened and ice water quenched specimens of impact polypropylene copolymer (hiPP). Magnifications are given above each photomicrograph. (From Reference 28 with permission from John Wiley & Sons, Inc.)

366

Polyolefin Blends

sion’’ for storage times in the melt of 5 s to 60 min at 193 C. Each scanning electron micrograph in Fig. 12.8 corresponds to a different specimen of the same material stored for the specified time. The commercial ex-reactor hiPP powder that was studied (molecular structure provided in Reference 28), of course, contained iPP and EPR components with broad molecular weight distribution and in the case of the EPR broad composition distribution. Therefore, the calculation of theoretical rate constants for coalescence and Ostwald ripening was not possible, without extreme assumptions. However, this system exhibited the theoretically predicted temporal dependence of coarsening, that is, r t1=3 , as can be observed in Fig. 12.9. Figure 12.9 shows the weight-average particle diameters versus time of storage in the melt from the data as in Fig. 12.8. The slope of the linear least squares fitted line in Fig. 12.9 is 0.32 and the coefficient of determination is R2 ¼ 0:97. The predictions of the coarsening theory that are embodied in Equation 12.3 are as follows. The temporal exponent for the rate of radial growth of the particle is 1/3. Therefore, the radius should vary with t1=3 . It is assumed that the total volume (VTot ) of the coarsening phase is invariant with time after the first-order transition (i.e., phase segregation). Therefore, the number of particles is proportional to VTot =r 3 ðtÞ. According to Equation 12.3 then, the number of particles per unit volume, N, is proportional to t1 . The number of particles per unit area NA (as observed in scanning electron photomicrographs) is related to the particles per unit volume N as follows: NA ¼ Ndn

ð12:7Þ

The number of particles per unit volume N as a function of storage time in the melt for the ice water quenched (from data as in Fig. 12.8), as well as another set of more

0.1 0 –0.1 –0.2

log dw

–0.3 –0.4 –0.5 –0.6 –0.7 –0.8 –0.9 0

0.5

1

1.5

2

log t

2.5

3

3.5

4

Figure 12.9 The weight-average dispersed phase particle diameter (dw ; mm) versus time in the melt (t s) for quenched specimens of the hiPP. The line is the linear least squares fitted to the data. (From Reference 28 with permission from John Wiley & Sons, Inc.)


367

3.5

log NA/dn

3

2.5

2

1.5

1 0.5

1

1.5

2

2.5

3

3.5

4

log t

Figure 12.10 The number of particles per unit volume N (where N ¼ NA =dn ) versus time in the melt (t s) for the (O) quenched and (&) bench-top cooled specimens. (From Reference 28 with permission from John Wiley & Sons, Inc.)

slowly cooled (bench-cooled) samples, is shown in Fig. 12.10, where NA =dn is plotted versus the time in the melt for these specimens (30). The initial slope of this plot is 0:16 and the slope at longer time is 1:02 (R2 ¼ 0:96). This appears to indicate that the expected N t1 behavior is not approached until longer time (100 s). It has been observed in this and other phase-segregated systems that the shortest time specimens (5 s in the melt) consistently exhibited larger particle diameters than expected from the t1=3 dependence. It may therefore be inferred that specimens that spend only 5 s in the melt have not yet entered the longtime coarsening regime. These short times may actually be in the short-time regime, that is, before coarsening, which is not well understood (33,34). The short-time regime may be complicated by simultaneous occurrence of the demixing process and the coarsening process in this time regime (21). The coarsening process is predicted to exhibit a self-similar nature at long time (33). That is, the particle size distribution is predicted to remain constant in the longtime regime. The particle size data for the ice water quenched specimens exhibited a trend of broadening particle size distribution from very narrow distributions at short time to slightly broader distributions at long time. The corresponding particle size distributions (PSDs) for the ice water quenched specimens are presented in Fig. 12.11 (30). The PSDs are plotted as reduced chord length versus frequency, where reduced chord length is the chord length (Li ) divided by the number-average chord length (Ln ). A trend of broadening PSD with increased time in the melt is clearly observed and may be due to the shift from the short-time regime to the longtime coarsening regime, as shown schematically in Fig. 12.7. Little is known about the nucleation and growth process and the PSD that is produced in the shorttime regime (33, 34). Further, the shift from the short-time regime to the asymptotic longtime regime, and its effect on PSD, has been little studied (33).

368

Polyolefin Blends

Figure 12.11 The particle size distributions for the quenched specimens plotted as frequency ½ f ðNÞ versus the reduced chord length ðLi =Ln ) for (—)5 s, () 45 s, (- - -) 2 min, (&) 8 min, and (o) 60 min storage time in the quiescent melt. (From Reference 28 with permission from John Wiley & Sons, Inc.)

12.4.3 Coarsening of Model Blends The coarsening of a model blend consisting of a hydrogenated poly(1,4-butadiene) with Mn ¼ 93,000,Mw ¼ 111,000, and Mw =Mn ¼ 1:2 and 100 ethyl branches/1000 total carbon atoms (equivalent to a poly(ethylene-butene-1) copolymer with mole fraction butene-1 of 0.205) and a commercial HDPE with Mn ¼ 22,000, Mw ¼ 159,000, and Mw =Mn ¼ 7:2 and 0 branches/1000 total carbon atoms was studied (31). As in the previous study (30), the one-phase melt was artificially produced (for this normally two-phase system) to establish a true ‘‘zero time’’ for the onset of latestage coarsening and this system exhibited the theoretically predicted temporal dependence of coarsening, that is, r t1=3 . Although this system was somewhat idealized, relative to the hiPP, it is still a multicomponent system, due to the broad molecular weight distribution of the HDPE. Using some approximations (described in Reference 29), the theoretical rate constant for Ostwald ripening was calculated from Equation 12.4 and was KOR ¼ 3:6 1018 cm3 s1 . The experimentally determined rate constant (K in Equation 12.3) was Kexp ¼ 4:8 1018 cm3 s1 . The agreement between the experimentally measured and the theoretically calculated rate constants is quite good. The ratio of these two rate constants is Kexp =KOR ¼ 1:3. This was taken as an indication that the coarsening occurred by Ostwald ripening, although the theoretical constant for coalescence was not calculated. However, the agreement may be somewhat fortuitous due to the uncertainty in the parameters used to calculate the theoretical rate constant. Nonetheless, the good agreement between the theoretical and experimental values strongly suggests that the theory is particularly applicable to this polymer blend system.


369

Each of the previous two systems (30,31) was polydisperse in at least one component, thereby causing the systems to be multicomponent as opposed to binary. The fact that these systems were not simple binary systems made the comparison of the experimental data to the theory more difficult and uncertain. Notwithstanding this fact, the agreement between experiment and theory was remarkably good. In another study (32), a binary blend was formed using an HDPE fraction, containing 0 branching (NIST SRM 1484) and Mn ¼ 100,500, Mw ¼ 119,600, and an HPB containing 100 ethyl branches per 1000 C atoms and Mn ¼ 96,000, Mw ¼ 111,000. Again, the one-phase melt was artificially produced (for this normally two-phase system) to establish a true ‘‘zero time’’ for the onset of late-stage coarsening and this system exhibited the theoretically predicted temporal dependence of coarsening, that is r t1=3 . This system was assumed to be a binary blend, since molecular weight and composition distributions were very narrow. All constants (defined in Section 12.4.1) for this system are presented in Table 12.4. Also presented in Table 12.4 are the theoretical rate constants for Ostwald ripening and coalescence KOR ¼ 0:86 1018 cm3 s1 and KC ¼ 3:6 1020 cm3 s1 , respectively, from calculations using Equations 12.4–12.6. It may be observed that these theoretical rate constants (KOR and KC ) differ by about two orders of magnitude. The experimentally determined rate constant was Kexp ¼ 1:23 1018 cm3 s1 and was in fairly good agreement with that for Ostwald ripening. The ratio of these two rate constants is Kexp =KOR ¼ 1:4, which is similar to that obtained in the previous study (31). Therefore, this result indicates that this binary system coarsens by Ostwald ripening and that coalescence is negligible. The results for the above, binary system, lent credence to the conclusion that the coarsening in all these polymer–polymer blends (30–32) with relatively high melt viscosity (i.e., high molecular weight) matrices was due to Ostwald ripening. This was also supported by the work of Crist and Nesarikar (29), which showed that

Table 12.4 Parameters for Calculation of Theoretical Rate Constants for Coarsening by Ostwald Ripening (Evaporation/Condensation) and Coalescence for HDPE (Fraction)/HPB Blenda. Ostwald ripening (evaporation/condensation) D cm2 s1 1:91 1011

Vm (cm3 mol1)

g (erg cm2 )

Cea (mol fraction)

T (K)

KOR (cm3 s1)

1:44 105

1.20

0.011

450

0:86 1018

Coalescence DBr (cm2 s1 ) 2:00 1013 a

h (poise)

f (vol fraction)

T (K)

KC ðcm3 s1 Þ

1:10 105

0.10

450

3:60 1020


370

Polyolefin Blends

polymer systems with high molecular weight matrices coarsen mainly by Ostwald ripening. Further, the excellent agreement of the experimental rate constants with that for Ostwald ripening supported this conclusion. Ratios of Kexp =KOR ¼ 1:3 and 1.4 appear to strongly support Ostwald ripening. This author considers the case to be decisive. However, alternate views do exist. Fortelny and coworkers argued that the theory of coalescence of Smoluchowski (24) is based on dilute systems, derived a theory of coalescence for concentrated systems (35), and analyzed literature data according to their derived theory (36). They argued that phase-separated polymer systems in the literature, such as discussed above, do not satisfy the criteria of the Smoluchowski theory and that for systems with greater than 10% of the droplet phase the average distance between particles is substantially smaller than the droplet radius. They described a coalescence mechanism for such systems in which the forces between neighboring droplets were comparable to those from Brownian motion, leading to drainage of the matrix film between particles followed by droplet coalescence. Literature data were claimed to fit their theory of coalescence for concentrated systems better than other theories. The main thrust of the work of Fortelny and coworkers is toward an alternate mechanism to Ostwald ripening, that is, the mechanism of coalescence of neighboring droplets described above. No new experimental evidence was provided by Fortelny and coworkers to support the alternate mechanism they proposed.

12.4.4 Interfacial Effects in Polypropylene Copolymer Systems Interfacial effects on multiphase polymer systems have been of interest to polymer scientists. For example, Koberstein and coworkers demonstrated the compatibilizing effects of block copolymers in ternary blends of polystyrene/polybutadiene/ poly(styrene-block-butadiene) (37). These workers showed that the block copolymer produced a sharp decrease in interfacial tension. Torkelson and coworkers applied this technology to HDPE/polystyrene blends compatibilized with styrene/ethylene– butylene/styrene (SEBS) triblock copolymers blended in an intensive solidstate shear pulverization process (38). They showed that the SEBS acted as a compatibilizer, manifested by a decreased quiescent coarsening rate: 3.5, 5, and 10 wt% SEBS resulted in a 10-fold and 30-fold decrease in rate, and cessation of coarsening, respectively. Similar results were obtained by Cavanaugh and Nauman for polystyrene/polybutadiene and polystyrene/polyisoprene blends compatibilized with polystyrene/poly(styrene-random-butadiene) diblock copolymer (39). These are examples of marked deceleration of coarsening or true ‘‘pinning’’ of the dispersed phase droplets against coarsening by the use of interfacial agents. The application of this strategy to commercially important polyolefin multiphase systems to produce ‘‘tuned’’ and/or stabilized morphologies is certainly attractive. However, the required interfacial agents for this purpose have


371

historically been unavailable or prohibitively expensive. Recently, this situation may have changed due to the development of polyolefin block copolymers by Dow Chemical Co. (40). It remains to be demonstrated that such olefin block copolymers (OBC), which Dow has introduced commercially as Infuse1 OBCs (41), may be effective in controlling the morphology of related multiphase polyolefin systems. The formation, optimization, and commercialization of polypropylene nanocomposites is presently a field of active development (42). A recent study of TPO

Figure 12.12 AFM phase images of (a) TPO-0 (0 wt% clay), (b) TPO-1 (0.6 wt% clay), (c) TPO-3 (2.3 wt% clay), (d) TPO-4 (3.3 wt% clay), and (e) TPO-6 (5.6 wt% clay). (From Reference 42 with permission from John Wiley & Sons, Inc.)

372

Polyolefin Blends

Figure 12.13 EPR number-average particle diameter versus weight percent clay in the TPO. (From Reference 42 with permission from John Wiley & Sons, Inc.)

organoclay nanocomposites by Mirabella and coworkers showed that the EPR particle morphology in the TPO exhibited a progressive breakup and decrease in particle size, as clay loading increased in the range from 0.6 to 5.6 wt% clay (43). The TPO was composed of about 70 wt% of iPP and about 30 wt% of EPR. The nanocomposites were prepared by blending the TPO with maleic anhydride (MA)grafted iPP with 1.0 wt% MA concentration, as a compatibilizer and Cloisite1 20A natural montmorillonite clay modified with a quaternary ammonium salt. Figure 12.12 shows AFM phase contrast micrographs of the EPR particle morphology as clay loading increased in the range from 0.6 to 5.6 wt% clay. The EPR particle size decreased with clay loading as shown in Fig. 12.13. The breakup of the EPR particles was suspected to be due to the increasing melt viscosity, which was observed as clay loading increased, and/or the interfacial activity of the accompanying chemical modifiers on the clay. Figure 12.14 shows transmission electron microscopy (TEM) micrographs of the same system as in Fig. 12.12. It may be observed in Fig. 12.14 that the EPR domains (elliptically shaped) are surrounded by clay platelets (dark rodlike structures) and that the clay platelets preferentially segregate to the EPR/iPP interface. This was taken as evidence for the proposal that the interfacial activity of the accompanying chemical modifiers on the clay reduces the interfacial tension with concomitant reduction in particle size. A study of a similar TPO/iPP-MA/organoclay system was reported in which AFM and TEM images of the EPR particle morphology revealed a systematic reduction in the EPR particles as clay loading increased (44). Further, these workers observed an increase in impact strength of the TPO nanocomposites with increasing clay loading, which is generally the opposite trend as that expected for such nanocomposites. They explained this unexpected behavior in terms of the reduction in particle size of the EPR elastomer. They suggested that the reduction in particle size was due to some


373

Figure 12.14 Transmission electron micrographs (varying magnification as indicated by scale bars) of (a) TPO-0 (0 wt% clay), (b) TPO-1 (0.6 wt% clay), (c) TPO-3 (2.3 wt% clay), and (d) TPO-6 (5.6 wt% clay). EPR-rich rubbery domains (elliptically shaped) surrounded by clay platelets (dark rodlike structures) are clearly seen. (From Reference 42 with permission from John Wiley & Sons, Inc.)

combination of the melt viscosity increase and the interfacial effects of the clay, as clay loading increased.

12.5

CONCLUSIONS

Polypropylene copolymers are commercially produced for applications especially requiring high impact strength. However, many other properties are offered by

374

Polyolefin Blends

such resins. The typical processes involve the production of iPP in the first reactor, followed by the production of EPR in the second reactor to yield hiPP and TPOs. The dispersion of the rubbery EPR component especially serves to produce high impact strength, while high stiffness is contributed by the iPP matrix. These copolymers offer a balance of many other desirable physical and mechanical properties. These PP copolymers are primarily used in injectionmolded parts for automotive, appliances, and other durable goods applications, as well as for extruded sheet and thermoforming. The wide range of physical and mechanical properties, relative ease of processing, and low density constitute these polypropylene copolymers as extremely attractive materials capable of competing with more expensive plastics in many demanding applications. The automotive industry has made TPOs the primary choice for an increasing range of interior and exterior applications. These TPO resins continue to replace other polymeric compositions in interior and exterior applications, due to their desirable balance of properties and safety attributes. Interior applications include instrument panels, consoles, door panels, and pillars. Exterior applications include bumpers, fascia, body side cladding, rocker panels, and cowl vent grilles. The reactor grades of impact polypropylene copolymers are often compounded with other components, especially other toughening agents to extend their applications. A characteristic of hiPP and its compounds is that these are never miscible systems that undergo phase separation. The morphology of the polymer blends is multiphasic from the start. The morphology of the reactor product, typically powder, is extremely heterogeneous with some regions containing no observable EPR and other region containing large ‘‘pools’’ of EPR in the rigid iPP matrix. The reactor product with very heterogeneous morphology is typically converted to pellets for shipment with homogeneous morphology due to the great efficiency of commercialscale corotating twin-screw pelletization extruders. The heterogeneous reactor product undergoes homogenization by droplet breakup and coalescence in the pelletization extruder. This droplet breakup and coalescence occurs in the molten state of the viscoelastic iPP and EPR, matrix and dispersed phases, in the extruder under a complex strain field, which is a combination of nonuniform, transient shear and elongational fields. Further, a variable temperature profile is used along the barrel of the extruder causing complex variation in the viscoelastic properties of these components. The dispersive mixing theory covering droplet breakup and coalescence is generally a modified one, based on the case of a Newtonian drop suspended in a Newtonian matrix, that is, Taylor’s theory. Observed average particle size was 100 times that calculated from the dispersive mixing theory for an hiPP containing 16 wt% EPR. The larger particle size observed compared to that calculated is due to the effects of flow-induced coalescence. At practical concentrations of the dispersed phase of 10–20% in hiPP and TPO resins, observed particle size reported is 10–100 times above that predicted by theory. The very crude nature of the Taylor limit predictions is evident. Experimental determination of particle morphology is necessary in dispersively mixed systems.


375

The structure of a commercial impact PP was determined by P-TREF, followed by analysis of the fractions by NMR in order to determine chemical composition and tacticity. It was found that the hiPP (MFR ¼ 6) was composed of about 75 wt% iPP with average tacticity of about 97% and exhibiting a broad tacticity distribution extending to very low values, 17 wt% EPR containing an average of about 48 wt% ethylene and exhibiting a broad composition distribution and about 8 wt% of an ethylene-rich copolymer containing 92 wt% ethylene. The hiPP had Mw ffi 380,000 and Mw =Mn ¼ 6:3, the iPP had Mw ffi 382,000 and Mw =Mn ¼ 4:3, and the EPR had Mw ffi 342,000 and Mw =Mn ¼ 14:2. Although, phase separation from a one-phase melt was not typically relevant to the polypropylene copolymer systems, coarsening of the dispersed phase particles is a potentially important effect in systems in which the two-phase melt experiences, at least partly, a quiescent state. The coarsening of the phase-separated system can occur by two mechanisms. Coalescence, by particle diffusion and collision, is one mechanism, while evaporation and condensation, called Ostwald ripening, is another. It was shown that these hiPP systems exhibited the theoretically predicted temporal dependence of coarsening, that is, r t1=3 , which holds for coalescence or Ostwald ripening. Coarsening of model blends exhibited the theoretically predicted temporal dependence, that is, r t1=3 , and theoretical rate constants for Ostwald ripening and coalescence were KOR ¼ 0:86 1018 cm3 s1 and KC ¼ 3:6 1020 cm3 s1 , respectively. The experimentally determined rate constant was Kexp ¼ 1:23 1018 cm3 s1 and was in fairly good agreement with that for Ostwald ripening. The ratios of the experimental to the theoretical rate constants were found to be Kexp =KOR ¼ 1:3–1.4 for two model blend systems and appear to strongly support Ostwald ripening as the coarsening mechanism. Interfacial effects on multiphase polymer systems have been of interest to polymer scientists. Application of this strategy to commercially important polyolefin multiphase systems to produce ‘‘tuned’’ and/or stabilized morphologies is certainly attractive. Recently, polyolefin block copolymers have been introduced commercially and may be effective in controlling the morphology of related multiphase polyolefin systems, such as hiPP and TPO. For the case of TPO/organoclay nanocomposites, which are attracting wide commercial interest, it was shown that the EPR particle morphology in the TPO exhibited a progressive breakup and decrease in particle size, as clay loading increased. The breakup of the EPR particles was suspected to be due to the increasing melt viscosity, which was observed as clay loading increased, and/or the interfacial activity of the accompanying chemical modifiers on the clay. The development of polypropylene copolymer multiphase systems is continuing at a robust pace. These polymer systems, based on simple and inexpensive polymer building blocks, are being improved for applications historically reserved for more expensive engineering thermoplastics. The understanding of the structure/property relationships in polypropylene copolymers will indeed be a driver for further innovation in the commercial application of these polymers.

376

Polyolefin Blends

NOMENCLATURE AFM an Cea Cloisite1 D DBr dn dw EPDM EPR hd hm G g HDPE hiPP HPB Infuse1 iPP Kc KOR Kraton Li MA MFR Mn Mw Mw/Mn N NA PP PSD f R rð0Þ rðtÞ SEBS T t TEM Tg TPE TPO

Atomic force microscopy Number-average particle diameter Equilibrium concentration (mole fraction) of the minor phase Montmorillonite clay modified with a quaternary ammonium salt Molecular diffusion coefficient Droplet diffusion coefficient due to Brownian motion Number-average particle diameter Weight-average particle diameter Ethylene–propylene diene monomer Ethylene–propylene rubber Melt viscosity of the dispersed phase Melt viscosity of the matrix phase Effective shear rate Interfacial tension High density polyethylene High impact polypropylene Hydrogenated polybutadiene OBC olefin block copolymer Isotactic polypropylene Coalescence rate constant Ostwald ripening rate constant Styrenic block copolymers Chord length Maleic anhydride Melt flow rate Number-average molecular weight Weight-average molecular weight Molecular weight polydispersity Number of particles per unit volume Number of particles per unit area Polypropylene Particle size distribution Volume fraction of the minor (droplet) phase Gas constant Particle radius at time t ¼ 0 Particle radius at time t Styrene/ethylene–butylene/styrene Absolute temperature Time Transmission electron microscopy Glass transition temperature Thermoplastic elastomer Thermoplastic olefin


TPV Vm VTot

377

Thermoplastic vulcanized elastomer Molar volume of the minor component Total volume of the coarsening phase

REFERENCES 1. N. F. Brockmeier, Gas-phase polymerization, in: Encyclopedia of Polymer Science and Engineering, Vol. 7, H. Mark, N. Bikales, C. Overberger, and G. Minges (eds.), 1987, p. 480. 2. G. Schweier, Poly. Div., Abstracts, 214th ACS National Meeting, Las Vegas, NV, 1997, p. 450. 3. T. Simonazzi, G. Cecchin, and S. Mazzullo, Prog. Polym. Sci., 16, 303 (1991). 4. D. R. Paul and S. Newman (eds.), Polymer Blends, Academic Press, New York, 1978. 5. F. M. Mirabella, Polym. News, 30, 1 (2005). 6. P. Galli, T. Simonazzi, and D. Del Duca, Acta Polym., 39, 81–90 (1988). 7. I. Urdampilleta, A. Gonzalez, J. J. Iruin, J. C. de la Cal, and J. M. Asua, Ind. Eng. Chem. Res., 45, 4178 (2006). 8. Y. Chen, Y. Chen, W. Chen, and D. Yang, Polymer, 47, 6808 (2006). 9. S. Middleman, Fundamentals of Polymer Processing, McGraw-Hill, New York, 1977. 10. F. M. Mirabella, Polymer, 34, 1729 (1993). 11. F. M. Mirabella, Structure and thermodynamic aspects of phase segregation of commercial impact polypropylene copolymers, in: New Advances in Polyolefins, T. C. Chung (ed.), Plenum Press, New York, 1993. 12. J. J. Strebel, F. M. Mirabella, C. Blythe, and T. Pham, Polym. Mater. Sci. Eng., 44, 1588 (2004). 13. S. Wu, Polym. Eng. Sci., 27, 335 (1987). 14. H. Shariatpanahi, H. Nazokdast, B. Dabir, K. Sadaghiani, M. Hemmati, J. Appl. Polym. Sci., 86, 3148 (2002). 15. V. E. Ziegler and B. A. Wolf, Macromolecules, 38, 5826 (2005). 16. G. I. Taylor, Proc. R. Soc. (London) A, 138, 41 (1932). 17. G. I. Taylor, Proc. R. Soc. (London) A, 146, 501 (1934). 18. U. Sundararaj and C.W. Macoscko, Macromolecules, 28, 2647 (1995). 19. J. J. Elmendorp and A. K. Van Der Vegt, Polym. Eng. Sci., 26, 1332 (1986). 20. F. M. Mirabella, J. Polym. Sci.: Appl. Polym. Symp. 51, 117 (1992). 21. T. Hashimoto, Phase Transitions, 12, 47 (1988). 22. W. Ostwald, Z. Phys. Chem., 37, 85 (1901). 23. I. M. Lifshitz and V. V. Slyozov, J. Phys. Chem. Solids, 19, 35 (1961). 24. M. von Smoluchowski, Phys. Z., 17, 557, 585 (1916). 25. V. G. Levich, Physicochemical Hydrodynamics, Prentice-Hall, Englewood Cliffs, NJ, 1962, pp. 207–211. 26. E. D. Siggia, Phys. Rev. A, 20, 595 (1979). 27. T. Hashimoto, M. Takenaka, and T. J. Izumitani, Chem. Phys., 97, 679 (1992). 28. B. Crist, Macromolecules, 29, 7276 (1996). 29. B. Crist and A.R. Nesarikar, Macromolecules, 28, 890 (1995). 30. F. M. Mirabella, J. Polym. Sci. B Polym. Phys., 32, 1205 (1994). 31. F. M. Mirabella and J. S. Barley, J. Polym. Sci. B Polym. Phys., 32, 2187 (1994). 32. F. M. Mirabella, F. M., Barley, and J. S., J. Polym. Sci. B Polym. Phys., 33, 2281 (1995). 33. P. W. Voorhees, J. Stat. Phys., 38, 231 (1985).

378

Polyolefin Blends

34. G. Venzl, Ber. Bunsenges. Phys. Chem., 87, 318 (1983). 35. I. Fortelny and A. Zivny, Polymer, 21, 4113 (1995). 36. I. Fortelny, A. Zivny, and J. J. Juza, Polym. Sci. B Polym. Phys., 37, 181 (1999). 37. S. H. Anastasiadis, I. Gancarz, and J. T. Koberstein, Macromolecules, 22, 1449 (1989). 38. Y. Tao, A. H. Lebovitz, and J. M. Torkelson, Polymer, 46, 4753 (2005). 39. T. J. Cavanaugh and E. B. Nauman, Polym. Sci. B Polym. Phys., 36, 2191 (1998). 40. D. J. Arriola, E. M. Carnahan, P. D. Hustad, R. L. Kuhlman, and T.T. Wenzel, Science, 312, 714 (2006). 41. http://news.dow.com/prodbus/2006/20060620a.htm. 42. F. M. Mirabella, Polypropylene and thermoplastic olefin nanocomposites, in: Encyclopedia of Nanoscience and Nanotechnology, J. A. Schwarz, C. I. Contescu, and K. Putyera (eds.), Marcel Dekker, New York, 2004. 43. S. Mehta, F.M. Mirabella, K. Rufener, and A. Bafna, Appl. Polym. Sci., 92, 928 (2004). 44. H. Lee, P. D. Fasulo, W. R. Rodgers, and D. R. Paul, Polymer, 46, 11673 (2005).

Chapter

13

Heterogeneous Materials Based on Polypropylene Jesu´s Marıá Garcıá Martıńez1, Susana Areso Capdepoń1, Jesu´s Taranco Gonza´lez1, and Emilia Pe´rez Collar1

13.1 INTRODUCTION The long-range elasticity, high strength, and high viscosity, all of the self-defining macromolecular states are deeply influenced by the intermolecular forces—a direct consequence of the size and constitution of the covalent structures of macromolecules. For the thermoplastic polymers family polypropylene belongs to, the great number of atoms involved in the primary intrachain bonds above a given critical value—that is, mean size or molecular weight—are able to induce interchain secondary interactions at such a strong level that the matter becomes a material, with enough structural integrity to be useful. Since the beginning, organic polymers have been combined with another substances, mainly inorganic, to lower costs. However, the reinforcement effect in the polymer-based material caused by the presence of a second component was soon ascertained (1). In mid twentieth century, it was surmised that the contact regions between both components, that is the interphases, play a crucial role in almost all the heterogeneous materials. Numerous efforts, mainly in the development of light-weight structural applications based on thermosetting polymers, were devoted to improve these interfacial regions. Since then began a 20th second revolution on the basis of heterogeneous materials, after the so-called industrial revolution. Once their production

1 Department of Physics and Engineering of Polymers, Polymer Engineering Group, Institute of Science and Technology of Polymers, CSIC, CL Juan de la Cierva 3, 28006 Madrid, Spain


379

380

Polyolefin Blends

processes were consolidated on an industrial scale, organic-based materials have been called to be the key to the new era. After the pioneering work by Helfand and Tagami (2), the basis to the study of interphases from a perspective of the organic matter-based materials and to control the properties of any given multicomponent systems, was established. So the option of designing with materials began to be surpassed by the option of designing multiphase materials, with at least one organic polymer obtained in a polymerization reactor being one of their components (3).

13.2 THE INTERPHASE: DEFINITION From the early 1980s, it was well accepted that the desired performance of a polymer-based heterogeneous material would pass through a well-optimized interaction level across the interphases between the components (4). The interphase is defined as the dynamic and finite spatial region placed between the border of each of the two different phases where momentum, mass, and energy transport phenomena may occur. It follows from this definition that transport phenomena across the interphases are governed by the morphology and even topography of the two phases, with dimensions of the nanoscopic scale, in such a way that atoms or small groups of atoms located on the border surfaces determine the flows of momentum, mass, and/or energy balances between both phases. Because only in a few heterogeneous systems (mainly in the highly selective bioactive systems) is it possible to measure such flows, given the high complexity inherent in the macromolecular systems, the transport phenomena or the changes in any given property across the interphases can be evaluated by considering the interface model as displayed in Fig. 13.1. According to this figure, the interface would be obtained by projecting all volumes between both phases sensitive to the desired property onto a single and finite surface defined by a critical thickness, where any measured property would exhibit the sharpest possible change from one phase to another. So, different critical thicknesses for different properties or response functions can be expected.

13.3 MAGNITUDE ORDERS IN THE INTERPHASE For some of the recent enthusiasts of nanochemistry, it would be convenient to remind that macromolecules of thermoplastic polymer materials remain together due to the high intensity of the secondary interactions derived from the very high number of atoms present in their polymer backbones. It is important to note that the primary bonds located on the main chain typically require energy of bonding values from 400 to 800 kJ mol1. Meanwhile, the order of magnitude for the secondary forces ranges from 4 to 40 kJ mol1. The latter means distances of 0.3 and 0.4 nm between the atoms involved in the secondary forces, that is, three times or more larger than the 0.1

Chapter 13 Heterogeneous Materials Based on Polypropylene

381

Figure 13.1 The interface model.

to 0.15 nm existing between the atoms that form a primary bond. The reach and intensities of these intermolecular forces acutely depend on temperature and time, in a much more sensitive way than in any other classical materials, for example, metals, concrete, or ceramics. In consequence, several assumptions valid to explain the solidstate behavior of these materials can hardly be accepted in the case of organic thermoplastic materials. Because both the softening as well as the molten state of a polymer require the strain phenomenon to occur before the flow takes place, it is logical that to induce flow in any organic macromolecular system not only temperature but also pressure is required. The coupled viscous dissipation due to the self-friction of the macromolecular chain segments hardly may be neglected when undertaking any study. Even it may occur that while the strengths of those secondary forces in the macromolecular systems would be still strong enough to maintain the macromolecules together, some primary bonds would begin to break in a local adverse environment. Figure 13.2 compiles the main variables and parameters defining the polymerbased heterogeneous materials from the macro- to nanoscales, which determine the performance of the material as a whole.

382

Polyolefin Blends

Order of magnitude in multiphase materials

Parameters

Structure assembling Length at molding stages Width at molding stages Length Width Fiber length in finished parts Fiber length at molding stages Fiber diameter Surface roughness Crystal aggregates Polymer crystals Chain length Macromolecular diameter –12

–10

–8

–6

–4

–2

Twelve orders of magnitude Interfacial interactions

0

Multiphase material

Reinforcement and/or dispersed phase

Polymer

2 4 Log units, m Macroscopical properties

Figure 13.2 Orders of magnitude at the interphase studies on polymer-based heterogeneous materials.

13.3.1 The Dispersed Phase The most frequent disposal of most of the polymer-based heterogeneous materials family takes place by the dispersed phase/matrix mode. So the dispersed phase components may be identified by the finite size of each of their domains, being surrounded by the continuous matrix. Both the size and the geometry of the particles featuring the dispersed phase together with their surface properties govern the transport phenomenon across the interphase between the dispersed particles and the continuous matrix. According to the interface approach defined in the previous section, it is obvious that the domain size and its distribution confine the interfacial volume available for effective transport flows between the matrix and the disperse phase. To control this, parameters of the highest relevance for studies of interfacial phenomena have been set, so that particle size and distribution remain constant, or almost constant, all along the experimental work. When the dispersed phase is constituted by rigid particles (5), the breakdown caused by external stresses may occur, yielding significant changes in both the mean particle size and the size distribution. However, if reinforcing particles do not break down during processing steps, these variables remain under control all along the experimental work. Hence, interfacial modifications are able to change the breakdown behavior of the solid particles dragged by the matrix flowing streams, for example, by lubrication, and introduce an additional experimental variability by way of


383

noncontrolled effects at the different stages of the experimental runs. Consequently, misleading or wrong conclusions on the reach of the interfacial changes may be obtained from such works. Otherwise, in the case of a deformable dispersed phase, that is, polyblends, when the polymer matrix flows, it occurs that the interfacial volume available for the transport flows between both phases could change in a way much more complex than when the dispersed particles are not deformable, not only in terms of changes in mean size and size distribution all along the processing steps but also in shape and preferential geometry of the dispersed phase domains. These changes would evolve as the primary function of the viscosity ratio between the components, and of the surface tension balance when interfacial modifiers are present (6). Even, if primary bonds connecting dispersed and continuous phases are formed through the interphase in whatever processing step, they can generate stable morphologies that remain intact even after other postreactive processing steps, yielding a controlled distribution of the dispersed phase into the matrix (7).

13.3.2 The Matrix The design criteria based on a well-defined performance of a material, with emphasis on its mechanical behavior, have greatly increased the application spectrum of the organic polymeric materials. The application varies from the neat thermosetting matrices for engineering devices to both the development of engineering thermoplastics and the composites based on them incorporating discrete reinforcements, where the polymer matrix properties have been designed to satisfy specific requirements (1,3,8). Not only have mechanical performance considerations and other specific properties of the material led to this situation, but also the economic benefits derived from faster processing operations for thermoplastics compared to those for thermosetting polymers, as well as environmental considerations dealing with easier recycling possibilities of thermoplastic matrices. This has emerged as the driving force for a change in designing criteria. Figure 13.3 compiles the main variables and parameters that define a semicrystalline thermoplastic polymer matrix that, in combination with the processing steps, optimize thermophysical and mechanical behavior of the material to be useful for any purpose. This is the well-known structure–processing–properties dynamic triangle.

13.3.3 The Interphase: Designing the Interface From the previous sections it follows that when designing the interface, starting on the desired performance of the heterogeneous material at the macroscopical scale, of 12 magnitude orders, may be the dimensional gap required to obtain the optimum material (9). The classification of primary or secondary bonds as a power function of the distance between each pair of atoms involved for each bond and the way the

384

Polyolefin Blends Rheology, processing steps

Molecular weight

Intermolecular parameters: chain volume diameter and length (end to end distance)

Intramolecular parameters: interaction parameters (space/volume ratio)

Semicrystalline polymers: ❖ Amorphous/crystal ratio ❖ Unit cell ❖ Lamellar thickness distribution ❖ Spaced and long spacing ❖ Crystalline aggregates

Tortuosity parameter (Length/2*radius of gyration) Matrix morphology Thermophysical: ❖ Glass transition temerature ❖ Melting temperatures ❖ Melting and crystallization heats ❖ Specific heat ❖ Thermal conductivity ❖ Coefficient of thermal expansion ❖ Moisture absorption

Properties

Mechanical: ❖ Strength and module (tensile, flexural and shear modes) ❖ Poisson coefficient ❖ Impact behavior ❖ Hardness ❖ Friction coefficient

Others required by performance

Figure 13.3 Structure–processing–properties relationship for designing materials based on organic thermoplastic matrices.

intensities of these forces decrease with distance is of the highest relevance at the nanoscale level the interfacial phenomena occur (10). In particular for the mechanical responses across the interphase, it would be obvious that long distance forces let higher strain levels across the interphase than those from short distances, giving rise to brittle materials. Moreover, permanent dipole–dipole interactions yield attractive forces that decrease as the third power of the distance; meanwhile, induced dipole–dipole interactions yield attractive forces that decrease as the seventh power of the distance between the atoms involved. Dispersive forces due to electronic displacements mean attractive interactions decreasing as the sixth power of the distance. For an ideal amorphous polymer, the specific (referred to as chain segment) free volume should vary only with temperature. According to the basic principle of ideal elasticity, the internal energy of a system remains constant during the mechanical energy absorption/dissipation processes. From here it follows that the specific free volume for an ideal amorphous polymer would be an index of the available local space for the dissipation of the absorbed mechanical energy through the segmental rotational motions of the macromolecular chains (11). When the polymer has the potential to reach the three-dimensional order (12), that is, the crystalline state, the fracture behavior becomes more complex. It is well known that the distance between the macromolecular chain segments in the crystal are shorter than those located in the amorphous phase. Hence, when the material is required to dissipate any kind of energy, this would be preferably dissipated through the amorphous phase and, even more, through the zone where the chain segments are significantly constrained, as it is an amorphous/crystal interphase.


385

Then, the latter forms the weakest region in the material as a whole, the region where the fracture phenomenon would first start. This phenomenon takes place at a so called nanoscale. Meanwhile, studies on bioactive species, ranging from living things to biochemistry as a concept, have been conducted to get a hierarchical range of responses, passing through the different organized systems such as organs, functional tissues, expertise cells and so on with their different inner devices based on macromolecules. In the study of materials, However there exists a deep lack of knowledge of the macroscopic performance of these materials as a whole and their relationship with their macro-, micro-, meso-, and nanoscale morphologies, each of them defined by the emerging property that determines their functionality at each scale (13). In the case of an organic thermoplastic polymer, it must be mentioned that the generated morphologies and the supramolecular organization possibilities of their molecular structures are developed under external fields, mainly thermal and stress/strain fields, acting during the processing steps. This appears to be very sensitive to any other external field acting over the polymer when becoming a final part. These unknown hierarchical response levels and the ways they are interconnected are undoubtedly the key when designing the interphase between the components of any organic polymer-based heterogeneous material. It is then obvious that establishing the above mentioned hierarchical range of the interconnection parameters between the macro-, micro-, meso-, and nanoscales is necessary before a great effort in the study for an efficient control of the ultimate properties of the material is made.

13.4 INTERFACIAL MODIFICATION OF HETEROGENEOUS MATERIALS BASED ON POLYPROPYLENE It is well accepted that the good properties of the isotactic polypropylene as an engineering polymer matrix in thermoplastic composite materials and engineering blends are seriously affected by the inability of this polymer to develop an adequate level of interfacial interaction with polar components such as mineral fillers (calcium carbonate) and reinforcements (talc, mica, wollastonite), synthetic reinforcements (glass fibers, carbon fibers, and nanotubes), or engineering polymers such as polyamide, aliphatic polyesters, and so on. In order to preserve the affinity with the nonpolar polypropylene matrix, the substitution of a small fraction of the matrix by another polypropylene with just a few polar groups grafted onto its backbone has been proven, in the last 20 years, to be a very effective strategy to promote isotactic polypropylene for engineering polymer applications. Among the different polar groups chosen to be grafted into polypropylene, maleic anhydride has proven to be of the highest efficiency as follows from the

386

Polyolefin Blends

reviews by Xu and Lin (14) (299 references), Jois and Harrison (15) (129 references), Naqui and Choudhari (16) (118 references), and Moad (17) (188 references). From the point of view of designing at the interface level, there is not yet available any mathematical model based on the physicochemistry of this spatial region for general application to develop any composite material based on organic matrices like polypropylene. At this point, it must be noted that the main goal, when building a mathematical model for materials design, is to transform a physical event connected to a real system into a mathematical structure that can be used to describe and understand such event. Furthermore, the model must be able to predict the magnitude of the changes in the system responses under new external stimuli, as well as to obtain the desired performance. Aris (18) defines a mathematical model as a system of equations formulated to express the laws of a prototype system, including in a much more general system and whose prototype is characterized by a set of specific features of main interest in the system under study. Hence, it follows that to build a model that is able to include the overall aspects of a real system while maintaining its accuracy and robustness is hardly possible. The complex nature of the organic macromolecular systems because of the nonlinear regime of responses of these materials appears as the best example of these considerations, especially when faced with classical materials such as ceramics, metals, and inorganics. The first critical step toward developing an efficient model begins with the identification of the key aspects to be considered and those to be discarded in the process, while looking for obtaining a model simple enough to be managed. However, discarding any essential feature of a real problem while looking for an easier resolution of an algorithm model would lead to forecasts of little use and sense. When building a model to predict the mechanical behavior of heterogeneous materials based on polypropylene, the basic principle of fracture mechanics should be considered that fracture is induced in a material when an energy threshold on the weakest zones is reached. It is also well accepted that such energy threshold consists of at least two components: necessary energy to initiate a craze and the necessary energy to propagate it. From the preceding sections it may be deduced that the generation of a craze would need much more energy in the case of a well balanced interphase than the unbalanced interphase. Hence, both craze growth and propagation across a much more continuous interphase are expected to be much faster than in a less continuous interphase. Indeed, the craze progress would be a critical factor for forecasting the failure of an initially strong interphase. Then not only would it be necessary to take into account the numerical values of the macroscopic mechanical properties, moduli, or strength, but also their hierarchical relationship with different mechanical energy dissipation possibilities at macro-, micro-, meso-, and nanoscales should be taken into consideration when building a general model to design the interface. They are necessary heuristic models in which any deficiencies once evidenced would be successively removed. These models would allow the researchers to build models


387

that would enable control of the behavior of the organic polymer heterogeneous materials on the basis of their optimal balance of interactions through the interphases between their components.

13.4.1 Composites: When the Dispersed Phase is Rigid Mathematical models may be classified into deterministic and stochastic models. For deterministic models, knowledge of the relationship between dependent and independent variables is necessary. Consequently, the complex nature of polymer-based heterogeneous materials is rather incompatible with such requirements. Hence, stochastic models become necessary either when the existing knowledge about the stimulus-response behavior of a system is not enough as to ascertain its behavior or when it is not possible to build an efficient deterministic model able to score the system response. Data, collected by random experimental runs using probability and statistical models, may be a good way of approaching the responses of the system inside the experimental space scanned (19). Then, this would be a preliminary consideration when undertaking a model for a disperse phase/continuous matrix systems where the dispersed phase domains are rigid. Once assured that reinforcing particles do not change their size by breaking during the processing steps, the design of a surface treatment for the rigid particles would be twofold. On the one hand, it would tend to avoid any agglomeration of the solid particles and on the other, it would tend to improve their physicochemical interaction level with the polymer matrix. From the early 1980s, attempts were to avoid the negative effects of hydroxyl groups located on the surface of the inorganic particles by treating them with the socalled coupling agents such as silanes, titanates, and so on (20–23). The substitution of hydroxyl groups by chlorine or amine groups onto the lamellar talc surface, finally yielding polypropylene/talc composites with good mechanical performance, was studied by various groups (24–26).

13.4.2 Blends: When the Dispersed Phase is Flexible As it is well known, for most of the polymer/polymer pairs because of their low values of mixing entropy, a multiphase system is always generated. The dispersed phase morphology is the result of the balance between the applied shear forces and the counteracting interfacial tension forces. As a consequence, the long-range morphological regularity that may be observed on block copolymers, sometimes claimed to be model polymer blends, is hardly obtained by mixing two or more different polymers. Large shear rates enhance deformation capabilities of the dispersed phase domains generally as droplets, flowing with the matrix during the mixing and further

388

Polyolefin Blends

processing steps. Evolution of the morphology may be followed as a function of the mixing time and also the mixing conditions affecting drop dimensions and stability (6,7). These may be expressed by the Weber number given by We ¼

ðg hm Dn Þ ð2GÞ

ð13:1Þ

where g is the shear rate, hm is the matrix viscosity, Dn is the dispersed phase size, and G is the interfacial tension. From Equation 13.1, the existence of a critical value of the Weber number emerges, and then a critical particle size for the dispersed phase below which there is no particle deformation. Furthermore, if only physical interactions exist between components, the remaining dynamic coalescence mechanism could change on further processing steps such as a nonstable postprocessing morphology. Any model capable of predicting the emerging morphologies and their stability must take these elemental principles into consideration. The understanding of the relationship all along the 12 magnitude orders between the structural response levels at the macroscopic scale, and the driving forces acting at the interfacial scale also at the molten state, lead us to approach the hierarchical sequence of possibilities of matter organization from macro-, to micro-, to meso-, and to nanoscales at different windows of response, where each of them controls the behavior of the material either as a whole or when overlapped their effects (13).

13.4.3 The Role of the Interfacial Modifiers from the Matrix Side From above discussion, it becomes obvious that interfacial modifications may be driven either by the dispersed phase or from the matrix perspective, or even both. But driving forces in both cases must be well balanced to achieve the desired performance of the material at the macroscopic scale. Nevertheless, because of their affinity with the matrix, an increase in the flow across the interphase due to the presence of the interfacial modifiers from the matrix side is always observed. Besides, a decrease in the critical particle size for the polymer/polymer systems could also be obtained due to the decrease in the Weber number values, caused by an increase in the interfacial area available. A saturation level at the interaction plane across the interphase and in the concentration of the interfacial modifiers also emerges from the finite dimensions of the interphase. Above a critical concentration, the interfacial modifier could form its own phase, and then a nth phase ought to be considered in the studies rather than a model based on modified interfaces. Following sections show some examples of the role of the interfacial modifiers from the matrix side for both rigid and nondeformable dispersed phase polypropylene/mineral reinforcement system and polymer/polymer binary system based on polypropylene and polyamide 6.


13.4.3.1

389

Polypropylene/Mineral Systems

Depending on the number of grafted moieties as well as on the place where these grafted groups are located all along the polymer backbone, and even in the chain ends, it is possible to obtain many different functionalized polyolefins as a function of the reaction conditions chosen to obtain them. The effectiveness of the functionalized polypropylenes as interfacial modifiers would be associated with the increase of the second-order interactions between the matrix and the dispersed phase, as well as with the higher mobility of polar groups bonded at the chain ends. These would increase the interaction possibilities with the surface of the discrete phase particles across the interface, which is especially useful to reduce the interfacial tension between the components of the heterogeneous system. The presence of mineral reinforcements such as talc or mica, as foreign solid particles embedded into a polypropylene matrix, usually induces a nucleation effect. A significant increase in the crystalline content of the polymer is evidenced if compared with the neat polymer when processed at the same setup conditions that are necessary to ensure a good accommodation of the solid particles into the amorphous phase of the polymer in order to obtain a material with a good mechanical performance (27). The comparison between PP/mica and PP/talc composites in terms of their mechanical behavior under dynamic conditions in the solid state agrees with the morphological features derived from their chemical structures of both minerals (28). In fact, if the width and thickness of a lamellar particle are defined as a fraction of its longest and main dimension, lr , it is easy to define the surface/volume ratio (S/V) for each particle as follows: S ¼ V

1 1 1þ þ n1 n2

þ

2 lr

ð13:2Þ

where n1 and n2 are the fractional numbers for, respectively, the width and the thickness of the particle relative to its longest dimension, lr . Because the S/V ratio is proportional to the specific surface of the mineral and lr being higher for mica than for talc, it follows that specific surface would always be lower for mica than for talc particles. Then for the same crystalline amount of the polypropylene matrix, a higher fraction of amorphous phase involved in the coating of talc particles than in the coating of mica particles would be expected. So, if the interfacial regions have to be considered, we must differentiate the amorphous/crystal polymer interface from the amorphous polymer/mineral interface (29). By DMA measurements, a decrease in Tg matrix from 7 C for the PP/talc composites and also for the neat PP processed under similar conditions while a decrease upto 13 C was found for PP/mica composites. A higher fraction of ‘‘free amorphous phase’’ on the PP/mica system than on the PP/talc composites was evidenced. This ‘‘free amorphous phase’’ appeared to participate in the cooperative segmental free-rotation motion, well accepted (30) to be responsible for glass transition for the polymer matrix as fully discussed in Reference 29.

390

Polyolefin Blends

Such kind of considerations about the finite dimensions of the interfacial area at the polymer/mineral interphase must be taken into account when we try to modify the interface by substituting a part (frequently a little part is enough) of the polypropylene matrix with a chemically modified polypropylene by grafting of polar groups. Also, the structural characteristics of such chemically modified polypropylene must be considered, mainly its molecular weight, if crystalline or not, and the nature and number of polar grafts. An example of the relationship between macroscopic responses of the composite material as a whole and the micro-, meso-, and nanomorphological scales involved can be found when a classical tensile loading test of this kind of material above Tg of the polymer matrix portion of the composite is performed. Consequently, two effects are involved in the evolution of the elastic modulus and the tensile strength at yield and at break too. On the one hand, the higher number of crystallites due to the nucleation effects that act as cross-linking regions between the interconnecting chain segments that belong to the so-called tie molecules. On the other hand, the self-contribution of those new crystallites to increase the elastic modulus because of the higher stiffness of the crystals compared to that of the amorphous phase. Strength at yield usually evolves in the same way as the elastic modulus does, while for strength at break, and depending on the amorphous phase present in the material, many differences can be expected at the ultimate stages of the overall strain process. Figure 13.4 shows the variations between the tensile test parameters for the PP/interfacial modifier/talc composites as a function of either the grafted group, succinic anhydride (SA) or succinyl-fluorescein (SF) attached to an atactic polypropylene, as well as the differences on tensile strength and strain levels at yield or at break points, depending on the amorphous or semicrystalline nature of the interfacial modifier as it was fully discussed elsewhere (29,31). Figure 13.5 compares the differences between the evolution of the elastic modulus and tan d from the DMA tests with both the kind of attached group at the atactic polypropylene and the level of grafting of the interfacial modifier present in the composites. The different effect of each interfacial agent is clearly concluded as discussed elsewhere (33,56). The capability of the succinyl-fluorescein grafted polypropylene used as interfacial modifier in the PP/talc composites to saturate the interfacial area available at the interphase is well observed on the left-hand plots of Fig. 13.6 where it can be seen how the amount of molten material during the second heating cycle in dynamical tests by DSC, compared to that previously ordered during the cooling step, evolves as a min–max curve in a very similar way that strength at break does (34,35). Right-hand plots in the same Fig. 13.6 allows to conclude the well-optimized material performance from the macroscopic up to the mesoscales by the correspondence between the elastic modulus component obtained from the tensile tests and the elastic component of complex modulus from the DMA measurements performed at room temperature and which ratio results to be independent of the grafting level in the interfacial modifier (32,55).


391

Figure 13.4 Macroscopic responses by changes at the interfacial level. Variations on tensile test parameters with the amount and the structure of the interfacial modifiers for the PP/talc system. (From Reference 29 with permission of Elsevier.)

13.4.3.2

Polypropylene/Polyamide 6 System

Both from an applied and from an academic point of view, the polypropylene/ polyamide 6 system is indeed one of the most studied polymer/polymer binary systems, together with the interfacial modification possibilities induced in the morphology of the blend either by mixing during the processing steps or by the presence of interfacial modifiers. The immiscibility between both polymers is an important drawback that obliges us to look for a well-balanced set of properties, namely, mechanical, vapor, and gas permeability for barrier applications. Most applications are as multilayer, or sandwich sheet structures (polymer A/adhesive layer/polymer B), obtained by coextrusion, lamination or coating operations. The high operation costs associated with these processing techniques and the inability to obtain complex shapes that can expand

392

Polyolefin Blends

Figure 13.5 Mechanical and dynamic mechanical responses on PP/mica system. Changes with the grafting level and the structure of the interfacial modifier. (From References 33, 56 with permission of John Wiley & Sons, Inc.).

application to markets are the main driving forces behind the study of these blends from a technological point of view. From an academic perspective, this family of blends appears as an excellent model of the most general polycondensate/polyolefin polymeric systems; focusing on the development and control of post-reactive processing stable morphologies can yield a material with the desired performance, that is, good processability, low water absorption and liquid and vapor permeability, improved dimensional stability, good impact strength, and improved chemical resistance to alcohol and glycols. In a recent paper (9), a micromechanical model is proposed that is mainly characterized by a shape factor, w, given by w¼

ðn1 n2 Þ ðn1 þ n2 Þ

ð13:3Þ


393

Figure 13.6 Examples of correlation between responses from different scales when the interfacial modifier is changed: Left, tensile strength at break point (up) and relative crystalline variation for the polypropylene matrix (down) versus the grafting level in the interfacial modifier; right, elastic moduli (tensile/DMA) ratio (up) and components of the complex modulus from DMA tests (down) versus the grafting level in the interfacial modifier. (From References 32 and 55 with permission of John Wiley & Sons, Inc. and Elsevier, respectively.)

where n1 and n2 are the fraction numbers that refer, respectively, to the width and the thickness of the dispersed phase particles to their longest dimension. Because of the deformability of the dispersed phase, this shape factor is associated with the processing history and allows us to define an approach to the morphology– mechanical properties relationship, starting on the two limit values for the shape factor, always between zero and one as proposed earlier (9). Optical microscopy on phase contrast mode allows observation of the different morphologies obtained for each PP/interfacial modifier/PA6 blend. By image analysis techniques, it is possible to carry out statistical field measurements not only of the mean number of particles on the dispersed phase but also of their preferential geometry, mean size, and size distribution.

394

Polyolefin Blends

Figure 13.7 PP/PA 6 binary system with modified interface. Example of morphological changes in the 85/15 w/w ratio blend with the amount (3% or 15% over PP phase) and with the grafted group (succinic anhydride, SA, or succinyl-fluorescein, SF) attached to the atactic polypropylene used as interfacial modifier.

By following the Box-Wilson experimental design methodology, we tried to fit all these different parameters looking for correlations and emerging functions between the response levels of the system with those obtained by mechanical (9), thermal (34), and diffraction (35) techniques. Figure 13.7 displays, as an example, the morphology observed by optical microscopy on negative phase contrast mode for the PP/interfacial modifier/PA6 system after compression molded on 100 mm thick sheets previously Banbury batch mixed. The phase contrast micrograph on top shows the 85/15 unmodified blend. Because the refractive index for the polyamide 6 (1.53) is higher than that for the polypropylene (1.49), the negative contrast condition leads us to observe the polypropylene matrix as a dark field, while the disperse domains of polyamide appear as bright regions. When comparing the micrographs for the modified blends with that for the unmodified one, in Fig. 13.7, it is easy to observe a sharp difference between the modified blends morphologies as a function not only of the amount (3% or 15% w/w)


395

of the interfacial modifier at the interphase, but also depending on the attached group (SA or SF) onto the atactic polypropylene used as interfacial modifier. Indeed, the polypropylene matrix appears as a more uniform dark field because of the lower sizes of the domains of polyamide 6. The changes in the preferential geometry of the dispersed phase domains in the modified blends are also clearly observable as well as their higher brightness that should agree with a much more clear interphase in the modified blends. This is because the first-order bonds formed at several points of the PP/PA6 interphase are able to improve the optical light path on the microscope. The stable postreactive processing morphology of some of these blends was preliminary discussed by authors over their dynamical thermal responses (34,36). Moreover, the correspondence with the solid-state behavior of these blends can be observed in Figs. 13.8 and 13.9. The latter showing the differences between both a-PP-SA and a-PP-SF as interfacial modifiers for the PP/PA6 system at, respectively, the macroscopic mechanical response level, Fig. 13.8, and at the

Figure 13.8 Variations in the tensile strength values at yield (s F ) and at break (s R ) for the 85/15 and the 15/85 composition ratio of the PP/PA6 binary system with modified interface.

396

Polyolefin Blends

Figure 13.9 Evolution of long spacing of the crystalline phase of the polypropylene all along the PP/ PA 6 binary system with modified interfaces.

nanoscopic diffraction response level taking place over the crystalline phase of polypropylene, Fig. 13.9. For example, in Fig. 13.8, strength values at yield and at break from tensile tests for the PP/PA6 blends 85/15 and the 15/85 w/w ratio, have been displayed over a square plot (9). For the blends modified with a-PP-SA and for both PP/PA6 ratios, the linear increase of yield and break strength values when the aPP-SA amount increases up to the 9% is clearly observable. Above such value, both tensile strength values decrease if the a-PP-SA increases. In the same plots, it also may be observed that the tendency of higher strength at yield than at break for the neat PP remains in those blends where polypropylene acts as the matrix. Meanwhile, in those blends where polyamide 6 is the matrix (and so polypropylene the dispersed phase), it exhibits much closer values for strength at yield and at break, just as the neat polyamide 6 does. The different roles played by each interfacial modifier when located at the interphase are evident when we observe the two opposite effects caused by the aPP-SF/SA in the 85/15 and the 15/85 PP/PA6 modified blends. For the former, where the polypropylene is the matrix, the presence of the lowest amount of a-PP-SF/SA is enough to increase the tensile strength values with respect to the unmodified blend, while both strength at yield and at break decrease if the a-PP-SF/SA amount continues to increase. When polyamide becomes the matrix in the blend, both tensile strength values improve with respect to the unmodified blend by the presence of a-PP-SF/SA and also with an increase in a-PP-SF/SA amount.


397

As observed for blends modified by a-PP-SA, the tendency of higher strength at yield than at break for the neat PP remains for the blends modified by a-PP-SF/SA where the polypropylene is the matrix. However, closer tensile strength at yield and at break values obtained for the neat polyamide 6 may be also observed in the modified blends by a-PP-SF/SA where the polyamide 6 acts as the matrix. These results would be coherent with those plotted in Fig. 13.9 that shows the higher long spaced values (L) of the crystalline phase of polypropylene present in the blends modified by the succinyl-fluoresceine grafted atactic polypropylene, than L for the blends modified by the succinic anhydride grafted atactic polypropylene. This would agree with the higher spatial volume of the former and its lower reorganizing possibilities at the highly constrained amorphous/crystal interphase (35). Consequently, the overall responses of the thermoplastic organic materials related to their applications are sharply affected by the thermal and strain/stress fields as they underwent before reaching the solid state, that is, the top-down approach to develop ways to control the performance of these materials seems indeed promising.

13.5 INTERFACIAL MODIFIERS BASED ON POLYPROPYLENE Chemical modification of polymers affords an opportunity to modify undesired properties that would limit or even invalidate the potential usefulness of these materials. Polypropylene polymers are a good example, because their applications depend on their stereoregularity. Atactic homopolymer, a by-product from industrial polymerization reactors, lacks good material properties. However, its chemical modification by grafting with polar groups, one can convert it into a useful new material, as suggested by Natta (37). As it has been demonstrated in the previous sections and also in numerous works available in the literature, the efficiency of the succinic anhydride grafted groups onto polypropylene backbone as interfacial modifiers on blends and composites based on polypropylene is indeed proved. In spite of its very high economic relevance, the process to obtain these new high value-added polypropylenes is far away to be realized. Indeed, the chemical modification of the polypropylene by grafting of polar groups is not well understood because of the complex nature of the process from the point of view of the chemical engineering. It means one needs to take into consideration the reactor where the process takes place, as well as the nature of the macromolecules as reactants or coreactants.

13.5.1 The Kinetic Approach: Basic Aspects Due to the complex nature of the macromolecular systems, theoretical models of the reaction mechanism are very difficult to build. In many cases, empirical models based on the experimental results obtained in any given conditions are implemented

398

Polyolefin Blends

for a specific process but they cannot be generalized. From this, semiempirical models based on theoretical considerations and supported by experimental data would be the most interesting way to approach the problem solution. Before building a semiempirical model, several basic aspects must be observed. First of all, for classical low molecular weight reactants the same reactivity for each individual molecule can be assumed because they must be almost identical. In the case of macromolecular reactants, like polypropylene, the existence of a molecular weight distribution makes this assumption invalid (38). However, the conversion in terms of modified and unreacted or unmodified macromolecules is not a trivial measurement. From here the analysis of the chemically modified polymer fraction and the variation of the modification level, that is, the yield of the desired reaction, as a function of the process conditions is not indeed an easy task, and the quality of the experimental results and procedures would determine the kind of kinetic studies. Regarding the radical mechanism that goes ahead through the reaction, it is important to note that the termination rates are proportional to the square of the radical concentration, while the relative contribution to the reaction yielding increases along with the radical concentration. A good control of the kinetic process requires fast initiation, looking for the same reaction probabilities for any active site, fast exchange possibilities between sites with different activity, and a low contribution of chain breaking reactions or any other undesired secondary processes. Next section describes the main findings by the present authors following these fundamental principles.

13.5.2 Chemical Modification of Polypropylenes by Grafting of Polar Monomers After preliminary studies conducted on isotactic polypropylene and maleic anhydride in the presence of dicumyl peroxide as initiator, we concluded, in agreement with several others, the existence of a maximum grafting level, or ‘‘ceiling,’’ whatever was the initial concentration of maleic anhydride and peroxide in the reaction media. By using a statistical design method for experimental runs, it was put into evidence that results needed to be carefully evaluated in order to find the true evolution of the process. According to initial findings on batch reactors, both in solution and in the molten state, the role of the reaction time in overall yield of the process was too far to be neglected (39) (Fig. 13.10). In order to check that point and also the role stereospecificity could play in the reaction extension, that is the molecular motion possibilities of the macromolecular coreactant, atactic polypropylene was chosen to be modified by grafting maleic anhydride in the same mode as before for isotactic polypropylene. The influence of the temperature of the process had also to be checked because the atactic polymer could not be processed at the same temperature as the isotactic one due to its nature. Results were fully discussed elsewhere (40). Figure 13.11 exhibits the dynamic and unsteady character of the grafting process.


399

Figure 13.10 Grafting level (w/w) on polypropylenes chemically modified by maleic anhydride on batch processes. The role of the reaction time. (From Reference 39 with permission of John Wiley & Sons, Inc.)

This was in good agreement with the chemisorption mechanism over a changing surface, which had been proposed in Reference (39) to be responsible for the desired reactions. As shown in Fig. 13.11 the highest yield on grafting over the atactic polymer is also observable. If the assumption that polyolefin was responsible for grafting reaction were true, then by changing the polar monomer to be bonded, the grafting results ought to be showing similar tendencies (41,42).

Figure 13.11 The influence of the stereoregularity and the reaction time on the w/w grafting level of polypropylenes chemically modified by maleic anhydride. (From Reference 40 with permission of John Wiley & Sons, Inc.)

400

Polyolefin Blends

Figure 13.12 Grafting level (w/w) on isotactic polypropylene chemically modified by p-phenylenbis-maleamic acid on batch processes. The role of the reaction time. (From Reference 41 with permission of Elsevier.)

Figure 13.12 exhibits the unsteady evolution of the p-phenylen-bis-maleamic acid grafts on isotactic polypropylene, while plots in Fig. 13.13 confirm the same on atactic polypropylene. One of the oldest aspects of discussion about the topic under consideration was the possibility of grafting of polysuccinic anhydride sequences into the polypropylene backbone. The condensation reaction occurring between resorcinol and single succinic anhydride groups, previously bonded to the macromolecule, following Scheme 13.1 leads to the conclusion of the nonexistence of such polysuccinic anhydride grafts, as well as to the finding of a new family of fluorescent additives

Figure 13.13 Grafting level (w/w) on atactic polypropylene chemically modified by p-phenylene-bismaleamic acid on batch processes. The role of the reaction time. (From Reference 42 with permission of John Wiley & Sons, Inc.)


401

Scheme 13.1 The reaction between resorcine and succinic anhydride grafted polypropylene to obtain the succinyl-fluorescein grafted polypropylene.

that has proven to be efficient as interfacial modifiers either in blends (9,34–36) or in polypropylene based composites (31–33,35,43). Taking into account the general chemisorption reaction scheme proposed in Reference 39 as reproduced in Fig. 13.14, new experimental results were obtained.

Figure 13.14 The Chemisorption process for maleic anhydride grafting in polypropylene. (From Reference 39 with permission of John Wiley & Sons, Inc.)

402

Polyolefin Blends

Recently a general balance of species for the unsteady grafting process of maleic anhydride into polypropylene in solution has been fully discussed in a recent paper (41), giving rise to the following general equation.

Scheme 13.2 General species balance involved in the reaction between maleic anhydride and polypropylene by a peroxide initiator in the presence of a solvent.

The pathway based on the general scheme for the classical consecutive reaction model, as shown in Fig. 13.15, contains all the relevant species involved in the chemical modification of polypropylene by a polar monomer, and agrees with a series of remarks associated with findings in previous studies (39–43,45). This implies the generation of free succinic anhydride (SA) as a reaction byproduct, and obtaining of the grafted polymer in a ratio of only 2/3 with respect to the remaining radicals active in the polymer bulk (unable to yield grafting reactions and obliged to lose their activity by b-scission processes leading to degradation of the polymer).

Figure 13.15 The consecutive reaction model for the unsteady chemisorption process of grafting of maleic anhydride in polypropylenes. (From References 44 and 46 with permission of John Wiley & Sons, Inc.)


403

The latter is confirmed by the fact that the reaction proceeds with no loss of efficiency when performed in the presence of radical traps, or the thermal stabilizers usually used in polyolefin processing operations (38–45). The low grafting yield was obtained, both in terms of the number of bonded polar groups and of conversion levels. However, the experimental evidence reveals higher grafting yields obtained when the process occur in the melt rather than in solution (38,40,45). Scheme 13.2 once particularized for the molten state reaction media (46) drive us to the following.

Scheme 13.3 General reaction scheme between maleic anhydride and polypropylene by a peroxide initiator in the molten state.

Here it can be seen that for each SA group grafted onto the PP backbone, a second group is attached but involving some other sequence of the ‘‘solvent’’ PP. Further, every two SA groups grafted onto a-PP-SA require that three PP radicals remain active, that is, transferring radical activity from the reactant mass, which will continue with increasing reaction time until it achieves full deactivation. These three active radicals are, by definition, unable to graft MAH. They therefore manifest their activity through, for example, degradation processes such as chain scissions (in the case of PP) or by reacting with other species in the reaction mass (with the exception of MAH) such as the thermal stabilizer molecules. This can slow down or even stop the degradation processes. The two grafted SA units obtained in the species balance for the batch solution process in Scheme 13.1 may be considered grafted onto equivalent reactive sites (propylene sequences on the PP backbone). However, they must be essentially different from the two grafted SA groups obtained in the molten state, whose reaction surroundings are different due to the absence of solvent molecules. Scheme 13.3 provides a coherent explanation for earlier findings (38,45) regarding the higher grafting levels of i-PP-SA obtained with the molten state process than with the batch solution process (always with the same concentration of peroxide and MAH with respect to polypropylene). These findings were confirmed when the macromolecular coreactant was a-PP, as discussed elsewhere (39,40,43). As Scheme 13.3 shows, the presence of three active PP radicals for each two grafted SA groups at any given reaction time is coherent with the unsteady character of the reaction proposed (39). The ability of resorcine to act as a molecular probe, able to stabilize the SA groups on the polypropylene backbone according to the condensation reaction displayed in Scheme 13.1, was used in Reference 46 to validate the proposed

404

Polyolefin Blends

Figure 13.16 Experimental validation of the intermediate reaction on the consecutive reaction model describing the grafting of maleic anhydride into polypropylene. (From Reference 46 with permission of John Wiley & Sons, Inc.)

pathway. Figure 13.16 clearly shows the very high risk of neglecting the early stages of the process, a problem of existing kinetic models for the grafting of polar monomers onto polyolefines. Indeed, both reactions in Schemes 13.2 and 13.3 shows that in the balances of the species involved in the solution and molten state processes, three polypropylene radicals remain active for every two grafted SA groups. These become involved in the termination of the radical step that could proceed, as it is well accepted by disproportion and recombination, as the two major processes involved in the termination step of the classical three-step radical processes. In light of experiment results fully discussed in a recent paper (46), termination by disproportion would seem to be the dominant mechanism for the grafted species obtained in molten state, while recombination between radical species yielding SA groups trapped between two polypropylene sequences may be responsible of some 40% of the SA units being unable to react with the resorcine molecules. These could not undergo the structural rearrangement shown in Scheme 13.1 and would remain as grafted SA units when the condensation step is complete. From these findings, it follows for both the solution and the molten state reactions, and due to the unsteady character of the process that a dynamic distribution of grafted groups all along the polypropylene backbone takes place. These groups would increase in number until an optimum reaction time is reached. Once this has passed, an increasing number of grafted groups would form end chains. If the reaction is allowed to continue, a critical time would be reached, after which degradation processes would be the only means of deactivating the radical activity


405

still remaining in the bulk of the reactant mass. In fact, as fully discussed by us (46), three different grafting population distributions for each of the a-PP-G samples produced by the batch solution and molten state processes were identified as welldefined fractions obtained by selective extraction in boiling n-heptane. The first was a nonsoluble fraction of a-PP-G unable to leave the Soxhlet cartridge; the second fraction from the ‘‘macroscopic’’ a-PP-G sample corresponded to soluble grafted chains able to leave the Soxhlet cartridge and recovered by classic precipitation in a nonsolvent such as methanol; and the third, soluble fraction—termed the soluble ad infinitum fraction—was recovered after the evaporation of the solvent–nonsolvent mixture and vacuum drying. While the insoluble species were characterized by the highest grafting levels obtained for a single a-PP-G sample close to the theoretically highest possible (according to sterical hindrance) for any given a-PP segment, the precipitant fraction showed the lowest grafting levels, far below the average macroscopic graft value; and the soluble ad infinitum fraction was characterized by grafting values similar to those of the macroscopic average, and almost identical to those of the a-PP-G samples obtained by the batch solution process. According to the experiment results for each of the solution batches, as displayed in Fig. 13.17, the three fractions showed the decreasing weight sequence soluble>insoluble>soluble1, while those produced by the molten state

Figure 13.17 The three different grafted species population distributions for the a-PP-G samples as revealed by solvent extraction in boiling n-heptane.

406

Polyolefin Blends

batches (with reaction conditions close to the optimum) followed the order soluble1>soluble>insoluble. Otherwise, if the initial concentration of maleic anhydride was far from optimum, either by default or because of an excess of the optimum coating on the chemisorption process (39), an inversion occurs between the weight populations of the two soluble fractions. The sequence then becomes soluble>soluble1o insoluble. The very low weight population of the insoluble fraction in all cases is in good agreement with a nonoptimum initial concentration of MAH. In the light of these findings, it may be hypothesized that the shortest chains, that is, those coming from disproportion termination reactions plus others produced by chain scissions processes (if there are any) should contribute to the (soluble)1 fraction. The inversion between the amounts of the different sequences in fractions

Figure 13.18 Modelization of grafted polar groups sequences on polypropylene for succinic anhydride (SA), succinyl-fluoresceine (SF), and p-phenylen-bis-maleamic acid ( p-PBM): HOMO (highest occupied molecular orbital) and LUMO (lowest unoccupied molecular orbital) draws and the potential surfaces for each group.


407

of samples produced by the solution process (the lowest weight population) and in those obtained by the molten state process when close to optimal conditions (the highest weight populations) should agree with the recombination termination processes proposed for the solution reactions, in contrast with the disproportion reactions that should be the dominant termination pathway when the graft processes are carried out in the molten state. The way the properties of the original unmodified polypropylenes are affected by the presence of the different polar groups bonded to their macromolecular structure remains an open line of work not only to the need characterization of the modified polyolefin, but to take the full control of the modification processes in terms of the number and localization of the sites where the polar groups are attached to the macromolecule. Further studies on modified polymer structure are in progress now in our laboratories following those about their molecular weights, the thermal behavior of the modified polymers in terms of the first- and second-order thermal transitions, respectively, assigned to the crystalline and to the amorphous phases (47–49), and those dealing with both their radiation behavior, that is, vibrational (50,51) and NMR (52) spectroscopies as well as their molecular modeling (53). Main goal would be to develop correlations between the experimental and theoretical structural data of the chemically modified polypropylenes and their different interfacial modification capabilities on polymer blends and composites, not only by the chemical structure of the attached polar group (Fig. 13.18), but also by the number of those attached groups and where they are alocated all along the polymer backbone (54–63).

13.6

CONCLUSIONS

As the twenty-first century goes ahead, the organic materials remain on the vanguard of the development of new and advanced materials and trends to approach to the Nature. Furthermore, it is evident that the new technological frontiers to be crossed lay on the development and understanding of the fundamentals that govern the properties of the material combinations, rather than on the deepest understanding of the application of a particular material. Environmental considerations and sustainable development criteria when applied to the materials design field should favor the applications based on the organic thermoplastic polymers (64). The high versatility of polypropylene has placed it beyond the border of commodities to become an engineering polymer. Indeed, new developments based on this polyolefin would be the key in the strategic area of the heterogeneous materials based on organic polymers, both from an academic as well as technical point of view. Because of the increasing demand on the full control over the material properties from macro up to the nanoscales, the top-down approach in the modeling and development of new, advanced materials makes it necessary to improve analysis and characterization tools together by interdisciplinary research teams, always faithful to the rigors of scientific inquiry.

408

Polyolefin Blends

ACKNOWLEDGMENT Results compiled and referred to in this chapter were partially financed and developed under the following Spanish Research Public Projects: MAT93-0115, MAT960386, and MAT2000–1499.

NOMENCLATURE a-PP-G b lr Dn d hm w g n1 n2 PP PA 6 S SA SF sF sR Tg V We G

Chemically modified atactic polypropylene by grafting of the polar group G Carbon atom onto polypropylene backbone where the scission process goes ahead Longest dimension in a particle Mean size of the particles in a dispersed phase Loss phase angle on DMA tests Matrix viscosity in a polymer blend Shape factor Shear rate Width/longest dimension ratio in a particle Thickness/longest dimension ratio in a particle Polypropylene Polyamide 6 Particle surface Succinic anhydride Succinyl-fluoresceine Tensile strength at yield Tensile strength at break Glass transition temperature Particle volume Weber number Interfacial tension between the components of a polymer blend

REFERENCES 1. F. N. Cogswell, Thermoplastics Polymer Composites, Butterworth-Heinemamm, Oxford, 1987. 2. E. Helfand and Y. Tagami, J. Polym. Sci., Polym. Lett., 9, 741 (1971). 3. D. V. Rosato, D. P. DiMattia, and D. V. Rosato, Designing with Plastics and Composites, Van Nostrand-Reinhold, New York, 1991. 4. L. E. Nielsen, Mechanical Properties of Polymer and Composites, Vols I and II, Marcel Dekker, New York, 1974. 5. H. S. Katz and J. V. Milewski, Handbook of Fillers and Reinforcements for Plastics, Van NostrandReinhold, New York, 1978. 6. L. A. Utracki, Polymer Alloys and Blends, Hanser, Munich, 1989. 7. W. Baker, C. Scott, and G. H. Hu (eds.), Reactive Polymer Blending, Hanser, Munich, 2001.


409

8. L. Mascia, Thermoplastics Materials Engineering, 2nd edition, Elsevier, London, 1989. 9. J. Mª. Garcıá-Martıńez, S. Areso, and E. P. Collar, J. Macromol. Sci. B Phys., B40, 387 (2001). 10. R. Toth, A. Coslanich, M. Ferrone, M. Fermeglia, S. Pricl, S. Mierturs, and E. Chiellini, Polymer, 45, 8075 (2004). 11. B. Wunderlich, Macromolecular Physics, Vols 1 and 2, Academic Press, New York, 1973. 12. J. M. Schultz, Polym. Eng. Sci., 24, 770 (1984). 13. P. W. Anderson, Science, 177, 393 (1972). 14. G. Xu and S. J. Lin, J. Macromol. Sci. Rev. Macromol. Chem. Phys., C34, 555 (1994). 15. Y. H. R. Jois and J. B. Harrison, J. Macromol. Sci. Rev. Macromol. Chem. Phys., C36, 433 (1996). 16. M. K. Naqui and M. S. Choudhari, J. Macromol. Sci. Rev. Macromol. Chem. Phys., C36, 601 (1996). 17. G. Moad, Prog. Polym. Sci., 22, 81 (1999). 18. R. Aris, Mathematical Modeling Techniques, Pittman, San. Francisco, 1978. 19. P. K. Freakley and G. A. W. Murray, J. Appl. Polym. Sci., Appl. Polym. Symp., 50, 150 (1992). 20. E. P. Plueddemann, Interfaces in Polymer Matrix Composites, Academic Press, New York, 1974. 21. A. Whelan and J. L. Craft, Developments in Plastics Technology, Vol. 2, Elsevier, London, 1985. 22. H. Ishida, Proc. NATO ASI Series, Akovali, G. (ed), Appl. Sci., 230, 169 (1993). 23. T. F. Cooke, J. Polym. Eng., 7, 197 (1987). 24. J. Taranco, O. Laguna, and E. P. Collar, J. Polym. Eng., 11, 359 (1992). 25. J. Taranco, J. M. Garcıá-Martıńez, O. Laguna, and E. P. Collar, J. Polym. Eng., 13, 287 (1994). 26. E. P. Collar, J. M. Garcıá-Martıńez, O. Laguna, and J. Taranco, J. Polym. Mater., 13, 111 (1996). 27. J. Mª. Garcıá-Martıńez, O. Laguna, and E. P. Collar, J. Polym. Mater., 15, 127 (1998). 28. N. H. Ray, Inorganic Polymers, Academic Press, London, 1978. 29. E. P. Collar, O. Laguna, S. Areso, and J. M. Garcıá-Martıńez, Eur. Polym. J., 39, 157 (2003). 30. D. W. Van Krevelen, Properties of Polymers, 3rd edition, Elsevier, Amsterdam, 1990. 31. J. Mª. Garcıá-Martıńez, O. Laguna, and E. P. Collar, J. Polym. Eng., 17, 269 (1998). 32. J. Mª. Garcıá-Martıńez, O. Laguna, S. Areso, and E. P. Collar, J. Polym. Sci. Polym. Phys., 40, 1371 (2002). 33. J. Mª. Garcıá-Martıńez, O. Laguna, S. Areso, and E. P. Collar, J. Polym. Sci. Polym. Phys., 38, 1554 (2000). 34. E. P. Collar, C. Marco, S. Areso, and J. M. Garcıá-Martıńez, J. Macromol. Sci. B Phys., B40, 369 (2001). 35. J. Mª. Garcıá-Martıńez, S. Areso, M. A. Go´mez, C. Marco, and E. P. Collar, Hasylab Annual Report, Part I, Digital Realease, 2003. 36. C. Marco, S. Areso, E. P. Collar, and J. M Garcıá-Martıńez, J. Polym. Sci. Polym. Phys., 40, 1307 (2002). 37. G. Natta, F. Severini, M. Pegoraro, and C. Tavazzani, Makromol. Chem., 119, 201 (1968). 38. J. Mª. Garcıá-Martıńez, J. Taranco, O. Laguna, and E. P. Collar, Int. Polym. Proc., 9, 246 (1994). 39. J. Mª. Garcıá-Martıńez, O. Laguna, and E. P. Collar, J. Appl. Polym. Sci., 65, 1333 (1997). 40. J. Mª. Garcıá-Martıńez, O. Laguna, and E. P. Collar, J. Appl. Polym. Sci., 68, 483 (1998). 41. J. Mª. Garcıá-Martıńez, A. G. Cofrades, O. Laguna, S. Areso, and E. P. Collar, Eur. Polym. J., 36, 2253 (2000). 42. J. Mª. Garcıá-Martıńez, A. G. Cofrades, S. Areso, and E. P. Collar, J. Appl. Polym. Sci., 88, 2202 (2003). 43. J. Mª. Garcıá-Martıńez, O. Laguna, S. Areso, and E. P. Collar, J. Appl. Polym. Sci., 70, 689 (1998). 44. J. Mª. Garcıá-Martıńez, S. Areso, and E. P. Collar, J. Appl. Polym. Sci., 102, 1182 (2006). 45. J. Mª. Garcıá-Martıńez, J. Taranco, O. Laguna, and E. P. Collar, Int. Polym. Proc., 9, 346 (1994).

410

Polyolefin Blends

46. J. Mª. Garcıá-Martıńez, S. Areso, and E. P. Collar, J. Appl. Polym. Sci. 104, 345(2007). 47. E. P. Collar, S. Areso, O. Laguna, and J. M. Garcıá-Martıńez, J. Therm. Anal. Calorim., 58, 541 (1999). 48. E. P. Collar, A. G. Cofrades, O. Laguna, S. Areso, and J. Mª. Garcıá-Martıńez, Eur. Polym. J., 36, 2265 (2000). 49. E. P. Collar, S. Areso, O. Laguna, and J. M. Garcıá-Martıńez, Eur. Polym. J., 35, 1861 (1999). 50. J. Mª. Garcıá-Martıńez, S. Areso, M. A. Go´mez, C. Marco, and E. P. Collar, Hasylab Annu. Rep., Part I, 845, 2002. 51. J. Mª. Garcıá-Martıńez, O. Laguna, S. Areso, and E. P. Collar, J. Appl. Polym. Sci., 73, 2837 (1999). 52. L. Garrido, Ma. M. Pe´rez-Meńdez, J. Aisa, J. M.a Garcia-Martinez, C. Marco, G. Ellis, S. Areso, E. P. Collar, Final Report MAT 1499-2000, Spanish Public Research Project, unpublished results, 2003. 53. Mª. M. Pe´rez-Meńdez, L. Garrido, J. Aisa, J. M.a Garcia-Martinez, C. Marco, G. Ellis, S. Areso, and E. P. Collar, Final Report MAT 1499-2000, Spanish Public Research Project, unpublished results, 2003. 54. G. Ellis, C. Marco, M. A. Go´mez, E. P. Collar, and J. M. Garcıá-Martıńez, J. Macromol. Sci. B Phys., B43, 251 (2004). 55. J. Mª. Garcıá-Martıńez, O. Laguna, S. Areso, and E. P. Collar, Eur. Polym. J., 38, 1583 (2002). 56. J. Mª. Garcıá-Martıńez, O. Laguna, S. Areso, and E. P. Collar, J. Appl. Polym. Sci., 81, 625 (2001). 57. E. P. Collar, S. Areso, O. Laguna, and J. Mª. Garcıá-Martıńez, J. Polym. Mater., 15, 363 (1998). 58. E. P. Collar, S. Areso, O. Laguna, and J. Mª. Garcıá-Martıńez, J. Polym. Mater., 15, 355 (1998). 59. E. P. Collar, S. Areso, O. Laguna, and J. Mª. Garcıá-Martıńez, J. Polym. Mater., 15, 237 (1998). 60. J. Taranco, O. Laguna, and E. P. Collar, J. Polym. Eng., 11, 345 (1992). 61. J. Taranco, O. Laguna, and E. P. Collar, J. Polym. Eng., 11, 335 (1992). 62. J. Taranco, O. Laguna, and E. P. Collar, J. Polym. Eng., 11, 325 (1992). 63. J. Taranco, O. Laguna, and E. P. Collar, J. Polym. Eng., 11, 315 (1992). 64. Opinion, Nature, 419, 543 (2002).

Chapter

14

Polypropylene/Ethylene– Propylene–Diene Terpolymer Blends Chang-Sik Ha,1 Subhendu Ray Chowdhury,2 Gue-Hyun Kim,3 and Il Kim1

14.1 INTRODUCTION Different polymers have different unique properties. To combine these unique properties of component polymers blending is an attractive means. There are a few methods to make polymer/polymer blends: solution blending, melt extrusion, in situ polymerization, among others. Compatibility usually plays a major role in the development of properties. The blends prepared by melt mixing of thermoplastic materials and rubbers have met industrial needs in recent years. Thermoplastic elastomeric materials have many important applications including cable and wire especially in mineral, electronic equipment, and automobile industries. The most commonly used method of obtaining thermoplastic elastomer in materials is to toughen plastics by blending rubbers and plastics. Among the most versatile polymer matrices, polyolefins such as polypropylene (PP) are the most widely used thermoplastics because of their well-balanced physical and mechanical properties and their easy processability at a relatively low cost that makes them a versatile material. PP has the disadvantage of becoming brittle at low temperature, however, because of its high transition temperature and high crystallinity. The best way to improve its impact strength is to blend PP with elastomers 1

Department of Polymer Science and Engineering, Pusan National University, Busan 609-735, Korea

2

Department of Materials Science and Engineering, Pennsylvania State University, University Park, PA

16802, USA 3

Division of Applied Bioengineering, Dongseo University, Busan 617-716, Korea


411

412

Polyolefin Blends

Figure 14.1 Young’s modulus of PP/EPDM blends plotted against EPDM content (blends were prepared with 20 rpm of a rotor speed of an internal mixer at 220 C). (From Reference 4 with permission from Elsevier Science Ltd.)

particularly ethylene–propylene copolymers and terpolymers (EPM and EPDM, respectively). The blend of PP and EPDM has been prepared by different means, melt mixing, dynamic radiation curing, and ultrasonic curing, among others. These PP/EPDM blends have been widely studied from different angles; that is, structure–properties relationship, morphology, mechanical properties, rheology, thermal properties, among others. In this sense, some monographs and reviews have been published on the PP/EPDM blends and related materials for these two decades (1–3). Thus, readers should refer to those review articles and book chapters to get general guidelines on the preparation and properties of PP/EPDM blends. In this chapter, we do not reproduce those general aspects of the PP/EPDM blends. Instead, we wish to review recent interesting reports on the PP/EPDM blends in terms of thermoplastic polyolefins (TPO) and thermoplastic vulcanizates (TPVs) as commercially important products. We also briefly touch the applications of the PP/EPDM blends for readers. A typical example of the effect of EPDM addition on the mechanical properties of PP can be seen in the work of DA Silva and Coutinho (4). They described the effect of EPDM amount and also processing condition on the mechanical properties of PP/EPDM blends. As EPDM contents increases, the impact strength of PP/EPDM blends increases but the tensile strength and Young’s modulus decrease and the elongation at break increases. Figure 14.1, for instance, illustrates the effect of EPDM contents on the Young’s modulus of PP.

14.2 PP/EPDM BLENDS 14.2.1 Toughness and Crystallization Behaviors of PP/EPDM Blends The toughness of PP/EPDM blend was investigated by Huang et al. (5) over a wide range of EPDM content and temperature. It is seen in Fig. 14.2 that the Izod impact strength of PP increases with increasing EPDM contents and temperature. It is

Chapter 14 Polypropylene/Ethylene–Propylene–Diene Terpolymer Blends

413

Figure 14.2 Variation of notched Izod impact strength with temperature for various EPDM content (weight fraction). (From Reference 5 with permission from Elsevier Science Ltd.)

worth noting that both toughness and brittle–ductile transition (BDT) of the blends were a function of the notch radius (R) for the Izod notch impact tests. (In Fig. 14.2, a–c three different notch tips are shown; Notch A is a 45 V-shaped notch with the tip radius (R) of 0.25 mm, notch B is a 45 V-shaped notch with the R of 1.0 mm, and notch C is a rectangular notch.) At test temperature, the toughness tended to decrease with increasing 1/R for various PP/EPDM blends. The brittle–ductile transition temperature (TBT) increased with increasing 1/R, whereas the critical interparticle distance (IDc) reduced with increasing 1/R. The different curve of IDc versus test temperature ðTÞ for notches reduces down to a M M –T, where TBT is the TBT of PP itself for master curve if plotting IDc versus TBT M a given notch, indicating that TBT –T is a more universal parameter that determined the BDT of polymers. The thermal and morphological behaviors of PP/EPDM blends were studied by Da Silva and Coutinho (6) using differential scanning calorimetry (DSC) and polarized optical microscopy (POM), respectively. Crystallization kinetics of PP/ EPDM blends were found similar. Ten to twenty weight percent addition of EPDM resulted in increasing of spherulite size (Fig. 14.3). Heat of fusion and crystallinity degree of PP/EPDM systems decreased when EPDM contents were increased.

414

Polyolefin Blends

Figure 14.3 PP spherulite size vs. EPDM content. (From Reference 6 with permission from John Wiley & Sons, Inc.)

14.2.2 Compatibilization of PP/EPDM Blends Sometimes, the compatibilization of PP/EPDM blends has been the key issue to improve the properties of the blends. Lopez-Manchado’s group (7) studied the effect of grafted PP on the compatibility and properties of PP-EPDM thermoplastic elastomer blends. They functionalized the isotactic PP (iPP) through grafting in Brabender plasticorder with two itaconate. The functionalization of iPP was performed by melt blending through grafting with two itaconic acid derivatives, monomethyl itaconate and dimethyl itaconate (MMI and DMI, respectively). Grafting was performed at 190 C using two different initial monomer concentration and 2,5 dimethyl 2,5 bis (tert-butyl peroxy) hexane as radical initiator. Some PP/EPDM blends were made from unmodified PP and modified PP (i.e., grafted PP with MMI or DMI) with EPDM. The flow properties analyzed by torque values, melt index, and rheological studies showed that the blends made with grafted PP have better processability, show lower viscosity with this effect being more significant in DMI-modified PP. Good interaction between two phases are evident from a dynamic mechanical analysis (DMA) and tensile properties. The functional polar monomer acted as a compatibilizer as well as nucleating agent for PP crystallization. It shows a substantial decrease in the halftime of crystallization, which is attributed to the presence of a greater number of nuclei in the crystallization process. Su et al. (8) studied the mechanical properties and morphological structure relationship of blends based on sulfated EPDM ionomer and PP. They synthesized Zn2þ neutralized low degree sulfated EPDM (Zn-SEPDM) ionomer and PP blends and studied their mechanical properties. They found that Zn2þ neutralized low degree sulfated EPDM ionomer and PP blends have better mechanical properties than those of PP/EPDM blend, as shown in Fig. 14.4. They explained the reason why mechanical properties are higher for Zn-SEPDM and PP than for PP and EPDM using scanning electron microscopy (SEM) (Fig. 14.5). Finer dispersed phase size and the shorter interparticle distances are the main reasons for the improved mechanical properties of the PP/EPDM blend.


415

Figure 14.4 Stress–strain curve of neat PP, PP/EPDM, and PP/Zn-SEPDM blends (a) neat PP, (b) PP/EPDM, (c) PP/0.03 mol% Zn-SEPDM, (d) PP/0.06 mol% Zn-SEPDM. (From Reference 8 with permission from John Wiley & Sons, Inc.)

Figure 14.5 SEM micrographs of fractured etched PP/EPDM blends: (a) PP/EPDM, (b) PP/0.03 mol% Zn-SEPDM, (c) PP/0.06 mol% Zn-SEPDM. (From Reference 8 with permission from John Wiley

416

Polyolefin Blends

Figure 14.6 The plot of Tg vs. blend composition in the m-PP/Zn-EPDM blends. (From Reference 9 with permission from Wiley Interscience.)

Similarly, Ha et al. (9) blended Zn-SEPDM with maleic anhydride grafted PP (PP-gMAH). Using light scattering, DMA, and FT-IR spectroscopy, they found that the compatibility of PP/EPDM blend was significantly improved by the use of both EPDM ionomer and maleic anhydride grafted PP. In Fig. 14.6, a sigmoidal trend in Tg of the PP-g-MAH/Zn-SEPDM blends as a function of a blend composition suggests that there exists strong interaction between the maleic anhydride grafted PP and the ionomeric EPDM and compatibilization was achieved between PP and EPDM.

14.2.3 Ternary Blends and Composites from PP/EPDM Blends In order to improve properties and compatibility of PP/EPDM blends, ternary blends and composites are sometimes prepared from the PP/EPDM blends. For instance, Sanchez et al. (10) prepared ternary blends of PP, high density polyethylene and EPDM with several blending ratios and investigated the melt rheological behaviors. They discussed the effect of the shear rate on the viscosity and flow curve in terms of the exponent of low power for a non-Newtonian liquid. They showed that addition of an elastomer to the polyolefin blends changes the shape of the viscosity–composition curve, and they discussed it in terms of the possible morphology of the blend. Similar works have been also reported by Ha et al. (11,12).


417

14.2.4 Application of Radiation A series of gamma-radiation cross-linking of polymer blends containing fresh or waste EPDM and PP was prepared by Zaharescu et al. (13,14). They figured out the optimal dose range for the efficient cross-linking of all EPDM/PP blends to be 40–180 kGy. Thermal stability of the studied mixtures was assessed in order to state the contribution of the components to the radiation compatibilization of investigated polymers. Figures 14.7 and 14.8 illustrate typical results. They investigated the effect of radiation on tensile strength and elongation at break of EPDM/PP blend. Fresh and waste PPs were separately compounded with EPDM. In spite of the improvement in their gel content at 150 kGy, the degraded component causes alternation in mechanical properties. The effect of ionizing radiation on thermal oxidation of PP/EPDM blend over the range of total gamma doses up to 250 kGy was also studied by Zaharescu and

Figure 14.7 Change of the elongation at break of (a) EPDM/fresh PP and (b) EPDM/waste PP samples (&,^) unirradiated blends, (&,*) irradiate blend (150 kGy). (From Reference 14 with permission from Elsevier Science Ltd.)

418

Polyolefin Blends

Figure 14.8 Change in tensile strength of (a) EPDM/(fresh) PP and (b) EPDM/(waste) PP samples, () unirradiated blends; (c) irradiated blends (150 kGy). (From Reference 14 with permission from Elsevier Science Ltd.)

Budrugeac (15). They studied the influence of irradiation dose on oxidation induction periods by oxygen uptake and thermal analysis on polymer samples containing various concentrations of components (100/0, 80/20, 60/40,40/60, 20/80, and 100/ 0 w/w). Drastic decrease in oxidation induction time was observed for low doses. The competition between cross-linking and scission has been examined on the basis of radical recombination on postirradiation time. Effect of specimen formulation on oxidation induction time has been discussed considering the antagonistic processes, such as crosslinking, and oxidative degradation. The effect of ultrasonic irradiation on the mechanical property, morphology, and crystal structure of PP/EPDM blends were examined by Chen and Li (16). Appropriate ultrasonic intensity can increase the toughness of PP/EPDM blends noticeably. SEM showed that with ultrasonic irradiation, the morphology of a well-dispersed EPDM phase is formed in the PP/EPDM blend. Ultrasonic irradiation interestingly


419

Figure 14.9 SEM micrographs of fractured surface at the temperature of liquefacient nitrogen and treated by xylene at 20 C of PP/EPDM ¼ 70/30 (A: 0þ0, B: 150þ0, C:150 þ 100). (From Reference 16 with permission from Wiley Interscience.)

brought the Tgs of PP and EPDM closer. Ultrasonic irradiation increases crystallization PP/EPDM blend and b-crystal of PP form in PP/EPDM blend, which is proven by X-ray diffraction. The authors again studied the morphology and compatibility of PP/EPDM blend change by ultrasound irradiation. Stable morphology with reduced dispersed phase size is obtained. Dynamic rheological analysis indicated more homogeneous internal structure of PP/EPDM blend. If the ultrasonic irradiation time is increased, the blend becomes more homogeneous due to the reduction of interfacial tension between PP and EPDM decreased (Fig. 14.9). Zaharescu (17) studied the compatibility of PP/EPDM blends prepared by gamma irradiation. He found that the composition of polymer specimens does not influence the relative contribution of constituents to the calculated heat capacity ðCp Þ values for tested blends exposed to the same dose.

14.3 DYNAMICALLY VULCANIZED PP/EPDM BLENDS (OR THERMOPLASTIC VULCANIZATES (TPVs)) The blends of cross-linked EPDM and PP can be prepared in a roll mill or extruder by the ‘‘dynamic vulcanization (DV) (or curing)’’ method where EPDM is vulcanized under shear with curing agent. The dynamically vulcanized thermoplastic elastomers (TPVs) have been widely used in the plastic industry because of their technical advantages in processing as well as their versatile end-use properties (18–22). The blends have important technical advantages in processing because of the thermoplastic nature of the melt, even though they contain a vulcanized elastomer as one

420

Polyolefin Blends

component. In general, vulcanized rubber cannot be reprocessed because of the formation of network structure. Such thermoplastic nature of the TPV might be ascribed to the fact that the network structures are formed in small rubber particles dispersed in the uncross-linked thermoplastic polymer matrix. They have a number of practical advantages over conventional rubber: a short mixing and processing cycle and low energy consumption; the scrap can be recycled; and properties can be easily manipulated by changing the ratio of the components. Unvulcanized EPDM/PP blends cannot replace thermoset rubber because of poor resistance to compression or tension set at elevated temperature or under prolonged deformation and poor oil resistance. Compared with unvulcanized EPDM/PP blends, the following properties are improved in EPDM/PP TPVs: oil resistance, permanent set, ultimate mechanical properties, fatigue resistance, heat deformation, melt strength, among others (3). As a result, EPDM/PP TPVs are quite adequate for most vulcanized rubber applications.

14.3.1 Effect of Cross-linking on the Properties of PP/EPDM TPVs Ishikawa et al. (23) examined the toughening mechanism of PP/EPDM blend before and after crosslinking, as shown in Figs. 14.10 and 14.11. They evaluated yield stress, strength of craze, and density of void, which are dominant factors for enhancing toughness in PP blends. The fracture and deformation mechanism were discussed. Ha et al. (24) studied the structure–property relationship of EPDM/PP blend. EPDM was cured with PP with dicumyl peroxide (DCP) at different shear conditions.

Figure 14.10 Izod impact strength of PP blended with EPDM following selective cross-linking in comparison with that of PP blend before cross-link. (From Reference 23 with permission from John Wiley & Sons, Inc.)


421

Figure 14.11 Variation of blending moment–displacement curve of the round notched bar of PP blended with EPDM with increase in EPDM content. (From Reference 23 with permission from John Wiley & Sons, Inc.)

They also cured EPDM without PP and then blended with PP. The effect of DCP concentration, intensity of the shear mixing, and rubber/plastic composition were studied. They found in blend cure that the melt viscosity increased with increasing DCP concentration in the blends of 75% EPDM and 25% PP but decreased with increasing DCP concentration in blends of 75% PP and 25% EPDM. Melt viscosity increased with increasing DCP concentration for all compositions in cure blend. With increasing intensity of the shear mixing, the melt viscosity decreased. Figures 14.12 and 14.13 show the results. In dynamically vulcanized EPDM/PP blends, the melt viscosity increased with increasing DCP concentration in blends of 75% EPDM and 25% PP but decreased with increasing DCP concentration in blends of 75% PP and 25% EPDM. It is clear that increasing DCP concentration for EPDM/PP (75/25) blend leads to the more molecular restrictions by the chemical cross-links in EPDM. The decrease in viscosity with DCP concentration for EPDM/PP (25/75) blend is attributed to the accelerated mechanochemical degradation of PP in the presence of peroxide during the dynamic vulcanization. It seems that with compositions of PP greater than 50%, mechanochemical degradation of PP becomes predominant, whereas at EPDM/PP (75/25) composition, the cross-linking effect of EPDM is dominant. Organic peroxide, sulfur, and phenolic resin system are generally used for the preparation of TPVs as cross-linking agents. However, all cross-linking agents have their own disadvantages: Organic peroxide leads to the degradation of PP greatly, sulfur causes the odor, and considerable concentrations of phenolic resin and accelerator required for the effective cross-linking deteriorate the impact strength of TPVs. Recently developed dynamically photocross-linked PP/EPDM blends showed improved mechanical properties (25). Especially impact strength was enhanced dramatically due to improved compatibility between PP and EPDM. This improved

422

Polyolefin Blends

Figure 14.12 Shear stress-shear rate curves for EPDM/PP linear blends at 200 C: (*) EPDM 100; (&) EL75PP25; (~) EL50PP50; (&) EL25PP75; (*) PP100. (From Reference 24 with permission from John Wiley & Sons, Inc.)

compatibility was ascribed to the possible existence of grafted chains of PP and EPDM. Jain et al. (26) studied the effect of dynamic cross-linking on the melt rheological property of PP/EPDM rubber blends. Rheological properties of vulcanized and unvulcanized EPDM/PP blends were reported. The melt viscosity increases with increasing EPDM concentration and decreased with increasing intensity of the shear mixing for all compositions. Vulcanized blend displays highly pseudoplastic behavior, which provides unique processability in injection molding and extrusion. The high viscosity at low shear rate provides the intensity of the extruders during extrusion and the low viscosity at high shear rate enables low injection pressure and less injection time. They also explained the property difference for vulcanizates with the help of morphology study. Cure characteristics of EPDM/PP blends were investigated by Sengupta and Konar (27). They calculated the state of cure in blends containing conventional sulfur curing system under variable time–temperature conditions. They found that the activation energy for the cross-linking is almost similar for the virgin EPDM and EPDM/PP mixtures. Cross-link densities in TPVs can be analyzed by swollen-state


423

Figure 14.13 Effect of DCP concentration (phr) on viscosity for EB75PP25 blend cure at 200 C (mixed speed 60 rpm): (*) 0.00; (*) 0.33; (~) 0.67; (&) 1.00; (~) 1.33. (From Reference 24 with permission from John Wiley & Sons, Inc.)

nuclear magnetic resonance (NMR) spectroscopy with a network visualization techniques (28,29).

14.3.2 Microstructure of PP/EPDM TPV The enhanced properties of TPV are ascribed to their specific morphology that consists of a continuous PP matrix with tiny cured EPDM particles dispersed throughout the matrix. As the EPDM particle size decreases, the ultimate elongation and tensile strength increase. Therefore, the physical properties depend on the morphology of the EPDM particle size. Their morphological behavior is different from the unvulcanized EPDM/PP blends. According to Abdou-Sabet et al. (30), before dynamic vulcanization, PP was the dispersed phase in the EPDM matrix for 80/20 EPDM/PP composition. However, as dynamic vulcanization progresses, the continuous rubber phase becomes elongated further and further and finally breaks up into rubber droplets. As a result, the PP becomes the continuous phase. The detailed mechanism of the morphological development was suggested for the dynamically vulcanized EPDM/PP (60/40 w/w) blends (31,32). Hot xylene

424

Polyolefin Blends

extraction revealed the formation of agglomerate structure by the rubber particles. As the vulcanization of the rubber proceeds, melted PP molecules adsorb onto the surface of semicrosslinked rubber particles through segmental interdiffusion mechanism. This would lead to the formation of an immobilized PP shell attached to the surface of the rubber particles. When the shelled particles are close enough, they would form an agglomerate structure. When the melt viscosity of PP is matched with the viscosity of EPDM, the formation of higher number but smaller size rubber particle aggregates was observed. This would result in better mechanical properties of the TPV. Figure 14.14 shows the SEM micrograph of dynamically vulcanized EPDM/PP (75/25) blend fracture surfaces etched by hot xylene vapor. The microdomains of EPDM have the shape of dumbbell-like microgel of about 0.8–1.0 mm in size, where the dark portions represent the PP phase extracted out by hot xylene vapor. The morphology of the microgel domain of EPDM reveals the reason why the dynamically vulcanized blend can be processed and the dynamic vulcanization prevents the cross-linking of EPDM phase from truly continuous network (24). PP/EPDM blends including dynamically vulcanized one show a higher rate of crystallization and higher Tc than those of PP homopolymer (33). It was reported that

Figure 14.14 SEM micrograph of EB75PP25 blend-cure with DCP concentration of 0.67 phr (dynamically cured at 60 rpm). (From Reference 24 with permission from John Wiley & Sons, Inc.)


425

such crystallization behavior was attributed to the role of EPDM to selectively extract the defective molecules within PP crystals and also increase the mobility of neighboring PP chains by the reduction of glass transition temperature (34).

14.3.3 PP/EPDM/Ionomer TPVs Kim et al. (35) reported on the control of miscibility for iPP and EPDM by adding polyethylene-co-methacrylic acid ionomer and by applying dynamic vulcanization (DV). Blending and curing were performed simultaneously, that is, EPDM was vulcanized with DCP in the presence of PP/ionomer. Addition of ionomer and the application of the dynamic vulcanization were effective in enhancing the miscibility of PP and EPDM. It was found that the addition of ionomer and application of dynamic vulcanization were effective in enhancing the miscibility of PP/EPDM binary blend. This was due to the formation of the thermoplastic interpenetrating polymer network (IPN) of the ternary blends (36). The structure and properties of the ternary blends differed depending on the types and contents of ionomer, that is, the ternary blend containing Na-neutralized ionomer did not show a thermoplastic IPN structure, even if the blends were prepared by dynamic vulcanization. When the contents of ionomer and DCP were 15 parts or 10 parts, respectively, the ternary blend containing Zn-neutralized ionomer clearly showed the behavior of a thermoplastic IPN. Although the Na-neutralized ionomer can form an interpenetrating network between PP and EPDM, the possibility is less than that of Zn-neutralized ionomer because of the monovalent nature of Naþ. The crystallization rate of ternary blends is slower than that of the binary blends, and the ternary blends, which include Zn-neutralized ionomer, showed slower crystallization rate than ternary blends that included Na-neutralized ionomer (35). According to the literature, IPNs that possess physical interlocking at interfaces, strongly restrict crystallization (16,37). This IPN structure is postulated for the dynamically vulcanized EPDM and ionomers, especially for the blends containing Zn-neutralized ionomer. When the ionomer content was 5 wt%, the PP and EPDM blends are incompatible, that is, their phases are separated and the domain of EPDM was peeled off from the continuous matrix of PP (35). For the dynamically vulcanized EPDM and PP/ionomer ternary blends with 15 wt% ionomer, compatibilization was achieved between the PP and EPDM phases. The Zn-neutralized ionomer showed a much better compatibilizing effect than Na-neutralized ionomer. The dynamically vulcanized EPDM and PP binary blends showed somewhat quasicleavage fracture topology, regardless of the DCP contents (38). Figure 14.15 shows SEM micrographs of the fractured surfaces taken around the crack-tip for the dynamically vulcanized EPDM and PP/ionomer ternary blends having 20 wt% ionomers when the DCP content is low. The ionomer-added ternary blend shows no clear fracture surface topology of tough materials even the ionomer contents are 20 wt%, regardless of ionomer types. The ionomer-added ternary blends with high DCP content, however, showed different fracture surface topologies. In Fig. 14.16, SEM micrographs of the fractured surface taken around the crack-tip are shown for the 5 and 20 wt% ionomer-added dynamically vulcanized

426

Polyolefin Blends

Figure 14.15 SEM microfractographs of lightly vulcanized ternary blends: (a) EPDM / PP / Naneutralized ionomer (40/40/20) and (b) EPDM /PP/ Zn-neutralized ionomer (40/40/20). The initial crack length was 8 mm and DCP content is 0.33 phr. (From Reference 38 with permission from Springer.)

ternary blends when the DCP content is high. The ionomer-added ternary blends show typical fracture surface topology of tough materials irrespective of ionomer types, and the trend is clearer when ionomer contents are higher. The micrographs reveal well dimple-ruptured topologies, which are usually observed in tough materials. Careful inspection of Fig. 14.16 shows that the fracture surface of 20 wt% zinc-neutralized ionomer-added dynamically vulcanized ternary blends has most clear dimple fracture topology, exhibiting the toughest characteristics among the blends. The application of dynamic vulcanization and the addition of


427

Figure 14.16 SEM microfractographs of highly vulcanized ternary blends: (a) EPDM/PP/Naneutralized ionomer (47.5/47.5/5), (b) EPDM/PP/Na-neutralized ionomer (40/40/20), (c) EPDM/PP/Znneutralized ionomer (47.5/47.5/5), and (d) EPDM/PP/Zn-neutralized ionomer (40/40/20). The initial crack length was 8 mm and DCP content is 1.00 phr (From Reference 38 with permission from Springer.)

428

Polyolefin Blends

ionomer played a synergistic role to enhance the fracture toughness of PP or PP/EPDM blend.

14.3.4 Mechanical and Rheological Properties With increasing cross-link density, a narrowing of EPDM domain size and an increase in the PP ligament thickness were observed by Ellul et al. (29). EPDM domain size is a very important factor for the mechanical properties of EPDM/PP TPV. Smaller EPDM domain provides higher strength and elongation (3). Tensile strength and tension set improves with higher cross-link density. Gupta et al. (39) has studied the effect of dynamic cross-linking on tensile yield behavior of PP/EPDM rubber blends. They prepared blends of PP/EPDM in internal mixer by simultaneous blending and dynamic vulcanization. Dimethyl phenolic resin vulcanized PP/EPDM blends showed higher yield stress and modulus than unvulcanized PP/EPDM blend (Fig. 14.17 and Table 14.1). They found the increase in

Figure 14.17 Change in shape of the stress–strain curves for unvulcanized PP/EPDM blends. (From Reference 39 with permission from John Wiley & Sons, Inc.)

429

Sample notation

33.25 23.50 20.45 16.20 13.00

24.65 20.75 17.80 13.94

00 10 20 30 40

10 20 30 40

25.00 21.60 18.00 14.50

35.00 24.95 23.00 18.02 14.15

Tensile Stress at yield MPa wt % EPDM (YTS) rubber Unaged Aged

24.65 20.75 17.84 13.90

33.25 23.50 20.45 16.20 13.00 25.00 21.60 18.00 14.50

35.00 24.95 23.00 18.02 14.15 21.15 18.03 16.80 13.40

23.95 20.85 18.25 14.60 12.45 22.00 19.40 18.00 13.40

24.62 21.35 20.06 17.25 13.60

Tensile Strength Tesnsile strength at max at break stress, MPa MPa Unaged Aged Unaged Aged

1164 1008 787 690

1240 1050 858 695 576

1195 1082 802 680

1320 1125 924 740 622

56 76 140 350

44 52 66 122 285

Tensile modulus, modulus, (Young’s), Ultimate MPa elongation, Unaged Aged % (UEL)

24 56 99 143

18 20 33 49 100

Area under yield, peak, au

Values of Tensile Parameters for Polypropylene (PP)/EPDM Blends (from Reference 39 with permission from John Wiley & Sons)

Unvulcanized Control blends PP100 PP90EL10 PP80EL20 PP70EL30 PP60EL40 Vulcanized blends PP90EB10 PP80EB20 PP70EB30 PP60EB40

Blend systems

Table 14.1

430

Polyolefin Blends

interfacial adhesion by three-dimensional network is the cause of improvement in mechanical property. Shapes and sizes of EPDM phase were studied by SEM. According to Jain et al. (40), the impact strength of PP increased dramatically when the optimum amount of EPDM was blended. The cross-linking of the EPDM reduced the optimum amount of EPDM. Increased interfacial adhesion due to crosslinking leaded to the reduced EPDM particle sizes and a more uniform particle size distribution. Compared with unvulcanized PP/EPDM blend, the tensile properties of TPV were significantly improved (39,40). It was reported that the deformation is concentrated on the equatorial PP region between the rubber domains perpendicular to the extension direction, and polar ligament PP region between adjacent rubber domains along the extension direction is undeformed (41). As a result, the stress passes through the polar ligament PP region and concentrates on the EPDM domains, and EPDM domains are highly deformed. The high drawability of EPDM domain is the origin of rubber elasticity with ductile thermoplastic matrix. There has been the question why the TPV materials with ductile thermoplastic matrix display rubber elasticity. Several models have been suggested to answer this question (41–47). Inoue group first analyzed the origin of rubber elasticity in TPVs (43). They constructed a two-dimensional model with four EPDM rubber inclusions in ductile PP matrix and carried out the elastic–plastic analysis on the deformation mechanism of the two-phase system by finite-element method (FEM). The FEM analysis revealed that, even at highly deformed states at which almost the whole matrix has been yielded by the stress concentration, the ligament matrix between rubber inclusions in the stretching direction is locally preserved within an elastic limit and it acts as an ‘‘in-situ formed adhesive’’ for interconnecting rubber particles. In a series of works on the TPVs, Boyce et al. also analyzed such unusual elasticity of TPVs by deformation analysis using stress–strain curve of TPVs with a constitutive model and/or simulation (44–46). In particular, Boyce et al. reported the important role of matrix ligament thickness in controlling the initial stiffness and flow stress of the TPVs (45); thinner ligaments lead to earlier matrix yielding and thus earlier formation of the pseudocontinuous rubber phase. Upon formation of the pseudocontinuous rubber phase, the matrix material is seen to accommodate the large straining of the rubber phase by nearly rigid body motion (rotation and translation) of the bulky domains of the matrix; the rubber phase is seen to undergo large contortions as it attempts to deform as an almost continuous network around the ‘‘rigid’’ domains of matrix material. Furthermore, the asymmetry together with the thin matrix ligaments greatly aids the recovery of the material during unloading. Upon unloading, the rubber phase attempts recovery in a rubber-like manner. The bulkier regions of matrix material simply rotate and translate with the recovering rubber domains. The thin ligaments also rotate, but eventually also undergo bending and buckling, which enables the large amount of recovery observed in TPVs. Similar elastic behavior was also reported for nylon 6/EPDM TPVs as well as PP/EPDM TPVs by Oderkerk et al. (47). Hardness and tensile set are the properties usually specified to choose the grade to be used for its end-use properties. TPV hardness can be easily adjusted by


431

Figure 14.18 Notched–Izod impact strength versus EPDM content of uncross-linked and dynamically photocross-linked PP/EPDM blends measured at 25 and 30 C: (From Reference 25 with permission from John Wiley & Sons, Inc.)

changing the PP/EPDM composition. The rubber-rich blends can be used as thermoplastic elastomers, and the plastic-rich blends can be applied as rubber-toughened plastics. The mutual interaction and bulk properties are dependent on the composition. Tang et al. (48) prepared a dynamically photocross-linked PP/EPDM rubber thermoplastic elastomer by exposing the elastomer to UV light while melt mixing in the presence of photoinitiation as well as a cross-linking agent. The effect of dynamic photocross-linking and blend composition on the mechanical properties, morphological structure, and thermal behavior of PP/EPDM blends were investigated. Tensile strength, modulus of elasticity, and elongation at break improved slightly after photocrosslinking (Figs. 14.18 and 14.19). The notched Izod impact strength was also enhanced compared to uncrosslinked blend. From SEM, it is found that for uncross-linked PP/EPDM blends, the cavitation of EPDM particles was the main toughening mechanism, whereas for dynamically photocross-linked blends, shear yielding of matrix became the main energy adsorption mechanism. DSC showed a new smaller melting peak at about 150 C together with a main melting peak at about 166 C for each dynamically photocross-linked PP/EPDM blend. Compatibility between EPDM and PP was improved by dynamic photocross-linking. This was studied by dynamic mechanical thermal analyzer (DMTA). Goharpey et al. (49) studied the relationship between the rheology and morphology of dynamically vulcanized PP/EPDM blends. They performed

432

Polyolefin Blends

Figure 14.19 Notched–Izod impact strength versus photocrosslinking time for PP/EPDM (70/30) measured at 25 and 30 C: (From Reference 25 with permission from John Wiley & Sons, Inc.)

morphological study by SEM on cryogenically fractured samples (Fig. 14.20). Rheological behavior and melt viscoelastic properties of the samples were studied by rheometric mechanical spectrometry (RMS) at a temperature of 220 C. Sample showed a significant viscosity upturn and a strong storage modulus that tended to plateau at low shear rates, for 60% EPDM containing sample (Fig. 14.21). These structure findings were attributed to a network structure resulting from agglomerates formed between the cured rubber particles, as evidenced by the morphological features of the samples. Xiao et al. (50) studied the miscibility of PP and EPDM by DMTA, transmission electron microscopy (TEM), and DSC. The result showed that a decrease in the PP content and an increase in the cross-linking density of EPDM in the EPDM/PP blends caused an increase in the glass transition temperature of EPDM, although there is no change of Tg of PP (Fig. 14.22). The degree of crystallinity is decreased (Table 14.2). They found that mechanical properties of blends prepared by single screw extruder is higher than that made by open mill. From TEM, PP phase is seen as a bright phase and EPDM as a dark phase. With the increase in cross-linking density, the interface between the EPDM and PP becomes less defined and EPDM gradually dispersed in the PP phase became a continuous phase. Two different ways of processing to improve interfacial adhesion of PP and EPDM by introducing MAH were attempted by Ha et al. (51); In one way, the in situ


433

Figure 14.20 Scanning electron microscopy of the dynamically crosslinked EPDM/PP blend samples of different composites: (a) 20/80; (b)40/60; (c) 60/40, ww. (From Reference 49 with permission from Wiley Interscience.)

434

Polyolefin Blends

Figure 14.21 Results of dynamic (&) viscosity h, storage modulus G0 , versus angular frequency v, of the dynamically crosslinked EPDM/PP blend samples of different compositions; 20/80 (^); 40/60 (!); 60/40 (*) w:w. (From Reference 49 with permission from Wiley Interscience.)

grafting and dynamic vulcanization (ISGV) were performed simultaneously from PP and EPDM with MAH in the presence of DCP in an intensive mixer. In another way, PP was first grafted with MAH and then the PP-g-MAH was blended with EPDM in the intensive mixer in the presence of DCP by the dynamic vulcanization. It was found that the glass transition temperatures (Tg) of both PP and EPDM phases were shifted to higher temperature as the EPDM content increased for the blends prepared by both IGSVand DV methods, mainly due to the cross-linking of EPDM. The higher Tgs and larger storage moduli were observed for the blends prepared by the ISGV

Table 14.2 Physical Properties of EPDM/PP Blends (from Reference 50 with permission from John Wiley & Sons). Rubber/plastica

Amount of cure agentsb

Properties

70/30

60/40

50/50

0/100

0.03

0.12

0.21

0.30

Tc, C Tm, C DHf, Jg1 Xc, %

137 161 24 11

134 161 30 14

132 163 33 16

122 164 92 44

139 159 36 17

137 160 34 16

135 160 32 15

134 161 30 14

a

with 5 phr curing agents EPDM60PP40

b


435

Figure 14.22 DMTA curves of EPDM/PP blends with 5 phr of phenolic resin. The mass ratios of rubber to plastic were (-.-) 0/100, (-) 50/50, (—) 60/40, and (. . .) 70/30. (From Reference 50 with permission from John Wiley & Sons, Inc.)

method than those prepared by the DV method, while the morphology showed that the size reduction of dispersed particles in latter blends was larger than that of the former blends. Wang and Cakmak (52) studied the development of structure hierarchy in tubular film blown dynamically vulcanized PP/EPDM blend. The blown films were found to exhibit an unusual asymmetric structure. The PP phase was found to fibrillate at all the outside surface, while the inner surface remained relatively featureless. This was attributed to disproportionally rapid cooling of the outside surface by the air steam blown externally onto the film being extruded. This, in turn, resulted in solidification of very thin PP surface layers that caused their fibrillation under the heavy stress they had to endure. Increase in the blow-up ratio was found to expand this web-like surface texture. As a result of this fibrillation mechanism, the increase of both the blow-up ratio and draw-down ratios was found to reduce the mechanical properties.

436

Polyolefin Blends

14.4 APPLICATIONS OF PP/EPDM BLENDS Because of low cost, high heat deflection temperature (HDT (104 C)), notched impact resistance, improved low temperature impact and flexibility, weather resistance, flame retardancy, and impact resistance, the PP/EPDM blend has got a widespread applications. These unique characteristics of this thermoplastic elastomer blend make it an attractive alternative to conventional elastomer in a variety of markets such as automobile industries, wire, cable insulator, automobile bumpers and fascia, hose, gaskets, seals, weather stripping, among others. The market of PP/EPDM blends has grown dramatically because of its recycling ability and processability by conventional thermoplastic processing equipment. The unique characteristics of thermoplastic elastomer made it an attractive alternative to conventional elastomers in a variety of markets. Liu et al. showed from the experimental blends (53) that materials cost reduction of between 30% to 50% is possible in comparison to commercial products if one applies the PP/EPDM blends to the construction of a basketball court, a tennis court, and a roller hockey rink, which were estimated around $7000, $14,000, and $40,000, respectively. The cost comparison took into account the percentage of rubber or PP used in experimental blend, the exponential factor for a scale-up process and the overall surface area of the specific applications. Among many possible application of this blend two readily feasible applications are roofing and flooring. It was reported that the strategies used by several producers to design commercial TPV formulations are very similar (54). All the EPDMs used in several commercial TPV formulations showed very similar Tg and similar EPDM: oil ratio. They are also believed to have ethylene content of about 60% as supported by the endothermic peaks right after the EPDM-rich phase Tg. Considerable amounts of oil were found in all the formulations, and oil is believed to be preferentially located in EPDM. Oil improves TPV processability, especially in grades with higher crosslinked EPDM parts. Some oil may also remain in the amorphous region of the PP phase, thus improving the elasticity of the blend. The Ellul group reported the shear flow behavior and oil distribution between phases in TPVs (55). The distribution of the high temperature oil between the PP melt and the EPDM was a key parameter because this affected the viscosity of the PP/oil medium. Several PP/oil mixtures were prepared and their viscosity curves were correlated with the neat PP melt viscosity curves by means of shift factors varying with oil concentration. The oil distribution between the PP and EPDM phases was estimated from TEM micrographs of the TPV blends. It was found that the PPs are mixed with oil in different proportions in different TPVs and that the viscosity curves of these mixtures exhibit the same trends in magnitude as the corresponding TPV viscosity curves. Hence, the shear flow of TPVs could be understood more readily in terms of the effective PP/oil medium flow behavior than in terms of the neat PP melt flow. Ellul also reported two important works on the plasticization of TPVs (56,57). She reported that since only the amorphous component of PP is plasticized, the crystalline fraction is not much affected and the upper service temperature range is maintained (56). The plasticized TPVs have an excellent balance of engineering


437

properties to 125 C and are much more elastic than unplasticized TPVs due to the suppression of yielding behavior. In another report (57), she found that the use of a PP phase with a high degree of long-chain branching can improve the elasticity of the TPVs with the condition that the branching index at molecular weight greater than 1,000,000 should be less than about 0.6. It was postulated that in the melt and at low frequencies, the long-chain branched PP behaves as a network. In the melt, the dynamically vulcanized alloy behaves as a dual network material: one network being the chemically cross-linked rubber phase and the other being the physical network arising from the high level of long-chain branching in PP. In the solid state, the cocontinuous morphology arising from the choice of longchain branched PP contributes to the enhanced elasticity of the TPVs. PP/EPDM TPV for automotive applications include rack and pinion bellows, air ducts, underhood tubing and connector, underhood hose, plugs, bumpers, grommets, prop rod shaft cover, radiator air deflector, air bag door covers and skins, grips, seals, mats and cupholders, windshield wiper motor cover, fuel line hose, suspension bellows, and weatherseals. EPDM/PP TPV was also used for construction such as door sweep, expansion joints, and window and door seals. Electronic applications include scroll wheel, printer roller, scanner lid, grip, access panel door, speaker surrounds, and holders and bumpers. Also, TPVs are widely used in business machines, power tools, motor mounts, and many other fields. Because of their unique properties and widespread applications, PP/EPDM blends have also been recently subjected to thermoplastic elastomer nanocomposites. Different kinds of nanofiller have been used to prepare nanocomposites like nanoclay, spherical nanoparticles, carbon nanotube, among others (58–72). In addition, the development of environment-friendly polymers is one of recent important issues. In this sense, the use of recycled blends is also an important task. PP/EPDM blends are no exception. For example, Pfaendner et al. (73) studied mechanical recycling of thermoplastics for high value applications, which is directly associated with restabilization. They described the state of the art in processing, long-term heat and light stabilization of recyclates of PP, PP/EPDM blends with examples from packaging, distribution, and automobile and construction industry. They found that because of predamage and impurities, recyclates degrade faster and differently compared with virgin polymers, and therefore specially designed stabilizer systems are required according to previous damage and subsequent application.

14.5

CONCLUSIONS

Thermoplastic elastomeric materials have many important applications including cable and wire especially in mineral, electronic equipment, and automobile industries. The most commonly used method of obtaining thermoplastic elastomer in materials is to toughen plastics by blending rubbers and plastics. Among the most versatile polymer matrices, polyolefins such as PP are the most widely used thermoplastics because of their well-balanced physical and mechanical properties and their easy processability at a relatively low cost, which makes them a versatile material. PP has the disadvantage of

438

Polyolefin Blends

becoming brittle at low temperature, however, because of its high transition temperature and high crystallinity. The best way to improve its impact strength is to blend PP with elastomers particularly EPDM. The blend of PP and EPDM has been prepared by different means, melt mixing, dynamic radiation curing, ultrasonic curing, among others. These PP/EPDM blends have been widely studied from different angles; that is, structure–properties relationship, morphology, mechanical properties, rheology, thermal properties, among others. In this chapter, therefore, we reviewed recent reports on the PP/EPDM blends in terms of thermoplastic polyolefins (TPO) and thermoplastic vulcanizates (TPVs) as commercially important products as well as environmentfriendly use of recycled EPDM or PP for the PP/EPDM blends. The advances made in the plasticization of PP by low Tg diluents through the TPO and/or TPVs, thus changing its low temperature brittleness characteristics, are also worth noting (74). The market of PP/EPDM blends has grown dramatically because of its recycling ability and processability by conventional thermoplastic processing equipment. The unique characteristics of thermoplastic elastomer have made them an attractive alternative to conventional elastomers in a variety of markets. Thus, we introduced some recent practical applications of PP/EPDM blends.

ACKNOWLEDGMENTS The work was supported by the Korea Science and Engineering Foundation (KOSEF) through the National Research Laboratory Program funded by the Ministry of Science and Technology (MOST) (No. M10300000369-06J0000-36910), the SRC/ERC of MOST/KOSEF program (grant #R11-2000-070-080020), and the Brain Korea 21 Project.

NOMENCLATURE BDT DCP DMA DMI DMTA DSC DV EPDM IDc IPN iPP ISGV MAH MMI POM PP PP-g-MAH

Brittle–ductile transition Dicumyl peroxide Dynamic mechanical analysis Dimethyl itaconate Dynamic mechanical thermal analyzer Differential scanning calorimetry Dynamic vulcanization Ethylene-propylene-diene terpolymer Critical interparticle distance Interpenetrating polymer network Isotactic polypropylene In situ grafting and dynamic vulcanization Maleic anhydride Monomethyl itaconate Polarized optical microscopy Polypropylene Maleic anhydride grafted PP


R RMS SEM TBT TBTM Tg TEM TPO TPV Zn-SEPDM

439

Tip radius of a notch for impact strength measurement Rheometric mechanical spectrometry Scanning electron microscopy Brittle–ductile transition temperature TBT of PP itself for a given notch Glass transition temperature Transmission electron microscopy Thermoplastic polyolefin Thermoplastic vulcanizate Zn2þ neutralized sulfonated EPDM

REFERENCES 1. E. N. Kresge, Rubber Chem. Technol., 64, 469 (1991). 2. M. T. Payne and C. P. Rader, Thermoplastic elastomers: A rising star, in Elastomer Technology Handbook, N. P. Cheremisinoff (ed.), CRC Press, New York, 1993, pp. 571–582. 3. A. Y. Coran and R. P. Patel, Thermoplastic elastomers based on dynamically vulcanized elastomer— Thermoplastic blends, in Thermoplastic Elastomer, G. Holden, N. R. Legge, R. Quirk, and H. E. Schroeder (eds.), Hanser, New York, 1996, p. 153, Chapter 7. 4. A. L. N. Da Silva and M. B. Coutinho, Polym. Test., 15, 45 (1996). 5. L. Huang, Q. Pei, Q. Yuan, H. Li, and W. Jiang, Polymer, 44, 3125 (2003). 6. N. Da Silva and E. M. B. Coutinho, J. Appl. Polym. Sci., 81, 3530 (2001). 7. M. A. Lopez-Manchado, J. M. Kenny, and M. Y. Pedram, Macromol. Chem. Phys., 202, 1909 (2001). 8. Z. Su, P. Jiang, Q. Li, P. Wei, G. Wang, and Y. Zhang, J. Appl. Polym. Sci., 94, 1504 (2004). 9. C. S. Ha, Y. W. Cho, W. J. Cho, Y. Kim, and T. Inoue, Polym. Eng. Sci., 40, 1816 (2000). 10. F. H. Sanchez, R. Olayo, and A. Manzur, Polym. Bull., 42, 481 (1999). 11. C. S. Ha and S. C. Kim, J. Appl. Polym. Sci., 35, 2211 (1988). 12. C. S. Ha, J. Kor. Inst. Rubber Ind., 25, 203 (1990). 13. T. Zaharescu, R. Setnescu, S. Jipa, and T. Setnescu, J. Appl. Polym. Sci., 77, 982 (2000). 14. T. Zaharescu, M. Chipara, and M. Postolache, Polym. Deg. Stab., 66, 5 (1999). 15. T. Zaharescu and P. Budrugeac, Polym. Bull., 49, 297 (2002). 16. Y. Chen and H. Li, Polym. Eng. Sci., 44, 1509 (2004). 17. T. Zaharescu, Polym. Deg. Stab., 73, 113 (2001). 18. W. K. Fischer, US Patent 3,758,643 (1973). 19. W. K. Fischer, US Patent 3,862,106 (1975). 20. A. Y. Coran and R. Patel, Rubber Chem. Technol., 54, 892 (1981). 21. A. Y. Coran and R. Patel, Rubber Chem. Technol., 56, 210 (1983). 22. L. A. Coettler, J. R. Richwine, and F. J. Wille, Rubber Chem. Technol., 55, 1448 (1982). 23. M. Ishikawa, M. Sugimoto, and T. Inoue, J. Appl. Polym. Sci., 62, 1495 (1996). 24. C. S. Ha, D. J. Ihm, and S. C. Kim, J. Appl. Polym. Sci., 32, 6281 (1986). 25. L. Tang, B. Qu, and X. J. Shen, J. Appl. Polym. Sci., 92, 3371 (2004). 26. A. K. Jain, N. K. Gupta, and A. K. Nagraz, J. Appl. Polym. Sci., 77, 1488 (2000). 27. A. Sengupta and B. B. Konar, J. Appl. Polym. Sci., 66, 1231 (1997). 28. M. D. Ellul, J. Patel, and A. J. Tinker, Rubber Chem. Technol., 68(4), 583 (1995). 29. M. D. Ellul, A. H. Tsou, and W. G. Hu, Polymer, 45(10), 3351 (2004).

440

Polyolefin Blends

30. S. Abdou-Sabet, R. Puydak, and C. Rader, Rubber Chem. Technol., 69, 476 (1996). 31. F. A. Goharpey, A. Katbab, and H. Nazockdast, J. Appl. Polym. Sci., 81, 2531 (2001). 32. H. Nazockdast, F. Goharpey, and A. Katbab, Rubber Chem. Technol., 76, 239 (2003). 33. Y. Yang, T. Chiba, H. Saito, and T. Inoue, Polymer, 39, 3365 (1998). 34. E. Martsuscelli, C. Silvestre, and G. Abate, Polymer, 23, 229 (1982). 35. Y. Kim, W. J. Cho, C. S Ha, and W. H. Kim, Polym. Eng. Sci., 35, 1592 (1996). 36. S. C. Kim, D. Klempner, K. S. Frisch, W. Radigan, and H. L. Frisch, Macromolecules, 9, 258 (1976). 37. Y. Kim, W. J. Cho, and C. S. Ha, Polymer(Korea), 18, 737 (1994). 38. Y. Kim, W. J. Cho, and C. S. Ha, J. Mater. Sci., 31, 2917 (1996). 39. N. K. Gupta, A. K. Jain, R. Singhlal, and A. K. Nagpal, J. Appl. Polym. Sci., 78, 2104 (2000). 40. A. K. Jain, A. K. Nagpal, R. Singhal, and N. K. Gupta, J. Appl. Polym. Sci., 78, 2089 (2000). 41. T. Asami and K. Nitta, Polymer, 45, 5301 (2004). 42. C. S. Ha and S. C. Kim, J. Appl. Polym. Sci., 37, 317 (1989). 43. Y. Kikuchi, T. Fukui, T. Okada, and T. Inoue, Polym. Eng. Sci., 31(14), 1029 (1991). 44. M. C. Boyce, K. Kear, S. Socrate, and K. Shaw, J. Mech. Phys. Sol., 49 (95), 1073 (2001). 45. M. C. Boyce, S. Socrate, K. Kear, O. Yeh, and K. Shaw, J. Mech. Phys. Sol., 49(6), 1323 (2001). 46. M. C. Boyce, O. Yeh, S. Socrate, K. Kear, and K. Shaw, J. Mech. Phys. Sol., 49(6), 1343 (2001). 47. J. Oderkerk, G. Groeninckx, and M. Soliman, Macromolecules, 35(10), 2002. 48. L. Tang, B. Qu, and X. Shen, J. Appl. Polym. Sci., 92, 3371 (2004). 49. F. Goharpey, N. Nazockdast, and A. Katbab, Polym. Eng. Sci., 201, 84 (2005). 50. H. W. Xiao, S. Q. Huang, T. Jiang, and S. Y. Cheng, J. Appl. Polym. Sci., 83, 315 (2002). 51. C. S. Ha, Y. Cho, J. H. Go, and W. J. Cho, J. Appl. Polym. Sci., 77, 2777 (2000). 52. M. D. Wang and L. Cakmak, Rubber Chem. Technol., 74, 761 (2001). 53. http//www.sperecycling.org/PDF%20Files/0157.PDF. 54. M. Montoya, J. P. Tomba, J. M. Carella, and M. I. Gobernado-Mitric, Eur. Polym. J., 40, 2757 (2004). 55. K. Jayaraman, V. G. Kolli, S. Y. Kang, S. Kumar, and M. D. Ellul, J. Appl. Polym. Sci., 93(1), 113 (2004). 56. M. D. Ellul, Rubber Chem. Technol., 71(2), 244 (1998). 57. M. D. Ellul, Rubber Chem. Technol., 76(1), 202 (2003). 58. E. P. Giannelis, Adv. Mater., 8, 29 (1996). 59. S. S. Ray and M. Okamoto, Prog. Polym. Sci., 28, 1539 (2003). 60. X. Li, T. K. Kang, W. J. Cho, J. K. Lee, and C. S. Ha, Macromol. Rapid Commun., 22, 1306 (2001). 61. S. R. Lee, H. M. Park, H. T. Lim, T. K. Kang, X. Li, W. J. Cho, and C. S. Ha Polymer, 43, 2495 (2002). 62. H. M. Park, X. Li, C. Z. Jin, C. Y. Park, W. J. Cho, and C. S. Ha, Macromol. Mater. Eng., 287, 553 (2002). 63. J. K. Mishra, I. Kim, and C. S. Ha, Rubber Chem. Technol., 78, 42 (2005). 64. G. J. Hwang, J. W. Park, I. Kim, C. S. Ha, and G. H. Kim, Macromol. Res., 14, 179 (2006). 65. H. M. Jeong, M. Y. Choi, and Y. T. Ahn, Macromol. Res., 14, 312 (2006). 66. S. M. Garces, D. J. Moll, T. Bicerano, R. Fibiger, and D. G. McLeod, Adv. Mater., 12, 1835 (2000). 67. K. Y. Lee and L. A. Goettler, Polym. Eng. Sci., 44, 1103 (2004). 68. J. K. Mishra, I. Kim, G. H. Kim, and C. S. Ha, J. Polym. Sci. Part B Polym. Phys., 42, 2900 (2004). 69. S. Mehta, F. M. Mirabella, K. Rufener, and A. Befna, J. Appl. Polym. Sci., 92, 928 (2004). 70. B. B. Khatua, D. J. Lee, H. Y. Kim, and J. K. Kim, Macromolecules, 37, 2454 (2004). 71. J. K. Mishra, I. Kim, and C. S. Ha, Polymer, 46, 1995 (2005). 72. L. Valentine, J. M. Kenny, and M. A. Lopez-Manchado, J. Appl. Polym. Sci., 89, 2657 (2003). 73. R. Pfaendner, H. Herbst, K. Haffmann, and T. Sitek, Angew. Makromol. Chem., 232, 4140 (1995). 74. M. D. Ellul, Plastic Rubber Compos. Process. Appl., 26(3), 137 (1997).

Chapter

15

Ethylene–Propylene–Diene Rubber/Natural Rubber Blends Soney C. George1 and Sabu Thomas2

15.1 INTRODUCTION The scientific and commercial progress in the area of polymer blends during the past decades has been tremendous and was driven by the realization that, by blending, new material can be developed and can be implemented more rapidly and economically. Blending of polymers is technological way of providing materials with full set of desired specific properties at the lowest price. Blending also benefits the manufacturer by offering improved processability, product uniformity, quick formulation changes, plant flexibility, and high productivity. Many elastomers that are dissimilar in chemical structure are blended to improve processability, performance, durability, and physical properties. The elastomer blends generally exhibit poor mechanical properties due to incompatibility and gross phase separation (1,2). In addition to the poor interfacial adhesion caused by the thermodynamic incompatibility, these blends usually present cure rate incompatibility because of the differences between the reactivity of the elastomers with the curing agents and/or differences in solubility of the curatives in each elastomer phase (3,4). Extensive research work has been carried out in the field of elastomer blends for the past few decades (5–10). NR/EPDM blends are one of the elastomer blends, which gained lot of commercial interest during these days due to their excellent properties. 1 Department of Basic Science, Amal Jyothi College of Engineering, Koovapally, Kottayam 686518, Kerala, India 2 School of Chemical Sciences, Mahatma Gandhi University, Priyadarshini Hills, Kottayam 686560, Kerala, India


441

442

Polyolefin Blends

Natural rubber vulcanizates exhibit good elasticity and good tensile strength besides high resilience and excellent wear resistance. On the contrary, natural rubber is not recommended for outdoor applications where maximum resistance to sunlight, ozone, oxygen, or heat aging are major factors. EPDM exhibits excellent resistance to ozone, oxidants, and severe weather conditions, thereby making it an outstanding material for outdoor applications. But the resilience and tensile properties of EPDM are lower than that of natural rubber. Blending a suitable amount of low unsaturated ethylene–propylene–terpolymer (EPDM) into a diene rubber has been found to improve both heat and ozone resistance (11–15) besides its improvement in chemical resistance, mechanical properties, and building tack properties along with decrease in compression set (16). Apart from property benefits, the blends of EPDM and NR are also attractive from an economic point of view. This is due to the relatively high expense of EPDM. However, the difference in olefin concentration of EPDM and natural rubber results in a cure-rate misbatch leading to an incompatible blend. This has been recognized to cause both inferior static and dynamic mechanical properties such as poor tensile strength, fatigue resistance, and high hysteresis in the rubber blend (17). The chapter will review the preparation, properties, and application of NR– EPDM blends. Several issues related to cure-rate mismatch, compatibility problems, morphology control by introduction of compatibilizers will be discussed.

15.2 MISCIBILITY, COMPATIBILITY, AND THERMODYNAMICS OF POLYMER BLENDING Elastomers with similar polarities and solubility characteristics can be easily combined to produce miscible polyblend (18). Miscible polymer blend is a polymer blend, which is homogeneous down to the molecular level and associated with the negative value of the free energy of mixing and the domain size is comparable to the dimensions of the macromolecular statistical segment. Complete miscibility in a mixture of two polymers requires that the following condition be fulfilled (19): DGm ¼ DHm TDSm < 0

ð15:1Þ

where DGm , DHm , and DSm are the Gibb’s free energy, the enthalpy and entropy of mixing at temperature T, respectively. The value of TDSm is always positive since there is an increase in the entropy on mixing. Therefore, the sign of DGm always depends on the value of the enthalpy of mixing DHm . The polymer pairs mix to form a single phase only if the entropic contribution to free energy exceeds the enthalpic contribution, that is, DHm < TDSm

ð15:2Þ

The miscibility may be achieved through specific interaction, for example, (i) repulsive, (ii) dipole–dipole, (iii) ion–dipole, (iv) ion–ion, hydrogen bonding, and (v) chemical reaction between active blend constituents (20). The commercially useful polymer–polymer combination is linked by intermolecular forces such as

Chapter 15 Ethylene–Propylene–Diene Rubber/Natural Rubber Blends

443

van der Waals forces or dipole moments and exhibit sufficient thermodynamic compatibility to prevent the polymer phases from separating during melt processing (21). Blends of elastomers having similar polarity and cure rate exhibit almost additive properties, whereas dissimilar elastomers result in blends with inferior properties (22). This property failure has been ascribed mainly to three types of incompatibilities that exist between dissimilar elastomers: (1) thermodynamic incompatibility involving phase separation on molecular scale (23,24), (2) viscosity mismatch causing delay or even preventing the formation of coherent blends (25), and (3) cure-rate mismatch due to imbalance in unsaturation levels of the elastomers. Among these, viscosity mismatch can be improved through proper blending processes by adjusting the raw polymer viscosities, extender oil, and filler concentrations. Thermodynamic incompatibility can be alleviated to some extent by reducing the interfacial energy through the creation of microdomains and subsequent adhesion between the elastomeric phases or by cross-linking the phases across the interfaces during vulcanization (26). Several attempts have been made to minimize phase separation and to increase interfacial adhesion. The addition of a third component, that is, a compatibilizer, which is able to act at the interface between the phases is also a good approach to obtain polymer blends with a more homogeneous morphology and improved mechanical properties (27). The compatibilization of elastomer blends may be successfully performed by using a functionalized polymer as a reactive compatibilizer (28–30).

15.3

BLEND PREPARATION

The basic materials used for the preparation of blends are EPDM and natural rubber. The structure and general characteristics of EPDM and NR are given in Fig. 15.1 (31) and Table 15.1. In general, the masticated NR and EPDM were mixed together with other ingredients including compatibilizers and homogenizing agents in an internal mixer or open mills. The rubber compound is cured in an electrically heated press at 160 C for optimum cure time, which is determined by rheometer. In order to study the

Figure 15.1 Molecular structure of (a) ethylene–propylene rubber and (b) natural rubber. (From Reference 31 with permission from Elsevier Science Ltd.)

444

Polyolefin Blends Table 15.1 General Characteristics of EPDM and Natural Rubber. Properties Tensile strength,ps: Tg C Specific gravity Tear resistance Adhesion to metal Abrasion resistance Compression set Rebound cold Rebound hot Permeability to gases Dielectric strength Electrical insulation Acid resistance Heat resistance Sunlight resistance Ozone resistance

EPDM

Natural rubber

Over 3000 0.58 0.85 Good Good Good Good Good Good Poor Excellent Good Good Excellent Excellent Excellent

Over 3000 0.75 0.93 Good Excellent Excellent Good Excellent Excellent Fair Excellent Excellent Fair Good Poor Fair

cure-rate mismatch the vulcanizates are reported to be prepared in two different methods (32–33). These include one-stage vulcanization and two-stage vulcanization processes. In the one-stage vulcanization process, NR and EPDM are first masticated separately and then mixed with each other. Additives such as ZnO, stearic acid, carbon black (or silica), and process oil are added. The mix thus obtained is allowed to cool to room temperature. Finally, coupling agent known as DIPDIS and sulfur are added to the mix on the cooled mill. The stocks are finally cured under pressure at 160 C (32–33). In the two-stage process, NR and EPDM are first masticated separately. Then, additives such as ZnO, stearic acid, DIPDIS, and sulfur are incorporated in the EPDM. The compounded EPDM mix is then heated at 160 C in the hydraulic press for the predetermined time to yield the grossly undercured mix. The undercured mix is then blended with NR to the required blend ratio. The blend compound is finally vulcanized to the optimum cure time values (32–33).

15.4 COVULCANIZATION Cure-rate mismatch is extreme when the blends constitute high unsaturated diene rubbers like NR, and low unsaturation rubbers like EPDM. It involves the migration of polar curatives from the low unsaturation phase to a more polar high unsaturation phase, further undercuring the low unsaturation phase (11–13,34) and it has been shown that unvulcanized EPDM exists in the vulcanized blend with NR (35). Several approaches have been made to obtain a cocured blend vulcanizate of NR–EPDM without sacrificing the physical properties by (1) increasing the unsaturation of EPDM elastomer so that the cure rate becomes at par with NR or other diene rubbers


445

(36–38); (2) curing with peroxide and polysulfide (38); (3) halogenating the rubber (39); (4) effecting partial prevulcanization (36,40); (5) using accelerators with long hydrocarbon chains (41,42); and (6) grafting accelerators or sulfur donors ( 43,44) or polydiene chains (45,46) in EPDM. Incorporation of lead dithiocarbamate into EPDM, before it is mixed with NBR, has been reported to yield an improved blend (11,13) vulcanizate. EPDM has been reported to react with N-chlorothioamides (47) to produce a macromolecular cure retarder, making it compatible with NR. EPDM modified with maleic anhydride (48) has been observed to produce blend vulcanizate of improved physical properties. The use of polyoctanomer (TOR) (27,49) and halobutyl rubber (50) as the compatibilizer for NR–EPDM has also been reported. It is also reported that proton NMR spectroscopy is a good tool in examining the curative migration and cross-link density within the NR phase of the NR/EPDM blends (51). The spectral line widths are found to increase smoothly with cross-link density. Analysis of this signal leads to a measure of line broadens H%. H% increases with the increase in cross-link density (52) and H% of any of the blends was found to be greater than the H% of the appropriate NR control vulcanizate. This analysis showed that NR components of the blends are utilizing 84–92% of the curatives in the compound, that is, about 80% of the curatives initially located in the EPDM phase are diffusing to the NR phase during vulcanization. Similarly, the swollen state FT– NMR spectroscopic method of blend analysis (53) has been reported to be effective in analyzing the cross-linking density in NR/EPDM blends. This study shows that the presence of chemical modification in the EPDM has a dramatic effect on the crosslinking in the EPDM phase, but only a minor on that in the NR phase. But the overall cross-link density in the blend is increased. Despite these changes, there remains a large imbalance in the cross-link distribution in favor of the NR phase in both modified blends, yet the physical properties are good. Ghosh et al. (32) had developed a new vulcanization technique to mitigate the cure-rate mismatch of NR/EPDM blends by introducing a multifunctional additive, namely, DIPDIS. In order to control the curative diffusion from the nonpolar to polar elastomer, EPDM has made more polar through its reaction with DIPDIS. Although EPDM used in the investigation has a low unsaturation content (5%), it will react with DIPDIS and yield rubber-bound intermediates as shown in Fig. 15.2 (32). It was

Figure 15.2 Reactions of bis(diisopropyl)-thiophosphoryl disulfide (DIPDIS) with ethylene– propylene–diene rubber (EPDM) and zinc oxide. (From Reference 32 with permission from John Wiley & Sons.)

446

Polyolefin Blends

reported that DIPDIS is capable of raising the maximum rheometric torque values of both NR and EPDM as well as the physical properties of the blend vulcanizate. The properties are further enhanced by the two-stage vulcanization. The grossly undercured material obtained in the first stage favorably contains a higher amount of reactive fragments compared with that in one-stage vulcanization. The intermediates thus formed are expected to combine with NR in the second stage of the procedure, the resultant effect being the generation of more interrubber linkages and formation of novel rubber blends of significantly improved physical properties. There is significant improvement in modulus, tensile strength, elongation at break, and cross-linking value over those obtained in one-stage vulcanization of the corresponding blends.There is more coherency and homogeneity in the blend composition of two-stage vulcanizates. The cure–rate mismatch problem could thus be solved through the formation of rubber bound intermediates with a DIPDIS, thereby restricting the curative migration from EPDM to NR. The blend morphology as revealed by SEM studies (Fig. 15.3) accounts for significant improvement in physical properties, particularly in two-stage vulcanizates.

Figure 15.3 Scanning electron micrographs of tensile fractured surfaces of the vulcanizates cured at 160 C; (a) 75:25 unsaturated natural rubber–ethylene–propylene–diene (NR–EPDM) blend (one-stage) at 500 ; (b) 75:25 NR–EPDM blend (two-stage) at 500 ; (c) 50:50 NR–EPDM blend (one-stage) at 750 ; (d) 50:50 NR–EPDM blend (two-stage) at 750 . (From Reference 32 with permission from John Wiley & Sons.)


15.5

447

FILLER DISTRIBUTION IN NR/EPDM BLENDS

The distribution of filler is a major problem in NR/EPDM blends due to its polarity difference. The effect of carbon black and silica on the physicomechanical properties of the covulcanized NR–EPDM rubber blends has been reported (33). The blends rich in NR content exhibit comparatively better results due to inefficient cocure by curative migration and filler transfer (33) as well as the lower unsaturation of EPDM compared with the high unsaturation of NR. All these factors lead to further weakening of the already weakly reinforced EPDM phase by the dearth of polar curatives. Evidently, vulcanizates with poor physicochemical properties are being produced. As discussed earlier, in two-stage vulcanization EPDM is more polar due to the pendant moieties of DIPDIS (Fig. 15.2). Incorporation of carbon black into that modified EPDM matrix restricts the transfer of filler to the NR phase. In all the cases, significant improvement in tensile strength and elongation at break are observed, compared with one-stage vulcanization; however, modulus and cross-linking values show an opposite trend because of the longer cure time used for one-stage samples. Though the cross-linking density is more in one-stage vulcanizates, the properties such as tensile strength and elongation at break values decrease due to the absence of interfacial cross-linking between the components of the blend. The blend vulcanizates exhibit highest modulus, tensile strength, elongation at break, and cross-linking values, and least weight loss in the swelling experiment. These better properties are as a result of the curative fixation on the EPDM backbone. This fixation, in fact, lowers filler transfer from EPDM to NR. Silica-filled samples exhibit faster cure rate compared with gum (32) and black-filled vulcanizates of similar composition. This difference is due to the reaction between the fragments of DIPDIS and the silanol (Si-OH) groups present at the silica surfaces (Fig. 15.4). The isopropyl moiety (-OR) in DIPDIS reacts with the silanol group of silica through the elimination of isopropyl alcohol (54) facilitating the rubber–filler interaction and thus resulting in faster curing. It is evident that inefficient interfacial cross-linking between NR and EPDM occurs in the case of silica-filled systems subjected to one-stage

Figure 15.4 Reaction of silica with pendant DIPDIS fragments. (From Reference 33 with permission from John Wiley & Sons.)

448

Polyolefin Blends

vulcanization, where both curative and filler migration from EPDM to NR phase occurs during vulcanization. Distribution of silica in the NR matrix is facilitated by the interaction of the protein component of NR with the hydroxylated surface of the precipitated silica. As the mixing progresses, the viscosity of the NR phase is reduced gradually. Thus, highly polar silica opts for its migration from high viscosity EPDM to low viscosity NR phase. In the early stage of vulcanization, pendant DIPDIS fragments might react with silica to form large EPDM–DIPDIS (fragment)–SiO2 aggregates, which in turn facilitates the reaction with NR. Improvement in the physical properties of the vulcanizates is thus the outcome of the interfacial cross-linking. Addition of silica in the modified EPDM makes the latter substantially polar and in this way restricts the filler transfer to the NR phase. All these facts lend support to the formation of coherent and homogeneous blend systems of practical importance. The SEM micrographs indicate interfacial chemical bridges in the blend vulcanizate and thus corroborates the results obtained.

15.6 MORPHOLOGY OF NR/EPDM BLENDS The morphology and properties of polymer alloys and blends can be controlled via phase separation or phase dissolution during cure (55). Phase dissolution in elastomer blending is the key factor for achieving optimal morphology having a synergistic effect in the polymer blends. The morphology and properties of NR/EPDM blends could be controlled through compatibilization and covulcanization to promote phase dissolution during cure. The morphology of NR/EPDM blends depends upon the type of compatibilizer added. Hsich reported (55) that a synergistic result in properties of blends was achieved by having interconnected phase morphology. The domain sizes of the blends in this study are 3–5 mm before cure and 1–2 mm after cure. These blends showed superior mechanical properties than those of the individual NR or EPDM, after a long term of high temperature aging. Figure 15.5a and b represents the phase morphology of NR/EPDM blends of 70/30 wt% without compatibilizer and with 2.5% EPDMSH compatibilizer. Figure 15.5c and d represents NR/EPDM blends 60/40 wt% without and with compatibilizer 2.5% EPDMSH (56). The light region corresponds to the NR phase, and the dark points are related to the unstained EPDM phase. In the case of NR/EPDM (60:40 wt%) blend, the presence of EPDMSH resulted in a more homogeneous morphology, but the differences are marginal. On the contrary, the addition of TOR reduced the domain size (27) of the dispersed EPDM particles and it becomes uniform and spherical (Fig. 15.6 b) compared to the uncompatibilized, that is, in the absence of TOR, the dispersed particles are generally large, the size distribution is broad, and the shape of particles is very irregular (Fig. 15.6a). The irregular large particles over 20 mm coexist with spherical small particles below 1 mm. The influence of various compatibilizers on the morphology of 50/50 NR/EPDM blends is shown in Fig. 15.7 (57).


449

Figure 15.5 SEM micrographs of NR/EPDM (a and b) 70/30 wt% and (c and d) 60/40 wt% blends. Micrographs (a) and (b) are related to compatibilized blends with 2.5 phr of EPDMSH. (From Reference 56 with permission from John Wiley & Sons.)

Figure 15.7a shows that the two phases are with irregular domain sizes and shapes. This indicates that the NR/EPDM blends were completely immiscible, large EPDM domains being dispersed in the NR matrix. The average domain size of the dispersed phase was 4.1 mm. The compatibility of the NR/EPDM system was improved by the addition of a compatibilizer, as can be seen in Fig. 15.7b–g; the treatment resulted in noticeable surface hardening, and the physical changes in the surface were expected to influence physically both the deformation and adhesion of the two rubbers, that is, the compatibilizers improved both the morphology and compatibility of the blends because of the reduction in the interfacial tension between EPDM and NR rubbers. The size of the dispersed phase (EPDM) domain decreased with the addition of compatibilizers, and no gross phase separation was present in the blends (Fig. 15.7). For NR/BR/EPDM, the domain size was approximately 3.8–1.26 mm; NR/PVC/EPDM, 2.7–0.75 mm; NR/chlorosulfonated PE/EPDM, 2–0.75 mm; NR/g-radiation/EPDM 4–1.5 mm; and NR/MAH/EPDM.1– 0.25 mm. These results are in agreement with the observations of Anastasiadas and Koberstein (58) and Meier (59), who reported that compatibilizers reduced the phase domain size.

450

Polyolefin Blends

Figure 15.6 SEM micrographs of the NR/EPDM/TOR blends: (a) 70/30/0; (b) 70/30/10. (From Reference 27 with permission from John Wiley & Sons.)

15.7 COMPATIBILIZATION OF NR/EPDM BLENDS The properties of NR/EPDM blends are not very good due to their incompatibility (60). Therefore, EPDM was modified by several modifying agents such as, TOR (27), EPDMTA, and EPDMSH (56,61), g-rays, EPDM-g-MAH, polybutadiene rubber (BR), chlorinated rubber, chlorosulfonated polyethylene (SPE), and poly(vinyl chloride) (PVC), (57,62) in order to make it more compatible with natural rubber. The effect of concentration of trans-polyoctene rubber as a compatibilizer in various properties of NR/EPDM was investigated (27). The blend composition was fixed at NR/EPDM 70/30 with varying concentration of compatibilizer as 0,5, 10,15, and 20. As TOR having much lower viscosity compared to NR and EPDM, it will locate at the interface of NR and EPDM phase. So the TOR at the interface region reduces the interfacial tension between the incompatible rubbers, which will facilitate the mixing of the rubber blends, thereby improvement in compatibility. The morphology of the compatibilized and uncompatibilized blends (NR/EPDM: 70/30) shown in Fig. 15.6 is in agreement with this observation. Compatibilization is more effective in this case


451

Figure 15.7 SEM micrographs: (a) 50/50 NR/EPDM without a compatibilizer, (b) NR/BR/EPDM, (c) NR/PVC/EPDM, (d) NR/chlorinated rubber/EPDM, (e) NR/chlorosulfanted PE/EPDM, (f) NR/gradiation/EPDM, and (g) NR/MAH/EPDM. (From Reference 57 with permission from John Wiley & Sons.)

452

Polyolefin Blends

compared to the use of EPDMSH (56). By using EPDMSH for compatibilization purpose, more homogeneous morphology was observed. The tensile modulus of the blends is increased gradually, with increasing TOR content; however, tensile strength and elongation at break of the blends are decreased in the presence of TOR. In the case of NR/EPDM blends with 2.5% EPDMSH compatibilizer, tensile strength, and elongation at break are improving in all compositions but in 70/30 composition blends with 5% of TOR having higher tensile strength. The effect of mercaptoand thioacetate-modified EPDM (61) on the curing parameters and mechanical properties of natural rubber/EPDM blends were reported. In this study, 2,20 dithiobisbenzothiazole (MBTS) was employed as a conventional accelerator, whose proportion in the rubber formulation has been varied, in order to observe the performance of these functionalized copolymers, not only as a compatibilizing agent but also as a secondary accelerator (61). Both EPDMTA and EPDMSH resulted in increase of the ultimate tensile strength of the vulcanized blends. Nevertheless, the efficiency of EPDMSH was superior, which was attributed to the combination of several factors: (a) increase of the cross-link density, (b) the vulcanization of some fraction of the EPDM phase, thus promoting the covulcanization in some extent, and (c) the reactive compatibilization, as a consequence of chemical reactions between the mercapto groups and the rubber matrix. EPDMTA was not so effective in improving the mechanical performance or cross-link density of the blends because of the lower reactivity of the thioacetate groups. However, it was efficient in increasing the aging resistance of the corresponding blends. g Rays at radiation doses of 6 and 8 k gray were found to be suitable for creating cross-links necessary for the mixing homogeneity of NR and EPDM (57). Similarly, great improvement in the homogeneity of the blends was achieved by the use of 10 phr of compatibilizers like EPDM-g-MAH, polybutadiene rubber, chlorinated rubber, chlorosulfonated polyethylene and poly(vinyl chloride) (57). This is also evident from the morphology of the compatibilized blends shown in Fig. 15.7a–g. EPDM was modified with maleic anhydride and blended with natural rubber. A concentration of 10 phr of MA-g-EPDM improved the compatibility of NR/EPDM blends and thereby led to finer morphology (63). The degree of compatibility of NR/ EPDM blend was improved by adding homopolymers and copolymers of acrylonitrile and N-(4-chlorophenyl) acrylamide (64). It was found that a fairly good compatibility had been achieved by using polyacrylonitrile. NR/EPDM compatibility was improved by blending EPDM with epoxidized natural rubber (65).

15.8 MECHANICAL AND VISCOELASTIC PROPERTIES 15.8.1 Mechanical Properties 15.8.1.1

Effect of Blend Ratio

Blend composition has a strong influence on the mechanical properties of NR/EPDM blends (66). Figure 15.8 shows that tensile strength; stress at 100% and yield strain is increasing with the increase in the concentration of EPDM in the blend. This is attributed to the reinforcing effect of EPDM domains in the NR matrix. The


453

Figure 15.8 The effect of blend composition on the tensile and yield strain of NR/EPDM blends. (From Reference 66 with permission from Sage Publications Ltd.)

properties such as fatigue life and Young’s modulus increased while strain energy and cross-link density decreased (Table 15.2). 15.8.1.2

Effect of Compatibilization

Compatibilizers have a key role in improving the compatibility between the blend components and thereby enhancing the mechanical properties. Several reports are available in connection with improving the properties of NR/EPDM blends by compatibilization. George et al. (67) has reported on the improvement of mechanical properties of NR/EPDM blends by precuring EPDM prior to blending. Blends of Table 15.2 Mechanical Properties of Uncompatibilized NR/EPDM Blends. NR/EPDM blend ratio Properties Stress at 100% strain, MPa Tensile strength, MPa Rupture strength, MPa Yield strain,% Rupture strain% Young’s modulus, N mm2 Equilibrium swelling,% Soluble fraction,% Cross-link density 104, g mol1 Strain energy, MJ m3 No. of fatigue cycles 102

100/0 0.3 2.75 2.73 334 345 0.81 304 3.4 1.18 1.28 32

75/25 0.33 3.48 3.45 405 363 0.97 420 3.8 0.629 1.05 40

[From Reference 66 with permission from Sage Publications Ltd.]

50/50 0.35 3.92 3.89 495 379 1.05 431 4.2 0.517 0.89 45

25/75 0.4 4.35 4.31 515 394 1.12 450 4.8 0.365 0.79 49

100/0 0.53 3.32 3.28 315 359 0.94 392 3.7 0.9 1.12 43

454

Polyolefin Blends

Table 15.3 Mechanical Properties of NR/EPDM Blends as a Function of Composition and Compatibilization s aB;MPa NR/EPDM 100/0 80/20 70/30 60/40 0/100

ebB;%

c

d

c

d

11.5 10.3 10.5 9.4 3.4

9.6 13.0 13.5 15.0 5.5

750 850 900 600 100

780 830 840 700 200

a

Ultimate tensile strength.

b

Elongation at break.

c

Blends without compatibilizer.

d

Blends with 2.5 phr of EPDMSH.

[From Reference 56 with permission from John Wiley & Sons.]

NR/modified EPDM, in which EPDM was modified by pendant sulfur, exhibited improved endurance to repeated stress over that of covulcanized EPDM–NR rubber blends (68). The effects of ethylene and diene contents in EPDM, blend ratio, dicumyl peroxide curing system on the physical properties, interfacial adhesion force, and dynamic crack growth were examined (69). As the ethylene and diene contents in EPDM increased, the physical properties, such as dynamic cut growth, adhesion to other component were also increased. The mechanical properties of the blends are compared to those of the pure components in Table 15.3 (56). The ultimate tensile strength of noncompatibilized blends is lower than that of pure NR, as expected since these blends are incompatible. The addition of 2.5 phr of EPDMSH resulted in an improvement of this property for NR/EPDM blends. The values found for the compatibilized blends were even higher than that observed for pure NR, indicating a synergistic behavior with the compatibilization. Concerning the elongation at break, the compatibilization did not affect substantially this property, except for the system containing 100% of EPDM, where a significant increase was observed. The improvement of the tensile strength of the blends with the addition of EPDMSH is attributed to the interfacial action of this component associated to an increase of the cross-link degree. Abou-Helal and El-Sabbagh (66) found that compatibilizers like EPDM-g-MAH, chlorinated rubber, polybutadiene rubber, chlorosulfonated polyethylene, and so on improved the mechanical properties of NR/EPDM. A regression analysis was employed to correlate the mechanical properties, fatigue life, and strain energy with respect to blend ratio. It was found to obey the formula Y ¼ A þ BX, where Y is the mechanical properties, fatigue life, or strain energy, X is the blend ratio, and A and B are material constants. In the presence of compatibilizers, equation becomes Y ¼ AX B , where Y represents mechanical properties, and X is the blend ratio, and A and B are material constants. The stress strain curves of 50/50 NR/EPDM blends without and with compatibilizers are shown in Fig. 15.9. The behavior goes from a polymer that almost


455

Figure 15.9 Stress–Strain curves for 50/50 NR–EPDM blend without and with different compatibilizers. (From Reference 66 with permission from Sage Publications Ltd.)

yields with maximum and present strain hardening to polymer that nearly yields and brakes at deformations giving a large stress for different compatibilizers. 15.8.1.3

Effect of Homogenizing Agent

The concentrations of Ultrablend 4000 homogenizing agent of 0, 3, 5, and 7 phr were added into NR/EPDM: 70:30 blends (70). It was found that the tensile strength of blends was increased when a small amount of Ultrablend 4000 (3 phr) was added. It can be seen that tensile strength increases in the blends containing 3 and 5 phr Ultrablend 4000. However, this decreased when more Ultrablend 4000 (7 phr) was added. This could be explained in terms of the compatibility gained by the addition of Ultrablend 4000, which improves the compatibility between the matrix phase (NR) and dispersed phase (EPDM). When the sample was stretched in the tester, the stress was transferred from the matrix phase (NR) to the dispersed phase (EPDM) through the homogenizing agent. The mechanical properties such as tensile strength and elongation at break were gradually improved when 3 and 5 phr of Ultrablend 4000 were added. On the contrary, when Ultrablend 4000 was further added to 7 phr, tensile strength was poor. This could be due to the agglomeration of the excess amount of Ultrablend 4000 to become another phase. This new phase induces slippage or weak points between the matrix and dispersed phases, yielding a lower tensile strength. The high value of tensile strength and elongation at break at 5 phr of homogenizing agent shows a good adhesion between the phases of NR and EPDM at this particular concentration.

456

Polyolefin Blends 40

Tear strength, N mm–1

Unaged 30 Aged 20

10

0 With Si 30 phr

With CB 3 phr

With Si 30+CB 3 phr

70:30 NR/EPDM blend with filters

Figure 15.10 Comparison of tear strength of the NR/EPDM blends containing carbon black (3%), silica (30%), and carbon black/silica (3:30). (From Reference 70 with the permission from John Wiley & Sons.)

15.8.1.4

Effects of Fillers

Addition of silica was reported to improve the mechanical properties, such as tear strength and hardness of NR/EPDM blends (Fig. 15.10) (70). Blends with 30 phr of silica show higher tear strength and hardness compared with blends with 3 phr of carbon black. Synergistic improvement in tear strength and hardness is observed in presence of both carbon black (3 phr) and silica (30 phr).When the amount of Ultrablend 4000 was kept constant at 5 phr, it was found that the silica particles absorbed a greater concentration of Ultrablend 4000 than did the carbon black particles, resulting from their higher surface area and thereby better properties.

15.8.2 Dynamic Mechanical Properties The dynamic mechanical properties of NR/EPDM/TOR (27) and NR/EPDM/ EPDMSH (56) are shown in Fig. 15.11a and b. In Fig. 15.11a, it is clear that by the addition of TOR, dynamic elastic modulus (E0 ) of a blend over the measured temperature range is increased whereas the hysteresis (tan d) is not increased. A single tan d peak is observed, despite the phase-separated structure, which is due to the similarity of the glass-transition temperature of the blend components. This indicates that TOR increases rigidity of blend vulcanizates without substantial change in heat buildup under dynamic stress. In Fig. 15.11b (A), noncompatible blend displays only transition at a temperature. This is because of the presence of noncross-linked rubber mainly constituted by EPDM phase due to its lower cure rate as compared to NR phase. Considering the higher chain mobility of the noncross-linked EPDM, the glass transition temperature is expected to be shifted toward lower temperatures when compared to the same vulcanized rubber. Therefore, the single tan d peak in this blend can be attributed to the vulcanized NR


457

Figure 15.11 (a) Dynamic storage moduli and tan delta for he NR/EPDM/TOR blends containing different concentrations of TOR. (b) Dynamic mechanical properties of vulcanized NR/EPDM (70:30 wt%) blends; (a) noncomptibilized blend and compatibilized with 2.5 phr of EPDMSH. (From References 27,56 with permission from John Wiley & Sons.)

phase together with the nonvulcanized EPDM phase whose transition occurs at similar temperature. But compatibilized blend exhibits a lower Tg value related to the NR phase and a lower damping (Fig. 15.11b (B)). These results are due to the effective compatibilization of this functionalized copolymer. The strong interactions between the mercapto groups along with the EPDMSH backbone and the NR phase decrease the mobility of this phase, giving rise to a decrease of the damping. The blend compatibilized with EPDMSH displays a second transition with low damping at temperature. The proportion of the vulcanized EPDM phase in

458

Polyolefin Blends

the NR/EPDM blend compatibilized with EPDMSH may be responsible for the second transition at higher temperature.

15.9 RHEOLOGICAL PROPERTIES The log–log plots of apparent shear stress versus apparent shear rate for STR5L/ EPDM and STR5L/BEPDM blends with various blend compositions are shown in Figs. 15.12 and 15.13, respectively (71). Flow curves of all the blends show reasonably straight lines, whose intercept K and slope n correspond to the power law equation (the Ostwald–de Waele equation) (72). Table 15.4 shows the power law index and the consistency of flow of STR5L/EPDM and STR5L/BEPDM blends. The values of nðn < 1Þ indicate the pseudoplastic nature of STR5L, EPDM, BEPDM, and their blends. Hence, the apparent viscosity of the two sets of blends decreased as the shear rate increased. It can also be seen that for the pure rubbers, EPDM had the lowest n value and STR5L has the highest n value. This accounts for the high pseudoplasticity, the highly shear thinning fluid in the modified BEPDM, and the more plug-like profile (73) consequently, the blends of STR5L/BEPDM tended to have a lower n value at a given blend composition, which increased with increasing levels of STR5L. It is observed that the modified EPDM by bromination reaction affects the

Figure 15.12 Effect of apparent shear rate on the apparent shear stress of STR5L/EPDM blends at various blend compositions. (From Reference 71 with permission from John Wiley & Sons.)


459

Figure 15.13 Effect of apparent shear rate on the apparent shear stress of STR5L/EPDM blends at various blend compositions. (From Reference 71 with permission from John Wiley & Sons.)

shear flow property. That is, at a given shear rate, a higher apparent shear stress of pure BEPDM compared to that of EPDM and STR5L was found. The highest shear viscosity of BEPDM was therefore observed at a given apparent shear rate. It indicates that the bromine substituents on the rubber main chain may increase the chain rigidity of the rubber, consequently increasing the ability to resist flow, whereas STR5L gave the lowest apparent shear viscosity because of its ease in molecular weight breakdown with mastication during sample preparation and with shear force during the capillary flow test. Generally, the true shear viscosity of a polymeric blend follows the log additive rule (74–76). For the miscible blends, rheological properties (e.g., viscosity and die swell) show a positive deviation from their additive values; whereas the immiscible blends give a negative deviation in rheological properties (76). Blends in all blend compositions were evaluated and found to be negative deviations relating to their additive values. It is therefore indicated that the blends of STR5L/EPDM and STR5L–BEPDM were the immiscible blends. It means that there is no specific interaction between the two components of both blends. This may be attributed to the dissimilar, low unsaturated structure of EPDM, and the polarity

Table 15.4 The Power Law Index (n) and the Consistency of Flow (K) for Various Blend Compositions. NR/EPDM blend 0/100 25/75 50/50 75/25 100/0

n

K, KPa

NR/EPDM blend

n

K, KPa

0.14 0.15 0.20 0.21 0.22

293.0 149.9 88.7 85.1 86.8

0/100 25/75 50/50 75/25 100/0

0.10 0.14 0.16 0.20 0.22

444.6 169.0 124.5 81.8 86.8

n and K are power law index and consistency constant, respectively. [From Reference 71 with permission from John Wiley & Sons.]

460

Polyolefin Blends

of the bromine substituents on the BEPDM chains compared to the unsaturated nonpolar structure of natural rubber. In presence of homogenizing agent (Ultrablend 4000), the shear rate and shear viscosity are measured and plotted against content of homogenizing agent (70). A positive deviation of the blends was observed only at 5 phr of the homogenizing agent and which confirmed the blend compatibility of NR and EPDM. Pseudoplastic behavior was observed in the flow of blends. That is, the apparent shear viscosity decreased with an increase in the apparent shear rate. Therefore, the shear stress produced became smaller when the rate of shear increased. This flow behavior indicates a pseudoplastic fluid or shear thinning behavior of the blends.

15.10 THERMAL PROPERTIES The thermal characteristics of NR, EPDM, and their blend (50/50 NR/EPDM) have been examined with a DSC technique from 100 to 50 C (57,62). The glasstransition temperatures (Tg s) of NR, EPDM, and their blend are examined and are given in Table 15.5. It is found that the 50/50 (w/w) NR/EPDM blend having two distinct glass transitions; the lower glass transition was due to the EPDM phase and the higher glass transition was due to the NR phase. The large melting endoderm was attributable to the high crystallinity of EPDM. It is found that the specific heat capacity depends on the type of rubber, the compatibilizer, and the concentration of the blend. A careful inspection of Table 15.5 shows that the mean Tg value of pure NR was 63 C, and this changed to 64 C in the blend; the Tg value of pure EPDM was 37 C, and this changed to 45 C in the blend. This may be due to some interaction between NR and EPDM at the boundaries of their phases forming a third phase. DSC thermographs show the compatibilizing effects of BR, PVC, EPDM-g-MAH, and gradiation on NR/EPDM. For each component in the blend, Tg showed a higher shift Table 15.5 DSC Results Obtained for NR, EPDM, and 50/50 NR/EPDM Without and With Compatibilizers. Contribution from NR Sample NR EPDM NR/EPDM (control) NR/BR/EPDM NR/PVC/EPDM NR/EPDM-g-MAH/EPDM NR/g-radiation/EPDM NR/chlorinated rubber/EPDM NR/chlorosulfonated polyethylene/EPDM

Contribution from EPDM

Tg C

Shift in NR Tg C

Tg C

63 — 64 58 62 60 59.6 48

— — — þ6 þ2 þ4 þ4.4 þ12

— 37 45 44 2.5 40 45.6 48

60

þ4

[From Reference 57 with permission from John Wiley & Sons.]

60

Shift in NR Tg C — — — þ1 þ2.5 þ5 .6 3 14


461

than that observed in the following order: NR/BR/ EPDM > NR/g-radiation/EPDM >NR/ MAH /EPDM>NR/PVC/EPDM. Only one Tg was detected after the addition of chlorinated rubber or chlorosulfonated PE to NR/EPDM blends, and this indicated the improved compatibility or dominance of these phases. However, when the compatibilizers were added to the blends, the glass transition became less distinct, and this indicated improved compatibility. By using O’Neill’s method (77), the specific heat capacity of NR/EPDM blends was determined (31). It is found that the law of reciprocal affinity, the linear contribution of components to the specific heat capacity is followed in EPDM/NR blends.

15.11

ELECTRICAL PROPERTIES

NR/EPDM blends with various carbon black concentrations (0–30 phr) were analyzed in terms of electrical resistivity, dielectric breakdown voltage testing, and physical properties so that these blends could be used for the high insulation iron cross arms (70). It is essential that the carbon black concentration used in these applications is investigated to obtain an appropriate reinforced and insulated rubber for coating on an iron cross arm. Usually, the materials for iron cross arm coating must have a volume resistivity higher than 1011 ohm cm. Thus, the carbon black concentration is critical, which indicates that at most 10 phr are required for a suitable insulation compound. The volume resistivities of natural rubber and EPDM are 1015 and 106 ohm cm, respectively. As a consequence, the concentrations of carbon black at 0–10 phr were studied along with high concentration and are given in Table 15.6. Electrical properties such as volume resistivity and surface resistivity of blend samples are furnished. Blend with 30 phr carbon black was found to be overloaded in terms of both volume resistivity and surface resistivity. That is, the resistivity of these formulations was lower than the limit of the testing equipment. This is an effect of the quasigraphitic microstructure of the carbon black; this makes the blend more electrically conductive. The higher the surface/volume resistivity, the lower the leakage current and the less conductive the material is. The major application of carbon black Table 15.6

Effect of Concentration of Carbon Black on Electrical Properties.

Properties Volume resistivity 105, ohm cm) Surface resistivity 105, ohm

NEC0

NEC3

NEC5

NEC7

NEC10

NEC20

NEC30

2.7

2.2

3.8

3.5

3.3

1.8

b

7.5

6.8

1.6

1.2

4.2

5.5

b

NEC0, NEC3, NEC5, NEC7, NEC10, NEC20, and NEC30 are 70:30 NR/EPDM blends with carbon black content 0, 3, 5, 7, 10, 20, and 30 phr, respectively. b—the resistivities were lower than the limit of the testing equipment. [From Reference 70 with permission from John Wiley & Sons.]

462

Polyolefin Blends

in low concentration is for antistatic protection. The small amount of carbon black used in this application is to achieve a suitable resistivity before a significant drop in resistivity is noted. After applying the required high voltage at 10 kVac for 1 min to the specimens, the NR/EPDM samples do not exhibit any electrical burn-through, punctures, or electrical discharges. The results of dielectric breakdown voltage tests indicated that the appropriate concentration of carbon black is lower than 10 phr for insulation compounds. Effect of curing systems like sulfur and peroxide on the electrical properties of NR/EPDM blends were carried out (78). There is an improvement in the volume resistivity of blend samples, which could be used for insulation of wires and high voltage cables. The dielectric properties of NR/EPDM blends were conducted in varying proportions with a reinforcement of semireinforcing furnace carbon black (79). The permittivity and dielectric loss were determined. The permittivity and dielectric loss of NR/EPDM blends were determined with different compatibilizers at different frequencies (66). By changing the compatibilizer, the permittivity increases for BR, SPE, and chlorinated rubber are much more than that for PVC. An abrupt increase in the permittivity and dielectric loss was observed (Fig. 15.14) at a concentration of 6 phr EPDM-g-MAH in all blend ratios.

15.12 AGING PROPERTIES 15.12.1 Thermal Aging The effects of silica, carbon black, and a mixture of silica and carbon black on the mechanical properties of aged NR/EPDM sample are reported (70). The presence of silica in the blends produces a marked increase in mechanical strength after aging by 120%. The main effect of silica additive alone in NR/EPDM is very attractive. More interestingly, the synergistic improvement in tensile strength of the blend was obviously seen in the presence of both carbon black (3 phr) and silica (30 phr), after aging, with another increase of tensile strength by 38%. The silica filler contains the hydroxyl functional group on its surface. The intermolecular bonding between the hydroxyl group in silica and the NR/EPDM might be able to take place. The network structure further developed when carbon black was added. It is obvious that silica enhances the tensile strength of NR/EPDM blends after thermal aging, because silica improves heat resistance of the material, and it does not promote curing. The synergistic effect of carbon black and silica is likewise seen in the tear strength and hardness. This effect enhances the blend heat resistance by a substantial increase in tear strength and hardness. Figure 15.15 presents the retention of the tensile properties of the blends after treatment in an air-circulating oven at 70 C for 72 h. The best aging resistance has been achieved with the addition of EPDMTA, except in terms of elongation at break in blends vulcanized with higher amount of MBTS (61). Concerning blends compatibilized with EPDMSH, the aging resistance decreases as the concentration of the accelerator in the blend increases. The aging resistance of EPDMSH compatibilized blends is lower than noncompatibilized blends for accelerated-sulfur curing system.


463

Figure 15.14 (a) The permittivity (e0 ) versus the frequency (log f) for EPDM containing different 00

concentrations of EPDM-g-MAH as compatibilizer; (b) dielectric loss (e ) versus the frequency (log f) for EPDM containing different concentrations of EPDM-g-MAH as compatibilizer. (From Reference 66 with permission from Sage publications Ltd.)

15.12.2 Ozone Resistance NR/EPDM blends were extensively analyzed for its ozone resistance property (70). The blend ratios of NR/EPDM used in this study were 100/0, 80/20, 70/30, 60/40, and 0/100. To improve the NR/EPDM blends for ozone resistance, the amount of EPDM

464

Polyolefin Blends

Figure 15.15 Retention of the tensile properties of NR/EPDM blends with thermal aging: (a) noncompatibilized; compatibilized with (b) EPDMTA and (c) EPDMSH. (From Reference 61 with permission from Elsevier Ltd.)

cannot exceed 40% by weight. The natural rubber specimens could not withstand the ozone gas, which is the nature of the NR polymer. One can see cracks in the vulcanized rubber after tests. When 20 phr of EPDM and higher concentrations were blended, the specimens could withstand the ozone gas. The specimens did not crack. The dispersed EPDM domains reduced the crack length and increased the critical energy for macroscopic cracks. The EPDM domains function as crack dissipation centers, which delays the crack appearance. In the case of silica-filled blends, they enhance both mechanical properties (as reinforcement filler) and static ozone resistance. The silicafilled 80/20 (NR/EPDM) blends could withstand ozone gas of 50 ppm. Losses in tensile strength after aging in unfilled blends were higher than 50%, which is a very good indicator that silica can also withstand heat deterioration.


465

Table 15.7 Ozone Resistance of NR/EPDM Blends. Property Critical stress, MPa Critical strain, % Critical stored elastic energy kJ m3

1

2

3

4

0.142 10.15

0.180 11.55

0.210 12.55

0.240 11.25

7.350

10.00

12.65

13.06

1, 2, 3, and 4 are 70/30:NR/EPDM blends with TOR content of 0, 5, 10, and 20 phr, respectively. [From Reference 27 with permission from John Wiley & Sons.]

The ozone resistance of the blend was determined quantitatively in terms of critical stress–strain parameters and the values are given in Table 15.7 (27). The ozone resistance of the NR/EPDM blend is increased by the amount of 80% upon the addition of 10 phr of compatibilizer, TOR. The improvement in ozone resistance for the TOR containing blend is attributed to the better dispersion of the EPDM particles in the NR matrix, which is aided by TOR. That is, more finely dispersed EPDM particles prohibit the growth of ozone cracks initiated in the NR matrix before the crack grows over the critical length. Once the ozone crack grows over the critical size, crack propagation cannot be stopped by EPDM particles.There are studies available in the literature with special reference to the 60:40 ratio of NR: EPDM blend, which can provide excellent ozone resistance even in the absence of any antiozonant (80). This ratio was found good in maintaining complete ozone protection in the blend while maintaining the good set and elasticity performance. Absolutely, no crack was observed when the extruded product was subjected to ozone exposure at 40 C for 72 h.

15.13

TRANSPORT PROPERTIES

The swelling behavior of NR/EPDM blends in motor oil under compression strain was investigated (81). The weight uptake percentage was determined and plotted against the square root of the swelling exposure time (minutes) in motor oil (Fig. 15.16). The weight uptake increased as the exposure time increased, and equilibrium swelling was achieved after 192 h of immersion in oil. The lowest weight uptake was recorded with a mix containing EPDM polymer only (0/100), whereas the 25/75 NR/EPDM blend ratio recorded a lower weight uptake value. Generally, diffusion curves show the same pattern irrespective of the compression applied. It is also noted that the curve levels depended on the degree of compression applied. At a short exposure time in motor oil, the weight uptake was independent of the applied compression, whereas at longer exposure times, a decrease in the weight uptake with the applied compression was much more pronounced. At low compression (3%), the compression recovery percentage for all blend ratios was positive. At high compression, the recovery had negative values. The highest recovery value was recorded for the 25/75 NR/ EPDM blend. However, the lowest recovery was observed with the EPDM vulcanizate (0/100).

466

Polyolefin Blends

Figure 15.16 Weight uptake (%) of motor oil for NR/EPDM vulcanizates versus the exposure time (t1=2 ) at 100 C and 35% compression. (From Reference 81 with permission from John Wiley & Sons.)

The compression recovery decreased with the increase in applied compression for all exposure periods. The 25/75 NR/EPDM blend was advantageous because of its high degree of elastic recovery and its lower weight uptake of motor oil. Interaction of EPDM/NR blends with aromatic penetrates (82) aldehydes and ketones (83) and chlorinated penetrants (84) were studied. The sorptivity, diffusivity, and permeability values in nitrobenzene, cholrobenzene, and bromobenzene are lower compared to other NR-based polymer blends discussed (82). This is attributed to the tightly packed structure of EPDM blend and exhibiting both toughness of the plastic and elasticity of the gum elastomer phase. The mechanism of transport in NR/EPDM blends is Fickian irrespective of the penetrant used. The values of D, S, and P are much higher in chlorohydrocarbons than the other systems discussed. This is because of the strong interaction between the chlorinated hydrocarbons and blends.

15.14 APPLICATIONS The technology of making the NR/modified EPDM blends has been shown to be suitable for a number of applications such as extruded profile weather strips


467

Figure 15.17 Applications of NR/EPDM blends.

(Fig. 15.17a) for automotive and molded products like grommet (Fig. 15.17b) and washing machine gasket (Fig. 15.17c). These products typically use 100% EPDM, hence, the inclusion of a certain percentage of NR would reduce the products cost without affecting the products quality and performance. NR:EPDM with a ratio of 60:40 is also widely used for making extruded profile weather strip shown in Fig. 15.17a. Examples of such extruded profiles are door liners of car and back window seal of a van. Another major utility of these blends are for making diving suits. A few examples are shown in Fig. 15.18. Pro-Am (Fig. 15.18a) is a rubber suit made from NR/EPDM blend. The main features are good stretch characteristics for comfort and three layer construction for durability. It is ideal for sports, military, rescue, and light commercial applications. The Pro-hd (Fig. 15.18b) is tough NR/ EPDM blend rubber suit made to endure the harshest conditions. USIA (Fig. 15.18c) is also a vulcanized rubber suit made from NR/EPDM blend and it is ideal for sports, military, rescue, and light commercial applications. NR/EPDM blends are also used for making rubber boots for agriculture. These boots will have more resistance against cracks caused by ozone. A ratio of 25:75 to 40:60 parts by wt of NR/EPDM is used for preparing these kinds of boots.

468

Polyolefin Blends

Figure 15.18 Applications of NR/EPDM blends.


469

NR/EPDM-based sidewalls of radial tires are also prepared to get more durability and appearance. These blends with an EPDM content varying from 30 to 70 phr are used for making cables and conductors.

15.15

CONCLUSIONS

This chapter brings out the recent development in the field of EPDM/NR blends.The development of blends of NR with EPDM combines the superior physical properties and competitive price of NR with the excellent resistance to weathering, in particular, attack by ozone of EPDM. However, NR and EPDM blends are heterogeneous dispersion of fast curing NR phase and a slow curing EPDM phase. The overall result is that the blend vulcanizate will be composed of overcured NR and undercured EPDM. This would adversely affect the properties of the blend. By the judicious selection of NR-to-EPDM ratio and the concentration of DIPDIS in the compound, one can improve the physical properties of the vulcanizates. These properties can further be improved by two-stage vulcanization. The incorporation of compatibilizers into the NR/EPDM blends greatly enhances their compatibility and greatly improves the overall properties of the blends. The compatibilizers are able to create a well-dispersed bicontinuous phase. The Tg s from the DSC analysis indicate that both NR/EPDM and NR/BEPDM blends are thermodynamically incompatible. NR/ EPDM blends with 3 phr of carbon black are suitable for cable applications since it exhibits high tensile strength and suitable volume resistivity. EPDM-g-MAH was found to be an effective compatibilizer, which brings together the two incompatible components into the compatible level. A new method known as ‘‘reactive mixing’’ has developed recently to increase the cure rate of EPDM by modifying the EPDM phase to make it more reactive toward curatives, using commercially available sulfur donors such as bis-alkylphenoldisulphide (BAPD), in combination with dithiocaprolactam (DTDC) and/or dithiomorpholine (DTDM). The refinement of reactive mixing process with cost effective sulfur donors is one of the challenges in the maximum utilization of these elastomer blends.

NOMENCLATURE BAPD BEPDM BR DIPDIS DTDC DTDM EPDM EPDMTA EPDMSH

Bis-alkylphenoldisulphide Bromianted EPDM Polybutadiene rubber Bis(diisopropyl)-thiophosphoryl disulfide Dithiocaprolactam Dithiomorpholine Ethylene–propylene–diene rubber Thioacetate-modified EPDM Mercapto-modified EPDM

470

Polyolefin Blends

EPDM-g-MAH MBTS NR STR5L PVC TOR DGm DHm DSm

A graft copolymerization of EPDM with maleic anhydride 2, 20 -Dithiobisbenzothiazole Natural rubber Natural rubber from Thailand Poly(vinyl chloride) Trans-polyoctene rubber Gibb’s free energy Enthalpy of mixing Entropy of mixing

REFERENCES 1. J. A. Pomposo, E. Calahorra, L. Eguiazbal, and M. Cortazar, Macromolecules, 26, 2104 (1993). 2. J. W. Barlow and D. R. Paul, Polym. Eng. Sci., 27, 1482 (1987). 3. E. H. Andrews, Rubber Chem. Technol., 40, 435 (1967). 4. W. M. Hess, C. R. Herd, and P. C. Vegvari, Rubber Chem. Technol., 66, 329 (1993). 5. T. Johnson and S. Thomas, J. Mater. Sci., 34, 3221 (1999). 6. H. Varghese, S. S. Bhagawan, and S. Thomas, J. Polym. Sci. B Polym Phys, 37, 1815 (1999). 7. S. C. George, K. N. Ninan, G. Groeninckx, and S. Thomas, J. Appl. Polym. Sci., 78, 280 (2000). 8. S. Sadhu and A. K. Bhowmick, Rubber Chem. Technol., 78, 321 (2004). 9. S. Sadhu and A. K. Bhowmick, Rubber Chem. Technol., 76, 860 (2003). 10. S. Sadhu and A. K. Bhowmick, J. Appl. Polym. Sci., 92, 698 (2004). 11. J. A. Davidson and M. E. Woods, Rubber Chem. Technol., 49, 112 (1976). 12. V. A. Shershnev, Rubber Chem. Technol., 55, 537 (1982). 13. W. H. Whittington, Rubber Ind., 9, 151 (1976). 14. E. H. Andrews, Rubber Chem. Technol., 40, 635 (1967). 15. L. Spenadel and R. L. Sutphin, Rubber Age., 102, 55 (1970). 16. A. Y. Coran, Rubber Chem. Technol., 64, 801 (1992). 17. K. C. Baranwal and P. N. Son, Rubber Chem. Technol., 47, 88 (1974). 18. G. N. Ghebrehiwet and S. R. Stephanie, Rubber World, 227, 26 (2003). 19. D. J. Walsh, Polymer blends, in: Comprehensive Polymer Science, Vol. 2, Collin Booth and Collin Price (eds.), Pergamon Press, Oxford, 1989, chapter 5. 20. R. N. Jana and G. B. Nando, J. Elastom. Plasti., 34, 349 (2002). 21. R. L. Jalbert (ed.), Modern Plastics Encyclopedia, McGraw-Hill, NewYork, 1984. 22. G. Kerrutt, H. Blumel, and H. Weber, Kautsch. Gummi. Kunst., 22, 413 (1969). 23. O. Olabisi, L. M. Roberson, and M. T. Shaw, Polymer–Polymer Miscibility, Academic Press, New York, 1979. 24. D. R. Paul and J. W. Barlow, in: Multiphase Polymers, S. L. Cooper and G. M. Estes (eds.), American Chemical Society, Washington, DC, 1979. 25. G. N. Avgeropoulos, F. C. Weissert, P. H Biddison, and G. G. A. Bohm, Rubber Chem. Technol., 49, 83 (1976). 26. R. F. Bauer and E. A. Dudley, Rubber Chem. Technol., 50, 35 (1977). 27. Y. –W. Chang, Y. -S. Shin, H. Chun, and C. Nah, J. Appl. Polym. Sci., 73, 749 (1999). 28. B. Boutevin, E. Fleury, J. P. Parisi, and Y. Pietrasnta, Makromol. Chem., 190, 2363 (1989).


471

29. F. Romani, E. Passaglia, M. Aglietto, and G. Ruggeri, Macromol. Chem. Phys., 200, 524 (1999). 30. U. Gorski, K. Maenz, and D. Stadermann, Angew. Makromol. Chem., 253, 51 (1997). 31. T. Zaharescua, V. Meltzer, and R, Vilcu, Polym. Degrad. Stabi., 70, 341 (2000). 32. A. K. Ghosh, S. C. Debnath, N. Naskar, and D. K. Basu, J. Appl. Polym. Sci., 81, 800 (2001). 33. A. K. Ghosh and D. K. Basu, J. Appl. Polym. Sci., 84, 1001 (2002). 34. F. Guillaumond, Rubber Chem. Technol., 49, 105 (1976). 35. M. vanDuin, J. C. J. Krans, and J. Smedinga, Kautsch. Gummi. Kunst., 46, 455 (1993). 36. K. H. Wirth, U.S. Patent, 3, 492, 370 (1970). 37. C. B. Shulman, Rubber Chem. Technol., 59, 180 (1986). 38. K. Hashimoto et al., Nippon Gomu Kyokaishi, 43, 652 (1970). 39. R. T. Morrissey, Rubber Chem. Technol., 44, 1025 (1971). 40. Japan Patent, 3967 (1968). 41. F. P. Baldwin and G. Ver Strate, Rubber Chem. Technol., 45, 709 (1972). 42. F. Itsuro and M. Masao, Sumitomo Chemical Co. Ltd., Ger Offen., 2,045,574 (1971). 43. A. J. Tinker, Proceedings of International Rubber Conference, Moscow, Russia, MRPRA Publication, 1511, 1994, p. 180. 44. MRPRA. Res. Discl., 362, 308 (1994). 45. Chemische Werke Huls, A. G. Netherlands, 7, 31, 1958 (1974). 46. S. Yasui, M. Hirooka, and T. Oshima, Sumitomo Chemical Co., U.S. Patent, 3,649,573 (1972). 47. R. J. Hopper, Rubber Chem. Technol., 49, 341 (1976). 48. A. Y. Coran, Rubber Chem. Technol., 61, 281 (1988). 49. J. Lohmar, in: Proceedings of International Rubber Conference, Stuttgart, Germany, 1985, p 91. 50. D. G.Young, E. N. Kresge, and A. J. Wallace, Rubber Chem. Technol., 55, 428 (1982). 51. P. S. Brown and A. J. Tinker, J. Natl. Rubber Res., 5, 157 (1990). 52. M. J. R. Loadman and A. J. Tinker, Rubber Chem. Technol., 62, 234 (1989). 53. P. S. Brown and A. J. Tinker, J. Natl. Rubber Res., 11, 227 (1996). 54. S. K. Mandal, R. N. Datta, P. K. Das, and D. K. Basu, J. Appl. Polym. Sci., 35, 987 (1988). 55. H. S.-Y. Hsich, Adv. Polym. Technol., 10,185 (1991). 56. A. S. Sirqueira and B.G. Soares, J. Appl. Polym. Sci., 83, 2892 (2002). 57. S. H. El Sabbagh, J. Appl. Polym. Sci., 90, 1 (2003). 58. G. J. Anastasiadas and J. K. Koberstein, Macromolecules, 22, 1449 (1989). 59. D. J. Meier, Hetero-Phase Polymer Systems, American Chemical Society, Washington, DC, 1990. 60. R. Joseph, K. E. George, and J. D. Francis, Int. J. Polym. Matter, 11, 205 (1986). 61. A. S. Sirqueira and B.G. Soares, Eur. Polym. J., 39, 2283 (2003). 62. S. H. El Sabbagh, Polym. Test., 22, 93 (2003). 63. S. H. Botros, Poly-Plast. Tech. Eng., 41, 341 (2002). 64. A. B. Shehata, H. Afifi, N. A. Darwish, and A. M. El Syed, Poly-Plast. Tech. Eng., 45, 165 (2006). 65. A. M. El Sayed and H. Afifi, J. Appl. Polym. Sci., 86, 2816 (2002). 66. M. O. Abou-Helal and S. H. El Sabbagh, J. Elast. Plast., 37, 319 (2005). 67. N. Suma, R. Joseph, and K. E. George, J. Appl. Polym. Sci., 42, 2329 (1993). 68. H. Kenjiro, M. Minoru, M.Takahide, and O. Harunori, Nippon Gomu Kyokaishi. 49(3), 246 (1976). 69. G. Jin-Hwan and P. S. Soo, Komu Hakhoechi., 29, 121 (1994). 70. S. Kiatkamjornwong and K. Pairpisit, J. Appl. Polym. Sci., 92, 3401 (2004). 71. C. Lewis, S. Bunyung, and S. Kiatkamjornwong, J. Appl. Polym. Sci., 89, 837 (2003). 72. J. A. Brydson, Flow Properties of Polymer Melts, Plastics Institute, London, 1970, p. 12.

472

Polyolefin Blends

73. M. G. McCrum, C. P. Buckley, and C. B. Bucknall, Principles of Polymer Engineering, Oxford University Press, New York, 1997, p. 308. 74. L. A. Utracki, Polym. Eng. Sci., 22, 96 (1982). 75. P. P. Kundu, A. K. Bhattacharya, and D. K. Tripathy, J. Appl. Polym. Sci., 66, 1759 (1997). 76. P. P. Kundu, D. K. Tripathy, and B. R. Gupta, J. Appl. Polym. Sci., 63, 187 (1997). 77. M. G. O’Neill, Anal. Chem., 36, 1238 (1964). 78. P. Radivoj, T. Milan, G. Ivan, and S. Latinka, Plastika i Guma., 10(1), 23 (1990). 79. A. M. Ghoneim and M. N. Ismail, Polym. Plasti. Tech. Eng., 38, 979, (1999). 80. M. E. Samuels, in: Ethylene Propylene Rubber, R. O. Babbitt (ed.), R. T. Vanderbilt Company Inc, Norwalk, Connecticut, 1978, p. 147, chapter 5. 81. S. H. Botros and A. M. El Sayed, J. Appl. Polym. Sci., 82, 3052 (2001). 82. Siddaramaiah, S. Roopa, and U. Premkumar, Polymer, 39, 3925 (1998). 83. Siddaramaiah, S. Roopa, U. Premkumar, and A. Varadarajulu, J. Appl. Polym. Sci., 67, 101 (1998). 84. Siddaramaiah, S. Roopa and K. H. Guruprasd, J. Appl. Poly. Sci., 88, 1366 (2003).

Chapter

16

Phase Field Approach to Thermodynamics and Dynamics of Phase Separation and Crystallization of Polypropylene Isomers and Ethylene–Propylene–Diene Terpolymer Blends Rushikesh A. Matkar1 and Thein Kyu1

16.1 INTRODUCTION Polyolefins, one of the largest commodity polymeric plastics in the market place, have been widely studied over six decades covering synthesis, structural, physical, as well as mechanical properties point of views. However, the study on blends of polyolefins is rather scarce relative to their neat forms. One of the polyolefin blends that gained considerable attention is thermoplastic polyolefins (TPO) due to the enhanced impact strength and toughness of polyolefins. A typical example is a blend of polypropylene (PP) and ethylene–propylene rubber (EPR). EPR has been incorporated into PP through reaction in batch reactors or physical blending. The PP/EPR

1

Department of Polymer Engineering, University of Akron, Akron, OH 44325, USA


473

474

Polyolefin Blends

blends formed by mixing in the batch reactor are already phase separated and thus thermodynamics may not play a role, but the emerged structure and properties of physically blended ones may be affected by thermodynamic phase diagrams as well as by dynamics of phase separation. These rubber-modified polyolefins greatly improve the toughness and impact strength of the composites due to the rubber inclusion, whereas polyolefin constituents afford good tensile properties of the composites and also melt processability. A certain functional group may be introduced to EPR to provide chemical sites for cross-linking. Such cross-linking reaction further affords rubber-like network properties, but often it occurs at the expense of the reduction in melt processability. To circumvent such short comings, mineral or paraffinic oil has been added as a mean of improving processability and controlling swelling properties of the blends. Depending on the chemical structure of the functional groups, such reactive polyolefin blends are often known as thermoplastic elastomers (TPE) or thermoplastic vulcanizates (TPV). A classical example of TPE is the blend of polypropylene (PP) and ethylene propylene diene monomer (EPDM) (1–5). Despite a slight reduction in the rigidity or stiffness, these PP/EPDM blends exhibit enhanced toughness and impact strength, good resistance to ozone, and UV radiation without losing flow properties. The addition of small amount of PP raises the modulus and tensile stress of iPP/EPDM as compared with neat EPDM, and thus it has been regarded as a stiffness modifier. However, adding small amounts of impact modifier such as EPDM improves the toughness and impact strength of PP/EPDM at a marginal loss in tensile strength of the PP. As can be expected, the mechanical and physical properties of PP/EPDM blends are intimately related to the internal phase separated domain structures of the constituent phases. The emerged morphologies vary from a sea-and-island type to a bicontinuous structure, which may be governed by thermodynamic phase diagram, if it exists, as well as kinetics of thermally induced or reaction-induced phase separation. In addition, the crystalline phase of PP in the blends has to be addressed in evaluating the blend performance that has been ignored in the literature (1–5). In practice, the melt blends of commercial grade PP and EPDM were perceived to be immiscible, which may be a consequence of melt-blending conditions in given mixing equipment or driven by chemical reaction. To investigate the miscibility of a polymer blend, solution blending is preferred although it is usually not a customary practice in most industrial settings. This immiscibility perception has changed recently when a lower critical solution temperature (LCST) was first reported for the iPP/EPDM solution blends; this LCST was located very close to the melting temperature of the neat iPP (6). Moreover, the LCST phase diagram of iPP/EPDM blend was intervened by the melting transition of iPP, thereby implicating the phase behavior. In order to decouple LCST and Tm of iPP, Ramanujam et al (7) switched iPP to syndiotactic polypropylene (sPP) because sPP is known to have a lower crystalmelting temperature relative to that of iPP. These authors found that the existence of combined LCST and upper critical solution temperature (UCST) phase diagrams in the sPP/EPDM blend in which the melting transition of neat sPP is located in between the LCST and the UCST.

Chapter 16 Phase Field Approach to Thermodynamics

475

The motivation of the present article is (i) to elucidate the governing mechanisms of the spatio-temporal development of blend morphology involving the competition between the phase separation dynamics and kinetics of crystallization and (ii) to reconcile the different opinions of the complete immiscibility perception and the aforementioned complex liquid–solid phase diagrams of PP/EPDM thermoplastic elastomer systems. This chapter describes theoretical modeling and simulation on establishment of thermodynamic phase diagrams of PP isomers/EPDM and dynamics of thermal quench induced-phase separation and morphology development during crystallization of PP isomers.

16.2

EXPERIMENTAL PHASE DIAGRAMS

16.2.1 Cloud Point Phase Diagram of iPP/EPDM Blends Polypropylene may be classified in three isomeric forms based on the tacticity, viz., isotactic, syndiotactic, and atactic. Both isotactic and syndiotactic forms are highly crystalline, whereas the atactic PP is completely amorphous. On the contrary, EPDM is an amorphous copolymer comprised of equal parts of ethylene and propylene with 45% of ethylidene norbonene to afford dynamic vulcanization. We briefly review the miscibility studies of solvent cast iPP/EPDM blends primarily based on the contributions of Kyu’s group (6,7) and references therein. The solvent cast blend films were prepared by first dissolving iPP powder in xylene at 130 C, then adding EPDM after lowering the temperature to 100 C and stirred thoroughly for about 90 min to assure thorough mixing. The film specimens were prepared by solvent casting at ambient temperatures in a fume hood and dried in a vacuum oven for 48 h at room temperature. The average thickness of these blend films was approximately 1020 mm. The solvent cast films appear turbid to the naked eye, which might be attributed to the phase separation of the blends and/or the crystallization of iPP. It is difficult, if not impossible, to decouple the crystal-melting and liquid–liquid phase separation especially when the melting temperature of iPP and the LCST coexistence curve of the iPP/EPDM blend are in close proximity or intersecting each other. That is to say, iPP molecules may have gained sufficient mobility during premelting, and thus it is possible that phase separation could start before the crystal melting is completed. In order to circumvent the aforementioned problem, the cloud point measurement was performed based on light scattering by first melting the iPP crystal phase at 170 C and rapidly cooling it to 140 C within 2 min. Subsequently, the sample was reheated to 190 C at a rate of 0.5 C min1. The scattered intensity was measured at an approximate 2u angle of 20 during this reheating cycle. It was found that the blend at 140 C showed no scattering of light, suggestive of the homogenous character of the blend. However, the intensity increased rapidly as liquid–liquid phase separation commences at about 155 C. This procedure is thermally reversible so long as the phase separation process has not advanced significantly or the blend is not degraded. On the basis of this methodology, the phase diagram for this iPP/EPDM blend was

476

Polyolefin Blends

Figure 16.1 Experimental phase diagram of the iPP/EPDM blend as obtained by light scattering and DSC, showing the intersection of LCST and the crystal-melt phase transition. (From Reference (5) with permission from Elsevier.)

established by Chen et al. (6) as depicted in Fig. 16.1. The observed phase diagram is characterized by a lower critical solution temperature (LCST) type, which is intersected by the melting transition of iPP crystals. This phase diagram is reminiscent of an inverted teapot phase diagram of a polymer/liquid crystal system, exhibiting a liquid–liquid coexistence region, a narrow solid–liquid coexistence region, and neat crystal region bound by the solidus and liquid lines (Fig. 16.1).

16.2.2 Cloud Point Phase Diagram of sPP/EPDM Blends To alleviate the complex phase diagram of iPP/EPDM, Ramanujam et al. (7) investigated the sPP/EPDM blend as a complementary study. The advantage of its choice of sPP is that the melting point of sPP is significantly lower than that of iPP, which could effectively decouple the mutual interference of crystallization versus liquid–liquid phase separation. In Fig. 16.2 is shown the entire phase diagram of the solvent cast sPP/EPDM blends as established by a combination of differential scanning calorimetry (DSC) and cloud point determination. In descending order of temperature, the observed phase diagram exhibits an LCST curve, followed by a crystal–liquid transition, and subsequently UCST that lies underneath it. The coexistence lines have been drawn to guide the eyes. There exists a small miscible gap between the LCST and the UCST. The thermal reversibility of the LCST can be


477

Figure 16.2 Experimental phase diagram of the sPP/EPDM blend as determined by a combination of DSC and light scattering techniques, exhibiting the combined LCST and UCST together with the melting-point depression. The UCST curve was determined after the blends were homogenized in the single phase below the Tm , but above the crystallization temperatures. The symbols represent the experimentally determined points and the lines are drawn by hand or polynomial fits to guide the eyes. (From Reference (6) with permission from Elsevier.)

confirmed easily. However, the interference of crystal melting makes the confirmation of UCST more difficult at least experimentally, and thus it was cautioned that the UCST should be regarded as tentative. To substantiate the significance of these phase diagrams of iPP/EPDM and sPP/EPDM blends, we have developed a self-consistent theory for describing a crystalline–amorphous polymer blend in order to predict all possible phase diagram topologies and to compare some of these predictions with the observed cloud point phase diagrams. It may be anticipated that the present theoretical approach is capable of reconciling the discrepancy between the above phase diagrams and the perceived immiscibility of the PP/EPDM blend.

478

Polyolefin Blends

16.3 THERMODYNAMIC FREE ENERGY DESCRIPTION OF CRYSTALLINE POLYMER BLENDS 16.3.1 Flory–Huggins Free Energy of Amorphous–Amorphous Blends The Flory–Huggins (FH) lattice theory (8,9) has been customarily employed to establish phase diagram of binary amorphous–amorphous polymer mixtures based on the incompressible assumption that reads, DGm ðfÞ f ð1 fÞ ¼ lnðfÞ þ lnð1 fÞ þ xaa fð1 fÞ N A kB T r1 r2

ð16:1Þ

where DGðfÞ is the free energy of mixing. The volume fraction f may be expressed as f ¼ n1 r1 =ðn1 r1 þ n2 r2 Þ, where n1 and n2 are moles of each polymer having characteristic segment lengths of r1 andr2 , respectively. The total number of polymer chain segments is given as n ¼ ðn1 r1 þ n2 r2 Þ. Moreover, xaa is the FH amorphous– amorphous interaction parameter determining the enthalpy contribution toward mixing (8,10), which is proportional to the net interchange energy, but it is inversely proportional to absolute temperature. In order to account for free volume effects involving nonideality and noncombinatorial mixing, it is customary to expressxaa in the context of an empirical expression in what follows: xaa ¼ ðA þ B=T þ C ln TÞðDf þ Ef2 þ Ff3 þ Þ

ð16:2Þ

where the first bracket represents athermal and thermal dependencies, whereas the second bracket term accounts for concentration dependencies (11). This modified model has been used extensively to determine a variety of phase diagram topologies such as UCST, LCST, a combined UCST–LCST, a closed loop, and/or an hour-glass phase diagrams. Establishment of phase diagrams can be accomplished by applying a common tangent algorithm via free energy minimization of the mixture. This algorithm determines the coexistence line that satisfies the equal chemical potential principle and the free energy minimization which is the solution to the equations (12), @f ðfÞ @f ðfÞ f ðfa Þ f ðfb Þ ¼ ¼ @f fa @f fb fa fb

ð16:3Þ

where f ðfÞ is the free energy of the mixture at a given composition f. Although the FH theory may be adequate for elucidating the empirical phase diagrams of amorphous–amorphous, its extension to the crystalline–amorphous or crystalline–crystalline polymer blends requires a considerable modification by incorporating a solidification potential of the solid (crystal)–liquid (melt) transition (i.e., crystallization) of the crystalline constituent(s).


479

16.3.2 Extension of the FH Theory to Crystal–Amorphous Blends Existing theories of crystalline polymer blends have been predominantly dealt with the lowering of the pure crystalline constituent due to its miscibility with the polymeric diluent. The Flory diluent theory is a classical example of such a theory (8,11,13,14). One shortcoming of the Flory diluent theory is the assumption involved regarding the complete rejection of the solvent from the crystal phase, namely, the crystal phase in the blend is completely pure. This assumption has its own merit because equating the chemical potential of the crystalline component in the liquid phase to its chemical potential in the pure crystal phase has led to the analytically tractable solution, known as the Flory melting point depression equation. However, the Flory diluent theory encounters several limitations (15) when applied to partially miscible systems that exhibit phase separation in the vicinity of the melting temperatures such as the present iPP/EPDM blend. Also for systems that exhibit complex coupling between crystallization and phase separation (16–22) such as the above iPP/EPDM blend (6), it is imperative to take into account the crystal– amorphous interaction in addition to the conventional amorphous–amorphous FH interaction (23). Such a modification may make the free energy equation analytically intractable. However, with the advent of high speed computing capability of desktop machines, it is no longer necessary to adopt ‘‘a priori’’ for the neat crystal phase in the blends. In order to elucidate the thermodynamic phase diagram and kinetics of morphology evolution of crystalline microstructures, we shall introduce a phase field model of solidification for crystalline polymer blends undergoing the solid (crystal)– liquid (melt) phase transition. We then seek the self-consistent solutions to establish the coexistence phase boundaries of a hypothetical phase diagram of the crystal– amorphous blend. Polymer crystallization has been described in the framework of a phase field free energy pertaining to a crystal order parameter c in which c ¼ 0 defines the melt and assumes finite values close to unity in the metastable crystal phase, but c ¼ 1 at the equilibrium limit (23–25). The crystal phase order parameter (c) may be defined as the ratio of the lamellar thickness (l) to the lamellar thickness of a perfect polymer crystal (l0 ), i.e., c ¼ l=l0 , and thus it represents the linear crystallinity, that is, the crystallinity in one dimension. The free energy density of a polymer blend containing one crystalline component may be expressed as f ðc; fÞ ¼ ff ðcÞ þ

f ð1 fÞ lnðfÞ þ lnð1 fÞ þ fxaa þ xca c2 gfð1 fÞ ð16:4Þ r1 r2

where f ðcÞ is the free energy of crystallization of the crystalline component expressed as a Landau expansion in c that is weighted by its concentration or volume fraction in the mixture. This weighting ensures the free energy of the mixture to approach its pure crystal limit, that is, f ðc; fÞ ! f ðcÞ when f ! 1. The natural log terms represent the entropic contribution, whereas xaa fð1 fÞ is the enthalpic contribution to the Flory–Huggins free energy of liquid–liquid demixing. The

480

Polyolefin Blends

quantityxca c2 fð1 fÞ representing the solid–liquid interaction is complimentary to the amorphous–amorphous interaction term xaa fð1 fÞ. The xca parameter is called the crystal–amorphous interaction parameter that is repulsive and may be evaluated from the heat of fusion of the crystalline constituent as will be discussed later. When the linear crystallinity c is multiplied by its volume fraction, their product fc signifies the bulk crystallinity. Moreover, ð1 fÞc represents the amount of the noncrystalline component interacting with the crystal phase. The physical interpretation of xca c2 fð1 fÞ would therefore be the crystal–amorphous interaction term. It should be noted that the present modified theory reverts to the original Flory diluent theory in the extreme limit of complete rejection of the polymeric solvent from the crystal phase, that is, when the repulsive interaction between the crystal solute and amorphous solvent, xca becomes very strong or a neat crystal phase is formed in the blend. The free energy density of the crystal solidification may be expressed in the context of the asymmetric Landau expansion, f ðcÞ as FðcÞ zðTÞz0 ðTm Þ 2 zðTÞ þ z0 ðTm Þ 3 1 4 ð16:5Þ ¼W c c þ c f ðcÞ ¼ kB T 2 3 4 where W is a coefficient representing the energy cost for the system to overcome the nucleation barrier zðTÞ and z0 ðTm Þ represents, the location of the nucleation hump on the c axis and the solidification potential both of which are melting temperature dependent. This kind of asymmetric Landau potential has been utilized in the phase field model for the elucidation of solidification phenomena such as metal alloys or polymer crystallization (25). It should be cautioned that the coefficient of the cubic order must be nonzero in order to apply the Landau potential to the first-order phase transition; otherwise, the potential is applicable only to a second-order phase transition or at equilibrium where the two minima are equivalent (Fig. 16.3). The uniqueness of the present theory of polymer solidification is that these model parameters W, z, and z0 can be related to the material properties of the individual components and the experimental conditions (25). This Landau-type free energy of solidification has been successfully applied to describing the spatial temporal emergence of polymer single crystals, dendrite growth patterns, and dense lamellar branching in spherulites (25–29). The establishment of phase diagrams using the phase field model of crystallization is accomplished by minimizing the free energy of the mixture with respect to the nonconserved order parameters and finding the minimum free energy of the mixture at each composition. We then minimize f ðc; fÞ with respect to c in order to find the roots, leading to

or

@f ðc; fÞ ¼ Wðc zÞðc z0 Þ þ 2xca ð1 fÞ ¼ 0 @c Wðc zÞðz0 cÞ ¼ 2xca ð1 fÞ

ð16:6Þ

For the neat crystal, that is, f ¼ 1, the lower right-hand side of equation 16.6 becomes zero, thus cc ¼ z signifies the unstable potential hump, and cc ¼ z0


481

Figure 16.3 The variation of free energy of crystallization as a function of crystal order parameter, c of a pure homopolymer, showing a symmetric double well at equilibrium between c ¼ 0 and c ¼ 1 representing the melt and the solid phase, respectively. During supercooling to various temperatures, the shape of free-energy transforms to asymmetric double wells having the crystal order parameter at the solidification potential less than unity, reflecting the imperfect crystal (i.e., crystallinity of less than 1).

represents the solidification potential well of the neat crystal. In the blend, that is, f < 1, the solidification potential minimum cc ¼ cmin will shift away from z0 . At a given temperature of crystallization and a given concentration, W representing the penalty for overcoming the unstable potential hump. Since z and z0 are known, xca can be estimated analytically from the heat of fusion, DHu of PP crystals in the blends in accordance with the relationship of equation 16.6, that is, xca / W ¼ 6½ðDHu =kB TÞð1 T=Tm0 Þð1=2 zÞ1 . Alternatively, the self-consistent solution to equation 16.6 can be obtained by means of the steepest descent method with a tolerance of 1e 7. Subsequent to this minimization, we further calculate the coexistence curves based on the common tangent algorithm in what follows: @f ðc; fÞ @f ðc; fÞ f ðcmin ; fa Þ f ð0; fb Þ ¼ ¼ @f c¼cmin ;fa @f c¼0;fb fa f b

ð16:7Þ

16.3.3 Prediction of Phase Diagram Topologies In this section, we shall describe the various topologies of possible phase diagrams using the proposed thermodynamic model for a crystal–amorphous blend. Most polyolefin blends exhibit either an LCST or a UCST or a combination of both. In order to model the combined LCST–UCST or hour-glass phase diagram, the

482

Polyolefin Blends

empirical expression of xaa in equation 16.2 may be treated only as a function of temperature, viz., xaa ¼ A þ B=T þ C ln T

ð16:8Þ

Applying the conditions that xaa ¼ xcrit at the critical temperature of the UCST and LCST, one can treat these coefficients to be a function of only a single adjustable parameter A, that is, 2 3 1 1 6 7 lnðTUCST Þ lnðTLCST Þ 7 B ¼ ðxcrit AÞ6 ð16:9Þ 4 5 1 1 TUCST lnðTUCST Þ TLCST lnðTLCST Þ and C ¼ ðxcrit AÞ

TUCST TLCST TUCST lnðTUCST Þ TLCST lnðTLCST Þ

ð16:10Þ

where xcrit

1 1 1 2 ¼ pffiffiffiffi þ pffiffiffiffi 2 r1 r2

ð16:11Þ

On the basis of Equations 16.8–16.11, we have solved self-consistently for various phase diagram topologies of a hypothetical crystal–amorphous polymer blend having a critical LCST temperature at 252 C and a critical UCST temperature at 202 C, but the melting transition temperatures and crystal–amorphous interaction parameters vary. Figure 16.4 exhibits the influence of melting point on the phase diagram in columns from the top to the bottom as well as the effect of the crystal– amorphous interaction energy on the phase diagram in rows from left to right. As can be witnessed in Fig. 16.4a, the coexistence of the LCST and UCST can be established in which the UCST is intersected by the crystal solid–liquid transition, displaying liquid–liquid, crystal–liquid coexistence regions, and the neat crystal gap shown by the solidus line at the high crystalline polymer concentrations. With increasing repulsive crystal–amorphous interaction parameter, i.e., the solidus line moves toward the pure crystal component axis and concurrently the neat crystal gap become narrower suggesting that more solvent is rejected out from the solidus phase (Fig. 16.4a–c). On the contrary, upon raising the melting transition temperature of the crystalline polymer constituent, the solidus region increases as the solidus line moves up (Fig. 16.4a and d). With continued increase of the melting point, the crystal-melt transition is now intersecting with the LCST, thereby widening the crystal–liquid coexistence region like an hour-glass phase diagram (Fig. 16.4g). This kind of hour–glass phase diagram further transforms to almost completely immiscible


483

Figure 16.4 Hypothetical phase diagrams for crystal–amorphous polymer blends exhibiting a combined LCST and UCST intersected by the crystal–melt transition gap bound by the liquidus and solidus lines, showing (a) effect of melting temperature of the pure crystal component increasing from top to bottom (182, 223, and 262 C) and (b) the effect of repulsive crystal–amorphous interaction energy increasing from left to right. xca parameter is varied from 0.01, 0.1, and 1 at the melting temperature by setting the UCST temperature to 182 C and the LCST temperature to 262 C. The melting enthalpy of the constituent crystal was taken as 1500 cal mol1.

crystal solid—amorphous liquid (i.e., tree-trunk) gap with increasing the repulsive crystal–amorphous interaction (Fig. 16.4i). These theoretical predictions indicate clearly that it is possible to discern intricate phase diagram topologies encompassing the LCST coexistence curve coupled with the crystal–melt transitions, the combined LCST/UCST phase diagram and the solid–liquid coexistence region bound by the solidus and liquidus lines, all the way to the complete immiscibility of various PP/EPDM blends. The present calculation strongly suggests that the intricate phase diagrams of the iPP/EPDM reported by Chen et al. (6) and that of sPP/EPDM by Ramanujam et al. (7) are indeed possible and also the immiscibility perception for the melt blends of commercial PP/EPDM reported in literature is consistent with the present predictions.

484

Polyolefin Blends

16.3.4 Comparison with Experimental Phase Diagrams of PP/EPDM Blends It is encouraging that diverse phase diagram topologies predicted by the present theory for the hypothetical crystal–amorphous blend captures the observed trends of the phase diagrams of PP isomers/EPDM blends, it is essential to test directly with the experimental phase diagrams reported by Chen (6) for the iPP/EPDM blend (Fig. 16.1) and the sPP/EPDM blend of Ramanujam et al. (7) (Fig. 16.2) by utilizing the actual material parameters and the experimental conditions of iPP and sPP and their blends with EPDM. The materials’ parameters utilized were the enthalpy of fusion of iPP, DHiPP ¼ 2110 cal mol1 (30), the statistical segment length, riPP ¼ 1800, and rEPDM ¼ 1000, respectively. The equilibrium melting point of iPP is taken as 162.5 C and the LCST was calculated by setting A ¼ 0:01. The xca value was estimated to be 0.8 based on the heat of fusion of iPP crystal via the analytical expression of xca / W described earlier. Note that the density of the iPP crystal is approximated as unity so that the volume fraction roughly corresponds to the weight fraction of the PP. The conversion of the material parameters to the model parameters may be found elsewhere (23). There are two possible scenarios to draw the phase diagram depending on the value of xca as demonstrated in Fig. 16.4g and i. With xca ¼ 0:1, the solidus and liquidus lines are coincided, which intersected with the LCST that fitted reasonably well with the way the melting transition points and the experimental cloud point phase diagram were drawn in the original paper. However, in view of the estimated xca ¼ 0:8 value, the phase diagram seems more like that in Fig. 16.4i with the solidus line being located right on the pure iPP ordinate. Figure 16.5 shows the comparison between the self-consistently solved coexistence curves and the experimental cloud points of the iPP/EPDM blends and the melting points of iPP in the blends. This observation suggests that some cloud point data (UCST) falling below the liquidus line may be already in the solid crystal–amorphous liquid coexistence region suggesting the nonequilibrium nature of the cloud point determination methodology adopted by Chen et al. (6), especially below the crystal–melt transition. In addition, this kind of theoretical knowledge was not available at the time of their cloud point experiments (6). To alleviate such a complex interplay between the liquid–liquid phase separation and the crystallization, one idea that was developed was to replace iPP with sPP by virtue of the lower melting temperature of sPP relative to that of iPP. Upon replacing iPP with sPP, Ramanujam et al (7) were able to decouple the LCST of the sPP/EPDM with a minimum at around 150 C from the melting transitions of the sPP crystalline constituent located at 127 C (Fig. 16.2). However, the sPP/EPDM blend exhibited the UCST peak at around 100 C, which was buried under the melting point depression curve of sPP. The self-consistent solution gives the combined LCST/UCST phase diagram for this system by determining the model parameters using the material parameters for the sPP/EPDM blend. The material parameters utilized were the enthalpy of fusion of sPP (DHsPP ¼ 1912 cal mol1 ), the equilibrium melting temperature for sPP was approximated as 127 C, and the


485

Figure 16.5 Comparison between the self-consistently solved coexistence line and the cloud points (filled triangle) of iPP/EPDM and the melting points (filled circles). The solidus and liquidus lines are virtually overlapped (dots), but the existence of both lines is manifested by the kink in the LCST coexistence line. The phase diagram was calculated using the material parameters, DHiPP ¼ 2110 cal mol1 , Tm ¼ 162:5 C, riPP ¼ 1800, rEPDM ¼ 1000, and xca ¼ 0:8 at Tm .

statistical segment length, rsPP ¼ 1800 and rEPDM ¼ 1000 that roughly correspond to the molecular weights of the constituent polymers. The value was calculated through Equations 16.8–16.11 by setting A ¼ 0:01. Again the density of the sPP crystal is taken as unity so that the volume fraction in the theoretical description and the weight fraction of the experiment can be used interchangeably. The model parameter xca is estimated to be around 0.8 at 127 C from the relation of xca / W ¼ 6ðDHu =kB TÞð1 T=Tm0 Þð1=2 zÞ1 given above, which fits reasonably well to the experimental cloud point curves (Fig. 16.6). As shown in Fig. 16.6, the coexistence line afforded by the self-consistent solution shows a good fit with the LCST cloud points. The calculated liquidus line also closely matches the crystal-melting curve that is found to situate between the LCST coexistence curve and the UCST spinodal gap. The solidus line is situated right on the pure sPP crystal axis. However, the UCST coexistence line cannot be discerned since it has merged with the liquid and solidus lines, leaving the spinodal gap representing the unstable liquid envelope that was buried underneath the liquidus line. The tail end of the liquidus line curving downward asymptotically to the EPDM axis manifests the influence of the UCST, which in turn indicates that the UCST is merged with the liquidus line. At 67 C, it can be noticed that the buried spinodal spans all the way to 10 wt% of sPP. This probably explains the occasional observation

486

Polyolefin Blends

Figure 16.6 Comparison between the self-consistently solved coexistence curves and experimentally observed LCST (open diamonds) and UCST spinodal gap (open diamonds) showing the liquidus (denoted by open circles) and solidus lines on the pure sPP axis. The material parameters utilized were the enthalpy of fusion of sPP DHsPP ¼ 1912 cal mol1, Tm ¼ 123 C, rsPP ¼ 1800, rEPDM ¼ 1000, and xca ¼ 0:8 at the melting temperature.

of the spinodal type phase separated domains found in the 10/90 sPP/EPDM blend that experience deep quenches at around 27–67 C.

16.4 PHASE FIELD MODELING ON POLYMER PHASE TRANSITIONS 16.4.1 Theory on Phase Separation Dynamics and Morphology Evolution To elucidate the spatiotemporal emergence of crystalline structure and liquid–liquid phase separation in these polyolefin blends, we employ the time dependent Ginzburg–Landau (TDGL) equations pertaining to the conserved concentration order parameter and the nonconserved crystal order parameter. The spatiotemporal evolution of the nonconserved order parameter c, known as TDGL model-A equation (31,32), may be expressed as @c df ¼ Gc @t dc

ð16:12Þ


487

where Gc is the rotational mobility that is related to the viscosity of the system with a unit of frequency (s1 ) and df =dc is the functional derivative of the free energy density with respect to the order parameter, which is analogous to the pseudopotential. The product of the mobility and the potential represents the flux. For the case of the conserved order parameter such as volume fraction or concentration order parameter (f), we use the TDGL model B equation also known as the Cahn–Hilliard equation (33) @f df ¼ rGf r @t df

ð16:13Þ

where Gf is the mobility of the system, which is related to the self-diffusion coefficients of each constituent that obeys the Onsager reciprocal relation having a unit of m2 s1 . The functional derivative, which is analogous to the chemical potential pertaining to an arbitrary order parameter &, is defined as d=d& ¼ @=@& rð@=@r&Þ. This functional derivative is particularly important for a system whose total free energy of the system, f , comprises both the local free energy as well as the nonlocal free energy contributions that accounts for the moving interfaces. The nonlocal free energy may be given as fnonlocal ¼ f

kc kf jrcj2 þ jrfj2 2 2

ð16:14Þ

where kc and kf represent the coefficients of the interface gradient of respective order parameters. Then the total free energy of the system f may be represented by the total sum of local, nonlocal, and the coupling free energy terms in what follows: h i f kc ð1 fÞ lnð1 fÞ þ xaa fð1 fÞ f ¼ f f ðcÞ þ jrcj2 þ lnðfÞ þ 2 r1 r2 ð16:15Þ kf þ jrfj2 þxca c2 fð1 fÞ 2 The functional derivative of the free energy with respect to individual order parameters can thus be calculated, that is, df ¼ f½WðcÞðc zÞðc z0 Þ þ 2xca ð1 fÞc kc r2 c dc df 1 þ ln f 1 þ lnð1 fÞ þ xaa ð1 2fÞ kf r2 f ¼ df r1 r2 zðTÞz0 ðTm Þ 2 zðTÞ þ z0 ðTm Þ 3 1 4 c c þ c þW 2 3 4 kc 2 2 þ jrcj þxca ð1 2fÞc 2

ð16:16Þ

ð16:17Þ

488

Polyolefin Blends

Inserting Equations 16.16 and 16.17 into Equations 16.12 and 16.13, respectively, and subsequently nondimensionalizing them, the following equations of motion can be deducted, @c ~ c ff½WðcÞðc zÞðc z Þ þ 2x ð1 fÞc ~kc r~2 cg ¼ G 0 ca @t 9 8 1 þ ln f 1 þ lnð1 fÞ > 2 > > ~ þ xaa ð1 2fÞ ~kf r f > > > > > > > r r 1 2 > > > > = < @f zðTÞz ðT Þ zðTÞ þ z ðT Þ 1 2 m m 0 0 ~ 2 3 4 ¼r þW þ c c c > > @t 2 3 4 > > > > > > > > > > ~ k 2 c > > 2 ~ ; : þ rc þxca ð1 2fÞc 2

ð16:18Þ

ð16:19Þ

where the nondimensionalization was carried out by introducing dimensionless quantities for time and space coordinates, viz., t ¼ Gf t=‘2 and ~x ¼ x=‘. This allows ~ c ¼ Gc ‘2 =Gf , ~kc ¼ kc =‘2 and ~kf ¼ kf =‘2 us to obtain dimensionless variables, G that govern the spatio temporal growth of domain morphology. The advantage of the present model is that these phase field model parameters can be related to the material parameters that themselves are all supercooling dependent.

16.4.2 Dynamics of Crystal Growth in a Phase Separating System: iPP/EPDM Blends In view of the complex phase diagram of iPP/EPDM blend, it can be anticipated that crystallization kinetics and morphology development in iPP/EPDM blends would be complicated by the aforementioned phase separation. However, the thermodynamic phase diagram depicted in Fig. 16.1 certainly serves as guidance for mimicking the trajectory of thermal jump experiments. The 2D simulation has been carried out at the 50/50 iPP/EPDM blend composition (i.e., a near critical composition) following a temperature jump from a single phase temperature of 142 C to a temperature of 155 C, which is below the crystalmelting temperature, but it is above the LCST. The thermodynamic parameters used for the iPP/EPDM simulation are W ¼ 10, z ¼ 0:1, z0 ¼ 0:98, xca ¼ 0:8, ~kc 1, ~kf 1, r1 ¼ 18, r2 ¼ 10, and xaa ¼ 0:25. We have chosen the kinetic parameter ~ c 10 to reflect the fast crystallization rate as seen in the iPP melt crystallization G studies. The numerical simulations were carried out on several grid sizes of 128 128, 256 256, and 512 512 with varying time steps of 0.001, 0.0005, and 0.0001 to ensure the stability of the simulation. The results of the 512 512 simulations are shown in Fig. 16.7 exhibiting the emerged bicontinuous structure that is seemingly driven by liquid–liquid phase separation through spinodal decomposition in the composition field (upper row). The spinodal process is known to be spontaneous, thus any unstable fluctuation can grow very rapidly as this SD process does not required any energy to overcome. As can be seen in the bottom row, the nucleation (solid–liquid) of spherulite develops rather late relative to the phase separation, but it catches up very rapidly and eventually outgrows the SD domains.


489

Figure 16.7 Temporal evolution of the crystalline microstructure in the 50/50 iPP/EPDM blend, following a T-quench from the isotropic melt to a supercooled temperature below both the UCST spinodal gap, showing the growth of spherulitic front in the concentration field, but the overgrowth of this spherulitic boundary on the bicontinuous SD domain structures can be seen clearly only in the enlarged version.

The spherulitic boundary can be discerned in the enlarged picture, which is outgrowing over the interconnected SD domains. Although the simulated growth dynamics show interesting behavior, we did not have such knowledge at the time of the experiments performed some 10 years ago. Although the experiment was not ideal to quantitatively test with the present theoretical simulation regarding the competition between the phase separation and crystallization, it is still worthwhile to revisit what was observed experimentally at that time. Figure 16.8 shows the development of spherulitic morphology in the 50/50 iPP/ EPDM blend (6) upon cooling from 230 C to ambient at a slow cooling rate of 0.5 C min1. As depicted in Fig. 16.8a, the polarized optical micrograph under the cross

490

Polyolefin Blends

Figure 16.8 Morphology development in a 50/50 iPP/EPDM during cooling from 230 C to room temperature at a slow cooling rate of 0.5 C min1. (a) Polarized optical micrograph under cross polarizer’s clearly showing the maltese cross pattern indicative of the spherulite structure. (b) Four-lobe clover leaf pattern in SALS in the Hv configuration confirming the existence of the spherulite texture. (c) Polarized optical micrographs under parallel polarizer’s showing the phase-separated morphology and (d) the ring pattern in the Vv configuration dominated by the concentration fluctuations of the phaseseparated domains.

polarizers clearly reveals the maltese-cross pattern indicative of the iPP spherulite structure. The corresponding light scattering study exhibits a four-lobe clover pattern in the horizontal–vertical (Hv) configuration, which further confirms the existence of the spherulite texture (Fig. 16.8b). In the unpolarized configuration, the structure in the optical micrograph (Fig. 16.8c) is reminiscent of the phase-separated interconnected SD morphology. The corresponding light scattering under the vertical– vertical (Vv) configuration in Fig. 16.8d reveals a large scattering halo suggestive of domination by the concentration fluctuations of the phase-separated domains over the orientation fluctuations. That is to say, the orientation fluctuation can be attributed to the emerged crystalline spherulitic morphology, the size of which is too large that it primarily contributes to the main beam. Nevertheless, the observed overgrowth


491

of spherulitic structure on the existing interconnected SD structures can be identified, which accords very well with the above simulated structure. Hashimoto et al. (34) have also shown a similar growth trend in the PP/EPR blends exhibiting the spherulites outgrowing the SD domains upon quenching the system from a high temperature two-phase state to a temperature that is lower than the crystal-melting temperature yet above the LCST temperature of the blend.

16.4.3 Dynamics of Crystal Growth in a Phase Separating System: sPP/EPDM Blends In Fig. 16.9, the morphology development in a 50/50 sPP/EPDM blend isothermally quenched from 128 C (a single phase) to 100 C (two phases under the UCST) for a prolonged period is shown (7). The samples were analyzed under both unpolarized and cross-polarized configurations as time progressed. Even though phase separation occurs at this temperature, the cross-polar micrographs show the occurrence of crystallization, that is, the PP-rich domains are stretched along the bicontinuous regions dictated by the SD structure. A similar temperature quench has been carried out on the 70/30 sPP/EPDM blend that shows a different trend (7). As evident in Fig. 16.10, the EPDM-rich phase is dispersed in the form of globular droplets within the matrix of the PP-rich region. Thus the sPP crystals grow along channels by meandering around these EPDM-rich droplet domains, forming EPDM islands in the sea of sPP crystalline continuum. It may be inferred that in both the 50/50 and 70/30 blends, the crystal growth of sPP is confined within the phase-separated microstructure. This correlation between the crystalline and phase separated structures may be attributed to the strong crystal amorphous interaction that was predicted in the theoretical phase diagram. A similar result has also been reported by Crist and Hill (35) during thermal quenching of the blends of polyethylene and hydrogenated polybutadiene near the critical concentration. The 2D simulation of the kinetics of phase separation and morphology evolution in sPP/EPDM blends confirms this conjecture. Initially the system was set at a temperature (127 C) that is slightly higher than the melt temperature, but it is lower than the LCST temperature of the blend. First, a thermal jump was carried out from 127 to 157 C that lies inside the spinodal envelope of the LCST. No crystallization of the PP component was involved during such a jump beyond the melting temperature. The morphology evolution is seemingly governed by the simple liquid–liquid phase separation. As shown in Fig. 16.11a, the formation of the spinodally decomposed structures can be witnessed in the concentration order parameter field. As phase separation continues, the coarsening of the phases takes place as evident by the increase in periodic wavelength of the phase-separated bicontinuous domains (Fig. 16.11b). Another experiment was carried out by quenching the 50/50 blend from the initial temperature of 127 C (i.e., the single phase) to a temperature of 77 C that lies underneath the liquidus line and the UCST spinodal. The same set of parameters for the iPP/EPDM blend was employed except for the thermodynamic parameterxca ¼ 0:8 to account for the buried UCST immiscibility gap and the kinetic

492

Polyolefin Blends

Figure 16.9 Optical micrographs obtained for a 50/50 sPP/EPDM blend isothermally quenched at 100 C. Left column depicts the evolution of phase separation under the unpolarized condition and right column indicates the growth of crystals under the cross-polarized condition.

~ c 0:1 to reflect the slower crystallization of sPP relative to iPP. parameter G Nucleation was triggered by thermal noise in the crystal order-parameter field. During the initial stages, the formation of the spinodal phase-separated structures can be witnessed in the concentration field, but the crystals have yet to emerge in the crystal order-parameter field (Fig. 16.12). With the progression of time, some nuclei formed in the PP-rich regions because the probability of the nuclei to survive is much


493

Figure 16.10 Optical micrographs obtained for a 70/30 sPP/EPDM blend isothermally quenched at 100 C. Left column depicts the evolution of phase separation under the unpolarized condition and right column indicates the growth of crystals under the cross-polarized condition.

greater than those in the PP-poor regions. It can be envisaged that the growing crystalline regions are strongly dictated by the spinodal template, thereby loosing the radial growth habit such as spherulitic growth. This observation is indeed what was observed in the actual experiment of the sPP/EPDM blends (Fig. 16.9). In Fig. 16.13, the simulated structures are shown for the 70/30 sPP/EPDM mixture under the same conditions as described in the preceding case. The asymmetry in the composition results in the change of mechanism of phase separation from the spinodal to the nucleation-growth mechanism. Now the minority EPDM-rich phase forms the droplets in the continuum of sPP-rich phase. As can be anticipated, the PP crystallization lags behind the liquid–liquid phase separation because phase separation must occur first in order for the sPP phase to reach or exceed its critical

494

Polyolefin Blends

Figure 16.11 Emerged bicontinuous structure following a temperature jump from the isotropic melt between the LCST and the melting transition into the LCST gap, which is presumably driven by liquid– liquid phase separation through spinodal decomposition in the 50/50 sPP/EPDM mixture: (a) 1000 s and (b) 3000 s.

concentration so that the crystal nucleation can occur. The crystallization of sPP in the blends proceeds by weaving around the discrete EPDM domains. On the contrary, the sPP forms the droplets in the continuum of EPDM in the case of 30/70 sPP/EPDM. It is striking to discern that sPP crystals are strictly confined in the sPP-rich droplets (Fig. 16.14). The crystallization of sPP lags behind the liquid–liquid

Figure 16.12 Competition between the liquid–liquid phase separation through spinodal decomposition and the crystalline structure formation in the 50/50 sPP/EPDM blend. The crystallization occurs with the preformed SD networks. The top and bottom rows represent the temporal evolutions of the concentration field and the corresponding crystal order field.


495

Figure 16.13 Competition between the liquid–liquid phase separation through nucleation and growth showing the droplet domains and the crystalline structure formation in the 70/30 sPP/EPDM blend. The crystalline sPP component being the major phase, the sPP crystallization occurs in the matrix by weaving around the EPDM domains. The top and bottom rows represent the temporal evolutions of the concentration field and the corresponding crystal order field.

Figure 16.14 Competition between the liquid–liquid phase separation through nucleation and growth and the crystalline structure formation in the 30/70 sPP/EPDM blend. The crystallization of sPP is confined to the sPP-rich droplets. The top and bottom rows represent the temporal evolutions of the concentration field and the corresponding crystal order field.

496

Polyolefin Blends

phase separation occurring through the nucleation and growth mechanism except that the former is governed by the solid–liquid phase transition as opposed to the latter case of liquid–liquid phase separation. Again, the sPP concentration must reach the threshold value for the nucleation to occur that can only be achieved through phase separation.

16.5 CONCLUSIONS We have demonstrated the interplay of solid–liquid phase separation and liquid–liquid phase separation in the blends of iPP/EPDM and sPP/EPDM showing the influence of PP tacticity on phase diagrams. In the establishment of the experimental phase diagrams for the iPP/EPDM blend, we observed the intersection of the solid–liquid coexistence curves with the liquid–liquid coexistence curves that prompted the study of the sPP/EPDM blend to decouple the two competing processes. We have developed a thermodynamic model based on the crystal–amorphous interaction in addition to the conventional amorphous–amorphous interaction of the FH theory. With this modification, one can predict various phase diagrams of crystalline–amorphous polymer blends exhibiting both LCST and UCST behaviors coupled with the melting transition of one of the components. The crystal–liquid gap was bound by the solidus and liquidus lines that cannot be achieved by the original Flory diluent theory. On the basis of the present modified free energy, the interplay between phase separation and crystallization is explicable in the frame work of the phase field approach based on the TDGL-Model C equations of motion for the concentration and crystal order parameters. Various morphological features have been simulated that are consistent with the observed crystal morphologies of iPP/EPDM as well as of sPP/EPDM blends.

NOMENCLATURE c f z kc Gc Gf kf z0 xaa xca r1 ; r2 Tm W

Crystal order parameter corresponding to linear crystallinity Concentration order parameter or volume fraction Peak position of the energy barrier for solidification Coefficient tensor of the interface gradient of the c field Mobility related to propagation of crystal-melt interface of the c field Mutual diffusion coefficient related to translational mobility of the constituents Coefficient of the concentration gradient of the f field Crystal order parameter at the equilibrium solidification potential Amorphous–amorphous interaction parameter Crystal-amorphous interaction parameter Statistical segment lengths of components 1 and 2 Melting point at the extrapolated zero heating rate Coefficient representing the free energy penalty to overcome the unstable potential hump pertaining to c

Note: Symbols with tilde signify the dimensionless quantities


497

REFERENCES 1. A. Y. Coran, Handbook of Elastomers: New Developments and Technology, A. K. Bhowmick and H. L. Stephens (eds.), Marcel Dekker, New York, 1988, p. 249. 2. L. A. Goettler, J. R. Richwine, and F. J. Wille, Rubber Chem. Technol, 55, 1558 (1982). 3. A. Y. Coran and R. Patel, Rubber Chem. Technol. 53, 141 (1980); Rubber Chem. Technol. 53, 781 (1980); Rubber Chem. Technol. 54, 892 (1981); Rubber Chem. Technol. 56, 210 (1983). 4. A. Y. Coran and R. Patel, US Patent 4,297,453 (1981); US Patent 4,310,638 (1982); US Patent 4,350,470 (1982). 5. D. J. Lohse and W. W. Graessley, in: Polymer Blends, Vo.. 1, D. R. Paul and C. B. Bucknall (eds.), 2000, pp. 219–237, chapter 8. 6. C. Y. Chen, W. Md. Z. Yunus, H. -W. Chiu, and T. Kyu, Polymer 38, 4433 (1997); M.S. Thesis, University of Akron, Akron, Ohio, 1993. 7. A. Ramanujam, K. J. Kim, and T. Kyu, Polymer 41, 5375 (2000); A. Ramanujam, M.S. Thesis, University of Akron, Akron, Ohio, 1998. 8. P. J. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca, NY, 1953. 9. P. J. Flory, J. Chem. Phys., 17, 223 (1949). 10. J. M. Prausnitz, R. N. Lichtenthaler, and E. G. de Azevedo, Molecular Thermodynamics of FluidPhase Equilibria, Prentice-Hall, NY, 1999. 11. R. Konigsveld and W. H. Stockmayer, Polymer Phase Diagrams, Oxford University Press, Oxford, 2001. 12. T. Kyu and H. -W. Chiu, Phys. Rev E 53, 3618 (1996). 13. T. Nishi and T. T. Wang, Macromolecules 8, 909 (1975). 14. W. R. Burghardt, Macromolecules 22, 2482 (1989). 15. J. C. Canalda, Th. Hoffmann, and J. Martinez-Salazar, Polymer 36, 981 (1995). 16. Y. W. Cheung and R. S. Stein, Macromolecules 27, 2512 (1994). 17. Y. W. Cheung, R. S. Stein, J. S. Lin, and G. D. Wignall, Macromolecules 27, 520 (1994). 18. P. M. Cham, T. H. Lee, and H. Marand, Macromolecules 27, 4263 (1994). 19. J. P. Penning and R. St. J. Manley, Macromolecules 29, 77 (1996). 20. J. P. Penning and R. St. J. Manley, Macromolecules 29, 84 (1996). 21. K. Fujita, T. Kyu, and R. St. J. Manley, Macromolecules 29, 91 (1996). 22. H. Tanaka and T. Nishi, Phys. Rev. A, 39, 783 (1989). 23. R. A. Matkar and T. Kyu, J. Phys. Chem. B. 110, 12728 (2006). 24. J. D. Gunton, M. San Miguel, and P. S. Sahni, in: Phase Transitions and Critical Phenomena, Vol. 8 C. Domb and J. L. Lebowitz (eds.), 1983, p. 269, chapter 3. 25. H. Xu, R. A. Matkar, and T. Kyu, Phys. Rev., E 72, 011804 (2005). 26. H. Xu, W. Keawwattana, and T. Kyu, J. Chem. Phys., 123, 124908 (2005). 27. R. Kobayashi, Physica D 63, 410 (1993). 28. A. Wheeler, W. J. Boettinger, and G. B. McFadden, Phys. Rev. A, 45, 7424 (1992). 29. T. Kyu, R. Mehta, and H. -W. Chiu, Phys. Rev. E, 61, 4161 (2000). 30. S. Brandrup and E. H. Imergut, Polymer Handbook, Vol. 5, Interscience, New York, 1975, p. 24. 31. S. -K. Chan, J. Chem. Phys. 67, 5755 (1977). 32. P. R. Harrowell and D. W. Oxtoby, J. Chem. Phys., 86, 2932 (1987). 33. J. W. Cahn and J. E. Hilliard, J. Chem. Phys., 28, 258 (1958). 34. N. Inaba, T. Yamada, S. Suzuki, and T. Hashimoto, Macromolecules 21, 407 (1988). 35. B. Crist and M. J. Hill, J. Poly. Sci. Phys., 35, 2329 (1997).

Part III

Polyolefin/Nonpolyolefin Blends

Chapter

17

Compatibilization and Crystallization of Blends of Polyolefins with a Semiflexible Liquid Crystalline Polymer Liliya Minkova1

17.1 BLENDS OF POLYOLEFINS (HIGH DENSITY POLYETHYLENE AND ISOTACTIC POLYPROPYLENE) WITH A SEMIFLEXIBLE LIQUID CRYSTALLINE POLYMER 17.1.1 Introduction Much attention has been paid recently to blends of commercial thermoplastics with liquid crystalline polymers (LCPs) (1). The main benefits expected from the use of LCPs as blend components are (1) the pronounced reduction of the melt viscosity, with consequent improvement of processability, and (2) the reinforcing effect granted by the immiscible LCP particles, which can attain oriented fibrillar morphology when the blend is processed under elongational flow conditions. In situ composites (2) have been prepared by the addition of LCPs into a great variety of flexible resins, such as polyamides, polyesters, polycarbonates, and so on, and their morphology–processing–property relationships have been studied. For the particular case of polyolefin/LCP blends the most available thermotropic LCPs, which belong to the classes of wholly aromatic copolyesters

1

Institute of Polymers, Bulgarian Academy of Sciences, Acad. G. Bonchev str. Bl.103A, 1113 Sofia, Bulgaria Polyolefin Blends, Edited by Domasius Nwabunma and Thein Kyu Copyright # 2008 John Wiley & Sons, Inc.

501

502

Polyolefin Blends

and copolyesteramides, show very poor compatibility and interphase adhesion toward polyolefins. Moreover, the processing temperatures of common aromatic LCPs are in the 300 C range and are, therefore, much higher than those used for polyolefin processing. In fact, the reinforcement of polypropylene (PP) with different LCPs has normally led to blends with no improvement in the tensile strength, although some modulus enhancements were sometimes observed (3–5). As for the blends of polyethylene (PE) with LCPs, the studies confirm that the poor compatibility of the two polymers, coupled with the strong differences between their melting/processing temperatures, prevents the attainment of good morphologies and enhanced mechanical properties (6–11). It has been assumed that a semiflexible LCP, containing aliphatic spacer in the main change and having lower melting point, will be more suitable to be blended with polyolefins. In fact, slightly better results were obtained by blending linear low density polyethylene (LLDPE) with a semiflexible liquid crystalline polymer: SBH 112 by Eniricerche, Milan, synthesized from sebacic acid (S), 4,40 -dihydroxybiphenyl (B), and 4-hydroxybenzoic acid (H) in the mole ratio 1:1:2 (12). For these blends, the phase dispersion was good and the size distribution of the LCP droplets was fairly narrow. Moreover, the LCP phase was shown to play a nucleating effect for the crystallization of the LLDPE matrix (12,13). The mechanical characterization showed no reduction of the tensile strength and a 50% modulus increase over the LCP concentration range 0–20%, whereas the elongation to break decreased markedly only for SBH contents higher than 10%.

17.1.2 Blends of High Density Polyethylene (HDPE) with LCP The possibility of reinforcing HDPE by blending it with an LCP is based on the successful improvement of phase compatibility and interfacial adhesion of these two structurally unlike polymers. The addition of compatibilizing agents into intrinsically immiscible polymer blends can have a substantial merit to solving the problems of poor dispersion and low adhesion. Among the substances exhibiting compatibilizing activity, block or graft copolymers made up of chain segments having chemical structure and/or solubility parameters similar to those of the polymers being blended are most promising. The approach that has been considered in our laboratories consists of the synthesis of PE–LCP block or graft copolymers and of their use as compatibilizing agents for PE/LCP blends. Two main routes have been used for the PE-g-LCP synthesis. The first one is melt polycondensation of sebacic acid (S), 4,40 dihydroxybiphenyl (B), and 4-hydroxybenzoic acid (H), carried out at temperatures up to 280 C in the presence of an oxidized low molar mass PE sample containing free carboxylic groups (PEox) (14). The second one is reactive blending of PEox and a semiflexible liquid crystalline polyester (SBH 1:1:2) (50/50 w/w), at 240 C in a Brabender mixer, in the presence of Ti(OBu), catalyst, for different mixing times (15, 60, and 120 min) (15). The formation of PE-g-SBH copolymers is shown schematically in Fig. 17.1a for the first route and in Fig. 17.1b for the second route.

503

Chapter 17 Compatibilization and Crystallization of Blends + S, B, H

+ CH 3COOH

COOH

COO (SBH) HOCO (CH 2) 8 COOH

CH 3 COO

=S =B

OCOCH3

CH3COO

COOH

=H

(a)

H 2O COOH

(SBH)n =

+ (SBH) n

CO (CH 2)8 CO

+

COO

n

O

O

n

(SBH) n – m

O

CH 3COOH (SBH) m

CO

n

(b)

Figure 17.1 (a) Reaction scheme for the production of COP. (b) Reaction scheme for the production of the reactive blend. (From References 14 and 15 with permission from John Wiley & Sons, Inc.)

A physical blend between PEox and SBH 112 50/50 w/w has also been prepared in a Brabender apparatus at 240 C for 6 min without a catalyst for comparison. The polycondensation product COP, the reactive blends COP15, COP60, and COP120, and the physical blend MIX have been extracted with PE solvents with increasing solubility parameters and boiling points (toluene and xylene), in order to gain information on the presence of a true copolymer in the different fractions. The soluble fractions and the residues have been analyzed by Fourier transform infrared (FTIR) spectroscopy, (nuclear magnetic resonance) NMR spectroscopy, thermogravimetry (TG and DTG), differential scanning calorimetry (DSC), and scanning electron microscopy (SEM). All analytical procedures concordantly show that PE-g-SBH copolymers have, in fact, been obtained. For COP, both PEox and SBH chain segments are present, with different relative ratios, in all fractions of the polycondensate. Moreover, IR analysis has shown that a fairly quantitative esterification of the PEox carboxyl groups takes place under the adopted conditions. For the reactive blends, PE-g-SBH copolymers with different compositions, arising from differences of either the number of PEox carboxylic groups entering the transesterification or the length of grafted SBH branches, are formed as a result of blending. Qualitative IR analyses and quantitative TG measurements have shown that the amount of copolymers increases strongly with the mixing time of the reactive blends. The calorimetric data in Table 17.1 also demonstrate that the soluble and insoluble fractions of MIX consist essentially of neat PEox and SBH, respectively,

504

Polyolefin Blends

Table 17.1 Temperatures and Enthalpies of Fusion/Crystallization of PE Phase of MIX, COP, COP120, and Their Fractionsa. Cooling Tc , C Sample MIX 115.1 MIX toluene 114.1 soluble MIX xylene — insoluble COP 107.4 COP toluene 108.2 soluble COP xylene 105.0 insoluble COP120 111.2 COP120 toluene 113.1 soluble COP120 xylene 103.2 insoluble

DHc , J g1 87.3 186.6

Heating Tm , C 131.0 131.1

SBH, wt% SBH, wt% DHm , J g1 (from DSC) (from DTG) 90.3 45–55 53 186.1 — 0

—

—

—

—

100

82.4 167.9

126.5 128.6

81.0 172.4

50–57 5–10

49 0

29.4

120.1

26.7

70–75

75

73.8 180.9

128.1 128.1

73.5 177.2

— —

51 0

36.0

116.3

33.2

—

73

a

From References 14 and 15 with permission from John Wiley & Sons, Inc.

whereas those of the polycondensation product and reactive blends contain significant amounts of the other component, too. The contents of PE chains in the insoluble residues, calculated on the basis of the melting/crystallization enthalpies, are in fair agreement with those found by TG (Table 17.1) (14,15). The SBH content from TG measurements has been calculated by the weight losses corresponding to the degradation processes of both components, which take place at different temperature intervals. The SEM micrographs of MIX and COP60 are shown in Fig. 17.2, together with those of their insoluble residues. Both blends have a two-phase morphology. The SEM micrographs of the insoluble fractions of MIX and COP60 (Fig. 17.2c and d) reflect their different compositions. The morphology of the xylene-insoluble fraction of MIX appears homogeneous and fibrous, as expected for neat SBH, and that of COP60 xylene-insoluble fraction, on the contrary, shows a two-phase morphology. It could be assumed that the matrix is made of practically pure SBH and the minor phase consists of PE-g-SBH copolymer. Better understanding of the crystalline structure and morphology of the PEg-SBH copolymers is necessary to confirm their potential of compatibilizers. In cases of different polymer blends with crystallizable components, the X-ray diffraction method is reliable to find the degree of order of the LCP phase in the blend and its influence on the crystallinity of the matrix (16,17) as well to examine the crystallizability of the copolymer segments (18). The X-ray diffraction patterns of PE-g-SBH copolymers obtained via both procedures (19) consist of reflections typical for the orthorhombic crystalline lattice of PE and the single reflection of the solid LCP. The lack of dhkl variations with respect to those of neat PEox and SBH

Chapter 17 Compatibilization and Crystallization of Blends

505

Figure 17.2 SEM micrographs of MIX (a), COP60 (b), xylene-insoluble fraction of MIX (c), and xylene-insoluble fraction of COP60 (d). (From Reference 15 with permission from John Wiley & Sons, Inc.)

indicates the absence of interactions in the crystalline phase or that of cocrystallization phenomena between the components of the PE-g-SBH copolymers. The analysis of the crystallinity degree and normalized amorphous and crystalline contributions to the diffraction patterns of the products suggests that both copolymer components are partly miscible in the amorphous phase. The extent of miscibility depends on the copolymer structure, namely on the length of PE segments and SBH grafts. PE segments in PE-g-SBH copolymers obtained by the reactive blending are longer and exhibit a higher crystallizability than those obtained via melt polycondensation. SBH grafts of the copolymers obtained by reactive blending are also longer than those in the products obtained via melt polycondensation (19). To prove the compatibilizing efficiency of the ad hoc synthesized PE-g-SBH copolymers, the latter have been employed as compatibilizing additives for blends of HDPE with SBH (14,15,20). The rheological and mechanical properties, as well as the morphology of the compatibilized blends, have been compared with the uncompatibilized ones. A commercial sample of high density polyethylene grafted with maleic anhydride has also been used as compatibilizer for the same blends for comparison (20). In Fig. 17.3, the SEM micrograph of 76/8/16 HDPE/COP120/ SBH ternary blend is compared to that of the 80/20 HDPE/SBH binary blend. The improvement of both phase dispersion and interfacial adhesion brought about by the addition of COP120 is evident. Similar results have also been found by the addition of the PE-g-SBH copolymer prepared by polycondensation. The melt viscosity of the compatibilized blends is, in general, slightly higher than that of the corresponding uncompatibilized ones, due the increased interfacial

506

Polyolefin Blends

Figure 17.3 SEM micrographs of 80/20 HDPE/SBH (a) and 76/8/16 HDPE/COP120/SBH (b and c) blends. (From Reference 15 with permission from John Wiley & Sons, Inc.)


507

Figure 17.4 (a) Viscosity curves of the pure components: PE (HDPE), SBH, HDM (maleic anhydridegrafted PE), COPR (COP), and COPM (COP120). (b) Viscosity curves of the 80/20 PE/SBH blends with 5 wt% of compatibilizing agents. (From Reference 20 with permission from John Wiley & Sons, Inc.)

adhesion, especially in the low shear rate range (Fig. 17.4a and b) (20). Despite this, the viscosity of the blends always remains lower than that of the pure polyolefin (Fig. 17.4a and b). Thus, polyolefin processability can actually be enhanced by the addition of LCP, even in the presence of compatibilizers. However, the effect of the addition of COP120 or COP on the mechanical properties of the HDPE/SBH blend is fairly modest. The tensile modulus is slightly decreased, while the elongation at break is increased almost two times (20). It should be noted that the commercial compatibilizer does not improve the elongation at break. This behavior confirms that

508

Polyolefin Blends

whereas the compatibilizing role of the MA-grafted PE is very modest, the potential of the COP120 and COP is certainly higher. The obtained results may be rationalized considering that the PE-g-LCP copolymers used by us consist of fairly short PE backbones with LCP grafts of various length. The molecules with longer LCP branches are thought to become mixed at the surface of the LCP particles and to give rise to fairly weak interaction with the PE matrix. It is argued that new PE-g-LCP copolymers synthesized from functionalized PE samples of higher molar mass might show much better compatibilizing performance (20). The dependence of the components’ molar mass and of the mixing conditions on the compatibilizing efficiency of PE-g-SBH copolymers has also been studied (21). The results indicate that the PE-g-SBH copolymers do, in fact, compatibilize the HDPE/SBH blends and that the effect is more pronounced with the lower molar mass PE matrix and with the SBH sample having lower viscosity. Moreover, the compatibilizing ability of the graft copolymer is improved, if the latter is first blended with either of the two main components (21). An attempt has been made to synthesize copolymers with better compatibilizing efficiency (22) using functionalized PE of a higher molar mass. A novel graft copolymer (PE-g-LCP) consisting of PE backbones and LCP branches was synthesized via reactive blending of an acrylic acid-functionalized PE (Escor 5000 by Exxon) with a semiflexible LCP (SBH 1:1:2 by Eniricerche S.p.A.) (23). The investigations of the fractions soluble in boiling toluene and xylene and of the residue have shown that crude product (COP-AA) contains unreacted Escor and SBH, together with the graft copolymer forming the interphase. The morphology of the copolymers is not homogeneous and reveals two types of structures characteristic for both copolymer segments, namely tiny PE spherulites and small and large LCP domains (24). That means LCP does not lose its mesophase behavior when bonded to the PE backbone. The morphological observations also confirm that the distribution of SBH grafts along the PE backbone is random (24). The two segments in the copolymers form separate crystalline phases typical for the neat components. Similar to PE-g-SBH copolymers obtained previously, there are no interactions in the crystal phase or cocrystallization between the copolymer segments. The addition of relatively small amounts of COP-AA into binary HDPE/SBH blends was shown to cause a considerable reduction of the dimensions of the SBH particles and an improvement of the adhesion between the phases (23). A slight enhancement of interfacial adhesion was also shown to take place in the molten state as well. The compatibilizing efficiency of this product is considerably higher than that of pure Escor and, supposedly, of other commercially available functionalized polyolefins. With respect to the latter compatibilizers, COP-AA has apparently the advantage of a strong affinity to the SBH domains of the blends. The results also confirm previous findings indicating that optimum compatibility between the PE segments of the PE-g-SBH copolymer and the PE matrix is realized when the latter has an appropriate molar mass. Melt spinning tests demonstrated that deformation of the SBH droplets into highly oriented fibrils can be obtained for the blends of lower


509

Figure 17.5 SEM micrograph of the fracture surface of the fiber of the 78/4/18 HDPE/COP-AA/SBH blend. (From Reference 23 with permission from John Wiley & Sons, Inc.)

molar mass PE, compatibilized with small amounts of the novel PE-g-SBH copolymer (23) (Fig. 17.5). The assumption has been made that the compatibilization activity of these copolymers toward HDPE/SBH blends is due not only to the identical chemical structure of the copolymer segments and the corresponding blend components but also to the similarity of their crystalline structure, crystallization behavior, and morphology.

17.1.3 Blends of Isotactic Polypropylene with LCP Reinforcement of PP with several LCPs has been attempted by Baird and coworkers (3,4) and by others (5,25,26). Several investigations have dealt with the preparation and characterization of PP/LCP blends containing different commercially available compatibilizers, such as maleic anhydride-grafted PP (PP-g-MA) (27–29) or an ethylene-based reactive terpolymer (30). The latter additives have been shown to improve the phase dispersion and the interfacial adhesion, but the enhancement of tensile modulus, tensile strength, and surface finish was generally modest (27–29). Miller and coworkers (31–33) have used an acrylic acid-functionalized PP (PPAA) and a PPAA-based graft copolymer for the compatibilization of PP/LCP blends and have observed an improvement of the tensile properties, thermal stability, and crystallinity of the fibers produced therefrom. An ethylene–glycidyl methacrylate copolymer (EGMA) (34) and a terpolymer of ethylene, ethylacrylate, and glycidyl methacrylate (EEAGMA) (35) have also been investigated as reactive compatibilizers for PP/LCP blends. The reaction between the epoxy groups of the compatibilizers and the LCP end groups has been found to lower the dimensions of the LCP domains and to improve the impact strength (34,35). However, this positive effect has been shown to be accompanied by a substantial reduction of the PP degree

510

Polyolefin Blends

of crystallinity and of the tensile modulus (34). It is well known that, among compatibilizing agents, block or graft copolymers made up of segments whose chemical structure and solubility parameters are similar to those of the polymers being blended appear best suited to the scope (36). These compatibilizers can migrate to the interphase and reduce the interfacial energy between matrix and dispersed phase, thus causing a reduction of the minor phase dimensions and a stabilization of polymer blend morphology. New graft copolymers consisting of PP backbones and LCP branches, to be used as compatibilizing agents for PP/LCP blends, have been synthesized (37). The PP-g-LCP copolymers have been prepared by polycondensation of the monomers of a semiflexible liquid crystalline polyester (SBH 1:1:2), carried out in the presence of appropriate amounts of a commercial acrylic acid-functionalized polypropylene. The polycondensation products, referred to as COPP50 and COPP70, having calculated PPAA concentrations of 50 and 70 wt%, respectively, have been fractionated with boiling toluene and xylene, and the soluble and insoluble fractions have been characterized by Fourier transform infrared and nuclear magnetic resonance spectroscopy, scanning electron microscopy, differential scanning calorimetry, and X-ray diffraction. All analytical characterizations have concordantly shown that the products are formed by intricate mixtures of unreacted PPAA and SBH together with PP-g-SBH copolymers of different composition. An example for the analytical characterization of the samples has been demonstrated by the solid-state 13C NMR spectra (Fig. 17.6) and the average values of proton spin-lattice relaxation time of the materials (Table 17.2) (37). The observed behavior indicates that the grafting reaction has actually occurred between the two components and that these components are bonded chemically in the copolymers and form separated domains with linear dimensions larger than several hundred angstroms (37). Exploratory experiments carried out by adding small amounts of COPP50 or COPP70 into binary mixtures of isotactic polypropylene (iPP) and SBH while blending have demonstrated an appreciable improvement of the dispersion of the minor LCP phase (Fig. 17.7), as well as an increase in the crystallization rate of iPP. The uncompatibilized blend is characterized by a dispersed SBH phase appearing as large (20–70 mm) unequal droplets whose smooth surface denounces the complete incompatibility of the two phases. As it is clearly demonstrated, the addition of either COPP always leads to a significantly finer dispersion of the LCP droplets (5–7 mm). COPP50 seems to develop its maximum efficiency at concentration of about 5%, whereas higher concentrations are needed for COPP70 (37). The calorimetric characteristics of iPP phase of the uncompatibilized blends show that the presence of the SBH dispersed phase leads to a slight increase of the PP temperature of crystallization (Fig. 17.8) (37,38). This result can be interpreted by a slightly increased nucleation rate of PP phase in the presence of SBH dispersed particles. As seen in Fig. 17.8, iPP temperature of crystallization increases drastically in the presence of COPP70 and COPP50. The analysis of the X-ray patterns (38) (Fig. 17.9a and b) and calculated parameters (dhkl values, degree of crystallinity, crystallite size, and intensity ratio) allows the assumption that PP segments of


Figure 17.6

511

13

C NMR spectra of mechanical blends MIXP70 and MIXP50, the polycondensation products COPP70 and COPP50, and the residue of COPP50 (RXCP50). (From Reference 37 with permission from John Wiley & Sons, Inc.)

PP-g-SBH copolymers cocrystallize completely with bulk iPP, which increases PP crystallinity degree; part of the SBH component (SBH grafts of PP-g-SBH copolymers and bulk SBH) enters the mutual amorphous phase of the blends, leading to a decrease of the intensity of the SBH peak. That means each part of the compatibilizer is miscible with the corresponding bulk phase of the blend. The compatibilized iPP/LCP blends display improved crystallization kinetics, enhanced degree of crystallinity, and improved interphase adhesion (37,38). Consequently, an improvement of the mechanical characteristics should be expected for these blends. In fact, the investigation of the Vickers microhardness of uncompatibilized and

512

Polyolefin Blends Table 17.2 Average Values of Proton Spin-Lattice Relaxation Time for the PPAA and SBH Components of Different Materialsa. Sample

TH1 (s) (SBH carbons)

PPAA SBH MIXP70 MIXP50 COPP70 COPP50 RXCP50

— 0.75 1.15 1.15 0.71 0.31 0.30

a

TH1 (s) (PPAA carbons) 0.9 — 0.95 0.95 0.92 0.89 0.79


compatibilized iPP/SBH blends shows that the addition of 2.5, 5, or 10 wt% of PP-gSBH compatibilizer leads to an increase in the microhardness of the blends (39). The strong positive deviation of the experimental hardness values from the additive ones has been interpreted by the increase in the degree of crystallinity of PP crystals and by the decrease in the surface free energy of PP crystals and of SBH LC domains (Table 17.3). Moreover, it is well known that there is a correlation between the microhardness (H) and the modulus of elasticity (E) (40). On the grounds of the mechanism of hardness indentation, namely that the loading cycle is elastic–plastic and the unloading cycle is elastic, a plot has been derived (41), which is an indicator

Figure 17.7 SEM micrographs of the fracture surfaces of (a) 80/20 iPP/SBH blend, (b) the same with 2.5% COPP50, (c) the same with 5% COPP50, and (d) the same with 10% COPP50. (From Reference 37 with permission from John Wiley & Sons, Inc.)

513

120

0.65

115

0.60

110

0.55

105

Crystallinity

o

Tc, C


0.50 COPP70 COPP50

100

0.45 0

2

4 6 Compatibilizer, wt %

8

10

Figure 17.8 Dependence of iPP crystallization temperature (Tc ) and degree of crystallinity on the content of the two compatibilizers in iPP/SBH 80/20 w/w blends. (From Reference 38 with permission from John Wiley & Sons, Inc.)

of the position taken by different materials in the elastic–plastic spectrum. According to the H/E values obtained for the blends, the compatibilized blend iPP/COPP50/SBH 77.5/5/17.5 (H ¼ 94 MPa, E ¼ 950 MPa, H=E ¼ 0:100) is positioned in the elastic– plastic spectrum much closer to elastic material than the uncompatibilized blend iPP/ SBH 80/20 (H ¼ 78 MPa, E ¼ 890 MPa, H=E ¼ 0:087). This is in perfect agreement with the values of elongation at break, which for the uncompatibilized blend is 9%, while for the compatibilized blend this value is 15%. These results confirm the compatibilizing efficiency of COPP, leading to enhanced adhesion between the blend components.

17.2 CRYSTALLIZATION BEHAVIOR OF BLENDS OF POLYOLEFINS WITH A SEMIFLEXIBLE LIQUID CRYSTALLINE POLYMER 17.2.1 Isothermal and Nonisothermal Crystallization of Blends of Linear Low Density Polyethylene with a Semiflexible Liquid Crystalline Polymer The LCP dispersed phase may sometimes act as a nucleating agent, thereby enhancing the rate and/or the extent of crystallization of crystallizable matrices such as

514

Polyolefin Blends

Figure 17.9 (a) X-ray diffraction patterns of uncompatibilized iPP/SBH 80/20 w/w blend and of the blends compatibilized with 2.5, 5, and 10 wt% COPP50 (data from Reference 38). (b) Fitted profile of iPP/SBH 80/20 w/w X-ray pattern. (From Reference 38 with permission from John Wiley & Sons, Inc.)


515

Table 17.3 The Experimental Microhardness Values (Hexp ), the Calculated Microhardness Values According to the Additivity Law (Hcal ), PP Degree of Crystallinity Derived from WAXS (a), the Experimental Values of Melting Temperatures (Tm ), the Values of b Parameter, and the Values of the Surface Free Energy of Neat PP and PP Component in Uncompatibilized and Compatibilized Blends. PP-g-SBH copolymer

COPP70 COPP70 COPP50 COPP70 COPP70 COPP70 COPP50 COPP50 COPP50 a

Composition of blends iPP/COPP/SBH 100/0/0 0/0/100 90/0/10 88.2/2.5/9.2 86.5/5/8.5 87.5/5/7.5 80/0/20 78.2/2.5/19.2 76.5/5/18.5 73/10/17 78.7/2.5/18.7 77.5/5/17.5 75/10/15

Hcal , Hexp , MPa MPa 89 — 85 87 88 89 76 — — — 79 82 82

89 27 83 92 96 95 78 89 92 92 88 94 89

$

a

Tm , K

b , s PP , ˚ erg cm2 A

0.50 — 0.52 0.54 0.55 0.56 0.49 — — — 0.53 0.55 0.55

431.0 — 430.6 433.4 433.4 434.3 430.8 431.8 432.8 431.2 433.3 432.2 432.8

8.1 — 8.1 7.5 7.5 7.1 8.1 7.8 7.6 8.0 7.5 7.7 7.6

79.6 — 79.6 73.5 73.5 70.0 79.6 77.2 74.7 78.4 73.5 75.9 74.7

From Reference 39 with permission from Springer.

poly(ethylene terephthalate) (42–45), poly(butylene terephthalate) (46, 47), aliphatic polyamides (48), poly(phenylene sulfide) (49,50), and so on. In other systems, a depression of the crystallization rate induced by the added LCP was found (51). These effects depend critically on the interactions between the blend components in the molten state. The crystallization kinetics of LLDPE/SBH blends has been investigated under nonisothermal and isothermal conditions, and the results have been discussed with reference to the blends morphology (52). A very good dispersion of minute SBH droplets has been observed for LCP concentrations up to 10%, whereas for the 80/20 LLDPE/SBH blend, extensive droplet coalescence with formation of large LCP domains has been found to occur. It has been demonstrated that the morphology of the blends influences their crystallization behavior. In fact, both nonisothermal and isothermal studies have shown that the dispersed LCP phase plays a nucleating role whose efficiency is maximum for the 90/10 blend. Apparently, the nucleating effect depends on the interfacial surface available. The latter increases with the SBH concentration in the 0–10% range, and then decreases when the droplet coalescence prevails. The nonisothermal crystallization rate has been evaluated by the crystallization rate coefficient (CRC) proposed by Khanna (53). If the cooling rate is plotted against the crystallization temperature, the slopes of the straight lines drawn through the experimental points can be taken as CRCs (Fig. 17.10) (52). While the CRC value of neat LLDPE is 87 h1 , the CRCs of the blends are all about 110 h1 , independent of the LCP concentration. The overall nonisothermal crystallization kinetics has been studied by Harnisch and Muschik’s method (54). The Avrami

516

Polyolefin Blends 25

LLDPE 95/5 90/10 80/20

15

o

Cooling rate, C min–1

20

10

5

0 80

85

90

95 100 105 110 115 120 125 130 o

T cr, C Figure 17.10 Dependence of the cooling rate on the crystallization temperature of LLDPE/SBH blends. (From Reference 52 with permission from Springer.)

exponents n have been determined according to the following equation, which is valid at T ¼ Tc : h i y1 y2 ln 1x2 ln 1 x1 n¼1þ ð17:1Þ f2 ln f1 where xi is the crystalline fraction calculated by integration of the DSC endotherm, yi is the derivative of xi , and fi is the cooling rate. The values of the Avrami exponents (n ¼ 2:5–2.6) suggest a spherulitic three-dimensional growth process controlled by heterogeneous nucleation (52). The isothermal crystallization kinetics has been analyzed by means of the Avrami equation: Xt ¼ 1 expðKn tn Þ

ð17:2Þ


517

where Xt is the fractional crystallization occurred at time t, Kn is the kinetic constant, and n is the Avrami exponent, which depends on the type of nucleation and on the crystal growth geometry. The Avrami exponents of the samples are close to 3, which is an indication of heterogeneous nucleation with threedimensional growth, and drop to 1–2 during secondary crystallization (52). The values of the Avrami exponents found from isothermal measurements are slightly higher than those determined under nonisothermal conditions. This has been attributed to the formation of sheaflike superstructures not fully developed into spherulites, and to spherulites impingement effects (55) during nonisothermal crystallization. The interpretation of the nonisothermal and isothermal data in terms of the Avrami exponents indicated that the nucleation mechanism is not altered by the presence of a dispersed SBH phase. The values of the kinetic constants for the blends especially that with 10% SBH are appreciably higher than that measured for neat LLDPE (52). The nucleation effect of SBH dispersed phase has been confirmed by the decrease of LLDPE spherulite’s size in the blends by 50% (52). The average dimensions of the spherulites were determined from Hv patterns of small-angle light scattering by means of the equation (56) Rsph ¼

2:05l np sin Qm

ð17:3Þ

where Rsph is the spherulite radius, l=n is the wavelength of the light in a medium of refractive index n, and Qm is the angle of the incident and scattered beams corresponding to the maximum pattern intensity (52). The determination of the thermodynamic equilibrium melting point (Tm ) of the crystallizable component of the blends by means of the dependence of melting temperatures of isothermally crystallized samples (Tm ) on the temperatures of crystallization (Tc ) has shown that Tm ¼ 130 C for neat LLDPE and LLDPE/SBH blends (52).

17.2.2 Crystallization Behavior of PE-g-LCP Copolymers The valuable properties of block and graft copolymer compatibilizers, containing crystalline components, come from their ability to undergo microphase separation (57, 58). There is no simple relation between the crystallization characteristics of the copolymer components and their molecular structure parameters. But, it is well known that in the block or graft copolymers the transition temperatures and enthalpies of the crystallizable polymer segments decrease in comparison to those of the corresponding homopolymer (59). The nucleation and crystal growth of the corresponding phases can be hindered since the components of the copolymers are covalently bonded (60–62). The crystallization kinetics of PE-g-SBH copolymers obtained by polycondensation (COP) or reactive blending (COP120) has been investigated under

518

Polyolefin Blends 1

2

2 1

2

1

1

0

ln[y/(1– x)]

–1

MIX

COP120

COP

–2

–3

–4

–5 108

110

112

114

116

118

120

122

124

o

Temperature, C Figure 17.11 The Harnisch plots of MIX, COP120, and COP. Curve 1—2.5 C min1 cooling rate; curve 2—5 C min1 cooling rate. (From Reference 63 with permission from Springer.)

nonisothermal conditions (63). The crystallization temperature (Tc ) of the PE component of COP and COP120 decreases steadily upon increasing the concentration of the SBH grafts (63). It was found that COP120 crystallizes at slightly higher Tc than COP and at a higher rate, confirmed by the determination of the crystallization rate coefficients (63). The results have been interpreted by the fact that the PE crystallizable segments and SBH grafts of COP120 are longer than those of COP. The overall nonisothermal crystallization kinetics has been studied by the Harnisch and Muschik equation (54). The Harnisch plots of MIX, COP120, and COP are presented in Fig. 17.11. Values of n ¼ 3:3–3.4 were obtained for the PE component of MIX, COP120, and COP (63). These results show that the mechanism of the crystallization of PE phase does not change; it changes (n ¼ 1:5–1.6) only when the SBH content overruns about 50%, due to the decrease of both nucleation and crystal growth rates. The decrease of the crystal growth rate of the copolymers in comparison to that of the corresponding homopolymers could be explained by the lower mobility of the crystallizable segments of the copolymers (64). The results have shown that the crystallization behavior and morphology of COP120 are more similar to those of the neat components of PE and SBH. This explains the fact that COP120 is more effective as a compatibilizer for PE/SBH blends than COP.


519

17.2.3 Effect of PP-g-LCP Compatibilizer on the Morphology and Crystallization of PP/LCP Blends Concerning compatibilized blends, the interfacial behavior of the compatibilizer has an important effect upon the crystallization of the blended components as it was shown for crystalline/crystalline polymer blends (60,65–67) and for crystalline polymer/LCP blends (32,37,38,68). For the latter blends, the enhanced phase interactions and improved interfacial adhesion could increase the abovementioned nucleation activity of the LCP toward the crystallizable matrices. In the particular case of using polyolefin-g-LCP copolymer compatibilizer, the crystallization of the two blend phases might have a reverse effect upon the compatibilizing activity. Moreover, the miscibility (69,70) and/or cocrystallization (60) between the bulk homopolymers and corresponding segments of the copolymer could strongly influence the crystallization behavior of the blends. The nonisothermal crystallization behavior and morphology of iPP/SBH blends compatibilized with 2.5, 5, or 10 wt% PP-g-SBH copolymers have been investigated by differential scanning calorimetry and optical microscopy (38). The migration of the compatibilizer to the interphase leads to a reduction of the interfacial tension and to a strong reduction of the dimensions of dispersed SBH phase. This leads to an increase of the iPP nucleation rate (or crystallization rate) (Fig. 17.8) and to a decrease of the iPP spherulite dimensions (Fig. 17.12), but the iPP crystal growth mechanism remains unchanged (38). The effect of the compatibilization leads to an increase of the iPP degree of crystallinity (Fig 17.8).

17.2.4 Isothermal Crystallization Kinetics of Compatibilized Blends of Polyolefins with a Semiflexible LCP The crystallization behavior and kinetics under isothermal conditions of iPP/SBH and HDPE/SBH blends, compatibilized with PP-g-SBH and PE-g-SBH copolymers, respectively, have been investigated (71). It has been established that the LCP dispersed phase in the blends plays a nucleation role for the polyolefin matrix crystallization. This effect is more pronounced in the polypropylene matrix than in the polyethylene matrix, due to the lower crystallization rate of the former. The addition of PP-g-SBH copolymers (2.5–10 wt%) to 90/10 and 80/20 iPP/SBH blends provokes a drastic increase of the overall crystallization rate of the iPP matrix and of the degree of crystallinity. Table 17.4 collects the isothermal crystallization parameters for uncompatibilized and compatibilized iPP/SBH blends (71). On the contrary, the addition of PE-g-SBH copolymers (COP or COP120) (2.5–8 wt%) to 80/20 HDPE/SBH blends almost does not change or only slightly decreases the PE overall crystallization rate (71). This is due to some difference in the compatibilization mechanism and efficiency of both types of graft copolymers (PP-g-SBH and PEg-SBH). The two polyolefin-g-SBH copolymers migrate to blend interfaces and

520

Polyolefin Blends

Figure 17.12 Optical micrographs, taken at room temperature after cooling of the samples, of iPP/ SBH 90/10 w/w blend (a) and of compatibilized blends with 5 wt% COPP50 (b) or with 5 wt% COPP70 (c). Magnification 200. (From Reference 38 with permission from John Wiley & Sons, Inc.)


521

Table 17.4 Isothermal Crystallization Parameters for Uncompatibilized and Compatibilized iPP/SBH Blends. PP-g-SBH copolymer

Composition of blends iPP/COPP/SBH

Tc ( C)

t0:5 (s)

n

100/0/0

127 124 121 130 127 124 130 127 124 130 127 124 130 127 124 130 127 124 130 127 124 130 127 124 130 127 124 130 127 124 130 127 124 130 127 124

574 293 153 583 245 144 91 46 29 57 47 23 93 51 32 190 118 56 78 61 35 82 46 26 74 39 26 108 53 30 114 65 32 86 47 28

2.1 2.1 2.1 2.4 2.3 2.1 2.2 2.2 2.1 2.1 2.1 2.0 2.5 2.4 2.1 2.3 2.3 2.2 2.2 2.2 2.0 2.2 2.0 2.1 2.2 2.1 2.0 2.3 2.4 2.3 2.1 2.1 2.1 2.4 2.4 2.1

90/0/10

COPP70

88.2/2.5/9.2

COPP70

86.5/5/8.5

COPP50

87.5/5/7.5

80/0/20

COPP70

78.2/2.5/19.2

COPP70

76.5/5/18.5

COPP70

73/10/17

COPP50

78.7/2.5/18.7

COPP50

77.5/5/17.5

COPP50

75/10/15

a

Kn (sn ) 1:2 106 4:0 106 1:1 105 1:2 107 3:1 106 2:2 105 3:0 105 1:6 104 7:2 104 1:6 104 3:2 104 1:4 103 9:7 106 5:4 105 4:7 104 5:5 106 1:1 105 1:1 104 8:0 105 2:0 104 5:4 104 4:7 105 2:9 104 7:6 104 6:3 105 3:5 104 1:2 103 1:2 105 4:3 105 2:5 104 3:3 105 1:1 104 6:8 104 1:4 105 7:8 105 6:6 104


miscibility between SBH grafts and bulk SBH is realized. But while there is a complete cocrystallization between PP segments of PP-g-SBH copolymers and bulk iPP, only some interactions occur between PE segments of PE-g-SBH copolymers and bulk HDPE (71).

522

Polyolefin Blends

17.2.5 Crystallization and Morphology of Fibers Prepared from Compatibilized Blends of Polyethylene with a Liquid Crystalline Polymer The previous results have shown that the addition of PE-g-SBH copolymer compatibilizer to HDPE/SBH blends in a nonoriented state does not lead to an increase of the crystallization rate (71). However, the overall PE crystallization rate under

Figure 17.13 Optical micrographs of the samples: (a, b) uncompatibilized fiber of the blend high molar mass HDPE/SBH; (c, d) compatibilized with 4% Escor; (e–h) compatibilized with 4% COP-AA. Parallel polaroids (a, c, e, g) and crossed polaroids (b, d, f, h). The width of the micrographs corresponds to 150 mm. (From Reference 72 with permission from John Wiley & Sons, Inc.)


523

nonisothermal and isothermal conditions increases for fibers prepared by melt spinning of compatibilized HDPE/SBH blends (72). The dispersed LCP phase in the latter fibers appears as fibrils with high aspect ratio due to the enhanced interfacial adhesion in the presence of the COP-AA (Figs. 17.5 and 17.13). As seen (Fig. 17.13e–h) the LCP domains in the blend fiber compatibilized with 4% COP-AA (draw ratio 28) appear as long SBH fibrils with high aspect ratio (10–50). The results have been interpreted by an increase in the interfacial area between the SBH fibrils and the matrix phase, which enlarges the number of heterogeneous nuclei and increases the PE nucleation rate. The results from isothermal crystallization experiments of lower molar mass HDPE/SBH blends and the analyses of the kinetic parameters by Avrami equation have shown that the dispersed LCP fibrils have a nucleating role in the presence of the compatibilizers in good agreement with nonisothermal crystallization data (72).

17.3

CONCLUSIONS

The investigation of the compatibilization and crystallization of blends of polyolefins with a semiflexible LCP leads to the following conclusions: the compatibilization of polyolefin/LCP blends has been realized successfully by the addition of ad hoc synthesized polyolefin-g-LCP copolymers. The compatibilization results into materials, characterized by a stabilized morphology, improved crystallization kinetics under nonisothermal and isothermal conditions, and enhanced mechanical properties. Moreover, polyolefin processability has been enhanced by the addition of LCP, even in the presence of compatibilizers. New high quality materials with improved processability have been produced by technologies, which are economic, friendly to the environment, and socially acceptable.

ACKNOWLEDGMENT The author is grateful to Professor P.L. Magagnini from University of Pisa, Italy, for the years of fruitful collaboration in the field of the herein presented studies.

NOMENCLATURE COP COP15, COP60, COP120 COP-AA COPP50, COPP70

PE-g-SBH copolymer prepared by polycondensation of SBH monomers in the presence of PEox PE-g-SBH copolymers prepared by reactive blending of PEox with SBH PE-g-SBH copolymer prepared by reactive blending of an acrylic acid-functionalized PE with SBH PP-g-SBH copolymers prepared by polycondensation of the monomers of an SBH, carried out in the presence of an acrylic acid-functionalized PP

524

Polyolefin Blends

CRC DSC DTG E fi FTIR H HDPE Kn LCP LLDPE l=n MIX n NMR PE-g-LCP PEox iPP PP-g-LCP Qm Rsph SBH

SEM t Tc TG Tm Tm xi yi Xt

Crystallization rate coefficient Differential scanning calorimetry Differential thermogravimetry Modulus of elasticity Cooling rate Fourier transform infrared spectroscopy Microhardness High density polyethylene Kinetic constant Liquid crystalline polymer Linear low density polyethylene Wavelength of the light in a medium of refractive index n Physical blend between PEox and SBH Avrami exponent Nuclear magnetic resonance spectroscopy Graft copolymer between polyethylene and LCP Oxidized low molar mass PE sample containing free carboxylic groups Isotactic polypropylene Graft copolymer between polypropylene and LCP Angle of the incident and scattered beams corresponding to the maximum pattern intensity Spherulite radius Semiflexible liquid crystalline polymer, synthesized from sebacic acid (S), 4,40 -dihydroxybiphenyl (B), and 4-hydroxybenzoic acid (H) in the mole ratio 1:1:2 Scanning electron microscopy Time Temperature of crystallization Thermogravimetry Melting temperature Thermodynamic equilibrium melting temperature Crystalline fraction Derivative of xi Fractional crystallization

REFERENCES 1. F. P. La Mantia (ed.), Thermotropic Liquid Crystal Polymer Blends, Technomic, Lancaster, 1993. 2. G. Kiss, Polym. Eng. Sci., 27, 410, (1987). 3. D. Done, A. M. Sukhadia, A. Datta, and D. G. Baird, SPE Technol. Paper, 48, 1857 (1990). 4. A. Datta, A. M. Sukhadia, J. P. Desouza, and D. G. Baird, SPE Technol. Paper, 49, 913 (1991).


525

5. Y. Yongcheng, F. P. La Mantia, A. Valenza, V. Citta, U. Pedretti, and A. Roggero, Eur. Polym. J., 27, 723 (1991). 6. T. C. Hsu, A. M. Lichkus, and I. R. Harrison, Polym. Eng. Sci., 33, 860 (1993). 7. T. Harada, K. Tomari, A. Hamamoto, S. Tonogai, K. Sakaura, S. Nagai, and K. Yamaoka, SPE Antec92, 376 (1992). 8. A. M. Lichkus and I. R. Harrison, SPE Antec-92, 2257 (1992). 9. T. C. Hsu and I. R. Harrison, SPE-93, 1183 (1993). 10. U. Pedretti, A. Roggero, F.P, La Mantia, and P. L. Magagnini, SPE Antec-93, 1706 (1993). 11. E. G. Fernandes, I. Giolito, and E. Chilellini, Thermochim. Acta, 235, 67 (1994). 12. F. P. La Mantia, C. Geraci, M. Vinci, U. Pedretti, A. Roggero, L. I. Minkova, and P. L. Magagnini, J. Appl. Polym. Sci., 58, 911 (1995) 13. L. Minkova and P. L. Magagnini, Colloid Polym. Sci., 274, 34 (1996). 14. P. L. Magagnini, M. Paci, L. I. Minkova, T. Miteva, D. Sek, J. Grobelny, and B. Kaczmarczyk, J. Appl. Polym. Sci., 60, 1665 (1996). 15. L. I. Minkova, T. Miteva, D. Sek, B. Kaczmarczyk, P. L. Magagnini, M. Paci, F. P. La Mantia, and R. Scaffaro, J. Appl. Polym. Sci., 62, 1613 (1996). 16. C. U. Ko, G. L. Wilkes, and C. P. Wong, J. Appl. Polym. Sci., 37, 3063 (1989). 17. M. Pracella, D. Dainelli, G. C. Galli, and E. Chiellini, Makromol. Chem., 187, 2387 (1986). 18. B. Wunderlich (ed.), Macromolecular Physics, Vol. 1, Academic Press, New York, 1973. 19. Ts. Miteva and L. Minkova, Colloid Polym. Sci., 275, 38 (1997). 20. F. P. La Mantia, R. Scaffaro, P. L. Magagnini, M. Paci, C. Chiezzi, D. Sek, L. I. Minkova, and Ts. Miteva, Polym. Eng. Sci., 37, 1164 (1997). 21. F. P. La Mantia, R. Scaffaro, P. L. Magagnini, M. Paci, L. I. Minkova, and Ts. Miteva, J. Appl. Polym. Sci., 71, 603 (1999). 22. Y. Lyatskaya, D. Gersappe, N. A. Gross, and A. Balazs, J. Phys. Chem., 100, 1449 (1996). 23. L. I. Minkova, M. Velcheva, M. Paci, P. L. Magagnini, F. P. La Mantia, and D. Sek, J. Appl. Polym. Sci., 73, 2069 (1999). 24. L. Minkova and P. L. Magagnini, Macromol. Chem. Phys., 200, 2551 (1999). 25. C. Federic, G. Attalla, and L. Chapoy, Eur. Patent 0,340,655 A2, (1989). 26. S. C. Tjong, S. L. Liu, and R. K. Y. Li , J. Mater. Sci., 31, 479 (1996). 27. A. Datta, H. H. Chen, and D. G. Baird, Polymer, 34, 759 (1993). 28. A. Datta and D. G. Baird, Polymer, 36, 505 (1995). 29. H. J. O’Donnel and D. G. Baird, Polymer, 36, 3113 (1995). 30. M. T. Heino and V. Seppala, J. Appl. Polym. Sci., 8, 1677 (1993). 31. M. M. Miller, D. L. Brydon, J. M. G. Cowie, and R. Mather, Macromol. Rapid Commun., 15, 857 (1994). 32. M. M. Miller, J. M. G. Cowie, J. G. Tait, D. L. Brydon, and R. Mather, Polymer, 36, 3107 (1995). 33. Y. Qin, M. M. Miller, D. L. Brydon, J. M. G. Cowie, R. R. Mather, and H. Wardman, in: Liquid Crystalline Polymer Systems, Technological Advances, A. I. Isayev, T. Kyu, S. Z. D. Cheng (eds.), American Chemical Society, Washington, DC, 1996, p. 98. 34. Y. P. Chiou, K. C. Chiou, and F. C. Chang, Polymer, 37, 4099 (1996). 35. R. M. Holsti-Miettinen, M. T. Heino, and J. V. Seppala, J. Appl. Polym. Sci., 57, 573 (1995). 36. X. Jin and W. Li, J. Macromol. Sci. Rev. Macromol. Chem. Phys., C35, 1 (1995). 37. P. L. Magagnini, M. Pracella, L. I. Minkova, Ts. Miteva, D. Sek, J. Grobelny, F. P. La Mantia, and R. Scaffaro, J. Appl. Polym. Sci., 69, 391 (1998). 38. Ts. Miteva and L. Minkova, Macromol. Chem. Phys., 199, 597 (1998). 39. L. Minkova, H. Yordanov, G. Zamfirova, and P. L. Magagnini, Colloid Polym. Sci., 280, 358 (2002).

526

Polyolefin Blends

40. F. J. Balta Calleja, Adv. Polym. Sci., 66, 117 (1985). 41. B. Lawn and V. R. Howes, J. Mater. Sci., 16, 2745 (1981). 42. E. G. Joseph, G. L. Wilkes, and D. G. Baird, Am. Chem. Soc. Div. Polym. Chem. Polym. Prepr., 24, 304 (1983). 43. E. G. Joseph, G. L. Wilkes, and D. G. Baird, in: Polymer Liquid Crystals, A. Blumstein (ed.), Plenum Press, New York, 1984. 44. S. K. Battacharya, A. Tendolkar, and A. Misra, Mol. Cryst. Liq. Cryst., 153, 501 (1987). 45. S. K. Sharma, A. Tendolkar, and A. Misra, Mol. Cryst. Liq. Cryst., 157, 597 (1988). 46. M. Kimura, R. S. Porter, and G. Salee, J. Polym. Sci. Polym. Phys. Ed., 21, 376 (1983). 47. M. Paci, C. Barone, and P. L. Magagnini, J. Polym. Sci. Polym. Phys. Ed., 25, 1595 (1987). 48. M. Takayanagi, T. Ogata, M. Morikawa, and T. Kai, J. Macromol. Sci. Phys., B27, 591 (1980). 49. L. I. Minkova, M. Paci, M. Pracella, and P. L. Magagnini, Polym. Eng. Sci., 32, 57 (1992). 50. A. Valenza, F. P. La Mantia, L. I. Minkova, S. De Petris, M. Paci, and P. L. Magagnini, J. Appl. Polym. Sci., 52, 1653 (1994). 51. M. Pracella, E. Chiellini, and D. Dainelli, Makromol. Chem., 190, 175 (1989). 52. L. Minkova and P. L. Magagnini, Colloid Polym. Sci., 274, 34 (1996). 53. Y. P. Khanna, Polym. Eng. Sci., 30, 1615 (1990). 54. K. Harnisch and H. Muschik, Colloid Polym. Sci., 261, 908 (1983). 55. L. C. Lopez and G. L. Wilkes Polymer, 30, 882 (1989). 56. P. H. Geil, Polymer Single Crystals, Wiley, New York, 1968. 57. J. A. Manson and L. H. Sperling, Polymer Blends and Composites, Plenum Press, New York, 1976. 58. A. Noshay and J. E. McGrath, Block Copolymers—Overview and Critical Survey, Academic Press, New York, 1977. 59. B. F. Mathot and B. F. Vincent, Calorimetry and Thermal Analysis of Polymers, Hanser Publishers, Munich, 1994. 60. T. Tang and B. Huang, J. Polym. Sci., B 32, 1991 (1994). 61. S. Datta and D. Lohse, Macromolecules, 26, 2064 (1993). 62. P. Jannasch and B. Wesslen, J. Polym. Sci., A33, 1465 (1995). 63. L. Minkova and Ts. Miteva, P. L. Magagnini, Colloid Polym. Sci., 275, 520 (1997). 64. B. Wunderlich, Macromolecular Physics, Vol. 2, Mir, Moscow, 1979, p. 344. 65. V. Flaris, A. Wasiak, and W. Wenig, J. Mater. Sci., 28, 1685 (1993). 66. U. Plawky and W. Wenig, J. Mater. Sci. Lett., 13, 863 (1994). 67. T. Tang, H. Li, and B. Nuang, Macromol. Chem. Phys., 195, 2931 (1994). 68. G. Poli, M. Paci, P. L. Magagnini, R. Scaffaro, and F. P. La Mantia, Polym. Eng. Sci., 36, 1244 (1996). 69. J Huang and H. Marand, Macromolecules, 30, 1069 (1997). 70. L. Z. Liu, W. Xu, H. Li, F. Su, and E. Zhou, Macromolecules, 30, 1363 (1997). 71. Ts. Miteva, L. Minkova, and P. Magagnini, Macromol. Chem. Phys., 199, 1519 (1998). 72. L. Minkova, M. Velcheva, and P. Magagnini, Macromol. Mater. Eng., 280/281, 7 (2000).

Chapter

18

Functionalized Polyolefins and Aliphatic Polyamide Blends: Interphase Interactions, Rheology, and High Elastic Properties of Melts Boleslaw Jurkowski1 and Stepan S. Pesetskii2

18.1 INTRODUCTION Blends of thermoplastic polymers combine best characteristics of their components in a single material, eliminate demerits, and often possess a set of properties unattainable for homopolymers while the range of new products widens quickly and economically favorably. Polyamide (PA) blends for engineering applications appeared much earlier in 1948, when composites were developed based on PA66 and polyvinyl acetate (2). However, a wide-range development of new materials by means of blending existing homopolymers and copolymers was initiated in the 1970s (1–5); beginning with the work of Ide and Hasegawa (6), there were expanded investigations related to the technology of reactive compounding of PA blends (7). In 1975, Du Pont presented an ultra impact strength PA alloy, Zytel-ST, produced by reactive processing of PA66 with maleinated EPDM (copolymer of ethylene, propylene, and 2-ethylidene-5-norbornene) (2). Then followed investigations of 1

Division of Plastic and Rubber Processing, Institute of Material Technology, Poznan University of Technology, Piotrowo 3, 60-950 Poznan, Poland 2 Laboratory of Chemical Technology of Polymeric Composite Materials, V.A. Belyi Metal-Polymer Research Institute of National Academy of Sciences of Belarus, 32a Kirov Street, 246050 Gomel, Belarus Polyolefin Blends, Edited by Domasius Nwabunma and Thein Kyu Copyright # 2008 John Wiley & Sons, Inc.

527

528

Polyolefin Blends

PA/polyolefin (PA/PO) blends, the main advantages of these in comparison with homopolyamides being increased impact strength, reduced water absorption, improved processability, and dimensional stability at varying humidity (9–11). Considering that among aliphatic PAs PA6 and PA66 are most abundant (in EU countries they account for 92% of all PAs produced), blended systems have been developed mostly based on them. Without making use of the reactive processing technology, it would in fact be impossible to create valuable engineering products based on PA/PO blends. Because of high density of cohesive energy (23.2,23.3,21.9, and 20.3 ðJ cm3 Þ0:5 for PA6, PA66, PA610, and PA12, respectively), aliphatic PAs cannot form compatible blends with polyolefins whose density of cohesive energy is considerably lower (16.1 ðJ cm3 Þ0:5 for PE and 16.3 ðJ cm3 Þ0:5 for PP). That is why PA/PO blends are characterized by a high surface tension (14–18 mN m1 ), weak interphase adhesion, and distinct phase separation (8,9). The properties of such blends—like those of many other systems with separated phases—strongly depend on interphase interactions and phase morphology. As the same is true for properties such as viscosity, strength, and high elastic properties of the blends, which are the subject of our further consideration, here is a brief analysis of certain aspects of the PA/PO interface and contemporary methods of reactive compatibilization for materials of this type.

18.2 COMPOUNDING AND INTERPHASE PHENOMENA IN PA/PO BLENDS In order to create morphology that is stable during use, the free energy of the blend (DGbl ) must be negative (4): DGbl ¼ DHbl TDSbl

ð18:1Þ

In Equation 18.1, the role of the entropy factor is negligible (DSbl ! 0), so a negative value of DGbl can only be reached if the process of blending is exothermic (DHbl < 0). In other words, the process must proceed with heat liberation, which may be caused by specific interactions between the blend’s components. Interactions that lead to strong covalent or ionic (base–acid) binding and to relatively weak bonds like hydrogen, ion–dipole, dipole–dipole, or donor–acceptor ones are quite probable (4,14,18). As in PA/PO blends mostly Van der Waals forces are feasible, these blends are always heterogeneous and show distinct phase boundaries. The following morphologies are most typical: dispersion of one polymer within the matrix of another one and cocontinuous two-phase morphology, and depend on the character of components, viscosity of their melts at blending, composition, and conditions of blending. The majority of investigations on PA/PO blends and their other types are dedicated to formation particularities of dispersed structures in them. The particularities of continuous morphologies formed and properties of materials containing coexisting continuous phases have been investigated to a lesser degree (19). Because

Chapter 18 Functionalized Polyolefins and Aliphatic Polyamide Blends

529

Figure 18.1 Methods for preparing PA/PO compatibilized blends.

of weak interphase adhesion in PA/PO blends, domains of the dispersed phase undergo aggregation and form spherical or elliptical micron-sized particles that are dispersed within the matrix of the basic polymer. When preparing specimens for morphological studies by ultramicrotoming under liquid nitrogen, some domains are torn out of the polymer matrix, which is a consequence of low adhesion between the phases (20). The key trend of creating valuable engineering PA/PO blends is compatibilization, which is a combination of procedures intended to improve the compatibility of the components either by specific means of blending or by introducing compatibilizers into the blend. Irrespective of the means of compatibilization, the blend shows a high dispersion degree of the dispersed phase, adhesion between the phases, resistance toward coalescence, and improved processing characteristics (2,13,14). The major methods of preparing compatibilized PA blends are given in Fig. 18.1 (9). Compatibilization of PA/PO blends is based on mechanochemical reactions that take place in a polymer melt during compounding; these reactions follow the radical mechanism and result in grafted or block copolymers (1,15–18). The regimes of blending play a decisive role in the development of phase morphology. An increased shear rate of the melt leads to smaller particles of the disperse phase. The particle size is inversely proportional to the shear stress applied. The ratio of viscosities of the molten components influences dispersion of components more than changes in shear stresses. For example, in the case of PA6/PE blends extruded through a shaping die, an increase of the shear stress from 19 up to 29 kPa causes a change in the morphology of PA6 particles from a spherical to a wavelike one (21). Application of the compatibilization method by means of melt blending is not decided as yet because of inadequate understanding of and difficulty in realization of real technologies, which require complex apparatus.

530

Polyolefin Blends

The effect of compatibilization of grafted or block copolymers as controllers of interphase interaction has been well known (1,28). Marosi and Bertlan (29), for example, described the compatibilizing effect of polybutylene terephthalate– polytetramethylene oxide block copolymers on PA6/HDPE blends. The addition of a block copolymer prevents separation in the blends during processing and increases the impact strength of the material several times. According to Paul and Newman (1), even 2 wt% of an A–B-type copolymer is sufficient for compatibilization of blends of A and B polymers if its molecular weight is 104. The block copolymers reduce the surface tension and raise the adhesion between phases. The latter may rise up to 100 times against the adhesion between the components (7). Despite the efficiency of the compatibilization with the help of block copolymers, their use in commercial blends has been on a rather limited scale since no profitable ways are available to synthesize such copolymers. The present authors are unaware whether the market offers any PA/PO blends compatibilized with block or grafted copolymers prepared in advance. A more advantageous alternative is to prepare compatibilized PA/PO blends, like many other types of blends, by reactive processing that implies creation of composites with a required level of interphase interaction in situ during compounding (18). Grafted polymers, whose macromolecules contain necessary functional groups, are most often used as compatibilizers. Reactive polymers, during blending, undergo chemical reactions similar to those in low molecular weight organic substances. Reactivity of functional groups depends little on the molecular weight (18,23–25). However, it must be taken into account that steric obstacles created by the main polymer chain, as well as restricted diffusion, reduce the reaction rate. Therefore, to realize reactions leading to compatibilization in a short period of blending (usually a few minutes), the grafted polymer (compatibilizer) must contain sufficient quantity of functional groups; the reaction must be fast and selective; the conditions of mixing must ensure a maximum contact surface between the interacting components (9). Reactions such as amidization, imidization, etherification, aminolysis, amide– ester exchange, ring opening, ionic bonding, and neutralization of carboxyl groups proceed fast enough in the polymer melts at the regimes of reactive compounding of PA/PO blends (9,23,25,26). Numerous compatibilizers can be produced by grafting monomers (containing some kind of functional groups) in melt onto homopolymers and copolymers of olefins or their blends (25). Reactive extrusion (26) is a basic process for this when the twin-screw extruder is used as a reactor of continuous action (27). When developing multicomponent blends, it should be taken into consideration that the greater the number of components (n), the greater the number of interfaces (N) between them: N ¼ nðn 1Þ=2. Interfaces are potential sources of initiation of damage spots (7,28,29). Therefore, the most important strategic trend in creating multicomponent blends is the introduction into a system of, at least, one ingredient with reactive groups, which can interact with the components of the blend and lead to its compatibilization.


531

It is worth mentioning that interphase interactions in blends proceed only within mesophases whose thickness for incompatible polymers is between 2 and 50 nm depending on the thermodynamic interactions of the phases, temperature, regimes of mixing, and some other factors (7,11,29,30). The mesophase thickness depends on pffiffiffiffiffiffi the miscibility of the components and in first approximation (7) is x12 , where x12 is the parameter of thermodynamic interaction. It is one of the most significant factors that determines mechanical characteristics of the blends, especially the energy of impact failure (30,31). A mesophase of considerable thickness (>6 nm) constructed by the hierarchical method (28,29) promotes energy dissipation of impact failure and leads to microcrack healing. Intensified interphase interactions lead to mesophases of great thickness and restricted molecular mobility (29). This fact reduces the entropy; in the case of short chains, however, this fact is of a secondary importance. That is why polymer systems contain, in zones of interphase contact, extra number of shorter chain segments (32) or low molecular weight substances (32–37). This situation can be easily understood while counting the quantity of chain conformations; this quantity is higher if the chain terminal is on the interface rather than on the averaged statistical segment. It is clear, therefore, that polymer–polymer interphase boundary represents the region for most intensive interactions in systems with separated phases. The interface becomes a reaction zone. The major concentration of low molecular weight substances in the mesophase is used to promote compatibilization of PA/PO blends. For example, compatibility is improved in a PA/PE system if compatibilizer was alkylmaleic monoamide, alkylmaleic monoester (38), or stearic acid (39). Low molecular weight compatibilizers being reactive surfactants (29) allow controlling the morphology of PA/PO blends. Their favorable influence on the mechanical characteristics of the materials, however, is weaker than that of high molecular weight additives. One of the explanations can be the formation of weak mesophases if low molecular weight compatibilizers concentrate in zones of interphase contract (7,28,29). Because of this, PA/PO blends with reactive compatibilizers of macromolecular nature, so-called in situ compatibilized materials, have wider applications (7,11,40,41). As mentioned above, they are prepared by reactive processing. The properties of such materials depend much on the interactions on the interfaces occurring, as a rule, through the acid–base mechanism and depending, in their turn, on the nature and concentration of functional groups included in the compatibilizer chains, and on the nature of the olefin polymer or copolymer containing functional groups (7,29). The sulfoacidic group interacts stronger with PA than the carboxyl group (42). The anhydride group is also more reactive than the carboxyl one (7). It should also be considered that amide groups in PA are less reactive than terminal primary amino groups (24,41). However, as the concentration of amide groups is significantly higher than that of amino groups, the contribution of interactions with participation of amide groups can be quite important. The reaction between amide and anhydride groups results in PA chain opening and splitting off water molecules (41). The latter can cause hydrolytic splitting of the chains (24). That is why the region of interphase contact becomes enriched with PA chains of a lower molecular weight in comparison with the volume.

532

Polyolefin Blends

Figure 18.2 Strength dependence of adhesion bonds for film specimens of PA6, HDPE (1), and HDPE-g-MAH containing 1 wt% of grafted maleic anhydride (2) on formation time of adhesion contact at 220 C.

Reactions between anhydride and amino groups predominate in the formation of grafted copolymers that are obtained during reactive processing in the presence of anhydride-containing compatibilizers (41,43–45). Reactions between anhydride groups and PA chains are most often used in obtaining compatibilized blends of PA with PO. Their major feature is high rate in PA melt, which is important for the reactive compounding process. The consequence of such reactions is the increased interphase adhesion observed even in an early moment of contacting between the reagents. This fact can be visually illustrated using data on the kinetics of adhesional contact formed between PA6, neat HDPE, and HDPE grafted with maleic anhydride (MAH) (Fig. 18.2). It is evident that MAH grafted onto HDPE promotes a sharp rise in adhesion between the phases. The kinetics of adhesional contact formed in a polymer blend can be shown still more clearly using the comparative analysis of values of the conditional growth rate of adhesion between the phases (A ) (46,47): dA A ¼ lim t!0 dt

ð18:2Þ

where t is the time of creation of an adhesional contact. As the A value characterizes, in fact, the adhesional strength of bonding at a momentary contact between the polymers (t ! 0), this parameter does not account for the contribution of diffusion events in the adhesional interaction. The A value takes into account only energy interactions of unlike macromolecules in zones of interphase contact (47).

Q5 Chapter 18 Functionalized Polyolefins and Aliphatic Polyamide Blends

533

Figure 18.3 Effect of maleic anhydride grafted onto HDPE on Conditional growth of adhesion strength (A ) for joints of PA6/HDPE-g-MAH; temperature of adhesion contact is 220 C.

It can be seen in Fig. 18.3 that 1.2 wt% of MAH introduced into the HDPE structure raises A value more than 30 times, implying an extremely high rate of interaction between anhydride groups and PA6. Thus, grafting of MAH onto PE leads actually to momentary generation of an interphase contact zone. Similar results have been obtained by Yukioka and Inoue (49), who investigated the generation of interfaces in blends of amorphous PA and copolymer of styrene and acrylonitrile (SAN). The mesophase thickness was about 29 nm. The grafted copolymer was concentrated in this zone and its thickness was in fact independent of the time at 200 C. The extreme pattern of concentration dependence of A (Fig. 18.3) can be explained by the fact that 1.2 wt% of MAH is sufficient for bonding of all nonassociated amino and amide groups of PA6 in zones of interphase contact. A reduction in A at MAH concentration of >1.2 wt% can be caused by restriction of molecular mobility of the components and resultant deteriorated microrheological conditions of formation of an adhesional contact, as well as by accumulation—in the adhesional contact zone—of low molecular weight products of the reaction; these products give weak mesophases and weaken adhesion between the phases. In most blend systems of PA/PO compatibilized with olefin polymers and copolymers, are grafted with MAH, therefore, the concentration of the latter does not exceed 1–3 wt%. The main aim of grafting is (8) to ensure compatibility of polyolefins with polar polymers, which can be reached at the level of PO grafting of 0.1–0.5 moles of polar groups per 100 monomer units. A high rate of formation of adhesional contact between the phases in a PA/PO-gMAH system is very important from the technological viewpoint, because fast interphase interactions during compounding of materials depend little on subsequent processing of the blends.

534

Polyolefin Blends

When choosing compatibilizers for PA blends, the reactivity of amine and amide groups, as well as terminal carboxyl groups, in PA should be considered (24,50). Therefore, despite predominant use of polyamide–anhydride reaction for compatibilization, the compatibility and related engineering properties of the blends can be improved by introducing into macromolecules PO carboxyl, neutralized carboxyl (55,56), ester (24,52), oxazoline (58,59), epoxy (24,60,61), isocyanate (62), or other functional groups (24). Insufficient reactivity of some functional groups, in comparison with anhydride ones, can be compensated at the expense of their higher concentration. The concentration of carboxyl groups is sometimes increased upto 10 wt% or even upto 20 wt%.

18.3 RHEOLOGICAL AND HIGH ELASTIC PROPERTIES OF PA/PO MELTS The knowledge of rheological and high elastic properties of PA/PO melts is rather important for optimization of regimes of their processing, and for obtaining information on the flow mechanism and its influence on the morphology and engineering properties of the materials. Of particular interest are insufficiently studied blends, which show high viscosity and strength in melts. The fact is that aliphatic PAs, while having narrow molecular weight distribution and low viscosity of melt (MFI of PA6, e.g., is usually within the range of 10–25 g/10 min at T ¼ 250 C and P ¼ 21:6 N), are unsuitable for extrusion technologies. However, POs characterized by increased viscosity and strength of melts have been widely used for processing by blow extrusion. It seems promising that introduction of highly viscous PO into PA can lead to composites with rheological and high elastic properties typical of the extruded materials. In a general case, the melt viscosity of a polymer blend (hbl ) depends on the melt viscosities of the blended components and blend’s composition (1,65). The character of mutual influence between the components in a polymer blend on hbl can be described using the logarithmic rule (18): log hbl ¼ Si ui log hi

ð18:3Þ

where ui and hi are, respectively, the volume share and melt viscosity of an ith component. There are four types of the polymer blends: (i) Additive blends whose melt viscosity follows Equation 18.3, (ii) Blends with a positive deviation of hbl from Equation 18.3. These include blends with strong interphase interactions. (iii) Blends with a negative deviation from the logarithmic additivity, which is typical of incompatible components with weak interphase interactions. (iv) Blends that show both positive and negative deviations of hbl from the additive values (such a relationship is typical of materials in which structural changes take place during flowing). The viscosity of blends varies not only with the composition but with flow conditions as well, which depend on the temperature and shear rate (20,41,65). A decisive effect on PA/PO rheology is caused by the chemistry of interphase processes (41). Blends of PA with ungrafted PO, or uncompatibilized blends, are characterized by a decreased melt viscosity in comparison with the additive


535

value of hbl determined from Equation 18.3. In another study (20), this fact was clearly demonstrated for the PA6/PP blend. On increasing the shear rate g from 101 up to 103 s1 , the deviation of melt viscosity for PA6/PP blend from its additive values became stronger over the whole range of composites. This finding supports a lack of compatibilization and active interactions between the phases in PA6/PP uncompatibilized blends. The influence of a shear rate on hbl can be explained by variations in the morphology of the dispersed phase during flowing. On rising g or shear stress, spherical particles of the dispersed phase become elongated and oriented in the flow direction. At a low g , particles of the dispersed phase undergo minor deformation, and then hbl depends mainly on the size of drops and interaction between the phases. Blending of PA with grafted PO (g-PO), or compatibilization of PA/PO blends by addition of g-PO, changes the dependence pattern of hbl with component ratio and technological factors. As PA/PO compatibilized blends have a finer dispersed phase with a developed interface and show more intense interphase interactions, flowing disturbs a little the phase morphology, and hbl often becomes higher than the additive values and viscosity of any of the blend’s components (6,20,66–69). The viscosity of PA blends containing PP-g-MAH correlates with the volume of the mesophase falling at a volume unit of the blend (70). It is presumed that a decisive influence on hbl is caused by the concentration of the grafted copolymer, which becomes immobile within the mesophase formed during blending. When assessing the rheological behavior of PA/PO blends, a strong effect of shear forces upon hbl should be considered. The reason is a qualitative difference between the flow curves for PO and PA. Aliphatic PAs show an extended Newtonian plateau typical of polymers with a narrow MWD (71). PA6, for instance, can retain the Newtonian pattern of flow (72) up to a shear rate of 103 s1 . The curve describing the relationship of h versus g for PO is typical of polymers with a wide MWD: the anomaly in viscosity (h decreases with increase in the shear rate) was observed at a much lower shear rate of 102 s1 . That is why the effects of viscosity’s growth—in the case of PA6/PO compatibilized blends—manifest themselves to the utmost at relatively low shear rates, upto 102 s1 . Such shear rates are typical of extrusion of polymer materials (72). Reactive compounding of PA6 with HDPE compatibilized with a mixture of HDPE/copolymer of octene and ethylene with grafted MAH (Fig. 18.4) gives an increased viscosity of the melts especially at low shear rates (66). An increased concentration of the compatibilizer in a blend between 15 and 30 wt% causes the viscosity to rise. Unlike pure PA6, PA6/g-PE blends show 10-fold or higher increase in viscosity at low shear rates. Consequently, to create PA/PO blends with a high melt viscosity it is advisable to use fully functionalized polyolefins. It can be expected that proper dispersion in melt of PA blended with high viscosity g-PO can yield composites with satisfactory structural homogeneity and high melt viscosity. This possibility is based on the above data showing that contact between the phases in a PA/g-PO blend is created instantaneously, and the development of contact zones between the polymeric phases depends much on the degree of mechanical dispersion of the components and not on the diffusion processes taking place in the mesophase.

536

Polyolefin Blends

Figure 18.4 Effect of shear rate on melt viscosity at 275 C of PA6, PE, and PA6/PE blend compatibilized by PE-g-MAH containing 0.27 wt% of grafted MAH. Ciphers on curves stand for compatibilizer concentration in wt%, PE stands for LDPE/LHDPE (88:12) blend; LHDPE is copolymer of ethylene and octene. Reproduced from Reference (66) with permission from John Wiley & Sons. Inc.

We proved this presumption on blends of PA6 with LDPE and HDPE and on a blend of PE with ethylene–propylene rubber (EPDM). The polyolefin components were functionalized by grafting 1 wt% of itaconic acid (IA) (51,53). Functionalization and preparation of the blends were performed in a static mixer assembled onto the Brabender plastograph (51,73). Grafting with simultaneous cross-linking of PO resulted in a sharp reduction of MFI values and h of their melt. It can be seen in Table 18.1 that g-POs possess much higher melt viscosities than PA6. The viscosity of LDPE exceeds h of PA6 by two decimal orders. Ea of PA6 is twice as low as that of g-PO. The values of hMFI and Ea in Table 18.1 were calculated from MFI values (74): hMFI ¼

P rc r tc þ n rc Þ MFI

2D2p ðLc

ð18:4Þ

where P is the load; rc is the capillary radius; r is the density of polymer melt at a definition temperature of MFI, r ¼ 4m=pD2p L; tc ¼ 600, it is constant; Dp is the piston diameter; Lc is the capillary length; n is the input correction factor (its variations have not been accounted for in the calculations) (74); L is the stroke distance of the piston; and m is the volume of the extruded melt. Ea ¼

R ln k T1 T2 T2 T1

ð18:5Þ


537

Table 18.1 Properties of Blend Components. Component

Tm , C

PA6

222

LDPE

106

g-LDPE

106

HDPE

126

g-HDPE

126

g-LDPE/g-EPDM (50/50) g-HDPE/g-EPDM (50/50) PP

105*

124* 165

g-PP

165

MFI, g=10 min at T C 5.3 (230 C) 7.7 (250 C) 5.2 (230 C) 7.6 (250 C) No flow (230 C) 0.14 (250 C) 8.5 (230 C) 11.1 (250 C) No flow (230 C) 0.6 (250 C) No flow (230 C) 0.21 (250 C) No flow (230 C) 3.1 (230 C) 4.4 (250 C) 15.3 (230 C) 15.5 (250 C)

hMFI 103 , Pa s

Ea , kJ mol1

2.7

40.7

6.26

41.5

286.8 4.26

98.2 29.2

86.9

93.0

135.8

91.0

126.5 8.11

90.3 38.3

2.19

2.8

Notes: Tm is the melting temperature; r is the density; hMFI and Ea are the melt viscosity at T ¼ 250 C and the effective activation energy of viscous flow evaluated from MFI values; asterisks denote values of Tm for the PE phase.

where R is the universal gas constant; T1 and T2 are the temperatures at which MFI is determined (T2 > T1 ; P ¼ constant); and k ¼ MFI1 =MFI2 (at P1 ¼ P2 ; T1 < T2 ; the residence time in the melting cylinder being constant). Determined from Equation 18.4, hMFI is non-Newtonian. Despite a wide range of g values observed during determination of MFI (approximately between 0.1 and 15 s1 ) (75), it does not fall at the region of intensive development of viscosity anomaly for PO, the more so for PA. It is clear, therefore, that variations in g — during determination of MFI—for different composites cannot be the cause of differences in the values of this property. The concentration dependences of MFI are represented in Fig. 18.5. For all types of the blends there is a sharp decrease in MFI at g-PO 30 wt%. For blends containing 30–40 wt% of g-PO, the value of MFI is between 0.2 and 0.8 g/10 min, which is 10–26 times lower than that of pure PA6; the level of values corresponds with the requirements imposed on extrusion-processed materials (74). It is of interest that irrespective of g-PO type, at a concentration of 30–40 wt%, the MFI values of PA blends are quite similar despite a great difference in melt viscosities of g-POs used for blend preparation (Table 18.1). This can probably be explained by the fact that the continuous phase, during melt flowing of a blend, is formed by lesser viscous PA6 (76), which dominates in the blend. The decisive influence on the flow development of such blends comes from interphase interactions that are alike for all

538

Polyolefin Blends

Figure 18.5 Concentration dependence of MFI for binary PA6/g-PE (a) and ternary PA6/(g-PE/gEPDM) (b) blends; P ¼ 21:6 N; T ¼ 250 C.

types of the blends as all of g-POs used have been functionalized similarly. The higher the g-PO concentration, the more intensive are interactions between the phases and the more the high viscosity g-PO phase gets involved in the flow process, which results in a sharp rise of the total melt viscosity of the blend. PA6 and PA6/g-PO melt viscosities (Table 18.2) varied with shear stresses acting within melts during their movement through a capillary. The studies have been done using the capillary viscosimeter of Instron 1115 machine (capillary diameter: 1.225 mm; length; 5 mm). The analysis of data in Table 18.2 shows that melts of pure PA6 and its blends with g-PE behave like typical non-Newtonian liquids. The melt viscosity of the blends containing 30–40 wt% of g-PO is so high that at a shear stress of 0.5 MPa it is impossible to press them through the viscosimeter’s capillary. Some information for comparison of rheological behavior of PA6 blends with pure and grafted PO is given in Table 18.3. Values of MFI

Polyolefin Blends

Polyolefin Foams

Polyolefin Composites

Blends Dyeing

Blends Dyeing

Biomedical Applications Polymer Blends

Blends of Natural Rubber

Advances in Polyolefin Nanocomposites

Polymer Blends and Alloys

Phase Behavior of Polymer Blends

Synthetic fibres: Nylon, polyester, acrylic, polyolefin

Polyolefin Characterization: The First International Conference on Polyolefin Characterization (Macromolecular Symposia)

Polymer Blends Handbook 2 vols set

Synthetic Fibres : Nylon, polyester, acrylic, polyolefin

Progress in Olefin Polymerization Catalysts and Polyolefin Materials: Proceedings of the First Asian Polyolefin Workshop, Nara, Japan, December

Thermodynamics of Polymer Blends (Polymer Thermodynamics Library , Vol 1)

Polyolefin Fibres: Industrial and Medical Applications (Woodhead Publishing in Textiles)

Mixing Essential Oils for Magic: Aromatic Alchemy for Personal Blends

Biodegradable Polymer Blends and Composites from Renewable Resources

Polymer Blends Handbook, Volume 1 and Volume 2

Biomedical Applications Polymer Blends (Advances in Polymer Science)

Thermal Oxidation of Polymer Blends: The Role of Structure

Handbook of thermoplastic polyesters: homopolymers, copolymers, blends, and composites

Nano- and Micro-Mechanics of Polymer Blends and Composites

Designing with Blends: Conceptual Foundations of Human-Computer Interaction and Software Engineering

Polymer Blends and Composites in Multiphase Systems (Advances in Chemistry 206)

Polymers, Polymer Blends, Polymer Composites And Filled Polymers: Synthesis, Properties, and Applications

Polyolefin Blends

Polyolefin Foams

Polyolefin Composites

Blends Dyeing

Blends Dyeing

Biomedical Applications Polymer Blends

Blends of Natural Rubber

Advances in Polyolefin Nanocomposites

Polymer Blends and Alloys

Phase Behavior of Polymer Blends

Synthetic fibres: Nylon, polyester, acrylic, polyolefin

Polyolefin Characterization: The First International Conference on Polyolefin Characterization (Macromolecular Symposia)

Polymer Blends Handbook 2 vols set

Synthetic Fibres : Nylon, polyester, acrylic, polyolefin

Progress in Olefin Polymerization Catalysts and Polyolefin Materials: Proceedings of the First Asian Polyolefin Workshop, Nara, Japan, December

Thermodynamics of Polymer Blends (Polymer Thermodynamics Library , Vol 1)

Polyolefin Fibres: Industrial and Medical Applications (Woodhead Publishing in Textiles)

Mixing Essential Oils for Magic: Aromatic Alchemy for Personal Blends

Biodegradable Polymer Blends and Composites from Renewable Resources

Polymer Blends Handbook, Volume 1 and Volume 2

Biomedical Applications Polymer Blends (Advances in Polymer Science)

Thermal Oxidation of Polymer Blends: The Role of Structure

Handbook of thermoplastic polyesters: homopolymers, copolymers, blends, and composites

Nano- and Micro-Mechanics of Polymer Blends and Composites

Designing with Blends: Conceptual Foundations of Human-Computer Interaction and Software Engineering

Polymer Blends and Composites in Multiphase Systems (Advances in Chemistry 206)

Polymers, Polymer Blends, Polymer Composites And Filled Polymers: Synthesis, Properties, and Applications

Recommend Documents