Thermal Degradation Of Polymeric Materials 6a5m11

Thermal Degradation of Polymeric Materials Krzysztof Pielichowski and James Njuguna

Thermal Degradation of Polymeric Materials

Krzysztof Pielichowski and James Njuguna

Rapra Technology Limited Shawbury, Shrewsbury, Shropshire, SY4 4NR, United Kingdom Telephone: +44 (0)1939 250383 Fax: +44 (0)1939 251118 http://www.rapra.net

First Published in 2005 by

Rapra Technology Limited Shawbury, Shrewsbury, Shropshire, SY4 4NR, UK

©2005, Rapra Technology Limited

All rights reserved. Except as permitted under current legislation no part of this publication may be photocopied, reproduced or distributed in any form or by any means or stored in a database or retrieval system, without the prior permission from the copyright holder. A catalogue record for this book is available from the British Library.

Every effort has been made to copyright holders of any material reproduced within the text and the authors and publishers apologise if any have been overlooked.

ISBN: 1-85957-498-X

Typeset, printed and bound by Rapra Technology Limited Cover printed by The Printing House, Crewe, UK

Contents

Contents Preface ................................................................................................................. ix 1

Introduction ................................................................................................... 1 1.1

Thermal Degradation Techniques ........................................................... 3 Abbreviations (Table 1.1) ....................................................................... 4

2

1.1.1

Thermogravimetry (TG) ........................................................... 10

1.1.2

Pyrolysis (Py) ............................................................................ 14

1.1.3

Thermal Volatilisation Analysis (TVA) ..................................... 18

1.1.4

Differential Scanning Calorimetry (DSC).................................. 19

1.1.5

Matrix-Assisted Laser Desorption/Ionisation Mass Spectrometry (MALDI) ............................................................ 22

1.1.6

Others ...................................................................................... 23

1.2

Ageing and Lifetime Predictions ........................................................... 27

1.3

Thermal Degradation Pathways ........................................................... 29

Mechanisms of Thermal Degradation of Polymers ....................................... 31 2.1

Side-Group Elimination ........................................................................ 31

2.2

Random Scission .................................................................................. 32

2.3

Depolymerisation ................................................................................. 32

3

Thermooxidative Degradation ...................................................................... 33

4

Kinetics of Thermal Degradation .................................................................. 37

5

4.1

Introduction ......................................................................................... 37

4.2

Kinetic Analysis .................................................................................... 38

Polymers, Copolymers and Blends ................................................................ 41 5.1

Polyolefins ............................................................................................ 41

iii


5.2

5.3

5.4

5.5

5.6

5.7

5.8

iv

5.1.1

Polyethylene (PE)...................................................................... 41

5.1.2

Polypropylene (PP) ................................................................... 47

5.1.3

Polyisobutylene (PIB)................................................................ 48

5.1.4

Cyclic Olefin Copolymers ......................................................... 50

5.1.5

Diene Polymers......................................................................... 50

Styrene Polymers .................................................................................. 53 5.2.1

Polystyrene (PS) and its Chemical Modifications ...................... 53

5.2.2

Styrene Copolymers.................................................................. 57

5.2.3

Acrylonitrile-Butadiene-Styrene Terpolymer (ABS) ................... 59

5.2.4

Polystyrene Blends .................................................................... 60

Poly(Vinyl Chloride) (PVC) .................................................................. 62 5.3.1

Poly(Vinyl Chloride) Homopolymer ......................................... 62

5.3.2

Poly(Vinyl Chloride) Blends ..................................................... 68

Polyamides (PA) ................................................................................... 72 5.4.1

Poly(Ester Amide)s ................................................................... 76

5.4.2

Liquid-Crystalline Polyamides .................................................. 77

5.4.3

Polyamide Blends ..................................................................... 78

Polyurethanes (PUs).............................................................................. 79 5.5.1

Thermoplastic Polyurethanes.................................................... 83

5.5.2

Polyurethane Foams ................................................................. 85

Polyesters ............................................................................................. 89 5.6.1

Poly(Ethylene Terephthalate) (PET) .......................................... 90

5.6.2

Biodegradable Polyesters .......................................................... 91

Acryl Polymers ..................................................................................... 97 5.7.1

Poly(Methyl Methacrylate) (PMMA)........................................ 97

5.7.2

Acryl (Co)Polymers ................................................................ 104

5.7.3

Acrylonitrile-Containing (Co)Polymers .................................. 110

Others ................................................................................................ 112 5.8.1

Poly(Vinyl Acetate) (PVAc) ..................................................... 112

5.8.2

Poly(Vinyl Alcohol) (PVOH) .................................................. 115

Contents

5.8.3

Vinylidene Chloride (VDC) Copolymers ................................ 115

5.8.4

Sulfone-Containing Polymers ................................................. 116

5.8.5

Sulfide-Containing (Co)Polymers............................................ 120

5.8.6

Poly(Bisphenol-A Carbonate) (PC) ......................................... 123

5.8.7

Poly(Butylene Terephthalate) (PBT) ........................................ 125

5.8.8

Poly(Ethylene Glycol Allenyl Methyl Ether) (PEGA) .............. 126

5.8.9

Poly(Ether Ketone)s (PEKs) .................................................... 126

5.8.10 Poly(Epichlorohydrin-co-Ethylene Oxide) .............................. 126 6

Natural Polymers ........................................................................................ 129 6.1

Starch ................................................................................................. 129

6.2

Chitin and Chitosan ........................................................................... 130

6.3

Cellulose............................................................................................. 133

6.4

Lignins ............................................................................................... 138

6.5

Poly(Hydroxyalkanoate)s (PHAs)....................................................... 140

6.6

Proteins .............................................................................................. 143

6.7

Natural Rubber .................................................................................. 144

6.8

Poly(Hydroxy Acid)s .......................................................................... 148

6.9 7

6.8.1

Poly(L-Lactic Acid) (PLLA) .................................................... 148

6.8.2

Poly(L-Lactic Acid) Blends ..................................................... 149

Poly(p-Dioxanone) (PPDO) ................................................................ 150

Reinforced Polymer Nanocomposites ......................................................... 153 7.1

Glass-Fibre-Reinforced Composites .................................................... 153

7.2

Carbon-Fibre-Reinforced Composites ................................................ 157

7.3

Unsaturated Polyester Resins Reinforced with Fibres ......................... 161

7.4

Reinforced Polyurethane Composites ................................................. 162

7.5

Polyamides with Natural Fibres.......................................................... 165

7.6

Other Composites .............................................................................. 167

v


8

9

Inorganic Polymers ..................................................................................... 173 8.1

Polysiloxanes ...................................................................................... 173

8.2

Polyphosphazenes............................................................................... 177

8.3

Polysilazanes and Polysilanes.............................................................. 180

8.4

Organic–Inorganic Hybrid Polymers .................................................. 184

High Temperature-Resistant Polymers ........................................................ 189 9.1

Aromatic Polyamides.......................................................................... 189

9.2

Aromatic Polycarbonates.................................................................... 192

9.3

Aromatic Polyethers ........................................................................... 193

9.4

Phenylene-Containing Polymers ......................................................... 194

9.5

Poly(Ether Ether Ketone) (PEEK) ....................................................... 195

9.6

Polybenzimidazoles (PBIs) .................................................................. 197

9.7

Polybismaleimides (BMIs) .................................................................. 199

9.8

Polybenzoxazines ............................................................................... 202

9.9

Other High-Temperature Polymers ..................................................... 203 9.9.1

Phenolic Resins....................................................................... 203

9.9.2

Epoxies .................................................................................. 206

9.9.3

Poly(Ether Imide) (PEI)........................................................... 207

10 Recycling of Polymers by Thermal Degradation ......................................... 209 10.1 Polyolefins .......................................................................................... 211 10.2 Polystyrene ......................................................................................... 215 10.2.1 Polystyrene in the Melt ........................................................... 216 10.2.2 Polystyrene in Solution ........................................................... 216 10.3 Poly(Vinyl Chloride) ........................................................................... 217 10.4 Polyamides ......................................................................................... 220

vi

Contents

10.5 Natural Polymers ............................................................................... 221 10.5.1 Poly(L-Lactic Acid) ................................................................ 221 10.5.2 Lignocellulose......................................................................... 222 10.6 Other Homopolymers ........................................................................ 224 10.7 Mixtures of Polymer Wastes ............................................................... 225 10.8 Thermal Degradation of Polymeric Materials – Ecological Issues ....... 230 10.8.1 Disposal Options and Sources of Information ........................ 230 10.8.2 Sustainable Development........................................................ 231 11 Thermal Degradation During Processing of Polymers ................................. 233 11.1 Polyethylene ....................................................................................... 234 11.2 Polypropylene and its Blends .............................................................. 235 11.3 Poly(Vinyl Alcohol) ............................................................................ 237 11.4 Other Polymers .................................................................................. 238 12 Modelling of Thermal Degradation Processes ............................................. 241 13 Concluding Remarks .................................................................................. 247 Author References ............................................................................................. 249 References from the Rapra Polymer Library ...................................................... 277 Index ................................................................................................................ 297

vii


viii

Preface

Preface

This book presents the most recent developments in the study of thermal degradation of polymeric materials, which is of paramount importance in developing a rational technology for polymer processing, in using polymers at higher temperature, and in understanding thermal decomposition mechanisms for the synthesis of fire-safe polymeric materials. The degradation of materials could either worsen the properties and therefore be undesired, or lead to a useful phenomenon in of compatibilisation and stabilisation of the polymer via degradation-induced cross-reactions or recycling of polymer waste through thermal degradation. Despite the plethora of literature on the subject, or because of it, we are often faced by a dilemma when asked to recommend a single textbook on thermal degradation of polymeric materials; this has convinced us that there is a definite need for a single up to date textbook aimed at thermal degradation of polymeric materials. This book is specifically designed to introduce this field to scientists, engineers and students who have previously studied polymer science, and thus it attempts to be as comprehensive as possible within the constraints imposed by the length of the book and the background of the readers. The book has also sought to reassert what we see as the most important issues raised by progress in polymer science and to contemplate their relevance, so we have attempted to provide an introduction to the thermal degradation of polymeric materials but also to show how these polymeric materials contribute to today’s lifestyle. We also intend to show that the diverse insight that has been gathered through the research reported in this book can be meaningfully utilised to better our environment. Developments in thermal degradation of polymeric materials have been accomplished by a proliferation of literature in the form of books, reports, journals and conference proceedings. The emphasis of this book is on the thermal degradation of polymeric materials and we have made no attempt to cover the detailed, and equally important, topics associated with the bio-, mechanical-, photo- or catalytic degradation of macromolecular materials. The purpose is therefore to present a concise and thorough basic understanding of thermal degradation of polymeric materials. It is hoped to achieve this by taking the reader step by step through the developments in various thermal degradation study techniques, including recent advances in the relevant characterisation field, a consideration

ix

Thermal Degradation of Polymeric Materials of the mechanisms and kinetics of polymer thermal (thermooxidative) degradation. The fundamentals of the thermal degradation of polymers, copolymers and blends, natural polymers, fibre-reinforced polymers, nanocomposites, inorganic polymers and highperformance plastics are pursued. A brief review of the general principles of thermal degradation during processing is also presented. Further, a review is given on thermal recycling and modelling, including ecological issues concerning the thermal degradation of the polymers discussed. The field of polymer thermal degradation illustrates the difficulties of defining clear boundaries between disciplines, and this book seeks to stimulate the reader to investigate the subjects that they may not have encountered in their fields. To this end, the book tends to balance and integrate knowledge from many fields linked to thermal degradation and avoid preoccupation with any specific topic or perspective. For readers who intend to specialise in thermal degradation, this text forms not only an introduction to the discipline but also a framework for subjects to be studied in greater detail. Many of the references noted at the end of the book have been carefully selected to direct the reader to the specialised works that provide depth and an indication of what is possible in the chosen subject. For those who do not intend to pursue studies in thermal degradation of polymers beyond a certain point, the discussions in this book will serve as an overview and introduction, so that they can know and understand the thermal degradation of polymeric materials, its applications in other disciplines, and its significance in today’s world. We are indebted to institutions and individuals in both private industry and government that have been most generous with advice and . Cracow University of Technology, Rapra Technology Limited and City University, London, are among the organisations that assisted with development of the book. One of the authors (JN) was also ed by a Marie Curie Fellowship of the European Community programme ‘Improving the Human Research Potential and the Socio-Economic Knowledge Base’ under Contract No. HPMT-CT-2001-00379. Finally we must acknowledge the invaluable assistance of the many individuals who have contributed to our efforts to complete this book. Dr. S. Humphreys of Rapra Technology Limited has been very helpful. Other sincere thanks go to Prof. J. Pielichowski and Prof. J.R. Banerjee for their generous help in the course of the book preparation. Finally, special thanks go to our wives – Kinga and Agnieszka – for their continuous during the editing period. Krzysztof Pielichowski James Njuguna

x

Introduction

1

Introduction

Degradation of polymers includes all the changes in the chemical structure and physical properties of the polymers due to external chemical or physical stresses caused by chemical reactions, involving bond scissions in the backbone of the macromolecules that lead to materials with characteristics different from (usually worse than) those of the starting material [a.1, a.2] {503329}. Polymer degradation in broader includes biodegradation, pyrolysis, oxidation, and mechanical, photo- and catalytic degradation. According to their chemical structure, polymers are vulnerable to harmful effects from the environment. This includes attack by chemical deteriogens – oxygen, its active forms, humidity, harmful anthropogenic emissions and atmospheric pollutants such as nitrogen oxides, sulfur dioxide and ozone – and physical stresses such as heat, mechanical forces, radiation and ablation. While trying to elucidate the general features of polymer degradation, including the mechanism of elementary reactions, it is important to consider the effects of various physical factors on the reactions. The degradation of materials could either worsen the properties and therefore be undesired, or lead to a useful phenomenon in of compatibilisation and stabilisation of the polymer via degradation-induced cross-reactions or recycling of polymer waste through thermal degradation {886353}. The thermal degradation of polymers refers to the case where polymers at elevated temperatures start to undergo chemical changes without the simultaneous involvement of another compound [a.2]. Thermal degradation of polymers is of paramount importance in developing a rational technology for polymer processing, in using polymers at higher temperature, and in understanding thermal decomposition mechanisms for the synthesis of fire-safe polymeric materials. Thermal degradation of polymers can be subdivided into three types. The first is characterised by complete degradation with breaking of the main chain. Rupture of side fragments along with formation of volatile products and char residues are peculiar to the second type. Crosslinked polymers belonging to the third type give a small amount of volatiles and char largely. Studies of degradation mechanisms have not only served as a basis for prolonging the lifetime of polymers, but have also aimed at enhancing the degradation rate of large-volume plastics, such as polyethylene, poly(vinyl chloride) (PVC), polyamides (PA) or polystyrene (PS), to overcome the rapidly increasing problems of landfills filling with slowly degrading waste plastic products.

1

Thermal Degradation of Polymeric Materials Polymers may also be subjected to fairly high temperatures during processing, and during this time thermal degradation may be far more important in modifying the properties of the original material than any thermal degradation occurring in general usage {868543} {882333}. Thermal degradation is likely to be responsible for serious damage to any polymeric material, and this effect is especially important for recycled polymers, as they suffer successive cycles of high and low temperatures [a.6, a.7]. In addition, polymers usually experience shear-related thermal degradation during the process of manufacture. Therefore, investigation of the effects of shear on thermal degradation is of great importance. Controlling degradation requires understanding of many different phenomena, including the diverse chemical mechanisms underlying structural changes in macromolecules, the influence of polymer morphology, the complexities of oxidation chemistry, the complex reaction pathways of stabiliser additives, the interaction of fillers and other ingredients as well as impurities, and the reaction–diffusion processes that often take place. Furthermore, there exist substantial differences between pure and industrial polymers that may have detrimental effects on the thermal degradation of the macromolecular material. For instance, comparing the chemically pure powder form of poly(methyl methacrylate) (PMMA) with industrial-grade PMMA, the absence of mass transport limitation due to small particle diameter has a substantial impact on degradation [a.4]. The reaction steps by the differential thermal analysis (DTA) method were all reported to be endothermic in both pure nitrogen and oxidative atmospheres for pure powdered PMMA, while in industrial-grade PMMA, due to the relatively large particle diameter, the reaction is endothermic in pure nitrogen but exothermic in the presence of oxygen. For powdered PMMA this effect was not observed; therefore, due to the small particle size, mass transport limitation was not the controlling mechanism in the degradation of pure powdered PMMA. The study of the thermal degradation of polymers is therefore important in understanding their usability, storage and recycling. Though various kinds of techniques have been proposed for the conversion of waste plastics, it is generally accepted that recovery of materials is not a long-term solution to the present problem, and that recovery of energy and/or chemicals is a more attractive option. Consequently, new technologies are being investigated for energy and chemical recycling of plastic wastes. In spite of its limitations, like high melt viscosity, poor heat transfer and excessive vapour release, pyrolysis has been the common method for polymer degradation. An interesting novel approach is degradation in a single phase, which has been proposed to lower the degradation temperature by better heat transfer [a.5]. Another important issue is the flammability of polymers – fires involving polymeric materials have led to many fatalities and serious injuries, mainly attributed to the smoke released during the burning of the polymers {814817}. However, it is increasingly realised that the toxic gases that are evolved during thermal decomposition/burning play

2

Introduction a significant role in the devastating effects of such fires, especially in closed spaces [a.1]. In addition, appreciable amounts of HCl are evolved on heating chlorine-containing polymers (particularly PVC) even without ignition {883236}. While taking these factors into consideration during the material design stage, special attention needs to be paid to the vapour phase of the oxidation, as this is even more difficult to inhibit than oxidation in the solid polymer. Meanwhile, there is evidence that the regulations governing the use of polymers are only going to get more stringent. For polymers to remain competitive, a proper understanding of their thermal degradation is a key factor in their advent {785601}. This work summarises recent developments in the thermal degradation of polymers. To start with, a brief overview of thermal degradation techniques is presented, followed by the mechanisms and kinetics of thermal degradation. Thermooxidative degradation is briefly mentioned. This is followed by a consideration of the thermal degradation of various polymers, copolymers, high-performance plastics, blends and composites. A brief overview on the general principles of thermal degradation during processing is also presented. Further, a quick review is given on recycling and modelling of the polymers discussed, with some concluding comments. Nevertheless, this work does not deal with compounded polymers or catalysed thermal degradation of polymers, as they are beyond the scope of this review. The kinetics of thermal degradation and thermooxidative degradation are also only highlighted, and an exhaustive review is not provided. The current work therefore deals exclusively with pure thermal degradation of polymers unless otherwise stated. Most of the polymer and other abbreviations used in this work are given in Table 1.

1.1 Thermal Degradation Techniques Thermal analysis methods have proved useful not only in defining suitable processing conditions for these polymers as well as drawing up useful service guidelines for their application, but also in obtaining information on the relationships between thermal properties and polymer chain structure. In recent years, several different configurations of instrumentation have been developed in order to accomplish degradation for both conventional qualitative and quantitative analysis. Thermal decomposition of polymers has been investigated by techniques like thermogravimetry (TG) and differential scanning calorimetry (DSC). Volatile products have been analysed on-line by mass spectroscopy (MS) and Fourier transform infrared spectrometry (FTIR). The ‘hyphenated’ thermoanalytical techniques, e.g., TG-MS or TG-FTIR, have been proved to be a powerful tool in studying structural features of complex organic materials, although the volatile organic compounds obtained normally for only ca. 50–70% of the original organic matter [a.7]. The principal aim of thermal degradation induced by thermal energy alone is to break the heterogeneous macromolecular structure by maximising the quantity of molecular fragments of decomposition products (structural building subunits) and to make them

3


Table 1. Abbreviations AA

acrylic acid

ABS

acrylonitrile-butadiene-styrene terpolymer

ACVA

4,4-azo-bis(4-cyanovaleric acid)

ADSC

alternating differential scanning calorimetry

AFM

atomic force microscopy

AIBN

azo(bisisobutyronitrile)

ANN

artificial neural networks

APP

ammonium polyphosphate

aPS

atactic polystyrene

BD

1,4-butanediol

BDGE

1,4-butanediol diglycidyl ether

BMI

bismaleimide

BMIDM

4,4-bis(maleimidodiphenyl)methane

BMIE

2,2-bis(4-maleimidophenyl)ether

BMIF

2,2-bis(4-(4-maleimidophenoxy)phenyl)hexafluoropropane

BMIM

2,2-bis(4-maleimidophenyl)methane

BMIP

2,2-bis(4-(4-maleimidophenoxy)phenyl)propane

CA

cellulose acetate

CB

carbon black

CL

chemiluminescence

CMPSF

chloromethylated polysulfone

CNF

carbon nanofibres

COC

cyclic olefin copolymers

C

cetylpyridinium chloride

D

3-chloro-l,2-propanediol

E

chlorinated polyethylene

U

chlorinated polyurethane

CRF

cold ring fraction

CRFC

carbon-fibre-reinforced composite

CRTA

controlled-rate thermal analysis

DAT

diaminotoluene

DGEBA

diglycidyl ether of bisphenol-A

4

Introduction

Table 1. Continued... DHC

dehydrochlorination

DLA

D-lactide

DMA

dynamic mechanical analysis

DNS

2,4-dinitrostyrene

DTA

differential thermal analysis

DTG

dynamic thermogravimetry

ECOSAR

Ecological Structure Activity Relationships (USA)

EGA

evolved gas analysis

EII

electron impact ionisation

ENR

epoxidised natural rubber

EPDM

ethylene-propylene-diene elastomer

EPR

ethylene-propylene rubber

ESO

epoxidised sunflower oil

ESR

electron spin resonance

ESRI

electron spin resonance imaging

EVA

ethylene-vinyl acetate

FTIR

Fourier transform infrared spectrometry

GMA

glycidyl methacrylate

GPC

gel permeation chromatography

HAS

hindered amine stabilisers

HDPE

high-density polyethylene

HIPS

high-impact polystyrene

HIPS-Br

brominated high-impact polystyrene

HMDI

1,6-hexamethylene diisocyanate

HTPB

hydroxyl-terminated polybutadiene

IPC

impact polypropylene copolymers

IPN

interpenetrating polymer network

iPS

isotactic polystyrene

L

liquid-crystalline polymer

LDPE

low-density polyethylene

LLA

L-lactide

LOI

limiting oxygen index

5


Table 1. Continued... MA

maleic anhydride or methacrylic acid

MALDI-TOF matrix-assisted laser desorption/ionisation–time of flight (technique) MBS

methyl methacrylate-butadiene-styrene terpolymer

MC

melamine cyanurate

MDI

4,4-diphenylmethane diisocyanate

MMA

methyl methacrylate

MPW

municipal plastic waste

MS

mass spectroscopy

MSW

municipal solid waste

MW

molecular weight

MWD

molecular-weight distribution

MWNT

multi-walled nanotubes

NBR

nitrile-butadiene rubber

NMR

nuclear magnetic resonance

OAPS

octa(aminophenyl)silsesquioxane

OIT

oxygen induction time

PA

polyamide

PAH

polycyclic aromatic hydrocarbons

PAMS

poly(-methylstyrene)

PAN

polyacrylonitrile

PANI

polyaniline

PB

polybutadiene

PBI

polybenzimidazole

PBT

poly(butylene terephthalate)

PC

polycarbonate

PCL

poly(-caprolactone)

PCLA

poly(-caprolactam)

PCS

polycarbosilane

PD

1,2-propanediol

PDLA

poly(D-lactide) or poly(D-lactic acid)

PDMDPS

poly(dimethyldiphenylsiloxane)

PDMS

poly(dimethylsiloxane)

6

Introduction

Table 1. Continued... PDO

poly(p-dioxanone)

PDPS

poly(diphenylsiloxane)

PDsBI

poly(di-sec-butyl itaconate)

PE

polyethylene

PEEK

poly(ether ether ketone)

PEGA

poly(ethylene glycol allenyl methyl ether)

PEI

poly(ether imide)

PEO

polyethylene oxide

PEPF

phenyl ethynyl phenol–formaldehyde resin

PES

poly(ethylene sulfide)

PET

poly(ethylene terephthalate)

PEU

poly(ether-urethane)

PHA

poly(3-hydroxyalkanoate)

PHB

poly(3-hydroxybutyrate)

PHEMA

poly(2-hydroxyethyl methacrylate)

PHV

poly(3-hydroxyvalerate)

PIB

polyisobutylene

PIPA

poly(isopropenyl acetate)

PLLA

poly(L-lactide) or poly(L-lactic acid)

PMA

pyromellitic anhydride

PMAN

polymethacrylonitrile

PMF

maleimide functional resin

PMMA

poly(methyl methacrylate)

PMPS

poly(methylphenylsiloxane)

PMSAN

poly(-methylstyrene acrylonitrile)

PMTM

poly(methylthienyl methacrylate)

poly(BPMA)

poly(p-bromophenacyl methacrylate)

poly(CyAMA) poly(3-(1-cyclohexyl)azetidinyl methacrylate) poly(MPMA) poly(p-methoxyphenacyl methacrylate) poly(PAMA)

poly(phenacyl methacrylate)

POSS

polyhedral oligosilsesquioxanes

PP

polypropylene

7


Table 1. Continued... PPDO

poly(p-dioxanone)

PPE

poly(phenylene ether)

PPG

poly(propylene glycol)

PPS

poly(phenylene sulfide)

PPy

polypyrrole

PS

polystyrene

PSF

polysulfone

PSS

poly(styrene sulfone)

PTFE

polytetrafluoroethylene

PTHF

polytetrahydrofuran

PU

polyurethane

PVA

poly(vinyl alcohol)

PVAc

poly(vinyl acetate)

PVB

poly(vinyl butyral)

PVC

poly(vinyl chloride)

PVDC

poly(vinylidene chloride)

PVME

poly(vinyl methyl ether)

PVOH

poly(vinyl alcohol)

PVTES

poly(vinyl triethoxysilane)

Py-GC

pyrolysis–gas chromatography

Py-GC-AED

pyrolysis–gas chromatography–atomic emission detection

Py-GC-FTIR

pyrolysis–gas chromatography–Fourier transform infrared spectroscopy

Py-GC-MS

pyrolysis–gas chromatography–mass spectrometry

Py-MS

pyrolysis–mass spectrometry

QSAR

quantitative structure–activity relationships

RDF

refuse-derived fuel

RHR

rate of heat release

SAN

poly(styrene-co-acrylonitrile)

SATVA

sub-ambient thermal volatilisation analysis

SAXS

small-angle X-ray scattering

SBR

styrene-butadiene rubber

SDNS

styrene-2,4-dinitrostyrene

8

Introduction

Table 1. Continued... SEM

scanning electron microscopy

SPME

solid-phase microextraction method

sPS

syndiotactic polystyrene

TEM

transmission electron microscopy

TEOS

tetraethoxysilane

Tg

glass transition temperature

TG

thermogravimetry

TG-FTIR

thermogravimetry–Fourier transform infrared spectroscopy

TG-MS

thermogravimetry–mass spectrometry

Ti

initial temperature of decomposition

TIC

total ion chromatogram

Tm

melting temperature

TMA

thermomechanical analysis

TMAH

tetramethylammonium hydroxide

Tmax

maximum rate decomposition temperature

TMCh

N,N,N-trimethylchitosan

TPMMA

thiophene-capped poly(methyl methacrylate)

TTD

terminal trisubstituted double bond

TVA

thermal volatilisation analysis

TVD

terminal vinylidene double bond

UHMW-PET ultra-high-molecular-weight poly(ethylene terephthalate) UMPIR

urethane-modified polyisocyanurate foam

VC

vinyl chloride

VDC

vinylidene chloride

VTES-MMA

vinyl triethoxysilane-methyl methacrylate

WAXS

wide-angle X-ray scattering

XPS

X-ray photoelectron spectroscopy

XRD

X-ray diffraction

9

Thermal Degradation of Polymeric Materials escape quickly from the reaction zone to reduce secondary thermal fragmentation. The combination of results from more than one thermal degradation technique provides a better understanding of the thermal decomposition mechanisms, considering the advantages associated with each type of analysis.

1.1.1 Thermogravimetry (TG) Thermogravimetry (TG) is a thermal analysis method in which the mass change of a sample subjected to a controlled temperature programme is measured. The use of isothermal and dynamic TG for the determination of kinetic parameters in polymeric materials has raised broad interest during recent years {871496} {852665} {848609} {845557}. Although TG cannot be used to elucidate a clear mechanism of thermal degradation, dynamic TG has frequently been used to study the overall thermal degradation kinetics of polymers because it gives reliable information on the activation energy, the exponential factor and the overall reaction order {497843} [a.9]. These degradation parameters manifest themselves in changes in the slope and shape of the TG curves (Figure 1). Consequently, the corresponding differential TG (DTG) curves exhibit multiple peaks or asymmetric peaks with more or less pronounced shoulders. Except when the mass losses corresponding to each decomposition step occur in different temperature ranges, e.g.,

Figure 1. Dynamic TG analysis of PA, PE, PS and PVC at 10 k/min under argon atmosphere 10

Introduction those for PVC and poly(sec-butyl methacrylate), the relative magnitudes of the mass of polymer lost by each mechanism change with the heating rate. The Flynn–Wall [a.50] kinetic method is often applied to polymers with complex degradation mechanisms, and the values of the activation energy may vary with mass loss [a.32]. However, when the extent of the degradation reactions depends on the heating rate and when successive reactions overlap, the determination of the activation energy becomes ambiguous. Very recently, a method has been proposed to overcome this difficulty by deconvolution of the recorded DTG curves and the reconstruction of individual sets of TG curves for the different processes. The Flynn–Wall method is then applied to these individual sets of reconstructed TG curves, whereby the apparent activation energies of the individual processes are obtained. The high reproducibility of results in dynamic mode assures the calculation of precise kinetic parameters, while the isothermal mode permits the simple calculation of the temperature-dependent constant of the model. By use of this method, it is possible to relate the weight loss of samples with the degree of degradation directly. Moreover, the possibility of using different thermal histories can provide further information on the kinetic nature of the degradation process {868537}. Another drawback is that TG does not provide clear information on thermal degradation mechanisms because of its insufficient ability to analyse the evolved gas mixture. Thus, direct analysis of gas compositions by continuous monitoring with the real-time TGMS technique has gained more attention in the identification of gaseous products in the pyrolysis of polymers, in particular for mechanism studies. Recently, the development of a TG-MS and TG-FTIR interface design has made a significant breakthrough in thermal degradation investigations [a.8]. On the other hand, TG and volatile product analysis essentially give information concerning degradation that produces volatile species. Non-isothermal analysis tends to reflect phenomena at higher temperature. However, solid residue analysis is limited by interference due to crosslinking of functional groups in the polymer structure, and thus interpretation is complex. Therefore, investigation of isothermal degradation and analysis of residual materials needs to be integrated into the study of thermal degradation of a particular polymer {876727}. Therefore, the two modes are complementary, and it is necessary to perform both kinds of measurements to get a complete description of the mechanism of the degradation process. The kinetics of thermal decomposition reactions of carbonaceous materials is complicated in that the decomposition of these materials involves a large number of reactions in parallel and in series. Although TG provides general information on the overall reaction kinetics, rather than on individual reactions, it could be used as a tool for providing comparison kinetic data of various reaction parameters such as temperature and heating rate. Other advantages of determining kinetic parameters from TG are that only a single sample and 11

Thermal Degradation of Polymeric Materials considerably fewer data are required for calculating the kinetics over an entire temperature range in a continuous manner. Although mathematical models of thermal decomposition, obtained by TG, make it possible to understand the kinetics of the whole process, it should be considered that the proposed models are generally based on an oversimplification of the chemical reactions involved in the degradation, especially the absence of oxygen [a.3]. Different authors have developed several methods; comparing the results obtained, wide variations were observed depending on the mathematical analysis used {769831}. It seems that methods based on several curves (at different heating rates) present a lower risk of creating errors than methods based on a single curve {887512}. It is worthwhile to mention that short-term isothermal experiments are usually performed to determine the kinetic parameters associated with the thermal degradation of polymers. But the apparent activation energy values and the corresponding degradation curves thus determined might not be suitable to represent the behaviour of polymers in service because they are usually obtained at temperatures near to (or, in some cases, above) the melting temperatures.

1.1.1.1 Thermogravimetry–Fourier Transform Infrared Spectroscopy (TG-FTIR) The combination of TG and FTIR provides a very useful tool for the determination of the degradation pathways of a polymer, copolymer or the combination of one of these with an additive [a.435]. The TG is normally coupled to the FTIR spectrometer via a glass-coated transfer line. This transports the volatile products evolved during the decomposition of the sample to the gas cell of the FTIR spectrometer. Both the transfer line and the gas cell are heated to prevent condensation of the decomposition products. The FTIR spectrometer measures the spectra of the gases in the cell rapidly at frequent intervals. TG-FTIR makes it possible to assign the volatile components under investigation to the decomposition stages detected by TG during an experiment. Based on the measurements conducted, it is possible to achieve a simultaneous quantitative and qualitative characterisation of the materials investigated. Afterwards, a spectral range characteristic for a particular functional group can be selected and the infrared (IR) absorption bands in this range integrated and displayed as a function of time. Gas components identified by means of comparative spectra in the temperature range under investigation are assigned to the quantitative stages of decomposition of the TG signal as chemigrams. This makes classification with regard to thermophysical characteristics and qualitative and quantitative composition possible, in order to compare similar materials and discern favourable material behaviour. Furthermore, by separating the evolved gas phase into individual components, it is possible to separate the superimposed decomposition reactions, which makes possible deeper insights into the reaction dynamics of thermal decomposition.

12

Introduction With high chemical specificity and high resolution (timescale), TG-FTIR provides for the direct identification of compounds and functional groups; with the addition of overlapping weight-loss data, qualitative interpretation is one of the key advantages {776088}. In a recent brilliant review on this ‘hyphenated’ technique, Wilkie [a.8] deduced that, because TG-FTIR samples only the vapour phase, it is important also to analyse the solid residue at several temperatures in order to ascertain the correlation between the evolved gases and the arrangements that occur in the condensed phase which permit this evolution.

1.1.1.2 Thermogravimetry–Mass Spectroscopy (TG-MS) TG-MS is a useful ‘hyphenated’ technique combining the direct measurement of weight loss as a function of temperature with the use of a sensitive spectroscopic detector. The TG is coupled to the MS via a heated metal or quartz glass capillary tube. One end of the capillary is positioned close to the sample in the thermobalance. Part of the evolved gases is sucked into the capillary by the vacuum in the MS. The MS repeatedly measures the entire mass spectrum or monitors the intensity of characteristic fragment ions (m/z, the mass-to-charge ratio). TG-MS features are high sensitivity and high resolution, which allow extremely low concentrations of evolved gases to be identified, together with overlapping weight losses that can be interpreted qualitatively [a.435]. In addition to the weight-loss information, MS permits temporal resolution of the gases that are evolved during thermal or thermooxidative degradation of a polymer in controlled atmospheric conditions. The characteristics of a broad variety of TG-MS instrumental solutions that depend partly on the sample characteristics and the desired conditions of thermal degradation are normally considered in relation to polymer characterisation. This technique thus provides information about the qualitative aspects of the evolved gases during polymer degradation that is otherwise unavailable for TG-only experiments. This technique is therefore used for the structural characterisation of homopolymers, copolymers, polymeric blends and composites and also finds application in the detection of monomeric residuals, solvents, additives and toxic degradation products {776088} [a.97] (Figure 2). Topical issues on the advantages and limitations of TG-MS with respect to other evolved gas analysis techniques have recently been summarised by Raemaekers and Bart [a.10] in a review on TG-MS thermal degradation of polymers. The advantageous applications of the technique in polymer science can be extended from qualitative thermal degradation analyses to thermooxidation, structural characterisation and chemical analyses, kinetics, solid-state reaction mechanisms, chemical reactivity and curing, quantitative analyses, and finally product formulation and development.

13


Figure 2. TG-MS ion current intensities of some volatile components evolved from PVC Reprinted from [a.97] with permission from Elsevier

1.1.2 Pyrolysis (Py) The pyrolysis process consists of a very complex set of reactions involving the formation of radicals. In the absence of specific reactions, such as elimination and retro Diels–Alder reaction, thermal decomposition proceeds via a radical mechanism, initiated by homolysis of a bond {891557} {889529}. This generates a pair of free radicals as primary products, which can then undergo a multitude of secondary reactions – rearrangement, -scission, elimination, hydrogen abstraction, olefin addition, etc. – and ultimately abstract hydrogen from unreacted molecules to initiate them into the pyrolysis reaction cycle [a.11]. The gasification of by-products may additionally be applied, which results in a high proportion of gaseous products and small quantities of char (solid product) and ash. If the purpose is to maximise the liquid product yield, process conditions are selected as low temperature, high heating rate and short gas residence time. For high char yield, low temperature and low heating rate are required. In order to produce high yield of gas product, high temperature, low heating rate and long gas residence time should be applied.

1.1.2.1 Pyrolysis–Gas Chromatography (Py-GC) Py-GC is mainly employed in structure analysis that includes the exploration of monomer arrangement in the (co)polymer system, such as the number-average sequence length and

14

Introduction stereoregular distributions. In addition to the traditional post-pyrolysis derivatisation, pre-pyrolysis derivatisation has been developed in order to reselect degradation pathways effectively. The development of derivatisation techniques continues to grow in order to match the demand of different types of applications {704611}. It is noteworthy that volatilisation tends to decrease with increasing polarity (O2 content) due to intermolecular forces and, thus, polar degradation products are weakly represented in pyrolysates and also poorly eluted from the gas chromatograph column. To overcome this problem, various derivatisation techniques using tetramethylammonium hydroxide have been developed to enable hydroxyl and carboxyl groups to be detected as methyl ethers and methyl esters, respectively. In addition, derivatisation is a well-known technique in chromatography used to enhance chromatographic separation and/or detection for those compounds not suitable for separation/detection. The same concept has been adapted to Py-GC or PyGC-MS analysis [a.7]. Pyrolysis is carried out either outside or inside the identification instrument. In the outside mode, the thermal degradation is carried out by a pyroprobe connected to the injection port of a GC using a selected technique – time-programmed or flash pyrolysis. GC-separated individual pyrolysates can be identified with different equipment connected to the GC, e.g., Py-GC-MS, Py-GC-FTIR or Py-GC-AED (atomic emission detector). Advances in Py-GC and the different configurations can be summarised as follows: •

Use of a programmable temperature vaporisation injector furnace to conduct the multi-step thermal desorption and programmed Py-GC experiments,

•

Creation of a dual-inlet (pyrolysis and auto-sampler) system, sufficiently flexible to use both kinds of injection system,

•

Use of a thermal extraction unit for a furnace-type pyrolysis interface,

•

Development of a Py-GC system with a movable reaction zone.

1.1.2.2 Pyrolysis–Mass Spectrometry (Py-MS) Direct pyrolysis–mass spectrometry (Py-MS) is applied to determine the primary structure of macromolecules and to investigate selective thermal degradation mechanisms. This technique allows the thermal decomposition products of the polymer sample to be observed directly in the ion source of the mass spectrometer, so that the evolving products are ionised and continuously detected by repetitive mass scans almost simultaneously with their formation {805917} {757742}. Since pyrolysis is accomplished under high vacuum, the thermal fragments are readily removed from the hot zone, and because of the low probability of molecular collisions and fast detection the occurrence of secondary reactions

15

Thermal Degradation of Polymeric Materials is reduced. Therefore, primary pyrolysis products bearing the structure of the decomposing materials are mainly detected [a.7]. It is possible to greatly expand the number of degradation products to be analysed by connecting the pyroprobe directly to the MS (thermal degradation in vacuum). In this case, thermal degradation occurs in the ion source of the MS. By the Py-MS method, using inductively or resistively heated wire filaments and even low-energy (12–17 eV) electron impact ionisation (EII), the content of some easily fragmenting molecules can be low or absent (decomposition to CO2), and pyrolytically formed molecular fragments higher than m/z = 350 are seldom observed. This is due to the extensive thermal fragmentation under flash pyrolysis and to the MS fragmentation of the pyrolytically formed fragments even with low-energy EII. To avoid certain difficulties connected with the low-energy PyEI-MS and to obtain larger thermal fragments by the Py-MS method from humic solutes (up to m/z = 500), some specific methods have been adopted for thermal degradation and low-energy ionisation of pyrolytically formed fragments [a.12]. On the other hand, none of the three major ionisation techniques, i.e., low-energy EII, chemical ionisation or field ionisation, is ideal and certain advantages and disadvantages will occur with each of these [a.12]. The low-energy EII is readily obtainable on most mass spectrometers, whereas the other techniques require special modification of equipment. Mass peaks obtained by the Py-MS reflect the molecular ion distribution of the pyrolysate and any fragment ions formed under ionisation conditions. This overlapping Py-EI-MS information may hamper unambiguous identification. On the other hand, low-voltage EII will avoid further mass spectrometric fragmentation of pyrolytically formed molecular fragments as much as possible [a.12]. At a given not too high resolution, a certain m/z value does not necessarily belong exclusively to one class of molecules since ions of different structures may have the same mass-to-charge ratio and can contribute to the same peak in the mass spectrum. Under these criteria, the ions observed by the Py-EI-MS may be considered, with certain qualifications, to be specific and indicative of the chemical structure of the pyrolysed complex polymer. Finally, thermal degradation and identification of the degradation products with analytical pyrolysis are extremely equipment-sensitive tasks accompanied by several pitfalls.

1.1.2.3 Pyrolysis–Gas Chromatography–Mass Spectrometry (Py-GC-MS) In recent years pyrolysis–gas chromatography–mass spectrometry (Py-GC-MS) has been widely used for the separation and identification of the volatile pyrolysis products of polymers and can be considered as the most convenient method to detect simultaneously the presence of decomposition products qualitatively and quantitatively {589987}. Evolved gas analysis (EGA) performed by using a GC coupled with a mass-selective detector offers a number of advantages for the decomposition study. The number of peaks seen in the total

16

Introduction ion chromatogram (TIC) represents the number of compounds detected by GC-MS. The relative intensity of each peak corresponds to the relative concentration of each compound. The identification of each decomposition product can also be confirmed either by using model compounds and/or by comparing the spectrum with those in a GC-MS library. Various kinds of Py-GC-MS unit have been used, but the principal techniques have involved either heating the sample on an inert metal filament in an in-line chamber, or heating the chamber itself by means of an external furnace. During the chosen pyrolysis time, the carrier gas sweeps volatiles into the GC column, where they are separated according to their different boiling points and polarities. The separated components are then measured and characterised by the mass spectrometer. Most existing pyrolysis units are designed for the degradation of solid or highly viscous materials such as polymers. However, if the sample is a volatile oil, it presents a severe problem for such units because it may evaporate without decomposition, even before it has reached the desired pyrolysis temperature [a.13]. Recently, an in-line pyrolysis unit for the MS study of the thermal stability of light oils or volatile liquid prepolymers has been presented [a.13]. A sealed glass ampoule is dropped into the furnace, which is preheated and controlled at the desired temperature. After the chosen pyrolysis time has elapsed, a piston is screwed down to break the ampoule, and the pyrolysis products are thereby released and impelled by helium carrier gas via a heated capillary into the mass spectrometric source – for GC-MS analysis the products directly through the GC column before entering the MS source. This new pyrolysis unit has small dead volume, is precisely temperature-controlled and contains no cold regions where products would condense. The heated filament, Curie point and furnace are three major types of pyrolyser generally used in the experiments. The standard configuration is the pyrolyser mounted on top of the GC injection port. Once the pyrolyser is attached, the GC is exclusively used for pyrolysis experiments. In this technique the mass spectrometer is used as the GC detector, for which its sensitivity is at least as good as that of flame ionisation detectors, and even more importantly provides the facility of characterising the components associated with the chromatographic peaks of the pyrolysis fragments. On the detection side of the GC, in addition to flame ionisation and mass-selective detectors, there are other GC detection methods, such as atomic emission, flame photoionisation and nitrogen–phosphorus detection. An additional advantage is that mass spectrometers detect and measure permanent gases and other small molecules, to which flame ionisation detectors are insensitive. Advances in this technique, such as in the design of pyrolysis units, the use of sufficiently small samples, and the appropriate design of experiments, have provided reliable quantitative data that can be used to obtain mechanistic information about polymer 17

Thermal Degradation of Polymeric Materials degradation processes and/or help in deducing the initial macromolecular structure [a.14] {704611}.

1.1.3 Thermal Volatilisation Analysis (TVA) Thermal volatilisation analysis (TVA) is a common method invented in the early 1970s that allows examination of the volatile products of degradation and gives the rate of volatilisation versus temperature (or time), as shown in Figure 3 for poly(isoprenyl acetate) (PIPA) [a.15] {687290}. TVA experiments consist of measuring the pressure of substances undergoing transfer from one point to another in an initially evacuated system, which is continuously pumped – the TVA system consists of an oven with strict temperature control connected to a vacuum line. In addition to rate profiling of the volatile product flux of thermal degradation under high-vacuum conditions through measurement of pressure in the vacuum line as a function of sample temperature, the TVA technique may afford a convenient method for isolation, on the basis of volatility under high-vacuum conditions, of product fractions of thermal degradation for subsequent spectroscopic analysis {711105} {582415}. During the degradation, on-line quadrupole MS allows identification of the more volatile components and also selective ion monitoring throughout the heating programme.

Figure 3. TVA curves for PIPA (continuous evacuation, 10 °C/min) Reprinted from {687290} with permission from Elsevier

18

Introduction Using TVA experiment as a capstone, all products of degradation can be isolated for analysis by ancillary methods. At the end of the experiment, three main product fractions can be further examined: the volatile products condensable in liquid nitrogen; the tar-wax fraction that collected on the water-cooled surface beyond the hot zone (referred to as the cold ring fraction, CRF), and the non-volatile residue remaining in the sample boat. The condensable volatile products may be further separated on the TVA line using sub-ambient TVA (SATVA) while trace monitoring the volatilisation, recording and products responsible for each curve collected as separate fractions [a.16]. Infrared and MS methods may then serve for identification of the more volatile components. The less volatile liquids may be subjected to GC-MS investigation. The CRF can be collected from the TVA tube for further investigation using a suitable solvent, transferred to a weighing bottle and the amount determined by evaporation of the solvent. From direct weighing of the liquid fraction, residue and CRF, a mass balance for the main product fractions on the basis of volatility could be obtained, gases being estimated by difference from the sample weight. In recent times, TVA has been coupled to FTIR to expand its capabilities via a modified vacuumtight long-path gas infrared cell, as an interface allowing for the application of infrared spectroscopy for the on-line analysis of volatile products of polymer degradation.

1.1.4 Differential Scanning Calorimetry (DSC)

1.1.4.1 General Techniques Differential scanning calorimetry (DSC) is a thermal analysis method in which the occurrence of a temperature difference between the sample and the reference is the primary effect due to a thermal event within the sample. Heat-flux DSC and power-compensated DSC are the two types of DSC that have been widely used in thermal degradation of polymers. DSC provides a rapid method for the determination of the thermal properties of polymeric material, including thermal history studies, oxidation induction time (OIT) testing (see ISO 11357-6:2002 [a.17]) and dynamic and isothermal kinetic studies. A review by Bair [a.18] discussed the technique whereby a polymer sample is heated at a programmed heating rate in oxygen, at atmospheric or elevated pressure, with the most effective antioxidant system producing the highest degradation temperature. The OIT of polyethylene, as a film, powder or insulated wire, has been determined and it was observed that thin films acted as the optimum sample. For OIT testing, it has been shown that, when proper conditions are selected, there is no significant difference either in the general nature of the results or in the actual induction times when similar samples are run in the DSC [a.20]. However, in DSC stability tests, the sample cover must be in place while the oxidation is carried out at the desired reaction temperature. For polyolefins, oxidation conditions caused by temperature or ageing are manifested as discoloration,

19

Thermal Degradation of Polymeric Materials loss of mechanical and optical properties and surface cracks. OIT as measured by DSC varies with the number of branches and the length of branches introduced into the main chain by copolymerisation. The energetics of oxidation and thermal degradation of polypropylene (PP) has been extensively examined by DSC [a.20]. In an oxidising atmosphere PP can undergo oxidative crosslinking reactions followed by polymer degradation via chain scissions. Thermal scission of the C–C chain bonds is accompanied by a transfer of hydrogen at the site of the scission, and followed by a decomposition of the polymer. For PP, the hydrogen on the tertiary carbon is more reactive and is abstracted. The methyl group is small and does not hinder a hydrogen transfer. This mechanism leads to chain fragmentation and rarely yields a monomer. The resulting hydrocarbon polymer residue is highly oxidised. On the other hand, polyamide PA-6,6 degrades, both thermally and oxidatively, to yield water, carbon dioxide and ammonia. This has led to suggestions that the following reactions occur: two carbonyl endgroups react with a certain enthalpy of reaction to form a condensation product, carbon dioxide and water. A carbonyl group can yield branched structures with =C=N and water [a.18]. Also two amino endgroups could react to form a condensation product and ammonia. It is also possible that the latter reaction could yield branching through the reaction of a secondary amine with a carboxylic endgroup. Riga and co-workers [a.21] paid special attention to OIT measurements by the DSC method (using the ASTM E1858 standard test method) to determine the oxidative behaviour of some reference polymers (commercial engineering plastics) at selected isothermal temperatures and oxygen pressures. According to the results obtained, poly(phenylene sulfide), polycarbonate and polysulfone were the most stable of the polymers studied, while PVCs, both flexible and rigid, were the least oxidatively stable. The urethane elastomers were significantly more oxidatively stable than the diene elastomers studied. The polyacetals along with high-density polyethylene were the most oxidatively stable under high-pressure oxygen at 175 °C as determined by pressure DSC. It was noted that these polymers were much more stable in nitrogen and strongly susceptible to attack by oxygen. The work suggested also that polypropylene and its composites could be considered as a source of heat energy when combusting these materials at high temperatures and pressures (Figure 4).

1.1.4.2 Temperature-Modulated DSC (TMDSC) Temperature-modulated differential scanning calorimetry (TMDSC) has seen an exponential growth in interest since its introduction in the early 1990s by Reading and co-workers [a.22]. The TMDSC method improves the ability of conventional DSC by providing additional advantages – higher resolution and sensitivity, in addition to being

20

Introduction

Figure 4. PP and PP composites oxidation induction time (OIT), peak temperature and heat of combustion by pressure DSC Reprinted from [a.21] with permission from Elsevier

able to separate overlapping phenomena. TMDSC differs from conventional DSC in that a low-frequency sinusoidal or non-sinusoidal (e.g., saw-tooth) perturbation ranging from approximately 0.001 to 0.1 Hz (1000–10 s period) is laid in the baseline of the temperature profile. The use of a complex modulation allows the response to multiple frequencies to be measured at one time. The last decade has seen TMDSC commercialisation in various forms, e.g., modulated DSC, alternating DSC and dynamic DSC. Much of the interest has been fuelled by a desire to find new applications and advantages for this promising new thermal analytical technique [a.23]. In some areas, such as the separation of overlapping transitions of a different nature in polymer blends [a.23] and the monitoring of the degree of cure during the crosslinking reaction of thermosetting polymers [a.22], the advantages of TMDSC over conventional DSC have been clearly demonstrated. Schawe and Winter

21

Thermal Degradation of Polymeric Materials [a.25], Pielichowski and co-workers [a.26] and Hutchinson and Montserrat [a.23] applied modulated DSC for studying the crystallisation and melting of polymer materials, while Sandor and co-workers [a.24] applied the same technique for the characterisation of polyanhydride microsphere degradation.

1.1.5 Matrix-Assisted Laser Desorption/Ionisation Mass Spectrometry (MALDI) Matrix-assisted laser desorption/ionisation mass spectrometry (MALDI) provides massresolved spectra, which allow the detection of oligomers of low (10–15 to 10–18 mol) quantities of sample with an accuracy of 0.1–0.01% (up to 30 000 Da and above). The study of thermal degradation phenomena by MALDI involves the partial degradation of a polymeric sample by keeping it under inert or oxidising atmosphere at a certain temperature and then collecting MALDI-TOF (time-of-flight) mass spectra of the sample to observe the structural changes induced by heat and/or oxygen {870570} {766615} [a.27]. The molecules of the partially degraded polymer sample are detected without any further fragmentation, generating a mass spectrum that may consist of a mixture of undegraded and degraded chains. This off-line method of analysis suffers an important limitation in that only the soluble part of the polymer residue generated in the degradation processes can be analysed, and this also limits the upper temperature of thermal degradation or the specific conditions at which the formation of a totally insoluble residue is observed. Furthermore, the thermal degradation of a polymer sample is performed for a prolonged time at atmospheric pressure so that only the most thermally stable degradation products may survive the heating. Direct Py-MS performed on-line and in a continuously evacuated system provides very short transport times of the pyrolysis compounds from the hot zone and may then complement the MALDI data by supplying information on less thermally stable pyrolysis products and on compounds generated at temperatures at which the pyrolysis residue becomes insoluble and therefore inaccessible to the MALDI analysis [a.27]. Moreover, direct Py-MS allows fractionation and continuous monitoring of the effluents, and therefore it becomes easier, with respect to the MALDI method, to detect the less abundant pyrolysis products eventually formed. Recently, it has been shown that, by using isothermal pyrolysis followed by MALDI analysis, it is possible to gain detailed information on the structure of the pyrolysis residue of poly(bisphenol-A carbonate) (PC) by the detection of sizable oligomeric chains (up to 25–50 mers, and above) produced in the heating process at atmospheric pressure under a nitrogen stream [a.27]. Also, there is an opinion that, in studies concerning the thermal degradation of polymers, the MALDI technique should be best used in parallel with the direct Py-MS method, the latter being unique in obtaining structural information on intractable, insoluble materials.

22

Introduction

1.1.6 Others Identification of the low-molar-mass degradation products is a prerequisite in establishing degradation mechanisms. Prior to identification, appropriate methods must often be used to separate the low-molar-mass products from the polymer [a.19]. Solid-phase microextraction (SPME) is a relatively new extraction technique based on a fused-silica fibre coated with a polymeric stationary phase. The fibre is introduced directly into aqueous samples or the headspace over the liquid or solid sample matrix. During the extraction the analytes partition between the fibre and the sample matrix according to their partition coefficients {642071}. Although the amount of polyanalytes recovered by SPME is relatively small compared to several other methods, there are no analyte losses due to sample handling and the entire extraction is desorbed into the injection port of the gas or liquid chromatograph. Additionally, several fibre materials with different polarities are commercially available. Unfortunately, the choice of an appropriate stationary phase affects the sensitivity of the method. SPME has also been successfully applied to the extraction of degradation products from low-density polyethylene (LDPE) and toxic compounds from soil [a.28]. Non-degradative approaches, such as proton nuclear magnetic resonance (1H NMR) (liquid state) and 13C NMR (liquid or solid state), applied to humic solutes will give very relevant information about their chemistry. NMR spectroscopy has long been a method to probe degradation in polymers {882869} {872766}. For example, NMR has been employed to investigate radiation-induced crosslinking in polymers, including high-resolution 13C NMR of model hydrocarbons [a.29]. However, non-degradative methods alone do not yield sufficient chemical information and they are suited to providing an indispensable chemical overview. Degradative methods are widely applied to different polymers to give more structural information by trying to simplify the complex humic solute aggregates to specific individual compounds. Dynamic NMR is the NMR spectroscopy of samples that undergo physical or chemical changes with time. The timescales can be from picoseconds to months and the techniques used for their study depend on the timescale. Fast submillisecond processes are completely averaged out on the NMR timescale (around a second) and yield a normal single spectrum. However, their equilibria are temperaturedependent. When each of the exchanging entities has a different chemical shift, and the difference in enthalpy is similar to the entropy difference, then the chemical shift will vary with temperature in a controllable manner. Medium fast (up to a second or so) exchanges cause line broadening. At the fast end of the range a single spectrum is broadened. As the exchange slows, the spectrum splits into two, then starts narrowing again till two sharp spectra are observed for slow exchange. Varying the temperature changes the exchange rate, allowing the determination of the thermodynamic constants of the transition state. Medium slow (up to about a minute)

23

Thermal Degradation of Polymeric Materials exchanges yield sharp separate spectra but also yield exchange peaks in an exchange spectroscopy/nuclear Overha effect spectroscopy (ES/NOES) spectrum. If short mixing times and long relaxation delays are used, the ES/NOES data are quantitative and the results at varying temperatures can be used to calculate the thermodynamic parameters of the transition state as in the case for medium fast exchange. The relative concentrations may be used to determine the thermodynamic properties of the equilibrium. For slow exchange (over a minute) one can start with a mixture not at equilibrium and observe the change in the concentration of each species over time at a fixed temperature. This is enough to determine the difference in free energy, but the process must be repeated at different temperatures to yield the enthalpy and entropy differences. Electron spin resonance spectroscopy (ESR) and ESR imaging (ESRI) measure the number of charges occupying deep traps in a crystal bandgap [a.30]. By measuring the change in absorption of microwave energy within a continuously varying strong magnetic field, ESR detects the number of ‘unpaired spins’ of electronic charges trapped at various defects in the material lattice {681645}. This approach is based on encoding spatial information in the ESR spectra via magnetic field gradients and provides information on the spatial properties of paramagnetic species in a non-destructive way. ESR has the additional advantage in that it is selective and specific for the detection of species containing unpaired electron spins. Furthermore, the ESR detectable phenomena occur at much higher frequencies than the NMR ones, and therefore the first technique is inherently more sensitive than its nuclear equivalent. This is based on the fact that most electrons exist in pairs with no net spin – ESR may be observed from the unpaired electrons that exist, for example, in free radicals. Next, ESRI has evolved as a method for spectral profiling [a.31]. The correlation between the spatial distribution of the radical intensity and lineshapes and the degradation process made possible the visualisation of the onset of decomposition, the detection of differences between ultraviolet (UV) and thermal degradation, and the understanding of the effect of temperature on the course of thermal degradation. The non-destructive ESRI method is sensitive to early stages in the decomposition process, and is expected to be complementary to existing profiling methods, for instance FTIR, which are normally applied to more advanced stages of degradation. A recent work has assessed the thermal degradation process of PP and ABS containing hindered amine stabilisers (HAS) studied by ESR and ESRI [a.30]. The intensity profiles of HAS-derived nitroxides were determined by onedimensional ESRI, and the spatial variation of the ESR lineshapes was determined by two-dimensional spectral–spatial ESRI. Together with the determination of the nitroxide concentration, the imaging data allowed the mapping of the temporal and spatial variation of the nitroxides, depending on the irradiation source, time and temperature. The intensity profile in the sample depth was deduced by one-dimensional ESRI, and the spatial variation of the ESR lineshapes (spectral profiling) was determined by twodimensional spectral–spatial ESRI. The ESRI technique is of special interest in polymers 24

Introduction with phase-separated morphology. In ABS, for example, ESRI studies have demonstrated a hierarchical variation of the HAS-derived nitroxide concentration: within the sample depth on the scale of a millimetre, and within morphological domains of ABS on the scale of a few millimetres [a.31]. As a result, it became possible to establish an elastomer profile as shown in Figure 5, which tracks the evolution of the elastomer properties as a function of sample depth, type and length of treatment, and temperature. Ultrasonic techniques are one of the methods devised to measure the velocity and the attenuation of an ultrasonic wave through a polymer melt during extrusion. A highfrequency acoustic pulse is repetitively generated by an emitting piezoelectric transducer, which is transmitted to the polymer through a metallic buffer rod {546565}. Because of the acoustic impedance mismatch between the metal and the polymer, the acoustic energy is partly transmitted. The same phenomenon takes place at the second interface and the transmitted wave is thus caused to reverberate back and forth between the interfaces, producing several echoes from the initial signal. The signal transmitted across the second interface is then detected at the end of the second buffer rod by a second receiving piezoelectric transducer. This non-destructive evaluation technique is also very useful in understanding the long-term ageing processes of polymers (thermal fatigue and stresses) and ageing in situ so that proper design criteria for end-products can be established. The cone calorimeter is considered to be a new-generation facility for studying heat release behaviour, smoke emission behaviour and fire decomposition of polymeric materials simultaneously {776025} {431880}. The main parameters obtained from a cone calorimeter are divided into three kinds: (i) heat release parameters, including heat release rate, total heat released and effective heat of combustion; (ii) smoke emission parameters, including smoke production rate, total smoke production and smoke extinction area; (iii) fire decomposition parameters, including mass loss rate and mass loss. Another group of techniques are microscopic methods – scanning electron microscopy (SEM), transmission electron microscopy (TEM) and atomic force microscopy (AFM) – which are different techniques that provide complementary information and are mainly employed at several different stages of thermal degradation. They have enabled the concomitant morphological changes to be tracked on the micrometre and nanometre scales {864006}{838693}. AFM, in particular, is an extremely surface-sensitive technique that is capable of atomic or molecular resolution in the most favourable cases and has proved its capabilities for the study of a wide range of materials. For example, the morphological changes in poly(m-phenylene isophthalamide) fibres during the different steps of pyrolysis are too subtle to be detected by more conventional microscopic techniques, such as SEM (as shown in Figure 6), but can be seen by AFM (see Figure 7) {838693}. Evidently, an advantage of the AFM over the SEM is that little sample preparation is required, as the sample is not exposed to a high vacuum, and electrically insulating materials 25


A

UVB

0

Le 1 ng 2 th ,m 3 m

3380 3360 3340 4

3300

e

gn

Ma

,G

ld fie tic

3320

B

UVB

0

Le 1 ng 2 th ,m 3 m

3380 3360 4

3320

et

gn

Ma

,G

eld ic fi

3340

Figure 5. Two-dimensional spectral-spatial ESRI contour (top) and perspective (bottom) plots of HAS-derived nitroxides after (A) 70 h and (B) 643 h of irradiation by a Xe arc in a weathering chamber, presented in absorption. The spectral slices a, b, c and d for the indicated depths are presented in the derivative mode; these slices were obtained from digital (non-destructive) sections of the 2D image. %F is shown for a, b, c and d slices in (A) and for a, c and d slices in (B). Both 2D images were reconstructed from 83 real projections, Hamming filter, two iterations, L - 4.5 mm, H - 70 G, and were plotted on a 256 x 256 grid Reprinted from [a.31] with permission from ACS

26

Introduction

(a)

(b)

Figure 6. SEM micrographs of poly(m-phenylene isophthalamide) fibres before and after pyrolysis: (a) fresh fibre, (b) final pyrolysed fibre (900 °C) Reprinted from {838693} with permission from ACDS

can be examined {884128}. Therefore, hydrated, solvent-containing specimens can be imaged. AFM, TEM and SEM are nowadays also used to investigate the morphological transformations undergone by solid polymer material during thermal degradation. On the other hand, micro-thermal analysis (micro-TA) is a more recent technique that provides a characterisation tool capable of imaging samples in a variety of modes, including those of current AFM technology, combining the imaging capabilities of AFM with the ability to characterise, with high spatial resolution, the thermal behaviour of materials [a.7]. A miniature heater and thermometer replaces the conventional AFM tip, which enables a surface to be visualised according to its response to the input of heat (in addition to measuring its topography). Areas of interest may then be selected and localised thermal analysis (modulated DSC and thermomechanical analysis). Localised dynamic mechanical measurements are also possible. Spatially resolved chemical analysis can be performed using the same basic apparatus by means of Py-GC-MS or high-resolution photothermal infrared spectrometry. Moreover, additional information on crystalline structures and their transformations during thermal degradation may be provided by X-ray diffraction (XRD), wide-angle X-ray scattering (WAXS) and small-angle X-ray scattering (SAXS).

1.2 Ageing and Lifetime Predictions Polymers are subjected to destructive factors such as mechanical stress, the presence of different chemicals, ultraviolet light, ablation and high temperatures throughout shelf and service lives {805719}. These factors cause degradation and ultimately affect performance 27


(a)

1.00

75.0 nm

0.75

37.5 nm

0.50

0.0 nm

0.25

0

0.25

0.50

0.75

(b)

0 1.00 μm 1.00

75.0 nm

0.75

37.5 nm

0.50

0.0 nm

0.25

0

0.25

0.50

0.75

0 1.00 μm

Figure 7. General appearance of poly(m-phenylene isophthalamide) fibre before and after pyrolysis. The images were obtained by tapping mode AFM: (a) fresh fibre, (b) fibre pyrolysed at 900 °C. Lateral size: 1 μm Reprinted from {838693} with permission from ACS

28

Introduction and lifetime of the polymers, which are sometimes stored for long periods of time. Therefore it is important to know how long and under what conditions the polymers may best be stored with minimum deterioration of their properties. According to the lifetime stages of polymers, the relevant processes are classified as melt degradation, long-term heat ageing and weathering based on the mechanisms involved, i.e., thermomechanical, thermal, catalytic and radiation-induced oxidations and environmental biodegradation {785601}. The products are different low-molecular-weight (low-molar-mass) additives or degradation products from the additives or the polymer itself. The diffusion of these low-molecular-weight products changes the properties of the material and shortens the lifetime. For safety reasons it is necessary to have a good understanding of the thermal resistance of polymeric materials and to identify precisely the products likely to be formed. In addition to temperature, the induction time and therefore the durability of polymers depend upon the physical and chemical structure of the polymer, the efficacy of the stabilising additives, the presence of metal catalysts, the presence of stress and the power of the oxidising agent {739308}. Forecasting changes in the properties of polymer materials with time is the task of predicting the performance {708171}. The forecasting can be either by determining the service life of the material in a given set of conditions or by determining the guaranteed period of required performance by products of a given type. The prediction can be approached at three levels [a.33]: •

Empirical, predicting results from testing a given material,

•

Semi-empirical, based on the assumption that the mechanism of degradation can be presented in the form of a simplified model and the parameters have a physical meaning,

•

Non-empirical, based on the chemical physics of the polymeric material.

The above points specify the principles and procedures for evaluating the thermal endurance properties of plastics exposed to elevated temperature for long periods. The study of thermal ageing is based solely on the change in certain properties resulting from a period of exposure to elevated temperature. The properties studied are usually measured after the temperature has returned to ambient. For industrial practice, ISO standards have been developed covering thermal ageing and environmental degradation [a.17].

1.3 Thermal Degradation Pathways The breaking of chemical bonds under the influence of heat is the result of overcoming the bond dissociation energies. Organic polymers are highly thermally sensitive due to the

29

Thermal Degradation of Polymeric Materials limited strength of the covalent bonds that make up their structures. Scission can occur either randomly or by a chain-end process, often referred to as an unzipping reaction. Volatile products may be clipped from the end of a polymer chain from the very beginning of reaction, with a distribution that is not random, or by a process of end scission or backbiting – a process of unzipping may regenerate the monomer. In addition to these cleavages, at the lowest reaction temperatures enlargement processes can occur that increase molecular weight (molar mass) and may also increase polymer branching. As a result, there are many stages of degradation as subsets of thermal degradation [a.1, a.2, a.33]. •

Random initiation: occurs in the middle of a polymer chain, at an unspecified point.

•

Depropagation: occurs very similarly to terminal initiation, but the process continues and monomers keep volatilising out of the medium.

•

Intermolecular transfer: a polymer and a polymer radical yield two polymers and a polymer radical.

•

Terminal (end) initiation: occurs at the end of a polymer chain, when a monomer is volatilised out of the reaction medium.

•

Unimolecular termination: a short polymer chain breaks up into products, but this rarely is ed for in the bulk phase.

•

Termination by disproportionation: two polymer radicals share radicals and yield two non-radical polymers.

•

Termination by recombination: two polymer radicals together to form non-radical products.

30

Mechanisms of Thermal Degradation of Polymers

2

Mechanisms of Thermal Degradation of Polymers

Depolymerisation and statistical fragmentation of chains are generally the two different mechanisms of degradation of polymers. The rate and extent of degradation may be monitored by changes in a sample’s mass and molecular weight, detection and quantification of reaction enthalpy changes, quantitative analysis of reaction by-products such as carbonyls and/or by measurement of consumption of oxygen. In a polymer there are usually many different bonds and types of bonds that can break – if this ensemble of different bonds were represented in a bulk material of small molecules, there would be a distribution of bonds broken. But with all of the bonds in a single polymer chain, there will not be a distribution of bonds broken in the initiation step since, once one bond in the polymer molecule breaks, the molecular weight of that polymer chain reduces and degradation begins {503329}. The bonds that tend to break first are the ones that form the weakest link(s) in the chain. This is why most polymers decompose at a temperature substantially lower than comparable small molecules when there are irregularities that can act as weak points where degradation starts. The factor that limits polymer thermal stability is the strength of the weakest bond in the polymer chain. Thermal degradation of polymers can follow three major pathways: side-group elimination, random scission and depolymerisation.

2.1 Side-Group Elimination Side-group elimination takes place generally in two steps. The first step is the elimination of side groups attached to a backbone of the polymer. This leaves an unstable polyene macromolecule that undergoes further reaction, including the formation of aromatic molecules, scission into smaller fragments, or the formation of char, e.g., in PVC. The first step of thermal degradation of PVC is the elimination of the side groups to form hydrogen chloride. With the side groups removed, a polyene macromolecule remains. This then undergoes reactions to form aromatic molecules, typically benzene, toluene and naphthalene.

31


2.2 Random Scission Random scission involves the formation of a free radical at some point on the polymer backbone, producing small repeating series of oligomers usually differing in chain length by the number of carbons. Fragmentation of polyethylene produce molecules with a double bond at one end, and molecules containing two double bonds located at either end of the molecule. Polymers that do not depolymerise, like polyethylene, generally decompose by thermal stress into fragments that break again into smaller fragments and so on. The degree of polymerisation decreases without the formation of free monomeric units. Statistical fragmentation can be initiated by chemical, thermal or mechanical activation or by radiation. Three classes of bond cleavage are recognised regardless of mechanism, i.e., breaking backbone, breaking C–C bonds and formal 1,3-hydrogen-shifts lead to new saturated and unsaturated endgroups [a.26]. If such random scission events are repeated successively in a polymer and its degradation products, the result is initially a decrease in molecular weight and ultimately weight loss, as degraded products, with a broad range of carbon numbers, become small enough to evaporate without further cleavage.

2.3 Depolymerisation Depolymerisation is a free-radical mechanism in that the polymer is degraded to the monomer or comonomers that make up the (co)polymer. Several polymers degrade by this mechanism, including polymethacrylates and polystyrene. The formation of a free radical on the backbone of the polymer causes the polymer to undergo scission to form unsaturated small molecules and propagate to the free radical on the polymer backbone. The mechanism of depolymerisation can occur under the same conditions (high temperature) as statistical fragmentation. The mechanism according to which monomeric units split off from the end of the polymeric chain is the reverse mechanism to polymerisation. Several polymers can be depolymerised until the equilibrium between monomer and polymer at a given temperature is reached in a closed reaction system.

32

Thermooxidative Degradation

3

Thermooxidative Degradation

Unlike thermal degradation, where polymer scission can occur randomly and/or at the chain end, oxidative degradation is characterised by random scission in the polymer backbone. For instance, the addition of free-radical initiators to polyolefins during extrusion is used industrially to improve the mechanical properties of the polymer {864583}. The most important issues in thermooxidative degradation of polymers are where oxidation takes place, which structure fragments are most vulnerable, how they should be protected, and what are the main principles of protection. Polymer degradation by adding peroxide is a common manufacturing technique because the controlled addition of peroxide to, for example, polypropylene leads to polymers with superior flow properties [a.34, a.35] {865102} {832500}. Addition of peroxide during the extrusion of polyethylene leads to an increase in the durability of the polymer [a.36]. Other studies include the addition of peroxides to blends of polyolefins and rubber to improve the mechanical properties due to the change in the polymer molecular-weight distribution (MWD) (molar-mass distribution) caused by the reaction with peroxide [a.34] {890265} {865094}. Many investigations of polymer thermal degradation have centred on determining the yield of monomer and the rate of change of average molecular weight. Research has shown that PVC by itself degrades slowly and takes on a colour changing from light yellow to reddish brown at longer times [a.37]. When followed by its curves of HCl evolution, a small induction time to degradation is observed. Temperature does not affect the amount of HCl evolved, but only the rate. During the evolution of HCl, double-bond sequences are formed in the polymer chains. Such polyenes are known to be responsible for the coloration of the resin and are supposed to be a labile site for oxidation. The appearance of long polyene sequences in the first steps of dehydrochlorination (DHC) has been mentioned where they do not increase their length with the degradation time, unless different conditions for degradation are maintained. However, very recently some authors [a.37] have proved that polyene sequences are formed at very low percentage of DHC and increase only their concentration (not their length) with degradation. Such sequences seem to be ‘stabilised’ against the action of oxygen, at

33

Thermal Degradation of Polymeric Materials least in the presence of ultraviolet radiation. PVC has the lowest thermal stability of all carbon chain polymers, and the main indicator of such degradation is the elimination of HCl, which is followed by coloration of the resin. It has been suggested that the process starts at structural defects generated during the polymerisation reaction [a.38]. A recent study on the thermooxidative degradation of PVC at 180 °C reported that the synergism between epoxidised sunflower oil (ESO) and metal soaps results from the reduction of the initial rate of DHC due to the reaction between HCl evolved at the early stages of DHC with ESO and metal soaps, which reduces its catalytic effect on the degradation of PVC as well as etherification and esterification reactions of labile chlorine atoms. This leads to the formation of short polyene sequences, which are responsible for the absence of initial coloration. It was thus found that ESO exerted a stabilising effect on the degradation of PVC [a.39]. Elsewhere, the SPME method allowed detection of trace amounts of products after the early stages of thermooxidation of polyamide PA-6,6 [a.40]. Low-molar-mass products formed during thermooxidation of PA-6,6 at 100 °C were extracted and then identified by GC-MS. A total of 18 degradation products of PA-6,6 were identified. In addition, some low-molecular-weight products originating from the lubricants were detected. Several unknown thermooxidation products of PA-6,6 were identified, including cyclic imides, pyridines and structural fragments from the original polyamide chain. 1-Pentyl2,5-pyrrolidinedione (pentyl succinimide) showed the largest increase in abundance during oxidation. The cyclopentanones were found to be present in the unaged material, with their amounts decreasing during ageing, and thus they are not formed during thermooxidation of PA-6,6 at 100 °C. The identified thermooxidation products formed as a result of extensive oxidation of the hexamethylenediamine unit in the polyamide backbone. This work concluded that the long-term thermooxidative degradation, just like thermal degradation and photooxidation of PA-6,6, starts at the N-vicinal methylene groups [a.19]. Oxidative degradation of poly(vinyl acetate) (PVAc) in the presence of benzyl peroxide at 70–125 °C in a batch reactor by dissolving PVAc in chlorobenzene was investigated [a.41]. The MWD was measured as a function of reaction time by GPC. Experimental data indicated that degradation occurs by random chain scission only, without crosslinking and repolymerisation; thus a radical mechanism for the oxidative degradation was proposed. An optimum temperature of 110 °C was observed for maximum degradation. The energy of activation of the random scission oxidative degradation rate coefficient, determined from the temperature dependence, was 20 kcal/mol. Features of the kinetics of the high-temperature oxidation of volatile polymer degradation products for some general classes of polymers and polymeric materials have been investigated [a.42]. For PMMA, isoprene and ethylene-propylene rubbers, the oxidation kinetics depended on the pyrolysis temperature and are in good correlation with the

34

Thermooxidative Degradation structure of the volatile degradation products. The oxidation of PMMA decomposition products with increase in pyrolysis temperature led to slight decreases of the activation energy and pre-exponential factor, which means that the rate constant decreased for hightemperature pyrolysis in comparison with low-temperature pyrolysis. This proves that the process of polymer ignition from low-calorie and high-calorie heat sources will follow various laws, as the capacity of a heat source with gas-phase oxidation will essentially differ owing to different kinetic laws of reaction of oxidation. Among PMMA pyrolysis products at moderate temperatures, methyl methacrylate monomer is prevalent. Increasing the temperature leads to a significant decrease in the monomer yield. Hence, at pyrolysis temperatures of 370–430 °C, the oxidation of gaseous products is by monomer oxidation, while at higher temperatures it is by oxidation of carbon monoxide and methane. The kinetic parameters of the oxidation of volatile degradation products can be used for characterisation of ignition and burning processes. Knowledge of the basic trends in the variation of these parameters can help to create fire-retardant polymeric materials. Degradation of polymers in solution is favourable since there is only a single phase, good temperature control and enhancement in the reaction rates leading to degradation at lower temperatures compared to pyrolysis [a.25]. Thus, oxidative degradation of the polymer occurs at temperatures much lower than conventional pyrolysis, resulting in considerable energy savings. Continuous distribution kinetics can provide more details of the degradation process by ing for the time evolution of the complete MWD and has been used to study the thermooxidative degradation of PS and PMMA in solution. The PMMA degradation rate coefficients were determined by analysing the MWDs at various reaction times [a.43]. It was observed that the reaction took place within the first 30 min, indicating that peroxide was consumed within this period. This is consistent with the results of different work showing that all the peroxide was depleted within the first 15 min and that the final MWD was reached in this time [a.44]. The activation energies for chain-end scission of the polymer are generally in the range of 8–15 kcal/mol. The activation energy obtained in this study confirms that the oxidative degradation of PMMA was by random chain scission, while the thermal non-oxidative degradation of PMMA was by chain-end scission [a.44]. The change of tacticity during the thermal treatment of commercial PMMA at 200 °C in air was studied by the NMR technique {884331}. The ratios of the three characteristic triads – isotactic, syndiotactic and heterotactic sequences – depend on the degradation time and approach a constant ratio of syndiotactic : heterotactic : isotactic = 3:4:3 at about 80% mass loss when starting with an initial ratio of syndiotactic:heterotactic: isotactic = 5:4:1. The correlation between the evaluated parameters and the degradation processes led to information on repolymerisation, which was dominant after about 50% 35

Thermal Degradation of Polymeric Materials weight loss, i.e., from about 20 h on, and afterwards in two consecutive steps with a rapid change in tacticity, when 50% of the residual material was converted into a new sequence distribution. This type of tacticity conversion has its maximum rate at about 80% weight loss in the second of the consecutive steps.

36

Kinetics of Thermal Degradation

4

Kinetics of Thermal Degradation

4.1 Introduction Generally, the thermal degradation of a polymeric material follows more than one mechanism. The existence of more than one concurrent chemical reaction accompanied by other physical phenomena such as evaporation and ablation introduce further complications for the modelling of degradation kinetics. The development of workable models able to describe the degradation kinetics of polymers has been the concern of many authors [a.1, a.2] {886353}. Kinetic study of thermal degradation provides useful information for the optimisation of the successive treatment of polymer materials in order to avoid or at least limit thermal degradation [a.3]. The analysis of the degradation process becomes more and more important due to an increase in the range of temperatures for engineering applications, recycling of post-consumer plastic waste, as well as the use of polymers as biological implants and matrices for drug delivery, where depolymerisation is an inevitable process affecting the lifetime of an article. Additionally, scission of macromolecules driven by thermal fluctuations at elevated temperatures provides a good example for the analysis of population dynamics in complex systems. This subject has therefore attracted substantial attention recently. A valuable approach for measuring thermal degradation kinetic parameters is controlledtransformation-rate thermal analysis (CRTA) – a stepwise isothermal analysis and quasi-isothermal and quasi-isobaric method. In this method, some parameters follow a predetermined programme as functions of time, this being achieved by adjusting the sample temperature. This technique maintains a constant reaction rate, and controls the pressure of the evolved species in the reaction environment. CRTA is, therefore, characterised by the fact that it does not require the predetermined temperature programmes that are indispensable for TG. This method eliminates the underestimation and/or overestimation of kinetic effects, which may result from an incomplete understanding of the kinetics of the solid-state reactions normally associated with non-isothermal methods. In particular, CRTA gives improved sensitivity and resolution of the thermal analysis curve since uniform conditions are maintained throughout the sample by means of an

37

Thermal Degradation of Polymeric Materials appropriate control of the reaction rate. This method has been applied to estimate the apparent activation energy for poly(ethylene terephthalate) (PET) and poly(butylene terephthalate) (PBT) without prior knowledge of the actual mechanism [a.45, a.46]. The kinetic parameters of these polyesters were estimated from both the controlledrate thermogravimetry (CRTG) curve and evolved-gas components, obtained from the simultaneous TG-MS system, and corresponding to a kinetic-model-ing random scission of the main chain. It was concluded that analytical techniques using the thermogravimetry traces obtained from different decomposition rates of CRTG are capable of establishing unique kinetic parameters. CRTA offers significant advantages in this field of study when dealing with the thermal decomposition of polymers. Another study addressed the uncertainty of extracting the kinetic parameters solely from thermograms [a.9, a.47]. Thus, modification of the Ozawa method [a.49] was introduced to tackle complex TG curves, since the traditional approach of the integral method has major limitations in extracting reliable kinetic parameters. The method implemented in extracting the multiple decomposition kinetics was based on subtraction of the mass of a specific event from the total mass loss, and then addressed the following event on the TG curve. For PMMA, the results demonstrated that decomposition could be described in of both depolymerisation and vaporisation, while for polytetrahydrofuran (PTHF) it can be described in of vaporisation only.

4.2 Kinetic Analysis Let us consider that, from the shape of TG profiles, whose character does not change with time, one can assume that there are no thermal effects that appear after a certain induction period, so there are no constraints to apply dynamic data for kinetic investigations of a polymeric material. Hence, the rate of reaction can be described in of two functions, k(T) and f(), thus: d/dt = k(T)f()

(1)

where is the degree of conversion, f() is the type of reaction and k(T) is the rate constant. By substitution of the Arrhenius equation, k(T) = A exp(–Ea/RT), the following equation results: d/dt = A exp(–Ea/RT) f()

(2)

After introduction of the constant heating rate = dT/dt and rearrangement, one obtains

38

Kinetics of Thermal Degradation ⎛ ⎞ dα ⎛ A ⎞ = ⎜ ⎟ exp ⎜ − E a ⎟ dT f (α) ⎝ β ⎠ ⎝ RT ⎠

(3)

where T is the temperature in kelvin, Ea is the activation energy, A is the pre-exponential factor, and R is the gas constant. A subsequent integration of equation (3) leads to the equation: α T ⎛ ⎞ dα A G(α) = = exp ⎜ −E a ⎟ dT (4) f (α) β ⎝ RT ⎠ 0

∫

∫

T0

which cannot be expressed by a simple analytical form since its right-hand side corresponds to a series of infinite functions. In mathematical practice, logarithms are taken: ⎛ AE ⎞ a ln G(α) = ln ⎜ ⎟ − ln β + ln p(x) (5) ⎝ R ⎠ and exponential integral p(x) is introduced: −x

p(x) = e − x

∞

−x

∫ ex

dx

(6)

x

where x = Ea/RT. Using an approximation of the exponential integral in a form proposed by Doyle [a.48] ln p(x) = –5.3305 + 1.052x

(7)

it is possible to determine the activation energy of the thermal process by following the specific heat flow of a process at several different heating rates: ⎛ AE ⎞ a ln β = ln ⎜ (8) ⎟ − ln G(α) − 5.3305 + 1.052x ⎝ R ⎠ Equation (8) generates a straight line when ln() is plotted against 1/T for isoconversional fractions, the slope of the line being equal to –1.052Ea/R during a series of measurements with a heating rate of 1, …, j at a fixed degree of conversion of = k. The temperatures Tjk are those at which the conversion k is reached at a heating rate of j. This method was developed independently by Ozawa [a.49] and Flynn and Wall [a.50]. Another isoconversional procedure, introduced by Friedman [a.51], uses as its basis the following relationship: ⎛ dα ⎞ E ln ⎜ ⎟ = ln f (α) + ln A − RT ⎝ dt ⎠

(9)

which makes it possible to find the activation energy value from the slope of the line (m = –E/R) when ln(d/dt) is plotted against 1/T for isoconversional fractions. In equation (1) the term f() represents the mathematical expression of the kinetic model. The most frequently cited basic kinetic models are summarised in Table 2. 39


Table 2. Kinetic model functions Model Symbol f() Phase boundary-controlled reaction (contracting R2 (1 – )1/2 area) Phase boundary-controlled reaction (contracting R3 (1 – )2/3 volume) Random nucleation. unimolecular decay law F1 (1 – ) Reaction nth order Fn (1 – )n Johnson–Mehl–Avrami JMA n(1 – )[–ln(1 – )]1–1/n Two-dimensional growth of nuclei (Avrami A2 2[–ln(1 – )1/2](1 – ) equation) Three-dimensional growth of nuclei (Avrami A3 3[–ln(1 – )2/3](1 – ) equation) One-dimensional diffusion D1 1/(2) Two-dimensional diffusion D2 1/[–ln(1 – )] Three-dimensional diffusion (Jander equation) D3 3(1 – )2/3/2[1 – (1 – )1/3] Three-dimensional diffusion (Ginstling–Brounshtein) D4 3/2[(1 – )–1/3 – 1] n-dimensional nucleation (Avrami–Erofeev An n[–ln(1 – )n](1 – ) equation) Reaction of first order with autocatalysis C1 (1 – )(1 + Kcat) Reaction of nth order with autocatalysis

Cn

(1 – )n(1 + Kcat)

Prout–Tompkins equation

Bna

(1 – )na

Non-isothermal curves of a thermal reaction can satisfy the kinetic equations developed for the kinetic analysis of ‘nth-order reactions’, even if they follow a quite different mechanism. The results of comparative studies led to the conclusion that the actual mechanism of a thermal process cannot be discriminated from the kinetic analysis of a single TG trace [a.52]. Besides, both activation energy and pre-exponential factor, given in equation (2), may be mutually correlated. As a consequence of this correlation, any TG curve can be described by an apparent kinetic model instead of the appropriate one for a certain value of the apparent activation energy. Therefore, the kinetic analysis of TG data cannot be successful unless the true value of the activation energy is known.

40

Polymers, Copolymers and Blends

5


5.1 Polyolefins Polyolefins are some of the largest-volume commodity polymers and are produced with a variety of processes. This results in a wide range of polyolefin grades, differing in tacticity, morphology, degree of branching, molecular-weight distribution and other properties such as thermal stability, which can require significantly different stabiliser formulations [a.1, a.2] {886353} {815959} {751920}. Thermal degradation of polyolefins has been described by a number of researchers, some of whose recent data are described in the subsequent sections.

5.1.1 Polyethylene (PE) The thermal degradation of polyethylene (PE) occurs by random chain scission {886343} {877171} {831619} {670581} [a.388] (Scheme 1), producing small amounts of the monomer (ethylene), and the degradation proceeds by a free-radical mechanism (Scheme 2) [a.53].

Scheme 1. Mechanism of thermal degradation of polyethylene Reprinted from [a.53] with permission from ACS

41


Scheme 2. Free-radical chain mechanism for PE degradation Reprinted from [a.53] with permission from ACS

42


Scheme 2. Continued...

Main-chain cleavage to form chain-terminal radicals is the initiation step. Intramolecular hydrogen abstraction by primary radicals, i.e., backbiting, occurs to an appreciable extent, which explains the preferential formation of some products such as C6, C10 and C14 hydrocarbons. However, the rates of the various elementary steps involved in the whole reaction have not yet been determined, and the details of hydrogen transfer during the formation of molecular fragments from a macroradical in the polymer melt are also not clear. The usual initiation, propagation and termination steps are illustrated in Scheme 2 with ‘weak links’ (step 1b) [a.53]. The net rate constant for initiation is f(ki), where f is the fraction of radicals that escape the cage in which they are formed. Cage escape in a viscous polymer melt, probably more by segmental than translational diffusion, will likely

43

Thermal Degradation of Polymeric Materials increase as MW decreases during pyrolysis, and the polymer configuration moves from the entangled to the random-coil state. However, f will not be simply related to macroscopic melt viscosity and is difficult to predict. Cage recombination achieves no chemical change, but cage disproportionation would give the same stoichiometric result as the ‘concerted’ process noted above and, if f << 1, could make a non-negligible contribution to the RS pathway even with a sizable kinetic chain length. One propagation pathway available to Rp• is -scission to form monomer (the UZ pathway) and regenerate itself, except for a decrease in degree of polymerisation (DP) by one (step 2). Second, abstraction of hydrogen from another P chain (intermolecular transfer or simply ‘transfer’) generates PANE and a sec on-chain radical (RS•) (step 3a). Its subsequent -scission (step 3b) leads to PENE and regenerates Rp• with DP for both reduced, on average, by half. Sequence 3 constitutes the RS pathway. Third, abstraction of hydrogen from the same polymer chain (intramolecular 1,x-hydrogen-shift or the BB pathway) generates a ‘nearend’ sec on-chain radical (Rs•) (step 4a). Its -scission forms either a volatile ENE and Rp• with modestly reduced DP (step 4b) or PENE with modestly reduced DP and a small alkyl radical (step 4c), which will ultimately be converted to a volatile ANE by transfer (step 4d). However, Rs• can also competitively undergo additional serial 1, x-shifts to move the radical centre further along the chain (not shown) or transfer (step 4e), which ‘cancels’ the effect of step 4a. For step 4a, x = 5 involving a six-membered ring transition state is kinetically favoured. C–C scissions achieved by step 1a or by sequence 3 are statistically random, but those achieved by steps 2, 4b and 4c are not. A final, seldom discussed propagation step would be a 1,x-shift much further along the polymer chain as an alternative intramolecular route to Rs• (step 5, large n). Its competition with intermolecular step 3a should depend on long-range polymer conformational factors that can place the radical centre in Rp• adjacent not only to C–H bonds in neighbouring P molecules but also to those in its own random coil. Chains are terminated by radical combination or disproportionation; self-reaction of Rp• is shown in steps 6a and 6b, cross-reaction of Rp• and Rs• in steps 6c–6e, and self-reaction of Rs• in steps 6f and 6g. Combinations involving Rs• would introduce LCBs or crosslinks into the residual polymer, while disproportionations would generate VL functionality. Radical–radical encounter in a viscous polymer melt is expected to be diffusion-limited, and kd will likely increase as MW decreases. Finally, a unimolecular termination mode is sometimes considered in which a small radical evaporates (step 6h) in competition with, for example, step 4d; this, however, begs the question of its ultimate product-forming fate downstream. Studies on thermal degradation of PE have reported hydrogen production in the fluidisedbed pyrolysis of a low-density polyethylene (LDPE) [a.54]. Hydrogen may be produced from the polymer melt and/or from thermal dehydrogenation of volatile products in the gas phase as a secondary reaction at the reactor temperature. From the results, it seems more possible that hydrogen is formed in the polymer melt, because, if more hydrogen were produced from thermal dehydrogenation of volatile products in the gas phase, increased

44

Polymers, Copolymers and Blends temperature would have promoted hydrogen production. However, the results did not indicate any increase of hydrogen production with increasing temperature. Further, more work is needed for elucidation of hydrogen transfer in the polymer melt, which is a key point to the understanding of the whole degradation of polyethylene [a.55]. Based on the kinetic analysis method described by Park and co-workers {769831}, which allows calculation of frequency factor/activation energy and prediction of the degradation of polyolefins at any time, a decrease in activation energies in thermal degradation was found to follow the order high-density PE (HDPE) > low-density PE > linear LDPE, unlike in other degradation processes, e.g., photooxidation, where degradation follows the order LDPE > HDPE > linear LDPE {889478}. These changes in activation energies for thermal decomposition were found to correlate with the respective rates of oxidation of the different polyethylenes. These results are also in agreement with the chemiluminescence (CL) {885413} data (Figure 8), where thermally degraded HDPE exhibited CL emission at lower temperatures than linear low-density PE (LLDPE) and metallocene PE (mPE), and an autocatalytic oxidation process was observed with ageing time. However, a higher light stability was determined for LLDPE, which exhibited no emission at lower temperatures when compared to HDPE and LDPE, which also showed CL emission below their melting points. Studies of thermal degradation have been conducted in heterophase propylene-ethylene copolymers (HPEC), which are known commercially as impact polypropylene copolymers and consist of polypropylene modified by an elastomeric component, typically ethylenepropylene rubber (EPR) [a.56]. The study emphasised that the EPR component plays an important role during degradation. The initial point of attack was thought to be the tertiary carbon atom in the propylene unit. The results indicated that the rate of ageing processes in HPEC is determined by the increased rate of oxygen diffusion and reactant mobility in polymers with higher EPR content. The higher proportion of amorphous domains and the corresponding higher amount of EPR ed for the stronger signals in the ESR spectra measured at 160 °C, which were related to higher mobility in this system. A series of chlorinated polyethylenes (E) based on HDPE was synthesised, having chlorine contents in the range 10–48% {708121}. TGA on E showed that the percentage weight loss was proportional to their chlorine contents at the end of the first stage of the two-stage decomposition. In contrast, PVC showed significantly greater weight loss (75%) as compared to its chlorine content of 57%. The 48% Cl-E and PVC had very similar degradation behaviours beyond a temperature of 500 °C. The DTG curves showed that there were four classes of decomposition profiles depending on the chlorine contents: PVC with 57% Cl; 48% Cl-E; 22–27% Cl-E; and 0–10% Cl-E. The work also demonstrated that it is possible to use the DTG curves of chlorinated polyethylenes to estimate the chlorine contents of E, at least in the range of 0–57% chlorine. 45


Figure 8. Chemiluminescence spectra under nitrogen of HDPE, LLDPE and mPE at different irradiation times (solar filter 300–800 nm, 550 W/m2) Reprinted from [a.388] with permission from Elsevier

46

Polymers, Copolymers and Blends Many problems with odour and taste in food packaging can be traced to degradation of the polymer packaging materials during processing. From this starting point, the degradation of polyethylene in a commercial extrusion coating process was studied by analysing the degradation products present in smoke sampled at the extruder die orifice [a.57]. More than 40 aliphatic aldehydes and ketones, together with 14 different carboxylic acids, were identified in the smoke. The highest concentration was found for acetaldehyde regardless of PE type and processing conditions. Increasing the extrusion temperatures in the range 280–325 °C increased the amounts of the oxidised products in the smoke. The extruded film thickness influenced the concentrations of the degradation products, with the thicker film giving higher amounts of product. The recycled polymer gave lower concentrations of degradation products compared with the virgin polymer. Differences in the product spectrum between the two virgin polymers were related to differences in the manufacturing process. Many of the identified compounds have very characteristic tastes and smells, which need to be carefully controlled in food packaging applications.

5.1.2 Polypropylene (PP) Polypropylene (PP) has been widely applied for commercial products in various forms despite the polymer being one of the most oxidatively unstable of the polyolefins {886162} {704378}. PP is known to be very vulnerable to oxidative degradation under the influence of elevated temperature and sunlight because of the existence of tertiary carbon atoms {883695} {720469}. PP degradation chemistry has been very extensively studied and recognised as a free-radical chain reaction, which leads to chain scission {800586} {751929}. It is generally accepted that this chain scission is responsible for PP degradation {749597}. The addition of stabilisers has been widely used to depress this radical reaction. However, it is difficult to maintain the long-term performance of stabilisers for various reasons, including volatility {776415}. It is therefore vital to find new methods to depress degradation during long-term use. The free-radical degradation of PP consists of initiation, propagation, chain branching and termination leading to non-radical products [a.58]. Initiation results from thermal dissociation of chemical bonds, whereas the key reaction in the propagation is the reaction of the polymer alkyl radicals with oxygen to form polymer peroxy radicals in a very fast reaction. The next propagation step is the abstraction of a hydrogen atom by the polymer peroxy radical to yield hydroperoxide polymer (POOH) and new alkyl radical. The chain branching of POOH results in the formation of very reactive polymer alkoxy radicals and hydroxyl radicals. The polymer oxy radicals can react further to form in-chain ketones or can be involved in termination reactions. The termination of PP radicals occurs by various bimolecular recombinations.

47


5.1.3 Polyisobutylene (PIB) Polyisobutylene (PIB) has been degraded thermally as thin film on a thermocouplecontrolled filament {565743}. The weight and number distributions of the oligomeric products observed were compared with those predicted statistically on the basis of random scissions. The results showed that the total pyrolysis could be interpreted exclusively in of parallel depropagation and random scission mechanisms. However, the partial pyrolysis results were not consistent with random scission statistics but instead implied that either some kinetically favoured scissions occur near the ends of the molecules or secondary reactions took place, which favour the production of lower oligomers {575698}. Earlier work on the thermal stability of PIB included measurements at a single specified temperature of the rate constants for the formation of a variety of volatile products from the thermal degradation of a liquid polymer {708135}. The work concluded that, whilst the rate constant for monomer evolution provides an inverse index of the thermal stability of the polymer at that temperature, the measured rate constants for the evolution of the individual oligomers could not be used for the purpose. It was claimed that the situation arises for several reasons, one of which is that volatile oligomeric products are produced not only by thermal decomposition, but also by direct evaporation of the original components from the lower end of the molecular-weight distribution of the polymer. Recent work by Lehrle and co-workers [a.59] has shown that degradation and evaporation behaviour can be distinguished by utilising the principle that thermal degradation produces components with structures that differ from those of components that are simply evaporated. The study was applied to the thermal degradation of PIB and showed this polymer to degrade in two steps, illustrated in Figure 9. In the first degradation region, the activation energies for three of the samples are rather similar, and indeed within experimental uncertainty are identical with those for onset behaviour (100–120 kJ/mol). This identity is perhaps not surprising, because the rate at onset is expected to correspond to the rate in the early stages of the first degradation process. However, a PIB sample prepared cationically using AlCl3 as the catalyst was exceptional in that its activation energy (onset) was ca. 85 kJ/mol, whereas the activation energy (first degradation step) was ca. 175 kJ/mol. The results obtained also indicated that the PIB-succinic anhydrides may be more able to withstand excessive temperatures, and were more stable than the corresponding PIB samples. This conclusion is ed by the results from the second degradation region, where the activation energies are higher for the PIB-succinic anhydrides. The researchers explained the results to be as a result of increased steric bulk around the end radicals in the PIB-succinic anhydrides that made backbiting (and thereby chain-end-initiated depropagation) less likely or else due to unstable hydrogen atoms in the starting PIB being oxidised during the conversion of PIB to PIB-succinic anhydrides.

48


Figure 9. Examples of selected ion currents for monomer and oligomers during the thermal decomposition of PIB Reprinted from [a.59] with permission from Elsevier

The chemical structures of the non-volatile oligomers (Mn = 2600–9000) isolated from the PIB residues obtained by thermal degradation of the polymer at 300 and 320 °C have been determined by NMR spectroscopy with regard to the reactive endgroups {604629}. The functional groups formed in the degradation process were the tert-butyl endgroup, isopropyl endgroup, terminal trisubstituted double bond (TTD), terminal vinylidene double bond (TVD) and non-terminal trisubstituted double bond. The average number of TTD and TVD per molecule was in the range from 1.46 to 1.64 and suggested that 53–67 mol% of the non-volatile oligomers were telechelic oligomers having both TTD and/or TVD, while 39–29 mol% of the oligomers were macromonomer-like oligomers having a double bond at one chain end. Both tert-butyl endgroup and isopropyl endgroup were produced by intermolecular hydrogen abstraction of primary and tertiary terminal macroradicals, and the subsequent -scission of the resulting on-chain macroradicals at the skeletal C–C bond of the main chain yields TTD and TVD. The formation of telechelic oligomers was interpreted by the major contribution of hydrogen abstraction of volatile radicals, in addition to a minor contribution of hydrogen abstraction of terminal macroradicals.

49

Thermal Degradation of Polymeric Materials Similar work by the same researchers concluded that the end initiation reactions from a terminal TTD and a TVD lead to the formation of a primary terminal macroradical and a tertiary terminal macroradical respectively, and also that the concentration ratio scarcely depends on the initiation reactions {615213}.

5.1.4 Cyclic Olefin Copolymers Cyclic olefin copolymers (COC) obtained with metallocene catalysts are engineering thermoplastics with some unique properties, such as high glass transition temperatures in combination with excellent transparency, low dielectric loss, low moisture absorption and good chemical resistance for high-performance optical, medical, electrical, packaging and other applications owing to their rigid cyclic monomer units [a.60] {641470}. Studies on thermal stability and degradation kinetics have reported that COC maintain their superior thermal stability of polyolefin materials despite their lower peak temperatures of degradation, narrower degradation temperature ranges and higher amount of residual weights at the end of degradation. These attributes have been related to the chemical structure and morphological features of COC as well as to steric effects, e.g., branching. The onset and peak temperatures of degradation for COC were reportedly lower than those for HDPE and scattered at about 410 and 430 °C respectively. COC have narrower temperature ranges of degradation than HDPE, which means that chain scission happens in shorter time. Another effect of branching is a possible change of the reaction mechanism; therefore, the bonds next to the side chain exhibit a higher breakage rate than normal PE bonds, which leads to a more pronounced maximum in the conversion rate curve. Because polyolefins consist of carbon and hydrogen elements, there is usually little or no residue once the degradation of polyolefin has ended; however, COCs have 2–5% black residual ash at the end temperature of degradation. This is probably due to some crosslinked structures formed from the reaction between radicals. The rate of formation of radicals increases with their stability, and therefore the content of the crosslinked structures is higher if the radicals formed during the pyrolysis process are more stable [a.60].

5.1.5 Diene Polymers The kinetics of thermal decomposition of styrene-butadiene rubber (SBR) have been investigated thermogravimetrically under various heating rates in either pure nitrogen or nitrogen mixed with 5–25% of oxygen [a.61]. The results showed that in pure inert gas the reaction involved only one stage, with an initial reaction temperature of 330–350 °C and apparent activation energy at ca. 210 kJ/mol, whereas under an oxidative atmosphere

50

Polymers, Copolymers and Blends two degradation steps were observed. The initial reaction temperature decreases, but the reaction rate and its temperature range increase when the heating rate was increased. Py-MS results of a styrene-butadiene-styrene block copolymer indicated that thermal decomposition of each block resembles that of the related homopolymer, giving the possibility to differentiate the block segments [a.62]. However, SBR degraded in a manner that is in between the thermal characteristics of the two homopolymers. In another work, the thermal depolymerisation of styrene-butadiene block copolymer under vacuum using programmed heating conditions showed that the initial reaction, which occurs between 300 and 400 °C, was cyclisation in the polybutadiene section of the polymeric chain, while at about 400 °C a limited amount of volatile products, mainly 1,3-butadiene and 4vinylcyclohexene, was formed [a.63]. At higher temperature or with prolonged isothermal heating, the primary products were styrene and toluene. Recently, the depolymerisation reactions of SBR have been performed in both batch and semibatch reactors at the conditions of supercritical and near-critical water, respectively [a.64]. The destruction efficiency and liquid product distribution were strongly dependent upon the operating conditions of reaction temperature, reaction pressure, oxidant concentration and flow rate. Strong two- and three-factor interactions among the parameters were observed in the semi-continuous reactions. These multiple-factor interactions in conjunction with the complex behaviour observed in the destruction efficiency strongly suggested the existence of multiple phases and the mass transfer resistances associated with multiple phases. Benzene, toluene, ethylbenzene, styrene, benzaldehyde, phenol, acetophenone and benzoic acid were detected as liquid products. Two parallel reaction mechanisms of oxidative and thermal degradation were proposed. Oxidative degradation appears predominant at the lower temperatures studied, while thermal degradation appears predominant at the higher temperatures in the depolymerisation of SBR based upon the liquid analysis. Carbon dioxide, carbon monoxide and water comprised the gas products. Nitrile-butadiene rubber (NBR), which is commonly used for gaskets and O-rings in fuel systems, is specified for use as fuel system gaskets. NBR rubbers release degradation products from the butadiene-rich areas, acrylonitrile-rich areas and interphase areas in the rubber where butadiene and acrylonitrile are aded. Alekseeva [a.65] reported that NBR rubbers could be identified on the basis of acrylonitrile, butadiene and ethenylcyclohexene in the pyrolysate. Hummel and co-workers [a.66] characterised a number of copolymer rubbers, including NBR rubber, using Py-MS. They proposed a fragmentation scheme to for many of the major ions in the mass spectrum of NBR rubber. In particular, the scheme ed for ions from areas of the rubber with ading acrylonitrile and butadiene segments. Recently Py-GC-MS has been used to identify the NBR rubber degradation products such as ethenylcyclohexene and acrylonitrile, as highlighted by Scheme 3, and the pyrograms obtained at elevated temperatures are shown in Figure 10 [a.67].

51


Scheme 3. A fragment of NBR rubber with ading acrylonitrile and butadiene molecules. Consideration of the bond strengths indicates that thermal cleavage takes place preferentially at tertiary carbon atoms and at bonds to the CN triple bond Reprinted from [a.67] with permission from Elsevier

Dynamic TG was used to investigate the thermal degradation kinetics of butadiene rubber (BR) in a nitrogen atmosphere at constant nominal heating rates over the temperature range 175–575 °C. Two distinct mass change stages in the TG curves indicated that the degradation of BR might be attributed to two reactions [a.67]. In the meantime, Gamlin and co-workers [a.68] studied the effect of ethylene/propylene content on the thermal degradation behaviour of ethylene-propylene-diene (EPDM) rubber and reported that onset and peak degradation temperatures increased linearly as

52


Figure 10. Pyrograms of NBR at 900, 800, 700 and 600 °C (from top to bottom, respectively). Time in min Reprinted from [a.67] with permission from Elsevier

the ethylene content increases above 40%. Dubey et al. [a.69] reported that the ignition point of volatile products emitted by natural rubber during thermal decomposition is 325–430 °C and the maximum rate of volatile formation occurs in the 300–500 °C pyrolysis temperature range. It has been shown that the pyrolysis of butyl rubber at 600 °C gives CH4, C2H6, C2H4, C3C6, (CH3)3CH and (CH3)2C=CH2 and isoprene, with maximum yields at 900 °C.

5.2 Styrene Polymers

5.2.1 Polystyrene (PS) and its Chemical Modifications Polystyrene (PS) is a large-volume, commodity polymer with a broad range of uses, e.g., in food packaging applications {815959} {428809}. Numerous studies of the

53


Scheme 4. Brominated PS thermal degradation via a free-radical mechanism

thermal degradation of PS with different tacticities (aPS, sPS and iPS) and their chemical modifications have been reported and the predominant mechanism is accepted to be that of random chain scission followed by intermolecular transfer, with smaller amounts of unzipping and intramolecular transfer (Schemes 4 and 5) {888059} {581327} {860868} {724282} {594526} [a.70]. Complete volatilisation is usual and, depending upon precise conditions, monomer yield may be up to 40%, with the balance made up mostly of dimers and trimers. Structural

54


Scheme 5. A probable intramolecular transfer (backbiting) step in the thermal degradation of polystyrene. Other backbites can occur with rather smaller probability Reprinted from {594526} with permission from Elsevier

55

Thermal Degradation of Polymeric Materials irregularities are thought to be important as initiation sites, and, in the more regular anionically synthesised polymer, initiation of degradation has been reported to occur at the terminal benzyl group [a.71]. During typical thermal degradation experiments, measuring the mass of remaining polymer and analysing the volatile products evolved serve to monitor reaction progress {7045893} {688694} {583706} [a.72, a.73]. Thus, the rates of multiple processes, such as random scission, chain-end scission, vaporisation, diffusion, repeated initiation/recombination and vapour-phase reactions, are usually lumped into one rate coefficient. Among structural features, regioregularity may play an important role – the head-to-head PS is stable over a wide range of temperatures whereas the nature of the degradation of the headto-tail polymer is strongly temperature-dependent. At low temperature (<300 °C) the initial degradation event is clearly scission of head-to-head linkages. The macroradicals thus formed undergo unzipping to evolve styrene monomer. At higher temperatures, the degradation is more complex and involves random chain scission and subsequent transformations as well as head-to-head scission {702952} {669883} [a.72]. Thermogravimetry results of PS, poly(p-chloromethylstyrene) (CMPS) and their corresponding derivatives substituted with 2-(N,N-dimethylamino)ethanol (HAPS, HAPSF) or triethylamine (APS, APSF) have showed that the substituted compounds undergo a thermal degradation involving a complex mechanism with three steps [a.74]. In the first step, degradation takes place with the removal of chloromethylated compounds as low-molecular-weight products (CH3Cl, HCl, Cl2, CH2Cl2, etc.). The second one is partially overlapped with the first stage and shows a more complex process, where there is an elimination of low-molecular-weight volatile compounds and also successive processes or some crosslinking after the elimination of low-molecular-weight fragments. The third step appeared to be the main degradation stage, the weight losses being between 40 and 65% in the series PS > CMPS > HAPS > APS. The researchers concluded that the chloromethylation reaction and subsequent substitution reaction of the chlorine atom with alkyl- and hydroxyalkylamines led to a significant decrease of the thermal stability of PS modifications [a.74]. In another work, it was found that during thermal degradation of poly(4-n-alkylstyrene)s (PAS-n) only volatile products were formed, allowing a complete analysis of different reactions that occurred at high temperature [a.75]. Studies undertaken to analyse these processes using data on the composition of the degradation products of PAS-n in isothermal conditions showed that the main thermal degradation process is depolymerisation, like in PS. Other thermal reactions involve fragmentation of alkyl side chains. The initiation process for the depolymerisation of PS-based polymers was identified as main-chain scission with the formation of two macroradicals. Apparently, the formation of cyclic or polyaromatic structures did not take place.

56

Polymers, Copolymers and Blends Further, bromination of PS is variously carried out to obtain graft polymers, highreactive modification and fire-resistant materials. Depending on the reaction conditions, bromination may lead to the introduction of Br solely into the ring or chain or in both areas of the PS structure [a.76]. Bromination of PS via an ionic route, conducted in solution in the presence of iron or aluminium chloride, leads to the introduction of Br into the aromatic ring. Studies have shown that PS with bromine on the ring undergoes one-stage thermal degradation at higher temperatures (350–450 °C) than PS with Br atoms on the chain or both chain and ring {524531}. The maxima of the thermal degradation rate, on the basis of the first derivative of the weight-loss curve, were found at 405–415 °C. Thus localisation of the bromine on the chain influenced the initial decomposition temperature, the integral procedural decomposition temperature and the char residue. An earlier study confirmed that PS containing bromine on the ring is described by thermal parameters whose values are generally similar to those of pure PS [a.70].

5.2.2 Styrene Copolymers An alternating styrene-maleic anhydride copolymer has been hydrolysed to obtain styrenemaleic acid copolymer [a.77]. It was found that alternating styrene-maleic acid copolymer degraded in three stages. FTIR spectra of a heated film of the copolymer as well as MS of the volatile products of the decomposition indicated that dehydration is the main reaction and takes place at the first stage of degradation. However, the decomposition is not only the simple regeneration of the maleic anhydride units. During heating at 140 °C the hydrolysed styrene-maleic acid copolymer lost its solubility in polar solvents, presumably due to crosslinking. The decomposition process of styrene-2,4-dinitrostyrene (SDNS) copolymers showed two stages [a.78]. The temperature range and weight loss of each stage depend on the copolymer composition. The characteristics of the thermal degradation of SDNS showed that increasing the content of DNS in the copolymer gradually decreases the stability. The initial degradation temperature of PS was higher than that of SDNS copolymers. The melting temperature (Tm) value for the first-stage pyrolysis decreased gradually as the 2,4-dinitrostyrene (DNS) content in the copolymer composition increased. The early stages of thermal degradation of poly(styrene-co-sulfone)s have been studied by Yang and co-workers [a.79]. The activation energy of thermal degradation was found to be in the range 180–300 kJ/mol. The activation energy decreased with increase in the content of sulfur dioxide in the polymer and with increase in the content of sulfur dioxide–styrene–sulfur dioxide triad sequences. The activation energies of thermal degradation of sulfur dioxide–styrene–sulfur dioxide and styrene–styrene–sulfur sequences were calculated as 175 and 390 kJ/mol, respectively. It was found that the sulfur dioxide–

57

Thermal Degradation of Polymeric Materials styrene–sulfur dioxide triad monomer sequence is the most sensitive microstructure in the thermal degradation of poly(styrene-co-sulfone)s [a.79]. Earlier research studied the macroscopic composition characteristics and thermal stability of several polysulfones of variable composition at a number of heating rates and environmental gas conditions and reported that the thermal stability depended on the content of sulfur dioxide in the polysulfones [a.80]. Thermal degradation of the styrene-isoprene-styrene (SIS) block copolymers is reported to occur in the temperature range of 190–235 °C under direct pyrolysis conditions [a.81]. The characteristic ions diagnostic to polyisoprene reached their maximum values at 213 °C, whereas the ones that could only be due to the decomposition of the styrene block had a maximum at 227 °C. Each block showed a very similar thermal behaviour to the corresponding homopolymer. Isoprene block degradation proceeded through random chain scissions at and positions, followed by cyclisation, yielding 1-methyl cyclopentene and 1-methyl cyclohexene. The splitting of monomers and low-molecular-weight oligomers was also detected. A radical mechanism was associated with the depolymerisation of styrene blocks. Indirect pyrolysis results indicated that secondary reactions were very effective, yielding mainly styrene, toluene, benzene, 1-methyl pentene and 1-methyl hexene, when degradation occurred in a closed reactor. Thermal stability and/or decomposition products arising from different blocks could not be differentiated with the use of indirect pyrolysis MS findings. However, the latter was used to the direct pyrolysis results. In a corresponding work, maximum thermal decomposition yields from polyisoprene and PS were detected at 220 and 230 °C, respectively, only a few degrees higher than the temperatures corresponding to the maxima present in the ion–temperature profiles of the thermal degradation products of the copolymer [a.82]. The batch pyrolysis of PS in the presence of poly(-methylstyrene) (PAMS) was investigated to determine the effect of the second polymer on the decomposition of polystyrene – two polymers with similar structures but different degradation behaviours. While PAMS degrades almost exclusively to its monomer, PS tends to form a significant amount of other products in addition to its monomer, styrene [a.83]. It was observed that the decomposition of polystyrene was dependent on the molecular weight of PAMS. Enhancement of the polystyrene degradation rate was achieved during binary mixture pyrolysis of lowmolecular-weight polymers, but rate inhibition was observed during degradation in the presence of higher-molecular-weight PAMS. It was proposed that the divergence arises from differences in the relative magnitudes of the enhancement caused by the production of PAMS-derived radicals and inhibition due to the incorporation of -methylstyrene monomer into depolymerising polystyrene chains. Variations in the initial concentration of the reactants had little effect on the yields of PS degradation products.

58


5.2.3 Acrylonitrile-Butadiene-Styrene Terpolymer (ABS) Polymer chain sequences with different repeat units can be chemically linked together through covalent bonds to form a block copolymer. When two immiscible constituents are selected to form the block copolymers, phase separation takes place and results in the formation of microdomains with sizes of ca. several tens of nanometres. To obtain good mechanical properties in practical applications, one of the constituents is normally in the glassy state (rigid segments) at the service temperatures and forms the dispersed microdomains. On the other hand, soft segments in the rubbery state intervene between the rigid microdomains and are responsible for the elastic behaviour. The rigid microdomains serve as fillers and play the role of physical crosslinkers as well. Thus, the novel block copolymers termed ‘thermoplastic elastomers’ can be regarded as filler-reinforced rubbery composites with the ability to flow at temperatures higher than the glass transition of the rigid phase. The thermal degradation issues of these complex polymer systems are strongly influenced by both bond dissociation energy and multi-segmented morphology – no single and comprehensive theory has been proposed so far to describe the behaviour of segmented thermoplastic elastomers under thermo(oxidative) conditions. The durability of acrylonitrile-butadiene-styrene (ABS) block terpolymers is important in many applications and depends on composition, processing and operating conditions, environmental weathering, heat ageing and installation damage {871997} {774204} {454425} [a.91]. ABS consists of a bimodal polymer system in which non-grafted polybutadiene particles are dispersed in a styrene-acrylonitrile copolymer (SAN). Degradation of the bulk polymer does not occur at ambient temperature due to limited oxygen diffusion. ABS thermally degrades with the formation of ammonia or very toxic hydrogen cyanide in the gas fraction and N-containing compounds in the oil fraction, which may lead to the corrosion of engine parts and the formation of harmful compounds such as HCN or NOx when the oils are used as fuel {639956} [a.85]. An early study revealed that the degradation of ABS is a radical process including both chain-end and random scissions [a.85]. Recent work has applied the TG-FTIR technique to investigate the degradation behaviours of ABS as well as the constituent polymers of ABS, namely PAN, PB, SAN and PS [a.86]. The investigation demonstrated that the evolution of butadiene commenced at 340 °C and of styrene at 350 °C, while the evolution of acrylonitrile began at 400 °C. Thermal degradation studies on ABS have shown that the kinetics and mechanism of degradation depend on the chemical structure of the copolymer and the experimental conditions [a.18] {760942}. Various studies have examined the changes that occur in the thermal properties of materials when ABS is grafted or blended with other polymers, e.g., PVC [a.87. a.88]. In addition there have been studies concerning the thermal behaviour of polyacrylonitrile and styrene-acrylonitrile binary copolymers [a.89].

59

Thermal Degradation of Polymeric Materials The degradation of ABS copolymer by a semi-batch operation at temperatures between 400 and 440 °C gave 50–63 wt% oil with 29–40 mg/mL concentration of nitrogen [a.90]. The degradation temperature significantly affected the rate of evolution and the amount and the quality of the degradation oil. Using an N2 dynamic atmosphere or changing the residence time of the products in the reactor also affects the products of ABS degradation, mainly NH3, aliphatic and aromatic nitriles. Heterocyclic compounds with one or two N atoms were identified in small amounts only. More than 50 wt% of the degradation oils consisted of hydrocarbons such as toluene, ethylbenzene, styrene, isopropylbenzene and methylstyrene, and as such it represented a possible hydrocarbon source or fuel provided the concentration of N-containing compounds can be decreased to an acceptable level. The ABS/bean oil system thermally degrades into an asphalt-like degradation residue (350–370 °C), which is soluble in common organic solvents such as tetrahydrofuran (THF), instead of the monomer and oligomers that are usually generated in the direct pyrolysis of ABS {824066}. Moreover, for the ABS/bean oil system the crosslinking reaction of ABS with bean oil takes place and forms a polymer network before the decomposition of ABS. Between the two reaction stages, the polymerisation or oligomerisation of sequences of adjacent nitrile groups occurs. The thermal degradation of ABS in bean oil was believed to be a radical process, which is dependent on the reaction conditions, especially the concentration of bean oil, reaction temperature and time. For aged ABS under an imposed stress, microcracks initiate from existing flaws in the degraded polymer surface layer [a.91]. When the degraded layer reaches a depth of 0.08 mm, these cracks are large enough to propagate into the bulk of the polymer, causing abrupt mechanical failure. Microindentation measurements suggest that an increase in Young’s modulus in this layer also promotes brittle behaviour. Degradation of the elastomeric polybutadiene phase in ABS is initiated by hydrogen abstraction from the carbon to unsaturated bonds, producing hydroperoxide radicals, leading to carbonyl and hydroxyl products. Crosslinking of polymer chains is facilitated by the free radicals that are produced.

5.2.4 Polystyrene Blends Mixing of different polymers has revealed a new realm of technically important materials. Varying the composition of the polymer blends can alter their properties. Polymer blends are clearly of great commercial significance and their thermal degradation has been the subject of many studies, but the diversity of situations and material combinations makes it difficult to generalise about behaviour {886353}. However, it is clear that the issue of greatest interest in blend degradation is whether the overall response of the system is simply the sum of the responses of the parts, or is influenced by component interactions

60

Polymers, Copolymers and Blends [a.92]. Although a very large number of combinations of polymers are possible, there are relatively few that lead to totally miscible systems from the thermodynamic point of view [a.105]. Results from analysis of volatile and condensable products of thermolysis have revealed interactions between component polymers in the polymer bulk with low-molecular-weight or free-radical products arising by thermolysis of macromolecules and migrating across the phase boundaries from one polymer to another [a.1, a.2]. Ultimately, the products of thermolysis either trigger degradation of the blend (destabilising effect) or act as stabilising species for any of the component polymers. The final effect may depend on the ratio of components or the temperature. Systems where the ultimate degradation rates are reduced or the decomposition temperatures of all component polymers are shifted to higher values have the optimum behaviour. The thermal behaviour of polymer blends generally shows some similarities with graft copolymers, but differs from those of random copolymers. Poly(2,6-dimethyl-1,4-phenylene oxide), more commonly referred to as poly(phenylene ether) (PPE), has a high melt viscosity, and as this makes processing difficult it has been found useful to blend it with atactic polystyrene (aPS) to give more readily mouldable compositions [a.92, a.436]. The two polymers are thermodynamically compatible over the complete composition range, and this allows materials to be tailored for particular combinations of mechanical properties and processability. aPS/PPE blends have shown that PS can be stabilised by PPE on the basis of the temperature at which maximum PS degradation occurs as obtained by TVA. Graft copolymers of aPS and PPE, which undergo phase separation, also showed stabilisation of the PS component, suggesting insensitivity of the degradation processes to the system morphology, though the domain size associated with graft copolymers was small relative to that arising from crystallisation. The explanation for stabilisation offered was that the readily available hydrogen atoms of the PPE divert and thus terminate the PS intermolecular transfer process. Syndiotactic polystyrene (sPS) is also fully compatible with PPE, and blends of these two materials may be of interest as engineering thermoplastics [a.90, a.91]. Pure sPS is typically about 50% crystalline and has a melting temperature of 262–272 °C, with a heat of fusion of ca. 50 J/g. PPE based on 2,6-dimethylphenol has a glass transition temperature of 210 °C and is amorphous. Blends of sPS and PPE are reportedly compatible in the melt state, but on cooling the sPS undergoes partial crystallisation, separating out from the blend. sPS is presumed to degrade in a similar manner to aPS, although variations may arise from reduced steric hindrance of the transfer reactions. The thermal degradation of sPS synthesised using a metallocene catalyst system is reported as not sensitive to polymer molecular weight, suggesting initiation by random chain scission [a.93]. In the blends investigated, the onset temperature of polystyrene degradation was typically 10–20 °C higher than that observed with pure sPS. Using a

61

Thermal Degradation of Polymeric Materials low heating rate to minimise diffusional hindrance of degradation volatiles gave an even greater temperature difference between blend degradation temperature and that of pure sPS. This indicated that the apparent stabilisation of the sPS was at least partly chemical. The onset of degradation of PPE in the blends was not observed by thermogravimetry, though the material endured two-stage degradation, and the beginning of the second stage of mass loss occurred at lower temperatures than in the pure PPE. This polymer appeared to be destabilised in the blends, in which the degree of destabilisation was a function of the concentration of sPS present. Infrared spectroscopy has shown that the polystyrene degradation residue changes little throughout the degradation, whereas the PPE undergoes main-chain rearrangement before mass loss occurred [a.94]. Spectra obtained with the blend degradation residues showed a similar rearrangement process to that observed with the pure PPE. The results obtained were consistent with a melt state interaction between the degrading polystyrene and the PPE. The observed degradation temperature range of the sPS is coincident with the Fries-type rearrangement in the PPE, and interaction between the free-radical species generated by the degrading polystyrene and PPE transient decomposition products was presumed. This stabilised the sPS through interruption of the intermolecular transfer process while it destabilised the PPE by partially hindering its rearrangement. The reaction possibilities are sPS macroradical + PPE macromolecule, through hydrogen abstraction, and sPS macroradical + PPE macroradical during the thermal rearrangement of the PPE. If the proposed mechanisms are correct, then some copolymerisation reactions may occur, but the similarities of the sPS and PPE spectra will make identification of any such species difficult. Recently it has been shown that degrading PS can abstract hydrogen from polycarbonate in aPS/PC blends, thus destabilising the polycarbonate component. Preferential interaction with cross-termination and mutual stabilisation has also been reported for aPS/PMMA blends [a.95].

5.3 Poly(Vinyl Chloride) (PVC)

5.3.1 Poly(Vinyl Chloride) Homopolymer Poly(vinyl chloride) (PVC) has enormous commercial applications, which makes it one of the most well-studied polymers. One of the problems associated with the processing and use of PVC is its low thermal stability despite a general agreement that normal PVC with head-to-tail structures should be quite stable {895401} {883236} {805671} This leads to the assumption that there exist several defect sites in the polymer chain that are responsible for the instability. Such possible defects in PVC are branching, chloroallyl groups, endgroups, oxygen-containing groups, unsaturations and head-to-head structures

62

Polymers, Copolymers and Blends {704210} {704203} [a.84]. It is therefore of great importance to clearly understand the thermal degradation process of PVC to facilitate its processing and usability. The thermal degradation of PVC is generally accepted to be a two-step process {893672} {871835} {830034} {755845}. The first step (up to 350 °C) mainly involves dehydrochlorination of the polymer (Scheme 6), resulting in the formation of conjugated double bonds that break during the second step (up to 550 °C). In the first step HCl is the main volatile product – the amount of the other products is very low, including quantities of benzene and some other hydrocarbons {883234}. The main labile sites for dehydrochlorination are the allylic and tertiary chlorines {865099}. Hydrogen chloride may be anticipated as a possible pyrolysis product from chlorinecontaining polymeric materials, and, in any quantitative kinetic study of their thermal degradation, precise measurements of HCl yield will be required. Benzene formation (Scheme 7) is a relatively low-temperature process starting at 220–230 °C with parallel HCl elimination, and Scheme 8 shows the formation of other aromatic hydrocarbons [a.96]. At high temperatures, this process is inhibited by polymer crosslinking {871835} {866651} [a.96, a.97]. Benzene formation seems to be a well-established intramolecular cyclisation process of the polyene chain. The reaction is essentially initiated at the chain ends – the mechanism consists of several steps, including the formation of cyclohexadiene as an intermediate, which is then converted into benzene. These cyclic dienes have been effectively identified as thermal degradation products of PVC [a.98]. In the second step, the degradation of the polymer (which has already become the dehydrochlorinated product) continues with cracking and pyrolysis to low hydrocarbons of linear or cyclic structure (more than 170 C1–C7 products have been identified) [a.39]. Loss of HCl leaves a residue with a conjugated polyene structure having both cis and trans arrangements. Polyenes undergo aromatisation and crosslinking and form a wide variety

Scheme 6. Scheme of dehydrochlorination of PVC Reprinted from [a.84] with permission from Elsevier

63


Scheme 7. Benzene formation initiated by tertiary Cl scissions Reprinted from [a.96] with permission from Elsevier

64


Scheme 8. Aromatic hydrocarbon formation from pseudo-unsaturated chain ends of PVC Reprinted from [a.96] with permission from Elsevier

65

Thermal Degradation of Polymeric Materials of hydrocarbon products via two competitive mechanisms [a.98]: first, an intramolecular cyclisation leading to unsubstituted aromatics, e.g., benzene, naphthalene and anthracene (mainly at a temperature range of 200–360 °C); and second, intermolecular crosslinking leading to alkyl aromatics, such as toluene and methylnaphthalene and char at 360–520 °C {878335} [a.98]. However, the mechanism of thermal degradation of PVC is still not very well understood. The basic mechanism processes in PVC thermal degradation are relatively slow initiation, fast allyl-activated propagation of the dehydrochlorination by HCl elimination and formation of polyenes, and termination. Most recent work has been devoted to the initiation mechanism, mainly in an effort to establish relationships between the degradation rate and the amount of irregular structures such as allylic chlorines, internal unsaturations or tertiary chlorines at branch points, as well as the influence on the degradation of labile atoms of special monomer unit conformations. Identification and quantification of toxic chemicals formed in trace amounts during the thermal treatment of pure PVC is a major breakthrough. Radical, ionic, molecular and polaron {671275} mechanisms have been proposed, including a mechanism consisting of an initiation reaction, cis–trans isomerisations, 1,3-rearrangements and propagation {461325}. During the thermal degradation of PVC, the formation of sequences of conjugated double bonds within the polymer chains has been observed [a.99]. With increased splitting off of HCl, the colour becomes more and more intense, but an exact quantitative relationship between colour and amount of HCl evolved is not yet exactly known. In parallel, worsening of the mechanical properties takes place with increasing number of polyene sequences. It has also been reported that for PVC, even after the first stage of decomposition, 10% of the Cl remains trapped in the polymer until higher degradation temperatures {490068}. Therefore, a significant amount of weight loss is due to the simultaneous destruction of the ‘regular’ polyene structure that begins to be formed [a.96]. This causes benzene along with some naphthalene and anthracene to be liberated with HCl in the first stage. The initial rates of PVC degradation at low conversions (0.1–0.3%) have been shown to correlate well with the allylic and/or tertiary chlorine content of PVC. It has been argued that, on of the low concentrations of these structural irregularities in normal PVC, initiation of thermal degradation also takes place at regular monomer units. In addition to the difficulties of identification and quantification of such a small amount of labile allylic and tertiary chlorine-containing structures within a normal PVC main chain, e.g., by the NMR method, it is still difficult to separate their effects on degradation from that of regular polymer units. Dadvand and co-workers [a.100] carried out measurements of the yields of hydrogen chloride evolved during the thermal degradation of PVC and deduced that Py-GC-MS can detect HCl down to at least 1.4 nM. The actual limit was found to be smaller because the

66

Polymers, Copolymers and Blends calculations assumed that HCl was obtained in quantitative yield from PVC on prolonged pyrolysis at 250 °C. The Py-GC-MS results also provide an absolute sensitivity calibration for HCl, and, in addition, permit calculation of the rate constant for HCl evolution from PVC at 250 °C. The specific rate value, as calculated from the initial rate, was found to be ca. 5 × 10–3 s–1 at 250 °C, and was independent of sample size over the range 10–40 μg, corresponding to sample thickness 2.5–10 μm on the pyrolysis filament. An example of the complete sequence of results obtained from the pyrolyses performed on a single sample is shown in Figure 11, and evidently the decay in the size of the HCl peaks showed that the evolution of this compound was virtually complete at the end of the sequence. A computer scanning procedure for the determination of total double-bond concentration resulting from PVC degradation has recently been presented {829658}. This high-sensitivity procedure makes it possible for the monitoring of discoloration on a very small surface and generates very large amounts of experimental data allowing corrections for surface irregularities. By this procedure, conjugated polyenes with seven and more double bonds can be observed even for a slow degradation of monomer units. The measured extinctions correlate with the concentrations of double bonds and allow the estimation of degradation rates. On the other hand, the measurement of stabilisation times and the reaction rates of stabilisers with allylic chlorides and HCl offer the capability to optimise the PVC stabilisation process.

Figure 11. Total ion current chromatograms of a sequence of pyrolyses of the same sample of PVC. The first peak corresponds to the HCl yield from a 5 s pyrolysis at 250 °C. After about 5 min the residue on the pyrolysis filament was then pyrolysed for 5 s to yield the second peak, and so on. Sequence pyrolyses of this kind were performed on samples of different initial size, as follows: 10 mg (46 pyrolysis stages), 30 mg (50 pyrolysis stages) and 40 mg (48 pyrolysis stages) Reprinted from [a.100] with permission from Elsevier

67

Thermal Degradation of Polymeric Materials Results obtained from thermal degradation of vinyl chloride-vinyl acetate copolymers (PVC-co-VAc) indicate that the degradation rate coefficients are higher and the activation energies are lower compared to those of the homopolymers [a.101]. This clearly indicated that the copolymers are less stable than the homopolymers, and therefore addition of VAc to the copolymer structure reduces stability consistently. A proposed reason was that addition of vinyl acetate changes the polarity of the chain and enhances rapid elimination of CH3COOH/HCl. The TG profile of the copolymer showed a sharp change at around 60 °C, indicating volatile impurities, with no further weight loss till 240 °C. The 1H NMR data indicated the presence of hydroxyl groups in the copolymer. These groups disappeared after the polymer was heated, and thus the volatile impurities were initially eliminated. In the case of random copolymers, stepwise degradation of individual homopolymer segments may merge into single-step decomposition, as observed in this report, confirming the investigated copolymers to be random copolymers.

5.3.2 Poly(Vinyl Chloride) Blends Great attention has been paid to PVC, since blending with other polymers improves the properties of this commodity polymer and enhances its application possibilities {767672} {732433}. Knowledge of the thermal behaviour of different blends containing PVC is therefore of industrial importance. Thermal dehydrochlorination of unstabilised PVC occurs at about 100 °C and is also an undesired process in stabilised PVC at processing temperatures (180–200 °C). Dehydrochlorination ing for the formation of conjugated double bonds leads to allylic activation in degrading PVC. Hydrogen atoms of the methylene groups in the allylic moieties are able to form hydrogen bonds with the functional groups (e.g., =C=O) of the component polymer. Sequences of conjugated double bonds may participate in PVC crosslinking [a.110]. As mentioned in an earlier chapter, thermal degradation of PVC is known to occur in two steps, essentially complete dehydrochlorination followed by the decomposition of the resulting polyene chain. Previous instances of Cl radical migration and hydrogen abstraction in the thermal degradation of PVC blends are, for example, the case in PVC/PMMA [a.102], where monomer production in the PMMA is induced at the dehydrochlorination temperature, which is much lower than that normally required to induce depolymerisation. Another example is polystyrene, which, in the presence of PVC, undergoes increased chain scission at the dehydrochlorination temperature [a.102]. These effects cannot be explained by the attack of HCl, which in fact tends to stabilise PMMA. It may also be commented that the recently advanced polaron mechanism cannot explain these striking observations in the degradation of immiscible PVC blends, whereas the radical mechanism involving Cl chain carriers provides a ready explanation. Studies have also shown that the HCl produced by the degradation of PVC can destabilise a second polymer in the same environment, as studied in PVC/PVA blends [a.102, a.103].

68

Polymers, Copolymers and Blends PVC is well known for its efficiency to form miscible systems with various low- or highmolecular-weight polymers, which act as plasticisers. Miscible blends of PVC include its blends with nitrile-butadiene rubber (NBR), chlorinated PE and epoxidised natural rubber (ENR). In PVC/NBR blends, NBR acts as a permanent plasticiser for PVC in applications such as wire and cable insulation, food containers and pond liners. Simultaneously, PVC improves the ozone and chemical resistance as well as thermal ageing characteristics of NBR [a.104]. The results obtained from DMA indicated an increase in the Tg of PVC/ENR blends during thermal ageing at 140 °C for 2 days [a.105]. The presence of NBR in the ternary blend provided a resistance to this large increase in Tg. The thermal ageing also caused an increase in the tan peak widths, which implied a drop in the molecular mobility in blends during ageing. 13C magic-angle spinning (MAS) NMR spectra showed the formation of large-membered rings in the ENR segments in PVC/ENR blends, causing an increase in rigidity of the polymer system. The formation of these ring-opened structures was reduced in the ternary blend at 80 °C due to the better miscibility of NBR with PVC at elevated temperature. Such protection was not observed at higher temperatures, as HCl initiates the ring opening. The 1,3-butadiene/chlorinated PE NMR peak intensity ratio indicated the increased rigidity of the rubber molecules after ageing and changes in the peak widths implying increased heterogeneity in the blends. There was evidence in the relaxation data to indicate that there is a change in the morphology during ageing in the domain size where spin diffusion proceeds with 1HT1 (rotating frame proton relaxation time), but not in the larger domain size where spin diffusion proceeds with 1HT1 (laboratory proton relaxation time). Binary blends of poly(vinyl chloride) (PVC) and poly(vinyl butyral) (PVB) prepared by solution casting showed a high degree of molecular mixing of these two polymers [a.106]. The blends exhibited one major Tg whose position on the temperature scale is lowered with increasing level of PVB. The results showed the effect of ‘dilution’ of PVC by the PVB molecules, which minimised the possible cross-dehydrochlorination reaction on the one hand, and the possible interference of some moieties of PVB with the PVC degradation products on the other. The thermal stability of the blends was found to increase with increase in the PVB content in the blend. The thermal degradation mechanisms of PVC/PVB (Scheme 9) were proposed too. It was claimed that the mechanism could be reasonable, as the degraded blends were totally insoluble in tetrahydrofuran, in contrast to the unblended PVC. This meant that the substitution of the labile chlorine atoms on the PVC chains with much more stable ether group through the intervention of PVB in the thermal dehydrochlorination reaction of PVC was likely to occur, which led to the crosslinking of the polymer (equation (3)). According to the suggested mechanism, the splitting of the acetoxy radical rather than the acetate radical resulting from the thermal degradation of PVB with the subsequent formation of acetyl chloride is more likely to occur. 69


Scheme 9. Proposed thermal degradation mechanisms of PVC/PVB blends Reprinted from [a.106] with permission from Elsevier

The thermal behaviour of PVC/chlorinated 2,4-toluene diisocyanate-based PU (PVC/ U) polymers has been examined [a.107, a.108]. It was found that the decomposition proceeded through a two-step route: the main, decisive degradation stage in the 200– 320 °C temperature range was found to be a result of parallel reactions of PVC and PU decomposition. This was also confirmed by Ozawa–Flynn–Wall kinetic analysis: the activation energy remained constant for degrees of conversion >0.3. The reasons for better thermal stability of some PVC/U blends was explained by analysis of specific interactions between the C=O groups of the urethane segments and the -hydrogen of the chlorinated polymer or dipole–dipole C=O···Cl–C interactions. On the other hand, the rate of diffusion of volatile products through the microphase domain structure differed due to changes in morphology arrangement, thus considerably affecting the overall decomposition route. A detailed elucidation of dehydrochlorination rates of PVC blends with high-impact PS (HIPS) containing 16% non-grafted PS, poly(styrene-co-acrylonitrile) (SAN) and acrylonitrile-butadiene-styrene terpolymer (ABS) containing 27% non-grafted SAN in an inert atmosphere at 180 °C revealed accelerated degradation of the PVC component. The increased content of acrylonitrile in SAN enhances PVC dehydrochlorination. The improved miscibility of thermally treated PVC/SAN blends was related to the formation

70

Polymers, Copolymers and Blends of hydrogen bonds between the allylic methylene group and the nitrile group in SAN. On the other hand, a destabilising effect manifested by the increased dehydrochlorination rate of PVC was observed in PVC/chlorinated rubber blends [a.110]. Another work reported methyl methacrylate-butadiene-styrene terpolymer (MBS) as one of the most effective modifiers of PVC impact strength when added in amounts up to 20% [a.111]. The MBS has a characteristic shell–core structure, which consists of a styrene-butadiene core and a styrene-methyl methacrylate shell that is compatible with PVC and thus works well as a processing aid. A systematic investigation of the effect of various poly(alkyl methacrylate)s on HCl evolution from PVC blends and the length of polyene sequences formed at 180 °C revealed that the resistance of organotin-stabilised PVC increased in the presence of a high concentration of poly(alkyl methacrylate)s [a.109]. The butyl ester was shown to be more effective in this sense than the methyl ester, although some destabilisation of PVC was observed in the presence of the methyl ester used in lower concentration. The concerted thermolysis of PVC and PVAc in PVC/PVAc blends s for the formation of hydrogen chloride and acetic acid [a.111]. These compounds migrated over phase boundaries into adjacent phases of the immiscible blend and cross-catalysed the dehydrochlorination and deacetylation. The co-reactivity of PS in blends with PVC, indicating the formation and reactivity of PS macroradicals, was found to be an important factor during the thermal degradation of PVC/PS blends, while the thermolysis of neat PS starts at 180 °C and is very rapid at temperatures above 250 °C. Unsaturations and low-molecular-weight volatile products that are formed in subsequent decomposition reactions of PS may interact with HCl and other gaseous products of PVC decomposition [a.112]. Polysiloxanes are substantially more stable than PVC, such that increasing the quantity of polysiloxane in the blends gradually increases the stability [a.103, a.104, a.113]. Blends of PVC containing 50% or more poly(dimethyldiphenylsiloxane) (PDMDPS) showed greater stabilisation, which was explained on the basis of crosslinking induced by the presence of PVC (Scheme 10) [a.102]. Since the two polymers are not miscible, the agent responsible for this must be capable of diffusion across the phase boundaries. The effect observed in studies of PVC blends suggested that either HCl or Cl radicals were involved [a.103]. Since crosslinking in PDMDPS (20% phenyl concentration) can be induced by free radicals, it appeared likely that Cl radicals diffusing from the PVC were responsible for abstraction of hydrogen atoms from the methyl groups in PDMDPS and that pairs of such PDMDPS macroradicals crosslinked. However, such crosslinking cannot be induced in poly(diphenylsiloxane) (PDPS) because hydrogen abstraction is not possible due to the absence of methyl groups in the chain. It was expected that the increasing amount of siloxane (PDPS, PDMDPS) results in the accumulation of HCl or Cl radicals, consequently decreasing the catalytic effect and decreasing the rate of dehydrochlorination, as suggested for other PVC blends,

71


Scheme 10. Crosslinking in PDMDPS induced by the chloride radical during thermal degradation of PVC blends with polysiloxane Reprinted from [a.102] with permission from Elsevier

e.g., PVC/poly(-methylstyrene acrylonitrile) (PMSAN) blends [a.113] or PVC/chlorinated PE (E) blends [a.437].

5.4 Polyamides (PA) Polyamides are a very attractive class of construction polymers and have been used for numerous engineering applications because of their excellent tensile properties, chemical and abrasion resistance, and high melting point and fatigue resistance {877977} {660978}. The

72

Polymers, Copolymers and Blends degradation of polyamides is a complex process {881010} {795751} {787552} and can lead to many different products according to a recent review {743554}. The principal degradation product from PA-6 pyrolysis is generally agreed to be the cyclic monomer (Scheme 11), caprolactam, but the question of which additional products are detected seems to depend upon the sample size and the experimental conditions used {886353} {758424}.

Scheme 11. Proposed monomer formation mechanisms in PA-6: (a) initial fast rate; (b) slow rate. The intermediates as drawn do not imply a concerted mechanism, but show the overall rearrangement of atoms and bonds Reprinted from {758424} with permission from Elsevier

73

Thermal Degradation of Polymeric Materials However, recent research agrees that the main route of thermal degradation of PA-6 is the formation of caprolactam with yields as high as 85% – the presence of oligomeric products with nitrile and vinyl chain ends, which are formed as a result of depolymerisation, has been confirmed {890179} [a.17, a.114]. The increase of reaction order of the overall decomposition of PA-6 above 420 °C is correlated with the formation of by-products. Especially, the formation of the cyclic dimer seems to be a second-order reaction, which is responsible for the increase of the overall reaction order. The observed first-order reaction of -caprolactam formation (Table 3) is consistent with the mechanism of cis-elimination suggested, whereby the cis-elimination proceeds via a six-membered intermediate product (Scheme 12) [a.114]; Table 3 [a.115] provides kinetic data.

Scheme 12. Degradation of PA-6 via cis-elimination Reprinted from [a.114] with permission from Elsevier

74


Table 3. Kinetic data on the thermal degradation of polyamides in nitrogen Reprinted from [a.115] with permission from Elsevier Parameter Ea1 (kJ mol–1) log10(k0)1 Order 1 log10(kcat)1 Ea2 (kJ mol–1) log10(k0)2 Order 2

PA-6 162 9.42 1.34 0.35 476 32.95 1.35

PA-6,6 91 2.62 4.57 2.03 310 21.08 1.44

PA-12 2208 160 8.10 –19.68 260 16.47 0.63

PA-6,12 164 8.67 1.97 1.35 400 27.25 1.08

PA-6,6 eliminates as main decomposition product cyclopentanone but also some hydrocarbons, nitriles and vinyl fragments. Cyclopentanone is formed by a cyclic degradation mechanism in the adipic acid unit. Initially, a polymer chain with cyclopentanone end functionality is formed. In a subsequent equilibrium reaction, an isocyanate is formed and cyclopentanone may split off. The resultant isocyanate reacts to form a carbodiamide and the cyclopentanone, causing crosslink reactions. In PA-6,6 these reactions lead to an increased tendency to crosslinking and non-soluble residue formation {861966} {799647}. The degradation of caprolactam led mainly to oligomers with different endgroups (–C–N amongst others) and the cyclic dimers of caprolactam. Degradation of mixtures of caprolactam with melamine and cyanuric acid did not lead to additional products. In these degradation experiments, caprolactam was found to be much less reactive than cyclopentanone. The formation of cyclopentanone during the degradation of PA-6,6 was studied via Py-GC-MS and it was reported that the degradation of PA-6,6 at 400 °C leads to different cyclopentane derivatives, cyclopentanone and ammonia condensation products [a.116]. Conversely, the thermal degradation reactions of polyamides such as PA-6,10, for example, produce mostly linear or cyclic oligomeric fragments and monomeric units. Primary polyamide chain scission (C(O)–NH or NH–CH2 bonds), hydrolysis, homolytic scission, intramolecular C–H transfer and cis-elimination are all suggested from the product distribution as possible mechanisms {502570}. For PA-12, it has been found that lactams (cyclic monomer and dimer) are the major primary products of the thermal degradation; however, olefinic nitriles and -olefins were also found. In the case of PA-6,12 the formation of cyclic oligomers has been found to occur as a result of thermal degradation, but also some dinitriles have been identified {502570}. Studies seeking an understanding of the possible chemical reactions of melamine cyanurate (MC) as a flame retardant investigated the reaction between cyclopentanone (as a model

75

Thermal Degradation of Polymeric Materials for the degradation products of PA-6,6) and caprolactam (as a model for the degradation products of PA-6) with melamine, cyanuric acid and MC. The results showed that degradation of cyclopentanone lead to mainly self-condensation products [a.116]. The degradation of mixtures of cyclopentanone with melamine, cyanuric acid and MC lead to the same products, but the reaction products of cyclopentanone with melamine (NH3, HN=C=NH) or cyanuric acid (NH3, HN=C=O) decomposition products were found at 400 °C. It was further found that the addition of MC to PA-6,6 or PA-6 had no influence on the type of products formed, as expected, since MC acts mainly as a source of NH3, which is also formed during the degradation of PA-6 and PA-6,6. These results showed that the reactions observed for cyclopentanone and caprolactam took place during the degradation of PA-6,6 and PA-6, respectively. The reactions of the decomposition products of MC with the decomposition products of PA-6,6 lead to increased formation of insoluble degradation products, whereas MC hardly influences the amount of insoluble degradation products of PA-6, therefore highlighting the difference in the degradation mechanisms of the two polymers. The degradation products formed in PA-6,6 (cyclopentanone) may crosslink with MC degradation products (mainly NH3), resulting in less flammable high-molecular-weight structures. PA-6 degrades to less reactive compounds that do not crosslink [a.116]. Results from modelling of thermal degradation of PA-6,6 showed that the first bond that breaks leaves a free methylene radical and a free carbonyl radical, and also that the carbonyl carbon is the part of the PA-6,6 chain that is most susceptible to free-radical attack [a.118]. When the methylene radical folded back and attacked this susceptible carbonyl carbon, then cyclopentanone was formed.

5.4.1 Poly(Ester Amide)s The segmented block copolymers – poly(ester amide)s – are of great interest because they possess good solvent resistance, good mechanical properties and a wide range of application temperatures {884042} {882869} {816008}. It is well known that the hard polyamide blocks remain relatively unaffected at the decomposition temperature in an inert nitrogen atmosphere, although some scission of –NH bonds may take place. The thermal degradation properties of polyesters can be improved with the introduction of amide groups in the main chain, since they can give rise to strong intermolecular hydrogenbond interactions {815984} [a.118]. Poly(ester amide)s constitute a group of polymers that can replace polyesters in certain applications in photographic emulsions, magnetic tapes, adhesives, dielectric materials, biomedicine, interfacial agents and additives for the paper industry {726737} [a.118, a.119].

76

Polymers, Copolymers and Blends Studies on poly(ester amide)s derived from 6-amino-1-hexanol and glutaric acid (i-PEAG6) indicate that polymorphism exists involving packing modes similar to those found on aliphatic polyamides. Structural studies indicate that i-PEAG6 preferentially adopts a hydrogen-bonded sheet structure with an antiparallel molecular chain arrangement [a.118]. Significantly, the poly(ester amide) can be processed from the melt state, since no evidence of decomposition can be detected at temperatures lower than 200 °C. Above that, imide ring formation becomes a significant degradation mechanism. Consecutive sheets are sheared along both the hydrogen-bonding and the chain-axis directions. The polymer is hydrolytically degradable through the cleavage of ester bonds, and the degradation process is not accelerated by an imide ring formation, as happens with some succinic derivatives [a.118]. Different studies on related poly(ester amide)s derived from succinic acid and amino alcohols demonstrated that the degradation proceeded quickly due to the rapid formation of succinimide rings in the isoregic polymers [a.119]. Thermal decomposition of segmented poly(ester amide)s in a nitrogen atmosphere occurs by a rapid depolymerisation process, and this depolymerisation takes place through a chain scission mechanism mainly initiated in the thermally unstable polyether blocks and follows a first-order rate law {886353}. Degradation occurs much faster with lower activation energy as the polyamide hard-block molecular weight decreases or the soft-segment concentration increases. After this initial drop, both the tensile strength and elongation at break, in general, change marginally with either ageing time or temperature for all the polymers. It is interesting to note that the percentage retention of physical properties is high with higher hard-block molecular-weight polymers. But as the polyamide-block molecular weight decreases, the percentage retention of properties decreases. Polymers with high hard-block molecular weight are less susceptible to ageing and subsequent degradation in properties than polymers with low hard-block molecular weight.

5.4.2 Liquid-Crystalline Polyamides Liquid-crystalline polyamides are commercially important materials due to their excellent properties at high working temperatures, besides their high strength and low coefficients of thermal expansion {748747}. Liquid-crystalline order is exhibited either in the melt (thermotropic liquid-crystalline polymers (L)s) because of the rigid character of the molecules or in solution (lyotropic Ls). In the main-chain Ls, the rigid groups are part of the polymer chain, while in side-chain Ls they are attached to a (flexible) polymer backbone. Main-chain Ls are potentially useful materials as the molecular orientation enables the development of strong products. The melting point of lyotropic Ls is above their decomposition temperature, so they can only be processed from solution. The melting point of thermotropic Ls is lower, so they can be processed from the melt. Because of this advantage, thermotropic Ls find a variety of applications in aerospace, electronics, etc.

77

Thermal Degradation of Polymeric Materials Decomposition investigations on the wholly aromatic thermotropic liquid-crystalline polymers – the polyimide made of 1,2,4,5-benzenetetracarboxylic dianhydride (PMDA) and 1,3-bis(4-(4-aminophenoxy)cumyl)benzene (BACB), and the polyamide made of terephthaloyl chloride and BACB – showed that the polyamide is much less thermally stable than the polyimide [a.120]. The evolved gases were found to be H2O, CO, CO2 and various hydrocarbon fragments. The substantial amount of CO2 detected during the decomposition is due to degradation of the carbonyl functional groups from the polyimide liquid-crystalline polymers. The activation energies for the initial thermal degradation of this polyimide in nitrogen and air were 236 and 201 kJ/mol, and those for polyamide were 207 and 219 kJ/mol, respectively. A jump in the activation energy was observed around 40 wt% losses, beyond which it decreased in the case of the polyimide. However, an unusual observation was made during the degradation of polyamide in that the apparent activation energy values were found to be higher in an environment of air than in nitrogen.

5.4.3 Polyamide Blends In general, the thermal degradation of blends does not follow a regular behaviour and is not much dependent on the compatibility of the blend system. The degradation of polymer blends is influenced by the degradation conditions, by the structures of the components of the polymer blends, and by potential co-reactivity between the component polymers and/or their degradation products, which may lead to new chemical species, e.g., grafted copolymers, and/or influence in either a positive or negative sense the final stability of the blend {888209} [a.6]. There is insignificant theoretical basis for correlating the thermal degradation parameters to compatibility. Systematic knowledge of blend stability and the kinetics of degradation of the blends may give rise to some idea of the extent of chemical interactions occurring between the components under decomposition conditions, their bond strength, activation energy and melting temperature as evidenced by changes in heat energy. As mentioned earlier, polyamides are useful engineering polymers, which, however, are very sensitive to and brittle at low temperatures. Blends of polyamides with rubber have been extensively studied in order to obtain new macromolecular materials with good impact properties {814817} {783791}. Thermal degradation of PA-6/natural rubber blends with maleic anhydride (MA) showed that the DTG decomposition peak increased with an increase of rubber content {497853}. Higher polyamide main-chain mobility was also expected due to the presence of the rubber particles. Although higher mobility was also expected to contribute to the increase of decomposition temperature, grafting reduced the intensity of degradation due to the reduction in chain mobility. MA-containing blends showed smaller DTG peaks at 78

Polymers, Copolymers and Blends this region, relative to the same natural rubber composition but without MA. This also suggested the occurrence of grafting reactions. Polyamide-containing materials presented a weight loss of approximately 3 wt% at around 100 °C due to loss of water, whereas MAcontaining materials presented a weight loss due to free MA sublimation at approximately 200 °C. Natural rubber (with or without MA) showed weight loss due to degradation at around 400 °C, whereas all polyamide-containing blends showed degradation weight loss at higher temperatures (around 500 °C). Fundamentally, after polymer degradation above 500 °C, only MA-containing materials showed a residue up to 800 °C. This evidence indicated that reactions took place during processing and caused the formation of both gel and graft copolymer. In situ compatibilisation of PA-6/natural rubber blends with MA resolved into rubber reticulation. However, the 15 wt% rubber blends showed the same residual amount as the neat natural rubber with 3 wt% [a.76]. In binary immiscible blends, good performance in of the mechanical properties, polymer compatibility and modification of blend morphology can be obtained by using a compatibilising component [a.121]. The thermogravimetry curves for (PA-12,10 or PA-6,10)/EPDM/EPDM-g-MA blends showed essentially the same profile as the pure polyamides. It is suggested that the thermal degradation of the aliphatic polyamides studied included first the scission of the weakest C–N and –C(O)–NH bonds. The breakdown of the strongest C–N bonds occurred at temperatures exceeding 400 °C for pure polyamides. The high activation energy values for pure polyamides as well as for the binary or ternary blends of ca. 225 kJ/mol were associated with random chain scission and the formation of volatile products that probably occurs with multiple competing steps during the degradation. Similar activation energy values have been reported for pure polyamides such as PA-6,6, PA-6,10, PA-6 and poly(ether ester amide)s {502570}. The decrease in the activation energy values of polyamides, due to the addition of a component or in a blend formation, was observed for mixtures of PA-6,10, PA-6,6, PA-11 and PA-12 with a fire-retardant ammonium polyphosphate and in blends of PA-6 with functionalised or non-functionalised polypropylene {502570}. The presence of EPDM-g-MA in the blend decreased the activation energy and significantly affected the absorption bands of solid residues in the FTIR spectra. This behaviour suggested a decrease in the thermal stability of polyamide due to the presence of functionalised EPDM.

5.5 Polyurethanes (PUs) Polyurethanes (PUs) composed of polyether or polyester soft segments and diisocyanatebased hard segments are well-known tough materials and are usually used as an additive to enhance the toughness of brittle materials as well as improve their thermal properties {894608}. Because of incompatibility between the hard segments and the soft segments, PUs 79

Thermal Degradation of Polymeric Materials undergo microphase separation resulting in hard-segment domains, soft-segment matrix and urethane-bonded interphase {831060}. The hard-segment domains act as physical crosslinks in the soft-segment matrix. The primary driving force for phase separation is the strong intermolecular interaction of the urethane units, which are capable of forming intermolecular hydrogen bonds [a.438]. Owing to such interactions, interconnected or isolated hard segments remain distributed in the soft-segment matrix, though the soft domain may contain some hard segments dissolved in it, which is evident from the hydrogen bonding of the urethane –NH groups with the oxygen of the ether or ester linkages {760903}. These kinds of PU are utilised mainly as water dispersions (coatings, adhesives) and also as biomedical devices, temperature-sensing elements, polymer electrolytes, etc. Recent technological interest has been concerned with studies on composites containing conductive polymers and an inert polymer matrix. The general PU decomposition mechanism is shown in Scheme 13 [a.122]. In general, there are three main pathways for the initial degradation of the urethane linkage, which are: dissociation to isocyanate and alcohol; dissociation to primary amine, olefin and carbon dioxide; as well as the formation of a secondary amine with elimination of carbon dioxide [a.1]. Polyurethanes degrade at low temperatures (200–300 °C) with the formation of a nitrogenfree residue and a nitrogen-containing yellow smoke {875580} {871999}. With increased temperature, the residue further decomposes to smaller compounds, and the yellow smoke yields nitrogen-containing products like HCN and acetonitrile. Extensive research on the thermal degradation of a 13C-labelled MDI-based polyurethane found that HCN and all the other nitriles generated during high-temperature decomposition originate in the thermal fusion of the aromatic ring, the nitrile carbon being the 2, 4, or 6 carbon of MDI [a.99]. Several authors report the evolution of nitrogen-containing compounds (acetonitrile, acrylonitrile, propionitrile, pyrrole, pyridine, aniline, benzonitrile, quinoline and phenyl isocyanate) during the thermal decomposition of polyurethanes as displayed in Scheme 14 for chlorinated PU [a.1, a.20, a.122, a.125]. Grassie and co-workers [a.123, a.124] found that under inert conditions at temperatures above about 210 °C the polyurethane linkage disappears without any volatile products being formed, and the initial degradation step is seemingly a simple depolymerisation reaction. The two monomers are the primary products, and all the other products, which include carbon dioxide, butadiene, tetrahydrofuran, dihydrofuran and water as volatile products and carbodiimide and urea amide in the condensed phase, are formed from the monomers in a complex set of secondary reactions while they are diffusing from the hot polymer. This is unlike under oxidative conditions, whereby the first step involves the scission of the polyurethane molecule into primary amine, carbon dioxide and propenyl ether species, the latter leading to propene formation. The mechanism is reduced to a

80


Scheme 13. Decomposition mechanisms of PU

depolymerisation process followed by radical breakdown of the polyol chain in conjunction with simple radical formation. The radicals formed can explain the formation of various gaseous species during the thermal decomposition.

81


Scheme 14. Decomposition products (including radicals) of chlorinated PU Reprinted from [a.122] with permission from Elsevier

82


5.5.1 Thermoplastic Polyurethanes Thermoplastic polyurethanes constitute an important group of polymers that have found wide applications due to their fundamental physical properties that make them a polymer of choice {867499} {662846}. Studies on the thermal degradation of ester-based and etherbased thermoplastic polyurethane elastomers synthesised from MDI reported that the two polymers have a weight loss taking place in the range 280–485 °C {794286} [a.125]. The poly(ether-urethane) loses 85% of its initial mass, while the poly(ester-urethane) loses 90%. The poly(ether-urethane) generated oligomers with the base peak at m/z = 71, as detected by TG-MS. In the case of the poly(ester-urethane) no such oligomers were identified, but the most abundant organic compound was found to be cyclopentanone, evolving from the poly(butanediol) adipate {865110}. The PUs yielded 1,4-butanediol, whose formation was favoured under nitrogen at lower temperatures. The formation of polycyclic aromatic hydrocarbons (PAHs) for both polymers seem to follow the precursor theory. The concentration of naphthalene was higher than of phenanthrene and pyrene. The PAH development was reportedly higher at high temperatures (950 °C) and the atmosphere did not seem to have any essential impact on it. The researchers concluded that PAH formation is not dependent on the structure of the long-chain diols. The properties of molten thermoplastic PU elastomers show a complex dependence on their molecular, structural and morphological nature {894608} {825464} {706910}. A theoretical approach to the kinetics of the thermal decomposition of PU under conditions of thermoplastic processing was described. The fundamental kinetic equation described the decomposition reaction and the reverse reaction (formation reaction) – which is dependent on the system of measurement and processing – as a function of the molecular weight (endgroup concentration) of the original product, determined from the rate constants for the decomposition reaction and back-reaction. The consideration of the limiting value for t is in agreement with the equilibrium constant. Consequently, the development of physical characteristic functions of thermoplastic polyurethane elastomers independent of the system of measurement was possible. Studies have been performed on the degradation of segmented PU elastomers based on MDI/1,6-hexamethylene diisocyanate (HMDI), polyoxypropylenediol and low-molecularweight chain extenders: 1,2-propanediol (PD) or 3-chloro-l,2-propanediol (D) [a.126, a.127]. It was found that introduction of 3-chloro-1,2-propanediol into the polymer structure changes both the morphology and the flammability behaviour. A possible mode of stabilising action of this internal flame retardant is discussed in of the relatively rough structure of polyurethanes containing D, which may play an important role by inhibiting polymer-specific condensed-phase reactions. The study suggested that some cooling and fuel-diluting effects of the evaporated CO2 or crystallised water may push the heat- and light-emitting zone of the flame further away from the pyrolysing polymer

83

Thermal Degradation of Polymeric Materials surface – increased distance cuts down heat transfer. It was further suggested that the rate of evolution of volatile products from the degrading polymer was probably controlled by diffusion, and the primary products of degradation may undergo secondary reactions while they are retained in the hot zone. Mequanint and co-workers [a.128] studied segmented polyurethane dispersions containing bound phosphate groups on the soft segment which have been synthesised from a phosphate-containing macroglycol, a diisocyanate and a chain extender. TG study revealed three stages of thermal degradation – the initial weight-loss temperature for the phosphated polyurethane dispersions was lower than the ones without phosphate, indicating that degradation started at the soft segment, which contains the phosphate groups. The degradation profile of the dispersions was also dependent on the neutralising base. Thermal degradation of polyurethanes based on hydroxyl-terminated polybutadiene (HTPB) and poly(12-hydroxystearic acid-co-trimethylolpropane) ester polyol (PEP) with varying compositions has been studied by TG and Py-GC techniques [a.129]. The polyurethanes were found to decompose in multiple stages and the kinetic parameters were found to be dependent on the method of their evaluation. The activation energy for the initial stage of decomposition was found to increase; for the main degradation stage it decreases with the increase in PEP content. Py-GC studies on ammonium perchlorate-filled polyurethanes (solid propellants) showed that the major products during the pyrolysis were C2 and C3 hydrocarbons and butadiene. The amount of C2 fraction in the pyrolysate increased with solid loading, as well as with the HTPB content in the polyurethanes. Wen and co-workers [a.130] evaluated the thermal degradation properties of polyanilinefilled (PANI) PU/PMMA interpenetrating polymer networks (IPNs) by comparing the decomposition temperatures at various percentage weight losses and integral procedural decomposition temperature, which is an index of thermal stability of the system based on the thermogram area [a.131]. It was observed that the initial thermal stability of PANI is enhanced after incorporation into the PU/PMMA matrix. This is due to PANI entrenchment inside the PU/PMMA matrix, which drastically reduces the moisture absorption behaviour of PANI. Further TG investigations were conducted for polypyrrole (PPy)/TPU composites, whereby TPU and chemically prepared PPy were analysed over the temperature range of 100–600 °C in an inert atmosphere at a heating rate of 20 °C/min. The composite thermal properties were found to be comparable to those of pure TPU, though a decrease in Tg of the soft segment was observed in the composite from DSC experiments. This effect was ascribed to interaction of PPy –NH groups with either carbonyl or ether oxygens of TPU, leading to phase separation of the hard and soft domains. Liao and co-workers [a.132] and Gregory and Liu [a.133] have performed DSC analysis of IPNs of PU and polythiophene with urethane substituted in the position. The former found an increase of Tg values with increase of polythiophene content, indicating the probability of permanent entanglement and interlocking significantly enhanced due to high

84

Polymers, Copolymers and Blends compatibility when polyether-type PUs were employed in the IPNs. The latter noticed that the conducting composites possessed good thermal properties, with Tg and Tm values of 62 and 265 °C, respectively. PU/PMMA IPNs have been found to be stable up to 200 °C; further degradation up to completion at ca. 500 °C was undergone via a three-step process [a.135]. In the case of the PANI-filled IPN system, a two-step weight loss was observed and as well the pristine PANI could not degrade completely at this temperature.

5.5.2 Polyurethane Foams Polyurethane foams are used widely for insulation in construction, transportation and industrial applications. They are applied in, for example, refrigerators, freezers, cavity walls, floor s and roofing materials [a.136]. The isocyanate–polyol stoichiometry plays an important part in determining the structure and thermal properties of PU foams as it controls the relative amounts of urethane linkages. An important characteristic is the amount of isocyanate used, commonly expressed as the isocyanate index – a measure of the amount of isocyanate used relative to the theoretical stoichiometric amount. As a result of a considerable excess of isocyanate, foams with very high isocyanate index have a large proportion of the trimeric cyclic isocyanurate structure, which imparts resistance and thermal stability to the polymer. Also, the isocyanurate linkage has an inherently higher thermal stability than that of the urethane linkage. Selected chemical characteristics and flammability of rigid urethane-modified polyisocyanurate (UMPIR) foams are shown in Table 4.

5.5.2.1 Rigid Polyurethane Foams Yoshitake and Furukawa presented Py-GC-FTIR studies on the thermal degradation mechanisms of ,-diphenyl alkyl allophanates and carbanilates as model compounds for crosslinking sites in polyurethane networks [a.137]. The primary degradation reaction was dissociation of allophanate into phenyl isocyanate and alkyl carbanilate,

Table 4. Characteristics of UMPIR foams based on PPG and different polyols Reprinted from [a.136] with permission from Elsevier Polyol type

Aliphatic polyester polyol Aromatic polyester polyol Polyether polyol Polyether polyol

UMPIR foam 1 2 3 4

Polyol molecular weight – – 200 2000

Polyol Isocyanate LOI Isocyanate weight weight index fraction fraction 0.25 0.75 22.6 340 0.25 0.75 22.4 490 0.25 0.75 20.4 220 0.25 0.75 22.0 2078

85

Thermal Degradation of Polymeric Materials followed by dissociation of the alkyl carbanilate into phenyl isocyanate and alcohol. Decarboxylation of the ethyl carbanilate fragment also took place slowly. A small amount of diphenyl carbodiimide was observed at the pyrolysis temperature of 450 °C. In addition, decarboxylation of the isopropyl carbanilate fragment took place at 550 °C, while a small amount of diphenyl carbodiimide was observed from 350 to 550 °C. Dick and co-workers [a.134] have reported on the use of both in situ solid-state 1H and 13C NMR to characterise the condensed-phase residues obtained upon degradation under inert and oxidative conditions of rigid poly(urethane-isocyanurate) foams based on polypropylene glycol (PPG) and MDI. The TG, DSC and pyrolysis experiments revealed that the biggest difference in the behaviour of the foams is under inert rather than oxidative conditions. It was concluded that the difference in the observed flammability of the samples derives from differences in the volatile release profiles upon degradation in an essentially inert environment. Both DSC and high-temperature 1H NMR results clearly indicated that there are two major scission processes occurring within the polymers. The lower-temperature process was attributed to the scission of the urethane links, whilst the higher-temperature process (which became increasingly significant as the isocyanurate content of the polymer increased) was ascribed to the scission of the isocyanurate linkages. In addition, 13C NMR data on the residues clearly showed that PPG is lost preferentially from those materials with the highest urethane-to-isocyanurate ratio. It was claimed that the lower thermal stability of the urethane links leads to facile depolymerisation to yield free PPG from those foams where urethane dominates over isocyanurate linkages, and also that the lower-molecularweight PPG from these foams is more volatile than that in the isocyanurate-dominated foams. Lastly, the work also claimed that the more rigid crosslinked network of the predominately isocyanurate-linked foams restricts the diffusion of volatile species formed by and subsequent to the scission of any urethane bonds or the glycol backbone. In efforts to understand the thermal degradation of rigid polyurethane foams, studies have been done on polyurethane foams prepared with different fire-retardant (FR) concentrations and blowing agents – 1,1-dichloro-1-fluoroethane (a member of the ‘freon’ family) and pentanes [a.106]. The standard flammability tests indicated an optimum fireretardant concentration of about 15 wt% for foams using 1,1-dichloro-1-fluoroethane as the blowing agent, while no optimum condition was determined with pentane. The percentage mass retained (PMR) values or char yields have a linear relationship with combustion flame temperature in both series of blowing agents. The solid-state 13C NMR studies clearly showed that pentane is chemisorbed during the polymerisation and is retained within the foam matrix. The chars had lower concentrations of methylene and oxygenated aliphatic carbons, but a subsequent increase in the amount of aromatics was observed. The fire retardant investigated preserved the chemical structure of the polyurethane foam, and, therefore, resulted in a higher PMR or char yield. The TG experimental data showed that the maximum combustion reactivities of the chars have a linear relationship with the FR concentration in the parent foams. Py-GC/MS results

86

Polymers, Copolymers and Blends indicated that the aliphatic oxygenated functional groups are the first to evolve during the pyrolysis and combustion of the polymeric structure. Finally, this study showed that the addition of fire retardant to the foam formulation results in lower concentrations of small molecules being volatilised, therefore preserving the original chemical structure of the parent foam. However, the fire retardant investigated was not effective for the pentane series, and gave higher char aromaticities and PMR values than those reported for the 1,1-dichloro-1-fluoroethane series. Investigations by Font and co-workers [a.138] on the thermal degradation behaviour of rigid polyurethane foams showed an increase of the yield of light hydrocarbons (methane, ethylene, etc.) as the pyrolysis temperature increased. ing the pyrolysis products through a furnace prior to detection identified the secondary products from the formation/cracking reactions of the various primary compounds. GC-MS was used to identify the volatile and semi-volatile organic compounds generated by the thermal degradation reactions. A two-step parallel reaction model was successfully implemented for the interpretation of the decomposition characteristics – activation energy was in the 133–190 kJ/mol range. Grittner and co-workers studied the pyrolysis of semi-rigid poly(ether-urethane) foams at 700 and 800 °C and reported weight yields of methane (16%), ethylene (4.8%) and benzene (4.6%) [a.139]. Increasing reaction temperatures lead to oils that are poorer in hetero-compounds and richer in aromatics. Modesti and co-workers worked on poly(isocyanurate-urethane)s stabilised with expandable graphite (EG) or a mixture of EG and triethyl phosphate (TEP) [a.431]. The results showed a considerable decrease in rate of heat release (RHR) with respect to unfilled foams. In particular, for EG/TEP foams, the higher the triethyl phosphate content, the higher the rate of heat release decreases. The only hazard observed is an increase of CO/CO2 weight ratio in the presence of very high content (25%) of expandable graphite – this effect was not shown when increasing the TEP amount. Pielichowski and co-workers [a.141] studied the effect of sodium dihydrogenphosphate, trisodium pyrophosphate and sodium aluminocarbonate on the thermal decomposition of rigid polyurethane foams, based on 4,4´-diphenylmethane diisocyanate, diphenyl2,2-propane-4,4-dioxyoligo(ethylene oxide) and oxyalkylenated toluene-2,6-diamine, blown with pentane. TG data showed that there was a stabilisation effect of additives in the initial stage of degradation and the decomposition proceeded in two steps up to 600 °C. For phosphate-stabilised rigid PU foams, the activation energies remained stable over a broad range of degree of conversion, while for carbonate-containing foam two regions of activation energies were observed. Further advanced kinetic analysis by a nonlinear regression method revealed the form of kinetic function that was the best approximation for experimental data – for a two-stage consecutive reaction, the first step was the Avrami–Erofeev nucleation-dependent model, and the second step was a chemical reaction (nth-order) model. It was further shown that the global stabilisation effect is a

87

Thermal Degradation of Polymeric Materials multi-stage process whose initial conditions are of critical importance in governing the nature of the entire process.

5.5.2.2 Elastic Polyurethane Foams Ravey and Pearce’s [a.143] study of the pyrolysis of a flexible commercial polyurethane foam showed that the composition of the products depends on the conditions of the pyrolysis. When the volatiles were removed rapidly from the system, they contained TDI. However, under confined conditions, the TDI was replaced by diaminotoluene (DAT). The results were explained by assuming two decomposition routes to be operative – one leading to the regeneration of the two source materials (TDI and the polyol), and the other leading to DAT and a double-bond-terminated polyether. The TDI route was faster; however, if the volatile TDI was confined to the pyrolysis zone, equilibrium was established between the urethane group and the TDI/polyol system. Such conditions favoured the slower, though irreversible, route leading to the formation of DAT. On pyrolysis, the urea groups in the foam dissociated into TDI and DAT, recombining in the vapour phase to form an aerosol of polyurea. TG results showed that the weight loss for both aliphatic and aromatic elastic poly(esterurethane) foams under nitrogen followed similar trends, with weight loss occurring in one step – the onset of degradation begins at 260 °C, followed by an increase in degradation rate, up to ca. 450 °C [a.136]. The aromatic polyester content in UMPIR resulted in a slightly greater weight loss below 420 °C, but above this temperature the UMPIR showed greater stability. At lower temperatures, poly(ether-urethane) foam with high MW showed greater stability, and together with aromatic poly(ester-urethane) foams dominated at higher temperatures with a char residue of 44 and 45%, respectively. Gao and co-workers related the fire behaviour of PU foams based on PPG-425 with the degradation of these materials on a molecular basis, as studied by laser pyrolysis time-offlight mass spectrometry [a.144]. They found that as the isocyanate index of the foams increased, the yield of polyol (the major volatile product) decreased whilst the yield of CO2 (found to be a product of the decomposition of isocyanurate) increased. These researchers thus ascribed the increase in limiting oxygen index (LOI) with isocyanate index to a reduction in the production of flammable polyol – an observation consistent with the known degradation behaviour of urethane-based polymers. Lefebvre and co-workers have described the combustion of flexible polyurethane foams under cone calorimeter conditions as a two-step process: first, the foam melts to give a carbonaceous part and a tar; and second, the tar burns with a relatively high production of heat [a.145]. Esperanza and co-workers studied the decomposition in an inert atmosphere of varnish wastes based on polyurethane [a.146]. Two main peaks (reaction zones) in the devolatilisation rate were observed. However, a mechanism based on three parallel reactions (activation 88

Polymers, Copolymers and Blends energies of 90, 187 and 341 kJ/mol) with a power-law dependence on the solid mass fractions was required to describe with sufficient accuracy measurements carried out at different heating rates.

5.6 Polyesters Polyesters are important thermoplastic resins, which can be obtained by the condensation reaction of glycols and acids or anhydrides {798163}. For example, the condensation polymerisation of ethylene glycol or butylene glycol and terephthalic acid produces the well-known polyesters, poly(ethylene terephthalate) (PET) or poly(butylene terephthalate) (PBT), respectively. The mechanism of thermal degradation of polyesters has been studied for many years {817822} {776328} {756176} {670828} with the help of model compounds. However, there is still some current controversy on the crucial point whether the primary thermal decomposition reactions that occur in polyesters involve radical or ionic processes {777330} {490068}. Studies on the thermal stability of the di-n-alkyl esters of poly(itaconic acid) show that depolymerisation is the main mode of degradation mostly initiated at the chain ends due to the presence of chain-end unsaturations [a.147, a.148]. The extent of this mode of initiation is much greater than for the corresponding esters of poly(methacrylic acid). This has been ascribed to the introduction of chain-end unsaturations during chain transfer to monomer in radical polymerisation [a.148]. Thermogravimetry studies performed on poly(di-n-propyl itaconate) (PDnPI), poly(diisopropyl itaconate) (PDiPI), poly(di-n-butyl itaconate) (PDnBI), poly(diisobutyl itaconate) (PDiBI) and poly(di-sec-butyl itaconate) (PDsBI) show that PDiPI and PDsBI thermally degrade in a similar manner, with deesterification, together with depolymerisation, being a significant mechanism [a.148]. PDnPI and PDnBI also degrade in a similar manner by depolymerisation, with chain-end initiation being more important than random main-chain scission initiation. The results for PDiBI indicate more similarity to the n-alkyl esters than to the other branched-chain esters, although there are indications that this polymer decomposition route is not an insignificant thermal degradation mechanism. TG studies on polycyanurate networks (exhibiting good to outstanding fire resistance, as the ignition and fire resistance of solid polymers are governed by short-term thermal stability) prepared by thermal polymerisation of monomers containing two or more cyanate ester functional groups showed that the thermal stability of the polycyanurates was essentially independent of monomer chemical structure, with the major mass loss occurring at about 450 °C for all materials [a.149]. Analysis of the solid-state and gas-phase thermal degradation chemistry indicated a thermal decomposition mechanism for polycyanurates that begins with hydrocarbon chain scission and crosslinking at temperatures between 400 and 450 °C with negligible mass loss. This is followed by decyclisation of the triazine 89

Thermal Degradation of Polymeric Materials ring at 450 °C liberating volatile cyanate ester decomposition products. The solid residue after pyrolysis increases with the aromatic content of the polymer and incorporates about two-thirds of the nitrogen and oxygen present in the original material.

5.6.1 Poly(Ethylene Terephthalate) (PET) Khemani’s {827029} studies on the thermal degradation of poly(ethylene terephthalate) (PET) at 280 °C over an extended period of time showed a gradual decrease in the amount of acetaldehyde evolved with time as the decline approached an asymptotic value. The degradation mechanism proposed showed that three different routes – the first involving the hydroxyl endgroups, the second the vinyl endgroups and the third mid-polymer chain scission – generated the acetaldehyde. It was further suggested that the hydroxyl- and vinyl-endgroup-based acetaldehyde generation routes depleted with time, leaving behind the inexhaustible chain-scission route. An earlier summary of results on the thermal degradation of PET concluded that a -CH hydrogen transfer is involved, leading to the formation of oligomers {706082} with olefin and carboxylic endgroups {845462} {826911} {784191} [a.150]. It has been suggested that a thermal oxidative degradation process at high temperatures starts with the formation of a hydroperoxide at the methylene group, followed by homolytic chain scission {730485} {695399} [a.151]. There are even suggestions that hydroperoxide groups not only play an important role in inducing the thermal and photooxidative degradation of PET, but are also important intermediates in such reactions {485571}. Also, reducing the initial carboxyl content in the resin has been shown to reduce PET hydrolytic stability [a.65]. The thermal degradation and melt viscosity of ultra-high-molecular-weight PET (UHMW-PET) were studied in the melt phase under a nitrogen atmosphere {592654}. The degradation rate reportedly increased with the molecular weight of the polymer. The high degradation rate of UHMW-PET was interpreted by the differences in the terminalendgroup concentrations. The activation energy of decomposition of UHMW-PET with Mw = 2.3 × 105 at 300 °C was determined as 41 kcal/mol. A comparative study has been conducted into the thermal and thermooxidative degradation of PET and PBT polymer films and their model compounds, ethylene dibenzoate and butylene dibenzoate, in an oxygen atmosphere at 160 °C {832485}. On the basis of the compounds identified by GC-MS, a mechanism was proposed for the degradation of the model compounds that involves oxidation at the -methylene carbon with the formation of unstable peroxides and carboxylic acids [a.153]. From the studies performed under N2 at 160 °C, it was concluded that benzoic acid and esters are products of the thermal degradation as illustrated in Schemes 15 and 16, while benzoic and aliphatic acids, anhydride and alcohols were due to thermooxidative degradation.

90

Polymers, Copolymers and Blends In contrast to the thermooxidative degradation of other polymers, for PET and PBT, thermal degradation plays an important role especially at the beginning, but generally PET is more stable towards degradation than PBT. It is known that, at the processing temperatures of 250–280 °C, thermal, oxidative and hydrolytic degradation of PBT may occur. Nevertheless, PBT has found applications especially in the field of thermoplastics for injection moulding {490068} {832485}.

5.6.2 Biodegradable Polyesters

5.6.2.1 Poly((R)-3-Hydroxybutyrate), Poly(L-Lactide) and Poly(-Caprolactone) Poly((R)-3-hydroxybutyrate) (PHB), poly(L-lactide) (PLLA) and poly(-caprolactone) (PCL) are biodegradable polyesters used in practical applications mainly because they combine remarkable physico-mechanical performances with biodegradability, compostability and compatibility with different forms of waste disposal [a.439]. As shown in Scheme 17(a), thermal degradation of PHB has been suggested to occur almost exclusively by a nonradical random chain scission reaction (cis-elimination) involving a six-membered ring transition state [a.150] {852663} {832485}.

a Hydroxylic groups were confirmed by derivatisation using tert-butyl dimethylsilylmethyltrifluoroacetamide (TBDMSTA) with 1% tert-butyldimethylsilyl chloride (TBDMSCl)

Scheme 15. Proposed thermal degradation mechanisms for ethylene dibenzoate during thermal degradation of PET Reprinted from {832485} with permission from Elsevier

91


a

Hydroxylic groups were confirmed by derivatisation using TBDMSTA with 1% TBDMSCl

Scheme 16. Proposed thermal degradation mechanisms for butylene dibenzoate during thermal degradation of PBT Reprinted from {832485} with permission from Elsevier

On the other hand, PCL has been suggested to decompose by a two-step mechanism [a.151, a.440, a.441]. The first step is polymer chain cleavage via cis-elimination and the consecutive second step is an unzipping depolymerisation from the hydroxyl end of the polymer chain (Scheme 17(b)). The degradation rate of PHB is faster than those of PCL and PLLA at a given temperature, indicating the lower thermal stability of PHB. The chemical structure of the monomeric unit in PHB is characterised by an activated C–H bond neighbouring a carbonyl group, which can participate in the cis-elimination reaction. Although a PCL unit also possesses an activated C–H bond, it cannot participate in a six-membered cyclic transition state.

92


(a)

(b)

Scheme 17. (a) Thermal degradation of PHB and (b) unzipping depolymerisation from the hydroxyl end of the polymer chain Reprinted from {852663} with permission from Elsevier

For PLLA, thermal degradation proceeds by a complicated mechanism, which does not give simple activation parameters [a.432] {607567}, {756177}. The tertiary C–H bond is acidified by the two neighbouring ester groups. However, only the non-activated C–H bond of the methyl group can form a six-membered cyclic transition state. Thus, the availability of cis-elimination in thermal degradation would determine the thermal stability of the polyesters. Moisture, hydrolysed monomers and oligomers, and residual metals affect the thermal stability of PLLA {589987}. Thus, the thermal degradation mechanism of PLLAs varies according to the structure and molecular weight of each polymer. The complexity of the thermal degradation of PLLA suggests that degradation at temperatures above 200 °C includes intramolecular transesterification leading to lactide and cyclic oligomers, ciselimination leading to acrylic acid oligomers, and fragmentation producing acetaldehyde and CO2 {607567}. The influence of parameters such as polyester molecular weight and the nature of the PCL endgroups have been vigorously pursued. However, there are still some disagreements in the published works. For instance, it has been reported that the shape of the PCL weight-loss curves does not change with variations in heating rate, but the temperature of the maximum mass loss shows a slight increase at higher heating rates, which is an indication of the activated character of the degradation process. Meanwhile, different work on the thermal degradation of PCL at 220 °C and under N2 has proposed that decomposition proceeds via initiation by dissociation into active ionic species followed by depropagation of the active species to form the monomeric -caprolactone via a zipper mechanism [a.152]. In contrast, it has been observed that, during the pyrolysis of PCL in a mass spectrometer at 220 °C and 10–6 mmHg, only insignificant amounts of monomeric -caprolactone are 93

Thermal Degradation of Polymeric Materials produced {883185}. Instead, a series of oligomeric ions was found whose formation was explained by a thermal cleavage of the ester bond and formation of ketene and hydroxyl endgroups and, with a lower intensity, the cleavage of the O–CH2 bond and the formation of carboxyl and pentenyl endgroups. These products can only be detected in high vacuum, which provides for the fast removal of these reactive species, whereas at higher pressure, a back-reaction becomes dominant, so that concurrent degradation to monomer is observed. Elsewhere, the thermal degradation of PCL suggested that the polymer degrades by random chain scission and specific chain-end scission in solution and bulk, respectively {883185}. According to the models based on continuous distribution kinetics, the activation energy of the processes in bulk (determined from the temperature dependence of the rate coefficients) was found to be significantly higher than the degradation in solution. Researches on thermal degradation of purified and well-defined (R)-hydroxyl- -isopropyl ester PCL have proposed a two-stage degradation mechanism – Scheme 18(1) displays the ciselimination mechanism and Scheme 18(2) shows the unzipping process {852663} [a.153].

Scheme 18. Thermal degradation of PCL Reprinted from [a.153] with permission from ACS

94

Polymers, Copolymers and Blends It was established that the first process implied a statistical rupture of the polyester chains via an ester pyrolysis reaction. The produced gases were identified as H2O, CO2 and 5hexenoic acid. The second step led to the formation of -caprolactone (cyclic monomer) as a result of an unzipping depolymerisation process. Figure 12 displays the obtained TG-MS results. The degradation rate drops significantly with PCL chain length, as expected. This behaviour was explained in of statistical chain cleavage triggered by the pyrolysis reaction in the first degradation process. Indeed, the probability of forming fragments of sufficiently low mass to be volatile at 340 °C increased for lower-molecular-weight

Figure 12. Mass spectrum of -caprolactone monomer obtained by TG-MS at 30 °C/min under He Reprinted from [a.153] with permission from ACS

95

Thermal Degradation of Polymeric Materials PCL chains. Acetylation of the hydroxyl endgroups proved to limit the occurrence of degradation by depolymerisation. However, the initial blocking of terminal hydroxyl functions cannot fully prevent the second degradation mechanism from taking place. Certainly, the water molecules generated in the first degradation step can also hydrolyse the polyester chains, yielding carboxylic acid and free hydroxyl endgroups. Finally, it was shown that substitution of air (or O2) for inert gas changes the thermal degradation behaviour considerably, with an activation energy twice as high as the value calculated for oxidative thermolysis. The use of lignocellulosic materials, such as sisal fibres, shows an unfavourable effect on PCL – a decrease in thermal stability is observed [a.154]. In consequence, processing of composites derived from lignocellulosic fibres and PCL has to be carried out at lower temperatures than in the case of pure PCL in order to prevent thermal degradation of PCL. On the other hand, thermal recycling of PCL and sisal fibre mixtures requires lower energy and should be considered as an economic eco-compatible strategy for energy recovery. The main PCL–cellulose interactions have been suggested to take place in the solid state between the char produced during cellulose degradation and PCL, and at the solid–gas interphase between PCL and the gaseous products evolved from cellulose degradation. In the solid state, hydrogen-bonding type interactions occurred between the hydroxyl group of cellulose and PCL carbonyl groups. These interactions delay the microcrystalline cellulose dehydration reaction, which takes place at lower temperatures than depolymerisation. As degradation proceeded, the effect of hydrogen-bonding interactions decreased, but the char and gases evolved from cellulose degradation (mainly CO, CO2 and aldehydes) interacted with solid PCL. On the other hand, the char formed due to the thermal degradation of cellulose acted as a thermal stabiliser of PCL [a.154]. Although poly(-caprolactam) (PCLA) is known to be non-biodegradable, poly(caprolactam-co-caprolactone) P(CLA-co-CL) polymers, whose properties vary from rigid to elastomeric depending on the amount of lactone monomer present in the starting polymerisation composition, have been reported to be biodegradable. Studies on the thermal degradation of P(CLA-co-CL) copolymers show that PCL and PCLA are degraded in wellseparated temperature ranges (PCLA normally degrades below 400 °C while PCL degrades at temperatures above 400 °C). The presence in the thermograms of degradation steps in the close vicinity of their respective degradation temperatures gives credence to the existence of at least enriched PCL and PCLA blocks in the copolymers. Degradations occurring at intermediate temperatures were attributed to CLA/CL hetero-sequences [a.155].

5.6.2.2 Diglycidyl Ether of Bisphenol-A/Poly(-Caprolactone) (DGEBA/PCL) Blends Uncured diglycidyl ether of bisphenol-A (DGEBA)/poly(-caprolactone) (PCL) blends were described as being completely miscible over the entire composition range, both in the melt

96

Polymers, Copolymers and Blends and in the amorphous state [a.156]. In contrast, when cured with 4,4-diaminodiphenyl sulfone, a multiphase structure was formed. Thus, the change in miscibility state allowed the influence of miscibility as well as crosslinking on the thermal stability of this system to be studied. It was distinguished that there are significant differences in the thermal degradation for uncured and cured DGEBA/PCL systems [a.157]. The uncured systems, which were miscible, displayed three different stages of degradation. Two of them were similar to the single degradation stage displayed by each pure component, while the third one arose from the combination of some fragments produced during the degradation of DGEBA and PCL. On the contrary, for the 4,4-diaminodiphenyl sulfone cured systems, only a single degradation stage was reported. Regarding the blends, the results obtained showed that, on curing, the thermal stability of the mixtures increased significantly when comparing blends of identical DGEBA/PCL ratio, no matter what stability parameter has been considered. While pure polymers displayed only a single decomposition step arising from a homogeneous sample, for uncured and cured blends, either more than one stage or complex degradation steps arose from a heterogeneous sample. Finally, it was noteworthy that, although some researchers have proposed the variation of the first stage of the thermal degradation of polymer blends with composition as a criterion to estimate the compatibility of homopolymers, associating deviations from the additivity rule with a certain degree of miscibility, some other experiments reveal that not only do miscible blends show negative deviations from additivity, but also immiscible blends display such behaviour [a.156, a.157].

5.7 Acryl Polymers Thermal degradation of an acrylic polymer, polymerised using a free-radical method, proceeds in three steps of mass loss: the first and easiest (Scheme 19(1)) is initiated by scissions of head-to-head linkages at about 160 °C (representing one type of defect at the polymer backbone); the second (Scheme 19(2)) by scissions at the chain-end initiation from vinylidene ends at around 270 °C; and the last (Scheme 19(3)) by random scission within the polymer chain (at the weakest bonds) {894653} {805908} {784305} {737106} {711105} {630026} {600788} [a.442].

5.7.1 Poly(Methyl Methacrylate) (PMMA)

5.7.1.1 PMMA Homopolymer Since this polymer is widely used, e.g., in orthopaedic surgery, fracture fixation, human body implantations and as a filler in irregularly shaped skeletal defects or voids, a

97


Scheme 19. Three chain scission steps leading to thermal degradation in acrylic polymers

considerable amount of work has been dedicated to the thermal degradation of PMMA {893911} {884331} {857011} {827195} {803123} {753900}. The rate of thermal degradation has been found to depend upon the initial degree of polymerisation of the polymer, the dependence of which has been used to identify the mechanisms of thermal degradation, e.g., a radical one (Scheme 20) [a.8]. It has also been reported [a.159] {863996} that some degradation occurs by side-group elimination, leading to the formation of unsaturated products. It has also been claimed that side-group elimination is a more dominant process than chain scission initiation, 98


Scheme 20. Degradation pathways for PMMA Reprinted from [a.8] with permission from Elsevier

due to the possibility of efficient recombination of the caged radical chain ends [a.18, a.161–a.163]. The wide variation in the thermal degradation of PMMA can be explained in of the structure of the PMMA used and by the experimental conditions employed for preparing the polymer. A two-step degradation process results if the polymer has been prepared in the presence of air due to copolymerisation with oxygen but not to weak links formed by terminal combination since these would be present in all free-radical polymerisations. PMMA polymerised thermally is as stable as polymers initiated by free radicals in the absence of oxygen and peroxide impurities. It has a higher molecular weight and

99

Thermal Degradation of Polymeric Materials consequently a lower concentration of labile endgroups, which may for slight differences observed. Anionically polymerised PMMA has a similar thermal stability to the polymers prepared thermally or with free-radical initiators in the absence of oxygen. Apart from unsaturated endgroups and oxygen copolymerisation, there is no evidence for weak links increasing the rate of degradation of PMMA {863996}. PMMA begins to degrade slowly at 220 °C, and then 40–47% degrades in the temperature range 220–270 °C, but subsequent heating to 305 °C leads to 100% degradation. Three degradation reactions have been reported, and suggested initiation by the scission of weak links and random chain scission [a.160]. The weak links are produced either during polymerisation by head-to-head monomer addition, or by disproportionation reaction during termination, which produces vinylidene chain ends. PMMA homopolymer prepared by free-radical polymerisation is known to begin degrading at approximately 175 °C and depropagates to monomer as a result of thermal degradation at higher temperatures. The degradation originates from the formation of sterically hindered linkages that result from head-to-head coupling during polymerisation. Unsaturated endgroups that are formed by disproportionation during polymerisation begin to degrade at 225 °C, while the other possible saturated endgroups are thermally stable in an N2 atmosphere up to 300 °C [a.160]. After formation of a polymeric radical, depolymerisation occurs for PMMA to form MMA monomer. Random chain scission occurs above 300 °C, leading to depolymerisation and monomer volatilisation {443515}. PMMA degrades predominantly by a depropagation process (the reverse of the polymerisation process), the rate of which was first-order, but this order is only accurately followed with time over an initial period. Thermally degraded polydisperse PMMA of molecular weight between 18,000 and 35,000 that had been polymerised using benzoyl peroxide as the radical initiator degraded in the same way, with a calculated activation energy of 130 kJ/mol at 220 °C [a.158]. A marked increase in activation energy was observed as degradation proceeded and it was also found that the nature of the endgroup had an effect on the rate of degradation, such that introduction of 1,4-diaminoanthraquinone endgroups prevented degradation at 220 °C, indicating that chain-end initiation was an important process. Another extensive study on the effect of molecular weight on the depolymerisation of PMMA prepared by free-radical polymerisation stated that two reactions were present in the temperature range of 250–350 °C, one initiated at unsaturated and the other at saturated chain ends [a.161]. Benzoyl peroxide-initiated PMMA is far less stable than PMMA that had been thermally polymerised in the absence of initiator. Activation energies of 140 and 230 kJ/mol were calculated for the depolymerisation reaction of the radical- and thermally polymerised PMMA, respectively. Thermal degradation of anionically polymerised PMMA (MW of 6,100 and 19,300) that was terminated with saturated endgroups has been reported [a.162]. This work considered 100

Polymers, Copolymers and Blends that these saturated endgroups were inactive and that depolymerisation was initiated only from random chain scission. The activation energy was found to be 260 kJ/mol, and the pre-exponential factor was 2 × 1016 s–1 as also given elsewhere for anionically polymerised PMMA of molecular weight 131,000 and narrow MWD. Moreover, an activation energy of 280 kJ/mol and a pre-exponential factor of 1.29 × 1019 s–1 for the depolymerisation step were reported [a.163], while a different collaborative team acquired results by varying the molecular weight of anionically polymerised PMMA from 12,700 to 1,520,000 [a.164]. Thermal degradation of a commercial free-radical-polymerised PMMA with molecular weight of 40,200 has been described too [a.165]. The researchers found that, during the thermal degradation, the first phase began below 200 °C and was thought to include the volatilisation of additives and residual monomer, while the second phase, with an activation energy of 220 kJ/mol and a pre-exponential factor of 2.9 × 1016 s–1, started at 290 °C. In a similar study on the degradation of PMMA prepared by free-radical initiators, benzoyl peroxide and AIBN, in the presence and absence of a transfer agent, benzene, it was found that the nature of the free-radical initiator had a small effect on the degradation mechanism, a fact attributed to initiation from scission of weak links and random chain scission {894653}. The weak links were produced either by terminal combination during polymerisation, i.e., by head-to-head monomer addition, or by disproportionation termination producing vinylidene chain ends. Still on free-radical-polymerised PMMA (molecular weight 996 000), thermal degradation was found to occur stepwise, beginning at 150 °C, before slowing down momentarily at 300 °C (40 wt% loss) and continuing in a second degradation step. This is consistent with the observations for benzoyl peroxideinitiated PMMA where the activation energies of the first and second steps were 60 and 190 kJ/mol, respectively [a.158, a.161]. However, this second step was not reported in a similar work where only a one-step process commencing at 290 °C, with activation energy of 210 kJ/mol and pre-exponential factor of 4.15 × 1014 s–1, was reported [a.166]. For fractionated PMMA polymerised with AIBN as initiator and laurylmercaptan as transfer agent, the rate constants obtained indicated a change in thermal degradation mechanism at approximately 400 °C, where depropagation to the polymer chain end began to compete with the first-order termination process [a.167]. This led to a kink in the Arrhenius plot of the rate constants, where below 400 °C the activation energy was approximately 130 kJ/mol and above 400 °C it was ca. 250 kJ/mol. Thermal degradation of PMMA polymerised using a free-radical method proceeded in three steps of mass loss. It is interesting to note that the first two mass-loss steps were not observed with ionically polymerised samples, which indicate that the first two steps are caused by the defects in the polymer. Although the existence of head-to-head linkages could not be demonstrated, 1H NMR detected the vinylidene ends in the polymer. No significant differences were seen in the thermal or oxidative degradation of the PMMA when it was polymerised with the free-radical method using two common initiators [a.168].

101

Thermal Degradation of Polymeric Materials It has recently been demonstrated that PMMAs initiated using lactams and thiols are more thermally stable and have higher activation energy for degradation than AIBN-initiated polymer {883154}. In such a system, thiols were used as initiators, which was different from the traditional chain-transfer agents. Polymers prepared using thiols were found to have a high degradation temperature due to the elimination of an unsaturated endgroup. Depolymerisation was reported as the most important reaction in the thermal degradation of the polymer, and only the main-chain scission of PMMA was initiated by thiols. The activation of the degradation process of PMMA initiated by thiols occurred at a higher energy level than that of an AIBN-initiated polymer; the differences are ca. 90–100 kJ/mol for activation energy and five orders of magnitude for frequency factors.

All in all, it can be concluded that there is dependence of the rate of thermal degradation of PMMA on molecular weight at low degradation temperatures and that thermal degradation is initiated by a mixture of chain-end and random chain scission initiation, followed by depropagation and first-order termination. Random scission is attributed to pre-oxidation of the polymer on storage at room temperature [a.164]. At higher temperatures, a change in molecular-weight dependence is observed, related to depropagation at the end of the polymer chain. The thermal degradation of PMMA also leads to the formation of char, which is produced by the elimination of methoxycarbonyl side chains. The amount of char produced increases with increasing concentration of endgroups and temperature. The degradation of isotactic and syndiotactic PMMA has been studied by TG with an emphasis on their behaviour in ultra-thin films on silica. Both PMMA tacticity and adsorbed amounts were found to affect the degradation. In bulk, syndiotactic PMMA (syn-PMMA) had higher thermal stability than isotactic PMMA (iso-PMMA) due to its lower chain mobility. The Tmax (maximum rate decomposition temperature) was reported to be lower than that of bulk samples at higher adsorbed amounts for both iso- and synPMMA. The Tmax value increased when the adsorbed amount on the silica surface decreased for syn-PMMA, with the degradation behaviour of adsorbed iso-PMMA becoming very complex at low adsorbed amounts [a.169]. Chiantore and Guaita [a.170] investigated the effects of tacticity on the degradation of PMMA and indicated that both isotactic PMMA (iso-PMMA) and syndiotactic PMMA (syn-PMMA) have similar decomposition pathways and activation energies. The isoPMMA decomposed at a lower temperature with a broader range compared to the synPMMA with the same chain ends and similar molecular weight. Also, iso-PMMA was more sensitive to electron-beam irradiation than syn-PMMA; thus the former degraded more easily than the latter [a.171]. In addition, these reports comply with a very recent one that observed a higher decomposition temperature in PMMA/SiO2 nanocomposites [a.172]. This was attributed to the barrier function of silica particles, which prevent the release of evolved degradation products that recombine to form thermally stable residue/char.

102


5.7.1.2 PMMA Blends Interactions between the degradation products of one polymer and others also occur as a result of the diffusion of small mobile molecules or radicals through the interfacial layer {845530}. In that case, the degradation products of one polymer may stabilise/ destabilise the other polymers in the blend [a.443]. Reactions between macromolecules or macroradicals with small molecules and small radicals diffusing across the phase boundaries are characteristic of the thermal degradation of polymer blends in an inert atmosphere {886353}. The thermal behaviour of polymer blends shows some similarities with that of graft copolymers. Depending on the reactivity of macromolecules and lowmolecular-weight fragments, the resistance of component polymers is either increased or reduced in comparison with neat, unmixed polymers. Direct interactions between the two component polymers may not be observed in high-temperature degradation [a.120]. McNeill and co-workers [a.15, a.16] found that the nature of the interaction between different polymers depends strongly on the physical state of the system (the nature of the polymer, the miscibility of the polymer composition, or the degree of phase dispersion). In a heterogeneous blend, interactions occur in the bulk of one or both domains and in the phase boundaries. In homogeneous samples, the degradation products of one polymer are directly in with the other polymers so that their combined effect on the thermal degradation is greater. Other works have concluded that the thermal behaviour of polymer blends is related to the miscibility of the respective components and to their interactions, where immiscible blends show better stability than miscible ones [a.173]. Blending has been reported to have a great influence on the thermal stability of polymers, as the thermal stability of blends depends strongly on the interaction between individual polymers. Block and graft copolymers have applications including surfactants, compatibilisation agents in polymer blends, adhesives, additives in high-impact materials and thermoplastic elastomers. PMMA is an important thermoplastic material in this respect {884544}. For instance, the incorporation of siloxane into PMMA has a large number of potentially interesting applications, including surface-modified materials and gas separation devices. Studies have demonstrated the feasibility of the synthetic route for and interesting thermal stabilities of poly(dimethylsiloxane)-PMMA block copolymers (PDMS-b-PMMA) [a.174]. The incorporation of PDMS segments improves the thermal stability of PMMA. Silicon functionalisation affects the initial thermal degradation of MMA segments in PDMS-b-PMMA copolymer with higher activation energy than that in PMMA. The thermal degradation of cellulose acetate hydrogen phthalate (CAP) and its blends with PMMA has been investigated by thermogravimetry [a.175]. The TG/DTG curves showed two decomposition stages for pure CAP. The decomposition behaviour was changed with the addition of PMMA. For 90/10 and 70/30 CAP/PMMA blends, there were three

103

Thermal Degradation of Polymeric Materials decomposition stages. With further addition of PMMA, the decomposition behaviour shifts towards that of pure PMMA. For 50/50, 30/70 and 10/90 CAP/PMMA blends and pure PMMA, only one decomposition stage is reported. The temperature at which 50% decomposition took place, T50%, was found to increase with increased PMMA content. The thermal degradation of poly(methylphenylsiloxane)-PMMA graft copolymer (PMPSg-PMMA) was studied by using TGA, and compared with that of unmodified PMMA [a.176]. The deconvoluted DTG curve of PMPS-g-PMMA showed four peaks, while that of PMMA showed three peaks. The least stable step (7.0 wt%) of PMMA was attributed to scissions of head-to-head linkages, the second step (17 wt%) to scissions at chain-end initiation from vinylidene ends, and the most stable step (76 wt%) to random scission within the polymer chain. The temperature of maximum weight loss increased with increasing PMPS content, whereas the rate decreased. This was attributed to the fact that the flexible siloxane oligomer has better heat dissipation, thus reducing the internal temperature of the copolymer [a.174]. The apparent activation energy of thermal degradation was evaluated – the value of the activation energy for MMA segments in PMPS-g-PMMA was larger than that for pristine PMMA. The PMPS char residue of 35 wt% at 700 °C (heating rate of 10 °C/min) was reported [a.176].

5.7.2 Acryl (Co)Polymers The degradation of lower polyacrylates indicated a close qualitative similarity in the thermal degradation behaviour of poly(ethyl acrylate), poly(n-propyl acrylate) and poly(n-butyl acrylate) [a.177, a.178]. Their major degradation products are saturated and unsaturated dimers, trimers and monomers. The general mechanisms of the thermal degradation of these polymers were random main-chain scission, depolymerisation, carbonisation and side-group reactions. At faster heating scans, the first sub-step was not noticeable in TG experiments because of its strong overlap with the second step, which corresponds to degradation initiated at the vinylidene endgroups [a.159]. Nevertheless, these two substeps were clearly seen in DTG analysis even if performed at faster heating rates. In the degradation of polyacrylates {766562} {686483}[a.442] and poly(methyl acrylate)s, the mechanisms are random main-chain scission, depropagation, intermolecular transfer and acyl-oxygen scission [a.158]. Steric effects, such as large pendant groups in the higher polyacrylates, affect the intermolecular reaction so that the major degradation products are different in the case of lower and higher polyacrylates. Scheme 21 [a.137] shows the degradation mechanism of higher alkyl polyacrylates, while Figure 13 [a.158] displays the chromatogram of mixed higher alkyl alcohol obtained during the thermal degradation of higher polyacrylates.

104


Scheme 21. Degradation mechanism of higher alkyl polyacrylate Reprinted from [a.158] with permission from Elsevier

Figure 13. Chromatogram of mixed higher alkyl alcohols obtained during thermal degradation of a higher polyacrylate (time in minutes) Reprinted from [a.158] with permission from Elsevier

105

Thermal Degradation of Polymeric Materials Monomers, dimers and higher-molecular-weight oligomers are the predominant decomposition products of the lower alkyl polymers. The major degradation products of higher alkyl polyacrylates are olefins, alcohols and acrylate monomers (see Table 5), which are different from those of the lower alkyl polyacrylates. The thermal degradation of two poly(p-substituted phenacyl methacrylate)s – namely, poly(p-bromophenacyl methacrylate) (PBPMA) and poly(p-methoxyphenacyl methacrylate) (PMPMA) – begins at about 250 °C, leading to the formation of anhydride ring structures in the chain [a.179]. Poly(t-butyl methacrylate) containing a branched side chain mainly undergoes ester decomposition, giving isobutylene and poly(methacrylic acid) residue [a.180]. Poly(2-sulfoethyl methacrylate) containing a side chain having a bond of relatively low dissociation energy (S–C bond) undergoes ester decomposition and sidechain scission, giving small products such as CO2, CO, SO2, ethylene and residue [a.181]. In the case of poly(n-butyl methacrylate) [a.182] and poly((2-phenyl-1,3-dioxolane-4yl)methyl methacrylate) {687301}, ester decomposition and depolymerisation proceed simultaneously, while depolymerisation dominates up to 350 °C in the degradation of poly(ethyl methacrylate), with ester decomposition becoming important at higher temperatures [a.14]. Investigation of the nature of the evolved products during the thermal degradation of poly(phenacyl methacrylate) (PPAMA) has been supplemented by studies of structural changes in the degrading polymer by FTIR [a.183]. Depolymerisation was identified as the main reaction in the thermal degradation of the polymer. The degradation produced anhydride ring structures in the chain at temperatures above 300 °C, and the total

Table 5. Degradation products and yield of higher and lower polyacrylates Reprinted from [a.158] with permission from Elsevier Product

Degradation Relative abundance products of higher of polyacrylate (%) alkyl polyacrylate at 550 °C Olefin n-octadecene 54.94 n-eicosene n-docosene Alcohol n-octadecyl alcohol 26.09 n-eicosyl alcohol n-docosyl alcohol Acrylate n-octadecyl acrylate 18.96 monomer n-eicosyl acrylate n-docosyl acrylate Dimer

106

0.00

Degradation Relative products of lower abundance of alkyl polyacrylate polyacrylate (%) propane 4.25

n-propanol

4.09

n-propyl acrylate

17.08

70.58

Polymers, Copolymers and Blends degradation to 430 °C produced many volatile products, such as monomer, acetophenone and benzaldehyde. The activation energy for the thermal degradation of PPAMA was reported as 89 kJ/mol in the first stage observed at 205–300 °C. The second stage of decomposition occurred at 300–430 °C and the energy of activation was 67 kJ/mol. The thermal degradation of poly(3-(1-cyclohexyl)azetidinyl methacrylate) (PCyAMA) began at low temperature (about 180 °C) by decomposition of the azetidinyl ring, producing some amine-based products {783965}. The degradation produced anhydride ring structures in the chain above about 300 °C as a result of a reaction between two neighbouring units. The monomer (m/z = 223) was detected as only 0.7% in the cold ring fraction (CRF) collected from the thermal degradation of PCyAMA to 345 °C. At the next step, side-chain decomposition reactions caused considerable crosslinking, as evidenced by the high amount of residue (about 15%) at 500 °C. Thermal degradation of poly(2-sulfoethyl methacrylate) also began at about 180 °C with side-chain decomposition – the monomer was not generated during the decomposition, and 16% residue was left at 600 °C [a.181]. Poly(2-hydroxyethyl methacrylate) (PHEMA) has been studied extensively due to its wide range of applications, e.g., for biomedical applications. The thermal degradation of poly(n-hydroxyalkyl methacrylate)s typically produces monomer as a result of depolymerisation, and/or cyclic anhydride-type structures are formed by intramolecular cyclisation. Studies on the thermal degradation behaviour of PHEMA and its deuterium derivative reported that the CRF was trapped at two ranges – from ambient temperature to 340 °C (first decomposition step) and from 340 to 400 °C (second decomposition step) {820840}. At both CRFs, the major product was monomer due to depolymerisation reaction. The side products arising from ester decomposition were a six-membered glutaric anhydride type of ring, an oxolane, 2-isopropenyl ethyl methacrylate, methacrylic acid and CO2. The activation energy for the thermal degradation of PHEMA was calculated as ca. 130 kJ/mol. The thermal degradation of poly(2-(3-(6-tetralino)-3-methylcyclobutyl)-2-ketoethyl methacrylate) (PTKEMA) (Scheme 22) was reported to produce anhydride ring structures in the chain at temperatures up to about 300 °C, with depolymerisation being the main reaction {894646}. The cleavage of ketone, aldehyde and tetralin compounds from the side chains of polymers is a common reaction for the polymers. The activation energy for the thermal degradation of PTKEMA was 210 kJ/mol in the first stage at 220–330 °C, while the second stage decomposition occurred at 330–430 °C with an energy of activation equal to 125 kJ/mol. The thermal degradation studies of poly(2-methacrylamidopyridine) (PMAPy) synthesised and polymerised via free-radical polymerisation at temperatures of 300–500 °C revealed that most of the decomposition products were volatile [a.184]. In particular, the CRF1 107


Scheme 22. Thermal decomposition path giving the main products of the degradation for poly(2-(3-(6-tetralino)-3-methylcyclobutyl)-2-ketoethyl methacrylate) Reprinted from {894646} with permission from Elsevier

108

Polymers, Copolymers and Blends showed that 2-aminopyridine was a major product, while the CRF2 showed that 2methacrylamidopyridine was also formed as a main product. The activation energy of the thermal degradation of PMAPy was calculated as 110 kJ/mol and the pre-exponential factor was reported as 5.0 × 1011 s–1. The TG profiles of vinyl triethoxy silane-methyl methacrylate (VTES-MMA) copolymers in N2 have been reported to be similar to that of PMMA [a.185]. However, the initial decomposition temperature (IDT) of the copolymers decreased with increasing VTES content in the copolymer. This behaviour was attributed to the decrease in the molecular weight of these copolymers during degradation. A similar tendency for TG was also reported for styrene-siloxane block copolymers synthesised with a living anionic initiator [a.186]. The effect of the composition and form of the polymer sample on the non-isothermal weight loss of sub-micrometric multilayer polymer particles has been examined [a.178]. The polymer beads consisted of a PMMA core, poly(butyl acrylate) copolymer (BAC) network shell, and a further layer of PMMA copolymer with butyl acrylate (P(MMA-coBAC)). The DTG curves showed two decomposition stages for multilayer polymer powders and for particle/polymer mixtures prepared by mixing emulsions before the freeze–thaw isolation of polymer. The first degradation step, due to PMMA depolymerisation, started at a temperature 45 °C lower compared with the degradation of PMMA alone. The polymer samples, which were processed by melt stirring, press moulded or isolated from emulsions as transparent films before the TG experiment, were found to be more stable than pure polymer powders. The weight loss of bulk samples proceeded smoothly in one stage like that of a single-type polymer; nevertheless, the PMMA was found to be more sensitive to thermal degradation than the BAC network. The researchers proposed that the stabilisation of PMMA in a blend with BAC in a bulk polymer sample is possibly due to the BAC compressed network, which prevented PMMA depolymerisation. The thermal degradations of thiophene-capped poly(methyl methacrylate) (TPMMA) and poly(methylthienyl methacrylate) (PMTM) were studied via direct Py-MS {891577}. No significant effects of the heating rate on the thermal degradation behaviour of the polymers under investigation were observed in the heating range studied. It was determined that the thermal degradation of TPMMA occurs via a depolymerisation mechanism, mainly yielding monomer as in the case of pristine PMMA, as the only main difference is the presence of endgroups. An analogous degradation mechanism was also proposed for PMTM, which thermally decomposes in a single stage with a major weight loss at 410 °C – the elimination of side chains and evolution of OCH2C4H3S groups were found to be effective. Vinyl-terminated PMMA (PMMA–CH=CH2) thermally degrades at a lower temperature (230–300 °C) than saturated PMMA (PMMA–H), which degrades between 300 and 400 °C [a.160]. The major thermal degradation mechanism of PMMA–CH=CH2 involves 109

Thermal Degradation of Polymeric Materials efficient radical transfer to the end of the vinyl chain [a.162] {443515}. Any radical, independent of the reaction in which it is produced, degrades a large number of polymer chains by the chain-transfer process. The radical that transfers the active site to the next chain is the species present at the initiation end of the PMMA–CH=CH2 chain. The polymer has a high thermal resistance if degradation does not occur via chain-end unzipping. Thus, introducing a saturated endgroup with high bond energy at the ends of the polymer may reduce the end initiation effects. Consequently, a random scission step occurs mainly during thermal degradation, and the thermal degradation temperature gets enhanced.

5.7.3 Acrylonitrile-Containing (Co)Polymers The mechanism of this process for the thermal decomposition of non-oxidised polyacrylonitrile (PAN) is similar to the oxidative atmosphere mechanistic pathway of reactions leading to hetero-aromatic clusters ultimately growing into a graphite layer [a.39, a.137] {699986} {594557} {584927}. The weight loss during the thermal degradation of PAN results mainly from the evolution of oligomers due to radical chain scissions, although some quantities of low-molecular-weight species such as hydrogen cyanide and ammonia are also formed. FTIR analysis of the gases evolved during PAN degradation (displayed in Figure 14) revealed that ammonia and hydrogen cyanide are the most abundant lowmolecular-weight species evolving in the range of temperature corresponding to the first peak of the DTG curve [a.89, a.187]. The second peak was mostly accompanied by the release of hydrogen cyanide – when the temperature exceeded 480 °C, methane dominated in the gaseous products. Polymethacrylonitrile (PMAN) prepared by free-radical polymerisation with the initiator 4,4-azo-bis(4-cyanovaleric acid) (ACVA) was reported to have chain ends from the initiator fragments which incorporate carboxyl groups {686094}. In contrast with PMAN prepared with AIBN as initiator, which degraded quantitatively to monomer, the ACVAinitiated polymers gave much reduced monomer yields, an important tar/wax fraction and a substantial amount of residue, amounting to 32–48% of the sample weight, dependent on the initial molecular weight of the polymer. The thermal degradation mechanisms of poly(styrene-co-methacrylonitrile) (P(S-co-MAN)) is reported in of the competition between the depolymerisation and backbiting reaction on the basis of the bond dissociation energies of the C–C and C–H bonds in the polymer chains [a.188]. The activation energy of pyrolysis obtained by Ozawa’s plot increased with the content of methacrylonitrile units in the copolymer chain, although the onset temperatures of loss of sample mass in the TG curves shifted to the lower temperature region. Yields of each monomer, dimer and trimer, and also those of hybrid dimers and

110


Figure 14. FTIR spectra of the pyrolysis residue of PAN as a function of temperature. The lowest spectrum was obtained at room temperature, and the temperature for each of the other spectra is indicated at the left of each spectrum. Wave numbers are given in cm-1 Reprinted from [a.89] with permission from Elsevier

trimers, changed remarkably depending on the copolymer composition and the pyrolysis temperature. These phenomena have been explained in of the change of the ratio of depolymerisation to backbiting reactions depending on the chemical structure and degradation temperature. Compared with the case of the corresponding homopolymers under the same pyrolysis conditions, the yield of styrene monomer from the copolymer increased, while the yield of MAN decreased. This tendency was remarkable in the lower-temperature region of 360–485 °C. In contrast, the yields of styrene dimers and trimers from copolymers decreased with increasing content of MAN. An abnormal phenomenon was observed in the case of the styrene-MAN copolymer containing 55 mol% of MAN, whose sequence distribution is highly alternating – not only the yields of MAN and styrene monomer, but

111

Thermal Degradation of Polymeric Materials also those of styrene dimers and trimers were much lower than those of the homopolymer and other random copolymers with different sequence distributions. However, the amounts of dimers and trimers increased remarkably – a fact attributed to the competition between depolymerisation and backbiting reaction depending on the kind of penultimate unit. In the case of highly alternating copolymers, the probability of backbiting increased due to the unique sequence distribution, and hence generation of hybrid dimers and trimers increased considerably [a.188]. Elsewhere, thermal degradation mechanisms in flash pyrolysis of P(S-co-MAN) and the corresponding homopolymers (PS and PMAN) have been reported to be affected by pyrolysis temperature and by the chemical structure of the polymers, for example, the composition and sequence distribution of copolymer [a.189]. Yields of each monomer, dimer and trimer, and also those of hybrid dimers and trimers, changed remarkably depending on the copolymer composition and the pyrolysis temperature. The authors interpreted these results from the standpoint of the competition between depolymerisation and backbiting reaction. Under flash pyrolysis conditions, depolymerisation is closely related to the C–C bond dissociation energies in the main chain and backbiting is related to the C–H bond dissociation energies for the abstraction of a hydrogen atom by the terminal radical. Thermal degradation studies of acrylonitrile-cellulose graft copolymers reported that they degrade to produce oligomers representative of the copolymer chain [a.190]. The degradation of graft copolymers followed a two-step pyrolysis corresponding to the consecutive decomposition steps. Pyrolysis temperatures differentiated the coexisting graft copolymer components. At the beginning of pyrolysis, the graft polyacrylonitrile side chains were removed in the first step at 240 °C. In the second step, the backbone cellulose chain decomposed at 290 °C, prior to the decomposition of the graft side-chain polyacrylonitriles at 315 °C. The pyrolysis products had mass spectra characteristic of the copolymer composition in that they contained the repeat units of the oligomers.

5.8 Others

5.8.1 Poly(Vinyl Acetate) (PVAc) The elimination process that occurs during the thermal degradation of poly(vinyl acetate) (PVAc) has been studied and it has been found that elimination of acetate groups initially begins slowly, but increases as degradation proceeds due to an additional process. The increase in rate was found to depend on the concentration of unsaturated groups in the polymer chain. The activation energy of the initial step was found to be 190 kJ/mol, while that for the additional process was 130 kJ/mol. The additional process of elimination

112

Polymers, Copolymers and Blends was considered to be due to a four-membered transition state, activated by double bonds adjacent to the acetate unit {851392}. Scheme 23 shows the free-radical mechanism for PVAc thermal degradation [a.191]. Poly(isopropenyl acetate) (PIPA) is closely related to PVAc, the difference being that PVAc has the major part of its acetate groups attached to tertiary carbon atoms and only a small fraction of them attached to quaternary carbon atoms, whereas the acetate groups in PIPA are exclusively attached to quaternary carbon atoms. Unlike PVAc, PIPA degrades in two stages almost without residue. The first stage of degradation occurs between 150 and 250 °C and consists of acetic acid loss, which corresponds to the maximum theoretical yield. At higher temperatures, the remaining methyl-substituted unsaturated backbone gradually breaks down, forming a mixture of aromatic hydrocarbons that make up a slightly volatile higher fraction and relatively non-volatile cold ring fraction {687290}. Similarly, previous instances of acetate radical migration and hydrogen abstraction in the thermal degradation of PVAc polyblends have been found, for example, for PVAc/PMMA, where monomer production in the PMMA is induced at the deacetylation temperatures, which are much lower than those normally required to secure depolymerisation.

Scheme 23. Free-radical mechanisms for thermal degradation in PVAc Reprinted from [a.191] with permission from Elsevier

113

Thermal Degradation of Polymeric Materials Another example is PVAc/PS, where styrene segments undergo increased chain scission at the deacetylation temperature of the VAc sequences. These effects cannot be explained by the attack of CH3COOH, which in fact tends to stabilise styrene segments. On this , it is therefore only the radical mechanism of PVAc polyblend degradation that provides a ready explanation. For PVAc/poly(dimethylsiloxane) (PDMS) blends containing 50% or more PDMS concentrations, higher (than for pure PDMS) thermal stability was explained on the basis of crosslinking induced by the presence of PVAc [a.191]. Free radicals (e.g., acetate radicals) diffuse from the PVAc phase and abstract hydrogen atoms from methyl groups in PDMS – pairs of such radicals then crosslink (Scheme 24).

Scheme 24. Crosslinking in PDMS induced by acetate radicals during thermal degradation of PVAc/polysiloxane blends Reprinted from [a.191] with permission from Elsevier

114


5.8.2 Poly(Vinyl Alcohol) (PVOH) The main thermal degradation pyrolysis product of poly(vinyl alcohol) (PVOH) below 300 °C is known to be water, produced by the elimination of hydroxyl side groups {825040}. Further studies at 240 °C show that, in addition to water, acetaldehyde, some unsaturated aldehydes and ketones, benzene and benzene derivatives were evolved during thermal degradation {787566} {651339}. It was also shown that the double bonds produced by elimination do not lead to the formation of appreciable amounts of conjugated structures. Also a molecular-weight study by viscometry showed that the thermal degradation of PVOH leads to an initial increase in the molecular weight, followed by a decrease. These observations were explained as being due to the combined effects of crosslinking and chain scission. Recent studies concluded that in the molten state decomposition of PVOH consisted of water elimination and chain scission, via a six-membered transition state, leading to the formation of volatile products including saturated and unsaturated aldehydes and ketones. In the solid state, thermal degradation of PVOH was exclusively by elimination of water [a.192]. Owing to the lack of conjugated structures, it was considered that the elimination in both solid and molten states was random, while the C=C bond formation did not activate elimination of adjacent hydroxyl units, unlike in other cases, e.g., PVAc {851392}. Elimination of water was observed to decrease the amount of hydrogen bonds in PVOH, and hence reduced the Tm. Also, above 30 mol% of elimination, double-bondcontaining structures disrupt the crystalline regions of the polymer.

5.8.3 Vinylidene Chloride (VDC) Copolymers Vinylidene chloride (VDC) copolymers have several outstanding properties, such as high crystallinity, resistance to non-basic solvents and, most importantly, extremely low permeability to a wide variety of gases; therefore, VDC copolymers have wide applications, e.g., in the barrier plastic packaging industry {502579}. However, a serious deficiency of these materials is thermal instability at melt-processing temperatures, which leads to the formation of poly(chloroacetylene) sequences and gives rise to objectionable colour. At processing temperatures, these materials tend to undergo degradative dehydrochlorination by a radical chain process (Scheme 25) [a.193]. Thermally induced carbon–chlorine bond homolysis gives rise to a carbon–chlorine radical pair. The chlorine atom so produced most typically abstracts an adjacent hydrogen atom to form hydrogen chloride and generate an allylic dichloromethylene unit in the polymer main chain, which may act as an initiation site for further rapid sequential dehydrochlorination {502579} [a.193].

115


Scheme 25. Mode of degradation of vinylidene chloride polymers Reprinted from [a.193] with permission from Elsevier

The initiation of degradation is largely unaffected by increasing butyl acrylate content of the VDC copolymers and produces a carbon radical–chlorine atom pair (Scheme 26). Intervention of the side chain to deliver a hydrogen atom to trap the chlorine atom before it is able to abstract an adjacent hydrogen atom from the main chain would effectively interrupt propagation of the degradative dehydrochlorination reaction and lend stability to the polymer [a.193]. However, increasing amounts of butyl acrylate in the copolymers suppresses the rate of propagation of the degradation process. This is most likely due to chain stopping of the degradative dehydrochlorination process as the length of the vinylidene chloride sequences in the copolymer decreases as a consequence of higher levels of acrylate in the polymer.

5.8.4 Sulfone-Containing Polymers The extremely high sensitivity of polysulfones to radiation-induced main-chain scission has found application in the field of microelectronics {829337} {769786} {705682}. The reaction of sulfur dioxide with an olefin or double-bond-containing species leads to a poly(olefin sulfone),

116


Scheme 26. Possible fates of radicals generated by butyl ester side-chain scavenging of chlorine atoms formed during the degradative dehydrochlorination of vinylidene chloride/butyl acrylate copolymers Reprinted from [a.193] with permission from Elsevier

whose lack of thermal stability {872441} has been used in positive resists for electron-beam lithography. Hence, it is important to study the thermal degradation of poly(olefin sulfone)s at high temperatures, which is useful for selecting a ‘prebake’ temperature – the temperature that the poly(olefin sulfone) requires, after spin coating on a suitable substrate, to remove excess solvent from the solid film, relieve strain in the film, and promote better adhesion to the substrate. There are scarce reports on the thermal degradation of poly(olefin sulfone)s. Early studies found that the decomposition rates of poly(olefin sulfone)s increase with the number of hydrogen atoms attached to the -carbon atom, implying -elimination as the initial step {870313}. A similar mechanism was also invoked by Bowmer and O’Donnell in their study of the effect of the olefin structure on the thermal degradation of poly(olefin

117

Thermal Degradation of Polymeric Materials sulfone)s [a.194]. Moreover, the thermal degradation rates for poly(1-butene sulfone) and poly(2-methyl-1-pentene sulfone) in the temperature range of 110–130 °C were established. Other work on the thermal degradation of poly(olefin sulfone)s over the temperature range 180–260 °C by TG suggested a -hydrogen elimination mechanism [a.195]. Most of the volatile products were the corresponding olefin and SO2. The activation energies of the early stages of thermal degradation of poly(1-butene sulfone) and poly(2-methyl-1-pentene sulfone) in the 0–0.05% conversion range have been calculated as 270 and 179 kJ/mol, respectively {870313}. Poly(1-butene sulfone) has been reported to be thermally unstable above 130 °C, liberating gaseous products that contained about 50% of sulfur dioxide. Above 200 °C, the kinetics was first order with respect to sample weight, with an activation energy of 201 kJ/mol. Studies into the thermal degradation of poly(olefin sulfone) in the temperature range of 110–170 °C found that the initial thermal decomposition occurred in weak bonds of the polymer [a.195]. The dissociation energy of the C–S bond, 272 kJ/mol, was lower than that of the C–C bond, 347 kJ/mol. This indicated that the mechanism of degradation of poly(1-butene sulfone) and poly(2-methyl-1-pentene sulfone) in the study followed random main-chain scission, which evolved SO2 along the poly(olefin sulfone) backbone. Other studies on the thermal degradation of poly(styrene sulfone)s (PSSs) suggested that the activation energy decreased with increase in the content of sulfur dioxide groups in the poly(styrene sulfone) {868537} {860011} [a.196, a.197]. The SMS triad monomer sequence (S = SO2, M = olefin) was reported as the most sensitive microstructure in the thermal degradation of poly(styrene sulfone) (Scheme 27) {852665}. From a comparison of the activation energies in the early stage, the heat stability of the poly(olefin sulfone)s

Scheme 27. The mechanism of poly(olefin sulfone) thermal degradation Reprinted from {852665} with permission from Elsevier

118

Polymers, Copolymers and Blends obtained by free-radical copolymerisation of different olefins, vinyl chloride (PVCS), acrylamide (PAAS) and styrene, with sulfur dioxide (PSS), decreases in the order PSS > PVCS > PAAS, which was attributed to the electronegative effect of the substituent in the side chain of the olefin {852665}. In the later stage, the heat stability increases in the order of PSS < PVCS < PAAS, thus reflecting the effect of olefin structure on the thermal stability. As the results showed poly(styrene sulfone) to be of a higher quality for prebaking, the researchers concluded that poly(olefin sulfone)s synthesised using styrene are better for processing in microelectronics than those using vinyl chloride or acrylamide. The FTIR spectra of the studied poly(styrene sulfone)s are presented in Figure 15. High-resolution thermogravimetry has been successfully utilised to investigate the thermal degradation of the bisphenol-A poly(sulfone) [a.198]. The degradation parameters of the bisphenol-A poly(sulfone), involving temperature, maximum decomposition rate, char yield at 800 °C, activation energy, reaction order and frequency factor, exhibit a

Figure 15. The infrared spectra for various polysulfones. The characteristic O=S=O asymmetric stretch at 1300 cm–1, symmetric stretch at 1135 cm–1 and the C–S–C asymmetric stretch at 770 cm–1 are visible. PAaS, poly(acrylamide sulfone); PStS, poly(styrene sulfone); PVCS, poly(vinyl chloride sulfone) Reprinted from {852665} with permission from Elsevier

119

Thermal Degradation of Polymeric Materials dependence on surrounding atmosphere and heating rate. The temperature and activation energy of the first-step degradation of the bisphenol-A polysulfone were larger in an inert atmosphere than in an oxidative atmosphere, indicating more thermally stable and slower degradation in an inert atmosphere. The kinetic parameters calculated on the basis of the high-resolution TG results were in good agreement with those obtained from the traditional isothermal and non-isothermal TG. The degradation mechanism of poly(ether sulfone) has been widely studied {776366} and shows sulfur dioxide to be the first measurable decomposition product as the temperature of the polymer increases beyond 400 °C, degrading mainly between 450 and 650 °C. Based on the initial products of degradation and the bond dissociation energy values, degradation was proposed to start by chain scission at the carbon–sulfur bond between the aromatic ring and the sulfone group, since it is the weakest link in the polysulfone repeat unit [a.199]. Also, sulfone-containing units were found to be more thermally stable than those containing ether units. Since the poly(ether sulfone) in this investigation contained both units in the main chain, its degradation behaviour was in between the two. Thus, other than chain scission, crosslinking is an important factor in polysulfone degradation. The same team also reported on the thermal degradation of poly(methylene sulfone)s at 275 °C under reduced pressure and proposed a mechanism in which the rate-determining step consists of a concerted attack of the sulfone group on the -hydrogen atom with elimination resulting in -olefins and a sulfonic acid. The copolymerisation of acrylamide (AA) with sulfur dioxide to form a variable-composition poly(acrylamide sulfone) with average AA:sulfone molar ratio n 0 has been examined too [a.180]. The thermal degradation of poly(acrylamide sulfone)s was carried out over the temperature range of 325–625 °C, with three heating rates of 5, 10 and 20 °C/min in an N2 atmosphere. The results indicated that the entire degradation of poly(acrylamide sulfone)s under the experimental conditions of this investigation consists of two distinctive stages. Based on TG data, it was found that the activation energy in the first stage of thermal degradation decreased with increasing amount of sulfur dioxide in poly(acrylamide sulfone) and differed slightly in the second stage of thermal degradation. A relation between residue weight and temperature was proposed to describe the TG/DTG profiles of the thermal degradation of poly(acrylamide sulfone)s. According to the reported results, the data and the relation appeared to be in excellent agreement, indicating the very good utility of the relation in analysis data for the two-stage thermal degradation of poly(acrylamide sulfone)s.

5.8.5 Sulfide-Containing (Co)Polymers The growing interest in thermal degradation studies of sulfide-containing polymers is due to their extensive applications in adhesives, sealants, insulators, etc., in addition to interest in understanding the primary and secondary thermal degradation mechanisms

120

Polymers, Copolymers and Blends {848090} {845557} {774199} {612251} [a.79, a.200]. Random block sulfide copolymers such as poly(ethylene sulfidex-co-styrene sulfidey) (x = y = 0.5; x = 0.74, y = 0.26) and their corresponding homopolymers – poly(ethylene sulfide) (PES) and poly(styrene sulfide) (PSS) – have been investigated for thermal decomposition and their thermal degradation mechanisms suggested (Scheme 28) [a.200]. The results showed that the amount of comonomer in the copolymer plays a vital role in thermal degradation analysis. This led to suggestions that oligomers formed from the ethylene sulfide block (up to hexamer for poly(ethylene sulfidex-co-styrene sulfidey) (x = 0.74,

Scheme 28. Thermal degradation mechanisms of PES, PSS and their copolymers Reprinted from [a.200] with permission from ACS

121

Thermal Degradation of Polymeric Materials y = 0.26) copolymer and tetramer for poly(ethylene sulfidex-co-styrene sulfidey) (x = y = 0.5) copolymer) and hetero-linkage segment up to tetramer and trimer, respectively. In poly(ethylene sulfidex-co-styrene sulfidey) (x = y = 0.5) copolymer, SE, S2, SE2 and S2E (where S = styrene sulfide and E = ethylene sulfide) were suggested, whereas SE and SE2 were detected in Py-GC/MS mode [a.79] {852665}. The absence of S2 and S2E was attributed to their thermal liability in Py-GC/MS mode, where the possibility of the primary pyrolysis products formed was likely to undergo secondary thermal degradation. However, the formation of diphenylthiophenes suggested that the initially formed styrene sulfide dimer, a primary degradation product, underwent secondary thermal degradation. It was reported that the incorporation of styrene sulfide into the copolymer backbone has the significant effect on the thermal degradation products. Further, the absence of substituted thiophenes in Py-GC/MS is explained by the presence of styrene sulfide units disturbing their formation during thermal degradation. The thiophene derivatives observed in PES were absent in poly(ethylene disulfide) (PEDS) and poly(ethylene tetrasulfide) (PETS), which indicated the influence of sulfur level on the thermal degradation products. Figure 16 displays the gas chromatograms of the pyrolysates of PES at 700 °C.

Figure 16. Gas chromatogram of the flash pyrolysates of PES at 700 °C Reprinted from [a.200] with permission from ACS

122


5.8.6 Poly(Bisphenol-A Carbonate) (PC) Poly(bisphenol-A carbonate) (PC) is an important engineering thermoplastic material, which is subjected to injection moulding operations at temperature above 300 °C. At this temperature, degradation reactions are likely to occur {732351}, and therefore the understanding of its thermal behaviour is of crucial importance in the end-use applications {895472} {891434} {888765} {882593} {856011} [a.71]. The thermal decomposition of PC {882592} {855971} (Scheme 29) has been investigated by heating isothermally at 300, 350, 400 and 450 °C and subsequent analysis of the pyrolysis residue by means of MALDI mass spectrometry {870570} {766615}.

Scheme 29. Thermal degradation processes occurring in poly(bisphenol-A carbonate): (a) intramolecular exchange reaction; (b) hydrolytic processes; (c) elimination; (d) disproportionation reaction; (e) thermal rearrangement Reprinted from {766615} with permission from ACS

123

Thermal Degradation of Polymeric Materials The MALDI mass spectrum (Figure 17) of the pyrolysis residues obtained at 300 °C showed a progressive reduction of the abundance of cyclic oligomers, whereas the relative abundance of the other compounds was unaffected. At 350 °C, the occurrence of an extensive hydrolysis reaction was responsible for the degradation of cyclic and linear chains bearing tert-butylphenyl carbonate endgroups with subsequent formation of abundant open-chain PC oligomers with phenol endgroups. Furthermore, at these two temperatures, cyclic oligomers did not disappear in the MALDI spectra, even at longer heating time of 1 h, suggesting the presence of an equilibrium between the rate of cleavage and the rate of formation of cyclic structures. PC chains terminated with phenol groups at both ends, together with pyrolysed chains bearing phenyl and isopropylidene endgroups, were generated by the disproportionation of the aliphatic bridge of bisphenol-A at 400 °C. Condensed aromatic compounds such as xanthones, which are considered to be the precursors of a graphite-like structure of the charred residue,

Figure 17. MALDI-TOF spectrum of the soluble fraction extracted from the pyrolysis residue of PC obtained after heating for 0.5 h at 300 °C (D, G, H and I represent oligomers identified) Reprinted from {766615} with permission from ACS

124

Polymers, Copolymers and Blends were also detected in the MALDI spectra of PC heated at 400 °C and which became the most intense species at 450 °C. The PC heated at temperature higher than 450 °C consisted of insoluble carbonaceous materials not suitable for MALDI analysis. The enthalpy relaxation behaviour of polycarbonate has been studied by alternating differential scanning calorimetry (ADSC) whereby the samples were first annealed at 125 °C, about 20 °C below their glass transition temperature, for periods up to 2000 h, and then scanned in the ADSC using different modulation conditions [a.201]. The data were analysed in of total, reversing and non-reversing heat flows, and also in of complex, in-phase and out-of-phase specific heat capacities and phase angle. The results showed very good agreement between the experimental results and the theoretical predictions in that the total heat flow closely corresponds to conventional DSC in respect of both peak endotherm temperature and enthalpy loss (derived from the area under the peak). In contrast, the non-reversing heat flow peak area did not provide a good measure of the enthalpy loss because the reversing heat flow (and complex specific heat capacity) depended significantly on ageing, the transition region becoming much sharper as the ageing time increased. Likewise, the phase angle (when appropriately corrected for the problem of heat transfer) also became sharper on ageing, and the (negative) peak moved towards higher temperatures. The out-of-phase specific heat capacity was calculated using the corrected phase angle, and it was shown that the area under this peak was essentially independent of ageing time – confirming another prediction from the earlier theoretical model that this area provides no information about the enthalpy loss that occurs during the ageing process.

5.8.7 Poly(Butylene Terephthalate) (PBT) Poly(butylene terephthalate) (PBT) has an excellent balance of mechanical and electrical characteristics and withstands use at high temperatures. It is thus widely used in automobile components, such as connectors {783959}. However, PBT is degraded progressively, depending on the temperature and environmental exposure due to thermal decomposition, thermooxidation and photooxidation. The thermal decomposition of PBT can be evaluated by measuring changes in the amount of COOH endgroups, and in molecular weight, caused by the scission of molecular chains. A study of primary thermal degradation mechanisms of PBT in a direct Py-MS has reported that CO–Ph–CO–O is formed as an anhydride in intramolecular reactions at a temperature of 600 °C {490068}. Recent research efforts by Manabe and Yokota {783959} reported kinetic reaction products and COOH endgroups in thermooxidation of PBT to be dominant in the range of 140–180 °C, while thermal decomposition was found to be dominant at 200 °C.

125


5.8.8 Poly(Ethylene Glycol Allenyl Methyl Ether) (PEGA) The thermal degradation behaviour of poly(ethylene glycol allenyl methyl ether) (PEGA), poly(ethylene glycol allenyl methyl ether) macromonomer (PEGA 590) and their copolymers with styrene at various compositions has been investigated by thermogravimetry. The homopolymers and copolymers exhibited one-step degradation. The thermal stabilities of the copolymers are intermediate between those of the two homopolymers. PEGA exhibits low thermal stability resulting from the thermal instabilities of the C–O bonds of the side chains and the C=C double bonds in the main chain [a.202].

5.8.9 Poly(Ether Ketone)s (PEKs) Poly(ether ketone)s (PEKs) are important engineering polymers that thermally degrade with the formation of a residue stable up to 800 °C, with comprehensive mass losses that decrease on increasing the scanning rate, at least at lower heating rates {887517}. According to the study, the TG curves of the degradation of poly(ether ketone)s show a first degradation stage at lower temperature, associated with a very sharp DTG peak, immediately followed by a second one characterised by little weight loss and broad shape of DTG curve. The researchers attributed the degradation to random chain scission (first stage) onto which branching and crosslinking became superimposed at higher temperatures (second stage). This was ed by the large mass loss associated with the first sharp DTG peak, while only a small weight loss was associated with the irregular and broad second one, with the formation of a stable residue. These results are in agreement with other literature data, in which the occurrence of the processes of random chain scission, branching and crosslinking has been observed during the thermal degradation of polymers having similar structures to poly(ether ketone)s {871496} {764038} [a.203]. The results led to suggestions that the substitution of the ketone group with the sulfone group in the polymer chain should increase the apparent activation energy of degradation, thus making the process more difficult from the kinetic point of view. In contrast, introduction into the repeat unit of a further ether group lowers the degradation activation energy value {887517}.

5.8.10 Poly(Epichlorohydrin-co-Ethylene Oxide) Poly(epichlorohydrin-co-ethylene oxide) has a well-balanced profile of physical properties together with a high resistance to solvents and oils at moderate temperatures. For these reasons it has found many applications in the aerospace and automotive industries. In addition, this copolymer has attracted attention in the field of batteries and electrical devices, mainly because, when complexed with an inorganic salt, the elastomer can be used

126

Polymers, Copolymers and Blends as a solid-state polymeric electrolyte with good ionic conductivity at room temperature. Several studies in this area have been reported, describing its use in batteries, capacitors, electrochromic displays and photo-electrochemical cells. For all these applications, good thermal stability of the polymer is desirable. The thermal degradation mechanism and kinetic parameters for the overall degradation of the poly(epichlorohydrin-co-ethylene oxide) have been investigated by the Py-GC-MS technique [a.16]. Amongst the m/z ratios observed, those corresponding to ions of m/z = 35, 36, 37 and 38 confirmed that Cl• and HCl are among the pyrolysis products. Selected ion current measurements for a wide range of other possible degradation products were examined in order to assess the general structures, and these revealed that a wide range of low-molecular-weight hydrocarbons and chlorohydrocarbons are formed on thermal degradation of the copolymer. The results suggested that a major mechanistic process is the depolymerisation of macroradicals, and that hydrogen abstraction from a carbon atom adjacent to a C–O bond is an important process in the formation of volatile products (Scheme 30) [a.14]. Using the total ion current values obtained from sequence pyrolysis experiments, quantitative kinetic evaluation of the overall rate of production of volatile products was performed. The data leading to this overall rate constant were interpreted according to the Ericsson, Guggenheim and Kezdy–Jaz–Bruylants methods. McGuire and Bryden [a.204] have studied poly(epichlorohydrin) and poly(epichlorohydrinco-ethylene oxide) elastomers by the Py-GC-MS technique and proposed that the pyrolysis products damage the capillary columns. This was attributed to the aliphatic alcohols produced during the pyrolysis being totally adsorbed on the columns, leading to irreproducible results. The major products from poly(epichlorohydrin) were in of the loss of CH3Cl, loss of HCl, and the presence of protonated epichlorohydrin {645123}. A spectrum similar to that of poly(epichlorohydrin) was obtained from the pyrolysis of the poly(epichlorohydrin-co-ethylene oxide) copolymer, except for the relative intensities of the ions of m/z = 29 and 45. The researchers believed that these ions were formed in some free-radical reactions that occurred during thermal degradation. However, from the study it was not possible to distinguish between ions characteristic of the thermal degradative mechanisms and those from electron-impact fragmentation.

127


Scheme 30. Two possible reaction pathways for the formation of the main volatile pyrolysis products of poly(epichlorohydrin-co-ethylene oxide) Reprinted from [a.14] with permission from Elsevier

128

Natural Polymers

6

Natural Polymers

Natural polymers constitute a wide class of important polymers with many commercial applications, including food packaging, fibres, fuel, coatings, automobile components, adhesives and genetic engineering materials among many others. The main categories of natural polymers are polysaccharides (starch, chitin, chitosan, cellulose and their derivatives), proteins (amino acids, enzymes and peptides) and polynucleotides (polyesters of phosphoric acid and nucleosides). Others include rubber, lignin, humus, coal, kerogen, asphaltenes, shellac and amber. With many diversified applications, natural polymers have attracted a lot of research interest, particularly in biochemisty and materials science engineering, thus making it prudent to discuss the thermal degradation of natural polymers. However, due to the wide scope of the topic, the thermal degradation of only starch, chitin, chitosan, cellulose, lignins, poly(3-hydroxyalkanoates) (PHA), proteins, natural rubber, poly(hydroxy acid)s and poly(p-dioxanone) (PPDO) are presented in this chapter.

6.1 Starch Starch is one of the main polymers present in Nature; its structure and properties have been investigated extensively in the past century. Increasing environmental concerns in recent years have led to biodegradable materials replacing petrochemical polymers in many applications. Starch has shown advantages and superior characteristics over other natural and synthetic biodegradable polymers, especially because of the low cost of the raw materials. Its applications have been extended from traditional food, paper and textile industries to packaging, controlled drug delivery and many other areas in either its native or modified forms. Along with temperature increases, the solid-state reactions of starch start with the combination of phase transitions such as melting, evaporation and sublimation as well as chemical condensation, decomposition and finally carbonisation at very high temperatures [a.444]. Thermal analysis techniques - TG and DSC - are normally applied to monitor the mass loss and the endothermic or exothermic nature of any physicochemical changes involved in thermal processes. These provide valuable insights when assessing the chemistry

129

Thermal Degradation of Polymeric Materials of thermal decomposition. The examination of the chemical structure changes that occur with increased temperatures and the thermal reaction pathways of starch are important, as the results play a fundamental role in understanding the thermal behaviour and physical properties of the material [a.205]. Studies on the thermal degradation of starch reported that thermal reactions for starch start around 300 °C with thermal condensation between hydroxyl groups of starch chains to form ether segments and liberation of water molecules and other small molecular species. Dehydration of neighbouring hydroxyl groups in the glucose ring also occurred, resulting in the formation of C=C bonds or breakdown of the glucose ring [a.445]. Aldehyde groups were formed at the same time possibly as endgroups when the glucose ring was fractured. Increasing temperature generated aromatic rings, such as substituted benzene and furan structures with either –CH2– or –CH2–O–CH2– as the main linkages between the aromatic groups. The starch structure disintegrated after heating to 400 °C, and above that temperature a highly crosslinked system was formed similar to thermally crosslinked phenol/benzene/furfuryl resins. Thereafter, the thermal reactions of the system followed similar reaction pathways as phenol–formaldehyde or furfuryl resins undergoing thermal crosslinking and decomposition at increased temperatures [a.205, a.206]. The carbonisation reactions of the system at temperatures above 500 °C increased the relative intensity of aromatic carbon resonances, with the intensities of aliphatic carbons decreasing [a.206]. Relatively large conjugated aromatic structures formed above 600 °C, and further heating generated amorphous carbon structures. The initial thermal reactivity (at around 300 °C) relied on factors such as molecular weight, pH or structural modification of the glucose units of starch and probably amylose content as well; however, the overall reaction pathway amongst these starch samples was similar.

6.2 Chitin and Chitosan Chitin is a naturally abundant mucopolysaccharide and a ing material of crustaceans, insects, etc., which is known to consist of 2-acetamido-2-deoxy--D-glucose through a (14) linkage. Chitin can be degraded by alchitinase and its immunogenicity is exceptionally low, in spite of the presence of nitrogen; and also it is a highly insoluble material resembling cellulose in its solubility and low chemical reactivity. On the other hand, chitosan is the N-deacetylated derivative of chitin, although this N-deacetylation is almost never complete – a sharp nomenclature with respect to the degree of N-deacetylation has not been defined between chitin and chitosan. Chitin and chitosan are of commercial interest due to their high percentage of nitrogen (6.89%) compared to synthetically substituted cellulose (1.25%), which makes chitin a useful chelating agent and chitosan a potential polysaccharide resource [a.207].

130

Natural Polymers The thermal degradation properties of chitosan and of lactic and/or glycolic acid-grafted chitosan studied by DSC and dynamic TG showed that the samples are thermally degraded easily after grafting the lactic and/or glycolic acid [a.208]. The initial activation energy of all grafted samples was much lower than that of chitosan and it varied with degree of conversion. The FTIR spectra of thermally degraded residues gave an indication of the chitosan polysaccharide ring degradation after 30 min at 280 °C, while glycolic acid-grafted chitosan degraded only after 15 min. In different works, the products of the reactions of chitosan with cyclic oxygenated compounds showed a decrease in stability, attributed to the removal of the free amino groups, while the reactions of chitosan with aromatic aldehydes gave Schiff-base polymers, showing that the resultant polymer is less stable than chitosan itself [a.209, a.210]. FTIR and elemental analysis results indicated that the pyrolysis of chitosan fibres at relatively low temperatures is an effective method for obtaining carbon fibres in a mechanism in which degradation takes place together with decomposition of the pyranose ring with partial dehydration and deamination [a.211]. Also at the oxidation stage some ester formation and aromatisation occurred. During pyrolysis, dehydration and deamination were completed accompanied by fusion of the aromatic rings formed – pre-treatment of fibres with NH4Cl improved the carbon yield. A study of the thermal degradation of chitosan and N,N,N-trimethylchitosan (TMCh) in nitrogen atmosphere showed that the methylation of chitosan brought about the decrease of the thermal stability of the polymer, which was more important the greater the degree of quaternisation (DQ) of the methylated derivative [a.212]. The dynamic study showed the decrease of the activation energy for the main stage of the thermal degradation of these methylated derivatives of chitosan with increasing degree of quaternisation and the same tendency was observed in the isothermal study for the degree of conversion = 0.19. The presence of O-methylated and +NH3Cl– sites in the chains of the TMCh/chitosan suggested the role of these functionalities in the polymer thermal behaviour as presented in Table 6, while Figure 18 shows the FTIR spectra obtained before and after thermal treatment. Holme and co-workers [a.213] examined the thermal depolymerisation of chitosan chloride in the solid state followed by measuring the apparent and intrinsic viscosities. The initial rate constants were determined from the intrinsic viscosity data and were found to increase markedly with increasing degree of acetylation, FA, showing that FA is an important parameter for the rate of thermal degradation. The activation energies of the three chitosan chlorides with degrees of acetylation FA = 0.02, 0.16 and 0.35 were determined to be 114, 112 and 109 kJ/mol, respectively. On the other hand, the initial rate constant for chitosan chloride prepared by freeze-drying of a solution at pH 4 was about 30 times greater than that of a sample freeze-dried at pH 6, showing that the pH of the chitosan is important for its ability to degrade. 1H and 13C NMR spectroscopy of the thermally degraded chitosan with FA = 0.35 was used to identify the specificity in the reaction. The rate of acid hydrolysis of the glycosidic bond in chitosan solutions

131


Table 6. Characteristic temperatures for the first two stages of the thermal degradation of the parent chitosan and TMCh samples from the corresponding TG and DTG curves Reprinted from [a.212] with permission from Elsevier

First stage Chitosan TMCh1 TMCh2 TMCh3 Second stage Chitosan TMCh1 TMCh2 TMCh3

DQ (%)

Range

Temperature (°C) Midpoint

Peak

Weight loss (%)

– 5.0 21.0 33.0

25–140 25–140 25–140 25–140

59.2 55.6 52.5 49.3

45.0 44.7 50.0 44.0

11.2 11.4 12.2 13.9

– 5.0 21.0 33.0

200–400 190–350 190–350 190–350

317.2 264.6 261.3 253.6

310.6 268.6 262.0 251.1

39.5 37.6 39.8 38.8

Figure 18. IR spectra of chitosan before and after heating at 280 °C for different times in a nitrogen atmosphere Reprinted from [a.212] with permission from Elsevier

132

Natural Polymers was found to be in the order A–A A–D >> D–A D–D as reported by Vårum and coworkers [a.214] for the thermal depolymerisation of chitosan – hydrolysis of the linear binary hetero-polysaccharide chitosan composed of (14)-linked 2-acetamido-2-deoxy-D-glucopyranose (GlcNAc; A unit) and 2-amino-2-deoxy--D-glucopyranose (GlcN; D unit) residues. The NMR spectra also indicated that hydrolysis of the N-acetyl bond (de-N-acetylation) at the new reducing ends occurs in addition to the cleavage of the glycosidic bond. The work further showed that acid hydrolysis is the primary mechanism involved in the thermal depolymerisation of chitosan chlorides in the solid state and that cleavage of the A–A and A–D linkages is mainly responsible for the degradation in the range of acetyl contents investigated.

6.3 Cellulose Polymers that are biodegradable are currently attracting high interest in materials science since they offer reductions of landfill space during waste management as well as new end benefits in various fields of applications {893098}. Cellulose degradation proceeds by two competing reactions, i.e., dehydration and depolymerisation {893098} {890075} {802281} [a.215] (Scheme 31). The first reaction progresses by forming CO, CO2, H2O and other volatiles as well as char with intra-ring scission of the glucose unit in cellulose chains {787558} {747400} [a.446]. The second reaction comprises transglycosylation and levoglucosan formation. This reaction is initiated by depolymerisation at higher temperature to afford a gaseous fraction containing CO, CO2 and others, a tar or heavy oil fraction containing volatile materials, and a char fraction [a.215]. The degradation of cellulosic materials has been studied in detail and the pyrolysis of cellulose and especially the mechanisms of pyrolysis reactions examined {709655} {512357} [a.454]. One work {547250} has proposed the latest theory for the mechanism of pyrolysis of cellulose while considering the degradation of cellulosic materials by application of flame-retardant treatment. Recent studies have shown that the addition of PVC or poly(vinylidene chloride) (PVDC) causes the degradation of cellulose at lower temperature and increased char, compared to pure PVC, PVDC and cellulose. Hence, cellulose is thus shown to interact with PVC and PVDC under pyrolysis conditions. Studies have suggested that cyclodextrins are suitable models for the thermal degradation of cellulose and, more generally, for polysaccharides, since the thermal degradation pattern of cyclodextrins is similar to that of cellulose [a.216]. In this respect, cyclodextrins are suitable low-molecular-weight model compounds to understand the mechanism of cellulose degradation because they are crystalline oligomers with the same structure as cellulose that can be accurately purified and characterised throughout the degradation

133


Scheme 31. Pathways of cellulose pyrolysis Reprinted from [a.215] with permission from Elsevier

134

Natural Polymers process. As a result, effective intumescent fire-retardant systems based on cyclodextrins can be developed because of the large choice of substituents available to modify their thermal behaviour. Furthermore, easy modification of the polarity of cyclodextrins may favour a tailor-made approach to chemical structure of the polymer materials by increasing cyclodextrin– polymer compatibility. A key factor in the effectiveness of this approach to fire retardancy is the control of char yield and rate of charring as a function of heating temperature, which has to be tuned to the specific polymer degradation behaviour. This requires detailed knowledge of the mechanism of thermal degradation of the char source of the intumescent system. Researchers have found that the temperature of decomposition and the amount of thermally stable residue strongly depend on the substituent [a.217, a.218]. In particular, the insertion of a substituent is able to increase the residue up to 300%, compared with the parent cyclodextrin. This knowledge would also be relevant to understanding the mechanism of the thermal degradation of cellulose, which is of paramount importance, for example, in wood combustion, chemical production from cellulosic waste and maintenance of electrical transformers. Py-GC and Py-GC-MS have been employed for studying the effect of (NH4)2HPO4 and (NH4)2SO4 on the pyrolysis of Pinus halepensis [a.219]. The results showed differences in the composition of the organic volatile products after treatment with ammonium compounds. The major ones were the formation of new products such as levoglucosenone due to the alteration of the thermal degradation mechanism, or due to the interaction of chemicals with the pine needle pyrolysis products like benzonitrile. These changes consisted of, first, the formation of new products such as levoglucosenone originating from the cellulosic material as well as aromatic nitriles and nitrophenyl compounds originating from the lignin content, and, second, the quantitative differentiation in the evolution of volatiles as in the case of phenol. The evolution of levoglucosenone – a dehydration product of levoglucosan – as a prominent product is of special importance because it shows that the retardants enhance the pyrolysis path of cellulose via levoglucosan rather than via dehydration as illustrated in Figure 19 [a.220]. From the above, it was concluded that, although pine needles are a complex natural product, their main components (as cellulose and lignin) preserve their own character under pyrolysis conditions. Analysis of the effect of cellulose derivatives on PP, PS and PE thermal degradation showed that the presence of cellulosic materials produced a slight increase in the degradation temperature associated with a change in the degradation mechanism of PP. The yield of monomer and trimer from the thermal decomposition of PS was reduced in the presence of cellulose derivatives, indicating that radical chain reactions are hindered by the presence of lignocellulosic char. On the other hand, the effect of PP on the thermal decomposition 135


Figure 19. Comparison of gas-phase infrared absorption spectra from the pyrolysis of cellulose and levoglucosan obtained under the same experimental conditions Reprinted from [a.220] with permission from Elsevier

of cellulose derivatives was negligible. However, in some cases, the polymeric matrix influenced the thermal degradation of cellulosic materials [a.221]. It has also been claimed that PVC and PVDC might affect the thermal degradation of cellulose. HCl evolved in the dehydrochlorination reaction of PVC and PVDC seemed to act as an acid catalyst to promote the dehydration reaction much more than depolymerisation in cellulose pyrolysis [a.215]. The FTIR spectra of chars produced during the pyrolysis of cellulose/PVC are shown in Figure 20. The thermal degradation characteristics of lignocellulosic materials are strongly influenced by their chemical composition (cellulose, hemicellulose and lignin contents). The proportions of these constituents in rice husks vary to some extent between varieties, which may influence their kinetic behaviour {547250}. The results of thermal degradation of rice husks show the two-step nature of the TG curves and the dual peak characteristics of the DTG curves, confirming the presence of two distinct reaction zones during

136

Natural Polymers

Figure 20. FTIR spectra of chars produced by pyrolysis of pure cellulose and PVC(5%)/ cellulose under helium. (A) and (D) 0.33 of conversion; (B) and (E) 0.61 of conversion; and (C) and (F) 0.83 of conversion Reprinted from [a.215] with permission from Elsevier

thermal decomposition [a.215]. At temperatures around 75–100 °C, small endotherms, corresponding to the evolution of water present in the samples and external water bounded by surface tension, were observed. However, at temperatures above 290 °C, very rapid degradation rates of rice husk varieties were observed in the first reaction zone, whereas lower thermal degradation rates were observed in the second reaction zone.

137

Thermal Degradation of Polymeric Materials At temperatures above 450–470 °C (depending on the rice husk variety), the thermal degradation profiles cease, leaving 15–22% residue. This is in great contrast to pyrolysis of rice husks in a fluidised-bed reactor and with zeolite catalytic upgrading {863996}. The pyrolysis oils were homogeneous, of low viscosity and highly oxygenated, which showed a decrease in yield with increasing pyrolysis temperature with a consequent increase in gas yield and decrease in char yield, while the liquid had a high proportion of associated water. Polycyclic aromatic hydrocarbons were present in the oils at low concentration and increased in concentration with increasing temperature of pyrolysis. However, the yield of oil was significantly reduced after catalysis with wood oils, with the oxygen content of the oils also markedly reduced. The conversion of the oxygenated species in the oils was largely to H2O at lower temperatures and to CO and CO2 at higher temperatures [a.222].

6.4 Lignins Lignins are natural polymers occurring in plant cell walls – wood and other plants {878940} [a.447]. It has been shown that the structures of lignins are very different from various sources, such as gymnosperm, dicotyledonous angiosperm and wheat straw (Scheme 32) among many others [a.223]. The gymnosperm lignin obtained from larch is built up mainly from the guaiacyl propane unit, and the dicotyledonous angiosperm lignin obtained from Manchurian ash is built up mainly from the guaiacyl propane unit and the syringyl propane unit. Monocotyledonous angiosperm lignin obtained from straws is built up mainly from the guaiacyl propane unit, the syringyl propane unit and 4-hydroxylphenyl propane unit. The thermal degradation behaviour of lignins is greatly influenced by their complicated structures and their isolation methods. The investigation of their thermal degradation is of interest for flame-retarded wood and char manufacture aspects. To start with, studies have been carried out in order to clarify the chemical composition of in situ lignin in various plants [a.223]. For example, Py-GC-MS in the presence of tetramethylammonium hydroxide (TMAH) was applied to kenaf (Hibiscus cannabinus) fibres. Peaks retaining the structural attributes of syringyl -aryl ether subunit dominated; the core pyrogram revealed peaks retaining the structural attribute of guaiacyl and syringyl -aryl ether subunits, and the bast profile. Both pyrograms lacked products derived from p-hydroxyphenyl -aryl ether subunit. The product distribution showed that the core in situ lignin comprises 1.5 parts of syringyl -aryl ether subunits, 1 part of guaiacyl -aryl ether subunits, and in the bast in situ lignin the syringyl -aryl ether subunits are present in greater quantities than the guaiacyl -aryl ether subunits. The syringyl to guaiacyl lignin unit (S/G) ratios determined by conventional pyrolysis and alkaline CuO oxidation

138

Natural Polymers

Scheme 32. Schematic representations of lignin molecules from wheat straw Reprinted from [a.223] with permission from Elsevier

139

Thermal Degradation of Polymeric Materials also ed the pyrolysis TMAH results. 4-Methoxy- and 3,4-dimethoxycinnamic acid methyl esters were present in the core pyrolysate and absent in the bast pyrolysate. Also, 4-methoxycinnamic acid methyl ester was present in greater quantities than 3,4dimethoxycinnamic acid methyl esters. TG-MS studies on lignins found that the intensity and the evolution profile of the products (especially water, formaldehyde, methane and methanol) reflected the severity of the isolation procedure and the origin of the lignin [a.224] [a.448]. Correlations were observed between the abundance of volatile products and the type and amount of functional groups. The terminal –CH2OH groups decomposed by the release of both water and formaldehyde, as demonstrated by the relationship between the aliphatic hydroxyl group content and the evolution of formaldehyde as well as water. The dependence of the methane yield on the methoxyl group content suggested that the scission of methoxyl groups resulted in the formation of methane and methanol. The correlations found allowed the assignments of the gaseous products to functional groups. The thermal degradation and charring of both larch lignin and Manchurian ash lignin in the condensed phase were comparatively investigated using TG, FTIR and X-ray photoelectron spectroscopy (XPS) [a.225]. TG experimental results showed that larch lignin produced more char residue than Manchurian ash lignin under pure nitrogen at high temperature, which demonstrated that the carbon backbone of larch lignin was more stable than that of Manchurian ash lignin. This was attributed to more carbon–carbon bonds existing in larch lignin than in Manchurian ash lignin. FTIR and XPS data indicated that the cleavage of aliphatic ester bonds took place mainly under pure nitrogen, and more aromatic rings remained in the condensed phase. Manchurian ash lignin showed a high crosslinking rate based upon the relative intensity of C 1s and C 1s (C–C) and an obvious increase of the ratio of carbon to oxygen.

6.5 Poly(Hydroxyalkanoate)s (PHAs) Poly(3-hydroxyalkanoate)s (PHAs) – namely, poly(3-hydroxybutyrate) (PHB), poly(3hydroxyvalerate) (PHV) and poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBHV) – are biodegradable and biocompatible polymers produced by various bacteria, such as Ralstonia eutropha {776127}. These naturally occurring biopolyesters are optically active bacterial carbon reserves and energy storage materials. However, since carboxylic esters bearing a -hydrogen decompose at high temperatures in the absence of solvent, PHAs are thermally unstable in their melting point range, 70–180 °C {760252} [a.226]. Lower-molecular-weight PHAs can be used as components to build macromolecular architectures such as block and graft copolymers. Depending on the choice of the other block component, the copolymers can have amphiphilic and biocompatible properties for

140

Natural Polymers drug delivery excipients or can be surfactants and compatibilising adjuvants. They can be prepared by various methods, e.g., by limited polymerisation of -butyrolactone, by degradation of high-molecular-weight PHA, via acid-catalysed methanolysis, by acid/base hydrolysis, or via thermal degradation [a.147, a.227]. The thermal degradation of microbial polyesters has been proposed as a source of vinyl oligoesters. The oligomers produced by thermal degradation of bacterial PHV present a well-defined structure, with a carboxylic acid end and an unsaturated end on each chain, the latter being predominantly trans [a.227]. The molecular weight of the produced oligomers is affected by the hydroxyvalerate content and can be tuned by the reaction temperature and the reaction time. The thermal degradation at 190 °C for PHB, PHV and different copolymers of PHBHV was found to follow a random scission model for the first hours of the reaction. It subsequently autoaccelerated, probably because of the inductive effect of the unsaturated endgroups on the neighbouring ester linkages, which would have progressively increased the rate of ester cleavages. The 1H NMR spectrum obtained for PHV is shown in Figure 21. Scale-up studies in the absence of solvent allowed the production of important quantities of PHB oligomers in satisfactory yields. Polymerisation of the oligomers via the crotonate endgroups using free-radical methods presents difficulties. Alkyl crotonates do not polymerise by free-radical initiation because the presence of the methyl substituents on the unsaturations result in steric hindrance with the R substituent, and also degradative chain transfer. In addition, the steric hindrance is all the more enhanced when an oligomer is used instead of a monomer. However, chemical modification of one endgroup or the other can be done to yield new terminal groups, which would be more prone to free-radical polymerisation. This approach is being actively studied to prepare comb polymers [a.227]. The thermal decomposition of poly(3-hydroxyoctanoate-co-3-hydroxy-10-undecenoate) (PHOU) and the completely epoxidised form of this polymer, poly(3-hydroxyoctanoate-co3-hydroxy-10,11-epoxyundecanoate) (PHOE), has been studied by TG and DSC [a.228]. The thermal curves of PHOE that differ in the contents of epoxy units display a three-step degradation process, while those of the initial PHOUs exhibit only a one-step degradation process. This degradation behaviour of the PHOE, which have a higher thermal stability as measured by weight loss, was probably controlled by crosslinking reactions of the pendant epoxide groups in the polymer, which occurred during the degradation process – the presence of such reactions could be assigned to the exothermic peaks in their DSC profiles. An isothermal study of these polymers at 250 °C for 1 h indicated that the residual weight correlated directly with the amount of epoxide groups in the PHOE. Similar results have been presented whereby the thermal degradation of the epoxidised PHAs was characteristic in behaviour and different from that of the unsaturated PHA, 141


Figure 21. (a) The 500 MHz 1H NMR spectrum of a sample of PHV thermally degraded at 190 °C for 3 h. (b and c) Multiplet peaks of the proton . Here , and are, respectively, the protons in the trans configuration of the double bond, the cis configuration, and the proton for trans-2-pentenoic acid, and probably small oligomers where the pentenoic end has the trans configuration. (d) Peaks of the proton . Here , and are respectively the protons in the trans configuration of the double bond, the cis configuration, and the proton for trans-2-pentenoic acid, and probably small oligomers where the pentenoic end has the trans configuration Reprinted from [a.227] with permission from ACS

142

Natural Polymers which followed the same type of degradation pathway as for PHB [a.229]. Thus, it can be assumed that the complex decomposition process of epoxidised PHA may be caused by the thermal crosslinking of pendant epoxide groups with terminal carboxyl (and resultant hydroxyl) groups, which were formed by the -elimination reaction.

6.6 Proteins Some of the proteins that have been studied as potential film-forming agents include sodium caseinate, whey protein concentrate and gelatin. Casein is the main protein of milk, representing 80% of the total milk proteins; it is a phosphoprotein that may be separated into various electrophoretic fractions, -casein, -casein, -casein and -casein, which differ in primary, secondary and tertiary structure and molecular weight. The low content of cysteine and consequently of disulfide crosslinks results in a molecule with an open and random coil conformation. Whey is the other protein fraction of milk and represents the remaining 20% of the total milk proteins. It is composed of -lactoglobulin, -lactoalbumin, bovine serum albumin and immunoglobulin. Gelatin is a protein resulting from partial hydrolysis of collagen. It is differentiated from other proteins by the absence of appreciable internal order and the random configuration of the polypeptide chains in aqueous solutions. The thermal degradation of edible films based on pure sodium caseinate, whey and gelatin, and for these proteins in the presence of sorbitol as a plasticiser, was studied by TG and FTIR [a.230]. The initial temperature of degradation of the pure edible protein films was in the range 295–300 °C, but the presence of sorbitol significantly reduced the activation energy of the degradation of the edible protein films, as shown in Figure 22. This behaviour was in agreement with the decrease of the initial and maximum temperatures of degradation observed by TG/DTG. The decrease in the thermal stability is apparently associated with the effect of sorbitol on the inter- and intramolecular hydrogen bonds of the proteins. The FTIR spectra showed that the effective degradation began at ca. 300 °C with the formation of gas products, such as CO2 and NH3, suggesting that the reaction mechanism included at the same time the scission of the C–N, C(O)–NH, C(O)–NH2, –NH2 and C(O)–OH bonds of the proteins. The suggested mechanism of reaction was ed by the high values of the activation energy (E > 100 kJ/mol), which the study associated with a process that occurred by random scission of the chain. A study on cottonseed proteins found them to be thermoplastic materials with a Tg ranging from 80 to 200 °C when the glycerol content varies from 0 to 40 wt% (dry basis). The thermal denaturation temperature of the proteins increased from 141 (without glycerol) to 195 °C in the presence of 40 wt% glycerol. The thermal degradation of the proteins

143


Figure 22. Plots of activation energy versus the weight-loss fraction for the edible protein films. SC, sodium caseinate; WPC, whey protein concentrate; G, gelatin; and S, sorbitol Reprinted from [a.230] with permission from Elsevier

occurred at 230 °C as shown in Figure 23 irrespective of glycerol content, with the release of a variety of compounds [a.231].

6.7 Natural Rubber The major degradation product of natural rubber is 1-methyl-4-(1-methylethenyl)cyclo hexene. The presence of this compound as the major degradation product along with 2methyl-1,3-butadiene (monomer) and groups of compounds containing 15 and 20 carbon atoms (three and four monomer units) in the pyrolysate of a rubber is sufficient to identify it as natural rubber. Similarly, the presence of 1-chloro-4-(1-chloroethenyl)cyclohexene and 2-chloro-1,3-butadiene, the cyclic dimer and monomer of poly(chloroprene) rubber, in the pyrolysate of a rubber identify it as poly(chloroprene) rubber. A correlation between the crosslink density and the product ratio of isoprene dimer species to isoprene formed from pyrolysis of natural rubber vulcanisates has been reported {697436} [a.232]. The major products of the isoprene dimer species were 1,4-dimethyl-4-vinylcyclohexene and

144

Natural Polymers

Figure 23. TG and DSC analysis of thermal degradation of cottonseed protein isolate and glycerol/isolate 40% Reprinted from [a.231] with permission from ACS

1-methyl-4-(1-methylethenyl)cyclohexene formed from natural rubbers having head-tohead and head-to-tail linkages, respectively. It was found that the relative abundance ratio of the isoprene dimer species to the isoprene decreased in proportion to an increase in the crosslink density of the vulcanisate. Vulcanised polyisoprene with various crosslink densities was also investigated, and the structure and composition of the degradation products were determined {596000} [a.233]. In one study, the pyrolysis of cis-1,4-polyisoprene exhibited mainly three exotherms and two endotherms in the DTA curve [a.233]. It was further found that major reactions during the pyrolysis are not affected seriously by temperature changes from room temperature to 430 °C. The major products were identified via Py-GC as dipentene, isoprene, trimeric isoprene, toluene, benzene and xylene (Figure 24) {594526}. Analysis of the products showed that cis-1,4-polyisoprene, when pyrolysed under an inert atmosphere, experienced a radical mechanism. The chain scission mainly occurred at the -position of carbon–carbon single bonds, which are adjacent to double bonds. The pyrolysis process of cis-1,4-polyisoprene was accompanied by dehydrogenation and aromatisation. However, dipentene was obtained as a major pyrolysis product below 430 °C and its yield decreased strongly with increasing temperature [a.234]. The obtained Py-GC results are displayed in Figure 25.

145


Figure 24. Pyrogram of natural rubber with schemes showing depropagation and intramolecular transfer process Reprinted from {594526} with permission from Elsevier

146

Natural Polymers

Figure 25. Py-GC chromatograms of cis-1,4-polyisoprene at different temperature ranges: (A) Pyrolysis of cis-1,4-polyisoprene from room temperature to 330 °C. (B) Pyrolysis of cis-1,4-polyisoprene from 331 to 390 °C. (C) Pyrolysis of cis-1,4-polyisoprene from 391 to 430 °C. (D) Pyrolysis of cis-1,4 polyisoprene from 431 to 600 °C. Peaks: 1, isoprene; 2, benzene; 3, toluene; 4, xylene; 5, dipentene; 6, trimeric isoprene Reprinted from [a.234] with permission from Elsevier

147

Thermal Degradation of Polymeric Materials Thermal degradation studies showed that isoprene and dipentene are formed in high concentration in natural rubber pyrolysis, leading to suggestions that both isoprene and dipentene are produced by depolymerisation from polymer radicals occurring by -scission at double bonds {783965}. The polymer radicals were liable to form six-membered rings, especially under mild pyrolysis conditions, so the dipentene was formed predominantly at lower temperature. Different works have suggested that decomposition followed radical generation via polymer chain scission resulting in the formation of isoprene, dipentene and other smaller compounds [a.235]. Another work has suggested that the other products of pyrolysis can be ed for by the thermal decomposition of isoprene and dipentene themselves [a.236]. Groves and co-workers [a.237] analysed the oil derived from the pyrolysis of natural rubber in a Py-GC at 500 °C. These researchers showed that the major products were the monomer, isoprene, and the dimer; dipentene, with other oligomers up to hexamer also being formed in significant concentrations. It was suggested that the isoprene monomer was formed via a depropagating mechanism in the polymer chain, and that dipentene dimer was formed either by intramolecular cyclisation followed by scission, or by monomer recombination via a Diels–Alder reaction. It has also been reported that dipentene (DL-limonene) was formed in trace quantities as the pyrolysis temperature was increased to 550 °C [a.238]. In the range from room temperature to 310 °C, the yield of dipentene was predominant, suggesting that it was the best temperature range to obtain dipentene, which means that limonene is formed in high concentration at low temperatures and degrades as the pyrolysis temperature is increased. Apparently, high pyrolysis temperature decreases the DL-limonene yield due to the cracking of the pyrolysis oil. Earlier studies showed that DL-limonene was obtained as the main product from rubber pyrolysis [a.238, a.239], but the yields of benzene were found to be relatively unaffected with increasing temperature. The yields of xylene increased with increasing temperature at 430–600 °C whereby high yields of chain hydrocarbons were obtained – representing the deep decomposition of the polymer [a.240].

6.8 Poly(Hydroxy Acid)s

6.8.1 Poly(L-Lactic Acid) (PLLA) Poly(hydroxy acid)s are an important class of degradable polymers for biomedical applications due to their biocompatibility and physiologically tolerable degradation products {886302}. Poly(L-lactic acid) or poly(L-lactide) (PLLA) has been used as a biomaterial for tissue engineering, bone fracture fixation and controlled drug delivery

148

Natural Polymers because of its biodegradability, biocompatibility and processability [a.176, a.241]. PLLA is generally prepared by the ring-opening polymerisation of L-lactide, a cyclic dimer of Llactic acid. This reaction is an equilibrium reaction in which the concentration of residual L-lactide depends on temperature. Therefore, the L-lactide is regenerated through the thermal degradation process of PLLA. However, the thermal degradation of PLLA is more complex than the simple reaction that gives L-lactide, and involves the generation of significant amounts of other volatile decomposition products during pyrolysis, e.g., cyclic oligomers, lactides, carbon dioxide, acetaldehyde, ketene and carbon monoxide {848609} {759798} {756177} [a.241]. Another report found that intramolecular transesterification was the dominant degradation pathway, and that pyrolysis behaviour was different between pure and Sn-containing PLLAs [a.242]. Metals such as Sn, Zn, Al and Fe have a great influence on the pyrolysis behaviour of PLLA. Babanalbandi and co-workers {756177} reported activation energy values for PLLA using isothermal methods and showed that at first the activation energy value decreased from 103 to 72 kJ/mol with increase in weight loss and then increased up to a value of 97 kJ/mol. It was postulated that the PLLA degradation process follows more complex kinetics, even at low conversion. Lastly, studies on the degradation behaviour of PLLA found that the activation energy values changed in the range 80–160 kJ/mol with change in weight loss and concluded that the pyrolysis of PLLA involved more than two mechanisms [a.243]. In short, the factors that influence PLLA thermal decomposition include moisture, residual and hydrolysed monomers, oligomers, molecular weight and residual metals. However, the degradation of PLLA and copolymers in an aqueous environment generally occurs through hydrolysis of the ester group in the main chain [a.244].

6.8.2 Poly(L-Lactic Acid) Blends PLLA and poly(D-lactide) (or poly(D-lactic acid) (PDLA)) and their equimolar enantiomeric blend (PLLA/PDLA) have been prepared and the effects of enantiomeric polymer blending on the thermal stability and degradation of the films were investigated isothermally and non-isothermally under nitrogen gas using thermogravimetry [a.245]. The enantiomeric polymer blending was found to successfully enhance the thermal stability of the PLLA/ PDLA system compared with those of the pure PLLA and PDLA films. The activation energy values of the PLLA/PDLA, PLLA and PDLA films were in the range of 205–300, 70–130 and 155–240 kJ/mol when they were evaluated at weight-loss values of 25–90% and the activation energy value of the PLLA/PDLA blend was higher by 82–110 kJ/mol than the average activation energy value of the PLLA and PDLA [a.246]. Isothermal measurements at constant holding temperature in the range of 250–270 °C showed the percentage remaining weight of the PLLA and PDLA as a function of degradation time.

149

Thermal Degradation of Polymeric Materials At 250 °C the remaining weights of the PLLA/PDLA started to decrease without any induction period and then decreased monotonically to zero at 120 min and to 4% at 200 min, while that of the PLLA/PDLA started to decrease after ca. 40 min of induction period and then decreased gradually to 23% at 200 min. It is interesting to note that the remaining weight of the PDLA was higher than that of the PLLA for all the degradation time; this is in marked contrast with the results in non-isothermal measurements in which actual degradation occurred at a temperature exceeding 290 °C. This result can be explained as follows: The PDLA had higher initial molecular weight than the PLLA and therefore the PDLA had a lower amount of terminal groups where cyclic oligomers and monomers are formed, resulting in a higher remaining weight compared with that of the PLLA {870313}. McNeill and Leiper [a.247] and Babanalbandi and co-workers {756177} evaluated the activation energies for thermal degradation of PDLA and PLLA to be 119 and 70–105 kJ/mol, respectively. It was reported that the terminal groups play an important role for thermal degradation; the PDLA having a higher molecular weight was expected to have a higher thermal stability than that of the PLLA having a lower molecular weight. It is plausible that the PDLA contains a higher amount of tin catalyst and/or lactide remaining after the purification by precipitation, which may have enhanced the thermal degradation of the PDLA film at temperatures exceeding 290 °C. On the other hand, another work found that a stereocomplex is formed from enantiomeric PLLA and PDLA due to the unusually strong interaction between PLLA and PDLA chains [a.248]. The stereocomplexed PLLA/PDLA blend a had melting temperature ca. 50 °C higher than those of pure PLLA and PDLA [a.248], and could retain strength in the temperature range up to the melting temperature of 230 °C. Moreover, the PLLA/PDLA blend has a higher hydrolysis resistance compared with that of the pure PLLA and PDLA even when it is made in amorphous fashion, due to the strong specific interactions between PLLA and PDLA chains [a.245].

6.9 Poly(p-Dioxanone) (PPDO) Poly(p-dioxanone) (PPDO) is well known as an easily hydrolysable material and has been used to make monofilament structures with good tenacity and knotting. It has been suggested that the polymerisation of 1,4-dioxan-2-one (p-dioxanone, PDO) is an equilibrium reaction. Because of its good physical properties and biocompatibility, and also because it can be produced by a simple production process from diethylene glycol, which is an inexpensive raw material, PPDO is viewed as a candidate not only for medical use, but also for universal uses such as films, moulded products, laminates, foams, nonwoven materials, adhesives and coatings. Thermal degradation of PPDO begins at above 200 °C and reaches almost quantitative decomposition at 320 °C [a.249]. The thermal decomposition behaviour of PPDO has been presented, in which the exclusive evolution

150

Natural Polymers of PDO by a zero-order unzipping depolymerisation reaction was found and a random degradation process in the initial stage was also suggested [a.250]. It is not clear whether the suggested random process in the initial stage is an induction process followed by the main process or merely a minor reaction. The induction process is important in the degradation, because it would considerably influence the following main process. Parallel studies reported that, though the cyclic monomer PDO was the most abundant of the pyrolysis products, some random scission and endgroup-derived products were also present, which suggests that some competitive minor reactions proceeded [a.251]. From the simulation studies, the pyrolysis was analysed as mainly proceeding by the zero-order unzipping depolymerisation; however, some random reactions also proceeded competitively in the initial stage, especially under rapid heating conditions.

151


152

Reinforced Polymer Nanocomposites

7


Polymer composites are composed of a polymer matrix in which another well-defined material with distinctive phase boundary is dispersed. Hence, the thermal properties of polymer composites depend on both the macromolecular matrix and the additives. Thermal analysis methods have proved useful not only in defining suitable processing conditions for these materials and drawing up useful service guidelines for their application, but also for obtaining information on thermal properties–polymer structure relationships. There is currently a high level of interest in using nanoscale reinforcing fillers for the preparation of polymeric nanocomposite materials with exceptional properties. An improvement in the thermal stability and flammability properties of polymers has been obtained with nanoscale additives, and these filled systems provide an alternative to conventional flame retardants. It is therefore important to explore how the asymmetry (aspect ratio) and other geometrical effects of nanoadditives influence the thermal degradation properties of polymer nanocomposites. Despite its significance, the topic of nanocomposites has only been highlighted in this work through a few representative cases and readers are directed to current specialised reviews available in the literature [a.252, a.253]. Nevertheless, the thermal degradation of some of the most common reinforced polymer composites is presented in the following sections.

7.1 Glass-Fibre-Reinforced Composites The thermal stability and high-temperature mechanical properties of silicate matrix composites reinforced by carbon and SiC-based fibres in oxidising environments have been investigated quite extensively in the past by conducting thermal ageing and thermal cycling experiments over a wide range of temperatures. Most of these works have focused on hybrid inorganic/polymer composites (hybrid glass and glass/ceramic matrix composites), e.g., barium magnesium aluminosilicate glass/ceramic composites containing both SiC– Nicalon™ fibres and SiC whiskers, cordierite glass/ceramic matrix composites containing SiC monofilament and SiC whiskers, borosilicate glass containing SiC–Nicalon fibres, carbon fibres and various ceramic particle fillers (i.e., alumina, zirconia or carbon), and aluminosilicate glass containing SiC–Nicalon fibres and SiC particles [a.254]. The results

153

Thermal Degradation of Polymeric Materials of investigations conducted at temperatures in the range 500–700 °C show a decrease of tensile and flexural strength of the composites. Figure 26 shows the XRD results that demonstrated that no new crystalline phases were formed during thermal ageing. It has also been shown that this is the consequence of oxidation of the fibres, in the case of carbon-fibre-reinforced composites, or of degradation of the fibre/matrix interphase, which is in fact a carbon-rich nanometric interfacial layer, in the case of SiC-fibre-reinforced composites. Works reporting on the thermal degradation of glass-reinforced composites are scarce in the literature, with the few that are available only covering thermooxidative degradation. In one of the works on thermal degradation, thermal ageing in argon of an SiC-fibre-reinforced glass matrix composite was investigated at temperatures in the range 500–700 °C for an exposure duration of up to 1000 h [a.255]. An inert atmosphere was used to study the effects of temperature alone; thus the effects of oxidation are minimised and may be neglected. The fracture toughness values determined by chevron-notch tests were little affected by the ageing conditions and were in the range 19–26 MPa/m2. The frictional interfacial shear stress was not affected by the ageing conditions either. For the most severe ageing conditions investigated (1000 h at 600 °C and 100 h at 700 °C), a significant loss

Figure 26. XRD patterns of (a) as-received and (b) thermally aged (700 °C/24 h) SiC– Nicalon fibres Reprinted from [a.254] with permission from Elsevier

154

Reinforced Polymer Nanocomposites of flexural strength and stiffness of the samples was detected, as displayed in Figure 27, which was ascribed to the microstructural changes that occurred in the material during ageing as a consequence of the softening of the inorganic polymer glass matrix. At these ageing conditions, a lower interfacial shear stress for fibre–matrix debonding initiation was measured, which was explained also by the occurrence of matrix softening and void formation. A few of the published works on the thermal degradation of glass-fibre-reinforced composites in air conditions are presented in the following paragraphs. In a recent work the thermal ageing of a glass matrix composite reinforced by short carbon fibres as well as by ZrO2 particles (hybrid composite) was investigated at temperatures in the range

Figure 27. Load–displacement curves of SiC-fibre-reinforced glass matrix composite obtained in chevron-notch tests for three samples aged at 600 °C for different durations Reprinted from [a.255] with permission from Elsevier

155

Thermal Degradation of Polymeric Materials 500–700 °C for exposure durations of 24 h in air [a.254]. The mechanical properties of as-received and aged samples were evaluated at room temperature by using the three-point flexure chevron-notch technique. The fracture toughness values of as-received specimens were in the range 2.5–6.5 MPa/m2. Fracture toughness was affected by the thermal ageing conditions. For thermal ageing at temperatures <700 °C, degradation of fibre/matrix interfaces occurred and therefore the apparent fracture toughness and flaw-tolerant resistance decreased. For the most severe ageing conditions tested (700 °C/24 h), the fracture toughness values dropped to 0.4 MPa/m2. Significant degradation of the material was detected for this ageing condition, mainly characterised by porosity formation in the matrix as a result of softening of the glass and oxidation of the carbon fibres. Sutherland and co-workers [a.256] investigated two silicate matrix composites, Pyrex/Nicalon and BaO–MgO–Al2O3–SiO2 (BMAS)/Tyranno, studying composite stability with respect to time at various temperatures and under applied stress. Samples aged in an oxidising atmosphere were tested in flexure at room temperature, and also by fibre ‘push-down’ to investigate the interfacial properties. Tensile tests were carried out from room temperature up to 1200 °C on the BMAS material, and it was found that a steady degradation in strength occurred from 500 to 1100 °C, with a small but significant increase up to 1200 °C. Boccaccini and co-workers [a.257] studied the evolution of thermal damage in SiC–Nicalon fibre-reinforced glass matrix composites under thermal cycling conditions in an oxidising atmosphere. The samples were alternated quickly between high-temperature (T = 700 °C) and room-temperature air for different numbers of cycles for long periods of up to 250 h. The results were that the flexural strength and Young’s modulus decreased, while the internal friction increased with increasing numbers of cycles. Material degradation was attributed to phenomena related to viscous flow of the glass matrix, and to oxidation of the fibre, which occurred as a consequence of the extended exposures at high temperatures. The microstructural damage observed included porosity formation (cavitation) within the matrix and at the fibre/matrix interfaces. The experimental results also suggested that degradation of the in situ fibre strength occurred as a result of fibre surface oxidation and damage, fibre displacement and consequent fibre-to-fibre . Under thermal shock conditions, the mismatch in the coefficients of thermal expansion of the matrix and the fibres can contribute significantly to microcracking and indirectly to degradation in fibre strength through oxidation. Consequently, the thermal histories of the material, as well as the susceptibility of the material to a thermally dynamic environment, play a pivotal role in determining the mechanical properties of the material. In this respect, Chawla and co-workers [a.258] have studied the Nicalon-fibre-reinforced hybrid composite with a matrix of barium magnesium aluminosilicate (BMAS) glass with silicon carbide whiskers subjected to thermal shock from elevated to ambient

156

Reinforced Polymer Nanocomposites temperatures. The combination of SiC whiskers and BMAS glass resulted in a hybrid matrix with a lower thermal expansion than that of the fibres, inducing tensile stresses in the fibres upon thermal shock. This stress state resulted in microstructural damage in the form of fibre cracking and cracking along the fibre/matrix interface, as opposed to the conventional matrix cracking that is typically observed in ceramic matrix composites. Significant damage in the composite was only observed after three thermal shock cycles. Flexural resonance measurements showed a reduction in Young’s modulus that correlated well with the onset of microstructural damage. Finally, fibre push-out tests, performed to evaluate changes in fibre/matrix interface strength after thermal cycling, indicated a slight decrease in interfacial strength, which was attributed to recession of the carbon-rich fibre surface during thermal shock.

7.2 Carbon-Fibre-Reinforced Composites Carbon-fibre-reinforced composites (CFRC) possess good mechanical properties and also retain their strength and modulus at high temperature. Therefore, they are primarily developed and designed for high-temperature structural applications. In fact, CFRC retain their strength in an inert atmosphere well above 1000 °C, where superalloys and even ceramics lose their strength. For example, the experimental results of Tzeng and Lin [a.259] indicated that the carbon/carbon composites with interfacial carbon layers possess higher fracture energy than that without carbon layers after carbonisation at 1000 °C. For a pitch concentration of 0.15 g/ml, the carbon/carbon composites had both higher flexural strength and higher fracture energy than composites without carbon layers. In addition, carbon/carbon composites have low thermal expansion and good wear resistance, as well as excellent biocompatibility. As a result, applications were also found as refractory materials, brake linings for high-speed vehicles and biomedical load materials [a.259]. Naruse and co-workers [a.260] studied the thermal degradation of unidirectional CFRC rings, and the respective degradation of fatigue strength was investigated by applying the Arrhenius equation to confirm the relation between temperature and time. Hindeleh and Abdo [a.261] have studied the effect of thermal exposure on Kevlar fibres when exposed to a controlled atmosphere of nitrogen gas at temperatures above 150 °C for a duration of 15 min. It has also been reported [a.262] that after exposures of 250 h at 250 °C and 6 h at 350 °C, respectively, Kevlar-49 fibres turned brittle and crumbled with handling and were not suitable for further tensile tests. Thermal ageing treatments under vacuum environments at temperatures of 100–300 °C for durations from 2 to 8 h have been made on Kevlar-29 yarns [a.263]. It was established that both the tensile strength and the tensile strain decreased with increase in treatment temperature. For treatment at constant temperature, the heating time did not appear to have any effect on the tensile strength and the tensile strain. It was also found that the Young’s modulus of single fibres was

157

Thermal Degradation of Polymeric Materials not affected by heat treatment under the conditions described above. Heat treatment in vacuum did not have any effect on the tensile strength. Investigations have been conducted into thermal-cycling-induced deformation and damage in a carbon-fibre-reinforced copper composite under various thermal cycling conditions [a.264]. The average thermal expansion coefficient of the copper/carbon composite over the temperature range of 50–800 °C was found to be 1 × 10–6 K–1. Upon heating, the longitudinal thermal strain of the copper/carbon composite levelled off at a temperature >400 °C, which was related to the yielding and/or creep of Cu. As the number of thermal cycles increased, the axial extension initially showed a rapid increase and then gradually approached a saturation value (see Figure 28). This phenomenon was attributed to fibre breakage and void formation in the interior of the composite. Unlike in tungsten-wire-reinforced copper composites, in which interfacial sliding was suggested as the major deformation mechanism, in the copper/carbon composite void formation and growth were identified as the predominant mechanisms to relax the internal stress induced during thermal cycling. Cakmak and co-workers [a.265] performed a study on the carbon fibre/PDMS/PPy composite and observed major weight loss at 400 °C (Figure 29) in a more gentle method

Figure 28. The longitudinal thermal strain in the unidirectional Cu/C composite varies with temperature in the first two cycles as the temperature cycles between 50 and 800 °C Reprinted from [a.264] with permission from Elsevier

158

Reinforced Polymer Nanocomposites of degradation as compared to pure PDMS. At this temperature only 18 wt% losses were observed – even after the composite was subjected to a temperature of 1000 °C, only 37 wt% loss occurred. It was suggested that the formation of strong adhesive bonds between carbon fibre and polymer matrix, which stabilised the composite against thermal decomposition, might explain the enhanced thermal stability of the composites. Advances made in polymers reinforced with nanomaterials suggest that optimum nanomaterial loading in reinforced composites greatly enhances the thermal degradation properties. For instance, both catalytically grown nanofibres (CGNF) and nanotubes (CGNT) stabilise the matrix, probably due to radical capture by the nano-object surface, against the first stage of degradation, even at low loading fractions [a.142]. Similar effects were observed in nanotube composites based on other polymers. The weights remaining after complete polymer decomposition were qualitatively consistent with the fraction of nanotubes present. Quantitatively, however, the values were distinctly low, presumably because most of the well-dispersed nanomaterials were physically lost from the sample as the polymer decomposed. During TG experiments, CGNF and CGNT showed similar

Figure 29. TG and DTG profiles of PDMS/carbon fibre/PPy (3 : 20 : 77 w/w) composite Reprinted from [a.265] with permission from Elsevier

159

Thermal Degradation of Polymeric Materials behaviours, with a sharp weight loss on increasing temperature, peaking at about 600 °C. The main feature of the entangled CGNT data was similar, but there was an earlier weight loss of about 5% at 400 °C, suggesting the presence of a small amount of amorphous carbon. The arc-grown nanotubes exhibited the highest thermal stability, with the maximum weight loss occurring at around 700 °C, reflecting the higher crystalline quality of the material. The pure PA-12 matrix shows a two-stage decomposition process: a weight loss of about 10% occurs at 350 °C, followed by a second process centred at 450 °C – as can be seen in Figure 30. In conclusion, as seen in this work, the advent of polymer nanocomposites has expanded the horizons of CFRC by a great deal. The current limitations for CFRC for structural use include the relatively immature design and analysis practices, manufacturing scale-up, the effect of service exposure and non-destructive inspection for bonded construction. Laser-induced ablation technology for CRFC is likely to improve the workability (cutting, drilling, etc.) for polymer materials with improved properties. Potential applications for CRFC calls for improvement in matrix chemistry, better control of the resin/fibre interface, and the use of novel reinforcement approaches, e.g., by applying novel nanoadditives. New

Figure 30. Weight losses as a function of temperature for pure PA-12 fibres and PA-12 nanocomposite fibres containing various filler weight fractions of carbon nanofibres (CNF) Reprinted from [a.142] with permission from Elsevier

160

Reinforced Polymer Nanocomposites development methods in polymer matrix chemistry are expected to lead to improvements in mechanical performance, processing techniques and long-term durability at high temperatures. There is a need to identify and/or optimise polymerisation and processing conditions to enable resin transfer moulding processability without sacrificing long-term durability and high-temperature performance (currently limited to ca. 300 °C).

7.3 Unsaturated Polyester Resins Reinforced with Fibres Most studies on the thermal degradation of phthalate-based polyesters have been made in the temperature range of 200–600 °C [a.266]. Nonetheless, several studies have identified phthalic acid anhydride as one of the major degradation products in phthalate-based polyesters. Benzoic acid, 2-propenyl ester of benzoic acid, cyclic ethers from the diol, cyclic diesters and a variety of mono- and diesters of phthalic acid were some of the other products identified. Reinforcement of unsaturated polyesters with glass fibre was found to change the crosslinking kinetics. The presence of glass fibre in the vinyl ester matrix limited copolymerisation of the vinyl ester prepolymers with styrene, resulting in an insufficiently crosslinked material with low thermal stability. The glass fibre reinforcement also increases the quantity of degradation products. The mechanisms and degradation products formed at high temperatures may vary from those formed at the storage and use temperatures. In an early study of monomeric esters, it was shown that esters containing at least one hydrogen on the -alkoxy carbon atom decompose mainly to an olefin and an acid, whereas those lacking hydrogen exhibit greater thermal stability and a more complex pyrolysis pattern [a.267]. By multivariate data analysis, it has been demonstrated that developed partial least-squares models show a good correlation between amount of identified products and degradation time. By GC-MS several low-molar-mass phthalates and alcohols were identified during the degradation of glass-fibre-reinforced unsaturated polyester composites. The polyesters were subjected to accelerated ageing at 40 or 60 °C and 80% relative humidity, after being stored for 20 years at ambient temperature. In most cases it results in the same products but in varying amounts that were present after different degradation times. Diallyl phthalate was the most abundant product in all the GC-MS chromatograms/mass spectra [a.267]. In addition, several other phthalates were identified, whose concentration decreased during ageing. At the same time, the concentration of degradation products from the phthalates, e.g., phthalic acid anhydride, isobutanol, allyl alcohol and 1-butanol, increased. The concentration of other phthalate degradation products, e.g., benzoic acid and 2-propenyl ester of benzoic acid, remained basically constant. Temperature had a large influence on the degradation of phthalates and the formation of alcohols. Only small amounts of alcohols were formed during six years at 40 °C, but they were the

161

Thermal Degradation of Polymeric Materials major product in the chromatograms after six years at 60 °C. Phenyl ester of benzoic acid was one of the most abundant products before accelerated ageing; it was most likely a recombination product of benzoyl peroxide that was used for curing the polyester.

7.4 Reinforced Polyurethane Composites The thermal degradation of short-polyester-fibre-reinforced polyurethane composites with and without different bonding agents showed that the degradation of the polyurethane took place in two steps while that of the composites took place in three steps [a.268, a.269]. With the incorporation of 30 phr of fibre in the matrix, the onset of degradation was shifted from 230 to 240 °C. The presence of bonding agents in the virgin elastomer and the composite gave improved thermal stability. Results of kinetic studies showed that the degradation of polyurethane and the reinforced composites with and without bonding agents followed first-order reaction kinetics. In recent times, much research interest has been focused on HTPB-based polyurethanes, which are used as solid composite propellants in space applications, coatings, adhesives and sealants. Hence a thorough study of the thermal degradation of these types of polyurethanes at high temperature is important to detect their service temperature as well as the probable degradation products to take measures against toxicity and pollution. Lee and Ko [a.270] have shown that the increase in chain extender concentration lowered the initial degradation temperature as revealed by TG. On the other hand, an optimised hard-segment concentration is required to get the maximum tensile strength of segmented polyurethanes. A recent work has shown that the thermal degradation of HTPB-based polyurethane and poly(urethane-urea) composites started in PU hard segments through depolycondensation reaction in the temperature range 200–350 °C as observed in TG and FTIR [a.433]. This involves dissociation of urethane and urea bonds as well as breakdown of allophanate and biurate linkages. Increase in crosslink density in poly(urethane-urea)s was associated with lower weight loss in the depolycondensation step. The activation energies associated with thermal degradation in different temperature ranges were calculated by the Coats–Redfern and Chatterjee–Conrad methods. Thermal characterisation of mica-filled thermoplastic polyurethane composites made it possible to observe that degradation started at a temperature of about 310 °C while those containing 5 wt% mica started to degrade at a later stage of 320–330 °C [a.271]. This was explained as being due to encapsulation of hard domains of the composite by mica and reordering of the hard segments in the presence of additive. However, the presence of additional hydrogen bonds between the hard segments of the thermoplastic polyurethane and mica was not ruled out. At higher concentration of mica, these additional bonds along with the shielding effect of mica raised the stability of the composites. It was also evident

162

Reinforced Polymer Nanocomposites that the rate of decomposition was much slower as the mica concentration increased. Also, the residue left after complete degradation of the composites increased as the mica content increased. Correa and co-workers [a.272] studied the thermal behaviour of short-fibre-reinforced PU composites by DSC and TG techniques and reported that the thermal resistance of aramidfibre-reinforced composites was greater than that of carbon-fibre-reinforced composites or the pure matrix polymer. The DTG results are presented in Figure 31.

Figure 31. Temperature dependence of the rate of weight loss for PU and its aramidfibre- (AF) (top) and carbon-fibre-reinforced (CF) (bottom) composites (at various percentages) obtained via an extrusion process Reprinted from [a.272] with permission from Elsevier

163

Thermal Degradation of Polymeric Materials Also, analysis of the kinetics and the glass transition temperature suggested greater interaction between aramid fibres and elastomer matrix. In particular, the degradation of PU or PU composites reinforced with aromatic PA or short carbon fibres followed first-order kinetics. The high promise for industrial applications of nanocomposites has motivated vigorous research, which has revealed concurrent dramatic enhancements of many materials properties by nanodispersion in polymeric systems – where the property enhancements originate from the nanocomposite structure, these improvements are generally applicable across a wide range of polymers [a.273]. Montmorillonite clay and polyhedral oligosilsesquioxane (POSS) additives have been added to the polyurethane in order to provide flame retardancy and thermal stability [a.274]. Results obtained with PU/clay and PU/POSS showed the great potential of using POSS for such applications. ive work has reported the Tg values of the segmented polyurethanes to be increased substantially (approximately threefold) in the presence of a small amount of tethered nano-sized layered silicates of montmorillonite compared with their pristine state [a.275]. Furthermore, the heat resistance and thermal stability of these PU/montmorillonite nanocomposites was also enhanced, as shown by TG. In particular, a 40 °C increase in the degradation onset temperature and a 14% increase in the degradation activation energy was found in polyurethane containing 1 wt% trihydroxyl group swelling agent-modified montmorillonite compared to that of the pristine polyurethanes. However, the variation of the crosslink density or crosslinking agents has a rather limited effect on the Tg of the resulting poly(urethane-ether) elastomers in the low-temperature region, as detected by the DSC method [a.276]. The resultant elastomers exhibited greatly enhanced thermal properties in comparison with those of the corresponding linear PU and analogous elastomers, which were crosslinked by 1,1,1-tris(hydroxymethyl)ethane. On the other hand, recent studies on PU nanocomposite foams have reported that the decomposition temperature decreases as the loading of SiC particles is increased to 3% [a.277]. These results were explained macroscopically as a simple colligative thermodynamic effect of an impurity on a bulk solution, which may be seen as the result of the perturbation that the SiC introduces into the three-dimensional structure of the polymer. This perturbation weakens the van der Waals interactions between the polymer chains, thus affecting the stability of the polymer, which was reflected in the lowering of the decomposition temperature. Correspondingly, in the case of TiO2 infusion, the improvement in thermal properties continued even up to 3% loading. This continued enhancement in thermal stability was related to the catalytic effect on the crosslinking of the polyurethane foam caused by the TiO2 nanoparticles. Wang and co-workers [a.278] reported that the thermal degradation properties of a chromophore were significantly enhanced due to intercalation into the layered aluminosilicate saponite, and also that the glass transition temperature of (chromophore)+–

164

Reinforced Polymer Nanocomposites saponite/PU nanocomposites proportionally increased with increased clay content. In two other ive reports [a.279, a.280], the addition of only a small amount of organo-clay was enough to improve the thermal (and mechanical) properties of PU nanocomposites. In addition, the clay intercalative route to nanocomposite synthesis also affected the thermal properties of the nanocomposites. POSS is known to interact with PS to promote intercalation and exfoliation. Along this line of interest, work on POSS–polystyrene/clay nanocomposites reported that the Tg value of the PS component in the nanocomposite was higher than that of pure PS and its thermal decomposition temperature was also significantly raised [a.281]. Hence, the presence of the POSS unit in the montmorillonite enhances the thermal stability of the polystyrene. In a comparative study, the POSS-intercalated clay was found to be relatively more thermally stable than the ammonium salt of cetylpyridinium chloride (C)-intercalated clay. For the latter, poor thermal decomposition and removal of surfactants were blamed for the observed weight losses of the C-intercalated clay–PS nanocomposites. It has also been shown that the introduction of POSS molecules chemically grafted to the polymeric chains of PS resulted in the formation of a nanocomposite material having enhanced mechanical performance, higher Tg and higher Tdec due to the absence of polar units in the POSS molecules used in the PS matrix. It was proposed that most of the enhancements were caused by the retardation of polymer chain mobility by the POSS molecules [a.282, a.283]. The homogeneity of ternary polymer hybrids has been found to be closely dependent on the degree of hydrogen-bonding interactions between each of the elements, and the hybrids were shown to have high solvent resistance and high thermal stability [a.284]. Parallel TG results of polyurethane–POSS nanocomposites displayed a broad weight loss beginning at 190 °C due the cleavage of urethane linkages, followed by evolution of amines and CO2 [a.285].

7.5 Polyamides with Natural Fibres Flame-retardant polyamide/cotton fibre blends have been degraded and a mechanism suggested for the interaction between blend components [a.286]. The decomposition data provided evidence for chemical interactions during degradation, which might be the reason for the anomalous degradation behaviour. It was noted that, if the thermal degradation of a non-flammable fibre takes place at a lower temperature compared with a flammable fibre, the volatile products formed from this fibre degradation in early degradation stages appeared to play an important role in retarding the flammability of a flammable fibre. Based on a later work, Fukatsu [a.287] has shown that the limiting oxygen index (LOI) values of aromatic polyamide and cotton fibre blends are significantly lower than the calculated values, and deviate from the average in a direction and to an extent that defy simple explanation. These blends caused changes also in the weight loss associated with

165

Thermal Degradation of Polymeric Materials the thermal degradation of the individual components from those calculated by averaging. The amount of cotton fibre required for the value of LOI that showed self-extinguishing behaviour in the vertical test was less than 30% for blends with aromatic polyamide fibre. When in these blends, the cotton starts to lose weight at lower temperatures. The volatile products from cotton were expected to play a role in some possible interactions occurring in this blend system, which appeared to for the difficulty in preparing the blends. The TG/DTG data (Figure 32) provided evidence for chemical interactions during thermal degradation, which was the reason for the residue weight during the first degradation stage, associated with the degradation of cotton fibre, being lower than predicted based upon calculations assuming no interactions.

Figure 32. TG and DTG curves for various aromatic polyamide/cotton fibre blends at a heating rate of 10 K/min in air atmosphere: (1) 100/0; (2) 30/70; (3) 50/50; (4) 70/30; (5) 0/100 aromatic polyamide/cotton fibre blends Reprinted from [a.287] with permission from Elsevier

166

Reinforced Polymer Nanocomposites This result suggested that cotton fibres in this blend accelerated the thermal degradation of aromatic polyamide fibres. On the other hand, compared to the calculated activation energies, an important decrease took place during the thermal degradation process, showing the formation of intermediate degradation structures with lower thermal stability.

7.6 Other Composites The thermal stability of polystyrene composites reinforced with short sisal fibres was found to be better than that of sisal fibre and the PS matrix [a.288]. The effects of fibre loading, length, orientation and modification on the dynamic mechanical properties of the composites were evaluated. Benzoylation, maleic anhydride coating of the polystyrene and acetylation of the fibres were fibre modifications and treatments that were carried out to improve fibre/matrix adhesion through specific interactions of the macrochains. The Tg values of the composites were lower than that of unreinforced PS, and this effect was attributed to the presence of some residual solvent in the composites entrapped during composite preparation. However, the composites with treated fibres showed better thermal degradation properties than those with untreated fibres. A study into the thermal degradation behaviour of PP/sisal composites with special reference to fibre content and fibre treatment has been conducted {887655}. It was found that, in the case of sisal fibres, most of the cellulose was decomposed at a temperature of 350 °C, whereas PP decomposed at a temperature of 400 °C. Further, it was observed that the thermal stability of the PP/sisal composites was higher as a result of better fibre/matrix adhesion. DSC investigations showed that the incorporation of sisal fibres in PP caused an apparent increase in the crystallisation temperature and percentage crystallinity. These effects were attributed to the fact that the surfaces of sisal fibres acted as nucleating sites for the crystallisation of the polymer, promoting the growth and formation of transcrystalline regions around the fibres. A study of the thermal degradation of linear low-density polyethylene (LLDPE)–wood fibres–ammonium polyphosphate (APP) composites reported that both wood fibres and APP influenced the thermal degradation behaviour of LLDPE and LLDPE–wood fibres composite, as illustrated in Figure 33 [a.289]. Wood fibres made the thermal degradation of LLDPE take place earlier {849787}, while APP stabilised LLDPE in LLDPE–wood fibres composite; these results were explained by free-radical stabilisation. APP decreased the initial temperature of thermal degradation, and promoted char formation of the composite. It was inferred that APP could catalyse esterification, dehydration and char formation of wood fibres. Scheme 33 illustrates the proposed thermal degradation mechanism of wood and LLDPE at high temperatures.

167


Figure 33. TG/DTG profiles of the LLDPE–wood fibre–APP composite system: (a) LLDPE–wood fibre–APP composite (experimental); (b) LLDPE–wood fibre composite; (c) APP; (d) LLDPE–wood fibre–APP composite (calculated) Reprinted from [a.289] with permission from Elsevier

TG studies performed by Rajeev and co-workers [a.290] on EPDM, maleated EPDM and nitrile rubber reinforced with melamine showed that the presence of melamine in the vulcanisates reduces the rate of decomposition, and the effect was pronounced in the presence of a dry bonding system consisting of resorcinol, hexamethylene tetramine and silica. Melamine fibres controlled the first degradation step of the vulcanisate, whereas the fibres as well as the matrix contributed to the second degradation step. An increase in fibre loading decreases the rate of degradation and weight loss in the second degradation step. The rate of decomposition of NBR vulcanisates is lower than those based on EPDM

168


Scheme 33. Thermal degradation of wood and LLDPE at high temperature Reprinted from [a.289] with permission from Elsevier

and maleated EPDM rubbers. The activation energy of decomposition of the vulcanisates was increased when the fibres were properly adhered to the matrix in the presence of the dry bonding system. The crosslinking system also affected the activation energy of decomposition, especially for the second degradation step. It was also reported that melamine fibres caused significant reduction in the thermal erosion rate of the vulcanisates. The fibre-filled composites, in the presence of the dry bonding system, displayed a lower thermal erosion rate compared to those containing no dry bonding system, showing that proper adhesion between the fibres and the matrix is important to achieve improved ablative properties. Among the three matrices, the vulcanisates based on nitrile rubber display the lowest thermal erosion rate. The electrical properties of some polymer composites – polyethylene/carbon black (PE/CB), polyethylene/carbon black modified by polypyrrole (PE/CB-PPy) and polyethylene/carbon black modified by polyaniline (PE/CB-PANI) – were investigated during thermal ageing brought about by slow cyclic heating and cooling. The conductivity in the composites was measured in heating/cooling cycles in the temperature range from 15 to 125 °C [a.291]. It was found that the thermal treatment resulted in a conductivity increase in the composites when heated below the melting point of PE. This effect was explained by increased crystallinity in the polymer matrix of thermally treated composites and confirmed by DSC analysis (Figure 34). Thermal ageing during heating above the melting point of the polymer matrix caused a decrease in the conductivity of PE/CB composites, but increase of conductivity in composites containing CB-PPy or CB-PANI as filler. The modified fillers created a more perfect and thermally resistant conducting network in the PE matrix. The decomposition temperatures of PE/CB-PPy and PE/CB-PANI composites were higher compared with that of the PE/CB composite as observed by TG.

169


Figure 34. Comparison of the DSC curves of (a) PE and (b) the PE/5% carbon black (CB) composite before and after thermal treatment to 100 °C Reprinted from [a.291] with permission from Elsevier

170

Reinforced Polymer Nanocomposites There is a high level of interest in using nanoscale reinforcing fillers for making polymeric nanocomposite materials with exceptional properties. Thermal studies on polypropylene containing 2 vol% multi-walled nanotubes (PP/MWNT) show that PP degrades with a large single peak starting around 300 °C in nitrogen. This large peak corresponds to the thermal degradation of PP initiated primarily by thermal scissions of C–C chain bonds accompanied by a transfer of hydrogen at the site of scission [a.292]. The results for the PP/MWNT also showed broad single peaks, but the temperatures at the DTG maximum peak rates were about 12 °C higher than that of PP. The amount of MWNTs in PP does not produce a significant enhancement in the thermal stability of this nanocomposite system in nitrogen for the temperature range investigated in this study. An increase in the temperature at the peak sample mass-loss rate has also been reported for the PP/PP-g-MA/clay (MA = maleic anhydride) system compared with PP/PP-g-MA {810866}. An increase of 17 °C with 10 wt% of clay in PP/PP-g-MA was reported. This effect was attributed to a barrier labyrinth effect of the clay platelets such that the diffusion of degradation products from the bulk of the polymer to the gas phase was slowed down. The temperature increase observed resulted from a similar barrier effect due to the hindered transport of degradation products caused by the numerous carbon tubes in the nanocomposites [a.293]. Another study showed that protective barriers are formed for a PA-6/clay nanocomposite during its thermal degradation, which slowed down its rate of degradation via a diffusion process [a.294]. According to the shapes of the degradation functions and of the kinetic laws, the coating formed by PA-6/clay nanocomposite was assumed to be more efficient than that formed by PA-6. This explained the improved fire properties of PA-6/clay nanocomposite compared with PA-6. The formation of a protective barrier in the case of PA-6/clay nanocomposite in fire conditions may also correspond to a phase change of the nanocomposite, from a delaminated structure to an intercalated structure. Thus, this phase change enables an improved slowdown of the escape of fuels. Other studies have investigated the decomposition of PMMA utilising TG and calorimetry. One study concentrated on polymer layered silicate nanocomposites {825036} and compared the degradation profiles of PMMA-filled nanocomposites to that of pure PMMA by using TG-DSC-FTIR and GC-MS. The results based on TG and DSC indicated enhanced thermal stability and higher glass transition temperature of filled PMMA nanocomposites with respect to that of pure PMMA. Nonetheless, in both cases the decomposition was described as a two-step reaction. Blumstein [a.295] showed that free-radical-polymerised PMMA inserted between the lamellae of montmorillonite clay (d-spacing increase of 0.76 nm) resisted thermal degradation under conditions that would otherwise completely degrade pure PMMA (refluxing decane, 215 °C, N2, 48 h). TG revealed that both linear PMMA and crosslinked PMMA intercalated into Na+ montmorillonite have a 40–50 °C higher decomposition temperature. Improvements in thermal stability similar to that reported by Blumstein for

171

Thermal Degradation of Polymeric Materials both PDMS and polyimide nanocomposites were also observed. In the case of PDMS, the nanocomposite was prepared by melt intercalation of silanol-terminated PDMS into dimethyl ditallow ammonium bromide to allow ammonium-treated montmorillonite to be properly arranged [a.296]. Despite the low clay content, the disordered–delaminated nanostructure showed an increase of 140 °C in decomposition temperature compared to the pure PDMS elastomer. Gilman and co-workers [a.297] reported a reduction in the peak of the heat release rate by 50–75% for PA-6, PS and poly(propylene-graft-maleic anhydride) nanocomposites. The experiment also found that the type of layered silicate, level of dispersion and processing degradation had an influence on the magnitude of the flammability reduction. In another development, Singh and Haghighat [a.298] made new organic/inorganic nanocomposite structures by substituting high temperature organic phosphonium cations for the standard compatibilising agent – alkyl ammonium cations. The thermal extension range was based on the innovative use of organically modified layered aluminosilicates that combined the layered silicate and the organic surfactant/compatibilising agent in a single chemical compound. The organic surfactant groups were bonded to the structural Si atom through thermally stable Si–C bonds, thus enhancing the thermal degradation properties of the overall system. Therefore, those materials provide unique inorganic layered silicate reinforcements having markedly more thermally stable surfactants ‘builtin’ to the chemical structure.

172

Inorganic Polymers

8

Inorganic Polymers

Inorganic polymers are macromolecular substances whose principal structural features are made up of homopolar interlinkages between multivalent elements other than carbon. Inorganic polymers do not preclude the presence of carbon-containing groups in side branches, or as interlinkages between principal structural , and are mainly found in Nature, e.g., mica, clays and talc. Polysiloxanes, polyphosphazenes, polysilazanes, polygermanes and polystannanes are the most important classes of inorganic polymers from the applications point of view. High-molecular-weight polymers with inorganic elements in their backbone are attractive and challenging, because of their physical and chemical differences from their organic counterparts. These polymers offer a unique combination of high-temperature stability and excellent low-temperature elastomeric properties. In the following sections, recent developments in the thermal degradation of polysiloxanes, polyphosphazenes, polysilazanes, polysilanes and organic–inorganic hybrid polymers are presented.

8.1 Polysiloxanes Polysiloxanes are the most common and one of the most important inorganic polymers used in polymer chemistry. The polysiloxanes are known for their useful properties, such as flexibility, high permeability to gases, low glass transition temperature and low surface energy. With such crucial properties, polysiloxanes are widely used in many applications; for example, the medical applications include prostheses, artificial organs, facial reconstruction, catheters, artificial skin, lenses and drug delivery systems, while the non-medical applications include high-performance elastomers, membranes, electrical insulators, water repellants, anti-foaming agents, mould release agents, adhesives, protective coatings, release control agents for agricultural chemicals, and hydraulic, heattransfer and dielectric fluids. Dai and co-workers [a.302] have recently proposed the thermal degradation mechanism of polysiloxanes shown in Scheme 34. The superior thermal stability of polysiloxanes has made them attractive candidates for use at elevated temperatures. The most common member of the family, poly(dimethylsiloxane)

173


Scheme 34. The pyrolysis pathway of a liquid-crystalline polysiloxane Reprinted from [a.302] with permission from Elsevier

(PDMS), has been shown to be thermally stable to 300 °C under vacuum [a.299]. The incorporation of methylphenyl- or diphenylsiloxane as a comonomer with PDMS has been shown to increase the onset temperature of degradation to nearly 400 °C. Because of their thermal stability, poly(dimethyldiphenylsiloxane)s (PDMDPS) have been considered for application as adhesives for high-temperature service, packing for chromatographic columns and lubricants. Others have proposed that these polysiloxanes may be used as vaporisable components for the production of either microporous organic/inorganic materials [a.300, a.301] or non-porous ceramics. To fully succeed in these and other high-temperature applications, the main thermal degradation issues must be understood – in parallel it should be noted that factors other than the chemical nature of the polymer backbone might influence the thermal degradation process. Polysiloxanes are reported to be substantially more stable than PVAc, such that increasing the quantity of polysiloxane in the blends with PVAc causes a gradual increase in the stability. On the other hand, blends of PVAc containing 50% and more PDMS and PDMDPS concentrations were reported to be more stable than the pure siloxane [a.303]. This behaviour was explained on the basis of crosslinking induced by free radicals, e.g., acetate radicals, diffusing from the PVAc phase, which are responsible for abstraction of hydrogen atoms from the methyl groups in PDMS and PDMDPS. The macroradicals thus formed then undergo further crosslinking reactions. Crosslinking cannot be induced in

174

Inorganic Polymers PDPS blends because hydrogen abstraction is not possible due to the absence of methyl groups in the chains. The thermal degradation of PDMDPS results in the evolution of free benzene and volatile cyclic siloxanes [a.299]. The predominant cyclic compounds contain three or four –Si–O– bonds, although higher-molecular-weight structures have been observed. While it is clear that multiple reactions occurred, the calculated activation energies of the polysiloxanes decreased with the incorporation of phenyl groups and were much less than the values reported for degradation of vinyl-terminated PDMS. The reduction in activation energy was most likely due to the incorporation of additives in the starting polymer material. The pre-exponential factor decreased upon incorporation of phenyl groups; this effect was believed to be due to a reduction in the polymer chain mobility with the incorporation of phenyl groups. Visser and co-workers have shown that cyclic stress also acts to accelerate the degradation of polysiloxanes [a.304]. The chemical nature of the end-cap has also been shown to enhance the degradation reactions for phenyl-containing siloxanes. In principle, terminal hydroxyl groups can participate in a ‘backbiting’ reaction through which benzene is liberated and Si–O chain branches are formed. The presence of other external factors such as acidic or basic impurities, oxygen, water, fillers and residual catalyst can influence the rate of thermal degradation [a.301, a.305, a.308]. In general, these contaminants have been shown to increase the rate of degradation. However, some additives have been shown to stabilise the siloxane structures against depolymerisation. Another work has shown that zirconium and caesium octoates stabilise PDMS and phenyl-containing siloxanes against thermal degradation [a.305]. It was proposed that the octoates act to promote bond rearrangement and crosslinking, which effectively arrests the degradation process. According to a recent study, hydroxyl-terminated PDMS depolymerises predominantly from chain ends at moderate temperatures with a low activation energy [a.306]. The decomposition products of vinyl-terminated PDMS at 360 °C were principally the cyclic oligomers, hexamethyltrisiloxane and octamethyltetrasiloxane, which is consistent with the other reported nucleophilic substitution reaction mechanisms of degradation. Conversely, the degradation of PDMS copolymer also resulted in the evolution of benzene in the initial stages of the reaction where no cyclic oligomers with phenyl substituents were observed [a.307]. NMR analysis of the pyrolysed phenyl-containing polysiloxane copolymers indicated the reaction mechanism for generation of benzene to be thermally induced random free-radical reaction. Prior studies on the thermal degradation of polysiloxanes have focused on the volatile products evolved and the temperatures at which decomposition occurred [a.308]. Camino and co-workers [a.309] combined kinetic formal treatments and computer simulations to analyse the thermal degradation of PDMS. It was shown that PDMS 175

Thermal Degradation of Polymeric Materials thermally decomposes to cyclic oligomers through Si–O bond scission in a chain-folded cyclic conformation energetically favoured by overlapping of empty silicon d orbitals with the orbitals of the oxygen and carbon atoms. Kinetic analysis showed that PDMS thermal volatilisation, as rate of heating increased, became dominated by rate of diffusion and evaporation of oligomers produced on its decomposition. At 100 °C/min the thermal degradation behaviour of PDMS is clearly modified, since the weight loss in nitrogen was by then a result of two overlapping processes. A small black residue was formed (silicon oxycarbide), which was produced by an alternative decomposition pathway leading to cyclic oligomers, made possible at high heating rate. The TG and DTG results of PDMS are displayed in Figure 35. Very little attention has been paid to the residual product remaining after the decomposition process of polysiloxanes. Notably exceptional results have shown that controlled decomposition of thin sheets of crosslinked PDMS results in a nanoporous solid product with increased mechanical strength and reduced susceptibility to organic solvents and temperature [a.308]. Elsewhere, the products of thermal decomposition of the alternating copolymers were predominantly cyclic oligomers containing both diphenylsiloxane and dimethylsiloxane units [a.101]. Most cyclic compounds had either three or four siloxane units. In the case of the random copolymers, cyclic oligomers with a degree of polymerisation of three or four remained the dominant evolution products. Nevertheless, the composition

Figure 35. TG and DTG curves of PDMS in nitrogen (solid lines) and in air (dotted lines) at a heating rate of 1 °C/min Reprinted from [a.309] with permission from Elsevier

176

Inorganic Polymers of the cyclic oligomers was much more widely varied than in the alternating copolymer case. Mixtures of dimethyl- and diphenylsiloxanes were observed in the decomposition gases of every random copolymer. Cyclic compounds with only dimethylsiloxane moieties were also observed, as were cyclic structures with only diphenylsiloxane units. Benzene was also noted as a significant product from the thermal degradation of diphenylcontaining siloxane copolymers [a.101]. Below 300 °C, the quantity of benzene evolved was less than the molar concentration of terminal hydroxyl units. At higher temperatures, a significant increase in benzene evolution was observed. Ultimately, the quantities of benzene produced exceed that which could be explained by the terminal hydroxyl reaction by factors of 10–100 depending on the process [a.307, a.310]. A free-radical sequence originally proposed by Sobolevskii and co-workers [a.310] has been widely accepted to explain this additional benzene evolution. Another team of researchers has demonstrated that the random copolymers are more thermally stable than block copolymers and that the thermal stability increased with the addition of diphenyl units up to about 20 mol% – however, further increases in the diphenyl structures content had no significant impact on stability [a.311].

8.2 Polyphosphazenes Polyphosphazenes are well-developed inorganic–organic hybrid materials, consisting of alternating phosphorus and nitrogen atoms in the polymer backbone. The polyphosphazenes are an important class of inorganic macromolecules, which have aroused great interest in modern polymer science from both basic and practical points of view. Fluorinated phosphazene polymers and copolymers are perhaps the most important class of phosphazene macromolecules synthesised in the past few decades [a.59]. To cite but a few, Allcock’s [a.312] thermal studies on poly(bis(trifluoroethoxy)phosphazene) demonstrated that depolymerisation occurs at temperatures above 150 °C, while at high temperature (450 °C) the polymer is completely volatilised. Gleria and co-workers [a.313] have lately shown that fluorinated polyphosphazene-g-polystyrene grafted materials thermally decompose in the temperature range between 300 and 450 °C. Two main classes of fluorinated polyphosphazenes are presently available and can be categorised into two groups. The first group contains phosphazene homopolymers, in which the fluorine atoms may be directly linked to the phosphorus atoms of the inorganic , attached to the skeletal phosphorus through an aliphatic and/or an aromatic structure. The other group is of phosphazene copolymers, in which two (or more) fluorinated substituents are attached to the phosphazene backbone. Two or more different fluorinated macromolecules are linked together to form linear, block copolymers and in some cases one of the block-forming polymers may be organic in nature, or may

177

Thermal Degradation of Polymeric Materials be formed by silicone macromolecules. Also, the fluorinated phosphazene material acts as the main substrate onto which organic macromolecules are grafted according to a variety of different procedures. The thermal degradation of poly(2,2-dioxybiphenylphosphazene) at temperatures from 100 to 200 °C during periods of time up to 250 h is reported to take place with neither loss of mass nor a noticeable change in the chemical structure of the polymer [a.314]. In the interval 200–250 °C, random cleavage of the polymeric chains gives lower-molecularweight polymers. Above 250 °C, depolymerisation to cyclic oligomers occurs, and finally, above 400 °C, complex intermolecular coupling reactions take place leading to pyrolytic black residues – thus establishing that thermal decomposition of polyaryloxyphosphazenes occurs in three steps. Molecular dynamics simulations performed on systems containing segments of poly(2,2-dioxybiphenylphosphazene) indicate that there are three allowed conformations – one trans and the two gauche, with a strong preference for trans. The chains exhibit a distorted helical structure. Characteristic ratios computed in THF solutions are smaller than those obtained in bulk, which confirmed that the THF solution is below theta conditions. Thermal degradation of methyl methacrylate polymers functionalised with phosphoruscontaining molecules is reported to display very different degradation behaviour [a.315]. The first step corresponds to the degradation of the PMMA-related parts (production of MMA), and then the degradation of the functionalised PMMA structures occurs at higher temperatures than for its homopolymer counterpart. It has been shown that functionalisation introduces high-temperature degradation stages (above 750 °C) and weak points in the polymeric chain lead to a complete cracking of the polymer, as no monomer molecules were found during the degradation, in contrast to PMMA. Results from this work revealed a major and common degradation stage near 300 °C leading to diethyl phosphite HPO(OC2H5)2 or diethyl phosphonate RPO(OC2H5)2, RCHO, CO, CO2 and C2H4. However, the thermal degradation of this copolymer did differ very much in air and argon atmospheres [a.315]. For the polymers –(CH2CH(ON3P3Cl5))n– and –(CH2CH(ON3P2SOPhCl3))n– it has been shown that at relatively low temperatures crosslinking occurs between the cyclophosphazene ligand and the hydrocarbon main chain, with elimination of HCl [a.316]. At higher temperatures, elimination of (N3P3Cl5)2O was observed for the polymer –(CH2CH(ON3P3Cl5))n–. The thermal behaviour of the homopolymers of gem-methyl(vi nylbenzyl)tetrachlorocyclotriphosphazene (N3P3Cl4(Me)(CH2C6H4CH=CH2)) (STP) and its fully amino-substituted derivative (N3P3(NMe2)4(Me)(CH2C6H4CH=CH2)) (STPN), of copolymers of STP with PMMA and styrene, and of copolymers of gem-isopropyl-2-(acetoxyvinyl)tetrachlorocyclotriphosphazene (N3P3Cl4(iPr)(C(OC(O)Me)=CH2)) (VA) with PMMA and styrene has been reported [a.317, a.318]. Upon heating under TG conditions, the highest char yield (64 wt%) was found for the homopolymer of STP.

178

Inorganic Polymers The char yields for the copolymers appear to increase with increasing amounts of phosphazene incorporated [a.318]. The one-step weight losses observed for the homopolymer of STP were mainly ascribed to elimination of HCl. The STP-styrene copolymers decomposed in one step, indicating that HCl elimination, ring degradation and depolymerisation took place simultaneously. The STP-PMMA copolymers showed two-step degradation. From XPS data, complete loss of chlorine took place in the first step and probably in combination with some depolymerisation of PMMA units. In the second step, phosphazene ring degradation was observed, accompanied by further carbonisation. The VA-styrene copolymers started to decompose about 100 °C lower than the STPPMMA copolymers, exhibiting also a two-step TGA curve. The first step was associated with breakdown of polymer chains at the C–C linkage between inorganic monomers. In the second step, depolymerisation of the styrene sequences, HCl elimination and ring degradation occurred. PP-styrene copolymers also lost weight in a two-step process, but polymers with the same composition as the VA-styrene polymers give lower char yields. All polymers showed an enhanced flame retardancy [a.317]. Polydichlorophosphazene with pendant aniline dimer groups showed good thermal degradation properties, which were thought to be due to the thermal stability of the organic molecules that were incorporated into polyphosphazene as side groups of the polymer and aniline dimer. Aniline dimer started to decompose at 140 °C and the polymer at 175 °C [a.319]. The Td increased by 40 °C due to the influence of the polydichlorophosphazene main chain. In addition, the polymer decomposed more slowly than the dimer, thus indicating that the polymer possessed better thermal stability. Figure 36 shows the TG profiles of polydichlorophosphazene with pendant aniline dimer groups.

Figure 36. TG profile of polydichlorophosphazene with pendant aniline dimer groups, for the polymer (continuous line) and the dimer (dotted line) Reprinted from [a.319] with permission from Elsevier

179

Thermal Degradation of Polymeric Materials Ma and co-workers [a.320] claimed that, since 2-thiazolidinethione possessed good thermal degradation properties, it might have contributed to the thermal stability of the organic molecules that were introduced into polyphosphazene as side groups to form poly(organophosphazene). The TG results showed that the ligand started to decompose at 110 °C and the polymer at 145 °C. The Td increased by 35 °C while, additionally, the polymer decomposed more slowly than the ligand, indicating that poly(organophosphazene) possesses superior thermal properties to the polyphosphazene random and block copolymers found in the literature [a.321].

8.3 Polysilazanes and Polysilanes Polysilazanes have been shown to be excellent polymeric precursors to amorphous silicon carbonitride (SiCN), silicon nitride, silicon carbide (SiC) and their composites. The actual chemical and phase compositions of the ceramic products depend on the polymer composition and pyrolysis conditions, such as temperature, time and atmosphere. Polymeric silazanes consist of amorphous networks, which transform to amorphous SiCN ceramics by pyrolysis under inert atmosphere at around 1000 °C. These ceramic products remain amorphous up to 1400 °C in an inert atmosphere [a.322]. However, at higher temperatures the non-stoichiometric SiCN matrix decomposes, with nitrogen loss, giving the thermodynamically stable phases, namely Si3N4 and SiC. Polysilanes, polycarbosilanes and polysilazanes are commonly used for the preparation of high-performance ceramics such as silicon carbide, silicon nitride and silicon carbonitride. Zhu and co-workers [a.140] studied the thermal degradation of the silazane polymers that are often used to prepare SiCN ceramic materials. The results showed that there is no significant weight loss below 200 °C, and three stages of degradation occurred in the course of the thermal decomposition of silazane polymers. The first stage occurred with a weight loss in the range <12 wt%. Although TG curves indicated that a thermal lag took place in this stage, the initial degradation temperatures at different heating rates were the same. In the third stage, which occurred in the temperature range of 525–850 °C, different yields of pyrolytic products were observed and they increased with decreasing heating rate. There was a second stage between the first and third stages, thus suggesting that the mechanism of thermal degradation of silazane polymers is not the same at different heating rates. On increasing the heating rate, the size of the DTG peak below 300 °C decreased while the peak above 300 °C increased. The activation energies of the first degradation stage were calculated using the Flynn–Wall method and are displayed in Figure 37, which clearly shows the change of activation energies of silazane polymers during degradation, suggesting that thermal decomposition is a very complex process. Polysilanes have a continuous backbone of silicon atoms. Characteristics include lower solubility, higher glass transition temperatures and, perhaps most importantly, unique 180

Inorganic Polymers

Figure 37. Activation energy of the silazane polymer degradation process Reprinted from [a.140] with permission from Elsevier

optical properties, including long-wavelength ultraviolet absorption, which intensifies as the degree of polymerisation increases. This is associated with delocalisation of silicon– silicon bonding and other orbitals. In studies on poly(vinyl triethoxysilane) (PVTES), the TG curves showed that PVTES starts to lose weight at 310 °C [a.185]. However, in an O2 atmosphere this temperature is significantly decreased. PVTES degrades in a single step in an N2 atmosphere, so confirming the important effect of reaction atmosphere on the mechanism of thermal and oxidative degradation of PVTES. The TG and DTG curves of VTES-MMA copolymers were observed to be similar to that of PMMA. However, the initial decomposition temperature of the copolymers was decreased in both N2 and O2 atmospheres with increasing VTES content in the copolymer. This behaviour was attributed to the decrease in the molecular weight of these copolymers. The activation energies of the degradation reactions of PMMA were found to be 192 and 115 kJ/mol and that of PVTES to be 108 and 98 kJ/mol in N2 and O2 atmospheres, respectively, as shown in Table 7. However, the activation energy of the thermal degradation reaction of copolymers in an N2 atmosphere was determined to be in the range of 180–105 kJ/mol and that of the same process in an O2 atmosphere in the range of 100–154 kJ/mol. The activation energies of the degradation reactions of the copolymers were decreased in both N2 and O2 atmospheres 181


Table 7. GPC results of VTES copolymers synthesised at various feed monomer concentrations and the activation energy of the thermal degradation reactions and oxidative degradation reactions of these copolymers Reprinted from [a.185] with permission from Elsevier Mole fraction of VTES in feed 0.18 0.34 0.43 0.54 0.82

Mole fraction of VTES in copolymer 0.02 0.06 0.07 0.08 0.11

Mw × 10–3

427 217 210 164 90

Mn × 10–3 Polydispersity Thermal Oxidative (Mw/Mn) Ea (kJ/mol) Ea (kJ/mol)

116 30 29 15 8

3.7 7.1 7.2 10.9 11.3

179 138 136 108 105

101 81 80 66 54

with increasing VTES content. The decrease in the thermal stabilities of the copolymers was attributed to a decrease in the activation energy for degradation. The thermal degradation behaviour of the hydroborated copolymer prepared from 2,4,6trimethyl-2,4,6-trivinylcyclotrisilazane and borazine (B3N3H6) has been investigated by TG and DTA [a.323]. The polymeric products exhibited a high ceramic yield of 75% at 1000 °C with a major weight loss at 400–700 °C, compared to 55% reported for SiCBN polymers [a.324]. In addition, mass loss at 70–200 °C was observed and related to the vaporisation of low-molecular-weight polymers, whereas a sample crosslinked at 200 °C showed almost no weight loss over the same region. The significant weight loss in the range 400–700 °C suggested that the hydroborated copolymer was decomposed or rearranged to transform the polymeric structure to the ceramic phase, which involved the thermal decomposition of the aliphatic bridges or methylene groups, as observed by 13 C and 29Si MAS-NMR. Moreover, as detected by DTA, the broad exotherm over the temperature range 100–600 °C was attributed to crosslinking hydroboration and pyrolytic decomposition. In the temperature range 700–1400 °C, the ceramic phase produced became thermally stable with only slight weight loss, and a broad endothermic peak at 1150 °C was observed, presumably due to the redistribution reactions, which led to the amorphous SiCBN phase – this was compared with the redistribution reaction of silicon carbonitride glass, which is observed at the lower temperature of 980 °C. Kwet Yive and co-workers [a.325] reported that for the oligomeric vinylsilazanes the Si loss value was low (0.5%) due to enhanced crosslinking compared with other polysilazanes, e.g., oligomeric methylsilazane, which loses almost half of its Si content after pyrolysis. The thermal degradation was found to be a three-step process. The first weight loss occurred between 150 and 220 °C and was caused by the distillation of light oligomers;

182

Inorganic Polymers the second one between 220 and 400 °C was associated with dehydrogenation and transamination; and the third one above 400 °C was caused mainly by the evolution of methane and hydrogen gases. A similar three-stage weight loss was found in other kinds of polysilazanes [a.326], implying that this may be the general pyrolysis behaviour for polysilazanes. Analysis of the Si loss values suggested that the first weight-loss step is the most important. Thus, the most significant effects for raising ceramic yields are gained by preventing the evolution of Si-containing oligomeric products at lower temperatures. On the other hand, enhanced weight loss at higher temperatures, which correspond to the third stage, was related to the formation of hydrocarbon gases and other non-siliconcontaining compounds. A study of the pyrolysis of polysilazane by thermogravimetry–evolved gas analysis reported a three-step process, with the first displaying very small mass loss (0.4%) up to 350 °C as shown in Figure 38 – this was ascribed to evolution of traces of ammonia, which were detected in the temperature range from 250 to 350 °C [a.327]. The release of ammonia indicated the existence of transamination reactions corresponding to further crosslinking of the polysilazane precursor. The major mass loss (17.6%) occurred between 350 and 800 °C. This mass loss corresponds to the escape of hydrocarbons and traces of silicon species in the temperature range from 350 to 650 °C. The masses of the

Figure 38. TG-EGA: mass change and ion current intensities of m/z = 2, 17 and 18 of the thermal decomposition of polysilazane Reprinted from [a.327] with permission from Elsevier

183

Thermal Degradation of Polymeric Materials detected ions suggest the occurrence of ethene, propene and differently substituted silanes. However, at 450 °C the evolution of the main gaseous products (hydrogen and methane) was observed, while above 800 °C an additional small mass loss occurred, corresponding to the release of hydrogen and traces of water. TGA performed under an argon atmosphere showed that unbranched silazane gave a very low ceramic yield <11 wt% at 1000 °C, while branched silazanes showed a higher ceramic yield in the range of 50–75 wt%. The lower yield for the unbranched polysilazanes was related to the evaporation of oligomers at T < 300 °C. The weight loss at this stage was reduced from >80 wt% for the unbranched precursor to 30 wt% for the branched one, with the lowest degree of branching in the group of the branched precursors. For the branched precursor, the weight loss decreases with decreasing degree of branching, and the major difference between them is the evaporation of oligomers around 300 °C [a.328].

8.4 Organic–Inorganic Hybrid Polymers Organic–inorganic hybrid ceramers fabricated through sol–gel methods have become a new type of material, and have attracted much interest in the past decade [a.330] {760942}. These materials combine the advantages of both organic and inorganic materials and are expected to possess new properties that individual organic or inorganic materials could not achieve {870944} {768933} {757341}. From the interactions between organic and inorganic phases, ceramers can be divided into two major classes. In the first, the inorganic precursors directly bond to the organic polymers and covalent bonds are formed between them. The most common system in this class is the PDMS/TEOS system. In the second class, only physical interactions such as hydrogen bonding or interpenetrating networks may be formed between the organic and inorganic phases. Numerous studies have been performed on the preparation and thermal characteristics of organic–inorganic polymers, e.g., POSS–polystyrene hybrid polymers [a.329] and olefin-substituted cyclophosphazenes [a.315–a.318]. These hybrid polymers are of interest because reaction with various nucleophiles can easily change their properties. In this way it is possible to meet the demands necessary for use in a wide variety of applications. Of these properties, their enhanced flame retardancy due to high char yields is the most important. The thermal degradation of the phenolic resin/silica hybrid ceramers with different component ratios has been investigated [a.330]. Pure phenolic resin and hybrid ceramers showed two stages in the kinetic scheme of thermal degradation evaluated by the Kissinger method. The activation energies of thermal degradation of ceramers in the first stage (117–155 kJ/mol) were lower than that of phenolic resin (200 kJ/mol). This was associated with the emission of water and alcohol, which causes extra weight loss in the ceramer system. From FTIR studies, the main degradation mechanism was not affected

184

Inorganic Polymers by the existence of inorganic silica, but it revealed that the formation of a silica network structure was restricted by intermolecular hydrogen bonding at lower temperature. However, the restrictions disappeared when the phenolic resin started to decompose at higher temperature. TG-MS results showed that the amount of formaldehyde liberated was increased as a result of oxidation of the alcohol in the ceramer system. Furthermore, changes in distribution and amount of CO2 and phenol were observed and an unknown fragment was detected in the ceramer system. For PMMA/silica hybrid materials, an obvious difference was found between untreated and heat-treated hybrid materials [a.331]. The pre-degradation stage of all the untreated organic–inorganic hybrid materials started at 150 °C and ended at 260 °C. It was close to the temperature of maximum degradation rate of untreated silica. Subsequently, major degradation occurred from 300 to 460 °C with extensive weight loss. However, following 180 °C heat treatment, hybrid materials did not exhibit two-step degradation but only a single degradation stage from 300 to 460 °C, according to TG measurements. Hence, an inference was drawn that further condensation of the hybrid materials with heat treatment proceeded in the 180 °C oven, and showed only a single degradation stage instead of twostep degradation as previously reported by Yano and co-workers [a.332]. The application field of porous ceramics has recently expanded to thermal and acoustic insulations, kiln furniture, catalyst s, and hot gas, water or molten metal filters. Since the distribution of size, shape and volume of the pore space in porous ceramics directly relates to their ability to perform a desired function in a particular application, the need to establish uniformity of the cell parameters in order to achieve superior properties has been strongly emphasised. During pyrolysis in an inert atmosphere, the polymer-to-ceramic transformation of the pre-ceramic occurs, yielding a silicon carbide for polycarbosilane (PCS) or a silicon oxycarbide ceramic for polysiloxane. Avoiding the collapse of pore structure is a central issue during both the pyrolysis and sintering processes. Silicon oxycarbide (SiOC) ceramic foams, produced by the pyrolysis of a foamed blend of a methylsilicone pre-ceramic polymer and polyurethane (PU), exhibit excellent thermal and mechanical properties. They also show high surface area, high permeability, low specific heat and low thermal conductivity which are of primary importance for a number of technological applications [a.328, a.329]. The reported TG curves shown in Figure 39 indicate that the conversion process happens in two separate steps [a.300]. In the first step, decomposition of PU and volatilisation of silicone oligomers occur, while in the second one the organic bonds in the silicone resin are broken, with release of methane (ceramisation). The data showed that PU decomposition leaves a carbonaceous residue (about 8 wt%) within the SiOC material, while the ceramic yield of a pure trimethylolpropane trimethacrylate material is about 84 wt% at 1200 °C. Calculated

185


Figure 39. TG curves of unpyrolysed trimethylolpropane trimethacrylate (SR 350) (crosslinked), foamed semi-rigid polyurethane, and unpyrolysed trimethylolpropane trimethacrylate/polyurethane foamed blends (SR 350/PU weight ratio = 5.25/1, low-PU; and 1/1, high-PU) under nitrogen Reprinted from [a.300] with permission from Elsevier

residual carbon from PU after pyrolysis in the low-PU and high-PU foams (5.25:1 and 1:1 weight ratio of ceramic : PU, respectively) was about 1 and 4 wt%, respectively. Analysis of the total carbon content of pyrolysed materials using an oxidation accelerator gave 14, 17 and 30 wt% for a bulk SiOC ceramic, a low-PU and a high-PU SiOC foam, respectively. Linear shrinkage upon pyrolysis, measured using a dilatometer, showed a contraction of about 35% (at 1200 °C) for high-PU foam, while low-PU foam at that temperature shrunk about 27%. The maximum dimensional changes occur in two steps (300–400 and 600–800 °C range); in accordance with the TGA findings, the first interval was related to PU decomposition while the second one derives from the ceramisation of the silicone resin. SEM micrographs illustrating the pyrolysed PU/ceramic foam are presented in Figure 40 [a.300]. Thermal studies of monofunctional epoxy-substituted POSS monomer that was incorporated into a network composed of two difunctional epoxy monomers, the

186

Inorganic Polymers

(a)

(b)

10 μm

10 μm

Figure 40. (a) Topographic SEM image of an unpyrolysed low-PU foam (SR 350/PU = 5.25/1 weight ratio). The cross-section of a strut is shown. (b) Compositional SEM image of an unpyrolysed low-PU foam (SR 350/PU = 5.25/1 weight ratio) Reprinted from [a.300] with permission from Elsevier

diglycidyl ether of bisphenol-A (DGEBA) and 1,4-butanediol diglycidyl ether (BDGE), at a DGEBA:BDGE 9:1 mole ratio, were performed [a.333]. The glass transition region was observed by DSC to broaden with an increase in weight percentage of the POSS (10 wt%), but there was no change in the onset temperature of the glass transition. The topological constraints provided by the presence of POSS reinforcements slowed the motion of the network junctions. Therefore, the time needed to reach structural equilibrium increased substantially relative to that for non-nanoreinforced networks. Elsewhere [a.334], the thermal stability of the eightfold alkyl-substituted silsesquioxanes was reported to vary with n value. It was shown that when n = 1 the silsesquioxane sublimated at ambient temperature, while when n = 2 the silsesquioxane had a relatively high melting point (212 °C). For n > 3 the derivatives showed a clear ‘odd–even’ effect, with the compounds with an odd number of carbon atoms exhibiting generally lower melting point. TG results under N2 conditions showed that the onset of the weight loss was found to shift to higher temperatures with increasing alkyl chain length. Although in air the onset of the decomposition was found at lower temperatures than in an N2 atmosphere, the total weight loss was lower than under N2 – a phenomenon that was attributed to the formation of a crosslinked silicate network. Thermal analysis of POSS-containing triblock copolymers indicated the presence of two clear glass transitions in the microphase-separated system, with strong physical ageing observed in samples annealed at temperatures near the Tg of the poly(3-(3,5,7,9,11,13,15heptaisobutyl-pentacyclo(9.5.1.13,9.15,15.17,13)-octasiloxane-1-yl)propyl methacrylate),

187

Thermal Degradation of Polymeric Materials P(MA-POSS) [a.335]. WAXS results indicated that rearrangement of POSS moieties observed in glassy domains further ed the occurrence of physical ageing. It was found that the Tg of the P(MA-POSS) phase from triblock copolymers sequestered in microphase-separated domains was nearly 25 °C higher than that of a P(MA-POSS) homopolymer of comparable molecular weight, therefore suggesting a strong confinementbased enhancement of Tg in this system. Similar reports based on DSC studies of cyclopentyl- and cyclohexyl-substituted POSS homopolymers reported the onset of decomposition occurring before the observation of the Tg [a.336]. This phenomenon was ascribed to retardation of segmental motion of polymer chains due to the presence of bulky side-chain groups at each repeat unit of the backbone. In addition, another study [a.337] on diblock copolymers of polynorbornene and polynorbornene-POSS showed that increasing the length of the polynorbornene-POSS block had no effect on the Tg of the polynorbornene-rich phase, with Tg of 55 °C, while the morphologies traversed the usual sequence of spheres–cylinders–lamellae. Other findings on POSS-methacrylate copolymers have shown that high-molecular-weight POSS-containing homopolymers (MW = 200 000) do not reveal a glass transition below thermal decomposition around 400 °C. A different study revealed an analogous sensitivity of polyethylene oxide (PEO) crystal stability on the matrix Tg of PEO-POSS cylindrical diblocks and reported a similarly large Tg difference in ageing behaviour [a.338]. However, in both studies, no definite explanations were given. The aniline groups of octa(aminophenyl)silsesquioxane (OAPS) offer versatility both as reaction sites to which other nano-building blocks can be added and as starting points for generating other functional groups, thereby providing access to diverse and novel nanocomposites [a.339]. A key problem with almost all of the materials explored to date is that the aliphatic components limit the thermal stability of the resulting nanocomposites, strongly influence (lower) Tg and decrease the potential mechanical properties. However, OAPS nanocomposites when reacted with diepoxides or dianhydrides were proposed to provide high-crosslink-density materials with good thermal stability, and good to excellent tensile and compressive strengths [a.339]. Reacting OAPS with pyromellitic anhydride (PMA) at T > 300 °C provided complete curing and a material that exhibited a 5% mass loss at a temperature of 540 °C (air and N2) and 75 wt% char yield at >1000 °C in N2.

188

High Temperature-Resistant Polymers

9


In the past few decades the demands of the household, automobile, construction and especially aerospace industries have provided the driving force for the development of new high-performance polymeric materials to be used in structural applications {755543}. The main goal of research in this area is the preparation of polymers possessing good thermal and oxidative stability, toughness, stiffness and retention of physical properties at high temperature. Most materials of this type are based on structures containing aromatic rings incorporated into the main chain – for such polymer systems, a balance must be achieved between thermal stability and processability. On the other hand, extensive research on aromatic thermoplastic polymers potentially offers the favourable properties that make them very suitable for such applications.

9.1 Aromatic Polyamides As a result of their outstanding physical and mechanical properties, aromatic polyamides are attractive materials for use in high-performance structural applications {853072} {833611} {820253}, including aircraft components or fire protection garments, as constituents of both traditional, i.e., fibre-reinforced composites {774110} {762844}, and molecular composites. One of these applications takes advantage of their thermal stability {881232} {825031} {763791} {755849} and allows the manufacturing of heat-resistant materials for fire protection {713906}. In a different context, aromatic polyamides (aramid fibres) have been proposed in the past few years as precursors of activated carbon materials with distinctive adsorbent properties (thermally stable molecular sieves). Aramid fibres, e.g., poly(m-phenylene isophthalamide), poly(p-phenylene terephthalamide) {709654}, etc., are a class of synthetic polymers that possess excellent thermal and oxidative stability, good flame resistance, and superior mechanical and dielectric behaviour. Recent studies have shown that poly(m-phenylene isophthalamide) fibres start to degrade with the cleavage of hydrogen bonds at approximately 335 °C, which leads to a disordering of the polyamide chains on the nanometre scale {838693}. The next decomposition step takes place between 355 and 465 °C, with the disruption of the amide bonds, the subsequent

189

Thermal Degradation of Polymeric Materials breaking of the polymer chains into smaller units, and their condensation into aromatic compounds. From 465 °C onwards, the reaction progresses by the dehydrogenation of the polyaromatic structures and their arrangement into graphite-like assemblies, resulting in the final fibrous carbon, which is obtained at ca. 650 °C. Wholly aromatic poly(amide-hydrazide)s, (CO–NH–Ar–CO–NH–NH–CO–ArO)n, where Ar is a meta- and/or para-substituted phenylene unit, are successfully used in various engineering fields for their ability to form fibres with high mechanical strength and modulus – they were commercialised as excellent salt rejection asymmetric membranes for water desalination and efficient semiconductors from their modified metal chelates. A series of thermal degradation studies have shown that these polymers undergo a thermochemical transformation into the corresponding poly(amide-1,3,4-oxadiazole)s by loss of water [a.340]. The resulting 1,3,4-oxadiazole-containing polyamides could be rightfully classified among the highly thermally stable linear polymers. The reasons for such stability of these polymers are expected to originate primarily from their chemical structure, which is composed of building units generally known to be highly resistant to increased temperatures, such as amide groups, aromatic moieties and 1,3,4-oxadiazolyl rings. In addition to this, their high-temperature stability is also further enhanced by their considerable crystallinity, which was promoted by establishment of strong hydrogen bonding between the amide groups of the neighbouring chain segments. DSC curves indicated a common thermal behaviour of wholly aromatic poly(amidehydrazide) rigid-rod polymers, which all exhibit two endotherms [a.340]. The first is small and characteristic of evaporation of adsorbed surface water, while the second is large and broad due to the thermally induced cyclodehydration reaction of the polymers into the corresponding poly(1,3,4-oxadiazolyl-benzoxazole)s. During the first TG weight-loss step, which occurred between 90 and 130 °C in both air and N2 conditions, the polymer exhibited relatively small losses of only about 1–3 wt%. The second step showed considerable losses and occurred in different temperature ranges for various polymers with residues of 72–56.5 wt% remaining. This step reflected the occurrence of the thermally induced cyclodehydration reaction. The amount of water evolved during the cyclodehydration reaction was 11–13.5 wt% (based on the weight of the dried polymer samples), which was in good agreement with the theoretical value (12 wt%) calculated for the expected poly(1,3,4-oxadiazolyl-benzoxazole) repeat units. The third rapid weight-loss step was associated with the decomposition of the intermediate polymeric structures containing 1,3,4-oxadiazole and benzoxazole rings that were formed in the second step. In the work of Liu and Tsai, TG thermograms revealed that all the aromatic polyamides with bulky cyclic groups containing phosphorus began to degrade at about 310 °C [a.341]. The decomposition temperatures of the polymers at 10% weight loss were 340–390 °C as shown on Table 8.

190


Table 8. Basic thermal properties data of polyamides Reprinted from [a.341] with permission from Elsevier Samples Yields Inherent (%) viscosity (dl/g)

Tg (°C)

PA-1 PA-2 PA-3 PA-4 PA-5 PA-6 PA-7 PA-2

230 245 243 235 235 251 241 289

97 86 97 92 85 98 90 98

0.57 0.81 0.33 0.38 0.52 0.47 0.67 –b

a

Temperature at 10% weight loss

b

Not soluble

Td a (°C) N2

Air

Char residue at 700 °C (%) N2 Air

359 385 341 377 390 336 400 430

355 384 345 391 392 343 402 452

51.5 54.4 40.8 43.9 48.0 44.3 57.2 60.0

43.7 55.9 3.8 49.8 50.3 5.4 57.8 0.0

Char residue at 800 °C (%) N2 Air 48.6 52.4 37.6 42.1 46.4 41.1 55.5 56.1

15.9 29.3 2.8 25.4 29.9 2.1 34.9 0.0

The thermal stability of the polyamide containing 9,10-dihydro-9-oxa-10-oxide-10phosphaphenanthren-10-yl (DOPO) in nitrogen was found to be about 100 °C lower than that of the common phosphorus-free polyamide. On the other hand, the DOPO-containing PA showed higher heat resistance and retarded weight-loss behaviour at temperatures higher than 400 °C. Besides, the phosphorus-containing PA exhibited a highly antioxidative property in the high-temperature region. After the decomposition of the DOPO group at about 410 °C, the DOPO-containing polyamides exhibited excellent thermal stability at 450–650 °C. A very slow weight-loss rate and a small amount of weight loss were observed in this temperature region. At temperatures higher than 700 °C, weight loss due to char oxidation occurred, although the weight-loss rate was still low. From the TG curves of the fluorinated aromatic polyamides derived from two novel monomers, i.e., 5-(4-trifluoromethylphenoxy)isophthaloyl dichloride and 5-(3,5-bistriflu oromethylphenoxy)isophthaloyl dichloride, one work found that these polymers did not show obvious weight losses until the temperature reached 420 °C in nitrogen, implying that no thermal decomposition occurred [a.342]. However, at temperatures over 450 °C the polymers showed rapid thermal decomposition. The fluorinated polyamides have onset decomposition temperatures in the range of 430–460 °C, and temperatures at 5 and 10% weight loss in the range of 440–460 and 480–500 °C, respectively. In addition, the fluorinated polyamides retained 35–60% of the original weight at 700 °C. Hu and co-workers [a.344] obtained aromatic polyamides based on a new diamine containing both arylene ether and bulky substituents – the TG results indicated 10% mass loss at

191

Thermal Degradation of Polymeric Materials decomposition temperatures in the range of 415–430 °C. The Tg of these polymers was in the range of 215–262 °C. In addition, the Tg decreases with increasing volume of the substituted group in the polymer backbone. These observations were attributed to the fact that the 2,5-di-tert-butylbenzene groups had two bulky pendant groups that could result in increasing steric hindrance – the groups increased the space between polymer chains (and therefore the free volume), making the rotational movements of the mainchain segments easier.

9.2 Aromatic Polycarbonates Aromatic polycarbonates are well known as one of the most useful super-engineering thermoplastics because of their excellent properties, such as high impact strength, heat resistance and high optical transparency. Polycarbonates are produced by the interfacial polycondensation of bisphenol-A and phosgene. The major drawbacks of the conventional phosgene process are environmental and safety problems involved in using the highly toxic phosgene as the reagent and copious amounts of methylene chloride as the solvent. For this reason, phosgene-free processes for polycarbonates have been proposed that employ bisphenol-A and diphenyl carbonate, with the latter synthesised also in a phosgene-free process [a.343]. Aromatic polycarbonates exhibit excellent thermal stability, especially in the absence of oxygen and water. The dry polymer may be heated to 320 °C for several hours, or for short times as high as 330–350 °C with only minimal degradation. At these high temperatures, thermal-oxidative degradation leads to slight yellowing, requiring colour stabilisation. Low levels (usually <5000 ppm) of stabilisers (phosphites, phosphonites, phosphines and organosilicon compounds) are usually added during processing. At temperatures above 400 °C, rapid decomposition and cracking occur. Liaw and Chang [a.345] reported on brominated fluorine-containing homopolycarbonates and copolycarbonates of varied unit ratio that were synthesised from 3,3,5,5tetrabromobisphenol-AF and bisphenol-A polycondensed with trichloromethyl chloroformate using a phase-transfer catalyst at 25 °C. The homopolycarbonate based on tetrabromobisphenol-AF had the highest Tg at 205 °C. The TG curves show that the Td of polycarbonates was in the range of 445–475 °C. The LOI of homopolycarbonates based on bisphenol-A and tetrabromobisphenol-AF were found to be 26 and 93, respectively. Investigations on the N,N-diphenyl-N,N-bis(3-methylphenyl)(1,1-biphenyl)-4,4-diamine (TPD)/bisphenol-A polycarbonate system showed that the glass transition temperature of the polycarbonate was reduced significantly in the presence of the TPD [a.346]. The depression of the Tg was more pronounced in the case of the cyclohexyl polycarbonate, which was conformationally more restricted. This was related in part to the large reduction in the conformational energy of the TPD/cyclohexyl polycarbonate pair relative to that with

192

High Temperature-Resistant Polymers bisphenol-A polycarbonate. Infrared spectra showed shifts of aromatic group absorption bands to lower frequencies when the polycarbonates were mixed with TPD, indicating molecular interactions involving the phenyl groups. The frequency shifts followed the same trend as the depression of the Tg and the extent of the shift was more in the case of cyclohexyl polycarbonate.

9.3 Aromatic Polyethers The thermal degradation of poly(arylene ether)s containing naphthalene units of the polymers of each group formed by the same moiety linked to different structures (–O–, –OArO–, –OArArO–, –OArSO2ArO– and –OAr(CH3)O–) used as a polymer composite matrix have been reported {755543}. The moiety present in the first group of polymers was the aromatic triketone –ArCOArCOArCOAr– moiety, while that present in the second group of polymers was the heterocyclic –ArCOQCOAr– moiety (Ar = 1,4-substituted phenylene; Q = 3,7-substituted quinoline). The thermal degradations of aromatic thermoplastics containing carbonyl, ether and sulfone linkages and in whose repeat units the –ArCONaphCOAr– moiety occurred (where Ar = 1,4-substituted phenylene; Naph = 2,6-substituted naphthalene) were presented under N2 flow in dynamic heating conditions {755543}. The results obtained suggested that degradation occurred through random chain scission followed by branching and crosslinking, which were partially superimposed on the initial process. The Kissinger method was used to determine the apparent activation energy value associated with the first degradation stage, and linear relationships, attributable to slow diffusion of degradation products in the melt, were found in some cases {603272}. A thermal stability classification among the studied polymers was made on the basis of the activation energy values obtained. Aromatic thermoplastic polyethers containing ketone and sulfone groups are economically accessible polymers that show favourable properties for use as matrices for highperformance composite materials [a.347]. The kinetics of the isothermal degradation of a series of high thermally stable aromatic poly(ether ketone)s was studied by both a longterm experiment at 270 °C (considerably lower than the temperature of fusion) and a set of short-term experiments at temperatures near the temperature of fusion. An induction period was observed at 270 °C, followed by two degradation stages – the first with an exponential increase of the weight-loss rate as a function of heating time, and the second showing constant weight-loss rate. Short-term experiments showed similar behaviour, but no induction period was observed. A study on the thermal degradation properties of aromatic copolyethers containing a hexamethylene spacer found that the presence of the aliphatic spacer led to microphase separation with a favourable influence on the supramolecular ordering in the solid

193

Thermal Degradation of Polymeric Materials state [a.348, a.349]. The conformational studies showed linear chain geometry, also favourable for microphase separation. Small differences between the starting points of the degradation were explained by the supramolecular ordering, which was more important than the chemical structure. The differences between the lengths of the comonomers did not disturb the microphase separation. Also, the polymers studied had low values of the polar surface, typically below 4%, reflecting that interchain interactions play a secondary role in the thermal degradation.

9.4 Phenylene-Containing Polymers Poly(phenylene sulfide) (PPS) is a semicrystalline polymer with high thermal durability commonly used in engineering. PPS resists organic solvents, chemical corrosion and ignition, and can be easily adapted for processing. This is the reason why it has been evaluated recently for the possibility to replace conventional thermally durable materials and components used by the packaging and automobile industries. The degradation study of PPS is important since processing at high temperature may produce changes that will affect the ultimate performance. Therefore, the studies of its pyrolysis behaviour and mechanism of thermal degradation are of crucial importance for processing and performance evaluation [a.350]. However, the relationships between the detailed mechanisms and kinetics of PPS pyrolysis are not fully clear due to the complexity of the degradation. Studies on the mechanism indicated apparently one-stage pyrolysis, mainly by depolymerisation, main-chain random scission and carbonisation {772093} {660936} [a.350]. The initial scission of PPS was of a depolymerisation type and main-chain random scission followed by evolution of benzenethiol and hydrogen sulfide as major degradation products. The chain transfer of carbonisation also occurred in initial pyrolysis and gradually dominated at higher pyrolysis temperature to form the high char yield of solid residue. The massive evolution of hydrogen sulfide from PPS was similar to the emission of HCl from selected halogencontaining polymers (e.g. PVC) during pyrolysis. Poly(phenylene ether) is inherently a difficult material to work with because rearrangement reactions take place before significant mass loss occurs over the range 450–550 °C with production of 2,6-xylenol, 3,5-xylenol and a range of monomer- and dimer-sized molecules. It is suggested that the changes in the distribution of reaction products that occur with increasing degradation temperature can be explained by a Fries-type rearrangement in which the ether links are broken and chain continuity is re-established through methylene bridges, using the methyl side group [a.351]. The importance of rearrangement reactions has been identified and the significance of these reactions recognised as an explanation for the range of degradation products obtained on pyrolysis. The behaviour of model compounds and kinetic analysis was used to proposals for a mechanism that involved the formation 194

High Temperature-Resistant Polymers of a stable benzyl radical through abstraction of a hydrogen atom, followed by scission of the ether link, and then migration of the phenyl radical to the methyl radical. The thermal degradation of poly(phenylene ether) (PPE) based on 2,6-dimethylphenol has been semiquantitatively studied by experiments and computer simulations [a.131]. The polymer was thermally aged under several different conditions prior to thermal degradation, because rearrangement during ageing was found to be one of the most important processes. Although thermal ageing increased the yield of dimeric scission products, there was no clear increase in monomeric products (Table 9) – the thermal degradation was too complicated to analyse without computer-aided modelling. The amounts of the four major monomeric products, which were approximately the same, were larger than those of the other minor compounds. D1 is the monomer unit of PPE. B1 could be generated by the elimination of a methyl group from D1. However, among the four major products, E1 and G1 could not be obtained by direct degradation of PPE. A simple degradation model was designed and the cleavage rates of the ether, the methylene bridge and the side chain were simulated in order to be able to explain the distribution of the scission products. The results were not totally conclusive, but the degradation rates gave information on the cleavage. The rearranged chain was calculated to be about 60–80% of all chains, and the cleavage rate of the ether bridge was about nine times lower than that of the methylene bridge; also, the cleavage rate of the side chain was almost the same as that of the ether bridge. The work of Lavrenko and co-workers [a.352] on the thermal degradation of poly(1,4phenylene-1,3,4-oxadiazole) found that the degree of coiling of the molecules of degraded products did not differ from that of undegraded poly(1,4-phenylene-1,3,4-oxadiazole) samples, and led to the conclusion that degradation is not accompanied by a noticeable change in the short-range interactions in the molecular chain understood as a random chain scission.

9.5 Poly(Ether Ether Ketone) (PEEK) The mechanism and kinetic parameters associated with the thermal degradation of poly(ether ether ketone) (PEEK) have been studied in the past {871758} {639346} {575553} [a.199]. Both component homopolymers show a single-stage thermal degradation at temperatures lower than 700 °C {456769}. According to Hay and Kemmish [a.203], degradation in PEEK is initiated by random homolytic scission of either the ether or the carbonyl bonds in the polymer chains; however, there is conflicting evidence about which bond is more stable. The radicals formed abstract hydrogen from adjacent phenylene units, or terminate by combination to produce crosslinks. The products of the scission, if sufficiently mobile, will volatilise, while cyclisation to benzofuran derivatives can also occur. The thermal stability of PEEK is greatly reduced in an oxidative environment. 195


Table 9. Weight percentages of low-molecular-weight scission products of PPE Reprinted from [a.131] with permission from Elsevier Thermal ageing temperature Thermal degradation temperature A1

– 300 °C 350 °C 565 °C 764 °C 565 °C 764 °C 565 °C 764 °C 1.00 2.81 1.28 2.92 1.38 2.99

B1

17.90

23.50

19.75

23.57

18.00

21.69

C1

1.60

3.50

1.70

3.12

1.56

3.34

D1

29.08

25.00

29.79

26.77

30.67

23.66

E1

24.90

24.54

23.10

22.74

22.69

26.40

F1

0.34

0.15

0.51

0.84

0.44

0.58

G1

24.25

19.65

21.58

18.62

23.44

19.98

H1

1.13

0.86

2.29

1.41

1.82

1.35

The thermal decomposition of blends of PEEK with poly(p-hydroxybenzoic acid-co-2,6hydroxynaphthoic acid) (PHAHNA) using thermogravimetry under dynamic conditions has been investigated [a.355]. The thermal analysis of the blends showed that thermal stability is clearly affected with respect to the unblended materials – blending accelerates the degradation process. In the case of PHAHNA, the liquid-crystalline polymer was

196

High Temperature-Resistant Polymers destabilised at high PEEK contents. Blending appeared to modify only the rate of degradation of the component polymers, whereas the mechanisms remained unchanged. The FTIR spectrum of PEEK studied is presented in Figure 41. PEEK/poly(ether sulfone) (PES) blends in nitrogen atmosphere are reported to show that, though PEEK is more stable than PES, based on their TG profiles and activation energy values, the degradation rate is much faster in PEEK [a.199]. Both neat polymers decompose essentially in a single-stage degradation. The blends also showed single-stage degradation with a broadening of the degradation step. Further, it was observed that the presence of one component influences the thermal degradation behaviour of the other component, and the resultant blend has a lower thermal stability. The destabilising influence at low concentration of PES in the blend was due to the chemical interactions of the degradation products of PES (which starts degrading at lower temperature) with PEEK and reduction in the viscosity of the medium. But the decrease in the thermal stability at low concentration of PEEK in the blend, though noted as unusual, was not explained. Complete mechanistic knowledge of the degradation process in the blend, using TG properly interfaced with a spectrometer, would give a better understanding of the observed behaviour. However, such results have not yet been published.

9.6 Polybenzimidazoles (PBIs) Polybenzimidazoles (PBIs) are high-molecular-weight, strong and stable polymers containing recurring aromatic units with alternating double bonds that are produced mainly from the condensation of 3,3,4,4-tetraaminobiphenyl (diaminobenzidine) and diphenyl isophthalate. PBIs exhibit a high degree of outstanding mechanical, thermal and dielectric properties at high temperature and are widely used for fibres that withstand very high-temperature conditions. In addition, PBIs are the only known available polymeric matrix capable of maintaining load-bearing properties for an adequate length of time at 600 °C. Factors that have retarded the use of PBIs for structural applications include difficult processing conditions, the need for techniques to control voids from off-gases, and special requirements in handling the prepolymer. Studies have shown that PBIs display a two-step decomposition process: the onset begins at about 160 °C, and the second step occurs at a temperature around 625 °C [a.356]. In the first decomposition stage, the polymer progressively loses weight until the temperature reaches around 320 °C; beyond this temperature, the loss progresses gradually. After 625 °C, a sharp reduction in weight occurs – in the first stage at temperatures up to 625 °C it was only 5 wt% loss, but in the second stage between 625 and 800 °C it was about 10 wt% loss. Concerning the flammability, the limiting oxygen index (LOI) of PBIs is 41, so these polymers do not burn in air. In addition, PBIs emit little smoke at extremely high temperatures. Ariza and co-workers [a.357] studied the thermal properties of a sulfonated PBI (sPBI) membrane

197


A

Figure 41. FTIR spectrum of (A) undegraded PEEK and the solid residues from the degradation process at (B) 5.2%, (C) 10.5%, (D) 15%, (E) 24.7% and (F) 40% weight loss Reprinted from [a.355] with permission from Elsevier

198

High Temperature-Resistant Polymers and showed that most of the sulfuric acid molecules are ionically bonded to the imide ring, that the thermally treated sPBI membranes retain more acid molecules after the washing process than the non-heated ones, and that the thermal stability of PBIs decreased with an increase in the degree of sulfonation.

9.7 Polybismaleimides (BMIs) Crosslinked bismaleimides form an important class of high-performance, thermosetting polymers – their remarkable characteristics (like excellent thermal and oxidative stability, flame retardation, low propensity to moisture absorption, ease of synthesis and costeffective processing) render them suitable for varied high-performance applications {695555} [a.449]. However, the aromatic nature and the high crosslink density of the cured network make them brittle and hence call for matrix toughening to make them widely useful [a.295]. To improve the brittleness of BMI resins, some toughening agents such as diamines, dithiols, diallyl bisphenols, allylamine and reactive elastomers were added to the curing formulations of BMI resins. Incorporation of flexible ether groups and long phenoxy chains into the BMI monomer structures was another approach towards increasing the toughness of BMI resins [a.358]. Torrecillas and co-workers [a.359, a.360] established that, during thermal degradation of bismaleimide and bisnadimide networks, water, carbon monoxide and carbon dioxide are the most important products formed (Table 10); the major organic compounds detected were aniline, polycyclic molecules and isocyanate products, as shown in Scheme 35. The LOI increased with the number of crosslinking points, either maleimide or nadimide types. The results also showed a high sensitivity to oxygen for these materials and they indicated that the nadimide structure increased material thermostability. Interestingly, Liu and Chen [a.358] reported a one-stage weight loss in nitrogen and a two-stage pattern in air of polybismaleimides having silicon groups. The cured BMIs were found to have high glass transition temperatures above 210 °C and good thermal stability over 350 °C. The weight-loss behaviour corresponded with the resulting high char yields of the polymer. The specific behaviour of BMI degradation was correlated with the condensed-phase mechanism of char formation and flame retardation, which are widely observed with phosphorus-containing polymers. The extremely high thermal stability of the degraded residue of BMI was manifested by almost no weight loss occurring at temperatures above 700 °C. The incorporated silicon groups contributed to the high thermal stability of the residue through the formation of a silica protecting layer, which was thermally stable and extraordinarily heat-resistant, on the residual surface.

199


Table 10. Percentage of the various products detected during thermal degradation of BMI at 500 and 600 °C under nitrogen Reprinted from [a.360] with permission from Elsevier Proposed molecule

Molecular Gaseous products Heavy products weight solubilised in acetone recovered resolubilised in acetone 500 °C 600 °C 500 °C 600 °C 93 36 21.3 – 19.2

106

5.5

6.4

–

–

107

–

–

–

2.7

133

44

47.3

–

29.3

147

10.4

10

–

6.2

175

–

–

15

1

189

–

–

51

2.7

A study of the adhesive characteristics of the BMI/diallyl bisphenol-A/formaldehyde resin (ABPF) blend has shown that the system can serve as a good high-temperature adhesive [a.361]. TG analysis of the addition cured blend of 2,2-bis(4-(4-maleimidophenoxy)phe nyl)propane (BMIP) with ABPF resin, of varying maleimide to allyl phenol stoichiometry, showed a similar pattern of single-stage decomposition [a.295]. The systems with higher BMIP content exhibited a marginal improvement in thermal stability. Similar to the polymeric BMIP/ABPF system, the monomeric BMI/diallyl bisphenol-A (DABA) system

200


Scheme 35. Formation of isocyanate derivatives during thermal degradation of BMI Reprinted from [a.360] with permission from Elsevier

showed single-stage decomposition but with reduced thermal stability. The 2,2-bis(4maleimidophenyl)methane (BMIM) and 2,2-bis(4-maleimidophenyl) ether (BMIE) systems had identical decomposition patterns with nearly equal decomposition temperatures and char residues. The BMIP system also showed a similar pattern, but the initial temperature of decomposition (Ti) and char residue were slightly less, which was reported as the result of lower crosslink density and early degradation occurring at the isopropyl moiety. Differing from the other three, the 2,2-bis-(4-maleimidophenyl) sulfone (BMIS) system showed two-stage mass loss and had the lowest char residue. Bibiao and co-workers [a.362] presented DSC and TG results of modified polybismaleimide resins, prepared via Diels–Alder reaction of 4,4-bis(maleimidodiphenyl)methane (BMIDM) with different bisfuranylmethylcarbamates, before and after curing. DSC measurement illustrated that the curing reaction shifted to high temperature when the BMIDM content increased; also, H series resins have relative lower curing temperature than T series resins when the BMIDM content is nearly the same. TG data show that all the cured modified resins underwent initial degradation above 340 °C and had a statistic heat-resistant index Ts above 210 °C, thus revealing that the cured modified polybismaleimide resins had moderate heat resistance and thermal stability.

201

Thermal Degradation of Polymeric Materials Lin and co-workers noted that both 2,2-bis(4-(4-maleimidophenoxy)phenyl)propane (BMPP) and 2,2-bis(4-(4-maleimidophenoxy)phenyl)hexafluoropropane (BMIF) showed lower melting points than BMIDM, probably as a result of the flexible linkage (such as ether and isopropylidene linkage) and the bulky group (such as hexafluoroisopropylidene) between the phenyl units of the long-chain bismaleimide monomer, which interfered with close packing between molecules and led to a decrease in melting point [a.363]. Both BMPP and BMIF showed a lower decomposition temperature than BMIDM due to the presence of long and flexible segments between cure sites, which caused a small reduction in thermal stability. The Tg of the cured bismaleimide resins was measured by TMA using an expansion method. The influence of the incorporation of flexible segments on the Tg was that the ether-containing bismaleimides (BMPP and BMIF) have lower Tg. Moreover, this was ascribed to the relatively long distance between the cure sites of BMPP and BMIF, which had relatively lower Tg than BMIDM, by decreasing the crosslink density of the cured resins. It was suggested that the mouldability of BMPP and BMIF resins could be improved by the incorporation of flexible groups, which would decrease the crosslink density of the cured resins without significantly reducing their thermal properties.

9.8 Polybenzoxazines Polybenzoxazines offer high mechanical performance with excellent thermal properties, including low flammability. Thermal degradation studies performed by GC-MS have detected eight different categories of compounds as degradation products [a.286]. The degradation products either were a direct result of polymer degradation or resulted from the recombination or degradation of compounds formed during the thermal decomposition. Benzene derivatives, amines, phenolic compounds and Mannich base compounds came directly from the decomposition of polybenzoxazines, while 2,3-benzofuran was derived from the further degradation of phenolic compounds. On the other hand, biphenyl compounds were obtained from the recombination between two phenyl radicals formed after the loss of substituents from benzene derivatives, amines and phenolic compounds. Isoquinoline and phenanthridine derivatives were formed when Mannich base compounds lost the hydroxyl group and underwent successive dehydrogenation. The presence of 2,3-benzofuran, isoquinoline and phenanthridine derivatives, as well as biphenyl compounds, was critical and responsible for the char formation of aromatic amine-based polybenzoxazine. Another study on the thermal degradation of the polybenzoxazine dimers established that the majority of the model dimer compounds, which contain Mannich bridges but have no phenolic hydroxyl groups, simply evaporate when exposed to high temperature [a.364]. For the benzoxazine dimers, there were two different fragmentation processes: cleavage of C–N and C–C bonds occurring simultaneously during the thermal decomposition.

202

High Temperature-Resistant Polymers Both fragmentations resulted in the evaporation of the amines. The radicals left after the evaporation of the amines combined with each other to form bisphenol and biphenyl compounds. The presence or absence of a substituent at the ortho position to the hydroxyl group made a significant difference in of the decomposition pattern and the char formation of the dimer. The unblocked ortho position facilitated the radicals to undergo aromatisation and crosslinking processes that finally led to the formation of char. For the modified benzoxazine dimers, in addition to the cleavage of C–N and C–C bonds, the breaking of the C–O bond was found to be another possible fragmentation during the thermal decomposition. As a result, secondary and tertiary amines were formed, instead of primary amines. This led to a shift of the onset of degradation to higher temperatures compared to those of the unmodified benzoxazine dimers. The synthesis of poly(p-phenylene benzobisoxazole) (PBO) involves polymerisation of 4,6-diamino-1,3-benzenediol with an intermediate obtained from the reaction of poly(phosphoric acid) and terephthaloyl dichloride [a.365]. In addition to its advantageous mechanical characteristics, PBO displays good chemical resistance and excellent thermal stability, which makes this polymer particularly suitable for high-temperature applications. Although the origin of such behaviour can ultimately be traced to the high conjugation and rigidity of its constituent units, a detailed understanding of the thermal degradation process of PBO becomes essential if its applications in special fields are to be fully realised. Studies have shown that at temperatures below 500 °C the polymer retains its original conformation and becomes stabilised by enhancement of its crystallinity. Decomposition takes place in a single step and the main changes occur within a very narrow temperature interval. The formation of aryl amides as intermediates in the decomposition process has been detected; these amide bonds subsequently degrade by homolytic breakage, yielding nitriles. The final carbonaceous residue was rich in nitrogen and retained a certain degree of anisotropy, a fact that was explained by the conservation of crystallinity at an intermediate decomposition stage.

9.9 Other High-Temperature Polymers Composites containing a polymer matrix are a valuable class of materials often used in high-temperature applications – phenolic resins and epoxies can be considered useful polymer matrix materials in that respect, as described in the subsequent sections.

9.9.1 Phenolic Resins Phenolic resins are thermosetting polymers with high chemical resistance and thermal stability but low toughness and mechanical resistance. Phenolic resols have intrinsic

203

Thermal Degradation of Polymeric Materials resistance to ignition, low generation of smoke and relatively low cost. On the other hand, a disadvantage is that they are characterised by a complex process of polymerisation (cure) with the generation of water and formaldehyde and consequent formation of voids. Therefore, the processing of phenolic materials requires careful temperature control and gradual heating to allow continuous elimination of volatiles and to reduce the number of defects in final components. Normally the time required for these operations is incompatible with common industrial process schedules [a.368]. Phenol–formaldehyde resins are known for their high-temperature resistance and high charyielding properties. Phenolic resins with improved thermal and pyrolysis characteristics are desirable for application in composites for thermostructural applications. The thermal decomposition of phenol–formaldehyde resins consists of breaking the bonds between the aromatic rings and the methylene bridges [a.366]. After ing hydrogen atoms to the formed radicals, phenol and methylphenols as well as the bis- and trisphenols, etc., and their methyl derivatives are formed. Isomers of bis(hydroxyphenyl)methane are found in considerable quantities only in the case of partially cured novolac resins that contain linear sequences in their structure. During the thermal degradation, the pyrolysis products of phenol–formaldehyde resins are condensed ring compounds – fluorene (diphenylenemethane), dibenzofuran, xanthene, anthracene and phenanthrene. These aromatics can be created in a cyclisation reaction with the participation of hydroxyl or methyl groups in the ortho position to methylene bridges. The presence of multi-ring aromatics suggests that some dehydration and dehydrogenation occurs, which may not be free radical in nature. The thermal behaviour of maleimide functional resin (PMF), phenyl ethynyl phenol– formaldehyde resin (PEPF), ethynyl phenyl azofunctional resin (EPAN) and propargyl ether resin (PN) has been investigated in comparative studies as a function of their structure [a.367]. From the results, the acetylene and propargyl ether groups reduced the cure temperature significantly in comparison to the phenyl ethynyl and maleimide groups. All systems exhibited improved thermal characteristics in comparison to conventional resol resin. The PMF resins exhibited the lowest thermal stability and the maximum was in the EPAN system. PN and PEPF showed intermediate thermal stability, while maximum char yield was obtained for the EPAN system. Non-isothermal kinetic analysis of the thermal degradation implied that the polymer undergoes degradation in at least two kinetic steps. The very low pre-exponential factors for all polymers led to the conclusion that the volatilisation process of the degraded fragments controls the kinetics of degradation. Isothermal pyrolysis studies showed that complete pyrolysis of the polymer is not achievable under some conditions. In the case of nitrogen-containing polymers, pyrolysis occurred via loss of nitrogenous products, which is conducive to enhancing the carbon content of the resultant char. The FTIR spectra of the pyrolysed samples confirmed the presence of C–O linkages in the char; XRD analysis of the partially carbonised polymers

204

High Temperature-Resistant Polymers showed that these polymers are generally amorphous except for the EPAN system, which contained a minor amount of the crystalline phase. The thermal degradation of phenolic resins is characterised by a complex mechanism with at least two different processes (Scheme 36) that lead to the formation of a stable and resistant char structure [a.368]. During degradation, the reactions of chain scission and further crosslinking occurred simultaneously, although it was not possible to uncouple the two contributions and to determine the activation energy of each partial reaction. The global activation energy of

Scheme 36. A possible reaction during thermal degradation of phenolic resins Reprinted from [a.368] with permission from Elsevier

205

Thermal Degradation of Polymeric Materials the whole process reportedly increased with temperature, suggesting an increase in the number of chemical bonds in the network that led to the char structure formation at the end of the process. In addition, blending with epoxy/amine blends has been reported as a suitable route to improve the mechanical properties of phenolic resins and to reduce the cure temperature [a.368]. The results reported demonstrate that the epoxy/amine content should be kept below 15 wt% to avoid a significant reduction of the thermal stability of the blend. Flame resistance experiments identified the aromatic diglycidyl ether of bisphenol-A-based (epoxy equivalent of 190 g/mol) epoxy/amine system as one of the best epoxy resins to produce thermally stable blends with the phenolic resol.

9.9.2 Epoxies Epoxy resins are one of the most important thermosetting polymers for high-performance composites, primarily for aerospace applications. For this use, epoxy resins must show appropriate thermal and fire resistance performances in addition to the required high mechanical properties [a.450]. Although thermosetting epoxies posses high tensile strength and modulus, and excellent chemical and solvent resistance, these materials are generally brittle due to the high crosslink densities [a.152]. Epoxies are usually blended with elastomers in order to improve their fracture toughness or to maintain the high Tg of the unmodified resin. Blending has been reported to have a great influence on the thermal stability of individual polymers since it can have profound and sometimes unexpected effects on thermal stability, which cannot simply be predicted on the basis of the behaviour of the components and their relative proportions. The thermal stability for epoxy blends depends strongly on interactions between the individual polymers, and thus improvements of mechanical properties by blending are sometimes at the expense of stability. It has been proposed that the thermal degradation of epoxy resins proceeds along several concurrent pathways [a.369, a.451]. The first one was reported as the homolytic cleavage of the bisphenol-A unit to yield isopropylphenol, ethylphenols, cresols and phenol; while the second one was the heterolytic cleavage of the bisphenol-A unit to yield isopropenylphenol and phenol; the last one was postulated to be the cyclisation product of the glycidyl ether side chain to yield C6H5–O–C3H3 or C6H4–O–C3H4. In addition, during the pyrolysis of DGEBA, some uncured resin was produced. The unpyrolysed resin remains as part of the high-boiling-point volatiles, thus hindering the study of its thermal stability. Fundamentally, a different work proposed a single-step degradation process for either aliphatic or aromatic epoxy resins [a.368]. Incorporation of hydroxyl-terminated poly(dimethylsiloxane) into epoxy resin enhances the thermal degradation temperature [a.370]. The presence of the siloxane skeleton in

206

High Temperature-Resistant Polymers the unmodified epoxy system delays degradation, and a high amount of thermal energy is required to attain the same weight losses when compared with that of unmodified epoxy systems. The delay in degradation caused by the siloxane moiety may be attributed to the stability of the inorganic siloxane structure, which may stabilise the epoxy resin from the heat. The high energy of the siloxane bond and its partial ionic nature has been clearly shown to be responsible for the substantial thermal stability of epoxies. The siloxane bond energy is significantly greater than those of carbon–carbon and carbon–oxygen bonds. Further, it was also observed that the type of curative and percentage concentration of siloxane in the unmodified epoxy have specific influences on thermal degradation. Jiang and co-workers [a.371] studied the thermal degradation of PP/maleic anhydridegrafted ethylene-propylene-diene elastomer (MAH-g-EPDM) matrix and dynamically cured PP/MAH-g-EPDM/epoxy blends. In the case of the PP/MAH-g-EPDM (75/25) and PP/epoxy (80/20) blends, the incorporation of 25 wt% MAH-g-EPDM into the pure PP decreased the initial degradation temperature from 370 to 335 °C, but the addition of epoxy resin increased the initial degradation temperature from 370 to 400 °C. The initial degradation temperature of the PP/MAH-g-PP/epoxy (55/25/20) blend was found to be similar to that of the pure PP – decomposition started at 370 °C and finished at 450 °C. The dynamically cured PP/MAH-g-EPDM/epoxy (55/25/20/0.8) blend has the higher initial degradation temperature compared to the PP/MAH-g-PP/epoxy (55/25/20) and PP/epoxy (80/20) blends. This showed that dynamic curing of epoxy resin in the blends could further improve the thermal degradation properties. The temperatures of the maximum rate of mass loss for PP, PP/MAH-g-EPDM (75/25), PP/epoxy (80/20), PP/MAH-g-PP/epoxy (55/25/20) and dynamically cured PP/MAH-g-EPDM/epoxy (55/25/20/0.8) blends were obtained from DTG curves, and Tmax had the same tendency as Ti. The degradation processes of all the blends were one-step processes similar to the PP decomposition process.

9.9.3 Poly(Ether Imide) (PEI) Poly(ether imide) (PEI), which is a high-performance engineering thermoplastic polymer, has characteristics such as high strength and rigidity at elevated temperatures, long-term heat resistance, dimensional stability and good electrical properties {870572} {757742}. Like other amorphous, high-temperature resins, PEI has outstanding dimensional stability and is inherently flame-retardant {638278}. PEI resists chemicals such as hydrocarbons, alcohols and halogenated solvents. Creep resistance over the long term allows PEI to replace metal and other materials in many structural applications. Crosslinking was reported to be the main mechanism during initial processing followed by chain scission {812988}. When the processing temperatures and dwell time were increased, Tg resulted in an increase of more than 30%. An increase in matrix molecular weight due to crosslinking caused an

207

Thermal Degradation of Polymeric Materials increase in viscosity of the material, which in turn required more processing cycles or higher consolidation pressure.

208

Recycling of Polymers by Thermal Degradation

10


The degradation of waste plastics into fuel for the recovery of the organic content of the polymeric material represents a sustainable way of saving valuable petroleum resources in addition to protecting and conserving the environment. Mechanical recycling by regranulation and melting of the used plastics [a.452], feedstock recycling and energy recovery are the three main alternatives for the management of plastic wastes in addition to landfilling [a.387]. Feedstock recycling of waste plastics has more advantages than mechanical recycling or energy recovery, as the energy consumption of the process is very low, since only about 10% of the energy content of the waste plastic is used to convert the scrap into petrochemical products, and also the volume of harmful gases produced is much lower than that of the incineration process. In addition, some of the formed byproduct gases from this process such as HBr and HCl can be recovered and utilised as raw materials [a.387]. HCl can be used, for example, for the synthesis of vinyl chloride monomer for PVC production. An obstacle to the resource recovery of post-consumer plastics into feedstock chemicals and fuels is the wide variety of polymers present in the municipal solid waste stream [a.83]. Separation of each polymer can be both expensive and time-consuming, and significant error is also generally observed during manual separation. Instead, a potential solution may be to apply a limited separation step to remove those polymers that may be readily isolated from one another or those that may cause adverse effects during processing. Nevertheless, before a commercial process for the degradation of used plastic products into feedstock chemicals is implemented, it is first necessary to understand the effects that each polymer can have on the degradation of other reactants through intermolecular reactions. A logical starting point is to study the degradation of a feed comprising two polymers that differ from one another in only a few attributes [a.372]. The main types of feedstock recycling processes can be summarised as: chemical depolymerisation, gasification and partial oxidation, thermal degradation, catalytic cracking and reforming, and hydrogenation {852663}. The current advanced thermal treatments are as follows [a.372].

209

Thermal Degradation of Polymeric Materials • Depolymerisation. Depolymerisation processes use high-energy microwaves in a nitrogen atmosphere to decompose organic materials. The waste absorbs microwave energy, increasing the internal energy of the organic material to a level where chemical decomposition occurs on a molecular level. The nitrogen blanket forms an inert, oxygen-free environment to prevent combustion. Temperatures in the chamber range between 150 and 350 °C. At these temperatures, metals, ceramics and glass are not chemically affected. Depolymerisation processes based on the chemical cleavage of polymer molecules aim to convert them back into the raw monomers. • Gasification. This is the thermal degradation of organic compounds, otherwise referred to as carbonaceous materials, at high temperatures (900–1400 °C) in a low-oxygen atmosphere, to produce a combustible gas, referred to as syngas, and an inert, possibly vitrified, solid residue. • Pyrolysis. Pyrolysis is the thermal decomposition of organic materials at temperatures in excess of 200 °C in the complete absence of air. The end-product of pyrolysis is a mixture of solids (char), liquids (oxygenated oils) and a combustible gas, or syngas, comprising methane, carbon monoxide and carbon dioxide, with proportions determined by operating temperature, pressure, oxygen content and other conditions. The process does not affect metals, ceramics and other inert materials. By means of pyrolysis, high-molecular-weight polymers are converted into useful end-products such as gasoline oil that could be upgraded to transportation fuels. • Plasma gasification. Plasma discharge uses extremely high temperatures in an oxygenstarved environment to decompose input waste material completely into very simple molecules in a process similar to pyrolysis. Products include combustible gas and a vitrified solid residue. Pyrolysis is becoming an important route for treating the increased amounts of waste plastics – mainly LDPE, HDPE, PP, PS, PVC, PU, PA and PET that for more than 90% of total plastic wastes {888003} [a.373]. The process parameters that have the largest influence on the pyrolysis products are the temperature and heating rate {886353}. The oil produced has a high heating value and may be combusted directly or refined for the recovery of speciality chemicals. The production of a liquid product has advantages in that it is easier to handle, store and transport and hence the product does not have to be used at or near the pyrolysis process plant. Some polymers, such as PMMA, PS and PTFE, can be readily pyrolysed and lead to the recovery of a substantial proportion of the original monomers, whereas polyolefins, especially PE and PP, yield low-molecular-weight fragments that can be used as fuels. Other major generic types such as polyesters and polyamides can be hydrolysed in the presence of strong acid or base, generating a great proportion of the original monomers. Unfortunately,

210

Recycling of Polymers by Thermal Degradation the polymers that can be really recycled form only a minor part of the polymer waste due to their high cost and the downgraded properties of the recycled products, and for some reasons within the materials themselves, such as impurities and crosslinking {847519} {824066}. One of the problems associated with the pyrolysis approach is the high energy needed for the macromolecules to undergo thermal degradation. Therefore, it is of practical importance to reduce the degradation temperature and promote the degradation efficiency. For deg pyrolysis procedures, the behaviour of polymers during thermal decomposition with regard to the decomposition products and the kinetics of decomposition must be known. Furthermore, changes in the kinetics of decomposition of polymers due to interactions of the polymers during thermal degradation have to be quantified and explained. Waste plastics, especially thermoplastics, can be regarded as being a cheap and abundant source of chemicals and energy. Furthermore, recycling of thermoplastics from waste products can contribute to solving the pollution problems associated with the landfilling and incineration of plastics. Though several methods have been proposed for recycling waste plastics, it is generally accepted that material recovery is not a long-term solution to the present problem, and that energy or chemical recovery is a more attractive one. Consequently, new technologies are being developed for the chemical recycling of plastic wastes. Recent research has emphasised reactor and process design to maximise product value, extraction of kinetic models from thermal decomposition studies, and evolution of the changes in molecular weight of the degrading polymer residue [a.53]. Much less attention has been paid to mechanistic modelling, particularly of the composition of the volatiles, at the molecular level.

10.1 Polyolefins During the thermal recycling of polyolefins, different hydrocarbons are formed – the process has been extensively studied {890883} {889249} {871062} {870655} {845098} {845096}. Although polyolefins can be recycled via gas pyrolysis to produce gasoline-like materials, olefins are easily polymerised into unusable compounds during storage and transportation. Thus the oils obtained by thermal degradation have been reported by several research groups as unfit for fuel oils. A fixed-bed reactor under argon flow has been employed to pyrolyse PP and atactic polypropylene (a-PP) [a.373]. a-PP is the side product of polypropylene production and, because the structure of the polymer chain is not regular, a-PP is not resistant to chemical attack and is thermally weaker than other types of PP such as isotactic PP (i-PP) and syndiotactic PP (s-PP). Thus, a-PP is an undesired and unemployed product in the petrochemicals industry. The maximum volatile product evolution temperature was 420 °C

211

Thermal Degradation of Polymeric Materials for a-PP and 425 °C for PP. The recovery of hydrocarbons was determined and pyrolysis products were classified – Figure 42 displays the GPC results obtained for PP. PP and a-PP decomposed into a large number of aliphatic compounds without a residue. Some 96 wt% of the carbon in PP and 97 wt% of the carbon in a-PP was converted to volatile organic compounds such as alkanes, alkenes and dienes. Major compounds are, for instance, C9 hydrocarbons, such as 2-methyl-1-octene, 2-methyl-2-octene, 2-methyl4-octene, 2,4-dimethyl-1-heptene and 2,6-dimethyl-2,4-heptadiene. A kinetic study of the thermal decomposition of polypropylene, oil shale and their mixture concluded that the polypropylene accelerates the decomposition of the organic

Figure 42. Gas chromatogram of organic products at the temperature of the maximum of their evolution during pyrolysis of PP Reprinted from [a.373] with permission from Elsevier

212

Recycling of Polymers by Thermal Degradation matter in the oil shale. The degradation of the studied materials was considered as a first-order reaction, as was indicated by the results of an isoconversional kinetic method. The calculated activation energy of organic matter decomposition of the mixture was 240 kJ/mol, substantially higher than that for the unmixed oil shale (60 kJ/mol) and close to that of polypropylene (250 kJ/mol). The results indicated that the characteristics of the process depend on the heating rate, and that PP acted as a catalyst in the degradation of the oil shale in the mixture [a.374]. Thermal degradation of blends based on recycled PP and an elastomeric additive, ethylene/ -octene copolymer (EOC) or synthetic rubber, in low concentration is usually a one-step process and can be kinetically described by using the autocatalytic model [a.3]. Figures 43 and 44 show the dynamic TG results obtained. In the case of recycled PP/EOC blends, the absence of residues after completion of the decomposition process indicates the small influence of additive, as displayed in Figure 45. The stability of recycled PP was not significantly altered by the addition of small amounts of additives. Therefore, as the thermal properties of the blends did not significantly change, the improvement in mechanical and impact properties as well as the increase in value of a solid urban waste made the blending economically viable. This is also the reason for using

Figure 43. Dynamic TG curves of recycled PP at different heating rates Reprinted from [a.3] with permission from Elsevier

213


Figure 44. Dynamic TG curves of synthetic rubber at different heating rates Reprinted from [a.3] with permission from Elsevier

Figure 45. Dynamic thermal degradation curves of recycled PP/EOC (Affinity™) blends with different composition (15 °C/min) Reprinted from [a.3] with permission from Elsevier

214

Recycling of Polymers by Thermal Degradation higher amounts of elastomeric additives, as the final product would be too expensive to permit commercial viability of the recycled product. Polymer degradation in solution has several advantages over melt pyrolysis. The degradation of LDPE occurs at much lower temperatures in solution (280–360°C) than in conventional melt pyrolysis (400–450°C) {744165} [a.375]. LDPE was dissolved in liquid paraffin and degraded for 3 h at various temperatures (280–360°C). Samples were taken at specific times and analysed with gel permeation chromatography for the MWD. The time evolution of the MWD was modelled with continuous distribution kinetics. The data indicated that LDPE followed random-chain-scission degradation. The rapid initial drop in molecular weight, observed up to 45 min, was attributed to the presence of weak links in the polymer. The rate coefficients for the breakage of weak and strong links were determined, and the corresponding average activation energies calculated to be 24 and 88 kJ/mol, respectively. Pure LLDPE resin undergoes thermal degradation, which is complete at temperatures up to 550 °C, and there is no residue at the end of the degradation [a.376]. Flame-retarded LLDPE composites degrade with about 30% residue left after the degradation. Only a one-step process in the thermal degradation of pure LLDPE in nitrogen was reported, whereas the degradation of LLDPE composites proceeded in a three-step process. Also, an LLDPE/magnesium hydroxide/red phosphorus (5:4:1) composite degraded much less than an LLDPE/magnesium hydroxide (1:1) composite at the same temperature in the almost whole degradation process, which indicated that the partial replacement of magnesium hydroxide with red phosphorus in the LLDPE/magnesium hydroxide composite resulted in the increase of thermal stability. In another development, a technology has been developed to manufacture wood-flake-reinforced high-density polyethylene {817912} {700203} ecocomposite materials, based on post-industry wood wastes and post-consumer plastics waste, aiming to provide an ecologically friendly and economically effective technology to re-use abundant recycled solid wastes into a new market [a.377].

10.2 Polystyrene Applied pyrolysis converts waste plastic into fuel; then the converted fuels or chemicals could be merged in standard petrochemical or petroleum refining industry operations or the recovered monomer could be used to produce new polymer {787548}. A TG study of waste expanded polystyrene and general-purpose PS revealed that pyrolysis is activated around 380 °C and the rate of thermal degradation was maximal at 430 °C [a.378]. Maximum styrene selectivity (75 wt%) with oil yield (95 wt%) was achieved during the pyrolysis of waste expanded polystyrene at 450 °C. The rate of thermal degradation of waste expanded polystyrene was found to be higher than that of general-purpose PS at

215

Thermal Degradation of Polymeric Materials 450 °C, and styrene production was also more by 11 wt% in the case of waste expanded polystyrene. Oil yield increased from 24 to 98 wt% with increase in reaction temperature from 350 to 480 °C, while styrene selectivity, which was about 76 wt% up to 450 °C, decreased sharply to 49 wt% at 480 °C. This decrease in styrene selectivity takes place with increase in styrene dimer formation from 4 to 10 wt% and production of other chemicals. Also the production of toluene, ethylbenzene and methylstyrene decreased with the same rise in pyrolysis temperature. Thermal degradation of PS was reported to have started with random initiation to form polymer radicals, the main products being styrene and its corresponding dimers and trimers. These results were comparable with studies reported elsewhere on oil yield of 99 wt% with 60 wt% styrene monomer and 25 wt% for other aromatics. Another work has reported on recovery of 58 wt% styrene from thermal degradation of PS at 350 °C after a time of 60 min [a.379]. Furthermore, oil yields of 82 wt% with 70 wt% styrene selectivity and 77 wt% styrene recovery at 580 °C have been reported [a.380].

10.2.1 Polystyrene in the Melt The pendant phenyl groups in polystyrene restrict rotation around single bonds in the backbone. Polystyrene is thus a stiff molecule that favours crystalline regions in the melt, whereas in a good solvent polystyrene molecules are dispersed and unaggregated. Such crystallinity or aggregation in the melt would hinder hydrogen transfer and, therefore, restrict the random degradation mechanism. A further possible explanation of the observations is the relatively restricted molecular mobility in the melt, which may keep newly formed radicals in close proximity. The cage effect would thereby increase the probability for radicals to recombine before they diffuse apart. Each permanent radical formation may therefore be the result of multiple initiation and recombination events. Sterling et al. {822782} observed differences in polystyrene degradation in solution and in the melt. The activation energy for polystyrene random scission is substantially smaller in solution than in the melt (29 kJ/mol compared to 188–235 kJ/mol). For other polymers, such as PMMA and poly(-methylstyrene) (PAMS), the activation energy for solution degradation was also lower than for pyrolysis. During pyrolysis of polystyrene melt, chain-end products of degradation (oligomers) are significant, whereas in solution with reflux the generation rate of such products is negligible compared to products of random scission.

10.2.2 Polystyrene in Solution Degradation in solution offers an alternative by keeping all products in a single phase of much lower viscosity, enabling better heat transfer, enhanced reaction rates, improved

216

Recycling of Polymers by Thermal Degradation residence time control, and a modified product profile, thus allowing the introduction of plastic wastes into convectional cracking technology [a.375, a.381]. Degradation of polymers in solution has been proposed to amend some of the problems encountered in pyrolysis {847519}. For instance, the activation energy for degradation of PS by a Lewis acid (AlCl3) in chlorobenzene at various temperatures and at various mass concentrations of AlCl3 was reported as 32 kJ/mol, and the degradation rate was proportional to the fourth power of the aluminium chloride mass concentration. Polystyrene degradation was investigated in the presence of aluminium chloride at 50 °C, and a linear decrease in molecular weight with degradation time was observed [a.382]. Investigations on the thermal degradation of monodisperse PS and a segmented homopolymer of ABS terpolymer in bean oil revealed that in the presence of bean oil PS thermally degraded into larger fragments of polymer chain instead of the monomer or oligomers that are usually generated in the direct PS thermal degradation process [a.383]. The average molecular weights of the degradation products of PS decreased and the polydispersity indices increased as the reaction time increased. As a hydrogen-donor substance, bean oil could ‘capture’ the radicals, especially the terminal ones, which redirect or terminate the macroradicals involved in the degradation of PS.

10.3 Poly(Vinyl Chloride) Recycling of used PVC needs a careful thermal characterisation of PVC waste {853278}. The analysis of the scrap, especially with respect to thermal stability and molecular weight, is necessary before reprocessing. Besides the material recycling of PVC, there have been several attempts to prepare low-molecular-weight products from PVC by chemical or thermal treatment. Most of the proposed processes use the rather easy dehydrochlorination of PVC under the influence of either heat or alkaline media. The chemical recycling is mostly based on the idea of converting polymers back into short-chain chemicals for re-use in polymerisation or other chemical processes. Four main process technologies under current consideration for chemical recycling of PVC are cracking, gasification, hydrogenation and pyrolysis, as mentioned earlier. In order to recover usable products from PVC pyrolysis, two problems must be solved. First, the HCl, which is corrosive to the pyrolysis equipment, must be isolated and recovered, and then the concentration of chlorinated hydrocarbons in the liquid phase must be controlled. Studies of PVC thermal decomposition at high temperatures aimed at the recovery of useful materials from the thermal degradation of PVC have been shown by most investigators to proceed in two distinct stages (Scheme 37) [a.385].

217


Scheme 37. PVC thermal decomposition under vacuum Reprinted from [a.385] with permission from Elsevier

Up to around 360 °C, the degradation almost exclusively involves dehydrochlorination to a polyene structure accompanied by the evolution of large quantities of HCl and small amounts of hydrocarbons, especially unsubstituted aromatics, such as benzene, naphthalene and anthracene. Above 360 °C, structural degradation of the backbone occurs, leading to the formation of toluene together with a small quantity of other alkyl aromatics, yielding a residual char [a.385]. It has been reported that, at low temperatures, PVC pyrolysis yields predominantly HCl, then oil and solid residue are formed (Figure 46). The main chlorine-containing hydrocarbons reported are methyl chloride, vinyl chloride, ethyl chloride and chlorobenzene, all at trace levels. PVC pyrolysis at higher temperatures up to 500 °C leads to the formation of several organochlorine compounds with a total yield of 1.75 wt% of the liquid fraction. There is still a lack of information about the production of chlorinated compounds formed after pyrolysis [a.96, a.384, a.385]. A kinetic model involving three apparent reactions has been proposed, which considers that each weight-loss stage on the TG curve corresponds to an apparent reaction. Both a consecutive and a parallel model have been applied. The kinetic parameters and the conversion factors obtained from both models were quite similar, giving activation energies of 198, 143 and 243 kJ/mol for the first, second and third apparent reactions, respectively [a.385].

218


Figure 46. Comparison of PVC pyrolysis yields with DTG curve at 10 °C/min Reprinted from [a.385] with permission from Elsevier

Two-stage thermal decomposition for the pyrolysis reaction of waste PVC has been reported [a.386]. The first stage was ed for by stoichiometric reaction to yield volatiles and intermediates, while the second step was associated with the thermal decomposition of intermediates competitively into gas, liquid and solid by-products. The effects of additives on the pyrolysis kinetics of waste PVC seemed to be significant, especially in the first-stage reaction, which was retarded. A merged peak at low temperatures was observed on the DTG curve (HCl evolution) instead of the two peaks usually observed for pure PVC resin. The first peak on the DTG curve of pure PVC resin shifted more, resulting in the complete overlap of the two peaks. The quantity of evolved HCl decreased because of interaction of the metal components of the stabilisers with either HCl or active chlorine atoms or both. The final residual fraction increased as a result of pyrolysis of organic additives to yield extra char. On the other hand, the second-stage reaction kinetics demonstrated a similar pattern to that of pure PVC resin, implying that the effects of additives were less significant in comparison with that in the first-stage reaction. Moreover, it was observed that a plastisol with optimum thermal stability could be produced by holding PVC at 145 °C for 10 min and using a range of plasticiser concentrations between 50 and 70 phr [a.387]. A comparison between the apparent activation energies of the thermal

219

Thermal Degradation of Polymeric Materials degradation process of traditional phthalate-containing plastisols with those obtained for these new formulations, based on polymeric additives, showed that these are slightly lower. But the addition of stabilisers provided higher activation energies and degradation temperatures. The effects of the curing time, curing temperature and concentration of plasticiser on the thermal stability were evaluated and enhanced the optimum conditions that could be used in industrial processing. The degradation behaviour obtained with PVC formulations where phthalate plasticisers were replaced by polymeric plasticisers clearly indicated the suitability of their use in industrial processes from the point of view of their thermal stability. Jaksland and co-workers have recently proposed an environmentally sustainable technology for thermochemical recycling of PVC wastes [a.353]. The new technology transforms PVC wastes into completely new chemical products/raw materials. The process is based on a combined thermal and chemical degradation of PVC – in the reactor, the chlorine from the PVC reacts with fillers, producing calcium chloride. The metal stabilisers (lead, cium, zinc and/or barium) are converted to metal chlorides. Exploiting the influence of pH, temperature and liquid-to-solid product ratio, metals and calcium chloride are sequentially extracted from the reaction product. This occurs in a downstream multi-stage extraction–filtration procedure. The products from the process are calcium chloride, which satisfies the specifications as thaw salt that may be further purified and re-used, in addition to coke and organic condensates that may be used as energy resources for the process. It should however be pointed out that pyrolysis of PVC suffers from a number of technical problems due to its high chlorine content. A large amount of hydrogen chloride is produced during the thermal decomposition of PVC, in addition to the formation of undesirable compounds such as chlorinated hydrocarbons. Unlike PE, PP and PS degradation, the degradation of PVC produces halogenated organic compounds that prevent the use of the waste plastic degradation oil as a fuel or chemical feedstock, necessitating the removal of the halogen content from the waste plastic-derived oil. The current focus is to establish the optimal reaction conditions for a large-scale PVC vacuum pyrolysis operation and to improve knowledge about the reaction mechanism aimed at reducing the formation of chlorinated organic compounds [a.96]. Various researchers have studied the formation of hazardous volatiles including HCl during the incineration of PVC-containing materials {889475} {886353}. However, there is still a lack of full information concerning the recovery of valuable pyrolysis products from PVC-containing waste polymers.

10.4 Polyamides Polyamides are intrinsically hygroscopic because of the polar nature of their chains, which makes them moisture-sensitive materials, and hence chemically unstable towards hydrolytic conditions. Prior to processing steps, polyamide composites need to be dried in order

220

Recycling of Polymers by Thermal Degradation to attain their good mechanical properties afterwards. Nevertheless, polyamide-based materials may undergo discoloration and some deterioration after long drying periods. During composite reprocessing, the polyamide chain size is generally reduced and the decrease can cause a loss of packing capability of the chains to form crystalline domains. In one report, the moisture content was evaluated from the weight loss observed near 150 °C – this temperature is higher than the boiling point of water, but smaller than the onset of decomposition temperature of the polymer [a.354]. The drying period exhibited a great influence over the final moisture content of the material. During the first 3 h of drying, the material lost 35% of its initial moisture. The moisture loss after 6 and 9 h of drying was 50 and 68.8%, respectively. The recovery of -caprolactam from waste PA-6 realises the high value of PA-6 and has therefore the potential to be economically competitive with the traditional synthesis processes; significant positive environmental impact should also be mentioned. A presently applied recycling process is the Zimmer AG process, which performs the depolymerisation of PA-6 with the help of steam and liquid catalysts such as phosphoric acid [a.17]. This process can be applied only for non-mixed PA-6 materials. A disadvantage of this process is the high yield of salts and traces of phosphoric acid in the recovered -caprolactam, which is a drawback for the production of fibres. The kinetics of thermal recycling reactions of coal with PA-6 by TG and DSC methods has been studied and overall kinetic parameters were determined [a.389]. Simultaneously, in the range of about 350–500 °C, the thermal degradation of coal preceded the evolution of gas and tar, and also the thermal decomposition of PA-6 occurred and -caprolactam and other products were formed. The -caprolactam formation is promoted by water and hydrogen from coal degradation – due to the high content of hydrogen, coal acted as a strong hydrogen donor. The thermal treatment of waste polyamides with coal was tested during co-pyrolysis in a stationary bed up to a final temperature of 900 °C at a heating rate of 5 °C/min. It was found that besides -caprolactam mainly carbon oxides, methane, aliphatic hydrocarbons, simple aromatics and stable oil were formed during co-pyrolysis – the yields of gas and tar/oil from co-pyrolysis with polyamides were higher in comparison with those from pyrolysis of coal alone.

10.5 Natural Polymers

10.5.1 Poly(L-Lactic Acid) Hydrolysis to L-lactic acid and thermal degradation (depolymerisation) to the cyclic dimer L-lactide (LLA) is the general procedure for PLLA recycling [a.246]. For depolymerisation,

221

Thermal Degradation of Polymeric Materials intensive studies have been carried out to reveal the thermal degradation mechanisms of PLLA [a.246, a.248] {607567}. However, for a closed system, there is limited information available on the effects of degradation temperature and time on the yield of lactides (LAs) by thermal degradation of PLLA, on the contents of L-lactide (LLA), D-lactide (DLA) and meso-lactide (MLA) in the formed LAs, and on the formation mechanism of DLA and MLA from optically pure PLLA. Even in a distillation system, it is reported that significant amounts of DLA and MLA are formed during depolymerisation of PLLA, which reduces the yield of LLA, and that the amounts of DLA and MLA depend on the degradation catalyst and temperature. It is expected that such formation behaviour of LLA, DLA and MLA can be readily traced in a closed system and that the obtained information should be useful to increase the yield of LLA even in a distillation as an open system [a.390]. For the former method, increasing the temperature is effective to hydrolyse PLLA rapidly to L-lactic acid. However, crystalline residues, which are formed as a result of selective hydrolysis of the chains in the amorphous regions, require a long period for their complete hydrolysis when the hydrolysis is carried out in the temperature range below the melting temperature of PLLA. Thereby the crystalline residues will prolong the total period required for the complete hydrolysis of PLLA or decrease the yield of L-lactic acid in a short period. Lately, a new strategy for recycling PLLA to L-lactic acid utilising hydrolysis in the melt in high-temperature and high-pressure water has been proposed [a.391]. Depolymerisation of PLLA has been carried out in a sealed tube as a closed system in the temperature range of 250–290 °C for 15 h without further addition of depolymerisation catalyst [a.392]. The highest yields of LAs and LLA were as low as 14 and 8%, respectively, which were much lower than the 89 and 88% for the open system, and than the 93 and 89% of the highest yields of lactic acids and L-lactic acid when PLLA was hydrolysed in the melt in high-temperature and high-pressure water. The fractions of LLA and of DLA and MLA decreased and increased, respectively, with increasing degradation temperature and time. The low yield of LA such as 14% was explained by polymerisation of the LAs formed in the closed system and by the formation of low-molar-mass compounds other than LAs, while the low yield of LLA such as 8% was attributable to the formation of high amounts of MLA and DLA, in addition to the low yield of LA.

10.5.2 Lignocellulose Biodegradable polymers have recently attracted much public and industrial interest because of the increasing waste disposal problem. Besides solutions such as incineration, recycling or re-use, biodegradable polymers can be entirely converted by microbial activity in a biologically active environment to biomass and biological by-products. For

222

Recycling of Polymers by Thermal Degradation highly cellulosic biomass feedstock, the liquid fraction usually contains acids, alcohols, aldehydes, ketones, esters, heterocyclic derivatives and phenolic compounds [a.393]. The pyrolysis liquids are complex mixtures of oxygenated aliphatic and aromatic compounds. The tars contain native resins, intermediate carbohydrates, phenols, aromatics, aldehydes, their condensation products and other derivatives. Pyroligneous acids can consist of 50% methanol, acetone, phenols and water. CH3OH arises from the methoxyl groups of uric acid and from the breakdown of methyl esters and/or ethers from decomposition of pectin-like plant materials. Acetic acid comes from the acetyl groups of hemicelluloses. The pyrolysis gas mainly contains CO, CO2, H2, CH4, C2H4, C2H6, minor amounts of higher gaseous organics and water vapour [a.393]. The main components of alkaline black liquor, which is a lignocellulosic waste resulting from digestion with NaOH of wood, straw or other fibrous plants, are organic and inorganic compounds deriving from the raw material used in the papermaking process and water [a.394]. Black liquor can also be considered as a by-product because approximately 35% of the total energy requirement in the pulp and paper industry comes from black liquor combustion. Two of the most important properties of black liquor are its tendency to swell when heated and its viscosity. Both properties depend strongly on the organic matter composition and influence the efficiency of black liquor combustion or gasification. Nowadays, black liquor is concentrated and burned in recovery boilers to generate energy and to recover the inorganic chemicals required in the papermaking process. Both pyrolysis (or thermal devolatilisation) and gasification are potential processes for using black liquor as a source to produce gaseous products for use as a fuel. Pyrolysis or thermal devolatilisation is not only important as a thermochemical process in itself, but also as the previous stage in gasification processes. Gasification would be more energy-efficient if more of the carbon content in black liquor were converted to combustible gases instead of char during pyrolysis. Both pyrolysis and gasification can be considered as alternatives to conventional boilers for recovering chemicals and energy from lignocellulosic wastes. Thus, a good understanding of the behaviour of black liquor during the pyrolysis or devolatilisation process is required for the design and development of new alternative combustors and gasifiers. Study of pyrolysis of lignocellulose from straw shows that the thermal decomposition of the organic matter fraction of black liquor takes place at temperatures below 550 °C in an N2 atmosphere [a.394]. The weight loss observed at temperatures higher than 550 °C has mainly been related to reduction reactions of alkaline compounds by carbon. The final solid conversion and the devolatilisation rate (Figure 47) are also noticeably influenced by the heating rate applied. The usual problems experienced with common recovery boilers are aggravated in processes involving alkaline black liquor derived from straw. It is known that different lignocellulosic wastes can behave very differently in the same gasifier under the same operating conditions

223


Figure 47. Devolatilisation rate versus temperature at different heating rates of lignocellulose at a final pyrolysis temperature of 800 °C Reprinted from [a.394] with permission from ACS

[a.395]. The consequent need to develop alternative processes for the use of black liquor for energy purposes, such as pyrolysis or gasification, requires a good understanding of the thermochemical behaviour of this complex substance. In particular, the influence of the final pyrolysis temperature (250–900 °C) and the heating rate (5–30 °C/min) on the product yields, gas composition and specific surface area of the resulting char in a fixed-bed reactor has been analysed. The results obtained show that both the energy recovery and the specific surface area of the char increase with a rise in the final pyrolysis temperature and the heating rate.

10.6 Other Homopolymers Studies on the waste from PET bottles hydrolytically depolymerised in a high-pressure autoclave equipped with a stirrer concluded that the degree of degradation of PET increases as the particle size of PET decreases, reaching a maximum of 24.61% at a size of 1 mm × 1 mm [a.396]. The rate constants of hydrolysis of PET were found to be greater at higher particle size, in contrast to lower particle size, if catalyst (lead acetate) is used. Further, 224

Recycling of Polymers by Thermal Degradation the rate decreases at lower concentration of PET and due to deposition of terephthalic acid (TPA) on smaller particles of unreacted PET {736642} {706905}. Studies on the effect of continuously added di-tert-butyl peroxide on the thermal recycling of PAMS by thermal degradation in trichlorobenzene have been reported to significantly alter the PAMS degradation mechanism, from one in which random homolytic initiation is followed by depolymerisation to produce monomer in purely thermal processes, to one in which hydrogen abstraction to renew the cycle of random degradation is also significant with added peroxide {822782}. Also, investigations have been conducted on the recycling of polystyrene in solution for small conversions, for which the random scission rate coefficient was independent of molecular weight {668871}.

10.7 Mixtures of Polymer Wastes Pyrolysis is one form of energy recovery process that has the potential to generate char, oil and gas products, all of which have a potential end-use. The installations used for the pyrolysis of mixed plastics include melting vessels, blast furnaces, autoclaves, tube reactors, rotary kilns, coking chambers or fluidised-bed reactors {886353} {887517}. Kinetic data obtained from high-resolution thermogravimetry and gradient-free reactor experiments confirm that different molecular structures of commodity plastics bring about different reaction mechanisms of thermal decomposition, different reaction rates and different temperature dependences of the decomposition rates. From that, stepwise pyrolysis of mixtures of plastics seems to be reasonable where the different components of the mixture are pyrolysed at different temperatures [a.397]. The process conditions of pyrolysis can be optimised to enlarge the production of the pyrolytic char, oil or gas, which have potential uses as fuels [a.222]. However, depending on the pyrolysis temperature, the char fraction contains inorganic materials ashed to varying degrees, any unconverted organic solid and carbonaceous residues produced from thermal decomposition of the organic components. The liquid fraction is a complex mixture of organic chemicals. Thermal decomposition of commingled plastics comprising high- and low-density polyethylene, polystyrene, polypropylene and poly(vinyl chloride) was investigated under vacuum conditions by dynamic thermogravimetry in the temperature range of 25–600 °C [a.398]. The results suggest that, when the municipal plastic waste (MPW) is pyrolysed together in order to recover valuable materials, the pyrolysis can be separated into two steps: The first step occurs at 375 °C, with a weight loss of ca. 10 wt%, part of which is attributed to HCl production and the rest refers to various volatiles formed from the polymers present in the mixture. The second step is between 375 and 520 °C, where ca. 85 wt% of all the liquid product is obtained, while solid residue is left behind; little contamination of the organic liquid with HCl occurs during this step. The kinetic models used for the study are presented in Table 11. 225

226

2 α 2 PS 2 ⎯ ⎯ → V2 + R 2

k

1 α 1PS1 ⎯ ⎯ → V1 + R 1

k

PP ⎯ ⎯ →V +R

k1

2 α 2 LDPE 2 ⎯ ⎯ → V2 + R 2

k

1 α 1LDPE 1 ⎯ ⎯ → V1 + R 1

k

HDPE ⎯ ⎯ →V +R

k1

Kinetic model

dt

d[PS 2 ]

dt = −A2 e

= −A1e RT

−Ea 2

RT

RT

n2

[PS 2 ]

[PS1 ]

n1

[LDPE 2 ]

n2

n1

[LDPE 1 ]

n1

[PP]

−Ea1

RT

−Ea1

= −A2 e

−Ea 2

RT

n1

[HDPE]

−Ea1

RT

−Ea1

= −A1e

= −A1e

= −A1e

d[PS1 ]

dt

d[PP]

dt

d[LDPE 2 ]

dt

d[LDPE 1 ]

dt

d[HDPE]

Differential equations

Ea2 = 185

Ea1 = 120

125

Ea2 = 120

Ea1 = 120

250

Ea (kJ/mol)

n2 = 0.76

n1 = 1.60

0.40

n2 = 1.40

n1 = 1.40

0.65

n

A2 = 2.32×1013

A1 = 1.06×108

2.04×108

A2 = 1.34×1015

A1 = 1.34×109

1.71×1017

A (min-1)

Table 11. Kinetic analysis of the thermal decomposition of PE, PP, PS and PVC Reprinted from [a.398] with permission from Elsevier

2 = 0.90

1 = 0.10

2 = 0.10

1 = 0.10

Yield coefficient


a

dt

d[R 2 ]

dt

n1

RT

n

− A3e

[SR 1 ] 3 )

n2

[I]

−Ea 3

RT

n1

[PVC]

RT

n

n

[I] 2 )

[SR 1 ] 3 )

RT

−Ea 2

−Ea 3

[PVC] − A2 e

RT

−Ea1

−Ea 2

= g(A3e

= e(A2 e

RT

−Ea1

= −A1e

= b(A1e

d[R 1 ]

dt

d[I]

dt

d[PVC]

Differential equations

Ea3 = 243

Ea2 = 143

Ea1 = 198

Ea (kJ/mol)

n3 = 1.58

n2 = 1.15

n1 = 1.04

n

A3 = 5.77×1016

A2 = 9.95×1010

A1 = 3.57×1018

A (min-1)

V, volatiles; I, intermediate; SR1 residue generated in the first stage; SR2, residue generated in the second stage

3 eR 1 ⎯ ⎯ → fV2 + gR 2

k

bI ⎯ ⎯ → cV1 + eR 1

k2

PVC ⎯ ⎯ → aHCl + bI

k1

Kinetic model

Table 11. Continued...

g = 0.06

e = 0.36

b = 0.52

Yield coefficient


227

Thermal Degradation of Polymeric Materials Studies on the pyrolysis of plastic wastes have confirmed that the stepwise decomposition of polymers for gasification and separation of plastic mixtures (PVC, PE and PS) in a cascade of cycled-spheres reactors is an alternative pyrolysis procedure [a.397]. For processing of mixtures containing up to 15 wt% of PVC, no further pre-treatment was reportedly necessary. The dehydrochlorination took place quantitatively in the first reactor of the cascade. The chlorine balance gave a rate of conversion of about 99.6%, which corresponded to the results of isothermal measurements. In spite of the high chlorine amount in the feed, the product gases from the third reactor of the cascade contained only 0.0044 wt% chlorine. The second reactor of the cascade was used to decompose PS and the monomer was obtained in high yield. Neither hydrogen chloride nor gaseous pyrolysis products of the decomposition of PE were detected in the second reactor. PE decomposed in the third reactor of the cascade into paraffins and olefins. The interactions of single polymers in the liquid polymer mixture provide an additional means to control the products from polystyrene pyrolysis. Depending on the composition of the mixture, the pyrolysis products of polystyrene shifted to monocyclic aromatic compounds with a yield up to 93% and the amount of styrene dimer and trimer decreased. Thermal degradation of plastics such as PE, PVC, PET and their mixtures (PE + PVC and PE + PET) has been studied at 430 °C by batch operation to analyse the conversion of waste plastics into fuel oil [a.399]. Products of degradation were classified into the three groups of gases, liquids and residues in the reactor. The degradation of PE produced liquid products that consisted of a C5–C25 fraction of hydrocarbons with a yield of 70 wt%. On the other hand, the degradation of PVC produced only 4.7 wt% liquid products, which consisted of a C5–C20 fraction of hydrocarbons; while the degradation of PET produced no liquid products. The effect of mixing PVC and PET with PE on the yield and compositions of liquid products was investigated too. As a result, the addition of either PVC or PET to PE decreased the overall liquid products yield although it promoted the degradation of PE into low-molecular-weight liquid hydrocarbons. In a parallel study, the thermal degradation of PE/PVC at 420 °C, PP/PVC at 380 °C and PS/PVC at 360 °C into oil has been carried out in a glass reactor under atmospheric pressure by batch operation to investigate the recycling process {889475}. The results showed the yield of products obtained from the degradation of PVC mixed plastics, and also showed the distribution of chlorine in the products and the chlorine content (both organic and inorganic) of the oil – for the degradation of various mixed plastics systems, liquid yield was highest (73%) for PP/PVC and lowest (61%) for PS/PVC. The residue, which consisted of carbonaceous materials and heavier hydrocarbons, was 19.3% in the case of the PS/PVC system. For chlorine balance, 88–96 wt% of chlorine content of the sample was evolved as gaseous HCl, 3–12 wt% as liquid, and less than 2 wt% remained as residue. The chlorine content in the oil was determined by oxygen flask method with a Cl– ion-selective electrode. The organic chlorine content of the oil from PP/PVC was

228

Recycling of Polymers by Thermal Degradation 13 500 ppm, which was the highest among the PVC mixed plastics. The oil obtained from the degradation of PE/PVC mixed plastics contained 3600 ppm of organic chlorine. This led to suggestions that, if only organic chlorine compounds had been formed by the degradation of PVC, all the chlorine content in the oil would have been the same. However, from the results, the chlorine content of the liquid products differed according to the type of degraded plastic; thus it was suggested that organic compounds are formed by a reaction between the products of the degradation of PVC and the products of the degradation of the other polymers just as reported elsewhere [a.400]. Pyrolysis of the refuse-derived fuel (RDF) from mechanically separated municipal solid waste (MSW) produced approximately 28% of oils, 30% of non-condensable hydrocarbon gases and 42% of solid residues at 410 °C [a.401]. The TG technique was used for quantitative prediction of the RDF pyrolysis rate. The overall pyrolysis reaction rate was calculated from key component fractions (LDPE, HDPE, PS, PVC and cellulose) of the RDF using the weighted sum method. Good agreement was found in the pyrolysis kinetics between the RDF itself and the weighted sum method of the polymer components in the RDF. Pyrolysis of RDF in a fixed-bed reactor also had a similar result. This approach allowed one to easily for RDF composition variations, thus rendering the model more generically valid. During the thermal degradation of polymer mixtures containing PVC, over 90% of feed chlorine has been recovered as HCl by a dehydrochlorination step at 350 °C for 1 h [a.402]. The use of N2 flow during both the dehydrochlorination and degradation steps decreased the intensity of the reaction between HCl and the degradation products of the polymer due to the effective removal of HCl from the reactor before the degradation of other polymers occurred. Red mud had a good effect on the fixation of evolved HCl as well as iron oxides sorbents. It also affected the composition of organic chlorine compounds derived from the degradation of PS/PVC and PP/PVC mixtures. The use of red mud in vapour-phase- mode led to more HCl fixation, but vapour-phase- mode increased the chlorine amount in the oil, because of the reaction between fixed HCl and adsorbed degradation products on red mud; it showed no effect on the cracking of both PVC-containing polymer mixture and single polymers. The oils derived from PVCcontaining polymer mixtures by thermal degradation contained a lower amount chlorine than the oils obtained by use of red mud. Waste commodity plastics, LDPE, HDPE, PET, PS and PVC, have high hydrogen-to-carbon ratio and molecular chain structures suitable for liquefaction. Therefore, direct liquefaction can be considered as a potential waste disposal option by generating oil [a.372]. Bockhorn and co-workers [a.403] studied complex plastic mixtures such as PVC/PS/PE, PS/PA-6/PE and PVC/PA-6/PS/PE. The researchers calculated the conversion degree of the mixtures, based on the HDPE/LDPE/PP/PS/ABS/PVC mixture. Thermal decomposition of LDPE

229

Thermal Degradation of Polymeric Materials and HDPE from a mixture of products (PE, PP, PS) at 300–500 °C resulted in linear- and branched-chain paraffins and olefins from methane to C30 and some aromatic hydrocarbons [a.404]. The oil products were found to consist of mainly C10–C15 and C16+ hydrocarbons, with a relatively small amount of C5–C9. It was observed that chain scission was primarily a random process for poly(-olefins) but not for PS. Trimers were the upper size limit for PS, while the cracked products for polyethylene had no size limit. Miranda and co-workers [a.385] focused on recovery of valuable products from the vacuum pyrolysis of MPW. Two stage vacuum pyrolysis performed at temperatures of 360 and 520 °C enabled 99.11 wt% of the chlorine to be captured as HCl, followed by polyene decomposition into 32.39 wt% oil, 0.34 wt% gas and 8.53 wt% solid residue (yields based on the initial PVC basis).

10.8 Thermal Degradation of Polymeric Materials – Ecological Issues

10.8.1 Disposal Options and Sources of Information Plastics offer a variety of environmental benefits. However, their production, applications and disposal present many environmental concerns. New environmental regulations, societal concerns and a growing environmental awareness throughout the world have triggered a paradigm shift in industry to develop products and processes compatible with the environment, such as capability to compost into non-toxic residues under both aerobic and non-aerobic thermal conditions. Indeed, some polymer products have been designed in the past with little consideration to their ultimate disposability. Of particular concern are plastics used in single-use disposable packaging, service-ware items, disposable non-woven plastics, coatings and marine plastics. This brings up the so-called ‘cradle-tograve’ issue of deg polymeric materials that integrate material design concepts with ultimate disposability and resource utilisation and conservation. Such designed material properties have the capacity to help resolve the fundamental ecological issues associated with the disposal of plastics, notably the visible impact of a hitherto indefinite lifespan, and especially that of disposing of processing waste. Sources of information that can be accessed for evaluating products include the Hazardous Substances Data Base, which contains a summary of the fate of the substance in the environment and can be accessed via MEDLARS (Medical Literature Analysis and Retrieval System) [a.405], the EPA AQUIRE (Environmental Protection Agency–Aquatic Life) data base, which contains a listing of aquatic toxicity data for many industrial chemicals [a.406], Syracuse University’s data base of fate studies [a.407], and quantitative structure–activity relationships (QSARs) such as the EPA programme ECOSAR (Ecological

230

Recycling of Polymers by Thermal Degradation Structure Activity Relationships) [a.408], where molecular modelling software has been developed that can accurately predict important physical properties such as thermal properties/recycling feasibility.

10.8.2 Sustainable Development Sustainable development is the driving force for acting responsibly to protect Nature for future generations – it consists of an equal-balance triad between environmental, economical and social aspects. Polymers and the polymer industry take a key role in contributing to sustainable development. This is proven at every product life stage – manufacturing, use and disposal [a.409]. At the production stage, modern plant technology leads to lower energy consumption and lower emissions into air, water and soil. The largest contributions to resource savings take place during the use of the final plastics product. The durability and longevity of plastics and the fact that plastics require no further maintenance the optimisation of the demand for resources. For instance, some identifiable ecological aspects of sustainable development for use of packaging plastics are energy efficiency, reducing raw materials consumption and recoverability. For economic aspects, one should include job creation, lower prices for consumers and competitiveness of industry; while the social aspects of sustainable development include affordability to more people, convenience to the consumer, hygiene and safety. Therefore, sustainability cannot be assessed on an environmental basis only, but, instead, a well-balanced evaluation of all the three pillars of sustainability must be carried out as an example of climate protection by the use of renewable resources analysed in a comprehensive approach. The use of renewable raw materials in the manufacture of biodegradable plastics opens up a potential saving of CO2 emissions, though these savings have been said to be currently at a low level. However, noticeable contributions can be achieved through secondary effects during the use phase. For instance, lower fuel consumption and thus saving of CO2 emissions result by reducing the rolling resistance of starch-containing elastomer fill materials in car tyres.

231


232

Thermal Degradation During Processing of Polymers

11

Thermal Degradation During Processing of Polymers

Owing to the nature of polymer processing (mainly casting, injection, extrusion or moulding), deteriorative reactions occur at this early stage when polymers are subjected to heat and mechanical stress, leading to further degradation during the useful life of the materials, when heat and oxygen are the most important degradative factors. In more specialised applications, degradation may be induced by high-energy radiation, ozone, biological action, hydrolysis and many other influences. However, there are also new developments in polymer technology in which degradation processes find positive applications, for instance, recycling of polymeric products, photodegradable plastics and fabrication of fibres such as carbon fibres from polyacronitrile or cellulose. In fact, the microelectronics industry is dependent upon polymer degradation in the manufacture of its circuitry. Besides, degradation and combustion studies are getting more and more involved in the definition of the fire hazards that are associated with polymeric materials. Polymer properties may also be improved by processes like curing and grafting, which may be induced intentionally during processing. The mechanisms and kinetics of all these reactions taking place during processing must be understood if the technology and application of polymers are to continue to advance. Polymer extrusion is one of the most widely used polymer processing methods. The melting zone during co-rotating twin-screw extrusion has been identified as the most significant determinant in the process in attaining the best properties of the polymeric materials. The axial location in the feed stream of the conversion of solid polymer resin to molten mass is associated with an abrupt rise in melt viscosity that provides sufficient shearing of material. The focal points of extrusion are a function of material, machine design and process variables and can be summarised in of three variables: material variables, i.e., melting point, heat capacity, enthalpy of phase transition; machine variables, such as screw design; and process variables, i.e., barrel temperature profiles, screw speed and feed rate. Hence the melting process entails both homogenisation and desired and/or undesired high shearing that are much like those in any functionalisation process based on initiation and grafted monomer addition. In the case of polymer blends, their properties are largely

233

Thermal Degradation of Polymeric Materials determined by the morphological structure of the system containing two (or more) macromolecular components. In of extruder design, this means that it is necessary to have models available for estimating the development of the morphology over the length of the screws. Since significant morphological changes are observed in the melting section, in particular, it is necessary to analyse not only the plasticising process for binary material combinations but also the initial formation and further development of the morphology in this section of the extruder.

11.1 Polyethylene Polyethylene undergoes important volume variations during expansion in the melting process and contraction due to pyrolysis {827220} {798556}. The heat transferred to the region of the sample where melting occurs during processing is consumed in this process, so no temperature increase is observed. The melting process increases the temperature gradients throughout the sample thickness. The thermal conductivity of molten PE is lower than that of solid PE, so the temperature gradients in the sample are higher. Another important cause of the significant increase in the temperature gradients was that the variable incident heat flux over the surface increased during the experiment [a.410]. The melting effects on the heat transfer behaviour were more evident for low heat fluxes. For a slightly higher heat flux, it was observed that pyrolysis of the surface started approximately when melting of the bottom surface had finished. For higher heat fluxes, the effects associated with melting can be hidden by the weight loss and volume reduction of the sample due to thermal decomposition. For these heat fluxes, the pyrolysis started at the surface when part of the sample was still undergoing melting. Once the pyrolysis temperature was reached, the material generated gases and vapours. A regression of the surface started such that the heat flux incised more directly onto inner regions of the polymer, increasing the rate of the global process accelerated by the decreasing mass of the remaining material. The pyrolysis was reported as endothermic though its enthalpy was relatively low compared to the radiant heat flux, which led to deductions that the heat consumption affected the temperature profiles only slightly. The results from thermal degradation work showed HDPE to be less prone to oxidation than PP {888522} {865214} {831619} {734805}. The data obtained with HDPE showed the melt flow and the low shear viscosity to decrease continuously with increasing extrusion es. However, the weight-average molecular weight determined by GPC showed a slight increase only after the first extrusion . From the second to the fifth , the molecular weight seemed to decrease slightly. Thus, the GPC results were found to be misleading. They were related only to the soluble or slightly branched part of the polymer present after processing. Chain branching and crosslinking, leading to an increase of the molecular

234

Thermal Degradation During Processing of Polymers weight of the polymer, were seen as the dominant reactions with the specific HDPE used in the studies. They were found to be favoured in comparison with thermooxidatively and thermomechanically induced chain scissions, which occurred simultaneously. The study of several LDPE resins of different molecular weights and vinyl group contents (‘irregular structures’) showed a relationship between the amount of crosslinking and the proportion of chains having vinyl groups [a.410]. This relationship was thought to hold at least for resins produced by the same process and having similar additive levels, which would be useful for predicting the resin types most susceptible to crosslinking. The variation of the torque was measured at various temperatures between 150 and 375 °C as well as the concentration of unsaturated groups and carbonyl groups. The emphasis was on the kinetics of the reaction of vinyl groups during processing of LDPE in the torque rheometer. A linear relationship was found between the trans-vinylene and carbonyl groups formed. Above 325 °C the vinyl concentration also increased with processing time. However, the vinylidene concentration did not change significantly below 250 °C, but did so above that temperature.

11.2 Polypropylene and its Blends Controlled degradation of PP in an extruder was one of the earliest methods of extrusion routinely used to produce target rheology grades [a.453]. The addition of peroxides causes chain scission and modifies viscosity [a.415] {639383} {635363}. The degree of thermal degradation changes with screw design and screw speed, with varying speeds being used in co-rotating and counter-rotating modes as a result of induced shear stresses and increased viscous heating during the operation {749582}. Degradation during processing also differs from the peroxide-induced degradation {567842}. This method has been swiftly taken over by the functionalisation of PP, which is a far newer method {868543}. During functionalisation, carboxylation of PP adds acid groups to the polymer that improve adhesion to many substrates and impart compatibility with many other polymers [a.434]. PP is normally carboxylated by grafting onto the backbone either methacrylic acid (MA) or acrylic acid (AA). The melt phase reaction of AA onto PP proceeds by a freeradical mechanism. The radicals generated by thermal decomposition of initiator abstract hydrogen from the PP backbone and initiate homopolymerisation of AA, resulting in products such as AA-grafted PP and poly(acrylic acid) homopolymer. MA grafts undergo homopolymerisation under conditions of high radical concentration. The radical-induced reaction of MA with PP yields a polymer containing individual MA units at the chain ends along with the degraded PP. The mechanism of MA grafting is related to the mechanism of MA homopolymerisation – in the reactive species is an excited dimer of MA. Since the excited dimer is capable of

235

Thermal Degradation of Polymeric Materials abstracting hydrogen from the backbone of PP, MA contributes to the degradation of PP. The -scission of the PP backbone leading to lower viscosity (‘vis-breaking’) and melt flow increase is an example of an undesirable side reaction. Also, the homopolymerisation of monomer to oligomers is a source of waste, as monosubstitution of functional group is preferred in enhancing adhesion of the polymer backbone to sized fibre glass and various surface-treated reinforcements. Because of the superposition of residence time and shear deformation aspects of this type of reactive extrusion process, the functionalisation of molten polypropylene resin is representative of a compounding methodology to modify PP resins in general. Investigations have been conducted into polypropylene/polyamide-6 (PP/PA-6) blends with small components (by weight) of the disperse PA phase whereby process conditions of screw speed, throughput and viscosity ratio were varied through the use of two different PP grades [a.411]. The degree of melting and the development of the morphology over the length of the screws were determined for the individual tests. It was determined that, on the second component, which melts at higher temperatures, a kind of melt film removal occurred at the surface of the granules as they melt. The drops of the second component in the melting section, which were directly adjacent to components that had not yet fully melted in some cases, had already assumed dimensions similar to those that are seen at the end of the extrusion process. This meant that, in the melting section of the twin-screw extruder, no volumes became detached from or were worn off the already-molten granule surfaces {776088}. An evaluation of SEM micrographs also showed that the degradation mechanisms, such as quasi-steady drop break-up, folding, end pinching and decomposition through capillary instabilities, took place in parallel in the melting section of co-rotating twin-screw extruders. Researchers interested in free-radical grafting of glycidyl methacrylate (GMA) onto PP and PE demonstrated that, when a monomer is to be grafted onto a polymer backbone by a free-radical mechanism in a twin-screw extruder, the grafting process occurs mainly, if not exclusively, in the plastification (melting) zone [a.412]. A co-rotating self-wiping twin-screw extruder was used, followed by adjustments of the position and length of the plastification zone, which allowed follow-up of the grafting not only at the die exit, but also in the plastification zone under different grafting conditions. The results showed that it is indeed in the plastification zone that the entire grafting process occurred. The HDPE powder melted before the kneading block because of the heat transfer from the 200 °C barrel, whereas the porous PP pellets were only partially molten before the kneading section. Further, plastification was more sensitive to screw speed than to the feed rate. Based on this work, any relevant analysis or model of a free-radical grafting process of polyolefins carried out in a screw extruder must be based on detailed information generated not only at the die exit, but also and most importantly in the plastification zone {825385} {760259}.

236

Thermal Degradation During Processing of Polymers Chen and co-workers {546565} used their results for an interpretation of the difference in the behaviour of PP during processing. The chain branching and crosslinking reactions were attributed to the addition of alkyl radicals to vinyl groups. Of course, the addition of an alkyl radical to a –C=C– bond is energetically favoured; however, the rate and regioselectivity of this reaction is, to a large extent, determined by steric factors. This fact explained why the methyl-substituted CH2=C(CH3)– bond, which is present or is formed on processing PP, does not participate in alkyl radical addition reactions. For these reasons, PP is essentially prone to chain scissions unlike other polymers such as PE, which have a pronounced tendency to molecular enlargement and crosslinking.

11.3 Poly(Vinyl Alcohol) Poly(vinyl alcohol) is an interesting water-soluble synthetic polymer with a broad range of applications. Mainly it is used as a sizing agent or stabiliser of dispersion systems. Owing to its solubility and biodegradability, PVA films are increasingly used as packaging materials. Two technologies are mainly used for PVA film production – casting from viscous water solution or blow extrusion from the melt. For packaging films, blow extrusion is the most frequently used technology despite the relatively high sensitivity of PVA to thermal degradation. Therefore, an increase of thermal stability of the melt during processing is an essential aspect that can positively influence the blowing technology of water-soluble PVA films. A considerable amount of research work in recent times has been dedicated to improving PVA processing stability [a.413, a.414]. Studies on the thermal and processing degradation behaviour noted that the pH value of the water solution could significantly affect the solubility of PVA film. Besides casting from water solutions, PVA processing from the melt seems to be more suitable, particularly for water-soluble films. Processing at high temperatures is very sensitive to PVA degradation, and problems with stability of the melt have resulted in a search for suitable additives such as silica [a.414]. A patented work claims extrudable PVA composition with improved thermal stability [a.413]. According to this patent, better stability of PVA blends was obtained after adding a specific quantity of mineral acid (preferably phosphoric acid). Alexy et al. [a.417] presented another ive report on the thermal and processing degradation behaviour of PVA and thermooxidative degradation of PVA films. The results obtained from TG analysis of PVA films confirmed that the thermal stability of PVA really depends on the pH value of the solution in which it is pre-treated. This finding is important particularly for the PVA production process because the basic hydrolysis reaction is usually stopped with acetic acid addition and PVA is separated from the formed suspension. Therefore, the pH value of the suspension or solution before polymer drying will influence its thermal stability. Using

237

Thermal Degradation of Polymeric Materials FTIR spectroscopy, the rate of increase of the concentration of degradation products was monitored as the parameter determining polymer degradation. Significantly better stability was expressly confirmed by the statistical analysis of the results for the film containing Mg(OH)2 as thermal stabiliser. The presented results confirmed that a good stabiliser must effectively eliminate acidic compounds in PVA melt to avoid degradation reactions.

11.4 Other Polymers Extrusion processing is used to process starch and other polymers in both the food and plastics industries. During extrusion, starch degradation is dependent on the processing parameters. Because the extent of degradation influences the properties of the final product, an understanding of degradation during extrusion is needed. It has been found that the degradation of starch is controlled by the cumulative amount of mechanical energy imparted during extrusion processing [a.416]. A mathematical relationship was found between the extent of degradation and the mechanical energy input through two extrusion es. The results provide insight into polymer degradation during extrusion processing, an area of interest to many polymer and food scientists. Shear stress versus shear rate and wall shear stress versus slip velocity relations of polymer melts are key material property functions needed in the design of polymer processing equipment. These properties are normally measured using capillary or slit viscometers. The presence of the slit greatly complicates the processing of the resulting viscometry data. A description of the conversion of the viscometry data of a number of PVC melts through slit dies with different gaps into these two material property functions has recently been provided [a.413]. The conversion procedure, based on Tikhonov regularisation, has the advantage that it is independent of the rheological constitutive equation and does not require the extrapolation of experimental data. It also has the ability to cope with the unavoidable noise in the experimental data. Consequently, the property functions thus obtained are likely to be closer to true material properties than those from some of the existing methods {793817}. Tate and Narusawa {592654} studied the thermal degradation and melt viscosity of ultrahigh-molecular-weight PET (UHMW-PET) with an intrinsic viscosity exceeding 2 dL/g in the melt phase under a nitrogen atmosphere. It was found that the degradation rate increased with the molecular weight of the polymer, which the researchers interpreted as a result of the differences in the terminal-endgroup concentrations. The activation energy for the degradation of UHMW-PET with an intrinsic viscosity of 2.3 dL/g was ca. 167 kJ/mol. The melt viscosity of UHMW-PET with Mw = 2.3 × 105 at 300 °C under zero shear was 8 × 105 P, signifying a reduction of molecular weight from 2 to 1.1 dL/g. As confirmed in this work, UHMW-PET processing must be conducted at low temperatures

238

Thermal Degradation During Processing of Polymers and with as short times as possible to maintain a high retention of the molecular weight during the treatment. Epoxy-amine thermosets/thermoplastic blends, precursors of porous polymer matrices, were prepared from initially homogeneous solutions of thermoset precursors and a thermoplastic polymer, poly(vinyl methyl ether) (PVME) [a.418]. The process involved the polymerisationinduced phase separation of PVME followed by thermal degradation (extraction) of PVME at high temperature. PVME was confirmed to be only partially extractible by thermal degradation. According to the thermoplastic content, different morphologies with sub-micrometre sized voids (<200 nm) were stabilised – either dispersed closed cell or bicontinuous open-cell morphologies. The critical concentration was estimated to be around 10 wt% of thermoplastic. Even in the region of spinodal demixing, the observed final morphology was spherical whenever too little thermoplastic was present (e.g., 10–15 wt%). From 20 to 25 wt% of thermoplastic, the morphologies were clearly bicontinuous. Concerning processing, measurable amounts of polymer degradation occur during the fabrication of objects from polymer-coated ceramic powders by selective laser sintering [a.418]. It was suggested that, because the binder is important in achieving strong green parts that can be handled with minimal breakage during post-processing operations, it is essential to minimise the extent of binder losses. As the first step towards understanding the mechanisms of binder degradation, these researchers developed a thermal model of the physical system and used the model to determine the most influential parameters affecting binder losses during fabrication from polymer-coated powders. It was predicted that adjustments to laser beam diameter, laser scanning distance and gaseous environment strongly affect polymer binder degradation during processing. This work went further and predicted that polymer degradation during selective laser sintering processing is not sensitive to the inherent degradation kinetics of the polymer. The high crystallinity of rigid fibres sets them apart from the usual fibrous precursors of low or intermediate crystallinity. In this case, the fibres are normally transformed to activated carbon fibres by a two-step process, i.e., pyrolysis and (physical) activation. In consequence, the study of the thermal transformation of aramid fibres has relevance to, on the one hand, understanding the mechanisms by which the fibre degrades and, on the other, being able to select the most advantageous conditions to prepare activated carbon fibres {838693}. In the production of fibres in a melt-spinning process, the transfer system between the melt source and the spinning position has a special importance due to the thermal degradation of the polymer. A recent invention has addressed this issue for polyester fibres [a.419]. The process is characterised by employment of two or more indexes whereby transfer elements and polymer flows in the transfer system may be correlated for minimum degradation of the polymer. The invention is of particular importance in a continuous process involving ester polycondensation coupled with melt spinning.

239


240

Modelling of Thermal Degradation Processes

12

Modelling of Thermal Degradation Processes

Computer-aided chemistry can be used to predict and gain insight into the thermal degradation mechanisms of polymers in a fraction of the time it would take to perform the necessary experiments {879069} {865095} {862952} [a.117]. The computer-aided chemistry approach also often provides new insights into the mechanism of thermal decomposition and the formation of breakdown products that are unavailable by experimental techniques {793664}. Modelling the thermal degradation of a material is thus an important task for the understanding of the mechanisms occurring in the condensed phase {783421}. This ability to explore the relationship between a molecule’s structure and its chemical and physical properties allows for a more systematic approach to the design of new processes and polymers with more desirable properties. Modelling all of the reactions would be necessary in order to simulate product distributions, but predicting thermal stability, in of the onset temperature observed in a TG experiment for instance, does not require simulating the entire degradation process. It is a matter of whether or not thermal degradation will start at a given temperature. Since none of the free-radical reactions can take place until after the bond homolysis step, the initiation reaction is the key to limiting thermal stability [a.11]. However, although numerical simulation based on degradation chemistry has some advantages, the major drawback of this technique is that it is time-consuming and requires relatively fast computers and a large amount of computer memory unless smartly implemented. In addition, implementation of this technique requires information associated with the endothermic or exothermic decomposition reactions. Most mathematical models developed for investigating polymer degradation generally consider the average properties of the polymer chain-length distribution or MWD. Population balance equations are often employed in fragmentation models to describe how the frequency distributions of different-sized entities, both parent and progeny, evolve. The advantage of these models is that they provide straightforward procedures to derive expressions for monomers of the frequency distributions. The MWD is a partial record of kinetics and its evolution mechanism; some population models are solved directly from the distribution, but more often the moments are computed and then utilised to construct

241

Thermal Degradation of Polymeric Materials the distribution, as discussed by Laurence and co-workers [a.420] and Dotson and coworkers [a.421] for polymerisation. Population balance equations are generally written for discrete or continuous MWDs [a.422]. Continuous kinetics is valid when the MWD allows integrals to represent averages of the distribution. This approach provides governing integro-differential equations that can be directly solved by the moment method [a.423]. Results from thermal degradation behaviour of polymers using the population model showed that pyrolysis time only influences the maximum peak of the distribution curve [a.422, a.423]. The MWDs calculated using three adjustable parameters agreed well with the experimental GPC data. The degree of polymer chain scission was also efficiently represented using the G value defined as the number of radiolysis events caused by the absorption of 100 eV of radiation. An attempt to calculate Arrhenius parameters for the thermal degradation of polyethylene from the fraction of bonds broken instead of the use of the mass conversion or weight loss of the sample has been made [a.55]. The fraction of bonds broken in the thermal degradation of polyethylene was estimated and the activation energy of the degradation was then calculated from the fraction of bonds broken. The work found that the activation energy thus obtained is close to the values for C–C bond scissions and C–H bond scissions of hydrocarbon radicals. The C–H bond scission of radicals was proposed as part of the mechanism of thermal degradation of PE, which provided a detailed understanding of hydrogen transfer in the polymer melt and paved the way for a route to hydrogen gas formation in the pyrolysis of polyethylene. Other works concluded that both the dynamic and isothermal degradation of polypropylene are unlikely to be described by a first-order reaction model {887512}. The appropriate reaction order should be determined according to the max value observed in dynamic measurements. The validity of the thus determined reaction order of n = 0.35 for the dynamic degradation was verified by the similarity of activation energy values in a comparison of the single heating rate plot with the isoconversional plot. The isothermal degradation of the PP was described by reaction order n = 0.35, while the activation energy and pre-exponential factor of the isothermal degradation were reported to be similar to those measured in the dynamic degradation. Thermal degradation modelling of the PP/PA-6/ammonium polyphosphate (APP)/EVA intumescent blend showed that the first step of degradation (about 20 wt%) occurs at a low rate at about 300 °C, with APP/PA-6 blend beginning to degrade at a temperature ca. 50 °C lower than virgin PA-6 [a.424, a.425] {451119}. This was attributed to the attack induced by the phosphoric acid species on the alkylamide bonds of PA-6, formed from the degradation of APP, that led to the formation of phosphoric ester and primary amide chain ends and then elimination of -caprolactam. A material (about 25 wt%) presenting a low degradation rate between 400 and 500 °C was then formed – its degradation under

242

Modelling of Thermal Degradation Processes nitrogen around 550 °C led to the formation of a stable 10 wt% residue. The major weight loss took place above 500 °C, causing the formation of a carbonaceous residue with a low degradation rate. It was assumed that the PP degradation and the reaction between APP and PA-6 occurred simultaneously in the considered temperature range. Moreover, the calculated activation energy associated with this step of the pyrolysis was higher than that associated with the second step of the thermooxidative degradation of the material. Thus, it was considered that the development of the intumescent shield, which occurred during the considered steps, was made easier by the presence of oxygen. Indeed, the material had to degrade rapidly and easily to form the material presenting the barrier properties. Quantitative analysis of the thermal degradation of poly(phenylene ether) (PPE) has been attempted by a computer simulation model named the ‘Molic mouse model’ {783960}. The model was designed to simulate individual steps, each corresponding to a chemical reaction in the thermal degradation. The rearrangement reaction of the main chain, the cleavage reactions of ether and methylene bridges between benzene rings, and the elimination of a methyl groups from side chains were selected. Each reaction was assumed to occur if a generated random number was greater than its probability, which had been determined by fitting the distribution of monomeric scission products in a prior simulation. The results were verified by predicting the distribution of dimeric scission products, whereby the rearrangement reaction proceeded prior to the thermal degradation. As a result, the ratio of the two bridges (ether/methylene) was 22/78 at the thermal degradation temperature. Once the thermal degradation temperature was reached, the cleavage of the main chain took precedence over that of the side chain by 20–70 times. The possible side-chain cleavages occurred simultaneously and caused the generation of many kinds of cleavage products. Denq and co-workers developed a parallel competitive reaction model based on the assumption that the rate constant at any weight-loss fraction is approximately equal to the rate constant of its neighbouring weight-loss fraction, which s for the type of bond scission and the state of scission of the polymeric chain at any time. A dynamic model based on a modified power law that s for the thermal degradation of the polymer at any time has recently been proposed. The method was applied to predict the thermal degradation of HDPE, LDPE and linear LDPE by a conventional non-isothermal thermogravimetry technique at several heating rates between 10 and 50 °C/min. By using this method, the apparent activation energies of the thermal degradation of HDPE, LDPE and linear LDPE were calculated to be 330–345, 185–200 and 220–230 kJ/mol and the reaction order increased with the extent of branching {769831}. The degradation of chemically pure PMMA in nitrogen and oxygenated environments concluded that mass-loss rate measurements in pure nitrogen can be modelled as a threestep reaction – the first and second steps were minor steps and the third was a major one [a.4, a.426] {893911}. As oxygen fraction increased to 5% O2 in N2, the degradation 243

Thermal Degradation of Polymeric Materials was described as three steps, though the presence of oxygen shifted the first step to higher temperature, and also caused the major second step to be less stable. A third step was a minor one, commencing at ca. 440 °C amd attributed to char oxidation. Further, increase in the concentration of oxygen, to 10 and 21% O2 in N2, showed a big impact on the degradation in that the first step appeared to be fixed at 255 °C; therefore, there was no effect of oxygen in of initiation of the degradation. For the second step, the presence of oxygen caused the degradation to be less stable. Shear has uncertain effects on the processes of thermal degradation: it may have no effect on the processes of thermal degradation, it may accelerate some or all of the processes, it may decelerate some or all of the processes, or it may affect some processes and not others. Betso and co-workers [a.427] investigated the shear-related thermal degradation of vinylidene chloride/vinyl chloride (VDC/VC) copolymer in air. Thermal degradation is characterised by time-dependent molecular-weight change, mainly resulting from chain scission and crosslinking reactions. However, the proposed kinetic mathematical model was limited for shear-related thermal degradation of polymers in air because shear stress has uncertain effects on the degradation process. Nevertheless, the results showed that the shear-related degradation of VDC/VC copolymer in air is characterised by early predominant chain scission with crosslinking and later predominant crosslinking with chain scission. Both chain scission and crosslinking were dependent on the shear rate and temperature. Furthermore, the shear-stress dependence of degradation was modelled by a kinetic expression that incorporated shear stress into the Arrhenius pre-exponential factor. However, the effects of shear rate and temperature on the degradation process were difficult to evaluate directly. Consequently, the effects of shear stress on degradation had to be investigated under constant shear rate and temperature. In advancements on Betsos and co-workers’ mathematical model, a method has been presented recently to model shear-related thermal degradation of a VDC/VC copolymer based on the artificial neural networks (ANN) approach [a.428]. The method involved the back-propagation ANN that is adopted to predict both the number-average molecular weight and the weight-average molecular weight of VDC/VC copolymer during shearrelated thermal degradation in air. The degradation of PS has been modelled at the mechanistic level using the method of moments to track structurally distinct polymer species {889529}. To keep the model size manageable, polymer species were lumped into groups, and within these groups, the necessary polymeric features for capturing the degradation chemistry were searched. The pyrolysis reactions incorporated into the model included hydrogen abstraction, midand end-chain -scission, 1,5-hydrogen-transfer, 1,3-hydrogen-transfer, radical addition, bond fission, radical recombination and disproportionation. From the evolution of the zeroth, first and second moments tracked for each dead species, polymer molecular-

244

Modelling of Thermal Degradation Processes weight distributions were constructed by summing the Schultz and Wesslau distributions for the polymer groups. Model results were compared to experimental data where PS that differed in the shape and breadth of their initial distributions were pyrolysed. The model was able to predict the formation of a bimodal distribution during the pyrolysis of PS (MW range of 10,000–500,000) with narrow unimodal MW distributions (polydispersity index < 1.1). This was accomplished by distinguishing the initial polymer from the polymer formed from midchain -scission reactions within the model. At high conversions, the PS investigated evolved to unimodal distributions, which were best described by the Schultz function. A mathematical model to describe the molecular weight and polydispersity index in PLLA thermal degradation has been developed {756177}. Based on the random chain scission mechanism, effects of temperature and time on the molecular weight and polydispersity index were included in the model. It incorporated the degradation and recombination reactions of PLLA during thermal degradation, while taking into the equalprobability assumption. The developments of molecular weight and polydispersity index of PLLA polymer in the thermal degradation process were investigated at temperatures ranging from 180 to 220 °C – the experimental data showed that PLLA reaches its thermal degradation equilibrium in 2 h. The simulated results of this model were compared with the measured molecular weight and polydispersity index of the PLLA polymer – they agreed satisfactorily. Research focusing on the application of molecular modelling techniques to identify factors that affect the thermal degradation chemistry of polymers in ways that result in a reduction in their flammability culminated in the development of a novel computer program, MD_REACT (molecular dynamics reactions) [a.429]. The basis of this model is molecular dynamics, which involves solving equations of motion for the 3N (where N is the number of atoms in the polymer) degrees of freedom associated with the model polymer {820329} {687814} {609091}. The forces were obtained as the negative gradient of a potential energy function that describes the variation of the molecular energy with changes in the internal degrees of freedom (i.e., bond distances and angles). The feature that distinguishes MD_REACT from other molecular dynamics codes is that it allows for the formation of new bonds from free-radical fragments that are generated when bonds in the polymer break and, thereby, s for the chemical reactions that play a major role in the thermal degradation process of polymers. Most recently a governing set of differential equations was derived for individual linear polymers undergoing scission at a random location, and were used to determine how the molecular-weight distribution evolves in time {699495} [a.430]. With the proposed model it was possible to investigate other cases such as -scission using Monte Carlo simulation by which the obtained results showed a good correlation with the experimental results.

245

Thermal Degradation of Polymeric Materials It was suggested that the proposed model may be used to predict rates of mass loss for a vaporising polymer and also that the model may be extended to investigate other bondbreaking processes such as simultaneous random and end-chain scission, or the inclusion of recombination during degradation.

246

Concluding Remarks

13

Concluding Remarks

Thermal degradation of polymeric materials is an important issue from both the academic and the industrial points of view. The analysis of the degradation process has become more and more important as a result of an increase in the range of temperatures for engineering applications, recycling of post-consumer plastic waste, as well as the use of polymers as biological implants and matrices for drug delivery, where depolymerisation is an inevitable process affecting the lifetime of an article. Thermal degradation of polymers is therefore of paramount importance in developing a rational technology of polymer processing, highertemperature polymer applications, polymer usability, storage and recycling, in addition to understanding the thermal decomposition kinetics and mechanisms for optimum synthesis of long-lasting fire-safe polymeric materials. Unfortunately, thermal degradation is likely to be responsible for serious damage to any polymeric material and this effect is especially important for recycled polymers, as they suffer successive cycles of high and low temperatures. Controlling degradation requires understanding of many different phenomena, including the diverse chemical mechanisms underlying structural changes in macromolecules, the influence of polymer morphology, the complexities of oxidation chemistry, the intricate reaction pathways of stabiliser additives, the interaction of fillers and other additives together with impurities, and the reaction–diffusion processes that often take place. Furthermore, there exist substantial differences between pure and industrial polymers that may have detrimental effects on the thermal degradation of polymeric materials, and this increases the complexities of the thermal degradation of polymers. Thus, thermal degradation is an extremely complex and important process during the processing, use, storage, application and recycling of polymers. This work covers in depth the recent developments in thermal degradation of synthetic polymers, copolymers, natural polymers, inorganic polymers and their respective blends, (nano)composites and high-performance plastics. However, due to the large scope of the topic, this work has not dealt with compounded polymers or catalysed thermal degradation of polymers as they are beyond the current scope. The kinetics of thermal degradation and thermooxidative degradation are also only highlighted, and an exhaustive review is not provided. Nevertheless, thermal degradation of polymers is covered intrinsically. 247


248

References

14

References

[a.1]

N.S. Allen and M. Edge, Fundamentals of Polymer Degradation and Stabilisation, Elsevier Applied Science, London, 1992, 1.

[a.2]

W. Schnabel, Polymer Degradation – Principles and Practical Applications, Akademie-Verlag, Berlin, 1981, 13.

[a.3]

R. Navarro, L. Torreb, J.M. Kenny and A. Jimenez, Polymer Degradation and Stability, 2003, 82, 2, 279.

[a.4]

S.M. Dakka, Journal of Thermal Analysis and Calorimetry, 2003, 74, 3, 729.

[a.5]

S. Sato, T. Murakata, S. Baba, Y. Saito and S. Watanabe, Journal of Applied Polymer Science, 1990, 40, 11–12, 2065.

[a.6]

J. Pospisil, Z. Horak, Z. Krulis, S. Nespurek and S. Kuroda, Polymer Degradation and Stability, 1999, 65, 3, 405.

[a.7]

D.M. Price, M. Reading, A. Hammiche and H.M. Pollock, International Journal of Pharmaceutics, 1999, 192, 1, 85.

[a.8]

C.A. Wilkie, Polymer Degradation and Stability, 1999, 66, 3, 301.

[a.9]

Z.S. Petrovic and Z.Z. Zavargo, Journal of Applied Polymer Science, 1986, 32, 4353.

[a.10]

K.G.H. Raemaekers and J.C.J. Bart, Thermochimica Acta, 1997, 295, 1–2, 1.

[a.11]

E. Chamot, Polymer Preprints, 2001, 42, 1, 396.

[a.12]

H.L.C. Meuzelaar, J. Haverkamp and F.D. Hileman, Pyrolysis Mass Spectrometry of Recent and Fossil Biomaterials – Compendium and Atlas, Elsevier, Amsterdam, 1982, 1.

249

Thermal Degradation of Polymeric Materials [a.13]

P. Almen and I. Ericsson, Polymer Degradation and Stability, 1995, 50, 2, 223.

[a.14]

M.A. Soto-Oviedo, R.S. Lehrle, I.W. Parsons and M.A. De Paoli, Polymer Degradation and Stability, 2003, 81, 3, 463.

[a.15]

I.C. McNeill, European Polymer Journal, 1970, 6, 2, 373.

[a.16]

I.C. McNeill, A. Shafique and J.G. Gorman, Polymer Degradation and Stability, 1999, 63, 2, 265.

[a.17]

ISO standards, International Organisation for Standardisation, http://www.iso.org

[a.18]

H. Bair, in Thermal Characterization of Polymeric Materials, 2nd Edn., Ed. E.A. Turi, Academic Press, San Diego, 1997, 2276.

[a.19]

M. Groning and M. Hakkarainen, Journal of Chromatography A, 2001, 932, 1–2, 1.

[a.20]

V.B.F. Mathot (Ed.), Calorimetry and Thermal Analysis of Polymers, Hanser Publishers, Munich, 1994, 1.

[a.21]

A. Riga, R. Collins and G. Mlachak, Thermochimica Acta, 1998, 324, 1–2, 135.

[a.22]

M. Reading, Trends in Polymer Science, 1993, 1, 248.

[a.23]

J.M. Hutchinson and S. Montserrat, Thermochimica Acta, 2001, 377, 1–2, 63.

[a.24]

M. Sandor, N.A. Bailey and E. Mathiowitz, Polymer, 2002, 43, 2, 279.

[a.25]

J.E.K. Schawe and W. Winter, Thermochimica Acta, 1999, 330, 1–2, 85.

[a.26]

K. Pielichowski, K. Flejtuch and J. Pielichowski, Polymer, 2004, 45, 4, 1235.

[a.27]

G. Montaudo, M.S. Montaudo, C. Puglisi and F. Samperi, Rapid Communications in Mass Spectrometry, 1998, 12, 1, 1.

[a.28]

R.A. Doong and P.L. Liao, Journal of Chromatography A, 2001, 918, 1, 177.

[a.29]

T.M. Alam, M. Celina, R.A. Assink, R.L. Clough and K.T. Gillen, Radiation Physics and Chemistry, 2001, 60, 1–2, 121.

250

References [a.30]

M. Lucarini, G.F. Pedulli, M.V. Motyakin and S. Schlick, Progress in Polymer Science, 2003, 28, 2, 331.

[a.31]

M.V. Motyakin and S. Schlick, Macromolecules, 2001, 34, 9, 2854.

[a.32]

K. Pielichowski, Solid State Ionics, 1997, 104, 1–2, 123.

[a.33]

N.M. Emanuel and A.L. Buchachenko, Chemical Physics of Polymer Degradation and Stabilization, VNU Science Press, Utrecht, 1987, 10.

[a.34]

F. Ramsteiner, G. Kanig, W. Heckmann and W. Gruber, Polymer, 1983, 24, 3, 365.

[a.35]

P.C. Cheung and S.T. Balke, Industrial and Engineering Chemistry Research, 1997, 36, 4, 1191.

[a.36]

D. Suwanda and S.T. Balke, Polymer Engineering and Science, 1993, 33, 24, 1585.

[a.37]

M. Veronelli, M. Mauro and S. Bresadola, Polymer Degradation and Stability, 1999, 66, 3, 349.

[a.38]

G. Allen, J.C. Bevington, C. Booth and C. Price, Comprehensive Polymer Science, Vol. 6, 1st Edn., Pergamon, Oxford, 1989, 1.

[a.39]

M.T. Benaniba, N. Belhaneche-Bensemra and G. Gelbard, Polymer Degradation and Stability, 2001, 74, 3, 501.

[a.40]

C.L. Arthur and J. Pawliszyn, Analytical Chemistry, 1990, 62, 19, 2145.

[a.41]

G. Madras and S. Chattopadhyay, Chemical Engineering Science, 2001, 56, 17, 5085.

[a.42]

S.M. Reshetnikov and I.S. Reshetnikov, Polymer Degradation and Stability, 1999, 64, 3, 379.

[a.43]

G. Madras and V. Karmore, Polymer Degradation and Stability, 2001, 72, 3, 537.

[a.44]

Y. Kodera and B.J. McCoy, AIChE Journal, 1997, 43, 6, 3205.

[a.45]

J.M. Creado, F.J. Gotor, C. Real, F. Jimenez, S. Ramos and J. DelCerro, Ferroelectrics, 1991, 115, 43.

251


S. Bordere, F. Rouquerol, J. Rouquerol, J. Esfienne and A. Floreancig, Journal of Thermal Analysis, 1990, 36, 1651.

[a.47]

E. Inoue, M. Tsuchiya, K. Ishimaru and T. Kojima, Journal of Thermal Analysis and Calorimetry, 2002, 70, 3, 747.

[a.48]

C.D. Doyle, Journal of Applied Polymer Science, 1962, 6, 24, 639.

[a.49]

T. Ozawa, Bulletin of the Chemical Society of Japan, 1965, 38, 1881.

[a.50]

J.H. Flynn and L.A. Wall, Polymer Letters, 1966, 4, 232.

[a.51]

H.L. Friedman, Journal of Polymer Science, 1965, C6, 175.

[a.52]

J. Opfermann and E. Kaisersberger, Thermochimica Acta, 1992, 203, 167.

[a.53]

M.L. Poutsma, Macromolecules, 2003, 36, 24, 8931.

[a.54]

P.T. Williams and E.A. Williams, Journal of Analytical and Applied Pyrolysis, 1999, 51, 1–2, 107.

[a.55]

Z. Gao, I. Amasaki, T. Kaneko and M. Nakada, Polymer Degradation and Stability, 2003, 81, 1, 125.

[a.56]

K. Kruczala, J.G. Bokria and S. Schlick, Macromolecules, 2003, 36, 6, 1909.

[a.57]

T. Andersson, B. Wesslen and J. Sandstrom, Journal of Applied Polymer Science, 2002, 86, 7, 1580.

[a.58]

S.H. Hamid (Ed.), Handbook of Polymer Degradation, 2nd Edn., Marcel Dekker, New York, 2000, 266.

[a.59]

R.S. Lehrle, R. Duncan, Y. Liu, I.W. Parsons, M. Rollinson, G. Lamb and D. Barr, Journal of Analytical and Applied Pyrolysis, 2002, 64, 2, 207.

[a.60]

C. Liu, J. Yu, X. Sun, J. Zhang and J. He, Polymer Degradation and Stability, 2003, 81, 2, 197.

[a.61]

K.S. Chen, R.Z. Yeh and Y.R. Chang, Combustion and Flame, 1997, 108, 4, 408.

[a.62]

J. Hacaloglu, T. Ersen, N. Ertugrul, S. Suzer and M.M. Fares, European Polymer Journal, 1997, 33, 2, 199.

252

References [a.63]

I.C. McNeill and T.K. Stevenson, Polymer Degradation and Stability, 1985, 10, 4, 319.

[a.64]

Y. Park, J.N. Hool, C.W. Curtis and C.B. Roberts, Industrial and Engineering Chemistry Research, 2001, 40, 3, 756.

[a.65]

K.V. Alekseeva, Journal of Analytical and Applied Pyrolysis, 1980, 2, 19.

[a.66]

D.O. Hummel, H.J. Dussel, G. Czybulka, N. Wenzel and G. Holl, Spectrochimica Acta A, 1985, 41, 1–2, 279.

[a.67]

J.A. Hiltz, Journal of Analytical and Applied Pyrolysis, 2000, 55, 2, 135.

[a.68]

C. Gamlin, N. Dutta, N.R. Choudhury, D. Kehoe and J. Matisons, Thermochimica Acta, 2001, 367–368, 185.

[a.69]

V. Dubey, S.K. Pandey and N.B. Rao, Journal of Analytical and Applied Pyrolysis, 1995, 34, 2, 111.

[a.70]

K. Pielichowski and L. Stoch, Journal of Thermal Analysis, 1994, 45, 5, 1239.

[a.71]

Y. Ito, H. Ogasawara, Y. Ishida, H. Ohtani and S. Tsuge, Polymer Journal, 1996, 28, 1090.

[a.72]

B.A. Howell, Y. Cui and D.B. Priddy, Thermochimica Acta, 2003, 396, 1–2, 167.

[a.73]

G. Madras, J.M. Smith and B.J. McCoy, Industrial and Engineering Chemistry Research, 1996, 35, 6, 1795.

[a.74]

G. Lisa, E. Avram, G. Paduraru, M. Irimia, N. Hurduc and N. Aelenei, Polymer Degradation and Stability, 2003, 82, 1, 73.

[a.75]

V.V. Zuev, F. Bertini and G. Audisio, Polymer Degradation and Stability, 2000, 69, 2, 169.

[a.76]

K. Pielichowski, A. Puszynski and J. Pielichowski, Polymer Journal, 1994, 26, 7, 822.

[a.77]

M. Switala-Zeliazkow, Polymer Degradation and Stability, 2001, 74, 3, 579.

[a.78]

M.J. Fernandez and M.D. Fernandez, Polymer Degradation and Stability, 1998, 61, 1, 165.

253


M.H. Yang, Polymer Degradation and Stability, 2002, 76, 69.

[a.80]

M.H. Yang, Polymer Degradation and Stability, 2000, 68, 3, 451.

[a.81]

J. Hacaloglu, M.M. Fares and S. Suzer, European Polymer Journal, 1999, 35, 5, 939.

[a.82]

M.M. Fares, T. Yalcin, J. Hacaloglu, A. Gongor and S. Suzer, Analyst, 1994, 119, 4, 693.

[a.83]

O.S. Woo, T.M. Kruse and L.J. Broadbelt, Polymer Degradation and Stability, 2000, 70, 2, 155.

[a.84]

D. Braun, Progress in Polymer Science, 2002, 27, 10, 2171.

[a.85]

N. Grassie and H.W. Melville, Faraday Society Discussions, 1947, 2, 378.

[a.86]

M. Suzuki and C.A. Wilkie, Polymer Degradation and Stability, 1995, 47, 2, 217.

[a.87]

M. Suzuki and C. Wilkie, Polymer Degradation and Stability, 1995, 47, 2, 223.

[a.88]

M. Day, J.D. Cooney, C. Touchette-Barrette and S.E. Sheehan, Journal of Analytical and Applied Pyrolysis, 1999, 52, 2, 199.

[a.89]

T.J. Xue, M.A. McKinney and C. Wilkie, Polymer Degradation and Stability, 1997, 58, 1–2, 193.

[a.90]

M. Brebu, M.A. Uddin, A. Muto, Y. Sakata and C. Vasile, Energy and Fuels, 2000, 14, 4, 920.

[a.91]

B.E. Tiganis, L.S. Burn, P. Davis and A.J. Hill, Polymer Degradation and Stability, 2002, 76, 3, 425.

[a.92]

N. Grassie, Developments in Polymer Degradation, Vol. 4, Applied Science, London, 1982, 1.

[a.93]

S. Stack, O. O’Donoghue and C. Birkinshaw, Polymer Degradation and Stability, 2003, 79, 1, 29.

[a.94]

J. Jachowicz and M. Kryszewski, Polymer, 1978, 19, 1, 93.

[a.95]

D.M. Bate and R.S. Lehrle, Polymer Degradation and Stability, 1997, 55, 2, 295.

254

References [a.96]

I.C. McNeill, L. Memetea and W.J. Cole, Polymer Degradation and Stability, 1995, 49, 1, 181.

[a.97]

G.M. Anthony, Polymer Degradation and Stability, 1999, 64, 3, 353.

[a.98]

G. Montaudo and C. Puglisi, Polymer Degradation and Stability, 1991, 33, 2, 229.

[a.99]

J. Chambers, J. Jiricny and C. Reese, Fire Materials, 1981, 5, 133.

[a.100] N. Dadvand, R.S. Lehrle, I.W. Parsons and M. Rollinson, Polymer Degradation and Stability, 1999, 66, 2, 247. [a.101] S. Chattopadhyay and G. Madras, Polymer Degradation and Stability, 2002, 78, 3, 519. [a.102] S. Zulfiqar and S. Ahmad, Polymer Degradation and Stability, 1999, 65, 2, 243. [a.103] I.C. McNeill and S. Bassan, Polymer Degradation and Stability, 1993, 39, 2, 145. [a.104] M.C.S. Perera, U.S. Ishiaku and Z.A.M. Ishak, European Polymer Journal, 2001, 37, 1, 167. [a.105] M.C.S. Perera, U.S. Ishiaku and Z.A.M. Ishak, Polymer Degradation and Stability, 2000, 68, 3, 393. [a.106] N.A. Mohamed and M.W. Sabaa, European Polymer Journal, 1999, 35, 9, 1731. [a.107] K. Pielichowski, Journal of Thermal Analysis and Calorimetry, 1998, 54, 1, 171. [a.108] K. Pielichowski and I. Hamerton, European Polymer Journal, 2000, 36, 1, 171. [a.109] D. Braun, B. Boehringer, W. Knoll, N. Eidam and W. Mao, Angewandte Makromolekulare Chemie, 1990, 181, 23. [a.110] I.C. McNeill, in Comprehensive Polymer Science, Vol. 6, Eds. G.C. Eastmond, A. Ledwith, S. Russo and P. Sigwalt, Pergamon Press, Oxford, 1989, 451. [a.111] J. Huang, D.C. Chang and R.D. Deanin, Advances in Polymer Technology, 1993, 12, 1, 81. [a.112] I.C. McNeill, M. Zulfiqar and T. Kousar, Polymer Degradation and Stability, 1990, 28, 2, 131.

255

Thermal Degradation of Polymeric Materials [a.113] T. Kovacic, I. Klaric, A. Nardelli and B. Baric, Polymer Degradation and Stability, 1993, 40, 1, 91. [a.114] H. Bockhorn, A. Hornung, U. Hornung and J. Weichmann, Thermochimica Acta, 1999, 337, 1–2, 97. [a.115] M. Herrera, G. Matuschek and A. Kettrup, Chemosphere, 2001, 42, 5–7, 601. [a.116] P. Gijsman, R. Steenbakkers, C. Furst and J. Kersjes, Polymer Degradation and Stability, 2002, 78, 2, 219. [a.117] D.A. Gallagher, Scientific Computing and Automation, June 1996. [a.118] M. Vera, A. Almontassir, A. Rodriguez-Galan and J. Puiggali, Macromolecules, 2003, 36, 26, 9784. [a.119] I. Villuendas, I. Molina, C. Reganó, M. Bueno, A. Martinez de Ilarduya, J. Galbis and S. Munóz -Guerra, Macromolecules, 1999, 32, 24, 8033. [a.120] K.P. Pramoda, T.S. Chung, S.L. Liu, H. Oikawa and A. Yamaguchi, Polymer Degradation and Stability, 2000, 67, 3, 365. [a.121] I. Vieira, V.L.S. Severgnini, D.J. Mazera, M.S. Soldi, E.A. Pinheiro, A.T.N. Pires and V. Soldi, Polymer Degradation and Stability, 2001, 74, 1, 151. [a.122] K. Pielichowski and D. Slotwinska, Thermochimica Acta, 2004, 410, 1–2, 79. [a.123] N. Grassie and M. Zulfiqar, Journal of Polymer Science A, 1978, 16, 1563. [a.124] N. Grassie and G.A. Perdomo-Mendoza, Polymer Degradation and Stability, 1985, 10, 267. [a.125] M. Herrera, G. Matuschek and A. Kettrup, Polymer Degradation and Stability, 2002, 78, 2, 323. [a.126] K. Pielichowski and D. Slotwinska, Polymer Degradation and Stability, 2003, 80, 2, 327. [a.127] K. Pielichowski, Polymer Journal, 1997, 29, 10, 848. [a.128] K. Mequanint, R. Sanderson and H. Pasch, Polymer Degradation and Stability, 2002, 77, 1, 121.

256

References [a.129] V. Sekkar, K.N. Ninan, V.N. Krishnamurty and S.R. Jain, European Polymer Journal, 2000, 36, 11, 2437. [a.130] T.C. Wen, S.L. Hung and M. Digar, Synthetic Metals, 2001, 118, 1–3, 11. [a.131] M. Kinoshita, T. Nemoto, T. Souda and K. Takeda, Polymer Degradation and Stability, 2000, 68, 3, 437. [a.132] D.C. Liao, K.H. Hsieh, Y.C. Chern and K.S. Ho, Synthetic Metals, 1997, 87, 1, 61. [a.133] R.V. Gregory and M. Liu, Synthetic Metals, 1995, 69, 1–3, 349. [a.134] C. Dick, E. Dominguez-Rosado, B. Eling, J.J. Liggat, C.I. Lindsay, S.C. Martin, M.H. Mohammed, G. Seeley and C.E. Snape, Polymer, 2001, 42, 3, 913. [a.135] T. Jeevananda and Siddaramaiah, European Polymer Journal, 2003, 39, 3, 569. [a.136] E. Dominguez-Rosado, J.J. Liggat, C.E. Snape, B. Eling and J. Pichtel, Polymer Degradation and Stability, 2002, 78, 1, 1. [a.137] N. Yoshitake and M. Furukawa, Journal of Analytical and Applied Pyrolysis, 1995, 33, 269. [a.138] R. Font, A. Fullana, J.A. Caballero, J. Candela and A. García, Journal of Analytical and Applied Pyrolysis, 2001, 58–59, 63. [a.139] N. Grittner, W. Kaminsky and G. Obst, Journal of Analytical and Applied Pyrolysis, 1993, 25, 293. [a.140] Q. Zhu, R. Guan, F. Meng and S. Feng, Thermochimica Acta, 2003, 402, 1–2, 193. [a.141] K. Pielichowski, K. Kulesza and E.M. Pearce, Journal of Applied Polymer Science, 2003, 88, 9, 2319. [a.142] J.K.W. Sandler, S. Pegel, M. Cadek, F. Gojny, M. van Es, J. Lohmar, W.J. Blau, K. Schulte, A.H. Windle and M.S.P. Shaffer, Polymer, 2004, 45, 6, 2001. [a.143] M. Ravey and E.M. Pearce, Journal of Applied Polymer Science, 1997, 63, 1, 47. [a.144] F. Gao, D. Price, G.J. Milnes, B. Eling, C.I. Lindsay and P.T. McGrail, Journal of Analytical and Applied Pyrolysis, 1997, 40–41, 217.

257

Thermal Degradation of Polymeric Materials [a.145] J. Lefebvre, B. Bastin, M.L. Bras, S. Duquesne, C. Ritter, R. Paleja and F. Poutch, Polymer Testing, 2004, 23, 3, 281. [a.146] M.M. Esperanza, A.N. Garcia, R. Font and J.A. Conesa, Journal of Analytical and Applied Pyrolysis, 1999, 52, 2, 151. [a.147] L. Katsikas, G. Boskovic, S.J. Velickovic, J.S. Velickovic and I.G. Popovic, European Polymer Journal, 2000, 36, 8, 1619. [a.148] I.G. Popovic, L. Katsikas, H. Weller, S. Schrotter and J.S. Velickovic, Journal of Applied Polymer Science, 1993, 50, 8, 1475. [a.149] M.L. Ramirez, R. Walters, R.E. Lyon and E.P. Savitski, Polymer Degradation and Stability, 2002, 78, 1, 73. [a.150] L.H. Buxbaum, Angewandte Chemie – International Edition, 1968, 7, 182. [a.151] H. Zimmermann, in Developments in Polymer Degradation, Vol. 7, Ed. N. Grassie, Applied Science, London, 1987, 35. [a.152] S. Iwabuchi, V. Jaacks and W. Kern, Makromolekulare Chemie, 1976, 177, 7, 2675. [a.153] O. Persenaire, M. Alexandre, P. Degee and P. Dubois, Biomacromolecules, 2001, 2, 1, 288. [a.154] R.A. Ruseckaite and A. Jimenez, Polymer Degradation and Stability, 2003, 81, 3, 353. [a.155] A.C. Draye, O. Persenaire, J. Brozek, J. Roda, T. Kosek and Ph. Dubois, Polymer, 2001, 42, 20, 8325. [a.156] P.M. Remiro, M.M. Cortazar, M.E. Calahorra and M.M. Calafel, Macromolecular Chemistry and Physics, 2001, 202, 7, 1077. [a.157] P.M. Remiro, M. Cortazar, E. Calahorra and M.M. Calafel, Polymer Degradation and Stability, 2002, 78, 1, 83. [a.158] J. Li, H. Xu, J. Shi, C. Li and C. Bao, Analytica Chimica Acta, 1999, 402, 311. [a.159] J.D. Peterson, S. Vyazovkin and A. Wight, Physical Chemistry B, 1999, 103, 38, 8087.

258

References [a.160] L.E. Manring, Macromolecules, 1989, 22, 6, 2673. [a.161] S.L. Madorsky, Journal of Polymer Science, 1953, 11, 491. [a.162] L.E. Manring, Macromolecules, 1988, 21, 2, 528. [a.163] H. Arisawa and T.B. Brill, Combustion and Flame, 1997, 109, 3, 415. [a.164] B.J. Holland and J.N. Hay, Polymer, 2001, 42, 11, 4825. [a.165] T. Hirata, T. Kashiwagi and J.E. Brown, Macromolecules 1985, 18, 7, 1410. [a.166] B.J. Holland and J.N. Hay, Thermochim Acta, 2002, 388, 1–2, 353. [a.167] T. Arii, S. Ichihara, H. Nakagawa and N. Fujii, Thermochimica Acta, 1998, 319, 1–2, 139. [a.168] T. Kashiwagi, A. Omori and H. Nanbu, Combustion and Flame, 1990, 81, 2, 188. [a.169] B. Zhang and F.D. Blum, Thermochimica Acta, 2003, 396, 1–2, 211. [a.170] O. Chiantore and M. Guaita, Polymer Bulletin, 1988, 20, 201. [a.171] V.K. Sharma, R.A. Pethrick and S. Affrossman, Polymer, 1982, 23, 12, 1732. [a.172] A.B. Morgan, J.M. Antonucci, M.R. Vanlandingham, R.H. Harris Jr. and T. Kashiwagi, Polymeric Materials Science and Engineering, 2000, 83, 57. [a.173] A.L. Margolin and V.Y. Shlyapintokh, Polymer Degradation and Stability, 1999, 66, 2, 279. [a.174] L. Bes, K. Huan, E. Khoshdel, M.J. Lowe, C.F. McConville and D.M Haddleton, European Polymer Journal, 2003, 39, 1, 5. [a.175] R. Vijayalakshmi Rao, P.V. Ashokan and M.H. Shridhar, Polymer Degradation and Stability, 2000, 70, 1, 11. [a.176] T.C. Chang, C.L. Liao, K.H. Wu and Y.S. Chiu, Polymer Degradation and Stability, 1999, 64, 2, 227. [a.177] S.A. Liebman, Pyrolysis and GC in Polymer Analysis, Marcel Dekker, New York, 1984, 21.

259

Thermal Degradation of Polymeric Materials [a.178] J. Pavlinec, M. Lazár and K. Csomorová, Polymer Degradation and Stability, 1997, 55, 1, 65. [a.179] M. Coskun, C. Soykan, M. Ahmedzade and K. Demirelli, Polymer Degradation and Stability, 2001, 72, 1, 69. [a.180] D.H. Grant and N. Grassie, Polymer, 1960, 1, 445. [a.181] S.L. Harley, M.L. Mittleman and C.A. Wilkie, Polymer Degradation and Stability, 1993, 39, 3, 345. [a.182] N. Grassie and J.R. MacCallum, Journal of Polymer Science, 1964, 2, 983. [a.183] C. Soykana and A. Ahmedzade, Polymer Degradation and Stability, 2002, 78, 3, 497. [a.184] M. Coskun, M.M. Temuz and K. Demirelli, Polymer Degradation and Stability, 2002, 77, 3, 371. [a.185] T. Caykara, M.S. Eroglu and O. Guven, Polymer Degradation and Stability, 1999, 63, 1, 65. [a.186] P. Bajaj, S.K. Varshney and A. Misra, Journal of Polymer Science A, 1980, 18, 295. [a.187] B. Grzyb, J. Machnikowski, J.V. Weber, A. Koch and O. Heintz, Journal of Analytical and Applied Pyrolysis, 2003, 67, 1, 77. [a.188] M. Yang, T. Tsukame, H. Saitoh and Y. Shibasaki, Polymer Degradation and Stability, 2000, 67, 3, 479. [a.189] Y. Shibasaki and M.J. Yang, Journal of Thermal Analysis, 1997, 49, 1, 71. [a.190] S.M. Badawy, A.M. Dessouki and H.M.N. El-Din, Radiation Physics and Chemistry, 2001, 61, 2, 143. [a.191] S. Zulfiqar and S. Ahmad, Polymer Degradation and Stability, 2001, 71, 2, 299. [a.192] B.J. Holland and J.N. Hay, Polymer, 2001, 42, 16, 6775. [a.193] B.A. Howell, Z. Ahmed and S.I. Ahmed, Thermochimica Acta, 2000, 357– 358, 103.

260

References [a.194] T.N. Bowmer and J.H. O’Donnell, Polymer Degradation and Stability, 1981, 3, 2, 87. [a.195] R. Daniel, S. Jardine, N. Nekula and J. Peter, Macromolecules, 1986, 19, 6, 1772. [a.196] J.H. Wang, M.H. Yang and R.J. Lee, Polymer Testing, 1997, 16, 2, 147. [a.197] M.H. Yang, A.B.O. Yang and J.H. Wang, Polymer Degradation and Stability, 2001, 73, 1, 23. [a.198] X.G. Li and M.R. Huang, Reactive and Functional Polymers, 1999, 42, 1, 59. [a.199] B. Nandan, L.D. Kandpal and G.N. Mathur, European Polymer Journal, 2003, 39, 1, 193. [a.200] S. Sundarrajan, M. Surianarayanan, K.S.V. Srinivasan and K. Kishore, Macromolecules, 2002, 35, 9, 3331. [a.201] J.M. Hutchinson, A.B. Tong and Z. Jiang, Thermochimica Acta, 1999, 335, 1–2, 27. [a.202] Y.A. Aggour, Polymer Degradation and Stability, 1996, 51, 3, 265. [a.203] J.N. Hay and D.J. Kemmish, Polymer, 1987, 28, 12, 2047. [a.204] J.M. McGuire and C.J. Bryden, Journal of Applied Polymer Science, 1988, 35, 537. [a.205] X. Zhang, J. Golding and I. Burgar, Polymer, 2002, 43, 22, 5791. [a.206] X. Zhang, M.G. Looney, D.H. Solomon and A.K. Whittaker, Polymer, 1997, 38, 23, 5835. [a.207] Majeti N.V. Ravi Kumar, Reactive and Functional Polymers, 2000, 46, 1, 1. [a.208] X. Qu, A. Wirsen and A.C. Albertsson, Polymer, 2000, 41, 13, 4841. [a.209] F.A.A. Tirkistani, Polymer Degradation and Stability, 1998, 61, 1, 161. [a.210] F.A.A. Tirkistani, Polymer Degradation and Stability, 1998, 60, 1, 67. [a.211] M. Bengisu and E. Yilmaz, Carbohydrate Polymers, 2002, 50, 2, 165.

261

Thermal Degradation of Polymeric Materials [a.212] D. de Britto and S. P. Campana-Filho, Polymer Degradation and Stability, 2004, 84, 2, 353. [a.213] H.K. Holme, H. Foros, H. Pettersen, M. Dornish and O. Smidsrod, Carbohydrate Polymers, 2002, 46, 3, 287. [a.214] K.M. Vårum, H.K. Holme, M. Izume, B.T. Stokke and O. Smidsrod, Biochimica et Biophysica Acta, 1996, 1291, 1, 5. [a.215] Y. Matsuzawa, M. Ayabe and J. Nishino, Polymer Degradation and Stability, 2001, 71, 3, 435. [a.216] F. Trotta, M. Zanetti and G. Camino, Polymer Degradation and Stability, 2000, 69, 3, 373. [a.217] A.M. Emsley, Polymer Degradation and Stability, 1994, 44, 3, 343. [a.218] S. Soares, G. Camino and S. Levchik, Polymer Degradation and Stability, 1998, 62, 1, 25. [a.219] N. Tzamtzis, A. Pappa and A. Mourikis, Polymer Degradation and Stability, 1999, 66, 1, 55. [a.220] S. Li, J. Lyons-Hart, J. Banyasz and K. Shafer, Fuel, 2001, 80, 12, 1809. [a.221] E. Jakab, G. Varhegyi and O. Faix, Journal of Analytical and Applied Pyrolysis, 2000, 56, 2, 273. [a.222] P.T. Williams and N. Nugranad, Energy, 2000, 25, 3, 493. [a.223] R. Sun, J.M. Lawther and W.B. Banks, Industrial Crops and Products, 1997, 6, 1, 1. [a.224] E. Jakab, O. Faix, F. Till and T. Székely, Journal of Analytical and Applied Pyrolysis, 1995, 35, 2, 167. [a.225] J. Li, B. Li and X.C. Zhang, Polymer Degradation and Stability, 2002, 78, 2, 279. [a.226] A.J. Anderson and E.A. Dawes, Microbiology Reviews, 1990, 54, 450. [a.227] S. Nguyen, G. Yu and R.H. Marchessault, Biomacromolecules, 2002, 3, 1, 219.

262

References [a.228] W.H. Park, R.W. Lenz and G. Goodwin, Polymer Degradation and Stability, 1999, 63, 2, 287. [a.229] W.H. Park, R.W. Lenz and S. Goodwin, Macromolecules, 1998, 31, 5, 1480. [a.230] P.L.M. Barreto, A.T.N. Pires and V. Soldi, Polymer Degradation and Stability, 2003, 79, 1, 147. [a.231] J. Grevellec, C. Marquie, L. Ferry, A. Crespy and V. Vialettes, Biomacromolecules, 2001, 2, 4, 1104. [a.232] S.S. Choi, Journal of Analytical and Applied Pyrolysis, 1999, 52, 1, 105. [a.233] J. Jin and H. Li, Journal of Analytical and Applied Pyrolysis, 1981, 3, 1, 49. [a.234] F. Chen and J. Qian, Fuel, 2002, 81, 16, 2071. [a.235] A.K. Bhowmick, S. Rampalli, K. Gallagher, R. Seeger and D. McIntyre, Journal of Applied Polymer Science, 1987, 33, 4, 1125. [a.236] J.C.W. Chieng and J.K.Y. Kiang, European Polymer Journal, 1979, 15, 11, 1059. [a.237] S.A. Groves, R.S. Lehrle, M. Blazso and T. Székely, Journal of Analytical and Applied Pyrolysis, 1991, 19, 301. [a.238] H. Pakdel, D.M. Pantea and C. Roy, Journal of Analytical and Applied Pyrolysis, 2001, 57, 1, 91. [a.239] F. Cataldo, Journal of Analytical and Applied Pyrolysis, 1998, 44, 2, 121. [a.240] A.M. Cunliffe and P.T. Williams, Journal of Analytical and Applied Pyrolysis, 1998, 44, 2, 131. [a.241] P.X. Ma and R.Y. Zhang, Journal of Biomedical Materials Research, 1999, 46, 1, 60. [a.242] F.D. Kopinke and K. Mackenzie, Journal of Analytical and Applied Pyrolysis, 1997, 40–41, 43. [a.243] Y. Aoyagi, K. Yamashita and Y. Doi, Polymer Degradation and Stability, 2002, 76, 1, 53. [a.244] K. Stridsberg and A.C. Albertsson, Polymer, 2000, 41, 20, 7321.

263

Thermal Degradation of Polymeric Materials [a.245] H. Tsuji, Polymer, 2000, 41, 10, 3621. [a.246] H. Tsuji and I. Fukui, Polymer, 2003, 44, 10, 2891. [a.247] I.C. McNeill and H.A. Leiper, Polymer Degradation and Stability, 1985, 11, 4, 309. [a.248] Y. Ikada, K. Jamshidi, H. Tsuji and S.H. Hyon, Macromolecules, 1987, 20, 4, 904. [a.249] H.R. Kricheldorf and D.O. Damrau, Macromolecular Chemistry and Physics, 1998, 199, 6, 1089. [a.250] H. Nishida, M. Yamashita, N. Hattori, T. Endo and Y. Tokiwa, Polymer Degradation and Stability, 2000, 70, 3, 485. [a.251] H. Nishida, M. Yamashita and T. Endo, Polymer Degradation and Stability, 2002, 78, 1, 129. [a.252] J. Njuguna and K. Pielichowski, Advanced Engineering Materials, 2004, 6, 4, 193. [a.253] J. Njuguna and K. Pielichowski, Advanced Engineering Materials, 2003, 5, 11, 769. [a.254] I. Dlouhy, Z. Chlup, D.N. Boccaccini, S. Atiq and A.R. Boccaccini, Composites A, 2003, 34, 11, 1177. [a.255] A.R. Boccaccini, J. Janczak-Rusch and I. Dlouhy, Materials Chemistry and Physics, 1998, 53, 2, 155. [a.256] S. Sutherland, K.P. Plucknett and M.H. Lewis, Composites Engineering, 1995, 5, 10/11, 1367. [a.257] A.R. Boccaccini, A.J. Strutt, K.S. Vecchio, D. Mendoza, K.K. Chawla, C.B. Ponton and D.H. Pearce, Composites A, 1998, 29, 11, 1343. [a.258] N. Chawla, K.K. Chawla, M. Koopman, B. Patel, C. Coffin and J.I. Eldridge, Composites Science and Technology, 2001, 61, 13, 1923. [a.259] S.S. Tzeng and W.C. Lin, Carbon, 1999, 37, 12, 2011. [a.260] T. Naruse, T. Hattori, H. Miura and K. Takahashi, Composite Structures, 2001, 52, 3–4, 533.

264

References [a.261] A.M. Hindeleh and S.M. Abdo, Polymer Communications, 1989, 30, 184. [a.262] H.V. Parimala and K. Vijayan, Journal of Materials Science, 1993, 12, 2, 99. [a.263] C.Y. Yue, G.X. Sui and H.C. Looi, Composites Science and Technology, 2000, 60, 3, 21. [a.264] M.H. Lin, W. Buchgraber, G. Korb and P.W. Kao, Scripta Materialia, 2002, 46, 2, 169. [a.265] G. Cakmak, Z. Kucukyavuz, S. Kucukyavuz and H. Cakmak, Composites A, 2004, 35, 4, 417. [a.266] J.A. Hiltz, Journal of Analytical and Applied Pyrolysis, 1991, 22, 1–2, 113. [a.267] M. Hakkarainen, G. Gallet and S. Karlsson, Polymer Degradation and Stability, 1994, 64, 1, 91. [a.268] F. Suhara, S.K.N. Kutty and G.B. Nando, Polymer Degradation and Stability, 1998, 61, 1, 9. [a.269] J. Njuguna and K. Pielichowski, Journal of Materials Science, 2004, 39, 4081. [a.270] H.K. Lee and S.W. Ko, Journal of Applied Polymer Science, 1993, 50, 7, 1269. [a.271] D. Baral, P.P. De and G.B. Nando, Polymer Degradation and Stability, 1999, 65, 1, 47. [a.272] R.A. Correa, R.C.R. Nunes and V.L. Lourenco, Polymer Degradation and Stability, 1996, 52, 3, 245. [a.273] J. Njuguna and K. Pielichowski K, Composites, submitted. [a.274] E. Devaux, M. Rochery and S. Bourbigot, Fire and Materials, 2002, 26, 4–5, 149. [a.275] Y.I. Tien and K.H. Wei, Journal of Applied Polymer Science, 2002, 86, 7, 1741. [a.276] L.Y. Chiang, L.Y. Wang and C.S. Kuo, Macromolecules, 1995, 28, 22, 7574. [a.277] H. Mahfuz, V.K. Rangari, M.S. Islam and S. Jeelani, Composites A, 2004, 35, 4, 453.

265

Thermal Degradation of Polymeric Materials [a.278] W.J. Wang, W.K. Chin and W.J. Wang, Journal of Polymer Science B, 2002, 40, 15, 1690. [a.279] B.K. Kim, J.W. Seo and M.O. Jeong, European Polymer Journal, 2003, 39, 1, 85. [a.280] J.H. Chang and U.K. An, Journal of Polymer Science B, 2002, 40, 7, 670. [a.281] D.R. Yei, S.W. Kuo, Y.C. Su and F.C. Chang, Polymer, 2004, 45, 8, 2633. [a.282] B.X. Fu, B.S. Hsiao, S. Pagola, P. Stephens, H. White, M. Rafailovich, P. Mather, H. Jeon, S. Phillips, J. Lichtenhan and J. Schwab, Polymer, 2001, 42, 2, 599. [a.283] P.T. Mather, H.G. Jeon, A. Romo-Uribe, T.S. Haddad and J.D. Lichtenhan, Macromolecules, 1998, 32, 4, 1194. [a.284] A. Provatas and J. G. Matisons, Trends in Polymer Science, 1997, 5, 1, 32. [a.285] D. Neumann, M. Fisher, L. Tran and J.G. Matisons, Journal of American Chemical Society, 2002, 124, 47, 13998. [a.286] K. Fukatsu, Journal of Fire Sciences, 1990, 8, 1, 194. [a.287] K. Fukatsu, Polymer Degradation and Stability, 2002, 75, 3, 479. [a.288] K.C.M. Nair, T. Sabu and G. Groeninckx, Composites Science and Technology, 2001, 61, 16, 2519. [a.289] B. Li and J. He, Polymer Degradation and Stability, 2004, 83, 2, 241. [a.290] R.S. Rajeev, S.K. De, A.K. Bhowmick and B. John, Polymer Degradation and Stability, 2003, 79, 3, 449. [a.291] M. Omastova, S. Podhradska, J. Prokes, I. Janigova and J. Stejskal, Polymer Degradation and Stability, 2003, 82, 2, 251. [a.292] S.L. Madorsky, Thermal Degradation of Organic Polymers, Interscience, New York, 1964, 4. [a.293] T. Kashiwagi, E. Grulke, J. Hilding, R. Harris, W. Awad and J. Douglas, Macromolecular Rapid Communications, 2002, 23, 13, 761. [a.294] F. Dabrowski, S. Bourbigot, R. Delobel and M. Le Bras, European Polymer Journal, 2000, 36, 2, 273.

266

References [a.295] A. Blumstein, Bulletin of the Chemical Society of Japan, 1961, 34, 899. [a.296] S.D. Burnside and E.P. Giannelis, Chemistry of Materials, 1995, 7, 9, 1597. [a.297] J.W. Gilman, C.L. Jackson, A.B. Morgan, R. Harris Jr., E. Manias, E.P. Giannelis, M. Wuthenow, D. Hiltona and S.H. Phillips, Chemistry of Materials, 2000, 12, 7, 1866. [a.298] A. Singh and R. Haghighat, inventors; Triton Systems Inc., assignee; US Patent 6057035, 2000. [a.299] G. Deshpande and M.E. Rezac, Polymer Degradation and Stability, 2002, 76, 1, 17. [a.300] T. Takahashi, H. Münstedt, M. Modesti and P. Colombo, Journal of the European Ceramic Society, 2001, 21, 2821. [a.301] N.S.M. Stevens and M.E. Rezac, Chemical Engineering Science, 1998, 53, 9, 1699. [a.302] R. Dai, L. Ye, A. Luo, R. Fu, S. Zhang, G. Xie and S. Jin, Journal of Analytical and Applied Pyrolysis, 1997, 42, 2, 103. [a.303] T. van Hoang and A. Guyot, Polymer Degradation and Stability, 1991, 32, 1, 93. [a.304] S.A. Visser, C.E. Hewitt and T.D. Binga, Journal of Polymer Science B, 1996, 43, 9, 1679. [a.305] S. Lin, in High-Temperature Properties and Applications of Polymeric Materials, Eds. M.R. Tant, J. W. Connell and H. L. N. McManus, ACS Symp. Ser. 603, American Chemical Society, Washington, DC, 1995, 37. [a.306] T.S. Radhakrishnan, Journal of Applied Polymer Science, 1999, 73, 3, 441. [a.307] G. Deshpande and M.E. Rezac, Polymer Degradation and Stability, 2001, 74, 2, 363. [a.308] I. Hagglund, K. Janak, L. Blomberg, A. Bergard, S.G. Claude and M. Lymann, Chromatographic Science, 1991, 29, 396. [a.309] G. Camino, S.M. Lomakin and M. Lazzari, Polymer, 2001, 42, 6, 2395. [a.310] M.V. Sobolevskii, I.I. Skorokhodor, V. Ditsent, L.V. Soboleskaya, E.I. Vovshin and L.M. Blekk, Polymer Science – USSR, 1974, 16, 840.

267

Thermal Degradation of Polymeric Materials [a.311] C. Chou and M.H. Yang, Journal of Thermal Analysis, 1993, 40, 2, 657. [a.312] H.R. Allcock, G.S. McDonnell, G.H. Riding and I. Manners, Chemistry of Materials, 1990, 2, 4, 425. [a.313] M. Gleria, R. Bertani, R. De Jaeger and S. Lora, Journal of Fluorine Chemistry, 2004, 125, 2, 329. [a.314] M.T.R. Laguna, M.P. Tarazona, G.A. Carriedo, F.J. Garcia Alonso, J.I. Fidalgo and E. Saiz, Macromolecules, 2002, 35, 19, 7505. [a.315] M. Cochez, M. Ferriol, J.V. Weber, P. Chaudron, N. Oget and J.L. Mieloszynski, Polymer Degradation and Stability, 2000, 70, 3, 455. [a.316] M. Zeldin, K.J. Wynne and H.R. Allcock, in Inorganic and Organometallic Polymers, Eds. M. Zeldin, K.J. Wynne and H.R. Allcock, ACS Symp. Ser. 360, American Chemical Society, Washington, DC, 1988, 1. [a.317] G. Bosscher and J.C. van de Grampel, Journal of Inorganic and Organometallic Polymers, 1995, 5, 3, 209. [a.318] G. Bosscher, A. Meetsma and J.C. van de Grampel, Inorganic Chemistry, 1996, 35, 23, 6646. [a.319] F. Li, J. Wang, J. Gao, X. Yu, J. Tang and X. Tang, Materials Chemistry and Physics, 2000, 62, 1, 81. [a.320] H. Ma, Y. Li, H. Zhang, J. Wang, F. Li and X. Tang, Materials Chemistry and Physics, 2002, 73, 1, 93–96. [a.321] M.L. White, R.A. Montague, K. Matyjaszewski and T. Pakula, Polymer, 1995, 36, 18, 3493. [a.322] Y.H. Han, R.S. Mishra, M.J. Gasch, H.B. Lee and A.K. Mukherjee, Korean Journal of Ceramics, 2000, 6, 245. [a.323] Q.D. Nghiem, J.K. Jeon, L.Y. Hong and D.P. Kim, Journal of Organometallic Chemistry, 2003, 688, 1–2, 27. [a.324] W.R. Schmidt, D.M. Narsavage-Heald, D.M. Jones, P.S. Marchetti, D. Raker and G.E. Maciel, Chemistry of Materials, 1999, 11, 6, 1455. [a.325] N.S.C. Kwet Yive, R.J.P. Corriu, D. Leclerq, P.H. Mutin and A. Vioux, Chemistry of Materials, 1992, 4, 1, 141.

268

References [a.326] T. Gerdau, H.J. Kleiner, M. Peuckert, M. Brueck and F. Aldinger, inventors; Hoechst AG assignee; US Patent 5066623, 1990. [a.327] T. Breuning, Journal of Analytical and Applied Pyrolysis, 1999, 49, 43. [a.328] J. Luecke, J. Hacker, D. Suttor and G. Ziegler, Applied Organometallic Chemistry, 1997, 11, 2, 181. [a.329] G. Cardoen and E.B. Coughlin, Macromolecules, 2004, 37, 13, 5123. [a.330] J.M. Lin and C.C.M. Ma, Polymer Degradation and Stability, 2000, 69, 2, 229. [a.331] C.K. Chan, S.L. Peng, I.M. Chua and S.C. Ni, Polymer, 2001, 42, 9, 4189. [a.332] S. Yano, K. Nakamura, M. Kodimari and N. Yamauchi, Journal of Applied Polymer Science, 1994, 54, 4, 163. [a.333] A. Lee and J.D. Lichtenhan, Macromolecules, 1998, 31, 15, 4970. [a.334] F.J. Feher and S.H. Phillips, Journal of Organometallic Chemistry, 1996, 521, 1–2, 401. [a.335] J. Pyun, K. Matyjaszewski, J. Wu, G.M. Kim, S.B. Chun and P.T. Mather, Polymer, 2003, 44, 9, 2739. [a.336] J.D. Lichtenhan, Y.A. Otonaria and M.J. Carr, Macromolecules, 1995, 28, 24, 8435. [a.337] T.S. Haddad, P.T. Mather, H.G. Jeon, S.B. Chun and S.H. Phillips, Materials Research Society Symposium Proceedings, 628, 2001, CC 2.6.1. [a.338] L. Zhu, B.R. Mimnaugh, Q. Ge, R.P. Quirk, S.Z.D. Cheng, E.L. Thomas, B. Lotz, B.S. Hsiao, F. Yeh and L. Liu, Polymer, 2001, 42, 21, 9121. [a.339] R.M. Laine, R. Tamaki and J. Choi, inventors; University of Michigan, USA, assignee; European Patent EP1328529 (A1), 2002. [a.340] N.A. Mohamed and A.O.H. Al-Dossary, Polymer Degradation and Stability, 2003, 79, 1, 61. [a.341] Y.L. Liu and S.H. Tsai, Polymer, 2002, 43, 21, 5757. [a.342] Z. Ge, S. Yang, Z. Tao, J. Liu and L. Fan, Polymer, 2004, 45, 11, 3627.

269

Thermal Degradation of Polymeric Materials [a.343] W. Kim, K. Park and J. Lee, Journal of Molecular Catalysis A, 2002, 184, 1–2, 39. [a.344] C.C. Hu, Y.C. Wang, C.L. Li, K.R. Lee, Y.C. Chen and J.Y. Lai, Desalination, 2002, 144, 103. [a.345] D.J. Liaw and P. Chang, Polymer, 1997, 38, 22, 5545. [a.346] F. Khan, A.M. Hor and P.R. Sundararajan, Journal of Physical Chemistry B, 2004, 108, 1, 117. [a.347] L. Abate, I. Blanco, O. Motta, A. Pollicino and A. Recca, Polymer Degradation and Stability, 2002, 75, 3, 465. [a.348] N. Hurduc, M. Prajinaru, B. Donose, D. Pavel and N. Hurduc, Polymer Degradation and Stability, 2001, 72, 3, 441. [a.349] C. Damian, N. Hurduc, N. Hurduc, R. Shanks, I. Yarovsky and D. Pavel, Computational Materials Science, 2003, 27, 4, 393. [a.350] L.H. Perng, Polymer Degradation and Stability, 2000, 69, 3, 323. [a.351] S. Takayama, T. Mathubara, T. Arai and K. Takedo, Polymer Degradation and Stability, 1995, 50, 3, 277. [a.352] P.N. Lavrenko, O.V. Okatova and B. Schulz, Polymer Degradation and Stability, 1998, 61, 3, 473. [a.353] C. Jaksland, E. Rasmussen and T. Rohde, Waste Management, 2000, 20, 5–6, 463. [a.354] A.G. Pedroso, L.H.I. Mei, J.A.M Agnelli and D.S. Rosa, Polymer Testing, 2002, 21, 2, 229. [a.355] M. Naffakh, G. Ellis, M.A. Gmez and C. Marco, Polymer Degradation and Stability, 1999, 66, 3, 405. [a.356] T. Sugama, Materials Letters, 2004, 58, 7–8, 1307. [a.357] M.J. Ariza, D.J. Jones and J. Roziere, Desalination, 2002, 147, 1–3, 183. [a.358] Y.L. Liu and Y.J. Chen, Polymer, 2004, 45, 6, 1797. [a.359] R. Torrecillas, A. Baudry, J. Dufay and D. Mortaigne, Polymer Degradation and Stability, 1996, 54, 2–3, 267.

270

References [a.360] R. Torrecillas, N. Regnier and D. Mortaigne, Polymer Degradation and Stability, 1996, 51, 3, 307. [a.361] C. Gouri, C.P.R. Nair and R. Ramaswamy, High Performance Polymers, 2000, 12, 4, 497. [a.362] J. Bibiao, H. Jianjun, W. Wenyun, J. Luxia and C. Xinxian, European Polymer Journal, 2001, 37, 3, 463. [a.363] T.K. Lin, S.J. Wu, J.G. Lai and S.S. Shyu, Composites Science and Technology, 2000, 60, 9, 1873. [a.364] K. Hemvichian, A. Laobuthee, S. Chirachanchai and H. Ishida, Polymer Degradation and Stability, 2002, 76, 1, 1. [a.365] K. Tamargo-Martinez, S. Villar-Rodil, J.I. Paredes, A. Martinez-Alonso and J.M.D. Tascon, Chemistry of Materials, 2003, 15, 21, 4052. [a.366] J. Hetper and M. Sobera, Journal of Chromatography A, 1999, 833, 2, 277. [a.367] C.P.R. Nair, R.L. Bindu and K.N. Ninan, Polymer Degradation and Stability, 2001, 73, 2, 251. [a.368] D. Puglia, L.B. Manfredi, A. Vazquez and J.M. Kenny, Polymer Degradation and Stability, 2001, 73, 3, 521. [a.369] L.H. Lee, Journal of Polymer Science Part A, 1965, 3, 859. [a.370] S.A. Kumar and T.S.N. Narayanan, Progress in Organic Coatings, 2002, 45, 4, 323. [a.371] X. Jiang, Y. Zhang and Y. Zhang, Polymer Testing, 2004, 23, 3, 259. [a.372] Toronto City Council – Canada, Overview of New and Emerging Technologies, www.city.toronto.on.ca [a.373] L. Ballice and R. Reimert, Chemical Engineering and Processing, 2002, 41, 4, 289. [a.374] J. Gersten, V. Fainberg, G. Hetsroni and Y. Shindler, Fuel, 2000, 79, 13, 1679. [a.375] G.S. Kumar, V.R. Kumar and G. Madras, Journal of Applied Polymer Science, 2002, 84, 4, 681.

271

Thermal Degradation of Polymeric Materials [a.376] Z. Wang, G. Wu, Y. Hu, Y. Ding, K. Hu and W. Fan, Polymer Degradation and Stability, 2002, 77, 3, 427. [a.377] R. Li, L. Ye and Y.W. Mai, Plastics, Rubber and Composites – Processing and Applications, 1998, 26, 8, 368. [a.378] J.J. Park, K. Park, J.S. Kim, S. Maken, H. Song, H. Shin, J.W. Park and M.J. Choi, Energy and Fuels, 2003, 17, 6, 1576. [a.379] S.O. Woo, N. Ayala and L. Broadbelt, Catalysis Today, 2000, 55, 1–2, 161. [a.380] S.Y. Lee, J.H. Yoon, J.R. Kim and D.W. Park, Journal of Analytical and Applied Pyrolysis, 2002, 64, 1, 71. [a.381] B.L. Fletcher and M.E. Mackay, Resources, Conservation and Recycling, 1996, 17, 2, 141. [a.382] V. Karmore and G. Madras, Industrial and Engineering Chemistry Research, 2002, 41, 4, 657. [a.383] D. Dong, S. Tasaka and N. Inagaki, Polymer Degradation and Stability, 2001, 72, 2, 345. [a.384] J. Brandrup, M. Bittner, G. Menges and W. Michaeli, Recycling and Recovery of Plastics, Hanser Publishers, Munich, 1996, 434. [a.385] R. Miranda, H. Pakdel, C. Roy, H. Darmstadt and C. Vasile, Polymer Degradation and Stability, 1999, 66, 1, 107. [a.386] S. Kim, Waste Management, 2001, 21, 7, 609. [a.387] A. Jimenez, J. Lopez, J. Vilaplana and H.J. Dussel, Journal of Analytical and Applied Pyrolysis, 1997, 40–41, 201. [a.388] T. Corrales, F. Catalina, C. Peinado, N.S. Allen and E. Fontan, Journal of Photochemistry and Photobiology A: Chemistry, 2002, 147, 213. [a.389] P. Straka, J. Nahunkova and Z. Brozova, Journal of Analytical and Applied Pyrolysis, 2004, 71, 1, 213. [a.390] M. Noda and H. Okuyama, Chemical and Pharmaceutical Bulletin, 1999, 47, 4, 467. [a.391] H. Tsuji, H. Daimon and K. Fujie, Biomacromolecules, 2003, 4, 3, 835.

272

References [a.392] H. Tsuji, I. Fukui, H. Daimon and K. Fujie, Polymer Degradation and Stability, 2003, 81, 3, 501. [a.393] A. Demirbas, Energy Conversion and Management, 2001, 42, 11, 1357. [a.394] G. Gea, M.B. Murillo and J. Arauzo, Industrial and Engineering Chemistry Research, 2002, 41, 19, 4714. [a.395] K. Whitty, R. Backman, M. Forssen, M. Hupa, J. Rainio and V. Sorvari, Journal of Pulp and Paper Science, 1997, 23, 3, J119. [a.396] V.S. Zope, S. Mishra, V.S. Patil, K.K. Agrawal, J.P. Mahajan and S.A. Firke, Institution of Engineers (India) Journal – CH, 2003, 9, 44. [a.397] H. Bockhorn, J. Hentschel, A. Hornung and U. Hornung, Chemical Engineering Science, 1999, 54, 15–16, 3043. [a.398] R. Miranda, J. Yang, C. Roy and C. Vasile, Polymer Degradation and Stability, 2001, 72, 3, 469. [a.399] Y. Sakata, M.A. Uddin, K. Koizumi and K. Murata, Polymer Degradation and Stability, 1996, 53, 1, 111. [a.400] Y. Shiraga, M.A. Uddin, A. Muto, M. Narazaki and Y. Sakata, Energy and Fuels, 1999, 13, 2, 428. [a.401] K.S. Lin, H.P. Wang, S.H. Liu, N.B. Chang, Y.J. Huang and H.C. Wang, Fuel Processing Technology, 1999, 60, 2, 103. [a.402] J. Yanik, M.A. Uddin, K. Ikeuchi and Y. Sakata, Polymer Degradation and Stability, 2001, 73, 2, 335. [a.403] H. Bockhorn, A. Hornung and U. Hornung, Journal of Analytical and Applied Pyrolysis, 1998, 46, 1, 1. [a.404] L. Ballice, M. Yuksel, M. Saglam, R. Reimert and H. Schulz, Fuel, 1998, 77, 13, 1431. [a.405] MEDLARS, The National Library of Medicine, Bethesda, MD, USA, http:// www.nlm.nih.gov [a.406] ENVIROFACTS, US Environmental Protection Agency, Washington, DC, USA, http://www.epa.gov

273

Thermal Degradation of Polymeric Materials [a.407] Syracuse Research Corporation’s Environmental Fate Data Base (EFDB), Syracuse Research Corporation, North Syracuse, NY, USA, http://www.syrres. com/esc/efdb.htm [a.408] ECOSAR Program, US Environmental Protection Agency, http://www.epa. gov/oppt/newchems/21ecosar.htm [a.409] S. Ingo, ICS-UNIDO, Expert Group Meeting on Environmental Degradable Polymers and Sustainable Development, Trieste, Italy, 5–6 September 2002. [a.410] R.T. Johnston and E.J. Slone, International Conference on Advances in the Stabilization and Controlled Degradation of Polymers, Lucerne, Switzerland, 21–23 May 1990. [a.411] H. Potente, M. Bastian, K. Bergemann, M. Senge, G. Scheel and T. Winklemann, International Polymer Processing, 2001, 16, 4, 341. [a.412] H. Catrier and G.H. Hu, Polymer Engineering and Science, 1998, 38, 177. [a.413] F.L. Marten, A. Famili and J.F. Nangeroni, inventors; Air Products and Chemicals, Inc., assignee; European Patent 0415357 B1, 1992. [a.414] J.L. Willett, M.M. Millard and B.K. Jasberg, Polymer, 1997, 38, 24, 5983. [a.415] M.C.G. Rocha, F.M.B. Coutinho and S.T. Balke, Polymer Testing, 1995, 14, 4, 369. [a.416] J.L. Willett, M.M. Millard and B.K. Jasberg, US Department of Agriculture, http://www.nal.usda.gov/ttic/tektran/data/000007/40/0000074054.html [a.417] P. Alexy, D. Kachova, M. Krsiak, D. Bakos and B. Simkova, Polymer Degradation and Stability, 2003, 79, 3, 511. [a.418] A.G. Loera, F. Cara, M. Dumon and J.P. Pascault, Macromolecules, 2002, 35, 16, 6291. [a.419] D. Pendlebury, inventor; Allied Chemical Corporation, assignee, US Patent 4072663, 1978. [a.420] R.L. Laurence, R. Galvan and M.V. Tirrell, Mathematics of Polymerization, in Polymer Reactor Engineering, Ed. C. McGreavy, Blackie Academic, London, 1994. [a.421] N.A. Dotson, R. Galvan, R.L. Laurence and M. Tirrell, Polymerization Process Modelling, VCH Publishers, Munich, 1996.

274

References [a.422] I.H. Kim, H.C. Cho, Y.C. Bae, H.W. Park and K.S. Chung, European Polymer Journal, 2003, 39, 7, 1431. [a.423] M. Wang, C. Zhang, J.M. Smith and B.J. McCoy, AIChE Journal, 1994, 40, 1, 131. [a.424] G. Camino, L. Costa and L. Trossarelli, Polymer Degradation and Stability, 1985, 12, 3, 203. [a.425] X. Almeras, F. Dabrowski, M. Le Bras, R. Delobel, S. Bourbigot, G. Marosi and P. Anna, Polymer Degradation and Stability, 2002, 77, 2, 315. [a.426] B.L. Denq, W.Y. Chiu and K.F. Lin, Journal of Applied Polymer Science, 1997, 66, 10, 1855. [a.427] S.R. Betso, J.A. Berdasco, M.F. Debney, G.L. Murphy, N.P. Rome, S.G. Richards and B.A. Howell, Journal of Applied Polymer Science, 1994, 51, 5, 781. [a.428] J. Liu, Industrial and Engineering Chemistry Research, 2001, 40, 24, 5719. [a.429] M.R. Nyden, Recent Advances in Flame Retardancy of Polymeric Materials, Vol. IX, BCC, Norwalk, CT, USA, 1998, 1. [a.430] J.E.J. Staggs, Recent Advances in Flame Retardancy of Polymeric Materials, Vol. XIV, BCC, Norwalk, CT, USA, 2003, 27. [a.431] M. Modesti, A. Lorenzetti, F. Simioni and G. Camino, Polymer Degradation and Stability, 2002, 77, 2, 195. [a.432] D. Garlotta, Journal of Polymers and Environment, 2001, 9, 2, 63. [a.433] T. Gupta and B. Adhikari, Thermochimica Acta, 2003, 402, 1–2, 169. [a.434] M. Ratzsch, M. Arnold, E. Borsig, H. Bucka and N. Reichelt, Progress in Polymer Science, 2002, 27, 7, 1195. [a.435] T. Provder, M.W. Urban and H.G. Barth (Eds), Hyphenated Techniques in Polymer Characterization, ACS Symp. Ser. 581, American Chemical Society, Washington, DC, 1994. [a.436] C. Saron and M.I. Felisberti, Materials Science and Engineering A, 2004, 370, 1-2, 293. [a.437] N. Stiov Vrandecic, I. Klaric and T. Kovacic, Polymer Degradation and Stability, 2004, 84, 1, 23.

275

Thermal Degradation of Polymeric Materials [a.438] S. Husic, I. Javni and Z.S. Petrovic, Composites Science and Technology, 2005, 65, 1, 19. [a.439] C. Sivalingam and G. Madras, Polymer Degradation and Stability, 2004, 84, 3, 393. [a.440] C. Sivalingam, R. Karthik and G. Madras, Polymer Degradation and Stability, 2004, 84, 2, 345. [a.441] H. Abe, N. Takahashi, K.J. Kim, M. Mochizuki and Y. Doi, Biomacromolecules, 2004, 5, 4, 1480. [a.442] S.C. Liufu, H.N. Xiao and Y.P. Li, Polymer Degradation and Stability, 2005, 87, 1, 103. [a.443] S. Su and C.A. Wilkie, Polymer Degradation and Stability, 2004, 83, 2, 347. [a.444] Z. Stojanovic, L. Katsikas, I. Popovic, S. Jovanovic and K. Jeremic, Polymer Degradation and Stability, 2005, 87, 1, 177. [a.445] R.M. van den Einde, M.E. van der Veen, H. Bosman, A.J. van der Goot and R.M. Boom, Journal of Food Engineering, 2005, 66, 2, 147. [a.446] M. Roman and W.T. Winter, Biomacromolecules, 2004, 5, 5, 1671. [a.447] C. Branca, C. Di Blasi and C. Russo, Fuel, 2005, 84, 1, 37. [a.448] Q. Liu, C. Lu, Y. Yang, F. He and L. Ling, Journal of Molecular Structure, 2005, 733, 1-3, 193. [a.449] X. Colin, C. Marais and J. Verdu, Composites Science and Technology, 2005, 65, 1, 117. [a.450] N. Chisholm, H. Mahfuz, V.K. Rangari, A. Ashfaq and S. Jeelani, Composite Structures, 2005, 67, 1, 115. [a.451] Y.L. Liu, W.L. Wei, K.Y. Hsu and W.H. Ho, Thermochimica Acta, 2004, 412, 1-2, 139. [a.452] F. Stangenberg, S. Agren and S. Karlsson, Chromatographia, 2004, 59, 1-2, 101. [a.453] H.M. da Costa, V.D. Ramos and M.C.G. Rocha, Polymer Testings, 2005, 24, 1, 86. [a.454] O. Paric, C. Zollfrank and G.A. Zickler, Carbon, 2005, 43, 1, 53.

276

References Available from the Polymer Library

15

References Available from the Polymer Library (www.polymerlibrary.com)

{428809} P. Carmiti, P.L. Beltrame, M. Armada, A. Gervasini and G. Audisio, Industrial and Engineering Chemistry Research, 1991, 30, 7, 1624. {431880} M.J. Scudamore, P.J. Briggs and F.H. Prager, Fire and Materials, 1991, 15, 2, 65. {443515} L.E. Manning, Macromolecules, 1991, 24, 11, 3304. {451119} S.V. Levchik, L. Costa and G. Camino, Polymer Degradation and Stability, 1992, 36, 3, 229. {454425} S.R. Salman, N.D. Al-Shama’a and M.M.F. Al-Jarrah, Polymer Plastics Technology and Engineering, 1992, 31, 3 & 4, 213. {456769}. Jae-do Nam and J.C. Seferis, Journal of Polymer Science: Polymer Physics Edition, 1992, 30, 5, 455. {461325} K. Patel, A. Velazquez, H.S. Calderon and G.R. Brown, Journal of Applied Polymer Science, 1992, 46, 1, 179. {485571} N.S. Allen, M. Edge, M. Mohammadian and K. Jones, Polymer Degradation and Stability, 1993, 41, 2, 191. {490068} G. Montaudo, C. Puglisi and F. Samperi, Polymer Degradation and Stability, 1993, 42, 1, 13. {497843} A. Jiminez, V. Berenguer, J. Lopez and A. Sanchez, Journal of Applied Polymer Science, 1993, 50, 9, 1565. {497853} J.P. Agrawal and K.S. Kulkarni, Journal of Applied Polymer Science, 1993, 50, 9, 1655.

277

Thermal Degradation of Polymeric Materials {502570} S.V. Levchik, L. Costa and G. Camino, Polymer Degradation and Stability, 1994, 43, 1, 43. {502579} P. Simon, Polymer Degradation and Stability, 1994, 43, 1, 125. {503329} K. Lederer, Polimeri, 1993, 14, 6, 253. {512357} Z. Ozturk and J.F. Merklin, Journal of Applied Polymer Science, 1994, 52, 6, 747. {524531} K. Pielichowski, A. Puszynski and J. Pielichowski, Polymer Journal (Japan), 1994, 26, 7, 822. {546565} C. Chen, K.T. Nguyen, B. Sanschagrin and L. Piche, Proceedings of ANTEC ‘94, San Francisco, CA, 1994, II, 2041. {547250} M.J. Antal and G. Varhegyi, Industrial and Engineering Chemistry Research, 1995, 34, 3, 703. {565743} M.R. Grimbley and R.S. Lehrle, Polymer Degradation and Stability, 1995, 49, 2, 223. {567842} B.J. Kim and J.L. White, International Polymer Processing, 1995, 10, 3, 213. {575553} P. Moulinie, R.M. Paroli, Z.Y. Wang, A.H. Delgado, A.L. Guen, Y. Qi and J.P. Gao, Polymer Testing, 1996, 15, 1, 75. {575698} T. Sawaguchi, T. Ikemura and M. Seno, Macromolecular Chemistry and Physics, 1996, 197, 1, 215. {581327} Y. Nagasaki, N. Yamazaki and M. Kato, Macromolecular Rapid Communications, 1996, 17, 2, 123. {582415} M. Bounekhel and I.C. McNeill, Polymer Degradation and Stability, 1996. 51, 1, 35. {583706} P.V. Zgonnik, V.V. Zuev, L.D. Turkova and L.A. Shibaev, Polymer Science Series B, 1995, 37, 11-12, 509. {584927} B.A. Keller, C. Loewe and R. Hany, Proceedings of Euradh ‘94, Mulhouse, , 1994, 582. {589987} F.D. Kopinke, M. Remmler and K. Mackenzie, Polymer Degradation and Stability, 1996, 52, 1, 25.

278

References Available from the Polymer Library {592654} S. Tate and H. Narusawa, Polymer, 1996, 37, 9, 1583. {594526} R.S. Lehrle, D.J. Atkinson, D.M. Bate, P.A. Gardner, M.R. Grimbley, S.A. Groves, S.A. Place and R.J. Williams, Polymer Degradation and Stability, 1996. 52, 2, 183. {594557} Y.M. Shulga, V.I. Rubtsov, O.N. Efimov, G.P. Karpacheva, L.M. Zemtsov and V.V. Kozlov, Polymer Science Series A, 1996, 38, 6, 636. {596000} M.P. Anachkov, S.K. Rakovski and A.K. Stoyanov, Journal of Applied Polymer Science, 1996, 61, 4, 585. {600788} H.G. Schild, Journal of Polymer Science: Polymer Chemistry Edition, 1996, 34, 11, 2259. {603272} L. Abate, S. Calanna, A. Pollicino and A. Recca, Polymer Engineering and Science, 1996, 36, 13, 1782. {604629} T. Sawaguchi and M. Seno, Polymer, 1996, 37, 16, 3697. {607567} F.D. Kopinke, M. Remmler, K. Mackenzie, M. Moder and O. Wachsen, Polymer Degradation and Stability, 1996, 53, 3, 329. {609091} M.R. Nyden, T.R. Coley and S. Mumby, Proceedings of ANTEC ‘96, Indianapolis, IN, 1996, III, 3058. {612251} G. Montaudo, C. Puglisi, J.W. de Leeuw, W. Hartgers, K. Kishore and K. Ganesh, Macromolecules, 1996, 29, 20, 6466. {615213} T. Sawaguchi and M. Seno, Polymer Degradation and Stability, 1996, 54, 1, 33. {630026} A.H. Ali and K.S.V. Srinivasan, Journal of Macromolecular Science A, 1997, A34, 2, 235. {635363} J. Verdu, Macromolecular Symposia, 1997, 115, 165. {638278} L. Torre and J.M. Kenny, Proceedings of ANTEC 97, Toronto, Canada, 1997, II, 1728. {639346} L. Abate, S. Calanna, A. Pollicino and A. Recca, Macromolecular Chemistry and Physics, 1997, 198, 5, 1437. {639383} T. Nozaki, Y. Maeda and H. Kitano, Journal of Polymer Science: Polymer Chemistry Edition, 1997, 35, 8, 1535.

279

Thermal Degradation of Polymeric Materials {639956} S.S. Parikh and L. Zlatkevich, Angewandte Makromolekulare Chemie, 1997, 247, 255. {641470} R.W.J. Westerhout, J. Waanders, J.A.M. Kuipers and W.P.M. van Swaaij, Industrial and Engineering Chemistry Research, 1997, 36, 6, 1955. {642071} M. Hakkarainen, A.-C. Albertsson and S. Karlsson, Journal of Environmental Polymer Degradation, 1997, 5, 2, 67. {645123} M.I. Chipara, M. Corciovei, T. Sarbu, J.R. Romero, M D. Chipara and F. Taran, Polymer Degradation and Stability, 1997, 57, 2, 211. {651339} R.T. Pogue, K.L. Ackley and V. Majidi, International Journal of Polymer Analysis and Characterization, 1997, 3, 3, 193. {660936} Zuowan Zhou and Qixian Wu, Journal of Applied Polymer Science, 1997, 66, 7, 1227. {660978} L.J. Broadbelt, M.T. Klein, B.D. Dean and S.M. Andrews, Journal of Polymer Science: Polymer Chemistry Edition, 1997, 35, 15, 3305. {662846} Xiaohua Liu, Maoqing Kang and Xinkui Wang, China Synthetic Rubber Industry, 1997, 20, 6, 377. {668871} G. Madras, J.M. Smith and B.J. McCoy, Polymer Degradation and Stability, 1997, 58, 1-2, 131. {669883} T. Sawaguchi and M. Seno, Journal of Polymer Science: Polymer Chemistry Edition, 1998, 36, 1, 209. {670581} R.W.J. Westerhout, J.A.M. Kuipers and W.P.M. van Swaaij, Industrial and Engineering Chemistry Research, 1998, 37, 3, 841. {670828} J.C.A. Martins, K.M. Novack and A.S. Gomes, Polimeros: Ciência e Tecnologia, 1996, 6, 3, 44. {671275} V.H. Tran, Journal of Macromolecular Science C, 1998, C38, 1, 1. {681645} J.C. Bernede, S. Touihri, K. Alimi, G. Safoula and D. Legoff, European Polymer Journal, 1998, 34, 2, 269. {686094} I.C. McNeill and T. Mahmood, Polymer Degradation and Stability, 1998, 60, 2-3, 449.

280

References Available from the Polymer Library {686483} M.-H. Yang, Polymer Testing, 1998, 17, 3, 191. {687290} I.C. McNeill, S. Ahmed, S. Rendall and J.G. Gorman, Polymer Degradation and Stability, 1998, 60, 1, 43. {687301} M. Coskun, Z. Ilter, E. Ozdemir, K. Demirelli and M. Ahmedzade, Polymer Degradation and Stability, 1998, 60, 1, 185. {687814} M.R. Nyden and J.W. Gilman, Computational and Theoretical Polymer Science, 1997, 7, 3-4, 191. {688694} M. Swistek, N.B. Ismail and D. Nicole, Polymer Recycling, 1997/98, 3, 1, 67. {695399} M. Dzieciol and J. Trzeszczynski, Journal of Applied Polymer Science, 1998, 69, 12, 2377. {695555} N.L. Hancox, Materials and Design, 1998, 19, 3, 93. {697436} W. Yao, B. Huang, M. Han and J. Song, China Synthetic Rubber Industry, 1998, 21, 5, 296. {699495} J.E.J. Staggs, Fire and Materials, 1998, 22, 3, 109. {699986} M. Surianarayanan, R. Vijayaraghavan and K.V. Raghavan, Journal of Polymer Science: Polymer Chemistry Edition, 1998, 36, 14, 2503. {700203} P. Rangarajan, D. Bhattacharyya and E. Grulke, Journal of Applied Polymer Science, 1998, 70, 6, 1239. {702952} A.I. Roland, M. Stenzel and G. Schmidt-Naake, Angewandte Makromolekulare Chemie, 1998, 259, 69. {704203} K. Pielichowski, J. Pielichowski and K. German, Polimery Tworzywa Wielkoczasteczkowe, 1995, 40, 5, 257. {704210} K. Pielichowski, J. Pielichowski and K. German, Polimery Tworzywa Wielkoczasteczkowe, 1995, 40, 6, 317. {704378} V. P. Nekhoroshev. L.P. Gossen, E.G. Balakhonov, Yu. P. Turov and Yu. G. Slizhov, Plasticheskie Massy, 1994, 2, 71. {704589} V.V. Zuev, P.V. Zgonnik, L.D. Turkova, V.V. Nesterov, T.A. Antonova, L.A. Shibaev and Yu. N. Sazanov, Polymer Science Series B, 1998, 40, 7-8, 236.

281

Thermal Degradation of Polymeric Materials {704611} D.M. Bate, R.S. Lehrle, C.S. Pattenden and E.J. Place, Polymer Degradation and Stability, 1998, 62, 1, 73. {705682} C. Vasile, E. Costea, M.C. Pasccu and A. Warshawsky, Journal of Thermal Analysis and Calorimetry, 1998, 52, 2, 569. {706082} M.T. de A. Freire, F.G.R. Reyes and L. Castle, Polimeros: Ciência e Tecnologia, 1998, 8, 1, 46. {706905} B. Krolikowski and R. Sojecki, Polimery Tworzywa Wielkoczasteczkowe, 1996, 41, 2, 92. {706910} S. Hirose, K. Kobashigawa, Y. Izuta and H. Hatakeyama, Polymer International, 1998, 47, 3, 247. {607567} F.D. Kopinke, M. Remmler, K. Mackenzie, M. Moder and O. Wachsen, Polymer Degradation and Stability, 1996, 53, 3, 329. {708121} A.J. Varma, S.V. Deshpande and P. Kondapalli, Polymer Degradation and Stability, 1999, 63, 1, 1. {708135} R.S. Lehrle and C.S. Pattenden, Polymer Degradation and Stability, 1999, 63, 1, 89. {708171} P. Budrugeac and E. Segal, Journal of Thermal Analysis and Calorimetry, 1998, 53, 3, 801. {709654} Xin-Gui Li and Mei-Rong Huang, Journal of Applied Polymer Science, 1999, 71, 4, 565. {709655} Xin-Gui Li, Journal of Applied Polymer Science, 1999, 71, 4, 573. {711105} I.C. McNeill and A. Alston, Angewandte Makromolekulare Chemie, 1998, 261/262, 157-72. {713906} E. Halasa and M. Heneczkowski, Polimery Tworzywa Wielkoczasteczkowe, 1998, 43, 4, 209. {720469} R. Cunko, M. Gambiroza-Jukic and E. Pezelj, Journal of Applied Polymer Science, 1999 71, 13, 2237. {724282} K. Demirelli and M. Coskun, Polymer Plastics Technology and Engineering, 1999, 38, 1, 167.

282

References Available from the Polymer Library {726737} C.S. Wang and C.H. Lin, Journal of Polymer Science: Polymer Chemistry Edition, 1999, 37, 7, 891. {730485} K.C. Khemani, Polymer Preprints, 1999, 40, 1, 625. {732351} X.-H. Li and M.-R. Huang, Polymer International, 1999, 48, 5, 387. {732433} G. Martinez, Revista de Plásticos Modernos, 1998, 76, 506, 164. {734805} S.M. Kurtz, L.A. Pruitt, D.J. Crane and A.A. Edidin, Journal of Biomedical Materials Research, 1999, 46, 1, 112. {736642} S.D. Mancini and M. Zanin, Polymer Recycling, 1997/8, 3, 3, 239. {737106} B.A. Howell and B. Pan, Proceedings of ACS Polymeric Materials Science and Engineering Meeting, San Francisco, CA, Spring 1997, 76, 401. {739308} J. Fiebig, M. Gahleitner, C. Paulik and J. Wolfschwenger, Polymer Testing, 1999, 18, 4, 257. {743554} S.V. Levchik, E.D. Weil and M. Lewin, Polymer International, 1999, 48, 7, 532. {744165} W.C. McCaffrey, D.G. Cooper and M.R. Kamal, Journal of Applied Polymer Science, 1999, 73, 8, 1415. {747400} Xin-Gui Li, Mei-Rong Huang and He Bai, Journal of Applied Polymer Science, 1999, 73, 14, 2927. {748747} K.P. Pramoda, T.S. Chung, S.L. Liu, H. Oikawa and A. Yamaguchi, Proceedings of ANTEC ‘99, New York City, NY, 1999, II, 1932. {749582} F. Duriau-Montagne, S. Pongratz and G.W. Ehrenstein, Proceedings of ANTEC ‘99, New York City, NY, 1999, II. 2383. {749597} S. Ding, M.T.K. Ling, A. Khare and L. Woo, Proceedings of Antec ‘99, New York City, NY, 1999, II, 2458. {751920} F. Aufauvre, Muanyag es Gumi, 1999, 36, 6, 183. {751929} O.F. Shlenskii, E.B. Krentsel, I. Kh. Musyaev, N.V. Minakova, V.G. Degtyarev, N.V. Medvedeva. E.M. Kulishova and Yu. V. Zelenev, Plasticheskie Massy, 1999, 3, 12.

283

Thermal Degradation of Polymeric Materials {753900} J.D. Peterson, S. Vyazovkin and C.A. Wight, Macromolecular Rapid Communications, 1999, 20, 9, 480. {755543} L. Abate, S. Calanna, G. Di Pasquale and A. Pollicino, Polymer, 2000, 41, 3, 959. {755845} A.L. Petre, P. Budrugeac and E. Segal, Journal of Thermal Analysis and Calorimetry, 1999, 56, 3, 1065. {755849} H. Abematsu, M. Tsuchiya, Y. Iseri and T. Kojima, Journal of Thermal Analysis and Calorimetry, 1999, 56, 3, 1093. {756176} E.-J. Choi, D.J.T. Hill, K.Y. Kim, J.H. O’Donnell, P.J. Pomery and A.K. Whittaker, Polymer International, 1999, 48, 10, 971. {756177} A. Babanalbandi, D.J.T. Hill, D.S. Hunter and L. Kettle, Polymer International, 1999, 48, 10, 980. {757341} T.C. Chang, C.L. Liao, K.H. Wu, H.B. Chen and J.C. Yang, Polymer Degradation and Stability, 1999, 66, 1, 127. {757742} S. Carroccio, C. Puglisi and G. Montaudo, Macromolecular Chemistry and Physics, 1999, 200, 10, 2345. {758424} R.S. Lehrle, I.W. Parsons and M. Rollinson, Polymer Degradation and Stability, 2000, 67, 1, 21. {759798} K.L.G. Ho, A.L. Pometto and P.N. Hinz, Journal of Environmental Polymer Degradation, 1999, 7, 2, 83. {760252} N. Galego and C. Rozsa, Polymer International, 1999, 48, 12, 1202. {760259} R. Bagheri, Polymer International, 1999, 48, 12, 1257. {760903} Yeong-Tarng Shieh, Hui-Tzu Chen, Kuan-Han Liu and Yawo-Kuo Twu, Journal of Polymer Science: Polymer Chemistry Edition, 1999, 37, 22, 4126. {760942} Mu-Hoe Yang, Polymer Testing, 2000, 19, 1, 105. {762844} Chin-Ping Yang and Jyh-An Chen, Polymer Journal (Japan), 1999, 31, 11, Part 1, 955. {763791} S.H. Park, J.W. Lee and D.H. Suh, Polymer Bulletin, 1999, 43, 4-5, 311.

284

References Available from the Polymer Library {764038} L.H. Perng, Journal of Polymer Science: Polymer Chemistry Edition, 1999, 37, 24, 4582. {766562} J. Sun, J. Lu, X. Zhu, K. Zhang, Z. Lu and J. Zhu, Journal of Thermal Analysis and Calorimetry, 1999, 58, 2, 301. {766615} C. Puglisi, F. Samperi, S. Carroccio and G. Montaudo, Macromolecules, 1999, 32, 26, 8821. {767672} F. Kubatovics and M. Blazso, Macromolecular Chemistry and Physics, 2000, 201, 3, 349. {768933} T.C. Chang, C.W. Yang, K.H. Wu, T.R. Wu and Y.S. Chiu, Polymer Degradation and Stability, 2000, 68, 1, 103. {769786} L.H. Perng, Journal of Polymer Science: Polymer Chemistry Edition, 2000, 38, 3, 583. {769831} J.W. Park, S.C. Oh, H.P. Lee, H.T. Kim and K.O. Yoo, Polymer Degradation and Stability, 2000, 67, 3, 535. {772093} M.S. Kim, D.J. Kim, I.R. Jeon and K.H. Seo, Journal of Applied Polymer Science, 2000, 76, 8, 1329. {774110} A. Horta, J. Coca and F.V. Diez, Advances in Polymer Technology, 2000, 19, 2, 120. {774199} L.G. Lage and Y. Kawano, Polimeros: Ciência e Tecnologia, 1999, 9, 4, 82. {774204} E.M.S. Sanchez, M.M.C. Ferreira and M.I. Felisberti, Polimeros: Ciência e Tecnologia, 1999, 9, 4, 116. {776025} B. Li, Polymer Degradation and Stability, 2000, 68, 2, 197. {776088} M. Herrera, G. Matuschek and A. Kettrup, Journal of Thermal Analysis and Calorimetry, 2000, 59, 1-2, 385. {776127} D.T. Shah, M. Tran, P.A. Berger, P. Aggarwal, J. Asrar, L.A. Madden and A.J. Anderson, Macromolecules, 2000, 33, 8, 2875. {776328} G. Boskovic, L. Katsikas, J.S. Velickovic and I.G. Popovic, Polymer, 2000, 41, 15, 5769. {776366} L. Abate, A. Pallalardo and A. Recca, Polymer Engineering and Science, 2000, 40, 5, 1114.

285

Thermal Degradation of Polymeric Materials {776415} Y.P. Baidarovtsev, G.N. Savenkov, V.V. Shevchenko and Y.M. Shulga, Polymer Science Series A, 2000, 42, 3, 287. {777330} A.C. de Albuquerque, J. Kuruvilla, L. Hecker de Carvalho and J.R. Morais d’Almeida, Composites Science and Technology, 2000, 60, 6, 833. {783421} M.J.P. Slapak, J.M.N. van Kasteren and A.A.H. Drinkenburg, Computational and Theoretical Polymer Science, 2000, 10, 6, 481. {783791} H. Soto-Valdez and J.W. Gramshaw, Journal of Materials Science Letters, 2000, 19, 10, 823. {783959} N. Manabe and Y. Yokota, Polymer Degradation and Stability, 2000, 69, 2, 183. {783960} T. Nemoto, S. Yonezawa, T. Soda and K. Takeda, Polymer Degradation and Stability, 2000, 69, 2, 191. {783965} M. Coskun, H. Erten, K. Demirelli and M. Ahmedzade, Polymer Degradation and Stability, 2000, 69, 2, 245. {784191} M. Dzieciol and J. Trzeszczynski, Journal of Applied Polymer Science, 2000, 77, 9, 1894. {784305} L. Shi, H. Saitoh and Y. Shibasaki, Journal of Polymer Science: Polymer Chemistry Edition, 2000, 38, 15, 2794. {785601} V.G. Rakova, N.M. Shchapenkova, E.A. Prudnikova and O.Y. Sabsai, International Polymer Science and Technology, 2000, 27, 6, T/28-30. {787548} T. Sawaguchi, Y. Suzuki, A. Sakaki, H. Saito, S. Yano and M. Seno, Polymer International, 2000, 49, 9, 921. {787552} B.J. Holland and J.N. Hay, Polymer International, 2000, 49, 9, 943. {787558} J. Rychly, L. Rychla and M. Strlic, Polymer International, 2000, 49, 9, 981. {787566} L. Huang and S. Wang, Journal of Applied Polymer Science, 2000, 78, 2, 237. {793664} P. Doruker, Y. Wang and W.L. Mattice, Computational and Theoretical Polymer Science, 2001, 11, 2, 155. {793817} J.L. Garcia, K.W. Koelling and J.W. Summers, Proceedings of ANTEC 2000, Orlando, FL, 2000, 86.

286

References Available from the Polymer Library {794286} R. Voelkel, T. Servay, H. Schmiedberger and S. Lehmann, Proceedings of Polyurethanes Conference 2000, Boston, MA, 2000, 191. {795751} A.I. Balabanovich, W. Schnabel, G.F. Levchik, S.V. Levchik and C.A. Wilkie, Fire Retardancy of Polymers, Royal Society of Chemistry, Cambridge, UK, 1998, 236. {798163} D.-J. Liaw, B.-Y. Liaw, J.-J. Hsu and Y.-C. Cheng, Journal of Polymer Science: Polymer Chemistry Edition, 2000, 38, 24, 4451. {798556} M.C. Cramez, M.J. Oliveira and R.J. Crawford, Proceedings of ANTEC 2000, Orlando, FL, 2000, 274. {799647} M.A. Schaffer, E.K. Marchildon, K.B. McAuley and M.F. Cunningham, Journal of Macromolecular Science C, 2000, C40, 4, 233. {800586} L.S. Semko, L.S. Dzyubenko and V.L. Kocherov, Journal of Thermal Analysis and Calorimetry, 2000, 62, 2, 485. {802281} S. Soares, N.M.P. Ricardo, F. Heatley and E. Rodrigues, Proceedings of Natural Polymers and Composites, Sao Pedro, Brazil, 2000, 222. {803123} Z. Shenmin, T. Guohua and Y. Deyue, China Synthetic Rubber Industry, 2001, 24, 1, 45. {805671} S.R. Deshmukh and F. Lai, Proceedings of ANTEC 2000, Orlando, FL, 2000, 663. {805719} S.H. Hamid, Proceedings of ANTEC 2000, Orlando, FL, 2000, 711. {805908} D. Wang and D.C.C. Lam, Journal of Materials Science Letters, 2000, 19, 23, 2109. {805917} V.V. Zuev, F. Bertini and G. Audisio, Polymer Degradation and Stability, 2001 71, 2, 213. {810866} M. Zanetti, G. Camino, P. Reichert and R. Muelhaupt, Macromolecular Rapid Communications, 2001, 22, 3, 176. {812988} L. Augh, J.W. Gillespie and B.K. Fink, Journal of Thermoplastic Composite Materials, 2001, 14, 2, 96. {814817} F. Gouin, Plastiques & Elastomères Magazine, 2000, 52, 7, 33.

287

Thermal Degradation of Polymeric Materials {815959} J.D. Peterson, S. Vyazovkin and C.A. Wight, Macromolecular Chemistry and Physics, 2001, 202, 6, 775. {815984} Der-Jang Liaw, Pei-Nan Hsu, Jao-Jin Chen, Been-Yang Liaw and ChiehYing Hwang, Journal of Polymer Science: Polymer Chemistry Edition, 2001, 39, 10, 1557. {816008} Guey-Sheng Liou and Sheng-Huei Hsaio, Journal of Polymer Science: Polymer Chemistry Edition, 2001, 39, 10, 1786. {817822} Xue-Song Wang, Xin-Gui Li and Deyue Yan, Polymer Testing, 2001, 20, 5, 491. {817912} C.N. Kartalis, C.D. Papaspyrides, R. Pfaendner, K. Hoffmann and H. Herbst, Polymer Engineering and Science, 2001, 41, 5, 771. {820253} S.H. Park, J.W. Lee, D.H. Suh and S.Y. Ju, Journal of Macromolecular Science A, 2001, A38, 5-6, 513. {820329} C.A. Wilkie, Polymer News, 2001, 26, 3, 92. {820840} K. Demirelli, M. Coskun and E. Kaya, Polymer Degradation and Stability, 2001, 72, 1, 75. {822782} W.J. Sterling, Y.-C. Kim and J. McCoy, Industrial and Engineering Chemistry Research, 2001, 40, 8, 1811. {824066} Mu-Hoe Yang, Polymer Testing, 2000, 19, 1, 105. {825031} T. Fujimura, N. Sarugaku, M. Tsuchiya, K. Ishimaru and T. Kojima, Journal of Thermal Analysis and Calorimetry, 2001, 64, 2, 425. {825036} Z. Gao, W. Xie, J.M. Hwu, L. Wells and W.P. Pan, Journal of Thermal Analysis and Calorimetry, 2001, 64, 2, 467. {825040} P.S. Thomas, J.P. Guerbois, G.F. Russell and B.J. Briscoe, Journal of Thermal Analysis and Calorimetry, 2001, 64, 2, 501. {825385} Q. Xiang, M. Xanthos, S. Mitra and S.H. Patel, Proceedings of ANTEC 2001, Dallas, TX, 2001, 289. {825464} L. Fambri, A. Pegoretti, C. Gavazza and A. Penati, Journal of Applied Polymer Science, 2001, 81, 5, 1216.

288

References Available from the Polymer Library {826911} M. Dzieciol and J. Trzeszczynski, Journal of Applied Polymer Science, 2001, 81, 12, 3064. {827029} K.C. Khemani, Proceedings of ANTEC 2001, Dallas, TX, 2001, 402. {827195} A.M. Rdissi and S.L. Simon, Proceedings of ANTEC 2001, Dallas, TX, 2001, 465. {827220} Yeh Wang, H.-C. Chan, S.-M. Lai, H.-F. Shen and Y.-K. Hsiao, Proceedings of ANTEC 2001, Dallas, TX, 2001, 490. {829337} Li-Hsiang Perng, Journal of Applied Polymer Science, 2001, 81, 10, 2387. {829658} R. Bacaloglu and U. Stewen, Journal of Vinyl and Additive Technology, 2001, 7, 3, 149. {830034} R. Bacaloglu and U. Stewen, Proceedings of ANTEC 2001, Dallas, TX, 2001, 602. {831060} D.J. Harris, R.A. Assink and M. Celina, Macromolecules, 2001, 34, 19, 6695. {831619} E.S. Trofimchuk, M.Y. Yablokova, N.I. Nikonorova, A.V. Antonov, A.L. Volynskii and N.F. Bakeev, Polymer Science Series B, 2001, 43, 7-8, 190. {832485} G. Botelho, A. Queiros, S. Liberal and P. Gijsman, Polymer Degradation and Stability, 2001, 74, 1, 39. {832500} L.M. Rincon-Rubio, B. Fayolle, L. Audouin and J. Verdu, Polymer Degradation and Stability, 2001, 74, 1, 177. {833611} E.V. Kalugina, T.N. Novotvortseva, M.B. Andreeva, A.B. Blyumenfel’d, Ya. G. Urman and V.V. Gur’yanova, Plasticheskie Massy, 2001, 38, 4, 11. {838693} S. Villar-Rodil, J.I. Paredes, A. Martinez-Alonso and J.M.D. Tascon, Chemistry of Materials, 2001, 13, 11, 4297. {845096} M. Watanabe, T. Adschiri and K. Arai, Kobunshi Ronbunshu, 2001, 58, 12, 631. {845098} T. Moriya and H. Enomoto, Kobunshi Ronbunshu, 2001, 58, 12, 661. {845462} S.A. Razzak and S.A. Jabarin, Polymer International, 2002, 51, 2, 164.

289

Thermal Degradation of Polymeric Materials {845530} Jyh Ming Hwu, G.J. Jiang, Zong Ming Gao, Wei Xie and Wei Ping Pan, Journal of Applied Polymer Science, 2002, 83, 8, 1702. {845557} Xin-Gui Li, Mei-Rong Huang and He Bai, Journal of Applied Polymer Science, 2002, 83, 9, 1940. {847519} V. Karmore and G. Madras, Industrial and Engineering Chemistry Research, 2002, 41, 4, 657. {848090} Xin-Gui Li, Mei-Rong Huang, He Bai and Yu-Liang Yang, Journal of Applied Polymer Science, 2002, 83, 10, 2053. {848609} A. Nalbandi, Iranian Polymer Journal, 2001, 10, 6, 371. {849787} Yu. A. Sangalov, N.A. Krasulina and A.I. Il’yasova, Plasticheskie Massy, 2001, 7, 39. {851392} B.J. Holland and J.N. Hay, Polymer, 2002, 43, 8, 207. {852663} Y. Aoyagi, K. Yamashita and Y. Doi, Polymer Degradation and Stability, 2002, 76, 1, 53. {852665} Mu-Hoe Yang, Polymer Degradation and Stability, 2002, 76, 1, 69. {853072} E.V. Kalugina, T.N. Novotortseva and M.B. Andreeva, International Polymer Science and Technology, 2002, 29, 4, T/51. {853278} Z. Czegeny, E. Jakab and M. Blazso, Macromolecular Materials and Engineering, 2002, 287, 4, 277. {855971} S. Carroccio, C. Puglisi and G. Montaudo, Macromolecules, 2002, 35, 11, 4297. {856011} A. Rivaton, B. Mailhot, J. Soulestin, H. Varghese and J.-L. Gardette, European Polymer Journal, 2002, 38, 7, 1349. {857011} Bing Zhang and F.D. Blum, Polymer Preprints, 2002, 43, 1, 484. {860011} Mu-Hoe Yang, Journal of Applied Polymer Science, 2002, 85, 8, 1698. {860868} B.A. Howell, Y. Cui and D.B. Priddy, Polymer Preprints, 2002, 43, 1. 360. {861966} A. Jain and K. Vijayan, Journal of Materials Science, 2002, 37, 13, 2623.

290

References Available from the Polymer Library {862952} V.M. Yanborisov and K.S. Minsker, Polymer Science Series A, 2002, 44, 5, 538. {863996} B.J. Holland and J.N. Hay, Polymer Degradation and Stability, 2002, 77, 3, 435. {864006} B. Fayolle, L. Audouin, G.A. George and J. Verdu, Polymer Degradation and Stability, 2002, 77, 3, 515. {864583} P. Mariani, G. Carianni and A. Monteverdi, Proceedings of 2001 8th European Polymers, Films, Laminations and Extrusion Coatings Conference, Barcelona, Spain, 2001, 5. {865094} L. Woo, C.L. Sandford and S.Y. Ding, Polymer Preprints, 2001, 42, 1, 394. {865095} E. Chamot, Polymer Preprints, 2001, 42. 1, 396. {865099} W.H. Starnes, V.G. Zaikov, L.B. Payne, Y. Li and X. Ge, Polymer Preprints, 2001, 42, 1, 404. {865102} L. Audouin, B. Fayolle and J. Verdu, Polymer Preprints, 2001, 42, 1, 412. {865110} E.B. Orler, D.A. Wrobleski, D.W. Cooke, B.L. Bennett, M.E. Smith. M.S. Jahan and M. King, Polymer Preprints, 2001, 42. 1, 428. {865214} S. Jipa, L.M. Gorghiu, I. Mihalcea, T. Zaharescu, R. Setnescu, T. Setnescu and M. Dumitru, Materiale Plastice, 2002, 39, 2, 81. {866651} V.M. Yanborisov and K.S. Minsker, Polymer Science Series B, 2002, 44, 56, 127. {867499} M.G. Lu, J.Y. Lee, M.J. Shim and S.W. Kim, Journal of Applied Polymer Science, 2002, 85, 12, 2552. {868537} Mu-Hoe Yang, Journal of Applied Polymer Science, 2002, 86, 7, 1540. {868543} T. Andersson, B. Wesslen and J. Sandstrom, Journal of Applied Polymer Science, 2002, 86, 7, 1580. {870313} Der-Kuen Tsay, Mu-Hoe Yang and Jenn-Hwa Wang, Polymer Degradation and Stability, 2002, 76, 2, 251. {870570} C. Puglisi, F. Samperi, S. Carroccio and G. Montaudo, Polymer Preprints, 41, 1, 680.

291

Thermal Degradation of Polymeric Materials {870572} S. Carroccio, C. Puglisi and G. Montaudo, Polymer Preprints, 2000, 41, 1, 684. {870655} R. Aguado, M. Olazar, B. Gaisan, R. Prieto and J. Bilbao, Industrial and Engineering Chemistry Research, 2002, 41, 18, 4559. {870944} S.K. Young, G.C. Gemeinhardt, J.W. Sherman, R.F. Storey, K.A. Mauritz, D.A. Schiraldi, A. Polyakova, A. Hiltner and E. Baer, Polymer, 2002, 43, 23, 6101. {871062} A. Karaduman and E.H. Simsek, Journal of Polymers and the Environment, 2001, 9, 2, 85. {871496} L. Abate, I. Blanco, A. Pollicino and A. Recca, Journal of Thermal Analysis and Calorimetry, 2002, 70, 1, 63. {871758} J. Kim, W.I. Lee and S.W. Tsai, Composites Part B: Engineering, 2002, 33B, 7, 531. {871835} V.M. Yanborisov, K.S. Minsker, G.E. Zaikov and V.G. Zaikov, Journal of Vinyl and Additive Technology, 2002, 8, 3, 176. {871997} H.-Y. Yen, F.-S. Lee and M.-H. Yang, Polymer Testing, 2003, 22, 1, 31. {871999} C.N. Caseaval, D. Rosu, L. Rosu and C. Ciobanu, Polymer Testing, 2003, 22, 1, 45. {872441} M.H. Yang, D.K. Tsay and J.H. Wang, Polymer Testing, 2002, 21, 7, 737. {872766} C. Soykan and M. Ahmedzade, Polymer Degradation and Stability, 2002, 78, 3, 497. {875580} L. Rosu, C.N. Cascaval, C. Ciobanu and D. Rosu, Materiale Plastice, 2002, 39, 3, 183. {876727} S. Jahromi, W. Gabrielse and A. Braam, Polymer, 2003, 44, 1, 25. {877171} A.A. Donskoi, M.A. Shashkina, G.E. Zaikov and R.M. Aseeva, Kauchuk i Rezina, 2001, 4, 11. {877977} A.C. de Souza, A.T.N. Pires and V. Soldi, Journal of Thermal Analysis and Calorimetry, 2002, 70, 2, 405. {878335} J.C. Michel, Proceedings of ANTEC 2002, San Francisco, CA, 2002, 500, 5.

292

References Available from the Polymer Library {878940} S. Hirose, T. Hatakeyama, Y. Izuta and H. Hatakeyama, Journal of Thermal Analysis and Calorimetry, 2002, 70, 3, 853. {879069} S.I. Stoliarov, P.R. Westmoreland, M.R. Nyden and G.P. Forney, Polymer, 2003, 44, 3, 883. {881010} R.P. Singh, S.M. Desai and G. Pathak, Journal of Applied Polymer Science, 2003, 87, 13, 2146. {881232} A. Horta, J. Coca and F.V. Diez, Advances in Polymer Technology, 2003, 22, 1, 15. {882333} M.M. Jackson and S. Bullard, Proceedings of ANTEC 2002, San Francisco, CA, 2002, 598. {882592} P. Lee-Sullivan, D. Dykeman and Q. Shao, Polymer Engineering and Science, 2003, 43, 2, 369. {882593} D. Dyleman and P. Lee-Sullivan, Polymer Engineering and Science, 2003, 43, 2, 383. {882869} M.J. Caulfield, X. Hao, G.G. Qiao and D.H. Solomon, Polymer, 2003, 44, 5, 1331. {883154} T. Kirchmann, Kunststoffe Plast Europe, 2003, 93, 2, 32. {883185} G. Sivalingam and G. Madras, Polymer Degradation and Stability, 2003, 80, 1, 11. {883234} Lecon Woo, Y.S. Ding and C.L. Sandford, Polymer Preprints, 2001, 42, 2, 866. {883236} I.W. Parsons and R.S. Lehrle, Polymer Preprints, 2001, 42, 2, 869. {883695} J.-C. Huang and X. Junke, International Journal of Polymeric Materials, 2003, 52, 3, 203. {884042} Xin-Gui Li, Mei-Rong Huang, Zhi-Liang Zhu, Yi Jin and Xue-Song Wang, Journal of Applied Polymer Science, 2003, 88, 4, 1065. {884128} S.H. Kim, W.K. Son, Y.J. Kim, E.-G. Kang, D.-W. Kim, C.W. Park, W.-G. Kim and H.-J. Kim, Journal of Applied Polymer Science, 2003, 88, 3, 595. {884331} K. Wondraczek, J. Adams and J. Fuhrmann, Macromolecular Chemistry and Physics, 2002, 203, 18, 2624.

293

Thermal Degradation of Polymeric Materials {884544} B.M. Ginzburg, L.A. Shibaev, V.L. Ugolkov and V.P. Bulatov, Journal of Macromolecular Science B, 2003, B42, 1, 139. {885413} M. Hamskog, G. Ahlblad, G. Faernert, P. Gijsman and B. Terselius, Polymer Testing, 2003, 22, 4, 363. {886162} T. Ishikawa, T. Ohkawa, M. Suzuki, T. Tsuchiya and K. Takeda, Journal of Applied Polymer Science, 2003, 88, 6, 1465. {886302} H. Yu, N. Huang, C. Wang and Z. Tang, Journal of Applied Polymer Science, 2003, 88, 11, 2557. {886343} S.O. Han, D.W. Lee and S.K. Woo, Polymer Preprints, 2002, 43, 2. 1359. {886353}

M.A. Brebu, Y. Sakata and M.A. Uddin, Handbook of Polymer Blends and Composites, Volume 3B, Rapra Technology Ltd., Shrewsbury, 2003, 561 & 627.

{887512} Z. Gao, T. Kaneko, I. Amasaki and M. Nakada, Polymer Degradation and Stability, 2003, 80, 2, 269. {887517} L. Abate, I. Blanco, A. Orestano, A. Pollicino and A. Recca, Polymer Degradation and Stability, 2003, 80, 2, 333. {887655} P.V. Joseph, K. Joseph, S. Thomas, C.K.S. Pillai, V.S. Prasad, G. Groeninckx and M. Sarkissova, Composites Part A: Applied Science and Manufacturing, 2003, 34A, 3, 253. {888003} L. Tang, H. Huang, Z. Zhao, C.Z. Wu and Y. Chen, Industrial and Engineering Chemistry Research, 2003, 42, 6, 1145. {888059} K. Saido, Y. Kodera, H. Taguchi, K. Tomono, Y. Ishihara and T. Kuroki, Polymer Preprints, 2002, 43. 2, 1162. {888209} K.P. Pramoda, T.X. Liu, Z.H. Liu, C.B. He and H.J. Sue, Polymer Preprints, 2002, 43. 2,1217. {888522} L.S. Shibryaeva, O.V. Shatalova, A.V. Krivandin, O.B. Petrov, N.N. Korzh and A.A. Popov, Polymer Science Series A, 2003, 45, 3, 244. {888765} S. Tajika, A. Sugita, K. Uemoto and S. Tasaka, Kobunshi Ronbunshu, 2003, 60, 3, 101. {889249} A. Karadauman, M.C. Kocak and A.Y. Bilgesu, Polymer Plastics Technology and Engineering, 2003, 42, 2, 181.

294

References Available from the Polymer Library {889475} C. Tang, Y.-Z. Wang, Q. Zhou and L. Zheng, Polymer Degradation and Stability, 2003, 81, 1, 89. {889478} Z. Gao, I. Amasaki, T. Kaneko and M. Nakada, Polymer Degradation and Stability, 2003, 81, 1, 125. {889529} Y.M. Kruse, H.-W. Wong and L.J. Broadbelt, Industrial and Engineering Chemistry Research, 2003, 42, 12, 2722. {890075} Naian Liu, Rowen Zhong, Lifu Shu, Jianjun Zhou and Weicheng Fan, Journal of Applied Polymer Science, 2003, 89, 1, 135. {890179} B. Havenith, Plastiques et Elastomères Magazine, 2002, 54, 7, 61. {890265} S. de Goede, R. Brüll, H. Pasch and N. Marshall, Macromolecular Symposia, 2003, 193, 35. {890883} F.P. La Mantia and N.T. Dintcheva, Macromolecular Symposia, 2003, 194, 277. {891434} F.P. La Mantia and A. Correnti, Progress in Rubber, Plastics and Recycling Technology, 2003, 19, 3, 135. {891557} J.M. Hutchinson, Journal of Thermal Analysis and Calorimetry, 2003, 72, 2, 619. {891577} J. Hacaloglu, I. Athar, L. Toppare and Y. Yagci, Journal of Macromolecular Science A, 2003, A40, 6, 605. {893098} M. Sairam, B. Sreedhar, D.V.M. Rao and S. Palaniappan, Polymers for Advanced Technologies, 2003, 14, 7, 477. {893672} K. German and K. Kulesza, Polimery, 2003, 48, 5, 337. {893911} S.M. Dakka, Journal of Thermal Analysis and Calorimetry, 2003, 73, 1, 17. {894608} W. Endres, M.D. Lechner and R. Steinberger, Macromolecular Materials and Engineering, 2003, 288, 6, 525. {894646} I. Erol, C. Soykan, Z. Ilter and M. Ahmedzade, Polymer Degradation and Stability, 2003, 81, 2, 287. {894653} R.A. Ruseckaite and A. Jimenez, Polymer Degradation and Stability, 2003, 81, 2, 353.

295

Thermal Degradation of Polymeric Materials {895401} Heng Lin, Jun Zhang, Qi Dong and Chunfang Zhang, China Synthetic Rubber Industry, 2003, 26, 4, 254. {895472} P.J. Yoon, D.L. Hunter and D.R. Paul, Polymer, 2003, 44, 18, 5341.

296

Index

Index 2,2-bis(4-(4-maleimidophenoxy)phenyl) hexafluoropropane 202 2,2-bis(4-(4-maleimidophenoxy)phenyl) propane 202 2,2-bis(4-maleimidophenyl)methane 201 2,2-bis(4-maleimidophenyl) ether 201

A Abbreviations 4 ABS degradation products 59 overview 59 with bean oil 60 with hindered amine stabilisers 24 Acetaldehyde 47 Acrylic copolymers 104 Acrylic polymers 97 Acrylonitrile-cellulose 112 Acrylonitrile copolymers 110 Activation energy 10, 39 Adhesives 76, 80, 103 Ageing 19, 59 Ammonium polyphosphate 167, 242 Aramid 157, 163, 189 spinning 239 Aromatic polymers 189 Arrhenius relation 38, 157 PE 242 VDC/VC 244 Artificial neural networks 244 Avrami-Erofeev model 40, 87

B Backbiting 43 Bean oil 60, 217 Biodegradable polyesters 91 Biodegradable polymers 140 Bisphenol-A polysulfone 120 Black liquor 223 Blowing agent 86 Bond cleavage 32 Bond dissociation energy 59 Bond scission 30 Butadiene rubber 52 Butyl rubber 53

C Carbon-fibre composites 157 Carbon black 169 Casein 143 Cellulose 133, 223 Cellulose acetate 103 Ceramics 180, 184, 185 Chain scission 47 Chatterjee-Conrad method 162 Chitin 130 Chitosan 131 Chlorinated polyethylenes 45 Coatings 80, 129 Coats–Redfern method 162 Cold ring fraction 19 Composites 153 Computer modelling 241 Cone calorimeter 25

297

Handbook of Biodegradable Polymers Controlled-transformation-rate thermal analysis 37 Controlling degradation 2 Copper/carbon composite 158 Corrosion 194 Cottonseed proteins 143 Cyclic olefin copolymers 50 Cyclodextrins 133

D Degradation mechanisms 236 Dehydrochlorination 218 PVC waste 229 VDC 116 Depolymerisation 32 Derivatisation techniques 15 Diene elastomers DSC 20 Diene polymers overview 50 Differential scanning calorimetry chitosan 131 cottonseed 145 overview 19 PA-6 recycling 221 PC 125 PE/carbon black 169 PMMA nanocomposite 171 poly(amide-hydrazide) 190 polybismaleimide 201 PU composite 163 starch 129 Diglycidyl ether of bisphenol-A DGEBA/PCL 96 Disproportionation 30, 44

E Ecological issues 230 Electrical properties 169

298

Electron impact ionisation 16 Electron spin resonance 24 Environment 1, 59, 192 EPDM 52 melamine reinforced 168 PA/EPDM 79 PP/epoxy blend 207 Epoxy-amine 239 Epoxy resin 206 siloxane blend 206 Ethylene-propylene rubber 45 Ethynyl phenyl azofunctional resin 204 Evolved gas analysis 12, 16 Expandable graphite 87 Extrusion 233 PP 235 Extrusion coating 47

F Fire-retardant 86, 135 Flammability 2, 83 Flynn-Wall method 11, 39, 70, 180 Food packaging 47 Fourier Transform Infrared Spectrometry cellulose 136, 137 chitosan 131 PAN 110 PEEK 197 phenolic resin 204 poly(styrene sulfone) 119 PPAMA 106 proteins 143 PS/PPE 62 PVA 238 Fuel recovery 223, 228

G G value 242 Gasification lignocellulose 223

Index Gas chromatography-mass spectrometry PA-6,6 34 unsaturated polyester 161 Gas constant 39 Gel permeation chromatography HDPE processing 234 PP 212 VTES copolymers 182 Ginstling–Brounshtein model 40 Glass-fibre composites 153 thermal cycling 156 unsaturated polyester 161 XRD 154 Glass transition temperature 50, 61, 164

H High density polyethylene DSC 20 extrusion 234 modelling 243 thermal recycling 230 High impact polystyrene PVC/HIPS 70 Hydrogen abstraction 49

I Inorganic polymers 173 Interpenetrating polymer networks 84 Ionisation techniques 16 Irregular structures 66, 235 Isothermal degradation 11

J Jander equation 40 Johnson-Mehl-Avrami model 40

K Kenaf 138 Kinetic analysis 10, 37 mixed plastic waste 225 PE 45 kinetic models 39 Kissinger method 193

L Lignin degradation products 140 molecular structure 139 overview 138 Lignocellulose 136 thermal recycling 222 volatilisation 224 Linear low density polyethylene composites 215 wood composite 167 Low density polyethylene degradation in solution 215 modelling 243 processing 235 thermal recycling 229

M MALDI-TOF overview 22 PC 124 PU foams 88 MAS-NMR PVC/epoxidised NR 69 MD_REACT 245 Melamine 168 Melamine cyanurate 75 Micro-thermal analysis 27 Microscopic methods 25

299

Handbook of Biodegradable Polymers Mixed plastics waste pyrolysis 225 Modelling 241 Molecular weight distribution 245 Molic mouse model 243 Monomers 30, 32, 58 Monte Carlo simulation 245 Montmorillonite 164 Multilayer polymer particles 109 Municipal solid waste 229

N Nanocomposites overview 153 PMMA/SiO2 102 PU 164 Nanofibres 159 Nanotubes 159 Natural polymers 129 Natural rubber 53 degradation products 144 mechanism 146 PA blend 78 PVC blends 69 Py-GC 145 pyrolysis oil 148 Nitrile-butadiene rubber degradation products 51 melamine reinforced 168 overview 51 pyrograms 53 pyrolysate 51 Nuclear magnetic resonance 23 PDMS 175 PHV 142 PIB 49

O Octa(aminophenyl)silsesquioxane 188

300

Organic-inorganic hybrid polymers 184 Oxidation induction time 19 Ozawa-Flynn-Wall kinetic analysis 70 Ozawa method 38, 110

P PA-12 degradation products 75 nanocomposites 160 PA-6 191 caprolactam 73 caprolactam recovery 221 cis-elimination 74 degradation products 73 nanocomposite 171 PA-6/natural rubber 78 PP/PA-6 Processing 236

thermal recycling 221 PA-6,10 degradation products 75 PA-6,6 20 cyclopentanone 75 degradation in products 75 Py-GC-MS 75 thermooxidation 34 PA-7 191 Peroxide 33, 235 PHBHV 140 Phase transitions 129 Phenol-formaldehyde resin 204 Phenolic resin 203 epoxy blend 206 mechanism of degradation 205 silica hybrid 184 pH value 237 Photooxidation 34, 90 Pine needles 135 Plasma 210 Plasticiser 219

Index Polyacetal DSC 20 Poly(acrylamide sulfone) 120 Polyacrylate chromatogram 104 degradation products 106 mechanism of degradation 104 Polyacrylonitrile 110 Polyamide 72, 191 activation energy 79 aromatic 189 blends 78 cotton fibre blends 165 dynamic TG 10 fluorinated 191 kinetic data 75 liquid crystal 77 PA/EPDM 79 thermal property data 191 thermal recycling 220 Poly(amide-hydrazide) 190 Polyaniline composites 169 Poly(arylene ether) 193 Polybenzimidazole 197 Polybenzoxazine 202 Polybismaleimide 199 Poly(bisphenol-A carbonate) 22 Poly(p-bromophenacyl methacrylate) 106 Polybutadiene 60 Poly(n-butyl acrylate) 104 Poly(t-butyl methacrylate) 106 Poly(butylene terephthalate) mechanism of degradation 90 overview 125 Polycarbonate aromatic 192 degradation products 124 DSC 20, 125 MALDI-TOF 124

mechanism of degradation 123 overview 123 PS/PC 62 Poly( -caprolactam) 96 Poly( -caprolactam-co-caprolactone) 96 Poly( -caprolactone) cellulose interactions 96 degradation products 93 DGEBA 96 mechanism of degradation 92,94 TG-MS 95 Polycyanurate 89 Poly(3-(1-cyclohexyl)azetidinyl methacrylate) 107 Polydichlorophosphazene 179 Poly(dimethyldiphenylsiloxane) 174 degradation products 175 Poly(dimethylsiloxane) degradation products 175 mechanism of decomposition 175 nanocomposite 172 stabilisers 175 TG 176 Poly(p-dioxanone) 150 Poly(2,2-dioxybiphenylphosphazene) 178 Poly(epichlorohydrin-co-ethylene oxide) 126 Polyester bacterial 141 mechanism of degradation 89 melt spinning 239 overview 89 Poly(ester amide) 76, 77 Polyether aromatic 193 Poly(ether ether ketone) FTIR 197 overview 195 PES blends 197 Poly(ether imide) 207

301

Handbook of Biodegradable Polymers Poly(ether ketone) 126 Poly(ether sulfone) 120 Poly(ethyl acrylate) 104 Polyethylene 41 breakdown products 47 carbon black composites 169 chemiluminescence 45 dynamic TG 10 free-radical degradation 41 hydrogen production 44 modelling 242 processing 234 thermal recycling 228 Poly(ethylene glycol allenyl methyl ether) 126 Poly(ethylene sulfide) 121 Poly(ethylene terephthalate) degradation products 90 hydrolysis of waste 224 mechanism of degradation 90 overview 90 processing 238 Polyhedral oligosilsesquioxane epoxy modified 186 PS 165 PU 164 Poly(hydroxyalkanoate) 140 Poly(3-hydroxybutyrate) 140 mechanism of degradation 91 Poly(2-hydroxyethyl methacrylate) 107 Poly(3-hydroxyoctanoate-co-3-hydroxy10-undecanoate) 141 Poly(3-hydroxyvalerate) 141 NMR spectrum 142 Polyimide 78 Polyisobutylene overview 48 random scission 48 Polyisoprene 144 Poly(isoprenyl acetate) 18

302

Poly(isopropenyl acetate) 113 Poly(itaconic acid) 89 Poly(D-lactide) 149 Poly(L-lactide) 148 blends 149 degradation products 149 depolymerisation 222 mechanism of degradation 93 modelling 245 thermal recycling 221 Polymeric binder 239 Polymer blends 21, 60, 97 Poly(2-methacrylamidopyridine) 107 Polymethacrylonitrile 110 Poly(p-methoxyphenacyl methacrylate) 106 Poly(methyl methacrylate) activation energy 100 blends 103 cellulose acetate/PMMA 103 degradation products 100 DTA 2 initiation 101 mechanism of degradation 99 modelling 243 nanocomposites 102, 171 overview 97 phosphazene copolymers 178 pure vs. industrial 2 PVC/PMMA 68 rate of degradation 98 silica hybrid 185 siloxane/PMMA 103 siloxane copolymer 104 tacticity 35,102 TG 102 thermooxidation 34 vinyl terminated 109 Poly(methylphenylsiloxane)-PMMA 104 Poly(-methyl styrene) thermal recycling 225

Index Poly(methylthienyl methacrylate) 109 Polyolefin grafting 236 heat ageing 19 overview 41 peroxide degradation 33 Poly(olefin sulfone) 117, 118 Poly(phenacyl methacrylate) 106 Poly((2-phenyl-1,3-dioxolane-4-yl)methyl methacrylate) 106 Poly(p-phenylene benzobisoxazole) 203 Poly(phenylene ether) 194 blends with PS 61 degradation products 195 modelling 243 Poly(m-phenylene isophthalamide) 189 SEM 25 Poly(1,4-phenylene-1,3,4-oxadiazole) 195 Poly(phenylene sulfide) 20, 194 Polyphosphazene 177 PMMA copolymers 178 Poly(n-propyl acrylate) 104 Polypropylene chain scission 20, 47 composite nanotube 171 sisal 167

EPDM/epoxy blend 207 extrusion 235 functionalisation 235 GPC 212 mechanism of MA grafting 235 modelling 242 overview 47 oxidation 20 oxygen induction time 21 PP/PA-6 processing 236

recycled 213 Polypropylene glycol 86 Polysilane 180

Polysilazane 180 TG EGA 183 Polysiloxane mechanism of pyrolysis 174 overview 173 PVC blends 71 Polystyrene 53 backbiting 55 bean oil 217 blends 60 blends with PPE 61 brominated 54,57 composite POSS 165 sisal 167

depolymerisation 56 dynamic TG 10 effect of poly(-methylstyrene) 58 initiation 56 mechanism of degradation 55 melt pyrolysis 216 modelling 244 monomer yield 54 poly(p-chloromethylstyrene) 56 PS/PC 62 PS/PMMA 62 regioregularity 56 solution degradation 216 TG 56 thermal recycling 215 Poly(styrene-co-methacrylonitrile) 110 Poly(styrene sulfide) 121 Poly(styrene sulfone) 57, 118 FTIR spectra 119 Poly(2-sulfoethyl methacrylate) 106 Polysulfone 58 DSC 20 FTIR spectra 119 overview 116 Poly(2-(3-(6-tetralino)-3-methylcyclobutyl)2-ketoethyl methacrylate) 107

303

Handbook of Biodegradable Polymers Polyurethane chlorinated degradation products 82

chlorinated PU 70 composites 162 montmorillonite 164 POSS 164 SiC 164 TiO2 164

decomposition mechanism 80 degradation products 80 elastomers 20 mica-filled 162 overview 79 pre-ceramic blend 185 thermoplastic - see thermoplastic PU 83 Polyurethane foams blowing agents 86 cone calorimeter 88 degradation products 87,88 flexible 88 FR effects 86 overview 85 rigid 85 Polyurethane, thermoplastic 3-chloro-l,2-propanediol flame retardant 83 composites 84 degradation products 83 Poly(vinyl alcohol) 115 processing 237 PVC/PVA 68 solution pH 237 stabilisation 237 Poly(vinyl acetate) blends 113 mechanism of degradation 113 overview 112 PVAc/siloxane 114 PVC/PVAc 71

304

thermooxidation 34 Poly(vinyl butyral) PVC/PVB 69 Poly(vinyl chloride) aromatic hydrocarbon formation 63 benzene formation 63 blends 68 defect sites 62 degradation products 63 dehydrochlorination 33,63 double-bond concentration 67 DSC 20 dynamic TG 10 epoxidised natural rubber blend 69 initiation 66 ion current chromatograms 67 mechanism of decomposition 217 overview 31,62 plasticiser 219 PVC/chlorinated PU 70 PVC/HIPS 70 PVC/PMMA 68 PVC/PVA 68 PVC/PVAc 71 PVC/PVB 69 Py-GC-MS 66 pyrolysis yields 219 siloxane blends 71 TG-MS 14 thermal recycling 217,228 thermooxidation 33 Poly(vinyl triethoxysilane) 181 Pre-exponential factor 39 Processing 233 Proteins 143 Prout-Tompkins equation 40 Pyrolysis 14 Pyrolysis-gas chromatography 14 polyisoprene 145 thermoplastic PU 84

Index Pyrolysis-gas chromatography-mass spectrometry 16 HCl detection from PVC 66 PA-6,6 75 PES 122 poly(epichlorohydrin) 127 poly(epichlorohydrin-co-ethylene oxide) 127 PU foams 86 Pyrolysis-mass spectrometry 15 acryl copolymers 109 PBT 125 with MALDI 22

R Random scission 32 Rate of reaction 38 Red mud 229 Refuse-derived fuel 229 Rice husks 136 Rubber 33, 78

S Scanning electron microscopy PP processing 236 PU/ceramic 186 Schultz function 245 Shear 133, 244 Side-group elimination 31 Silica hybrids 184 Silicon carbonitride 180 Silicon oxycarbide 185 Siloxane crosslinking 114 epoxy blend 206 PMMA copolymer 104 PVAc/siloxane 114 siloxane/PMMA 103 Sisal 167

Sodium caseinate 143 Solid-phase microextraction 23 Solution degradation 215 Stages of degradation 30 Starch 129 degradation products 130 processing 238 Styrene-butadiene rubber degradation products 51 overview 50 oxidative degradation 51 TG 50 thermal degradation 51 Styrene-2,4-dinitrostyrene copolymer 57 Styrene-isoprene-styrene copolymer 58 Styrene-maleic acid copolymer 57 FTIR 57 MS 57 Sulfide-containing polymers 120 Supercritical water SBR depolymerisation 51 Sustainability 231 Synthetic rubber 213 TG 214

T Temperature-modulated DSC overview 20 Thermal analysis methods 3 Thermal degradation 1 definition 1 types 1 Thermal volatilisation analysis overview 18 Thermochemical recycling 220 Thermogravimetry ABS 59 overview 10 PE/carbon black 169

305

Handbook of Biodegradable Polymers PU composites 162 Thermogravimetry-evolved gas analysis polysilazane 183 Thermogravimetry-Fourier Transform Infrared Spectroscopy 12, 59 Thermogravimetry-mass spectrometry lignin 140 overview 13 PCL 95 thermoplastic PU 83 Thermooxidative degradation 33 Thermoplastic PU 83 3-chloro-l,2-propanediol flame retardant 83 composites 84 degradation products 83 Thiophene-capped poly(methyl methacrylate) 109 Total ion chromatogram 16 Toxicity 162, 230 Triethyl phosphate 87 Trimethylchitosan 131 Twin-screw extruder 233, 236

U Ultrasonic techniques 25 Unsaturated polyester degradation products 161 Urethane elastomers DSC 20

V Vacuum pyrolysis mixed plastics waste 230 Vinylidene chloride copolymers 115, 116 Vinylidene chloride-vinyl chloride copolymer 244 Vinyl chloride-vinyl acetate copolymer NMR 68

306

TG 68 Viscometry 238 Volatile oligomers 48

W Weathering 29, 59 Wide-angle X-ray scattering POSS copolymers 188 Wesslau function 245 Wood fibre 167

X X-ray 27 X-ray photoelectron spectroscopy lignin 140 STP-PMMA 179 X-ray diffraction glass-fibre composites 154

Z Zimmer AG process 221

ISBN: 1-85957-498-X

Rapra Technology Limited Rapra Technology is the leading independent international organisation with over 80 years of experience providing technology, information and consultancy on all aspects of rubbers and plastics. The company has extensive processing, testing and analytical facilities. It provides testing to a range of national and international standards and offers UKAS accredited analytical services. Rapra also undertakes commercially focused innovative research projects through multi-client participation. Rapra publishes books, technical journals, reports, technological and business surveys, conference proceedings and trade directories. These publishing activities are ed by an Information Centre which maintains and develops the world’s most comprehensive database of commercial and technical information on rubbers and plastics.

Shawbury, Shrewsbury, Shropshire SY4 4NR, UK Telephone: +44 (0)1939 250383 Fax: +44 (0)1939 251118 http://www.rapra.net

Thermal Degradation Of Polymeric Materials 6a5m11

Overview 26281t

More details 6y5l6z

Related Documents 3h463d

Thermal Degradation Of Polymeric Materials 6a5m11

Degradation Of Materials 1h3t3q

Degradation Of Materials 1h3t3q

Fundamental Principles Of Polymeric Materials 696a6a

Thermal Stress Strength Of Materials Review c5l1d

More Documents from "Juan David Ramirez" 2o71p

[free-scores.com]_lecuona-ernesto-gitanerias-de-suite-espagnole-635.pdf 96924

Manual_rnec_supernumerarios.pdf 4z402y

Pequenos-preludios-y-fugas-bach.pdf 3u396c

Guitarra Clasica 13705e

Thermal Degradation Of Polymeric Materials 6a5m11

Formula Para Sacar La Nota 6f1p49