Green Composites: Polymer Composites and the Environment

i

Green composites Polymer composites and the environment Edited by Caroline Baillie

CRC Press Boca Raton Boston New York Washington, DC

WOODHEAD

PUBLISHING LIMITED Cambridge England

ii Published by Woodhead Publishing Limited, Abington Hall, Abington Cambridge CB1 6AH, England www.woodhead-publishing.com Published in North America by CRC Press LLC, 2000 Corporate Blvd, NW Boca Raton FL 33431, USA First published 2004, Woodhead Publishing Ltd and CRC Press LLC © 2004, Woodhead Publishing Ltd The authors have asserted their moral rights. This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. Reasonable efforts have been made to publish reliable data and information, but the authors and the publishers cannot assume responsibility for the validity of all materials. Neither the authors nor the publishers, nor anyone else associated with this publication, shall be liable for any loss, damage or liability directly or indirectly caused or alleged to be caused by this book. Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming and recording, or by any information storage or retrieval system, without permission in writing from the publishers. The consent of Woodhead Publishing and CRC Press does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific permission must be obtained in writing from Woodhead Publishing or CRC Press for such copying. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation, without intent to infringe. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. Library of Congress Cataloging in Publication Data A catalog record for this book is available from the Library of Congress. Woodhead Publishing ISBN 1 85573 739 6 CRC Press ISBN 0-8493-2576-5 CRC Press order number: WP2576 The publisher’s policy is to use permanent paper from mills that operate a sustainable forestry policy, and which have been manufactured from pulp which is processed using acid-free and elementary chlorine-free practices. Furthermore, the publisher ensures that the text paper and cover board used have met acceptable environmental accreditation standards. Typeset by Replika Press, Pvt Ltd, India Printed by TJ International Ltd, Cornwall, England

Contents

1

Contributor contact details

viii

Why green composites?

1

C . BAILLIE , Queen’s University, Canada

1.1 1.2 1.3 1.4

Introduction An environmental footprint and life cycle assessment Drivers for change The structure of this book: a life cycle approach

1 1 3 5

2

Designing for composites: traditional and future views

9

C . ROSE , University of Brighton, UK

2.1 2.2 2.3 2.4

3

Introduction: design thinking The three principles of development and the value system The big challenge: the future of material consumption, utilisation and innovation The use of composite materials through the ages: design, form and structure Sources of further information References

9 11

20 22 22

Life cycle assessment

23

14

R . MURPHY , Imperial College London, UK

3.1 3.2 3.3 3.4 3.5

Introduction Life cycle assessment: methodology LCAs of composite materials Future trends: making use of LCA Conclusions Sources of further information Acknowledgements References

23 24 35 43 46 46 46 47 iii

iv

4

Contents

Natural fibre sources

49

T. NISHINO , Kobe University, Japan

4.1 4.2 4.3 4.4 4.5 4.6 4.7

5

Introduction The microstructure of natural plant fibres The crystal structure of celluloses The crystal modulus of natural fibres The mechanical properties of cellulose microfibrils and macrofibrils Natural fibre/sustainable polymer composites Future trends References

49 49 54 56 65 68 74 76

Alternative fibre sources: paper and wood fibres as reinforcement

81

P . PELTOLA , Tampere University of Technology, Finland

5.1 5.2 5.3 5.4 5.5

6

Introduction and definitions Wood fibres: structure, properties, making pulp and paper fibres Recycling of paper Wood and plastic composites and the theory of fibre reinforcement Composites made of wood or wood fibre and plastics Acknowledgements References Alternative solutions: recyclable synthetic fibre–thermoplastic composites

81 83 87 90 92 98 98

100

R . A . SHANKS , RMIT University, Australia

6.1 6.2 6.3 6.4 6.5 6.6 6.7

Introduction and definitions Green composites and the structure and function of composites Natural material sources: reconstitution of thermoplastic polymers and the effect of water Synthetic recyclable composites Processing innovations and mineral-filled composites Properties of single polymer fibre–matrix composites Future trends Sources of further information and advice Acknowledgements References

100 101 104 105 108 114 119 120 120 121

Contents

7

Natural polymer sources

v

123

D . PLACKETT , Risø National Laboratory, Denmark and A. VÁZQUEZ,

Universidad Nacional de Mar del Plata, Argentina

7.1 7.2 7.3

7.4 7.5

7.6

8

Introduction: biocomposites and biodegradable polymers Polylactides: polylactic acid (PLA) synthesis, properties, biodegradation, processing and applications Polyhydroxyalkanoates: polyhydroxyalkanoate (PHA) synthesis, properties, biodegradation, processing and applications Starch-based polymers: properties, biodegradation, processing and applications Bio-based composites: mechanical properties, processing, characterisation, modification, water absorption, biodegradation and reinforcement Future trends Sources of further information References

123

135 147 148 149

Optimising the properties of green composites

154

124

128 132

S . H . AZIZ and M . P . ANSELL , University of Bath, UK

8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8

9

Introduction Thermosetting matrices versus thermoplastic matrices: a comparison Selecting natural fibres for composites: stress transfer and physical characteristics Case study: natural fibre composites with thermosetting resin matrices Mechanical properties of composites as a function of design Dynamic mechanical thermal analysis (DMTA) of long fibre composites Environmental stability of natural fibre composites Discussion and conclusions Sources of further information and advice Acknowledgements References

154

173 176 177 178 178 178

Green fibre thermoplastic composites

181

155 161 164 165

M. SAIN and S. PANTHAPULAKKAL , University of Toronto,

Canada

9.1 9.2

Introduction: biofibre production Green fibres for composite production

181 183

vi

Contents

9.3 9.4

Thermoplastics for natural fibre composites High performance fibres: thermal, chemical and mechanical treatments Processing of natural fibre-filled composites The performance and durability of natural fibres Environmental benefits of using natural fibre-reinforced thermoplastics Future trends References

201 202 203

Clean production

207

9.5 9.6 9.7 9.8

10

187 189 192 197

N . TUCKER , University of Warwick, UK

10.1 10.2 10.3 10.4 10.5 10.6

11

Introduction: clean processing Energy saving in the manufacture and production of composites Limiting the environmental impact of processing The use of additives End-of-life disposal strategies Future trends References

207 210 215 222 226 230 231

Applications

233

M . HUGHES , University of Wales, UK

11.1 11.2 11.3 11.4 11.5

12

Introduction and definitions Historical applications of green composites Contemporary applications of green composites Future trends Conclusions Sources of further information and advice References

233 235 237 244 247 248 249

Re-use, recycling and degradation of composites

252

A . HODZIC , James Cook University, Australia

12.1 12.2 12.3 12.4 12.5 12.6 12.7 12.8

Introduction Recycling of polymers and composites Recycling of thermoplastic composites Recycling of thermosetting composites Degradation of polymers: UV light and biodegradation Recycling of composites in the automotive industry Utilising green composites and incinerating polymers Conclusions and future trends References

252 254 255 257 260 263 264 267 268

Contents

13

Reprocessing

vii

272

J . C . ARNOLD , Swansea University, UK

13.1 13.2 13.3 13.4 13.5

Introduction Management of waste plastics and composites Methods of sorting and separating plastics and polymers Methods of recycling plastics Future trends Sources of further information References

272 273 277 283 294 295 296

Index

301

Contributor contact details

(*= main point of contact)

Chapter 1 Professor C. Baillie Faculty of Applied Sciences Fleming Hall Stewart Pollock Wing, Room 307 Queen’s University Kingston Ontario Canada K7L 3N6 Tel: 1 613 533 6249 Fax: 1 613 533 2721 Email: [email protected]

Chapter 2 C. Rose, M. Des. RCA Three Dimensional Design & Materials Practice School of Architecture & Design University of Brighton Grand Parade Brighton BN2 2JY UK Tel: +44 (0)1273 643080 Fax: +44(0)1273 694561 Email: [email protected]

viii

Contributor contact details

Chapter 3 Dr Richard J. Murphy Department of Biological Sciences Sir Alexander Fleming Building Imperial College London South Kensington Campus London SW7 2AZ UK Tel: +44 (0)20 7594 5389 Fax: +44 (0)20 7594 5390 Email: [email protected]

Chapter 4 Dr Takashi Nishino Department of Chemical Science and Engineering Faculty of Engineering Kobe University Rokko, Nada, Kobe 657-8501 Japan Tel: +81 (0)78 803 6164 Fax: +81 (0)78 803 6205 Email: [email protected]

Chapter 5 Piia Peltola, M.Sc. Tampere University of Technology Materials Science, Plastics and Elastometer Technology P.O. Box 589 33101 Tampere Finland Email: [email protected]

Chapter 6 Professor Robert A. Shanks Department of Applied Chemistry, RMIT University GPO Box 2476V Melbourne 3001, Victoria Australia Tel and Fax: +61 3 9925 2122 Email: [email protected]

ix

x

Contributor contact details

Chapter 7 Dr David Plackett* Danish Polymer Centre Risø National Laboratory Frederiksborgvej 399 4000 Roskilde Denmark Tel: +45 4677 5487 Fax: +45 4677 4791 Email: [email protected] Dr A. Vázquez Research Institute on Material Science and Technology (INTEMA) Engineering Faculty Universidad Nacional de Mar del Plata Juan B Justo 4302 7600 Mar del Plata Argentina Tel: +54 223 481 6600 Fax: +54 223 481 0046 Email: [email protected]

Chapter 8 Dr Martin P. Ansell* Department of Engineering and Applied Science University of Bath Bath BA2 7AY UK Tel: +44 (0)1225 386432 Fax: +44 (0)1225 386098 Email: [email protected] Ms Sharifah H. Aziz Malaysian Institute for Nuclear Technology Research (MINT) Bangi, 4300 Kajang Malaysia Tel: +60 (0) 3 8925 1510 Fax: +60 (0) 3 8920 2968 Email: [email protected]

Contributor contact details

Chapter 9 Dr M. Sain* and Dr S. Panthapulakkal Centre for Biocomposites and Biomaterials Processing Faculty of Forestry University of Toronto 33 Willcocks Street, Toronto Canada M5S 3B3 Email: [email protected]

Chapter 10 Dr Nick Tucker International Automotive Research Centre Warwick Manufacturing Group University of Warwick Coventry CV4 7AL UK Tel: +44 (0) 24 7652 2499 Email: [email protected]

Chapter 11 Dr Mark Hughes The BioComposites Centre University of Wales, Bangor Gwynedd LL57 2UW UK Tel: +44 (0) 1248 370 588 Fax: +44 (0) 1248 370 594 Email: [email protected]

Chapter 12 Dr Alma Hodzic School of Engineering James Cook University Townsville QLD 4811 Australia Tel: +61 7 4781 5082 Fax: +61 7 4781 4666 Email: [email protected]

xi

xii

Contributor contact details

Chapter 13 Dr Cris Arnold Materials Research Centre School of Engineering University of Wales Swansea Singleton Park Swansea UK Tel: +44 (0) 1792 295 749 Email: [email protected]

1 Why green composites? C. BAILLIE Queen’s University, Canada

1.1

Introduction

Often when pursuing research into green composites we say that we are protecting the environment, that we are working for nature. We may as well stop kidding ourselves – nature will be fine; nature will work out OK and adapt to the changes. It’s humans that will cease to exist if we continue the way we are at present. Some scientists and engineers have realised that they need to take responsibility for the outcome of their work. Researching ways of creating faster machines and bigger toys, without due consideration of the effects on the environment or on people, is irresponsible. This book represents some of the workers who have, over the last 10 years or so, decided to change the direction of their research to address some of these issues. We have recently been seeing an increase in the number of researchers working in this area and it is time to reflect on the progress and purpose of our work to make sure that we are in fact doing what we say we would like to do.

1.2

An environmental footprint and life cycle assessment

In this context we are defining green composites as composites that are designed with the lowest environmental ‘footprint’ possible. Furthermore, we are focusing on fibre-reinforced polymer composites in this book as these are the most abundant material group of the composite family in use. In this chapter we will explore some of the assumptions we make and consider the life cycle of such materials, not only from ‘cradle to grave’ but beyond the grave into the after life. What do we mean by an environmental footprint? What factors must be considered? We consistently hear the terms ‘green’, ‘eco’, ‘sustainable’, ‘environmentally conscious’, ‘life cycle’, ‘clean’ and assume we know what is meant by them. We also often label our work with these terms in order to 1

2

Green composites

generate funding from governmental bodies who in turn use the terms to satisfy their promises. But is anyone actually making any difference to the damage we are doing to our planet? We need to consider the impact that our material choice and design will have on the society and the environment (Rose and Baillie – Navigating the Materials World, Academic Press, 2003). Life cycle assessment is defined as ‘an objective process to evaluate the environmental burdens associated with a product, process or activity by identifying energy and materials used and wastes released to the environment, and to evaluate and implement opportunities to effect environmental improvements’ (Society of Environmental Toxicology and Chemistry, SETAC, Code of Practice, 1991). I dispute the objective part of this definition. Life cycle assessments are so difficult, often so subjective in their evaluations and so complex, that they are frequently ignored, or taken as an add on at the end of a project. Many companies insist on carrying out an LCA or life cycle analysis before their design can be realised. An environmental LCA helps us to quantify how much energy and material are used and how much waste is generated at each stage of a product’s life. The analysis takes place first but, after this, the life cycle assessment needs to take place, which is where the interpretation and value judgements come in, e.g. ∑ ∑ ∑ ∑

Is it worse to use up more energy in transport or to produce more factories? Is it worse to burn and create harmful gases or to create landfill? Is it worse to dump or to use up energy in recycling? Is it worse to have the risk of food poisoning or waste food or increased packaging?

Assessment of the impact on the environment is therefore considered at each stage: resources, production, distribution, use, disposal or re-use. Many such LCAs have actually proved the project to have been a waste of energy in itself. In a recent report from a CRAFT project to produce natural fibre composites in which I was involved (CRAFT project European Commission report (BES2-3163), 2000) we compared natural mat- (NMT) and glass mat-reinforced thermoplastics (GMT). A comparison was made between GMT and NMT manufactured by a current production method of prepreg followed by compression moulding into an automotive and nonautomotive part. The LCA analysis was performed for both the non-automotive and automotive part and for three different types of performance requirements, i.e. stiffness, strength and impact resistance. The results of LCA showed that, for most cases, the environmental impact of NMT material is higher than the reference GMT. Despite their ‘green’ image, natural fibre composites are not necessarily an environmentally friendly alternative to glass fibre

Why green composites?

3

composites in applications where strength and impact are the performance drivers. The reasons for this include the need for pesticides and other chemicals to grow the flax fibre and the higher fibre loadings needed to fulfil the impact and strength criteria. Even in the case of stiffness-based non-automotive applications, where no higher fibre loadings are needed to meet the performance requirements, the differences in environmental impact between GMT and NMT are very small and in the current analysis the poorest performance was shown for NMT due to the negative effect of pesticides. In the case of automotive and stiffness critical applications NMT composites do seem to perform better than GMT composites. This reduction in environmental impact for NMT composites is, however, mainly due to the lower weight of natural fibre composite parts, which leads to lower fuel consumption during the use of composites and not so much the result of the use of ‘green’ natural fibres. Hence, the advantages of natural fibre composites are relevant if weight savings are obtained over glass fibre composites.

1.3

Drivers for change

From the above it would seem as if only those researchers who are altruistic would be involved in this game. In fact, there are many drivers for the change we currently observe. Global concerns are considered by the Kyoto Protocol, national concerns by government legislation, and local companies in turn make a response to the legislation. All of this will influence funding mechanisms and availability of funds for researchers. Global responses to climate change are critical at the time of writing where countries are deciding upon the Kyoto Protocol. Reports of a lecture by Michael Grubb of Imperial college to the Royal Institute of international Affairs (The Independent, 5 November, 2003) quote him as saying: the US is starting to pour billions of dollars into research on technologies like carbon sequestration and hydrogen. Unfortunately, pursuing climate technology while eschewing emission caps is like designing a fancy car while opposing all efforts to put an engine in it... . Governments are not good at delivering industrial technologies: there has to be a market for them. The argument may be said to be true of green composites in automotive and other applications. In Canada at present there is a push to make a new profitable market from the use of agricultural fibres in composites. Economic arguments alone will not cause this technology to take off. Rumour has it that local industry may not become interested in changing to more ecofriendly products because, even if the Canadian Government encouraged them to do so with legislation, they would simply sell ‘over the border’. One of the largest markets for natural fibre composites is the US automotive

4

Green composites

sector who currently do not appear to have the same drivers as the European market. The EU Directive for automotive parts has meant that many companies in Europe have started to consider environmentally friendly alternatives to fuel and materials for production. The Directive stipulates that re-use and recycling of end-of-life vehicles must increase to 95% by 2015. Further details are given by Tucker in Chapter 10. He tells us that disassembly is a concern and costs associated with dismantling plastic components from cars are too high at present. Tucker would suggest ‘design for disassembly’. The European Community approach to waste management is based on two complementary strategies: avoiding waste by improving product design and increasing the recycling and re-use of waste. The EU Landfill Directive (1999) states that, by 2010, the amount of biodegradable municipal waste going to landfill must be reduced by 75% of the total produced in 1995. The EU Packaging Waste Directive (1997) states that there must be 50–60% recovery and 25–45% recycling by 2006. For a city like London, this means that there must be alternative routes for waste of between 2 and 4 million tonnes per year. The UK has responded by developing its Government Waste Strategy (2000). This states that 30% of all waste produced must be recycled or composted by the year 2010. Furthermore, the UK has responded to the EU Directives by bringing in a landfill tax system that forces companies to think about end of use. Hence, response to EU Directives creates a huge driver and support for research in the green materials area. In the US the Environmental Protection Agency Office of Solid Waste ‘supports’ reaching a goal of 35% recycling by 2005 (American Plastics Council, 2001, National Post Consumer Plastics Report). Implementation of legislation is, of course, another huge factor. Hence it might be that Liverpool in the UK cannot ask its citizens to take responsibility for the environment as they do not have a collection system in place and the locals would use more energy in fuel driving to the nearest collection point in Chester. Kingston, Ontario in Canada, on the other hand, represents one town that has a well-established curb side collection system in place. However, it is rumoured that much of the plastic waste gets sent to other countries as an export. Those countries obviously do not have enough of their own waste.

1.3.1

Swings and roundabouts

We cannot pour money into environmental research until legislation reinforces the changes that ensue. We cannot encourage participation of profit driven corporations unless legislation supports such changes. Furthermore, we cannot create environmental schemes that rely on the participation of communities of people who produce waste when the infrastructure is not there to support the schemes. The drivers for, and the national and global economic,

Why green composites?

5

environmental and social impact of, new directions in research must all be considered before seriously trying to effect change in material usage.

1.4

The structure of this book: a life cycle approach

In the light of the above we have decided to arrange this book with a life cycle approach. We consider first the choice of materials that iterates with the design and function or the application. We consider the factors affecting the life cycle analysis. We look at possible fibres to use as reinforcement, as well as potential polymer matrices. In this latter category we need to consider thermoplastics as a source which may be recycled, or as a nonvirgin source; composites as a means of upgrading recycled polymers as well as thermosets which need to be re-used or biodegradable thermosets which degrade. We will look at processing of the composite; its production into components, trying in the meantime not to forget in our discussions the transport of raw materials and parts at each stage and the energy consumed in the production run. We consider a range of applications – replacement for GFRP (glass fibre-reinforced polymers) as well as upgrading of polymer components. Finally, we look at the re-use, recovery and recycling of the composites that we have made. We start the process by reflecting upon ‘Design thinking’. Rose takes us on a philosophical pathway through notions of progress and a reflection on our values to the idea of learning from nature about ‘ecologically responsible design’. Murphy takes pity on the reader and helps us place this in a more solid context by leading us through those issues which affect the life cycle of a natural fibre composites product. Read this chapter even if you dip in and out of the others. It is necessary to consider all issues impacting our ‘footprint’ if we are to take ourselves and our research seriously when we say we are working on ‘green composites’. The choice of fibre is determined by all of the usual selection criteria such as ultimate mechanical properties, interfacial adhesion with the matrix (high or low for strength or toughness applications respectively), cost, availability of resources, chemical properties, resistance to moisture, etc. It is also determined by the ‘green’ criteria – whatever those might be. Nishino discusses the potential for natural fibres in composites. They are divided into wood, vegetable, animal and mineral. We have seen different trends in different parts of the world for the uptake of certain fibres. Abundant good quality fibre sources will determine this to an extent and we see far more wood fibres used in North America and more agricultural fibres in Europe. However, it is also lifestyle and marketing that will determine these directions. Hughes discusses the fact that the predominant market for wood plastic composites (WPC) is in decking in North America, whereas the European market for vegetable or agricultural fibres is the automotive sector.

6

Green composites

Nishino’s Table 4.6 shows us the main merits and demerits of using natural fibres. Sustainability of natural resources – or the ability to re-grow on an annual basis, together with low energy consumption and biodegradability make them a good environmental or ‘green’ choice so long as the remainder of the life cycle is properly assessed. For example, we need to think about transport of fibres, moisture resistance, interfacial adhesion, etc. Often, we see arguments comparing glass with natural fibres and suggesting that the only way for natural fibres to be adopted is if their properties can be proved to be ‘as good as’ or better than glass. However, it is also stressed that, unless the new system is cheaper, it will not be adopted. In other words people are not prepared to change anything. This is not the way of nature, which continually adapts to new conditions. The benefits of natural fibres are quite different to those of glass fibres and they should be used to create materials that will better suit the needs of our environment and communities of the future. If we look to see how an ecosystem works, we see that everything is used at its optimal function and no more. There is no waste of material or energy. Peltola focuses more on paper fibres. He states that in the US over 300 kg per capita of paper is used per year. In Europe, of the 35 million tonnes produced in 2002, 3 million tonnes are recycled and 14 million tonnes are used as energy. His argument is that we should divert paper fibre from landfill to a useful composite of waste fibre and plastic. An alternative environmental solution for the choice of fibres is presented by Shanks. He presents the work of researchers who are working with fibre/ matrix of the same thermoplastic material so that, at the end of life, the entire mass can be melted to form a new product. All polypropylene (PP) composites, for example, have great potential environmentally. Even though they do not come from a renewable source – composites made from all PP have at least one more life before incineration or energy recovery. Now we need to turn to the matrix. Of course, it is never possible to deal with the fibre and matrix independently, and all authors do address many other aspects than the one they have been asked to put in focus. In consideration of the choice of matrix, in a way similar to that for the fibre we can select natural sources for sustainability or recyclable polymers. We can also use biodegradable polymers. Thermoplastics are the first matrix of choice as they are recyclable. However, once we have added fibres, they are often rendered non-recyclable so the argument is not well made. Only in the case of all polymer composites such as all PP can the end-of-life be recycling. Adding natural fibres will be better than glass as the composite could be incinerated for energy recovery. However, a better alternative, as discussed earlier, is the use of recycled polymer as source. The addition of fibres will upgrade the polymer. The main thermosets considered are biodegradable. Plackett explains, however, that natural polymer matrices are at varying

Why green composites?

7

stages of development and much work is needed to improve stability, processability, mechanical properties and degradability. Alongside the choice of fibre and matrix will come the choice of processing method – very different for thermoplastics and thermosets. As considered by Aziz and Sain, thermoplastic injection moulded parts can be complex but fibres are short and stiffness low as a result. Furthermore, high temperatures are often needed for adequate viscosity and natural fibres can be damaged. Aziz argues that thermosetting composites have better properties: stiffness can be increased by using longer fibres; interfacial properties will be better because of longer processing times and better fibre wetting. These factors must be weighed against production costs (parts per hour produced) and environmental impact. Natural thermosetting polymers seem to come out quite well in many of the arguments and Aziz considers cashew nut shell liquid resin as an example. Clean processing is discussed more thoroughly by Tucker. He uses Thorpe’s (1999 in Tucker) definition: ‘a way to reverse our current non-sustainable use of materials and energy’. He tells us that we should move manufacturing from a linear use of resources into the cyclical use of resources that do not produce waste products. He also points out that cleaner production methods must satisfy the triple bottom line: environmental, social and economic. He quotes Thorpe’s suggestions for clean production, helping us think about the need rather than the desire for products in the first place, the need for durability rather than for recycling, and reduction of consumption whilst maintaining quality of life. He really seems to balance the three aspects above, but will companies in the current economic system wish to respond? Once we have selected the materials and production method appropriate for our application, and the properties required, whilst keeping our environmental impact as low as possible, we must consider the potential for end of use. Already, we have seen this consideration but we now bring this aspect into focus. If we find we will be producing something with one life before landfill, we need to think again. Hodzic reminds us that plastic waste may be reduced to its monomers, recycled or biodegraded. Thermoplastics are not without their problems as they require a high degree of purity of recyclate as well as separation of the matrix from the reinforcement except in the case of all PP, for example. Thermosets are more problematic, as discussed previously, but some chemical additives can help to degrade polymers, although Hodzic warns against possible side effects to the environment. Chemical additives must always be considered in the light of our ecosystem approach. Wherever possible it would be better to use something present within the system to carry out the desired function, as nature would do, rather than adding an alien element that will have a counter effect. Arnold takes us through the whole process of recycling and reprocessing.

8

Green composites

His concern echoes my own – that increases in recycling rates (or use of green composites) will not occur simply by market forces. He stresses that ‘design for recycling’ is becoming more widespread where legislation is impending as in the automotive sector. This takes us back to the beginning of the book, to Rose’s chapter. The cycle begins again, but hopefully next time we will think and act more responsibly and gradually reduce our own footprint as researcher, manufacturer, engineer. . . .

2 Designing for composites: traditional and future views C. ROSE University of Brighton, UK

2.1

Introduction: design thinking

What visual and sensory signatures distinguish ‘green’ design principles? What ‘misplaced’ imagery or features confuse or obscure these? Achille Castiglione, the great Italian designer, has observed that aesthetics show you the intention of the designer. Our aesthetic response to objects, features and structures in our built environment is largely a matter of conditioning and of a conditioned reaction to history. At a deep level, aesthetic ‘language’ is a component of constructed meaning, and it is this constructed meaning that we experience alongside our actual sensory experience in the moment of using or encountering things. The work of the designer is that of combining the above challenges with the practical issues of material availabilities and properties, of manufacturing techniques, of creative and problem-solving concepts, and with the many apparently conflicting pressures within the total picture of the business of product design. This phenomenon makes it a difficult challenge to answer the question, ‘what does ecologically responsible design look like?’ sadly because there is little history connected to it. Difficult, because the history of design is the history of repeated metaphors, of images and appearances copied from one era to the next in differing guises. Alternatively, if not copied, then reacted against. Reacting against something does not guarantee improvement, but rather just changing appearances, often leaving the underlying and flawed processes unaddressed. Old metaphors will not help us. The history of design imperatives has been, we now see, the history of narrow perspectives, of exploitation, and of grandiosity. You would think that the eighteenth-century orthodoxy regarding ‘nature’, i.e. a raw, immoral force to be subdued and contained, or where preferred, eliminated altogether, by burning, felling, killing, poisoning, etc., would be an orthodoxy no longer in fashion. This convenient characterisation betrays the crudeness of those means of intervention available to people of that persuasion. The character of this mode of engagement with ‘nature’ is in 9

10

Green composites

keeping with the attitude that underpins it. That these activities continue largely unconstrained, not to say actively promoted by some governments, is a shocking contradiction to any notion we may cherish that things are ‘not as bad now’; we can see that the old stupid orthodoxy continues actually to underpin the action involved in our ‘use’ of nature, despite all the talkingbased developments of ‘eco-awareness’. The debate is way ahead of the action. We have been hearing about waste, about misuse of resources and about our dependence upon the living environment for the whole of our lives, yet there is almost no action. Buckminster Fuller observed that an important idea for society took 25 years to register with people, and that a really important idea took 50 years. He said that in 1974, so it seems we have time yet. It is time now, if ever, to notice the accuracy; the style; the fitness for purpose; the ‘form follows function’; the drama; the delight; and above all the value systems in nature; to emulate them, to profit from them, to be informed by them. We begin by noticing what is there. It has been shown that we see with our brains and our memory and ideas as much as we see with our eyes. Let’s get our ideas shaping up to what we are looking at, when we look at natural examples. In fact, technology is giving us the tools to see these attributes of nature and to be so much less ignorant, but we have to change our thinking and our imperatives truly to see with these tools, otherwise we will continue only to see what we always have; that is, how to make a quick buck at someone else’s expense. Except now, the someone else is you and me. Just as we have the power to understand nature’s materials in greater complexity, we will always be faced with the same choices – will it be ‘slash and burn’ but on a nano-scale? Will we really study at this level or simply subvert it before attempting to understand it? All spheres of knowledge present us with a blank canvas upon which to display our value system. In the material world our track record is not good. In the present state of industrial consumption, it is time for a change in more than just the appearances of things. In 1972 the Ecologist (www.theecologist.org) publication set out a ‘Blueprint for Survival’ and in 2002 the 30th anniversary edition of the same (vol. 32, no. 7) outlined many measures against which it could be established that conditions were significantly worse than in 1972. This, despite a number of notable ‘advances’ and isolated incidents of successful ecological improvement. However, bluntly, the main thrust of manufacture and consumption is unquestionably set for nonsustainability, and more significantly for expansion of the non-sustainable model in the ‘developing’ world. In contrast, natural systems have specialised in creating solutions, some defiant, some fantastical, some quiet and near invisible, all from conflicting factors – there is no other way in nature. What can we learn from this?

Designing for composites: traditional and future views

11

One of the elementary exercises at art school is to study drawing an object, say a stool, by drawing only ‘everything-but’ the stool, i.e. the shape of the spaces around the subject. This is a powerful method of developing genuine observation, relying as it does upon an activity. We are so familiar with the stool in everyday life that we tend to draw a stylised representation of something we already hold in memory, which process defeats the ability to observe and draw what in fact is in front of you at the present moment. In this sense, thinking only in symbols separates us from the subject and the present, connecting us instead only to a pre-existing attitude. By analogy, the principle of this study technique applies to our research activities, which must explore ‘around’ a subject in order to begin to understand the relational properties and the ‘life’ the subject has. In this way a more informed, less thoughtless design process can be supported. The significance of this concerning our conventional value system for materials is explored below. But what would be the obvious and stark result of applying this to everyday examples of contemporary materials consumption, in an attempt to establish a conceptual framework that could lead us towards a more lifesupporting value system regarding materials?

2.2

The three principles of development and the value system

2.2.1

Three principles of development

Now that we are developing the technology to design composite material properties across the entire spectrum from nano- to micro- to macro-scale, in terms of fabrication and assembly techniques, we have less excuse to fail to address the key challenges of conventional production, namely: ∑ no waste, a discipline applied at all stages of design, appraisal, manufacture, use, re-use, recycling and disposal ∑ environment considered at every stage (constraints observable from living processes) ∑ ‘valuable’ as a term has to gain a revised definition old definition – rare, exotic, not easy to obtain, special, prized, trade gains, kill to get it new definition – preserves or enhances the environment and life processes. Attempting to apply these concepts and deal with the consequences will surely produce the conflict with our present knowledge framework that is required to jump start creativity, utilise our senses and map out the response space available to us.

12

2.2.2

Green composites

An obsolete value system

Despite a generation of commentary and research to the contrary, we remain stuck with an obsolete value system when it comes to the manufacture and distribution of materials, products and produce. The exotic material fixation that is a leftover of empire remains to this day. The working philosophy of ‘think global – act local’ has hardly begun to make an impact. Much ecomarketing cynically manipulates that genuine interest that exists in the public sphere, in order to continue the old polluting or poisoning processes for maximum profit. What profit? Here is the remarkable paradox, where in an isolated value system the term ‘valuable’ applies to short-term money within a rigged gameplan, ignoring all the attendant consequences. Here is where money is valued more than the environment. The balance sheet looks good because of all the actual, ‘real world’ consequences whose value or significance has been discounted, and whose destructive and uncontrollable effects are ignored or excused by an increasingly self-absorbed value system. Something can be regarded as ‘too expensive’ in this scheme – because it is being compared to something detached from the actual ecosystem, for the convenience of managing a process within an outdated set of parameters. The very process of ‘birthing’– of life creation within natural energy systems, is being noticeably curtailed within this business model. It is a contemporary phenomenon that metaphors of progress sourced from the rapid changes in manufacturing processes that have occurred since the nineteenth century are now generally applied, quite uncritically, across most social and working situations. Much misery and stress among employees and users is created by these means. It is not only the environment that suffers from our present manufacturing ethos. Its very underpinning philosophy is capable of subverting our thinking. ‘Doing more with less’ is a great idea when applied to improvements in the rate at which a certain product can be turned out. But the metaphor fails when applied to, say, a Beethoven string quartet. It took four people and 41 minutes to perform in 1826, the same in 1958, and no doubt will be much the same in 2014. The string quartet in question (Beethoven, Late String Quartet, Op. 132, 1825) reached a steady state of technological development, and no amount of innovation or creativity will bring about a condition in which 2.8 people can do the same job in 13 minutes. Metaphors are evidently inescapable in the language of research, but knowing where they do, and where they do not apply is remarkably challenging. ‘Prolonged empathetic study’ as described by Goethe1 is necessary. For lessons in ‘what is valuable’ or ‘how expensive’ something is, we need to study natural processes. Natural processes of fabrication of substance, of components, of structures, are all responsive, i.e. there are relational properties at play in the varying factors of the environment affecting the resulting patterns, formations, shapes and compositional factors. This can

Designing for composites: traditional and future views

13

be seen as having a consequence for each of the ‘principles’ above (no waste, relate to environment, local available resources support life). Secondly, as established by Goethe through his principle of ‘prolonged empathetic study’ there is both an appropriate timeline for all life processes and a patterned set of adjacent phenomena without which an understanding of the subject is not possible. An intelligent study takes these formation and deformation sequences and dynamic contexts as characteristics of the research process, which must ‘track’ these characteristics in order to question our experience associated with them. The ‘subject’ is seen more as process made up of characteristic stages, not a conceptual snapshot of an object in isolation from such a process or ‘fate’. A leaf is a good example of this principle. A real leaf is an adventure of changing form, changing process, of response to harsh conditions and to damage and attack, of aerodynamics and thermal control, of chemistry and physics, of deployment and retraction, of disposability and recycling, even of environmental roles for other life forms after its connection with the tree is ended. Natural systems are multipurpose. Does its connection with the tree ever end? Does the tree leave the leaves or do the leaves leave the tree? The history of the making of ‘things’ has given us all too often a preoccupation with the object itself – its style, identity and possession of it, rather than its role in a continuing narrative. The implication is that of independent existence, i.e. (the business plan) is constructed in a manner devoid of any connection with life processes, and fits perfectly the desire for ‘identity’ to be captured within the unique purchasable product. This could be seen as the endgame of the Victorian notion of nature as ‘foreign’ to the rightly minded person, now clearly an essentially vain construct. One of the main examples of this, and an indication of just how archaic this value system is, is the notion that as the possessions that were once the preserve of the royal figurehead are now available for public admiration, so society must be getting better now that we can all have those things. Thus the ‘exotic’ artefact, by way of ever more crude iterations, becomes the ubiquitous possession. The royal beneficence flows down the cascade of society, enriching all. Further than this, now that brand image itself has been established as a ‘virtual’ attribute, devoid of material substance, it is possible to coerce individuals and groups into purchasing literally nothing, yet continue to generate pollution and waste in the process. Any beneficial content that was supposed to be present in this trading system has become a chimera, despite the fact that the burden of waste upon life processes continues in no less real a manner than at any time previously. What a remarkable transformation in the meaning of the term ‘industry’ this is. The appalling consequences of the narrative that has unfolded from this outdated and unthinking value system is only now impinging upon our consciousness; the mountains of computer waste in China, the daily barges

14

Green composites

of post-consumer waste queuing to escape Manhattan island en route for the ocean, to make space for more ‘pound-shop’ produce, the increasing familiarity with the term landfill in our news, you cannot miss it. When you think of the term ‘land-fill’, what effect does it have upon you? The bizarre phenomenon of the disposable drink container that lasts longer than a house yet has 18 seconds of primary use, if nothing else, could make you wonder what we could achieve if we concentrated on a conceptual picture for product design that included all the players and their timeline in a more integrated scenario.

2.3

The big challenge: the future of material consumption, utilisation and innovation

The big challenge for designers concerning contemporary imperatives in material consumption, utilisation and innovation, is that a century of attitudes towards materials – the ‘exotic’, the ‘rare’, the ‘special’ – (epithets that belong to an age of exploitation and vanity, with attempts to emulate royalty, conquest, etc.,) has left us with an inappropriate vocabulary for contemporary material science. The easy reiteration of accepted orthodoxy, i.e. to do what we did before but do it faster, while being one of the features of computer-aided manufacture, is not the feature we are bound to adopt. The choice is ours. In a now dated yet significant critique of overly contrived ordering in city life, author Richard Sennett (in The Uses of Disorder 2) makes a significant observation about the differing relationships possible between ourselves and the materials we use: In pre-industrial workshop production systems, the experience of making a product was more important than a predetermined standard image of the ‘whole’ to be made. These craftsmen conceived that to define in advance what a thing should look like would interfere with their notion of ‘efficiency’, that is, with the freedom of the craftsman to exploit the materials and forms during the making process. In an industrial situation the product to be made is conceived beforehand so that its realisation is a passive routine, not an active experience of exploration’. [My italics-CR] An ‘active experience of exploration’ is just exactly what is required now with new material alternatives, materials which do not arrest living processes, do not poison the air, the water and the ground, and do not subvert life processes irreversibly. Our present state of working with ‘green’ composite materials parallels in some respects the nature of those craft processes that drove innovation in the industrial revolution. The application of complex new materials and their translation into new components and products requires a slow, studied and skilful engagement of the new generation of materials ‘craftspeople’. We are at a decisive stage of ‘reflective practice’ again, where material possibilities

Designing for composites: traditional and future views

15

extend beyond previous experience, references and conceptual metaphors. The circumstance is analogous to the period in which cast iron structures were being designed based upon the carpentry techniques of prior experience. Despite all the conceptual baggage associated with this approach, it was the only available platform from which to extend into new material territory. Those craft skills and interpretations, by being stretched and challenged, provided the literal bridge into what became new material territory, bringing with it new forms, new methods, new structures, new products and new environments. We can be certain that a distinctive cultural and aesthetic language will emerge from an understanding of the applications and consequences of intelligently formed materials. However, it will not emerge from a theoretical position but from a series of actions and events in society. If ecological responsibility is to revalue what is local, and to understand the web of relational properties that must inform responsible design, manufacturing and consumption, and indeed living properties themselves, this new aesthetic language needs stimulating in order to create a place in consumers’ understanding of these new materials. This has to go beyond the use of carbon-fibre look-alike decals that make the things they are stuck to look ‘technical’, or the overprinting of eco-fashion unbleached colours onto conventional bleached paper to trick the consumer. Polite home-style labelling is no guarantee of home cooking. This major challenge in materials intelligence lies equally in the realms of design, marketing, manufacture, information, distribution, retailing and waste control. Why would a biodegradable product be convincingly seen as superior to its energy-inefficient and polluting predecessor? How will a material that respects life processes be perceived as truly ‘valuable’. Let’s apply those three principles: (i) no waste; (ii) environment considered at every stage; (iii) redefine the term ‘value’ in terms of life support. How do products ‘speak’ about creation and destruction – about living processes and intelligent design? It would be a mistake to regard these issues as only applicable to luxury goods or to the retail mall, or to imagine they can only be tinkered with at the margins of society. The value system that brings us this apocalyptic picture of waste, short-term thinking and abuse of the living environment continues to pervade the majority of manufacturing and materials consumption in its various spheres of influence. Whilst there are examples to be found of top-down and bottom-up environmental intelligence in contemporary business organisation, the 20-year window of opportunity for sustainable values so comprehensively described by Paul Hawken3 in The Ecology of Commerce in 1993 has not been realised. As a designer, I have to look to nature to try and find the departure points to help with this.

16

2.3.1

Green composites

How to think about composite materials

In an attempt to begin to answer some of the questions posed above, we can look at the relationships between material nature and ‘formal’ design (i.e. the form things take) and understand more about the paradigms of the natural examples we choose. The following is a list of a few observable visual aspects of the physical formations and patterns that are characteristic of natural material engineering. Rather than provide illustrations, which could fall into the trap described above of drawing the stool, you, the reader will hopefully recall in memory those features that are recognisable and be able to add some of your own. This kind of memory tends be ‘holographic’ in nature and so inherently richer than selective line drawings allow. Making your own drawings of these and other such features is a potent tool for appreciating key principles. The natural material world is full of ‘self-illustrating’ phenomena, and we are surrounded by source material for such study. The word list is intended as a provocation and a starting point. The key questions are: why are they like that? How do material properties relate to the shapes of surfaces, of components, and of substructures? ∑ Network construction (bird nest, leaf, web) ∑ Blending of forms avoids abrupt transition of shapes (bones, branch, wing) ∑ Abrupt transitions in shape accompanied by local thickening or other reinforcement ∑ Proper location of openings (crab shell, skin structures, bodyforms) ∑ Apertures, fixing points, projections, to spread load (pelvis, skull, basket design) ∑ Interface issues: differential properties arranged in three dimensions to accommodate complex characteristics (tendon, spider web, foot) ∑ Volume, skin, strut, cell (e.g. water-retaining structures in plants, ‘vessel’ forms and ‘vessel’ containment) ∑ Corrugation: the zone between two dimensions (2D) and three dimensions (3D) ∑ Profiles: harmonic shapes, moiré patterns ∑ Stem–branch relations ∑ Swelling of forms: exterior/interior relations ∑ Curvature: simple and compound ∑ Braiding, twisting, binding ∑ Thick edges, thin films ∑ Doming: compression and tension in domed forms ∑ Stem and plane (support structures for surfaces) ∑ Perforation ∑ Twisting of plane shapes

Designing for composites: traditional and future views

∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑

17

Orientation of fibres, layers of fibre orientations, simple and complex Combination different fibre materials; combining of properties Fibres following forms, skirting holes or point loads Fibre–matrix relation (rationing of) Granulation, granular networks Inclusion of locator or bearing components: contrast materials Modification of form or material at fixing points Different density/particle structure; properties across section Outer and inner forms relate to material properties ‘Product architecture’: structural metaphors (‘modular’, ‘frame’, ‘clamshell’, ‘skeleton’, etc.) help us conceptualise the principles of our design.

A useful formal metaphor for composite sheet material or surface materials, for example, is that of drapery, or fabric forms that are ‘frozen’. Other such relationships with our common experience can be useful in studying natural materials. Here are six images (Figures 2.1 to 2.6) that give some examples of opportunities to learn from natural material structure, in which the visible shapes, not only of the object but of surface detail and texture, and structural features, display some of the ‘signature’ shapes of essential functional attributes from the short listing above.

2.1 Leaf networks: repetition and diversity. This type of recognisable pattern has that special combination of fluid variation on a theme that is key to natural systems. It is pleasing aesthetically because it suggests movement and rhythm without being coldly repetitious. (Photograph by C. Rose.)

18

Green composites

2.2 Basket made by Jenny Crisp (UK). Natural material aesthetics. The fibrous structure of the willow can be highly distorted whilst still retaining sufficient integrity. The design of the whole basket has to define edges and volumes with differing techniques, analogous to the sophistication of natural engineering. (Photograph by C. Rose.)

2.3 Crabshell architecture: inner complexity. Very light inner 3D network of compartmented forms developing from the outer shell significantly improves damage control of the whole while providing an interface for inner organic structures. (Photograph by C. Rose.)

Designing for composites: traditional and future views

19

2.4 Crabshell edge: where it gets vulnerable. The open rolled edge is resistant to impact, defines the form (the form that we recognise) and provides location for contrasting tissue structures. Edges are always vulnerable. Many natural structures avoid them or make them multiple. (Photograph by C. Rose.)

2.5 Palm leaf packaging: natural food wrapper. Corrugation folds extra length into the apparent surface and retains complex flexibilities, allowing the enclosed volume to be variable, and for the very challenging distortion of the 2D surface into the 3D form. Where are the edges? (Photograph by C. Rose.)

20

Green composites

2.6 Lessons in design: multipurpose structure of cactus ‘skeleton’. 3D corrugation, perforation and ‘braiding’ of outer shape, provides a characteristic organisational definition of the whole structure and its living processes. Clockwise and anticlockwise spirals at differing slopes are typical, giving a formal ‘bias’. (Photograph by C. Rose.)

2.4

The use of composite materials through the ages: design, form and structure

Designing and using composite materials has been the principal manner of making functional products for millennia. Restricted to natural materials, tremendous sophistication and technical expertise characterise much indigenous and pre-industrial object making, developing as it has over many generations who evolved, refined and perfected designs into what could be termed a ‘stable’ technology. From Inuit clothing based on the assembly of skins, through the Tudor oak-framed house in England, which gets tougher with every passing decade, the bushman recycled eggshell water-carrier and the composite longbow combining those properties of bone and wood that can cope with extremes of energy storage and release. Physical struggle with the myriad material properties of available materials was the only research method available, and examples of these artefacts are prized as collectable products redolent of human connection with nature. Such material sophistication was

Designing for composites: traditional and future views

21

not, however, limited to reworking organic materials. The bronze casting technology of Benin in the sixteenth and seventeenth centuries has never been bettered. The fact that this advanced science, its location and practitioners, did not comfortably fit with a colonial world view led to attempts to suppress, destroy or intellectually marginalise its significance, a phenomenon only partially redressed by twentieth-century scholarship. Echoes of such technoimperialism deeply affect us today in the value systems attributed to various ‘classes’ of product, often blinding us to the sophistication inherent in certain traditional ‘evolved’ designs. The conventional styling associated with many ‘modern’ products has no connection with objective measures of genuine sophistication of material and method. Certain ancient Egyptian chariot wheels combined heat-formed wood, laminated and bound with skin, which made a tough and resilient composite and biodegradable vehicle part 3000 years ago. Some of the earliest MiddleEastern engineering feats utilised wood and fibre constructions in the form of giant water wheels (noria) for irrigation, extending the reach of wood (‘nature’s fibreglass’) into massive components that were able to perform a difficult task. Possibly the most iconic of all structural forms – that of the upright bound papyrus column – not only survives its many reinterpretations to the present day (a phenomenon that has been ascribed by some writers to Brunelleschi’s studies in Egypt) but is in itself an object lesson in design for composite materials, combining as it does both outer and inner form and the multiplying factor of smaller elements. Different ‘forms’ (shapes) can reference a variety of recognisable principles, from the cultural icon to the structurally prototypical. Design practice explores this multiplicity of ‘visual language’ and renews or reinterprets the familiar, dealing with issues of appropriateness, interpretation and use. The bundle form, of which the column is a notable architectural example, appears everywhere, from ancient ocean-going rafts, through the cast iron connecting rods of industrial revolution machines to the protective chrysalis enclosing emergent life in nature. In my view this ‘form’ therefore merits study in any consideration of materials, their behaviours and preferences. Water has a preference for ‘going globular’. Goethe considered that the physical forms of plants, composed largely of water, showed a preference for embodying the tendency of water to generate fern-shaped turbulent flow vortices when disturbed. Further study of these ‘flowforms’4 is highly relevant to new applications for certain composites and their manufacture – not only in terms of mechanical processing but in terms of material formation, biomimetic properties, etc. Since our technical ability to model and visualise in three dimensions complex mathematics such as flow, pressure gradients, intricate networks, etc., has become generally accessible in the last 10 years, we have the tools to match up two powerful processes in our understanding, namely our ability

22

Green composites

to recognise ‘complexity’ because of our human interest in pattern and our lifelong experience in reading such patterns in our sensory and social environment, and the ability to arrest, simulate and visualise the patterns of complex structures and model their various attributes as part of design processes that go ‘below the surface’ and potentially encompass differing scales from micro to macro in an integrated approach closer to the engineering of nature. From composite molecular design to composite buildings, the sophistication of future structures, products and materials holds out a promise as impressive, and as dangerous, as the longbow.

Sources of further information ∑ Lightness: The Inevitable Renaissance of Minimum Energy Structures, Beukers, Adriaan; van Hinte, Ed 010 publishers Rotterdam 2001. ∑ Design Engineering, Cather, H.; Morris, R.; Philip, M. and Rose, C; Butterworth Heinemann 2001. See chapter on ‘Creative Design Practice’. ∑ ITDG (Intermediate Technology Development Group) London. Extensive publication list. ∑ Fourth Door Review; ‘A Green Cultural Review for the Twenty-first Century’. www.fouthdoor.co.uk.

References 1 Seamon, D. and Zajonc, A. (1998). Goethe’s Way of Science, a Phenomenology of Nature, State University of New York Press. 2 Sennett, R. (1992). The Uses of Disorder : Personal Identity and City Life, New York: W.W. Norton and Co. 3 Hawken, P. (1994). The Ecology of Commerce; a Declaration of Sustainability, Harper Collins. 4 Wilkes, J. (2003). Flowforms, The Rhythmic Power of Water, Floris Books, Edinburgh.

3 Life cycle assessment R. MURPHY Imperial College London, UK

3.1

Introduction

Awareness that human activities can damage the Earth’s capacity to sustain life has been growing since the publication of Rachael Carson’s book Silent Spring in 1962 and the Club of Rome’s The Limits to Growth in 1972 (Meadows et al., 1972). More recently, the concepts of Sustainable Development (The World Commission on Sustainable Development, Brundtland, 1987) and Industrial Ecology (see Frosch and Gallopoulos, 1989) have emerged to help guide ways in which human society can be organised so that this capacity is preserved. Assessment of the environmental impact arising from activities such as construction, packaging or transport is an essential activity when attempting to design sustainable approaches to our development needs. A number of tools are now available for such assessments. Amongst these, life cycle assessment (LCA) has emerged over the last 10 to 15 years as a widely respected technique for evaluating the environmental aspects associated with a wide variety of products, processes or activities throughout their entire life cycles.

3.1.1

Life cycle assessment

LCA is a systems analysis tool for evaluating environmental impacts over the whole life cycle of a product, process or activity from the ‘cradle’ (raw material acquisition) to the ‘grave’ (disposal or recycling) (UNEP, 1996; ISO, 1997). The emphasis in LCA is to generate transparent and complete assessments of environmental impact resulting from all stages of the life cycle of the product or activity in question and to use this to evaluate its environmental attributes in a holistic way. This approach has the potential to avoid mistakes or distortions engendered by ‘single issue’ or ‘single focus’ environmental assessments. LCAs are being used in fields as diverse as construction, energy systems, composite materials, packaging, agricultural production and green chemistry 23

24

Green composites

amongst others (Marteel et al., 2003). It is known to policy makers and forms the basis for a number of eco-labelling schemes. The capacity to conduct LCA work is also supported by the availability of a number of commercial LCA software packages (see Sources of further information at the end of this chapter) as well as in-house systems in various research groups around the world.

3.1.2

Aims and objectives

This chapter outlines the LCA method, highlights the availability of LCA studies on composites and presents some current perspectives in LCA research with these materials, including the use of LCA in product development systems.

3.2

Life cycle assessment: methodology

In the case of a product, the life cycle embraces all activities needed for its manufacture, use and disposal. Typically, these comprise extraction of raw materials, design, formulation, processing, manufacturing, packaging, use (reuse), and disposal. Over the course of its development, different LCA approaches and methodologies have given rise to various terminologies, often for very similar concepts. The terms life cycle assessment and life cycle analysis are frequently used interchangeably, although other terms have also been used to represent environmental impact assessments that are very similar to, or synonymous with, LCA. These include: cradle to grave analysis; ecobalance; ecoprofile; life cycle balance; resource and environmental profile analysis; product line analysis; and integrated substance chain analysis. All these terms refer to studies that analyse and assess environmental impacts of products, processes or activities over partial or entire life cycles. To maintain consistency with the ISO 14040 series of standards for LCA, the term life cycle assessment (LCA) is used here. The first studies to examine life cycle aspects of materials and products date from the late 1960s and focused on raw materials consumption, energy efficiency and, to a more limited extent, waste disposal. Initially, energy was considered a higher priority than outputs or waste and, because of this, there was little distinction between inventory development (resources going into a product) and the interpretation of total associated impacts. However, after the oil crisis of the early 1970s, the prominence of energy issues declined and outputs gained more significance (Jensen et al., 1997). By the mid1980s and early 1990s, a much greater general interest in LCA was shown by a broad range of industries, who by the time of the 1992 UN Earth Summit, regarded LCA as among the most promising new tools for a wide range of environmental management tasks (Jensen et al., 1997).

Life cycle assessment

25

Although the field continued to develop, progress was somewhat sporadic, partly due to a lack of experience of the discipline and a lack of accepted international standards underpinning the technique. Undue expectations of LCA outcomes, together with a sense that some practitioners were conducting LCAs simply to reinforce existing positions rather than to fully understand and respond to real issues, led to a period of disillusionment with LCA (Jensen et al., 1997). However, the consumer concerns of the late 1980s created market pressures and encouraged a growing number of large organisations to carry out their first environmental audits and to prepare environmental performance reports. Many of the same companies then undertook assessments of the upstream environmental performance of their suppliers as well as downstream product stewardship programmes, designed to ensure the safe use and disposal of their products. This placed an increasing demand for life cycle data on materials and products from suppliers, as well as the services of contractors, and helped to put LCA on the agenda of many small- to medium-sized companies. Until the early 1990s, there was no universal methodology for carrying out LCAs. The first initiative to harmonise the LCA methodology was taken in 1993, when the Society of Environmental Toxicology and Chemistry (SETAC), published a practice for conducting LCAs, based on a three-part (inventory – interpretation – improvement) process. Around the same time, the Society for Promotion of Life-cycle Assessment Development (SPOLD), an association of leading industries interested in accelerating the development of LCA as an accepted management tool for the necessary restructuring of company policies towards sustainable development, mobilised considerable financial resources to accelerate the development of sound LCA methodology (Jensen et al., 1997). In the late 1990s, the LCA methodology was consolidated and moved into a stage of maturity. This was facilitated by the introduction of the ISO 14040 series of standards beginning in 1997, which provided a clear overview of the practice, applications and limitations of LCA. While confidence and optimism in the usefulness of LCA has generally grown, some still point to its inaccessibility due to the complexity of the technique. Thus a balance has had to be drawn between the competing needs of ‘user friendliness’ and clarity, against validity, based on adequate data and scientific rigour in order that the credibility of results can be verified. In recent years, computer software designers have responded to this challenge, resulting in the development of powerful, modern LCA software.

3.2.1

Basic principles for LCA under the ISO 14040 series of international standards

The International Standards Organisation (1997) defines LCA as

26

Green composites

A technique for assessing the environmental aspects and potential impacts associated with a product, by compiling an inventory of relevant inputs and outputs of a system; evaluating the potential environmental impacts associated with those inputs and outputs; interpreting the results of the inventory and impact phases in relation to the objectives of the study. LCA studies the environmental aspects and potential impacts throughout a product’s life cycle, usually from raw material acquisition, through production, use and disposal. Whilst LCA is often based on an assessment of the entire life cycle (cradle-to-grave) of a product, process or activity, in practice many studies have a more restricted scope to a partial LCA (sometimes called a cradle-to-gate LCA), mostly due to time constraints or limitations in data availability. Some LCA studies are also designed to address only specific issues and may not require an assessment of the entire life cycle. Internationally agreed standards on LCA under ISO form part of its 14000 series of voluntary standards and guideline reference documents, including ISO 14040–14043 (1997) covering the various aspects of the application of LCA methodology. These ISO guidelines and standards are summarised below: ∑ ISO 14040: provides an overview of the practice, applications and limitations of LCA to a broad range of potential users and stakeholders, including those with a limited knowledge of life cycle assessment. ∑ ISO 14041: details special requirements and presents guidelines for the preparation, conduct, and critical review of life cycle inventory analysis phase of LCA that involves the compilation and quantification of environmental relevant inputs and outputs of a product system. ∑ ISO 14042: offers guidance on the impact assessment phase of LCA – that phase of LCA aimed at evaluating the significance of potential environmental impacts using the results of the life cycle inventory analysis. ∑ ISO 14043: provides guidance on the interpretation of LCA results in relation to the goal definition phase of the LCA study, involving review of the scope of the LCA, as well as the nature and quality of the data collected. According to the ISO 14040 guidelines to the LCA methodological framework, a life cycle assessment shall include a goal and scope definition, inventory analysis, impact assessment and an interpretation of results. Figure 3.1 illustrates the ISO 14040: LCA framework and some potential applications of LCA. The LCA methodology is thus a structured framework that specifies the required data, methods of calculation and the procedure of its interpretation. This involves description of the system, production of an inventory of inputs and outputs associated with that system, translation of this inventory data into potential environmental impacts, and finally, evaluation of the results in

Life cycle assessment

27

Framework Direct applications

Goal and scope definition Interpretation

∑

Inventory analysis ∑ Impact assessment

∑ ∑

Product development, improvement Strategic planning Marketing Other

3.1 Life cycle assessment framework according to ISO 14040:1997. (Source: http://www.boustead-consulting.co.uk/iso14040.htm.)

order to facilitate decision making. The following classification and description of the separate phases of the LCA methodological framework presents a broad overview of key elements of the technique.

3.2.2

LCA goal and scope definition

The first phase of the ISO LCA framework, the goal and scope definition, is the planning phase of the LCA. This is a critical part of any LCA as at this stage the specifications of the study are defined. This includes the following main issues: the goal of the study; the scope definition; the functional unit; the system boundaries; the quality of data; and the critical review process. 3.2.2.1

Goal of the study

The goal of a study states the intended application of the results including the reasons for carrying out the study and the target group or intended audience. Additionally, the goal also determines the complexity of the study and requirements for reporting. An appreciation of these issues is important in order to make the appropriate decisions throughout the study and achieve the stated goal (Jensen et al., 1997). 3.2.2.2

Scope definition

Scoping defines the system boundaries, study assumptions and limitations of an LCA study. It also defines what activities and impacts are included or excluded in the study and why. The scope is defined to ensure that the breadth and depth of the analysis are compatible with, and sufficient to address the goal of, the LCA. At this stage, functions of the systems being

28

Green composites

studied (the functional unit, system boundaries, allocation procedures, data requirements, types of impact, methodology of impact assessment and interpretation, assumptions, limitations, type and format of the report, and type of critical review) must be clearly stated. 3.2.2.3

Functional unit of study

It is important in LCA studies to define the unit to be studied. This functional unit can be defined as ‘specification of the unit size of a product or system, on the basis of which subsequent environmental scores are calculated’ (UNEP, 1996). Definition of the functional unit or performance characteristics is the foundation of an LCA because this sets the scale for the measurement of performance that a system (product) delivers. In addition, all data collected in the inventory phase will be related to the functional unit. 3.2.2.4

System boundaries

The system boundaries have a significant influence on the outcome and the informative value of LCA studies. They determine the relative assessment area. The system boundaries define the processes and operations, and also the inputs and outputs to be considered in the LCA. At this stage decisions are made about which unit processes would be included and which environmental releases would be evaluated and whether the study would require a complete analysis (cradle-to-grave) or a partial analysis (cradle-to gate). Although difficulties associated with data availability, time and resource constraints usually necessitate applying limits to the assessment area, defining the system boundaries is often a subjective process. 3.2.2.5

Data quality

The quality of data used at the inventory stage is naturally reflected in the quality of the final LCA. It is therefore important that, at the goal and scope definition phase, the data used for the LCA is described and assessed in relation to geographical, time-related and technological coverage. In addition, descriptions of the sources of the data, whether collected at specific sites or from published sources, and whether the data is measured, calculated or estimated, should be outlined. Often, both published and site-specific data are used to develop an inventory (Jensen et al., 1997). 3.2.2.6

Critical review process

Transparency is essential in any LCA study. The purpose of the critical review is to ensure the quality of the LCA. The review can be either internal,

Life cycle assessment Inputs Energy

Raw materials

29

Outputs Raw materials acquisition Water effluents Manufacturing, processing, and formulation

Airborne emissions

Distribution and transportation

Other environmental releases

Solid wastes

Use/re-use/maintenance

Recycle

Waste management

Use of products

Cradle-to-grave system boundary

3.2 Example of life cycle inventory template flow diagram. (Source: SETAC, 1993.)

external or involve interested parties, as defined within the goal and scope definition phase. This definition clearly identifies why the critical review has been undertaken, what it covers, to what level of detail, and those involved in the process (Jensen et al., 1997). Peer review is especially important for LCA studies intended to support external, comparative assertions about the products or services under examination.

3.2.3

LCA inventory analysis

Inventory analysis is the second phase of the LCA framework under ISO 14040, and is the core of any LCA. At the inventory phase, a process flow diagram of the system is assembled, which offers a simple way of representing the linkages between processes or sub-systems involved in a product life cycle, supports data collection and also facilitates reporting and transparency of the LCA. It is crucial to the success of the LCA that all relevant processes are included. Relevant data such as energy (MJ), minerals consumed (kg), and emissions such as CO2 (kg) and losses of PO 3– 4 to water are then collected. The system boundaries can be redefined where necessary and calculations are performed in order to quantify the relevant inputs and outputs of the system being studied (Jensen et al., 1997). Figure 3.2 illustrates an example life cycle inventory template. 3.2.3.1

Data collection

One important stage in the inventory analysis phase is the collection and

30

Green composites

treatment of data required for inputs (materials and energy) and outputs (products and environmental releases) specified in the process flow diagram. Data collection is generally the most time consuming aspect of an LCA study. Data has to be collected for all single processes in the life cycle. Due to difficulty involved in data collection, both previously published data as well as primary (data originally collected specifically for the purposes of that study) are often used (UNEP, 1996). 3.2.3.2

Refining of system boundaries

The system boundaries are defined as part of the scope definition procedure. However, after the initial data collection, the system boundaries can be refined to include or exclude particular life cycle stages, unit processes or material flows which may or may not have significant effects on the results (Jensen et al., 1997). This is part of the iterative process of LCA. 3.2.3.3

Calculation

No formal demands exist for calculation in LCA except for those described for allocation procedures. Various LCA software programs have been developed for this purpose, which can be selected according to the type and amount of data to be managed (Jensen et al., 1997). 3.2.3.4

Validation of data

Systematic data validation is carried out in order to highlight specific areas where data quality must be improved or substituted, and to improve overall data quality. This involves a permanent and iterative check on data collected (Jensen et al., 1997). 3.2.3.5

Relating data to the specific system

For each unit process identified in the goal and scope definition, an appropriate reference flow is determined. This could be in terms of energy consumption (e.g. 1 MJ for energy, 1 kg for material), and the quantitative input and output data of the unit process is calculated in relation to this reference flow. In determining these reference flows it becomes possible to relate data to the actual product (Jensen et al., 1997). 3.2.3.6

Allocation

At the allocation stage of inventory analysis, the collected data is quantified in relation to a defined unit of output. For multi-output processes (unit processes,

Life cycle assessment

31

which produce more than one useful product), or multi-input processes (processes in which strict quantitative causality between inputs and outputs does not exist), an allocation procedure that is clearly defined and justified is normally used for sharing environmental impacts amongst the products. Allocation may also be required when dealing with open-loop recycling (recycling process in which the waste material leaving the system is used as a raw material by another system) (Jensen et al., 1997). In cases where allocation by mass or volume of unit output is not practical, according to ISO 14041, economic allocation may be applied.

3.2.4

LCA impact assessment

The purpose of the impact assessment phase is to translate the results from the inventory phase into potential environmental impacts. The life cycle impact assessment specifically uses impact categories and associated indicators to simplify inventory results into environmental issues such as global warming, eutrophication, acidification, etc. The impact assessment phase comprises both mandatory and optional elements which are described below. 3.2.4.1

Selection and definition of impact categories

Selection and definition of impact categories is the first mandatory element of the impact assessment phase. It involves the sorting of inventory results into selected impact categories that describe impacts caused by the product (system) being studied. The selected impact categories should be consistent with the goal and scope of the study. Impact categories can be selected to align with frequently applied categories such as global warming, acidification, human toxicity and resource depletion (UNEP, 1996). Impact categories can also be defined to represent specific issues such as noise or odour. 3.2.4.2

Classification

Classification is the second mandatory element of the life cycle impact assessment phase. At the classification stage, environmental impacts listed in the inventory results are assigned to the selected impact categories. For example, carbon dioxide and methane are assigned to global warming. The impacts can be classified into global, regional or local scales. 3.2.4.3

Characterisation

Characterisation is the last mandatory element of the impact assessment phase which involves assigning the relative contribution of each input and output to the predefined impact categories. This is mainly a quantitative step

32

Green composites

based on scientific analysis of the relevant environmental processes and an estimation of their impacts (Jensen et al., 1997). 3.2.4.4

Normalisation

Normalisation is an optional element of the life cycle impact assessment framework which involves relating and transforming all impact scores of a functional unit to a reference situation (commonly the total loading of each category). Normalisation can be used to check for errors in the inventory data and to provide a better interpretation of the characterisation data (Jensen et al., 1997). 3.2.4.5

Valuation/weighting

This is an optional element within impact assessment, in which the results of characterisation/normalisation are weighted against each other in order to make the impact information more ‘decision friendly’. This involves qualitative and quantitative valuations not necessarily based on natural sciences, but on political or ethical values (Jensen et al., 1997). 3.2.4.6

Data quality assessment

Data quality assessment is another optional element in the life cycle impact assessment. Techniques for assessing data quality in LCA studies include gravity analysis, uncertainty analysis, contribution analysis and sensitivity analysis, amongst others.

3.2.5

LCA interpretation of results

Interpretation is the fourth phase in LCA. It is a systematic technique which includes three main steps: (i) identification of significant environmental issues; (ii) conducting a qualitative and quantitative check and evaluation of information from the results of the inventory analysis and/or the impact assessment phases of the LCA; and (iii) formulation of conclusions and recommendations based on (i) and (ii) as a method of improving the overall reporting and transparency of the study. Interpretation is the phase in which choices and assumptions made during the course of the study and the results of the analysis are judged as to their accuracy, consistency and completeness. The findings of this interpretation may take the form of conclusions and recommendations to decision makers, consistent with the goal and scope of the study. The interpretation phase may involve the iterative process of reviewing and revising the scope of the LCA, as well as the nature and quality of the data collected consistent with the defined goal. The findings of

Life cycle assessment

33

the interpretation phase should reflect the results of any sensitivity analysis that has been performed (ISO, 1997).

3.2.6

Limitations of the LCA approach

LCA is an effective environmental analysis tool for evaluating the environmental impacts of a process, product or activity throughout its entire life cycle. This holistic approach to environmental analysis although a major strength, is also a limitation. A complete LCA requires the inclusion of inputs and outputs from extraction of raw material to disposal. However, in practice, it is very difficult to include all aspects adequately and this can lead to oversimplification of certain aspects of the life cycle to suit a particular goal of the LCA. One main limitation of the LCA tool is the large volume of data required to develop an inventory. Owing to the difficulty involved in collecting data, data quality and accuracy are common problems limiting the informative value of the results of most LCA studies. In addition, lack of spatial and temporal dimensions in the inventory results introduces uncertainties in the impact assessment results. In essence, LCA attempts to provide as comprehensive an account as possible of the environmental impact of the products, processes or services under investigation. This is often a very demanding task and LCA reports warrant careful consideration of elements such as the goal and scope, system boundaries and data quality declarations when considering the outcomes of the study.

3.2.7

LCA issues of special relevance to bio-based composites

There are a number of considerations that are of particular relevance in LCA studies on green composites, especially those derived from bio-based materials. These are considered briefly below but several are areas of current research in LCA and as such one can expect a number of developments to occur in how they are approached or resolved. Clearly, bio-based materials, e.g. plant and wood fibres are of interest as feedstocks for ‘green’ composites. They are grown in agricultural and forestry (natural and plantation) systems. As such, they require areas of land to be allocated to their cultivation, and ‘land use’ is an environmental impact category that is expressed in a number of LCA studies. However, there is considerable controversy over the definition, measurement and classification of land use in LCA. In forestry, for example, temporal considerations can have significant influences as the rotation of tree crops can take several decades during which substantial changes will be exhibited in the forest ecosystem. Furthermore, forest and agricultural land management can change

34

Green composites

significantly over time, and recent attention on ecosystem services as coproducts or by-products of such land uses has supported new approaches to such management. The representation of aspects such as the extraction of biotic resources from ecosystems (e.g. forest products) and the potential impacts on biodiversity, land classification and land restoration in LCA is not mature at present and may indeed remain a specialist element of ‘conventional’ LCA. However, even if such issues are not dealt with through the characterised output of an LCA, a fully transparent study would make narrative reference in the systems description to such elements of the life cycle. Further aspects of the agricultural production of fibres (e.g. hemp, flax) such as the use of fertilisers, agro-chemicals, fuels for tillage, etc. are aspects of LCA that must be included in the inventory calculations. It should be noted that there can be complex allocation issues between co- and byproducts of such agricultural production, e.g. oils, straw and soil conditioning, that need to be transparently represented in LCA studies. An additional allocation issue arises when considering low-value bulk by-products of other crops (e.g. cereal straws derived from food grain production) as opportunities for using such materials in new ways, including composite manufacture, are continually being researched. The allocation of upstream environmental impacts to main products and such by-products can be done on various bases. Whilst earlier there was a preference for allocation procedures based on mass or other physical causality, there has been an increasing use of economic allocation by LCA practitioners in recent times as this is considered to better reflect real world ‘cause and effect’. Whatever the system chosen for a given LCA, the basis of any allocation procedure must be clearly and transparently stated in the full LCA report. Bio-based composites frequently possess a wide range of end-of-life possibilities such as incineration, recovery/recycling and composting. As plant fibre materials sequester atmospheric CO2 in their growth, they are often considered to be ‘carbon neutral’. This concept refers to the carbon dioxide equivalent taken up into the plant fibre biomass and its release back to atmosphere through incineration or decomposition. However, in order to properly account for this in balance with other carbon emissions in the life cycle (e.g. from fossil fuel combustion), it is necessary to represent appropriate waste management scenarios for the product at the end if its life. This can become especially important for bio-composite materials if they are likely to enter a disposal situation where there is risk of anaerobic decomposition leading to methane emission to the atmosphere due to its high greenhouse gas activity (approximately 21 times greater than CO2). It must be noted that such end-of-life scenarios will be missing from cradle-to-gate LCA studies. In view of the uncertainty in data for a number of end-of-life waste management scenarios (e.g. emission from landfill, composting emissions) it is useful to include a number of alternative ‘scenarios’ and/or conduct sensitivity analysis

Life cycle assessment

35

for the effects of this phase of the life cycle on the overall conclusions for studies on ‘green’ composites.

3.3

LCAs of composite materials

Most LCAs of composite materials to date have focused on wood-based composites in construction and plant fibre-based composites in automotive applications. Commonly, these materials are compared with alternative components manufactured from materials like glass fibre, steel and concrete. There have also been a number of LCAs conducted in related areas of biobased polymers, although these are not considered further here (see Patel et al., 2003).

3.3.1

LCA of wood-based composite materials

LCA has been applied to the manufacture and use of panel products made from wood, natural fibres and other materials in several studies. Data and assessments from these are now available in published reports (see, for example, Mundy et al., 2000; Esser and Robson, 1999) and commercially available LCA databases (e.g. IVAM, 2003; Sima Pro, 2003; amongst others). In the UK, recent work co-ordinated by the Building Research Establishment (BRE) through the Partners in Technology programme of the DETR has produced a detailed assessment of the environmental profile of the UK woodbased panel production (and forestry and sawmilling) sectors. Such data are being incorporated into design tools such as BRE’s Envest to enable environmental data to be incorporated into the design of buildings and structures. 3.3.1.1

Incorporation of post-consumer recovered timber into woodbased panels

Several wood-based panels can be and are manufactured from post-consumer wood recovered as part of waste management systems. This recovery of wood materials can be expected to increase in the future and the utilisation in reconstituted wood-based panel products is a significant potential use for such recovered wood. At the same time, it should be recognised that alternative uses of such wood biomass are available (e.g. composting, incineration with energy recovery) and LCA can be usefully applied to evaluate such options. A recent study by Speckels et al. (2001) has highlighted several interesting considerations regarding the waste management and potential use options for recovered wood. They conducted an LCA on three common ways of waste wood management in Germany:

36

Green composites

Table 3.1 Overview of LCA results on options for use of recovered wood Class H1 in Germany LCA impact category

Incineration with energy recovery

Particleboard manufacture

Landfilling

Global warming potential

+

–

–

Eutrophication potential

+/–

+/–

+/–

Photochemical ozone creation potential

+

–

–

Acidification potential

+

–

–

Human toxicity potential

–

+/–

+/–

Aquatic toxicity potential

+

–

–

Terrestrial toxicity potential

+

–

–

Land use

+

–

–

+ = process is optimal, +/– = processes are similar, – = process is not preferable. After Speckels et al. (2001).

∑ thermal utilisation for heat and/or electricity ∑ material recycling into new particleboard ∑ landfill. The LCA work was based on 1 tonne of waste wood and involved a system approach incorporating wood production from forestry, particleboard manufacture, energy generation processes (either from recovered wood or fossil fuels) and landfilling. The results of the LCA for uncontaminated (Class H1) recovered wood in Germany are summarised in Table 3.1. Six of the eight LCA impact categories favoured the use of recovered wood for energy generation rather than particleboard manufacture or landfilling. Clearly, there are benefits to be gained in terms of overall environmental impact in the use of renewable energy (such as waste wood as here) and substitution of fossil energy sources. The growing of new wood for panel product manufacture is a relatively low environmental impact process and may also have benefits in terms of carbon sequestration. The use of the wood twice – once for its value as a material and then, secondly, on recovery as a fuel – presents attractive benefits in terms of environmental impact. Ultimately, the selection of waste management processes depends upon many factors such as economics, transport and infrastructure and social considerations and the environmental assessment presented above will form only one element of the decision. However, an important aspect of the LCA process is to support a more thorough consideration of the extent to which

Life cycle assessment

37

combinations of processes and actions affect the overall environmental profile of such decisions through ‘life cycle thinking’. In this example, a presumption of recycling of wood waste into new particleboard manufacture is challengeable on environmental grounds and strongly suggests that further specific studies should be conducted in this area where relevant local circumstances can be taken into consideration. 3.3.1.2

Life cycle design (LCD) for wood-based panel products

The following example is based on a study conducted by the author in collaboration with Dr Martin Ansell and colleagues from the University of Bath. The aim of the study was to evaluate the environmental impact of potential design improvements to oriented strand board (OSB) manufacture in comparison with the ‘standard’ OSB board (Nishimura and Ansell, 2001, 2002; Nishimura et al., 2001). OSB is a wood-based panel product manufactured in the UK and elsewhere in standard sizes under heat and pressure from wood strands, bonded with a synthetic adhesive and wax. Type F2 boards (BS5669 Part 3), which use adhesives that are inherently moisture resistant and confer higher levels of durability were analysed here. Panels are generally loaded in bending in construction uses, e.g. flooring, walls and roofs. Flexural properties are therefore critical and, in addition to panel density and resin bonding, these depend on the disposition of strands at the panel surfaces. In the factory, strands are delivered to a forming mat on a conveyer, where the aim is to orient the top and bottom layers along the conveyer axis and the core strands are oriented at 90º to this direction. The surface strand orientation, size and shape must be optimised to maximise mechanical strength in flexure. The product life cycle and the panel design, in terms of wood and resin inputs, strand dimensions and orientation, are therefore closely linked. A full LCA of the OSB manufacture and its use in a model flat roof construction was undertaken to evaluate the environmental implications of design modifications to the panels. Goal of the study Scope of the study Functional unit

Assess the environmental impact of the ‘standard’ and modified manufacturing of OSB panels. Assess the effects in the context of OSB use in a ‘typical’ building application. Provision of a weatherproof outer layer to a flat roof, of the cold-roof type (as specified in Agreement Certificate BBA 92/2781), with an area of 3 ¥ 4 m (12 m2) above a habitable dwelling for a period of 50 years (see Fig. 3.3). This included sealants, mastic, fixings, supports, coverings plus installation/maintenance of a weatherproof barrier and full replacement at 25 years (for the 50 years’ service).

38

Green composites Nailing at 150 mm centres along supported edges, and at 300 mm centres intermediately

10 mm perimeter expansion gap

Nogging support at long edges to allow ventilation

Plasterboard ceiling

All edges to be fully supported (20 mm minimum bearing) Joists at 400 mm centres

3.3 Layout for OSB in the domestic flat roof.

System boundaries Data sources Assumptions

Zero order processes. Direct measurements at factory, design changes by factory and University of Bath. Transport within UK derived from UK average statistics. Underlying ceiling components (plasterboard, vapour barrier, thermal insulation) not considered to be part of the roof and their impacts excluded. Insulation properties of roof assumed to be dominated by the added insulation and so no thermal differences were assigned to the different OSBs.

OSB and design changes: The ‘standard’ OSB is based on manufacture with a phenol formaldehyde (PF), in powdered form, added at a rate of ~2.5% (dry wood basis). A paraffin slack wax is added at a rate of 1.5%. The timber used is 85% scots pine. The remaining 15% is composed of varying proportions of spruce, larch, douglas fir and other mixed conifers. All timber used is between 15 and 40 years growth, derived from first and second thinnings and tops. The final density of the board is between 600 and 650 kg/m3. Installation of the boards requires cutting on-site, and the use of half sized boards to minimise waste, as is common practice, with cuts is assumed. Fixings are galvanised mild steel nails.

Life cycle assessment

39

Product design improvements based on flake orientation studies at the University of Bath and the manufacturer’s process engineering improvements were examined. The study: (a) developed an LCA for ‘typical’ PF resin bonded OSB (b) analysed the effect of an optimised flake orientation developed in research at Bath giving board density reduction of 5% compared with the ‘typical’ OSB for the same mechanical/physical property specification (c) analysed the effect change from PF to PMDI resin.

% change in normalised scores against typical OSI

The study produced a number of striking findings concerning wood-based materials and panel products. Both design changes modelled in the manufacture of OSB produced a reduction in mass of wood and other materials in the panel for equivalent mechanical/physical properties compared with standard OSB. In this case, density reduction had a negative effect (~8% increase) upon climate change (global warming potential) (see Fig. 3.4). The disadvantage in terms of climate change value for the modified OSBs compared with the original (and heavier) OSB was due to the lower mass of wood material in the modified OSBs. This is nevertheless a striking demonstration of the fact that the use of wood and plant fibres that absorb atmospheric CO2 during their growth phase and lock up this carbon for long periods of time (as in construction) can have positive environmental attributes through sequestration. 12 10 8 6 4 2 0 –2 –4 –6 –8 –10

Total

Human toxicity

Ecotoxicity Eutrophication Low level ozone creation

Flake orientation

Climate change (100 years)

Fossil fuel Transport depletion pollution and congestion

MDI OSB

3.4 LCA profile of modified OSB in comparison with ‘standard’ PFbonded OSB.

40

Green composites

Overall, it is also clear that both design modifications to the OSB had a relatively small effect overall on the environmental impact of the OSB (+/– 3%). However, change of the resin (in this case to PMDI) exhibited a negative influence whereas the processing ‘efficiency’ gain on the standard OSB from the modelled improvement in flake orientation was all positive with the exception of climate change discussed above. Patel et al. (2003) have proposed that, at least in the context of bio-based polymers, reductions of the order of 20% in most environmental impacts (and specifically a saving of 20 MJ per kg of polymer and avoidance of 1 kg CO2 per kg of polymer) would be a useful guide for a good environmental improvement target. This proposal by Patel et al. is useful and suggests that neither of the changes modelled in the OSB LCA study have a significant influence on the overall environmental impact of the functional unit investigated. The changes in environmental profile for the OSB case discussed above do indicate clearly, however, that even relatively small changes to a product can result in shifts in overall environmental impact and that these can be sometimes unexpected.

3.3.2

LCA of plant fibre-based composites

Plant fibre-based composites are increasingly seen as being ‘green’ or ‘ecocomposites’. Whilst such assertions can be supported by LCA studies, the perception of the greenness of plant fibre composites is often taken for granted in association with general concepts of renewability. It is instructive to consider the following examples in order to consider the potential environmental impacts of such plant fibre-based composites. The environmental impacts of a composite garage door system manufactured containing a door skin based either on glass fibre or non-woven hemp fibre mat were compared on a cradle-to-gate basis using LCA. The results are shown in Figs. 3.5 and 3.6. It is clear from this analysis that there is little difference between the two door types in this analysis with the hemp mat door performing slightly better than the glass fibre equivalent in only one category (solid waste). Further analysis of the process for the hemp door illustrates the reason for the lack of a significant difference between the two door types (see Fig. 3.6). It is clear that very little of the impact in this cradle-to-gate life cycle resides in the hemp mat component of the door (5% or less for any category) – the great majority of the impact is associated with the manufacturing process and the polyester resin and gel coat components. These latter items are the same for the hemp or the glass fibre door and dominate the LCA. Thus, if an improvement to the environmental profile for this product is desired, the LCA results indicate that focus should be placed on the manufacturing stage and the resin/finishing components of the system offer far more potential for impact reduction than the fibre component. When considering eco-composites based on plant fibres or glass fibre, it is clearly

Life cycle assessment

41

100 80 Glass door Hemp door

(%)

60 40 20

– – – –

kg

J

ste

M

wa

gy er So

lid

En

HT

PO

kg

kg CP

kg

kg

AP

A EC

OD

GW P

kg P

kg NP

NP ODP ECA GWP

m3

0

Impact categories AP – Acidification potential POCP – Photochemical oxidant creation potential HT – Human toxicity potential

Neutrophication potential Ozone depletion potential Ecotoxicity potential Global warming potential

3.5 Relative environmental profile of hemp mat door to the benchmark glass fibre door.

necessary to maintain a sense of perspective as to the role of the fibre material in the overall product application. It should also be noted in this example that we are considering a cradle-to-gate LCA here and that performance 100 90 80 70

(%)

60 50 40 30 20 10

– – – –

Neutrophication potential Ozone depletion potential Ecotoxicity potential Global warming potential

M

kg

So li

d

gy

wa ste

J

kg HT

En er

kg PO

CP

kg

Impact categories

Truck highway/rural Aluminium trihydroxide Gelcoat NP ODP ECA GWP

AP

kg GW

P

g m3 m

EC A

P OD

NP

kg

kg

0

Hemp mat MEKP Unsaturated polyester resin Hemp door AP – Acidification potential POCP – Photochemical oxidant creation potential

HT

– Human toxicity potential

3.6 Relative environmental impact of different life cycle stages in production of the hemp mat door.

42

Green composites

Table 3.2 Consumption of primary non-renewable energy by a Miscanthus plant fibre/ polypropylene composite pallet and a glass fibre/polypropylene composite pallet as a function of transport distance Distance travelled in 5 years of use

Miscanthus pallet Glass fibre pallet

5000 km

100 000 km

200 000 km

700 MJ (21%) 1300 MJ (14%)

3500 MJ

6700 MJ (91%) 9300 MJ (87%)

5000 MJ

( ) = proportion of total primary non-renewable energy of the life cycle consumed in the use phase. Adapted from Corbière-Nicollier et al. (2001).

in use and especially at disposal has not been evaluated. Furthermore, this application relates to a construction situation and this contrasts somewhat with other applications such as automotive and transport. Full LCAs of plant fibre-based composites used in automotive or transport applications have often revealed potentially substantial benefits compared with glass fibre equivalents. These benefits are often in the order of 15 to 50% saving of environmental impact by use of the plant fibre composite and these savings often derive from lighter weight of these materials which has strong benefits during the use phase of the vehicle (or transport), sometimes referred to as ‘secondary savings’ as pointed out by Patel et al. (2003). This is shown for a comparison of transport pallets made from fibreglass-reinforced polypropylene or China reed (Miscanthus sinensis) reinforced polypropylene in a study by Corbière-Nicollier et al. (2001) (see Table 3.2). It should be noted that the comparison in Table 3.2 is based upon an expected service life of 5 years and includes an end-of-life scenario of incineration with energy recovery. The pallets are assumed to be functionally equivalent in terms of strength and durability. The data in Table 3.2 indicate that the Miscanthus polypropylene composite pallet would use between 30 and 50% less primary energy than the glass fibre equivalent. This was mirrored for several other environmental impact categories (data not shown). Corbière-Nicollier et al. noted that the energy consumption in the use phase dominate from about 27 000 km for the Miscanthus pallet and 38 000 km for the glass fibre pallet. If a transport distance in the use phase of 200 000 km over 5 years is modelled then this would account for approximately 90% of the total primary non-renewable energy consumed over the entire life cycle. This indicates the great importance of the use phase for transport (and automotive) applications of materials and, as shown in Table 3.2, the benefits that can be achieved by weight reductions associated with plant fibre composites in comparison with glass fibre

Life cycle assessment

43

composites. In the present case a saving of some 2600 MJ of primary energy was indicated for the Miscanthus composite pallet over a 200 000 km usage scenario when compared with the glass fibre equivalent. Garkhail (2002) has also used LCA to study the environmental impact of plant fibre composites in comparison with glass fibre. A flax/polypropylene compression moulded composite was modelled for both automotive and non-automotive (e.g. office chair) applications in comparison with the glass fibre equivalent. In this study the flax fibre composite only had a superior environmental performance in stiffness critical automotive applications where its lower mass gave advantage. In impact strength or tensile strength critical automotive applications a higher mass of the flax composite was needed to provide the same functionality as the glass fibre component and this led to a poorer environmental performance for the flax composite than the glass fibre composite. These examples indicate that LCA results for composites can be very sensitive to the specific applications and that it is difficult to draw broad, generic conclusions about such materials. This is especially true when comparing automotive/transport applications where fuel savings during the use phase of the life cycle due to the use of light weight composites can be highly beneficial. In construction applications mass can be a less important parameter and instead durability/maintenance and embodied energy can be critical. Insulation value is also a highly important property for construction materials that may be used in building envelopes. This discussion has focussed on the environmental impacts of composites as demonstrated through LCAs. Often of equal and frequently of greater importance for products are economic and regulatory issues. In this context, composites based on plant fibres may have advantages at the end-of-life as the options for waste management can offer ready alternatives to landfilling (e.g. recycling, waste-to-energy, composting). This can deliver both a direct, practical advantage for these materials and benefits in the LCA as a whole.

3.4

Future trends: making use of LCA

3.4.1

LCA essential information

The principal uses of LCA for informing manufacturers about the environmental profile of their products and providing environmental information for specifiers and decision makers to consider alongside economic and technical data are likely to continue for the foreseeable future. As more life cycle inventory data are acquired and become available, especially authoritative public access datasets, LCA studies will become less time consuming to conduct. This is not to say that they will become ‘easier’ or simpler to conduct because careful attention to the selection of appropriate data for the goal and scope of the study will always be a key issue in LCA demanding care and thought.

44

Green composites

In particular, it is necessary to maintain up-to-date information on many industrial and environmental processes in order to conduct reliable LCA studies. Scientific and technical knowledge on the effects of agricultural, forestry and industrial systems on land quality, stability and classification will continue to develop and these issues will remain significant in LCA both from a methodological as well as a data perspective (see Schweinle et al., 2001).

3.4.2

LCA in product development and design

A key aspect of success in applying LCA to design projects is to achieve close integration and understanding of the LCA activities with the technical design and development activities. Attempts to do this (such as p. 37) often reveal a need to adapt the ‘conventional’ LCA procedure to the realities of the design and re-design process. The conventional principles, procedures and methods of LCA are now strongly based on the terminology and structure of the ISO tools and standards on LCA (ISO 14040 series). This establishes a strict sub-division of tasks and procedures needed for conducting an LCA. However, in the development of a new or a re-design, the range of technical possibilities to be considered varies depending on the type of development project, but the most innovative and valuable solutions may come from the greatest degree of design freedom. In such cases, the applicability of the full LCA process is very limited, since it is based upon the full specification of the process chain. These details, by the nature of the design process, are in flux, and may encompass an impracticably large range of possibilities. Furthermore, the design process need not even begin from a specific functional application (which is an absolute requirement for an LCA), but may start, for example, as a search for workable applications for new material or process possibilities. The design process by which that range of possibilities is reduced is also not generally led by environmental data. As experimental and development data accrues, technical feasibility, economic and market issues will all be significant drivers for rejecting possibilities. There is therefore a need for a structured procedure in which the greatest practical amount of environmental data is provided at all stages of the design process for examination alongside technical and economic data. As the range of possibilities is reduced, the detail and completeness of environmental data escalates, making possible full LCA data for the final options. A procedure of ‘incremental LCA’ to do this which is not a streamlined or abridged LCA, but rather a gradual route to a full LCA is one possibility. ‘Incremental LCA’ aims to enhance the application of life cycle thinking and LCA to LCD for engineered timber products. Its structure and framework (see Fig. 3.7) maximises the value of LCA within the design process and minimises the redundancy of data that results from changes in a design during its development.

Life cycle assessment

45

Full LCA

Design idea

Role of environmental data

Range of design options Elimination/reduction

Prioritisation

Elaboration

3.7 Stages in the movement from a design idea to a full LCA.

The incremental LCA procedure encompasses the design process from a concept to a market-ready product. In re-design, where the application of the product and possibly large parts of its production and life cycle are already known and fixed, it is possible to omit the first elimination/reduction phase of the incremental LCA process and commence directly with the activities of the prioritisation phase. In addition to the use of LCA to provide environmental information supporting individual product design there have been strong developments in the construction area for integrated design tools for structures. These developments have been led by the ATHENA™ Sustainable Materials Institute in Canada and, more recently, BRE in the UK (see Sources of further information). In both cases, LCA data is used to enable more comprehensive environmental assessments of various construction options and mixes of materials than is readily undertaken by ‘conventional’ LCAs. It is expected that as such tools develop they will be more widely used in the architectural and civil engineering areas

3.4.3

LCA for environmental product declarations (EPDs)

Credible and high-quality LCAs provide a sound basis for application in eco-labelling schemes, environmental product declarations (EPDs) and support environmental management systems (EMS). In particular EPDs are likely to develop as a ‘shorthand’ way of transferring LCA based information in an agreed format for business to business and, possibly business to consumer communication. The network GEDnet and ISO TR 14025 acts as an important focus for interest and development in EPDs at the present time.

46

3.5

Green composites

Conclusions

LCA studies have demonstrated that a whole life cycle perspective can be a valuable adjunct to technical and economic development advances in composite products. The general benefits of timber as relatively low-impact material for a variety of engineering applications have been exemplified in the case studies presented here. The climate change benefit in using wood material has been strikingly and unusually demonstrated in the design ‘improvement’ LCA study of OSB, where the lower density giving lower transport emissions (in the construction of the flat roof) was still not sufficient to overcome the climate change benefit arising from carbon sequestration in the wood product. A new concept of ‘incremental LCA’ is proposed whereby LCA activity can be efficiently incorporated into product development and improvement. Such advances in LCD are expected to benefit the fibre, construction and engineering industries and raise the profile of composite products as competitive and environmentally friendly materials.

Sources of further information Some LCA software tools SIMAPro 5, Pre Consultants, http://www.pre.nl/ TEAM, Ecobalance UK, http://www.ecobalance.com GaBi, PE Europe Gmbh and IKP University of Stuttgart, http://www.gabisoftware.de Boustead, http://www.boustead-consulting.co.uk/ PIRA, http://www.pira.co.uk Organisations ATHENATM Sustainable Materials Institute, http://www.athenasmi.ca Building Research Establishment (BRE), http://www.bre.co.uk International Organisation for Standardisation (ISO), http://www.iso.ch United Nations Environment Programme (UNEP), http://www.unep.org/ and http://www.uneptie.org/pc/sustain/lca/lca.htm Society of Environmental Toxicology and Chemistry (SETAC), http:// www.setac.org/ and http://www.setac.org/WEB/lca.html Environmental product declaration information Global Type III Environmental Product Declarations Network (GEDnet), http://www.gednet.org/

Acknowledgements The author wishes to acknowledge financial support for LCA research from the following organisations: EPSRC (GR/L74996/01), Commission of the

Life cycle assessment

47

European Communities (Contract FAIR-CT95-072, DETR (PIT Contract C138/19/133). I am are also indebted to Dr Martin Ansell and his colleagues at the University of Bath for many useful discussions and collaboration, colleagues from the COST Action E9 working group, partners in the EC Life-Sys Wood project and colleagues at the BRE for valuable exchanges and discussions on LCA. I am especially grateful to Bill Hillier for many stimulating and informative discussions on LCA over many years.

References Brundtland, G. (ed.) (1987). Our common future. In The World Commission on Environment and Development. Oxford: Oxford University Press. Carson, R. (1962). Silent Spring, New York: Houghton Mifflin (2002 edn). Corbière-Nicollier, T., Gfeller Laban, B., Lundquist, L., Leterrier, Y., Månson, J.-A.E. and Jolliet, O. (2001). Life cycle assessment of biofibres replacing glass fibres as reinforcement in plastics. Resources, Conservation and Recycling, 33 (4), 267–87. Esser, P. and Robson, D.J. (1999). Life-Sys Wood: Consistent Life Cycle Analysis of Wood Product. TNO Center for Timber Research: Delft, The Netherlands (Final Report for EC R&D Contract FAIR-CT95-072). Frosch, R.A. and Gallopoulos, N.E. (1989). Strategies for manufacturing, Scientific American, 261 (3), 144–52. Garkhail, S.K. (2002). Composites Based on Natural Fibres and Thermoplastic Matrices. PhD Thesis, University of London. ISO (1997). Environmental Management Life Cycle Assessment – Principles and Framework EN ISO 14040. The International Standards Association. IVAM (2003). LCA Data 4. Database of LCA processes. IVAM: Amsterdam, The Netherlands. www.ivam.uva.nl. Jensen, A., Hoffmann, L., Møller, B., Schmidt, A., Christiansen, K., Elkington, J. and van Dijk, F. (1997). Life Cycle Assessment, A Guide to Approaches, Experiences, and Information Sources, Environmental Issues Series No. 6, Copenhagen, European Environment Agency, Marteel, A.E., Davies, J.A., Olson, W.W. and Abraham, M.A. (2003). Green chemistry and engineering: drivers, metrics, and reduction to practice. Annual Review Environment Resources: 28, 401–28. Meadows, D.H., Meadows, D.I., Randers, J. and Behrens, W.W. (1972). The Limits to Growth. A Report to The Club of Rome, London: Pan Books. Mundy, J.S., Thorpe, W.M.H., Bonfield, P.W., Hillier, W. and Murphy, R.J. (2000). Environmental Assessment of UK Forestry, Sawmilling and Panel Production. Final Report for DETR Partners in Technology Contract C138/19/133, Watford: Building Research Establishment. Nishimura, T., Amin, J. and Ansell, M.P. (2001). Image analysis of the shape and size of strands in model OSB panels and correlation with MOE and MOR. Proceedings of the Fifth European Panel Products Symposium, Llandudno, October 2001. Nishimura, T. and Ansell, M.P. (2001). The relationship between the arrangement of wood strands at the surface of OSB and the modulus of rupture determined by image analysis. Wood Science and Technology, 35(6), 555–62. Nishimura, T. and Ansell, M.P. (2002). Monitoring fibre orientation in OSB during production using filtered image analysis. Wood Science and Technology, 36 (3), 229–39.

48

Green composites

Patel, M., Bastioli, C., Marini, L. and Würdinger, E. (2003). Life-cycle assessment of bio-based polymers and natural fibre composites. In Biopolymers, Vol 10, General Aspects and Special Applications, Ed. Alexander Steinbüchel, John Wiley & Sons Inc. Schweinle, J., Doka, G., Hillier, W., Kaila, S., Köllner, T., Kreibig, J., Muys, B., Quijano, J.G., Salpakivi-Saloma, P., Swan, G. and Wessman, H. (2001). The Assessment of Environmental Impacts Caused by Land Use in the Life Cycle Assessment of Forestry and Forest Products; Guidelines, Hints and Recommendations. Final Report of Working Group 2 “Land Use” of COST Action E9. Mitteilung der BFH Nr. 209, Hamburg, Germany. SETAC (1993). Guidelines for Life-Cycle Assessment, a “Code of Practice”. Society of Environmental Toxicology and Chemistry, Brussels, Belgium. Sima Pro (2003). SIMA Pro LCA LCA software and databases. PRé Consultants, Amersfoort, The Netherlands www.pre.nl. Speckels, L.G., Fruhwald, A., Scharai-Rad, M. and Welling, H. (2001). Ecological aspects of waste wood utilisation in Germany. In Achievements of COST Action E9 Working group 3: end of life: recycling, disposal and energy generation, Ed. Jungmeier, G., Joanneum Research Report IEF-B-11/01, 170pp. UNEP (1996). Life Cycle Assessment: What It Is and How to Do It, Paris: United Nations Environmental Programme.

4 Natural fibre sources T. N I S H I N O Kobe University, Japan

4.1

Introduction

The utilisation of biomass has gained increased importance due to threats of uncertain petroleum supply in the near future and concerns about environmental pollution. Green, ecologically (environmentally) friendly, sustainable, renewable, biodegradable, composites from plant-derived fibre and cropderived plastics are among the most keenly required materials of the twentyfirst century (Bledzki and Gassan, 1999; Mohanty et al., 2002; Netravali and Chabba, 2003). Most sustainable plastics cannot compete economically with conventional petroleum-derived plastics in their present state. Economically favourable composites, therefore, are expected to be made from costly sustainable plastics in combination with inexpensive natural reinforcement fibres. Cellulosic materials are the most abundant form of biomass and the form most likely to be used as reinforcement fibres, not only for ecological and economical reasons, but also because of their high mechanical and thermal performance. To utilise and design materials successfully for industrial applications, it is first imperative to determine material properties that will affect performance. In this chapter, the hierarchical structures and mechanical properties of cellulose and its derivatives/composites will be described from microscopic molecular level to macroscopic level.

4.2

The microstructure of natural plant fibres

Natural fibres can be classified into several categories as shown below.

Natural fibres

Wood fibres Soft and hard woods (many spices) Vegetable fibres Cotton, hemp, jute, ramie, kenaf, etc. Animal fibres Wool, silk, spider silk, feather, down, etc. Mineral fibre Asbestos, inorganic whiskers, etc. 49

50

Green composites

Natural vegetable fibres exist as hairs (cotton, kapok, baobab), bast fibres (also called soft fibres; kenaf, ramie, flax, hemp, jute, papyrus, cordia, indian malow), hard fibres (sisal, abaca (manila hemp), raffia, pineapple, New Zealand flax), stem (bamboo, bagasse, banana stalk, cork stalk), fruit (coconut), straw (rice, corn, wheat) and others (seaweeds, palm). In the section below, the structure of kenaf, from a microscopic molecular level to a macroscopic molecular level, is described as an example. Figure 4.1 is a photograph of kenaf (Hibiscus cannabinus. L, family Malvaceae), grown in Japan. Kenaf is well known as a cellulosic source with economical and ecological advantages: in the three months after sowing the seeds, it can grow under a wide range of weather conditions to a height of more than 3 m and a base diameter of 3–5 cm. The growing speed may reach 10 cm/day under optimum ambient conditions (Rowell and Han, 1999). The stem is unbranched and straight and is composed of an outer layer (bark) and a core. It is easy to separate the stem into the bark and the core by chemical and/or enzymatic retting (Ramaswamy, 1999).

4.1 Photograph of kenaf (Hibiscus cannabinus. L, family Malvaceae).

Natural fibre sources

51

People have skilfully made use of kenaf since ancient times, traditionally as rope, canvas and sacking. Recently, kenaf has been used as an alternative raw material to wood in the pulp and paper industries to avoid the destruction of forests (Han et al., 1999). It has also been used to make non-woven mats in the automotive industry, textiles and fibreboard. Kenaf has been actively cultivated in recent years for several reasons. One reason is that kenaf absorbs nitrogen and phosphorus found in the soil and/or in wasted water. The average aborption rate for kenaf is 0.81 g/m2/day for N and 0.11 g/m2/day for P; rates which are several times higher than those for a variety of trees (Abe and Ozaki, 1998). The second is that kenaf accumulates carbon dioxide at a significantly high rate. The photosynthesis rate of kenaf (23.4 mg CO2/dm2/ h) is much higher than the photosynthesis rates of conventional trees: 8.7 (Konara: Quercus serrata Thunb.) and 2.5 (Kunugi: Acutissima carruth) mg CO2/dm2/h, respectively, under a photon density of 1000 m mol/cm2/s (Lam and Iiyama, 2000). Kenaf is also not abrasive during processing, has a low density and high specific mechanical properties and is biodegradable (Inagaki, 2002). The price of kenaf was $400 per tonne in 1995 and $278–302 per tonne in 2000 (Liu, 2000). From the viewpoint of energy consumption, it takes 15 MJ of energy to produce 1 kg of kenaf, whereas it takes 54 MJ to produce 1 kg of glass fibre (Mohanty et al., 2002). These figures indicate that kenaf has both ecological and economical benefits and that it can be a good alternative to glass fibres. Figure 4.2 shows an optical micrograph and a scanning electron micrograph (SEM) of the cross-section of the interface between the bark and the core of the kenaf stem. The bark constitutes 30–40% of the dry weight of the stem and shows rather dense structure. On the other hand, the core is wood-like, and makes up the remaining 60–70% of the stem. Figure 4.3 shows the SEM images of the bark and the core of kenaf along the stem (stem direction is shown by an arrow) together with X-ray diffraction photographs. The core reveals an isotropic and almost amorphous pattern; however, the bark shows an orientated high crystalline fibre pattern. By grinding the bark, its cell wall can be separated as shown in Fig. 4.4(a). The schematic representation of the cell wall of a natural plant is shown in Fig. 4.4(b). This structure is often called the macrofibril (in other words, microfibre or primary/elementary fibre). The macrofibril size and chemical content of kenaf stem is shown in Table 4.1 (Khristova et al., 1998; Inagaki, 2002). The cell wall consists of a hollow tube, which has four different layers: one primary cell wall and three secondary cell walls and a lumen. The lumen is an open channel in the centre of the macrofibril. Each layer is composed of cellulose embedded in a matrix of hemicellulose and lignin, a structure that is analogous to that of artificial fibre reinforced composites. Hemicellulose is made up of highly branched polysaccharides including glucose, mannose, galactose, xylose, etc. Lignin is made up of

52

Green composites

4.2 Optical and scanning electron micrographs of the cross-section of the interface between the bark and the core of the kenaf stem.

4.3 Scanning electron micrographs of the bark and the core of kenaf, together with the X-ray diffraction photographs. Stem direction is shown by an arrow.

Natural fibre sources (a)

(b) Macrofibril

53

(c) Microfibril

Lumen

S2 layer

Secondary cell wall

Crystalline region

Microfibril angle

Amorphous region Primary cell wall

ca.

Cellulose molecule

5–

30

nm

4.4 (a) Scanning electron micrograph of kenaf bark fibre, and schematic representations of (b) macrofibril and (c) microfibril of natural plant. Table 4.1 Macrofibril size and chemical content of kenaf stem Bark Fibril length, L (mm) Fibril width, W (mm) L /W Lumen diameter (mm) Cell wall thickness (mm) Cellulose (%) Lignin (%) Hemicellulose (%) Ash content (%)

Core

2.22 17.34 128.0 7.5 3.6

0.75 19.23 39.0 32.0 1.5

69.2 2.8 27.2 0.8

32.1 25.1 41.0 1.8

Source: Khristova et al. (1998); Inagaki (2002).

aliphatic and aromatic hydrocarbon polymers positioned around the fibres. The structure and contents of the cell wall change widely between different spices and between different parts of the plants. The primary (outer) cell wall is usually very thin (< 1 mm), but the secondary cell wall is composed of three layers. Of these, the S2 layer is very much the thickest and is the major contributor (80%) to the overall properties. The S2 layer is formed with microfibrils, which contain large quantities of cellulose molecules. The microfibrils run fairly parallel to each other and follow a steep helix around the cell (Hearle, 1963; Bos and Donald, 1999). Furthermore, the microfibril is composed of crystalline regions and amorphous regions alternately, as is shown in Fig. 4.4(c). The crystallite size is about 5–30 nm in the lateral direction and up to 20–60 nm along the axis and therefore the

54

Green composites

cellulose molecules pass through several crystallites along the axis (von Frey-Wyssling and Mühlethaler, 1963; Haigler, 1985). This structure is called a fringed micelle structure. This structural model is classical and may become old-fashioned when chain folding for synthetic polymers is found. For now, however, it is still believed to be valid for cellulose.

4.3

The crystal structure of celluloses

Cellulose is a natural linear homopolymer (polysaccharide), in which Dglucopyranose rings are connected to each other with b -(1 Æ 4)-glycosidic linkages. O HO

CH2OH O HO

HO HO O

O CH2OH

n

Cellulose is a crystalline polymer and there are several crystal modifications. Cellulose I type modification is obtained from natural plants and woods. Some kinds of bacteria (Acetobacter) and shell (tunicin) also produce native cellulose. Recently, it has been argued that cellulose I can be further divided into two crystalline modifications, so-called cellulose Ia (triclinic) and Ib (monoclinic) (VanderHart and Atalla, 1984; Sugiyama et al., 1991). Natural products are the mixture of these two modifications. Cellulose Ib is dominant for plants and woods, and the percentage of cellulose Ib present varies. For example, the cellulose Ib content of cotton linter and ramie is 77%. Cellulose Ia is transformed into Ib by a hydrothermal treatment in an alkali solution (Yamamoto and Horii, 1993) or by heat treatment at 280 ∞C in an inert gas (Wada et al., 2003). For example, the cellulose Ib content is increased to 90% by heat treatment at 260 ∞C in 0.1 mol/l NaOH (Horii, 2001). This indicates that cellulose Ib is thermodynamically more stable than cellulose Ia. Almost pure cellulose Ib is obtained from tunicates (Halocynthia roretzi). On the other hand, cellulose Ia is reported to be the major component of bacterial and algal cellulose. The structural difference between Ia and Ib can be distinguished using solid state 13C-NMR. The Ia content (fa ), determined by NMR, can also be correlated with the absorption ratio of the FTIR spectrum (Yamamoto et al., 1996). fa = 2.55 (A750 /A710) – 0.32 where A750, A710 are the integrated absorption at 750 cm–1 and 710 cm–1, respectively.

Intensity (au)

Natural fibre sources

55

Ia(010)

I b (1 10)

Ib(110)

Ia(100)

12

13

14 15 16 17 Diffraction angle 2q (∞)

18

19

4.5 X-ray diffraction profile of the bacterial cellulose. Diffraction peaks were curve resolved and indexed.

Figure 4.5 shows the X-ray diffraction profile of the bacterial cellulose. Several diffraction peaks could be curve resolved and indexed as shown in the Figure. The fa value can be calculated as follows (Wada et al., 1997): fa =

Ia (100)

Ia (100) + Ia (010) + Ia (010) + I b (110)– + I b (110)

where I is the integral intensity of each diffraction peak. By applying this method, the fa value is obtained as 0.60 for the sample shown in Fig. 4.5. Another evaluation method was also reported from the lattice spacing. The parameter Z is defined as follows (Wada et al., 2001): Z = 1.324d1 – 1.329d2 + 1.520d3 + 3.429d4 – 1.585 where di is lattice spacing, d1 = 6–6.2 Å, d2 = 5.2–5.5 Å, d3 = 3.8–4.0 Å, d4 = 2.5–2.6 Å, respectively. When Z > 0, the sample is cellulose Ia form rich type (when Z < 0, it is cellulose Ib form rich type). For the specimen in Fig. 4.5, d1 = 6.18 Å, d2 = 5.31 Å, d3 = 3.94 Å, d4 = 2.61 Å, so the Z value is 19.8. This indicates that this cellulose can be recognised as cellulose Ia form rich type. Cellulose II can be obtained by swelling cellulose I samples with alkali (mercerisation: for example, 21.5% NaOH aq. solution at 20 ∞C for 24 h) or by regenerating cellulose solutions. Cellulose IIII and IIIII are converted from the corresponding cellulose I and II by immersing them in liquid ammonia

56

Green composites Mercerisation

Alkali cellulose I Cellulose I Hot water

liq. NH3

Cellulose IIII 260 ∞C Cellulose IVI

Alkali cellulose II

Regeneration

Cellulose II

Hot water

liq. NH3

Cellulose IIIII 260∞C

Cellulose IVII

4.6 Crystal transformation map of a series of celluloses.

(–78 ∞C), respectively. These crystal modifications, IIII and IIIII, can easily be returned to their original form (I or II) by boiling water treatment. Cellulose IVII is obtained by annealing cellulose IIIII at high temperature (260 ∞C). The crystal transformation map of a series of celluloses is summarised in Fig. 4.6. Figures 4.7 and 4.8 show the equatorial and meridional X-ray diffraction profiles of cellulose polymorphs (Nishino et al. 1995b). These modifications are said to have the same skeletal conformation as cellulose I; that is, a fairly extended zigzag conformation. The chain packing, chain direction and intra/ interhydrogen bonds are different from one another, however, and this is reflected in the diffraction profiles. Table 4.2 summarises the unit cell parameters of cellulose polymorphs. The unit cell parameters are well refined for cellulose I series (Ia (Sugiyama et al., 1991), Ib, (Nishiyama et al., 2002), IIII, IVI (Takai and Tajima, 2000)). However, those of cellulose II series (II, IIIII, IVII) vary largely depending on the origins and treatment conditions (O‘Sukkivan, 1997; Takai and Tajima, 2000).

4.4

The crystal modulus of natural fibres

The elastic modulus of polymer crystal provides us with important information on the molecular conformation in the crystal lattice. The elastic modulus (crystal modulus) of the crystalline regions in the direction parallel to the chain axis has been measured for a variety of polymers by X-ray diffraction (Nakamae and Nishino, 1991). Examination of the data so far accumulated enables us to relate the crystal modulus, namely, the extensivity of a polymer

Natural fibre sources

57

Equatorial

Intensity (au)

IIIII

II

IIII

IVI

I 10

20 Diffraction angle 2q (∞)

30

4.7 Equatorial X-ray diffraction profiles of cellulose polymorphs.

molecule, both to the molecular conformation and the mechanism of deformation in the crystalline regions. Furthermore, knowledge of the crystal modulus is of interest in connection with the mechanical properties of polymers, because the crystal modulus gives the maximum value for the specimen modulus of a polymer. The initial slope of the stress–strain of the crystalline regions gives the crystal modulus, when the changes in the crystal lattice spacing under a constant stress are monitored by X-ray diffraction. Figure 4.9 shows the stress–strain curve of the crystalline regions of cellulose I (Nishino et al., 1995b). The open and half-filled circles are the observed results for the (004) and the (008) planes, respectively, whose normals are parallel to the chain axis. Thus the vertical axis directly corresponds to the cellulose chain extension by a tensile stress, which shows linear function with the stress through the origin. By assuming that the stress on the crystalline regions is equal to that on the whole specimen, the inclination gave the crystal modulus of 138 GPa for cellulose I. Table 4.3 shows the crystal modulus of ramie, regenerated cellulose and mercerised cellulose when absorbing water (Sakurada et al., 1966). Ramie is a kind of natural plant fibre and its crystal is assigned as cellulose I. On the

Green composites

Meridional (002)

(004)

IIIII

(006)

(008)

(002) (004)

II

(006)

(008)

(006)

(008)

(006)

(008)

(006)

(008)

(002)

Intensity (au)

58

(004) IIII

(004) (002) IVI (004)

(002) I 10

20

30

40 50 Diffraction angle 2q (∞)

60

70

4.8 Meridional X-ray diffraction profiles of cellulose polymorphs. Table 4.2 Unit cell parameters of cellulose polymorphs a

b

c

a

Å Ia Ib II IIII IIIII IVI IVII

6.74 7.84 7.09 10.25 9.97 8.03 7.99

5.93 8.22 9.22 7.78 7.65 8.13 8.10

b

g

degree 10.36 10.40 10.30 10.34 10.24 10.34 10.34

117 90 90 90 90 90 90

113 90 90 90 90 90 90

81 96.8 118.3 122.4 120.1 90 90

Sources: Nishiyama et al. (2002), O’Sukkivan (1997), Sugiyama et al. (1991), Takai and Tajima (2000).

80

Natural fibre sources Cellulose I

0.3

Strain (%)

59

0.2

0.1

El = 138 GPa 0 0

100

200

300

Stress (MPa)

4.9 Stress–strain curve for (open circle) the (004) and (half-filled circle) the (008) planes of cellulose I. Table 4.3 Crystal modulus of ramie, regenerated cellulose and mercerised cellulose when absorbing water Water absorption

El

wt%

Yl GPa

Ramie Ø Ø Ø

6.2 7.5 9.0 32

138 135 127 127

Regenerated Ø Mercerised Ø

13.1 53 13.3 46

88 88 88 88

34 27 24 12 10 5 5 1.7

Source: Sakurada et al. (1966).

other hand, the crystal lattices of the latter two fibres belong to cellulose II. Thus the large differences in the crystal modulus, El values, are reasonable, because these belong to the different polymorphs of each other. The macroscopic specimen modulus Yl, of the fibres decreased as water was absorbed. This is because the amorphous region of cellulose was swollen with water. When the stress is unhomogeneous within the fibre, in other words, when the cellulose fibre can be expressed by a parallel model as shown in Fig. 4.10(a), the softening of the amorphous region will bring the stress concentration onto the crystalline regions. In this case, the El value should decrease with the water content. On the other hand, stress on the crystalline regions is equal to the macroscopic stress for the series model shown in Fig. 4.10(b). As is clear from Table 4.3, the El values of cellulose I and II were independent of the water content, respectively. This suggests that cellulose fibres are expressed by a mechanical series model between crystalline and amorphous regions as shown schematically in Fig. 4.10(b).

60

Green composites

Crystalline regions Crystalline regions

Amorphous region

Amorphous region

(a) Parallel model

(b) Series model

4.10 Schematic representations of mechanical (a) parallel and (b) series models.

The crystal modulus can also be obtained by calculation. The first value was calculated by Meyer and Lotmar in 1936, and they reported 77–121 GPa. Recently, intramolecular hydrogen bonds have been found to play an important role in the crystal modulus of cellulose. Figure 4.11 shows the values calculated by Northolt (Kroon-Batenburg et al., 1986) and Tashiro and Kobayashi (1991), together with the skeletal conformation of cellulose. The main chains are aligned parallel to one another and are linked by intra- and intermolecular hydrogen bonds. There are two

El = 136 GPa

El = 168 GPa

O(6¢) El = 89 GPa O(2)

El = 64 GPa

Kroon-Batenburg et al. (1986)

O(3¢)

El = 162 GPa O(5)

El = 72 GPa

Tashiro and Kobayashi (1991)

4.11 Crystal modulus E1 of cellulose calculated by Northolt and coworkers (Kroon-Batenburg et al., 1986) and by Tashiro and Kobayashi, together with skeletal conformation.

Natural fibre sources

61

series of intramolecular hydrogen bonds: O(6¢)—O(2) and O(3¢)—O(5). Based on Northolt’s calculation, the crystal modulus decreases drastically from 136 GPa to 89 GPa because of the lack of O(6¢)—O(2) hydrogen bonds. In contrast, Tashiro and Kobayashi insist on the importance of O(3¢)—O(5) hydrogen bonds on the other side, and report that the crystal modulus decreases when these hydrogen bonds vanish. The lack of intramolecular hydrogen bonds decreases the crystal modulus, and the skeletons of cellulose are not hard to elongate to their axial direction intrinsically without intramolecular hydrogen bonds. Figure 4.12 shows the relationships between the crystal modulus El and the maximum specimen modulus Ymax of various natural and synthetic polymers so far reported in literature (Nakamae and Nishino, 1991). The crystal modulus and Ymax values of polyethylene (PE), poly(p-phenylene benzobisthiazole) (PBZT) (Nakamae et al., 1999) and poly(p-phenylene benzobisoxazole)(PBO: commercialised as Zylon from Toyobo, Co.) (Nishino et al., 2001) are extremely high. Apart from these values, the crystal modulus of so-called high performance

478 350 GPa 372

Maximum specimen modulus Ymax (GPa)

PBO 330 GPa PBZT Steel

200

PE

Kevlar 149 Kevlar 49 Glass

100

Al Cellulose II It.PP Silk

Cellulose I Technora Vectran Kevlar

PVA

Rodrun Ekonol

Chitin

Nylon 6

PET PEN

0 0

100 200 Crystal modulus El (GPa)

372

478

4.12 Relationship between the crystal modulus El and the maximum specimen modulus Ymax of various natural and synthetic polymers so far reported in the literature.

62

Green composites

polymers such as Kevlar (poly(p-phenylene terephthalamide)) is relatively high, and the ratio Ymax to El (Ymax/El) is more than 85% (Nakamae et al., 1987). These offer proof of the success of synthesis and processing in obtaining high modulus and high strength polymers from aromatic polyamides and polyesters. Among them, the crystal modulus (138 GPa) and Ymax values of cellulose I are comparable to those of high performance synthetic fibres, and even higher than those of aluminium and glass fibre. The advantages of natural fibres over traditional reinforcing fibres such as glass or carbon fibres are their low cost, low density and biodegradability, together with their high specific properties. The ultimate tensile strength of cellulose is reported to be 17.8 GPa: that is seven times higher than that of steel (Ito, 1990). The high elastic modulus and tensile strength (not specific modulus and specific strength) demonstrate that cellulose possesses a potential ability to replace glass fibre, and that it can be a good candidate for a reinforcement fibre of the composite, without taking each density into consideration. Table 4.4 summarises the crystal modulus El, cross-sectional area S, of one molecule in the crystal lattice, f-value and fibre identity period (FIP) of natural fibres: cellulose polymorphs (Nishino et al., 1995b), silk (Nakamae et al., 1989; Nishino et al., 1992), chitin, chitosan (Nishino, et al., 1999), cellulose triesters (Nishino et al., 1995a) and polyethylene (Nakamae et al., 1991). The f-value is defined as a force required to stretch a molecule by 1%, Table 4.4 Crystal modulus El, cross-sectional area S of one molecule in the crystal lattice, f-value and fibre identity period (FIP) of natural fibres (cellulose polymorphs, silk, chitin, chitosan), cellulose triesters and polyethylene El GPa Cellulose I II IIII IIIII IVI Cellulose triesters Acetate R = CH3:CTA Propionate R = C3H7:CTP Butylate R = C4H9:CTB Valelate R = C5H11: CTV

a-Chitin Chitosan Silk fibroin Bombyx mori Antheraea perni Polyethylene

138 88 87 58 75

f

S Å

2

10

FIP –5

dyn

Å

31.9 32.5 33.9 33.4 32.7

4.40 2.86 2.95 1.94 2.45

10.38 10.33 10.34 10.24 10.37

71.1 86.5 97.0 113

2.24 1.86 1.71 2.03

10.54 15.08 10.30 10.43

41 65

44.8 34.0

1.83 2.27

10.32 10.21

23 20

21.7 25.0

0.50 0.50

6.97 6.92

235

18.2

4.28

2.53

33.2 21.6 17.6 17.9

Natural fibre sources

63

and could be calculated using the El and the S values. As described above, the crystal modulus of cellulose I (natural cellulose) is relatively small compared with that of polyethylene. However, when a comparison is made based on an f-value, the f-value (4.40 ¥ 10–5 dyn) of cellulose I is almost equal to that of polyethylene (4.28 ¥ 10–5 dyn). This indicates that the extensivity of the cellulose I molecule itself is intrinsically the same as that of PE, and that the lower El value of cellulose I is attributed to its large S. On the other hand, the crystal moduli of cellulose triesters (which can be synthesised by esterification of three hydroxyl groups in the glucopyranose ring of cellulose with corresponding n-aliphatic acid) are smaller than that of cellulose I. The small crystal moduli for cellulose triesters could be explained by both the large S and the lack in intramolecular hydrogen bonds. The crystal moduli of other cellulose polymorphs (II, IIII, IIIII, IVI) are smaller than that of cellulose I. These can be explained by the chain contraction in the crystal lattice. Figure 4.13 shows the relationship between the f-value and the fibre identity period for a series of cellulose polymorphs (open circles) and cellulose triesters (closed circles). The almost constant f-values for cellulose triesters indicate that the chain contraction itself does not affect the f-value very much. On the contrary, hydrogen bonds play an important role, judging from the difference between the crystal moduli of celluloses and cellulose triesters. The effect of the hydrogen bonds seems to be diminished for the contracted chain. The f-value decreased drastically with the chain contraction for cellulose polymorphs. This contraction is thought to be associated with the internal rotation around the main chain ether linkage between the glucopyranose rings, which should also affect the intramolecular hydrogen bonds. It is well Chain contraction 5 I

f -value (10–5 dyn)

4

Intramolecular hydrogen bonds IIII

3

II Chitosan CTP

1 0 10.0

IVI Chitin

2 IIIII

CTV

CTA

CTB

10.2 10.4 Fibre identity period (Å)

10.6

4.13 Relationship between the f-value and the fibre identify period for a series of (open circle) cellulose polymorphs and (closed circle) cellulose triesters.

64

Green composites

Cellulose I

1.30

008 spacing (Å)

Elastic modulus El (GPa)

150

100

50

1.29 0 0

100 Temperature (∞C)

200

4.14 Relationships between the crystal modulus, the fibre identify period of cellulose I and temperature.

known that the vibrational frequency of intramolecular hydrogen bonding changes depending on the crystal modifications (Atalla, 1983). The crystal moduli for chitin and chitosan are low for the same reasons. Accordingly, the skeletal conformation and/or intramolecular hydrogen bonds change with the crystal transition, and these modifications are completely different from each other from the mechanical point of view. Figure 4.14 shows the relationships between the crystal modulus, the fibre identity period of cellulose I and temperature. The fibre identity period is almost constant from room temperature up to 200 ∞C. This reveals that the cellulose molecule does not show any thermal expansion or contraction. This is in contrast to any other solids, including metals, ceramics and polymers. For example, an iron crystal expands about 0.275% from 0 ∞C to 200 ∞C – a much larger expansion than that of the cellulose skeleton. The crystal modulus is also temperature independent, which shows that the cellulose skeleton (including intramolecular hydrogen bonds) is intrinsically thermally stable. Silk fibroins are not polysaccharides, but natural polypeptides. The skeletal structure of silk fibroin is well known to be an antiparallel pleated-sheet structure (b-sheet) (Marsh et al., 1955b; Takahashi et al., 1991). The skeleton of the silk molecule contracts 5.6% from the fully extended planar zig-zag conformation. This contraction is the reason for the relatively low crystal modulus of silk. The crystal modulus of A. perni is slightly smaller than that of B. mori (Marsh et al., 1955a). The S value is larger though for A. perni. The crystalline regions of B. mori consist of glycine, alanine and serine residues (3 : 2 : 1), while those of A. perni are mainly composed of alanine

Natural fibre sources

65

residues. The bulkiness of the methyl group of alanine is responsible for the increment of S and the decrement of the crystal modulus for A. perni.

4.5

The mechanical properties of cellulose microfibrils and macrofibrils

4.5.1

Cellulose microfibrils

The angle between the cellulose microfibrils and the longitudinal cell axis is called the microfibril angle, as shown in Fig. 4.4(b). In this spiral structure, the microfibril angle is one of the major factors in determining the mechanical properties of the fibre (McLaughlin and Tait, 1980; Reiterer et al., 1999). The average microfibril angle ranges from ca. 6–11∞ in flax (Wang et al., 2001) to ca. 30∞ in cotton, and to more than 40∞ in coir and some selected leaf fibres. Figure 4.15 shows the stress–strain curve of a single wood pulp fibre with different microfibril angles (Page and El-Hosseiny, 1983). The curves show yielding followed by plastic deformation until the breakage at 20% elongation for fibres with a high microfibril angle. On the other hand, for fibres with a low microfibril angle, the curve is steep and linear. The tensile strength and Young’s modulus decreased with the increase in the microfibril angle. The

1∞

Black spruce 45% yield kraft 1.2 mm span

1.5

Stress (GPa)

12∞ 22∞ 1.0

24∞

32∞ 39∞

42∞

0.5 46∞

0 0

5

10

15

20

Strain (%)

4.15 Stress–strain curve of the single wood pulp fibre with different microfibril angles (Page and El-Hosseiny,1983).

66

Green composites

tensile strength reached up to 1.7 GPa, and the initial inclination gave a macroscopic Young’s modulus, Yl, of ca. 90 GPa. This single fibre contains 70–80% cellulose microfibrils. Thus the Yl value of the 100% cellulose fibre can be estimated to be 113–128 GPa. Hepworth and Bruce (2000) reported a maximum elastic modulus of 130 GPa, when 1 cm cubed pieces of potato tissue were compressed. These values are very close to the crystal modulus of cellulose I shown above.

4.5.2

Cellulose macrofibrils

Table 4.5 illustrates the mechanical properties of natural cellulose so far summarised (Bledzki and Gassan, 1999; Kimura, 2001). The values are scattered widely and a variation in characterstics has been observed from one fibre to another. The stress–strain behaviour is strongly influenced by conditions such as (i) materials: microscopic: crystallinity, microfibril angle, crystal modifications; macroscopic: fineness, porosity, size and shape of lumen, history: source, age, retting and separating conditions, geographical origin, rainfall during growth, or (ii) measurement conditions: tensile speed, initial gauge length, moisture, temperature. For cellulose fibre, the tensile strength usually increases with moisture and decreases with temperature, while the Young’s modulus decreases when absorbing water (Ohya and Muraoka, 1978; Stamboulis et al., 2001). Figure 4.16 shows (a) Young’s modulus, (b) the tensile strength and (c) the elongation at break from numerous tensile tests on ramie single fibres. The tensile tests were performed at the initial length of 20 mm and tensile speed of 20 mm/min at 25 ∞C. The fibres were dried at 120 ∞C prior to the tensile tests. There are wide distributions in mechanical properties and this suggests the non-homogeneity of natural fibres. Two measurement methods can be used to give the cross-sectional area of a single fibre: weight, length and density can be measured (cross-sectional area = weight/length/density) Table 4.5 Mechanical properties of natural cellulose from the literature so far reported

Cotton Jute Flax Hemp Ramie Sisal Coir

Strength

Modulus

Elongation

MPa

GPa

%

287–597* 393–773 345–1035 690 400–938 511–635 175

–** 400–800 800–1500 550–900 – 600–700 –

5.5 –12.6* 26.5 27.6 – 61.4 –128 9.4 –22.0 4.0 –6.0

*Bledzki and Gassan (1999), **Kimura (2001).

– ** 10–30 60–80 70 44 38 –

–* 8 6–10 6 8 10–25 30–49

67

Frequency

Frequency

Natural fibre sources

0.2 0.4 0.6 0.8 1.0 1.2 (b) Tensile strength (GPa)

Frequency

0 20 40 60 80 (a) Young’s modulus (GPa)

0

1

2

3

4

(c) Elongation at break (%)

4.16 (a) Young’s modulus, (b) the tensile strength and (c) the elongation at break from tensile tests on ramie single fibres.

or the fibre can be directly observed using scanning electron/optical microscopy. The latter method is not very useful because of diameter fluctuation along the natural fibre. Therefore, in order to avoid this possible cause of experimental error, the Young’s modulus and the strength are often expressed in terms of gram per denier (g/d). Many efforts have been made to obtain high modulus/high strength viscose rayon and cuprammonium rayon. Polynosic fibre is one kind of viscose rayon with a Young’s modulus of 14 GPa and a strength of 5 g/d (Ohya and Muraoka, 1978). Recently, Northolt et al. (2001) reported regenerated cellulose fibre spun from an anisotropic phosphoric acid solution. This fibre is said to possess a Young’s modulus of 45 GPa and a strength of 1.3 GPa. However, the elastic modulus of the regenerated cellulose fibres should be intrinsically lower than those of natural cellulose because the crystal modulus of cellulose II is lower than that of cellulose I as shown above.

68

4.6

Green composites

Natural fibre/sustainable polymer composites

Conventional and traditional fibre reinforced composites are composed of carbon fibres and glass fibres, which are incorporated into unsaturated polyester or epoxy-resin. These composites show high mechanical and thermal properties, so they are widely used in various applications from aerospace to sports, but they must be disposed of by incineration, causing environmental problems. Thus, in order to overcome these problems, environmentally friendly composites are keenly sought after and can be made by utilising natural fibres as reinforcements combined with sustainable polymer as matrices. Kenaf fibre reinforced poly(L-lactic acid) (PLLA) composite is described below as an example (Nishino et al., 2003). PLLA is one of the most popular sustainable polymers (Drumright, et al., 2000). Sustainable polymers are synthetic high polymers originating from natural products: they are stable in their lifetime during use and storage, but degrade microbially and/or environmentally after disposal (Tsuji, 2002). PLLA is known to possess a relatively high melting point (usually around 160 ∞C) and a high mechanical performance. Kenaf/PLLA composite can be manufactured by both wet and dry impregnation processes. The latter process involves the stacking of the kenaf sheets and the PLLA sheets alternately, followed by a compression moulding. Here, the wet process is described, as shown schematically in Fig. 4.17. Kenaf sheet is made by chemically retting bast fibres of kenaf. After drying at 120 ∞C, the kenaf sheet is immersed into PLLA dioxane solution (10 wt%) under vacuum. After impregnation, it is dried at room temperature for 24 h, then further dried under vacuum until it reaches the plateau weight. Figure 4.18 shows the effect of the kenaf fibre content on Young’s modulus and the tensile strength of the kenaf–PLLA composite. As-cast PLLA film showed a low Young’s modulus (1.3 GPa) and a low tensile strength (21 MPa). Although they can be improved by mechanical drawing and annealing (Fambri et al., 1997; Sawai et al., 2003), these mechanical properties indicate that PLLA is in the category of so-called conventional plastics from the mechanical point of view. Both properties increased with an increase of the fibre content, and showed the maximum values (Young’s modulus: 6.4 GPa, and tensile strength: 60 MPa) around the fibre content of 70 vol%. These values are comparable to those of the conventional composites. The decrease in mechanical properties for a composite with a fibre content above 70 vol% is due to the insufficient impregnation of the matrix resin. Figure 4.19 shows the stress–strain curves of the kenaf sheet, the PLLA film and the kenaf–PLLA composite with fibre content of 70 vol%. Young’s modulus and tensile strength were higher for the composite than for the matrix resin and the kenaf sheet. This reveals that incorporation of kenaf fibres into the matrix is quite effective as a reinforcement. The mechanical

Natural fibre sources Kenaf sheet (thickness 40 mm) drying at 120 ∞C 15 min

drying at r.t. 24 h

69

Kenaf/PLLA biodegradable composite (Kenaf content: 70 vol%)

PLLA in dioxane solution (10 wt%)

CH3 CH

C O

O n

Poly (L-lactic acid) (PLLA)

4.17 Schematic route for the preparation of kenaf/PLLA composite through a wet process. 7 60

Tensile strength (MPa)

Young’s modulus (GPa)

6 5 4 3 2

40 30 20 10

1 0

50

0

20

40 60 80 Fibre content (% vol)

100

0

0

20

40 60 80 Fibre content (% vol)

4.18 Effect of the kenaf fibre content on Young’s modulus and the tensile strength of the kenaf/PLLA composite.

properties of the composite were enhanced even compared with the properties of the kenaf sheet itself. When the sheet was stretched using the tensile tester, the angular change of the kenaf fibre in the direction of the stretching was considered to be the main deformation mechanism for the sheet. This type of deformation is restricted for the composite because matrix resin was impregnated into the interfibrillar regions. This is considered to be one reason for the reinforcement. Another reason is that it enables a good stress transfer from the matrix to the incorporated kenaf fibres. Figure 4.20 shows the dynamic storage modulus (E¢) and mechanical tan d for the kenaf sheet (open circles), the PLLA film (closed circles) and the kenaf–PLLA composite (shaded circles), plotted against temperature. PLLA

70

Green composites 70

60 Kenaf/PLLA composite

Stress (MPa)

50

40

30 Kenaf sheet

PLLA

20

10

0 0

1

2

3

4 5 Strain (%)

6

7

14

15

4.19 Stress–strain curves of the kenaf sheet, the PLLA film and the kenaf/PLLA composite with the fibre content of 70 vol%.

shows an abrupt drop in the E¢ value above its glass transition temperature. On the contrary, the E¢ value for the kenaf sheet is almost temperature independent up to 200 ∞C. For the composite, the absolute value of the storage modulus is high and this high modulus is maintained up to the melting point (Tm = 161 ∞C from the DSC measurement) of PLLA. Though the tolerance was limited by the intrinsic property (melting point) of the matrix resin, this shows the relatively high thermal resistance of this composite. Furthermore, for the composite, the main tan d peak, assigned as the glass transition of PLLA, shifted to a higher temperature, and the absolute intensity decreased. This suggests a strong interaction between the kenaf fibre and the PLLA resin, which emphasises the mechanical reinforcement achieved by the addition of kenaf fibre. Figure 4.21 shows the relationship between the elongation of the PLLA film, the kenaf–PLLA composite under 0.12 MPa and temperature. With increasing temperature, the thermal expansion coefficient, a , of the PLLA film changed drastically around 41∞C due to the glass transition of PLLA. Above Tg, a of PLLA was positive (ca. 2.5 ¥ 10–4 K–1) up to 110 ∞C, but the composite did not seem to show any thermal expansion/contraction. This shows the high dimensional and thermal stabilities of the composite. Figure 4.22 shows the scanning electron micrograph of the kenaf sheet. The single fibres with their diameter of ca. 13 mm were observed to be preferentially orientated in the horizontal direction of the micrograph. This fibre orientation is considered to influence the mechanical properties of the composite.

Natural fibre sources

Storage modulus E ¢(Pa)

1010

71

Kenaf/PLLA composite

Kenaf sheet 109

PLLA 108

107

Tan d

1

Kenaf/PLLA composite 0.1

PLLA

Kenaf sheet

0.01 20

60

100 140 Temperature (∞C)

180

4.20 Dynamic storage modulus (E ¢) and mechanical tan d for (open circle) the kenaf sheet, (closed circle) the PLLA film and (shaded circle) the kenaf/PLLA composite, plotted against temperature.

2.0

Expansion (%)

1.5

PLLA

1.0

0.5 Kenaf/PLLA composite

0 –0.5 30

40

50

60

70 80 90 Temperature (∞C)

100

110

120

4.21 Relationship between the elongations of the PLLA film, the kenaf/ PLLA composite under 0.12 MPa and temperature.

72

Green composites

4.22 Scanning electron micrograph of the kenaf sheet (¥ 200).

Figure 4.23 shows the mechanical properties of the kenaf sheet, the PLLA film, and the kenaf–PLLA composite in directions both parallel and perpendicular to the preferential fibre orientation. The cast PLLA film was isotropic; however, the kenaf sheet showed large mechanical anisotropies. This anisotropy is reflected in the mechanical properties of the composite, as the composite showed large anisotropies in Young’s modulus and the tensile strength. Large anisotropy brings unidirectionally high mechanical performance as shown in Fig. 4.19. On the other hand, pseudo-isotropic composite can be expected to be obtained by laminating four plies of the composite sheets, changing their orientation direction every 45 degrees. Figure 4.24 shows the anisotropy of Young’s modulus for the laminated composite together with that for the single-ply composite. As described above, the single ply of the kenaf/PLLA composite showed large anisotropy. In contrast, the laminated composite showed almost the same modulus in all directions. Thus, the pseudo-isotropy for the mechanical properties of the

Natural fibre sources 70

6

Tensile strength (MPa)

Young’s modulus (GPa)

7

73

5 4 3 2 1 0

60 50 40 30 20 10 0

Kenaf

Kenaf/PLLA

Kenaf

PLLA

Kenaf/PLLA

PLLA

Perpendicular

Parallel

4.23 Mechanical properties of the kenaf sheet, the PLLA film and the kenaf/PLLA composite in the directions both parallel and perpendicular to the preferential fibre orientation.

composite was achieved by the lamination. As described above, the microfibril angle varies between different species of natural plant and between the different parts of the plant. The results shown in Fig. 4.24 suggest that the mechanical properties of the composite can be designed artificially by laminating multiplies based on a biomimetic idea.

Young’s modulus (GPa) 0∞

1 GPa

6.3 GPa 45∞

–90∞

3.8 GPa

90∞

3.6 GPa 3.7 GPa –45∞

4.24 Anisotropy of Young’s modulus for the laminated composite, together with that for the single-ply composite.

74

4.7

Green composites

Future trends

Recently, natural fibre-based composites have received considerable attention. They have an increasing market demand, especially among automobile companies looking for light-weight materials with sound damping properties (Peijs, 2001). In 1941, the Ford Motor Company, USA, investigated composites, which were soybean oil based. The Toyota Motor Corporation, Japan, made a commercialised vehicle with door trim panels made of kenaf–PP composite and a cover for a spare tyre made of kenaf–PLLA composite. Very recently, the Araco Corporation, Japan, made an electric vehicle with a body totally made of plant-based composite as shown in Fig. 4.25 (http://www.araco.co.jp/ english/e-index.html). The body of this vehicle is made of kenaf fibre and lignin-based matrix extracted from kenaf. As shown above, the utilisation of natural cellulose today has been mainly restricted to macrofibril-order level. Cellulose, with its diameter on the nanometre level, has a microfibril structure (Taniguchi and Okamura, 1998). The decrease of fibre diameter is expected to bring several advantages to the reinforcements. The specific surface area increases with the decrease of fibre diameter, which will enhance the interfacial connection to the matrix. Furthermore, the decrease in diameter itself may cause an increase in surface energy, and a decrease in inter-fibre distance. The former produces a strong interfacial interaction between the fibres and the matrix, and the latter will bring about a delay in crack propagation across the fibres. Yano and Nakahara (2003) repeatedly defibrillated kraft pulp using a high pressure homogeniser. After the microfibrillation process was repeated 30 times, the width of kraft pulp was reduced to 10 nm, which corresponds to

4.25 Photograph of electric vehicle with its body totally made of plantbased composite produced by Araco Corporation, Japan (2003).

Natural fibre sources

75

400 Mg alloy 350 300

Stress (MPa)

MFPM-starch 250 200 150 GFRP, chopped

100

Polycarbonate

50 0 0

0.01

0.02

0.03 0.04 0.05 Strain (mm/mm)

0.06

0.07

4.26 Stress–strain curves of microfibrillated pulp sheets, together with those of Mg alloy, chopped GRFP and polycarbonate.

a bundle composed of 5 to 10 microfibrils. Compression-moulded sheet was produced from the microfibrillated pulp slurry blended with the oxidised tapioca starch. The stress–strain curves of these sheets are shown in Fig. 4.26 together with those of Mg alloy, chopped GFRP and polycarbonate. The bending modulus is 12.5 GPa, and the bending strength is 320 MPa for the composites with starch. These values are comparable with those of Mg alloy used for electronic devices. High mechanical performance could also be achieved for the microfibrillated pulp itself without the matrix (bending modulus: 16 GPa, bending strength: 250 MPa). These remarkable mechanical properties are ascribed to the formation of the microfibrils’ web-like network tightly connected by hydrogen bonds in a nanometre scale. Bacteria also produce pure cellulose without lignin and hemicellulose. A uniform sheet can be prepared by squeezing and drying the bacteria cellulose gel (Nishi et al., 1990). This sheet showed high modulus (up to 30 GPa) twodimensionally, which is much higher than those of conventional polymers (up to 10 GPa), and is comparable with that of aluminium. High modulus brings high acoustic propagation velocity (5000 m/s) to this film. Together with this property, the low internal loss of the bacterial cellulosic sheet is suitable for the diaphragm of audio loudspeakers and headphones. These are commercialised by SONY Corporation, Japan. Algal cellulose from seaweed is also considered to be a good candidate for the reinforcement fibres of the green composites. Algal cellulose has a porous structure with high crystallinity and high specific surface area (Stromme et al., 2002; Wu and Berglund,

76

Green composites

Table 4.6 Merits and demerits of using natural fibres for reinforcements Merits

Demerits

Green image Sustainability CO2 neutral Biodegradability Low energy consumption No residues when incinerated Properties Low density High mechanical properties No toxicity No abrasiveness Less flammability Low cost

Against water Poor water resistance High moisture absorption Poor dimensional stability Interface Properties Toughness Processing Supplies Inhomogeneous quality Demand and supply cycles

Source: Stamboulis et al. (2001) and some additions.

2003). Finding some relevant use for algae has ecological benefits from the viewpoint of removing red tide from seawater. Silk-based materials have attracted much attention both from the academic and industrial sides (Osaki, 1996). Especially, spider silk outperforms synthetic fibres in its combination of strength (1.15 GPa) and elasticity (elongation at break: 39%) despite being spun at ambient conditions using water as a solvent (Vollrath and Knight, 2001). In contrast, natural silkworm silk is presumed to be much weaker and less extensible than spider silk. However, artificial reeling of silk from immobilised silkworms also produced fibres with high tenacity (1.3 GPa), high modulus (~20 GPa), high breaking energy (12 ¥ 104 J/kg) and high extensivity (37%), which are comparable to those of spider silk (Shao and Vollrath, 2002). These outstanding properties are expected to produce a green composite with greater toughness. Table 4.6 shows the merits and demerits of using natural fibres for reinforcements (Stamboulis, et al., 2001). So far, natural fibre composites are favoured mainly because of their green image and sustainability. They also exhibit excellent mechanical and thermal properties, and low density as revealed above. Other disadvantages such as water resistance properties should be able to be overcome in the near future. In addition to their environmentally friendly characteristics, green composites should provide excellent economical performance for acceptance in large quantity markets.

References Abe, K. and Ozaki, Y. (1998). Comparison of useful terrestrial and aquatic plant species for removal of nitrogen and phosphorus from domestic wastewater. Soil Sci. Plant Nutr., 44, 599–607.

Natural fibre sources

77

Araco Corporation, Japan (2003) http://www.araco.co.jp/english/e-index.html. Atalla, R.H. (1983). The structure of cellulose: Quantitative analysis by raman spectroscopy. J. Appl. Polym. Sci., Appl. Polym. Symp., 28, 659–69. Bledzki, A.K. and Gassan, J. (1999). Composites reinforced with cellulose based fibres. Prog. Polym. Sci., 24, 221–74. Bos, H.L. and Donald, A.M. (1999). In situ ESEM study of the deformation of elementary flax fibres. J. Mater. Sci., 34, 3029–34. Drumright, R.E., Gruber, P.R. and Henton, D.E. (2000). Polylactic acid technology. Adv. Mater., 12, 1841–6. Fambri, L., Pegoretti, A., Fenner, R., Incardona, S.D. and Migliaresi, C. (1997). Biodegradable fibres of poly(L-lactic acid) produced by melt spinning, Polymer, 38, 79–85. von Frey-Wyssling, A. and Mühlethaler, K. (1963). Die elementarfibrillen der cellulose, Makromolekulare Chemie, 62, 25–30. Haigler, C.H. (1985). The functions and biogenesis of native cellulose. In Nevell, T.P. and Zeronian, S.H., Cellulose Chemistry and Its Applications, John Wiley, NY, pp. 30–45. Han, J.S., Miyashita, E.S. and Spielvogel S.J. (1999). Properties of kenaf from various cultivars, growth and pulping conditions. In Sellers, T. and Reichert, N.A., Kenaf Properties, Processing and Products, Mississippi State Univ., pp. 267–83. Hearle, J.W.S. (1963). The fine structure of fibers and crystalline polymers. III. Interpretation of the mechanical properties of fibers, J. Appl. Polym. Sci., 7, 1207–23. Hepworth, D.G. and Bruce, D.M. (2000). A method of calculating the mechanical properties of nanoscopic plant cell wall components from tissue properties, J. Mater. Sci., 35, 5861–5. Horii, F. (2001). Structure of cellulose: recent developments in its characterization. In Hon, D.N.-S. and Shiraishi, N., Wood and Cellulosic Chemistry, Marcel Dekker, New York–Basel, pp. 83–107. Inagaki, H. (2002). Kenaf; plant of environmental conservation and fiber materials, Kobunshi, 51, 597–602 (in Japanese). Ito, T. (1990). Strength and modulus of fibers. In Society Polymer Science, Japan ‘High Performance Polymer Composites’, Maruzen, Tokyo, pp. 7–68 (in Japanese). Khristova, P., Bentcheva, S. and Karar, I. (1998). Soda-AQ pulp blends from kenaf and sunflower stalks, Bioresource Technol., 66, 99–103. Kimura, T. (2001). Natural fiber composites. In Jpn Soc. Composite Materials, Applications of Composite Materials, Sangyo-Tyosakai, Tokyo, pp. 828–835 (in Japanese). Kroon-Batenburg, L.M.J., Kroon, J. and Northolt, M.G. (1986). Chain modulus and intramolecular hydrogen bonding in native and regenerated cellulose fibres, Polym. Comun., 27, 290–2. Lam, T.B.T. and Iiyama, K. (2000). Structural details of kenaf cell walls and fixation of carbon dioxide. In Abstract of the 2000 International Kenaf Symposium, 14. Liu, A. (2000). World production and potential utilization of jute, kenaf, and allied fibers. In Abstract of 2000 International Kenaf Symposium, 7. Marsh, R.E., Corey, R.B. and Pauling, L. (1955a). The structure of tussah silk fibroin (with a note on the structure of b-poly-L-alanine), Acta Crystallogr., 8, 710–15. Marsh R.E., Corey R.B., Pauling L. (1955b). An investigation of the structure of silk fibroin, Biochim. Biophys. Acta, 16, 1–34. McLaughlin, E.C. and Tait, R.A. (1980). Fracture mechanism of plant fibres, J. Mater. Sci., 15, 89–95.

78

Green composites

Meyer, K.H. and Lotmar, W. (1936). The elasticity of cellulose. IV. The constitution of the crystallized cellulose portion, Helv. Chim. Acta, 19, 68. Mohanty, A.K., Misra, M. and Drzal, L.T. (2002). Sustainable bio-composite from renewable resourses: opportunities and challenges in the green materials world, J. Polym. Environ., 10, 19–26. Nakamae, K. and Nishino, T. (1991). Crystal moduli of high polymers and their temperature dependence. In Lemstra, P.J. and Kleintjens L.A., Integration of Fundamental Polymer Science and Technology-5, Elsevier Science, UK, pp. 121–30. Nakamae, K., Nishino, T., Shimizu, Y. and Matsumoto T. (1987). Experimental determination of the elastic modulus of crystalline regions of some aromatic polyamides, aromatic polyesters, and aromatic polyether ketones, Polym. J., 19, 451–9. Nakamae, K., Nishino, T. and Ohkubo H. (1989). Elastic modulus of the crystalline regions of silk fibroin, Polymer, 30, 1243–6. Nakamae, K., Nishino, T. and Ohkubo, H. (1991). Elastic modulus of crystalline regions of Polyethylene with different microstructures – experimental proof of homogeneous stress distribution-, J. Macromol. Sci.-Phys., B30, 1–23. Nakamae, K., Nishino, T., Gotoh, Y., Matsui, R. and Nagura, M. (1999). Temperature dependence of the elastic modulus of the crystalline regions of poly(p-phenylene benzobisthiazole), Polymer, 40, 4629–34. Netravali, A.N. and Chabba, S. (2003). Composites get greener, Materials Today, 6, 22– 9. Nishi, Y., Uryu, M., Yamanaka, S. et al. (1990). The structure and mechanical properties of sheets prepared from bacterial cellulose. Part 2: Improvement of the mechanical properties of sheets and their applicability to diaphragms of electroacoustic transducers, J. Mater. Sci., 25, 2997–3001. Nishino, T., Nakamae, K. and Takahashi, Y. (1992). Elastic modulus of the crystalline regions of Tussah silk, Polymer, 33, 1328–9. Nishino, T., Takano, K., Nakamae, K. et al. (1995a). Elastic modulus of the crystalline regions of cellulose triesters, J. Polym. Sci., Part B, Polym. Phys., 33, 611–18. Nishino, T., Takano, K. and Nakamae K. (1995b). Elastic modulus of the crystalline regions of cellulose polymorphs, J. Polym. Sci., Part B, Polym. Phys., 33, 1647–51. Nishino, T., Matsui, R. and Nakamae, K. (1999). Elastic modulus of the crystalline regions of chitin and chitosan, J. Polym. Sci., Part B, Polym. Phys., 37, 1191–6. Nishino, T., Kotera, M., Okada, K. et al. (2001). Elastic modulus of the crystalline regions of poly p-phenylene benzobisoxazole using synchrotron radiation, Mater. Sci. Res. Int. Special Technical Pub., 1, 378–81. Nishino, T., Hirao, K., Kotera, M., Nakamae, K. and Inagaki, H. (2003). Kenaf reinforced biodegradable composite, Composite Sci. Technol., 63, 1281–6. Nishiyama, Y., Langan, P. and Chanzy, H. (2002). Crystral structure and hydrogen-bonding system in cellulose I b from synchrotron X-ray and neutron fiber diffraction, J. Am. Chem. Soc., 124, 9074–82. Northolt, M.G., Boerstoel, H., Maatman, H., Huisman, R., Veurink, J. and Elzerman, H. (2001). The structure and proeperties of cellulose fibres spun from an anisotropic phosphoric acid solution, Polymer, 42, 8249–64. Ohya, S. and Muraoka, Y. (1978). Raw Materials of Fibers, Aikawa Shobou, Tokyo (in Japanese). Osaki S. (1996). Spider silk as mechanical lifeline, Nature, 384, 419–21. O’Sukkivan, A.C. (1997). Cellulose: the structure slowly unravels, Cellulose, 4, 173– 207.

Natural fibre sources

79

Page, D.T. and El-Hosseiny, F. (1983). The mechanical properties of single wood pulp fibres. Part VI: Fibril angle and the shape of the stress-strain curve, J. Pulp Paper Sci., 9, 99–100. Peijs, T. (2001). Composites turn green, e-Polymer, Special Issue, 59–63. Ramaswamy, G.N. (1999). Processing kenaf bast fibers: chemical retting. In Sellers T., Reichert N.A., Kenaf Properties, Processing and Products, Mississippi State Univ., pp. 91–6. Reiterer, A., Lichtenegger, H., Tschegg, S. and Fratzl, P. (1999). Experimental evidence for a mechanical function of the cellulose microfibril angle in wood cell walls, Phil. Mag. A, 79, 2173–84. Rowell, R.M. and Han, J.S. (1999). Changes in kenaf properties and chemistry as a function of growing time. In Sellers, T. and Reichert, N.A., Kenaf Properties, Processing and Products, Mississippi State Univ., pp. 33–41 Sakurada, I., Ito, T. and Nakamae, K. (1966). Elastic moduli of the crystal lattice of polymers, J. Polym. Sci., C15, 75–91. Sawai, D., Takahashi, K., Sasashige, A., Kanamoto, T. and Hyon S.H. (2003). Preparation of oriented b-form poly(L-lactic acid) by solid-state coextrusion: effect of extrusion variables, Macromolecules, 36, 3601–5. Shao, Z.Z. and Vollrath, F. (2002). Surprising strength of silkworm silk, Nature, 418, 741. Stamboulis, A., Baillie, C.A. and Peijs, T. (2001). Effects of environmental conditions on mechanical and physical properties of flax fibers, Composites A, 32, 1105–15. Stromme, M., Mihranyan, A. and Ek, R. (2002). What to do with all these algae?, Mater. Lett., 57, 569–72. Sugiyama, J, Vuong, R. and Chanzy, H. (1991). Electron diffraction study on the two crystalline phases occurring in native cellulose from an algal cell wall, Macromolecules, 24, 4168–75. Takahashi, Y., Gehoh, M. and Yuzuriha, K. (1991). Crystal structure of silk (Bombyx mori), J. Polym. Sci., Polym. Phys. Ed., 29, 889–91. Takai, M. and Tajima, K. (2000). Crystal polymorphs of cellulose. In Society Cellulose Japan, Encyclopedia of Cellulose, pp. 93–9, (in Japanese). Taniguchi, T. and Okamura, K. (1998). New films produced from microfibrillated natural fibres, Polym. Int., 47, 291–4. Tashiro, K. and Kobayashi, M. (1991). Theoretical evaluation of three-dimensional elastic constants of native and regenerated celluloses: role of hydrogen bonds, Polymer, 32, 1516–26. Tsuji, H. (2002). Polylactides. In Doi, Y. and Steinbuchel A., Biopolymers, vol. 4 (Polyesters 3), Wiley–VCH, Weinheim, Germany, pp. 129–77. VanderHart, D.L. and Atalla, R.H. (1984). Studies of microstructures in native cellulose using solid-state 13C NMR, Macromolecules, 17, 1465–72. Vollrath, F. and Knight, D.P. (2001). Liquid crystalline spinning of spider silk, Nature, 410, 541–8. Wada, M., Okano, T. and Sugiyama, J. (1997). Synchrotron-radiated X-ray and neutron diffraction study of native cellulose, Cellulose, 4, 221–32. Wada, M., Okano, T. and Sugiyama, J. (2001). Allomorphs of native crystalline cellulose I evaluated by two equatorial d-spacings, J. Wood Sci., 47, 124–8. Wada, M., Kondo, T. and Okano, T. (2003). Thermally induced crystal transformation from cellulose Ia to Ib, Polym. J., 35, 155–9. Wang, H.H., Drummont, J.G., Reath, S.M., Hunt, K. and Watson, P.A. (2001). An improved fibril angle measurement method for wood fibres, Wood Sci. Technol., 34, 493–503.

80

Green composites

Wu, Q. and Berglund, L.A. (2003). Algae: from environmental treat to nanocomposites, Abstract 2nd International Conference EcoComposites. Yamamoto, H. and Horii, F. (1993). CP/MAS 13C NMR analysis of the crystal transformation induced for Valonia cellulose by annealing at high temperatures, Macromolecules, 24, 1313–17. Yamamoto, H., Horii, F. and Hirai, A. (1996). In situ crystallization of bacterial cellulose. II. Influences of different polymeric additives on the formation of celluloses Ia and IIb at the early stage of incubation, Cellulose, 3, 229–42. Yano, H. and Nakahara, S. (2003). Bio-composites produced from plant microfiber bundles with a nanometer unit web-like network, J. Mater. Sci., 39, 1635–8.

5 Alternative fibre sources: paper and wood fibres as reinforcement P. P E LT O L A Tampere University of Technology, Finland

5.1

Introduction and definitions

Paper and cardboard are the most important and versatile commodities in the world. Paper consumption is one of the criteria for measuring the standard of living. In the USA over 300 kg per capita of paper are used per year and in Finland the consumption is about 250 kg per capita. Most of the paper in the world is used in packaging and in printing papers (Häggblom-Ahnger and Komulainen, 2000). Paper fibre in this context means the raw material used in paper processing industry and is in the shape of a fibre. It is mechanically ground or chemically made organic material from trees. Both mechanically made and chemically made cellulose fibres are called wood fibres. Mechanically made wood fibres contain lignin but the lignin has been removed from the cellulose fibres. Both fibre types are often mixed in the processing of paper in order to combine their properties. The mechanical properties/density of wood fibres can be compared with those of synthetic fibres such as glass fibre. These properties are compared in Table 5.1. The specific strength is the strength of the fibre divided by the density of the fibre, and the specific modulus is the modulus of the fibre divided by the density of the fibre.

Table 5.1 Comparison of the properties of wood fibre and glass fibre Property Density (g/cm3) Tensile strength (GPa) Specific strength (GPa cm3/g) Modulus of elasticity (GPa) Specific modulus (GPa cm3/g)

Wood fibre 0.6–1.1 0.98–1.77 1.63–2.95 10–80 17–133

Glass fibre (E-glass) 2.6 3.5 1.35 72 28

81

82

Green composites

5.1.1

Composites made of wood fibres and plastics

One of the most important properties of the paper fibres used to reinforce, for example, polyolefins is their polarity. In mechanically made fibre the lignin decreases their polarity. Cellulose fibres are polar because of the OHgroup in their structure. This is the problem in composites containing cellulose and polyolefin. The non-polar polyolefin cannot wet the polar fibre without coupling agents. On the other hand, lignin in mechanically made fibre improves the adhesion between fibre and polyolefin. Because of this, mechanically made fibres seem to work better in paper–polyolefin composites. Without the coupling agent cellulose fibres work mainly as the filler in the composite. From the point of view of recycling, the use of paper fibres and, for example, polyolefins is necessary because of the regulations concerning packaging materials. The amounts used in the world per year are huge. In Europe it is estimated that 35 million tonnes of plastic were processed in 2002. Only 3 million tonnes of plastic waste are recycled and 14 million tonnes are used as energy or otherwise utilised. In recycling plastics the cleanness of the waste is a problem. If it were possible to use plastic packages containing, for example, a label in processing, cleaning would not be necessary and the paper could work as a reinforcement (Peltola, 2002). Recycling of paper and plastic is desirable for social and many other reasons. ∑ Paper and plastic are available in abundance and are two of the largest landfill contributors in the Western world. ∑ In Finland the law currently prescribes that a minimum of 15% of plastic packages have to be recycled. ∑ Paper fibres contribute favourably to many properties of composites, the advantageous properties of paper fibres being renewability, biodegradability, low price, abundance and strength. ∑ Recycled fibre and the regenerated plastic binder are advantageous and environmentally friendly. ∑ At the end of the circle, the products can be burnt for energy recovery.

5.1.2

Definitions

In this text there are many words used that need defining. Collection

Paper Paper recycling

Separate collection of paper and paper products from different sources, for example, from industrial outlets, from households and offices for recovery. Overall term of all grades of paper and board unless otherwise mentioned. Reprocessing of recovered paper in a production

Alternative fibre sources

Recycled paper/fibre Waste

Noil

83

process for its original purpose or for other purposes. Composting is included but the energy recovery is excluded. Recovered paper, used paper or board that has been classified by European standadard EN 643. Any substance that is discarded or intended or required to discard. Once the paper products have been collected and processed for recycling, they become valuable secondary raw material and are no longer considered as waste. Flour, fines (zero fibre)

The term ‘paper’ in most articles is used for recycled newsprint made of mechanical pulp. In addition to wood fibres, this kind of paper contains natural binders and lignin. Small amounts of printing ink and other additives are also present.

5.2

Wood fibres: structure, properties, making pulp and paper fibres

In Europe and Asia the same kind of coniferous trees are used for fibres. The trees used are pine, spruce and larch, although larch has such a large content of extractives that it is not as suitable for pulp as pine and spruce. Coniferous trees in Canada and the USA differ in some ways from the coniferous trees in Europe and in Asia. In Canada there are about 13 different types of pine trees. From the point of view of pulp industry the coniferous trees are not so different in chemical structure but differ greatly in the extractive contents (Häggblom-Ahnger and Komulainen, 2000). In central and southern Europe the most common deciduous trees used in the pulp industry are beech, white beech and oak. There are many useful deciduous trees growing in north America, for example, aspen, beech, linden and oak. In north Europe (e.g. Finland) the most common deciduous raw material is birch. The main competitor for birch in the pulp industry is Eucalyptus from Asia, southern Europe, south America and Australia (Häggblom-Ahnger and Komulainen, 2000). Of the world’s paper, 5–6% is made of non-wood fibres. The processing is concentrated in the areas where there is a shortage of wood. In the cross-section of a piece of wood a dark pith can be seen and around it the cell wall substance. In the surface of the tree there is a thin layer called the cambium that creates a new wood layer inside it each year. On the outside the cambium creates a bast layer that, in turn, creates the bark. An Annual ring consists of the light earlywood (grown in the spring) and the dark latewood (grown in the summer). A cross-section of a pine tree is

84

Green composites

5.1 Cross-section of a pine tree. The annual rings can clearly be seen as well as light earlywood and dark latewood.

shown in the Fig. 5.1 and the annual rings can clearly be seen. The difference between earlywood and latewood is, for example, the chemical structure and the density. The latewood has a higher cellulose content than the earlywood (Häggblom-Ahnger and Komulainen, 2000). Fibres in wood are long cells. They vary in size and function. Each cell forms a fibre. A fibre consists mainly of cellulose, hemicellulose and lignin. The cellulose forms the frame of the cell wall; hemicellulose and lignin, on the other hand, form the surrounding intercellular substance. Cellulose and hemicellulose are hydrophilic chain polymers that have hydroxyl groups (-OH) (Häggblom-Ahnger and Komulainen, 2000). The wall of the wood fibre is built up of an outer layer (primary wall) and a three-layered inner layer (secondary wall). The fibres are surrounded by lignin. The outer layer consists mainly of lignin and has only a little cellulose. The inner layer consists mostly of cellulose and hemicellulose and has only some lignin (Häggblom-Ahnger and Komulainen, 2000). As mentioned earlier, cellulose is the main component of wood fibre. It therefore determines the properties of the fibres and makes it possible to use them, for example, in the paper-making industry. Having a hydrophilic nature, their ability to absorb water is strong and therefore water will affect the properties of the fibres. The humidity can change the dimensions but also affect the mechanical properties (Häggblom-Ahnger and Komulainen, 2000). Hemicellulose molecules are smaller than those of cellulose. They can be

Alternative fibre sources

85

divided into three groups: hexocane, pentosane and polyurenoids. Hemicellulose affects the ability of fibres to form bonding between each other (Häggblom-Ahnger and Komulainen, 2000). Lignin is a polymer that bonds fibres and gives wood its stiffness. The lignin content depends on the type of tree. In coniferous trees there is about 28% (dry weight) of lignin and in deciduous trees about 20% (dry weight) of lignin. If the cellulose and hemicellulose are hydrophilic materials, the lignin is hydrophobic (Häggblom-Ahnger and Komulainen, 2000).

5.2.1

Differences between fibres

There are some differences between coniferous trees and deciduous trees in their cell structures. The coniferous trees contain two types of cells: tracheid cells 90–95% and parenchyma cells. Tracheids are long, narrow and their diameter tapers towards the ends. Their length to diameter ratio is 100 : 1 and their length is 2–5 mm. The diameter varies from 20 to 30 mm depending on the type of tree. Parenchyma are shorter, being about 200 mm in length (Vikman, 1999). Deciduous trees consist of a variety of different cells. The main types are tracheids, tube cells and parenchyma and also their intermediate forms. It is hard to say which kind of cells they are. Usually, all the supportive cells are just called fibres. The tracheids and parenchyma content is quite small. Fibres are the biggest cell group in deciduous trees and, for example, in the birch about 60% of the cross-section area contains fibres. They have a thick wall and are tube-like. Their length is 1–2 mm and length to diameter ratio is 100 : 1. The second largest cell group is the trachea. They are quite short and thick-walled fibres. Usually, they are under 1 mm long and their diameter varies from 60 to 300 mm (Vikman, 1999; Häggblom-Ahnger and Komulainen, 2000).

5.2.2

Pulp-making processes

Pulp for the paper industry can be made by chemical or mechanical processes. Chemical pulp can be made by two different methods: sulphate or sulphite methods. The process is based on chemical fibrillation, and the lignin that bonds the fibres together is dissolved. The strength of the sulphite pulp is much less than the strength of the sulphate pulp. Lignin is not the only material that is dissolved in the chemical process, since part of the hemicellulose also dissolves. This means that the yield reduces and is usually around 50%. For 1 tonne of pulp, two tonnes (5–6 m3) of wood are needed. As a result of a chemical process, fairly pure pulp cellulose is gained (Häggblom-Ahnger and Komulainen, 2000; Vikman 1999). The two basic methods for processing mechanical pulp are grinding and

86

Green composites

mechanical pulping in a refiner. The lignin is softened by using heat and dynamic stress. In this way the fibres are separated but none of the components is diluted. The grinding process will continue until the pulp has the required degree of fineness (Canadian standard freeness (CSF)). For mechanically made pulp, there are many different methods in use (Häggblom-Ahnger and Komulainen, 2000). The properties of the cellulose fibres and mechanically made fibres are different depending on the processing method. The cellulose fibres are long, whole and flexible and the mechanically made fibres consist of many differentsized particles: long fibres, medium-sized fibres and fines. Mechanically made fibres have the lignin still on them and this makes them stiff and the bonding poor. The mechanical properties are inferior compared to those of the cellulose. These type of fibres are used, for example, in printing papers (mechanical), the inner layer in the cardboards, in some fine papers and as newsprint (Häggblom-Ahnger and Komulainen, 2000).

5.2.3

Recycled fibres in the paper-making industry

Recycled paper plays an important role as a raw material in the papermaking industry around the world. The paper that is collected must be dispersed, sorted out and cleaned. After these phases it is de-inked and ground mechanically. The recycled fibres are added to the pure fibres. The paper cannot be made merely from the recycled fibres since the properties of the fibres will decrease in the recycling process. The fibres will tolerate recycling four to six times before the properties become too weak. Recycled fibres are used mainly in newsprint, cardboard and fine papers but also in magazine paper (Häggblom-Ahnger and Komulainen, 2000; www.paperrecovery.org).

5.2.4

Paper fibres

The lignin content is higher and the hemicellulose content lower in coniferous trees than in deciduous. Therefore, deciduous trees have a high hemicellulose content that remains in the pulp. They give an abundant yield and this affects the properties of the pulp. Despite the differences in fibre type and content of cellulose, hemicellulose and lignin, the length of the fibres has a greater affect on the differences between coniferous and deciduous trees. Fibre length affects the mechanical strength properties of the pulp and paper, and the width and thickness of the fibres affects their flexural properties. For example, if the pulp is made of long fibre coniferous pulp, a stronger paper can be produced but only up to a certain point after which the strength begins to decrease (Häggblom-Ahnger and Komulainen, 2000). Early wood is considered better in the paper-making industry, since the thickness of the wall is less. Thin fibre is easier to flatten and the fibres will

Alternative fibre sources

87

bond better to each other. This will lead to a denser and stronger paper (Häggblom-Ahnger and Komulainen, 2000). In the paper-making industry, fibre that has been chemically made is called cellulose and the lignin is removed in the sulphate or sulphite and heat treatments. Therefore, cellulose fibres are more flexible, reactive and elastic. In mechanically made fibres, the lignin remains in the fibre (Vikman, 1999). Because of the chemical process, cellulose fibres are longer than mechanically made fibres, which are ground. The mechanical pulping process leads to the pulp having some splinters that have not been fibrilised in the process. These ‘sticks’, on the other hand, will stiffen the paper (Vikman, 1999). Paper is a three-dimensional fibre net in which the fibres are bonded together. There are also different types of additives and fillers present to make bonding easier and to fill the pores between the fibres. The most common fillers are talc, calcium carbonate (CaCO3), kaolin and titanium anhydride (TiO2). They are used to lower the price and improve the optical properties of the paper. Additives are used to improve the retention and formation in the wire and to improve the mechanical properties of the paper (Vikman, 1999). Paper pulps consist of fibres of different length. The most important parts are the long fibres and the noil that includes parts of the cell walls and the microfibrils. Cellulose pulp has lots of long fibres and just a little noil, whereas in the mechnical pulp there are fewer long fibres and more noil, (Vikman, 1999). Both chemical and mechanical pulp have their strengths and applications. Cellulose gives strength and mechanical pulp improves the printing and optical properties in the paper. Usually both pulps are used together in paper. For example, newsprint consists of the following parts (Vikman, 1999): ∑ ∑ ∑ ∑

80–100% mechanical pulp 0–20% cellulose pulp 0% fillers pigments if used.

Light weight coated paper (LWC) in turn consists of the following parts: ∑ 1/3 mechanical pulp ∑ 1/3 cellulose pulp ∑ 1/3 pigments.

5.3

Recycling of paper

Paper for recycling can be collected from different sources. Usually the sources are divided into household waste and industrial waste. It is evident

88

Green composites

that the collection methods depend on the source of the waste. The amount of recycled paper from industry is about 52%, from offices about 10% and from households 38% (www.paperinkerays.fi). The biggest source of recycled paper is industry. Recycled paper from industry consists of cardboard boxes, corrugated papers, cardboard, unbleached papers, scrap paper from printing industry, unsold magazines, etc. (www. paperinkerays.fi). Household paper is very waste varied and the cleanness of it is hard to control. Therefore, it is important that people follow given instructions. Recycled paper from households consists mainly of newsprint, envelopes, copy paper, magazines, cardboard, corrugated cardboard, etc. Usually paper is collected separately from other household waste. There are different practices in different countries in how the different types of papers are sorted out. In some countries paper, cardboard, plastics, etc. are collected together, whereas in others, such as Finland, newsprint and magazine paper are collected separately from other paper, cardboard and corrugated cardboard (www.paperrecovery.org, www.paperinkerays.fi). Recycled paper can be used in many different applications. For example, cardboard paper can be used in core stocks in the paper-making industry, whereas paper from households can be used as newsprint and for fine paper such as toilet paper and tissues. The most important use is for making new paper. It is possible to recycle paper four to six times before the properties and length of the fibres are reduced too much. This is why an amount of pure fibre material has to be added to recycled fibres in the paper making process. Table 5.2 shows how paper and cardboard are recycled (www.paperinkerays.fi). In 2000 the European Recovered Paper Council ERPC gave the European Declaration on Paper Recovery. It was announced that 56% of all paper and cardboard in Europe should be collected and recycled by 2005. For industry this would mean an increase of 10 million tonnes into recycling. In 2001 alone in the EU countries the use of recycled paper was 39.5 million tonnes and board consumption was 76 million tonnes. This means in percentage tonnes that consumption of paper decreased by 4% but the consumption of recycled fibre increased by 0.8%, meaning that recycling increased from 49.7% to 52.1%. For the paper industry the raw material in Europe consisted of 42% recycled fibre and 43% pure raw material (wood fibres). In Table 5.3 the collection rate of paper in different countries in 2001 is presented. As can be seen, the highest rates are in Germany, Finland and south Korea. In Finland and in other countries where most of the paper produced is exported, the amount of pure material will be high, although most of the paper waste is collected and recycled. For Finland this is presented in Fig. 5.2 (metsäteollisuus ry 2002; www.paperinkerays.fi). In 2001, 2.7 million tonnes of recycled paper were collected from CEPI countries and exported to third countries for recycling purposes.

Alternative fibre sources

89

Table 5.2 New life of paper and cardboard Type of paper

Source of the paper

How is it separated?

Newspapers, Household, magazines, printing houses, etc. offices Office paper

Companies, other communities

Corrugated cardboard

Industry, stores

Cardboard

Households, restaurants, schools

Cartons

Households, restaurants, schools

Different types of papers are separated, cleaned, baled and restored and sent to industry for raw material. Secret papers are usually shredded before baleing.

What happens in paper factory?

Products

The printing ink is removed.

Newsprint, reference books, catalogues, fine papers. Fine papers, for example toilet paper, kitchen paper, Kleenex

The printing ink is removed.

Fibres of the corrugated cardboard are separated from other possible layers. Fibres of the cardboard are separated from the other layers. Plastic coating and aluminium lining are separated from fibres.

Cardboard for packaging and cases and the inner layers of the corrugated cardboard. Re-used as a raw material.

Plastic for energy and aluminium is re-used.

Source: www.paperinkerays.fi/tietoa/tietopankki/kierokulku. Table 5.3 The collection rate in different countries in 2001 Country

Collection rate (%)

Germany Finland South Korea Switzerland Norway Austria Sweden The Netherlands Taiwan Japan Spain Western Europe Belgium Australia France Hungary

74 74 70 69 66 65 65 64 61 58 55 55 51 48 47 46

Source: www.paperinkerays.fi.

90

Green composites 800 700

1000 tonnes

600 500 400 300 200 100 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Year Collection rate

Utilisation

5.2 The collection and utilisation of the recycled paper in Finland in 1990–2002 as raw material (source: www.paperinkerays.fi).

5.4

Wood and plastic composites and the theory of fibre reinforcement

Composite materials are affected by the compatibility of the phases. In fibrereinforced composites the composite strength is determined by the strength of the fibre and by the ability of the matrix to transmit stress to the fibre. Transmission of stress to the fibre is affected by fibre orientation (as opposed to stress direction), geometry (e.g. diameter) and interfacial bond between fibre and matrix. The critical fibre length is the minimum length necessary for effective transmission of stress from matrix to fibre. It is possible to decrease this critical length by improving the interfacial bonds. The composite with natural wood fibres contains a wide distribution of fibres, both longer and shorter than the critical length, making a theoretical calculation of the composite properties difficult. Furthermore, for example, the fibre orientation in the injection-moulded test specimen is not random, making comparison with theoretical results still more vague. Some theories have been introduced in the literature. In wood fibre reinforced thermoplastics theoretical strength calculations are quite difficult to perform for the following reasons (Sanadi et al., 1994). ∑ The constant of orientation is hard to define for the injection moulded parts.

Alternative fibre sources

91

∑ The boundary surface of the fibre and matrix is hard to define because of the irregular topography of the fibre. ∑ There is a wide length distribution of the fibres from both sides of the critical length. ∑ The strength of the fibre varies depending on the source and age. The tensile strength of the fibre reinforced plastic composites can be estimated by modifying the rule of compounding for plastic compounds (Sanadi et al., 1994). sc = Vm s *m + Vf s f K1K2, where sc = sf = Vm = Vf = K1 = K2 = s *m =

[5.1]

tensile strength of the composite tensile strength of the fibre volume fraction of the matrix volume fraction of the fibres orientation constant of the fibres constant that is dependent on the adhesion between fibre and matrix the effect of the tensile strength of the matrix on strain at break of the composite.

Constant K1 is dependent on the fibre orientation. For unidirectional fibres the constant is 1. If the fibres are in three directions, the constant will be 1/5. K2 is dependent on the adhesion between fibre and matrix and on the length of the fibres. The critical length of the fibres, lc, can be defined for the reinforcement. If the length is smaller than lc the fibres are unable to gain the maximum strengthening effect and fracture will occur due to some other mechanism. For fibre lengths less than the critical length, constant K2 can be defined: K2 = l/2lc,

[5.2]

where l = length of the fibre. In this case the maximum value of K2 is 0.5. If the fibre length is the same or longer than the critical length, then K2 is between 0.5 and 1, as can be seen in equation (3) (Sanadi et al., 1994; Vikman, 1999). K2 = l – lc /2l

[5.3]

If the adhesion between matrix and fibre is adequate, the stresses should disappear by the fibre breaking when the fibre length is longer than the critical length.

92

5.5

Green composites

Composites made of wood or wood fibre and plastics

The composite in this case means the wood or wood fibres used as reinforcement in the plastic material. There have been, and are, various wood-based composites in use for a long time in different applications. Wood can be used in many different forms as reinforcement, e.g. in the form of wood sheets, wood flours or fibre (paper). Balsa is one of the wood materials used traditionally as reinforcement in sheet composites. It is used mainly in sheet form as core material in sandwich structures. The fibres are perpendicular to the face of the composite. The strength of the balsa divided by its weight is much better than, for example, that for the other commonly used core material – expanded polystyrene (PSE). Balsa is easy to work with and easy to glue. The disadvantage of balsa is that it absorbs the resin used, which will increase the weight of the sandwich structure. As always with natural materials, the quality of the material can vary greatly and the strength will vary with the quality of the wood (Airasmaa et al., 1991). Wood and wood fibres have many good qualities. The mechanical properties/ density of the wood fibres can be compared with those of synthetic fibres such as glass fibre as was shown earlier in Table 5.1. The properties of the composites are affected by many factors: by materials, content of fibres and processing parameters. In composites some problems are inevitable and depend on the matrix material. The fibres are polar materials and so there is poor adhesion between fibres and non-polar polymers such as polyolefins. Cellulose fibres are polar because of the OH-group in their structure. This presents a problem in composites containing cellulose and non-polar polymers that cannot wet the polar fibre without coupling agents. On the other hand, the lignin in mechanically made fibres improves adhesion between fibre and polyolefin. Without a coupling agent, cellulose fibres work mainly as a filler in the composite. The problem with thermoplastics is their viscosity and their melting or processing temperature. The viscosity of thermoplastics is so high that it is almost impossible for them to wet the fibres and to form adequate adhesion between matrix and fibres. The processing temperatures of thermoplastics are well above 100 ∞C but the wood fibres will at least be partly destroyed at lower temperatures. Both these problems (wetting ability and processing temperature) are not so great with thermoset polymers. The viscosity and processing temperature are lower so there is good adhesion between matrix and fibres. Nowadays, the main interest in the field of wood fibre plastic composites is in thermoplastics. The wood fibre form, dimensions, treatment of fibres and the possible fillers and impurities will effect the properties of the composites. The effect of fibre length is not unambiquous. From the theory of fibre reinforcement,

Alternative fibre sources

93

the longer the fibre, the better the strength of the composite should be. On the other hand, shorter fibres should disperse more easily and the common impression is that good dispersion increases properties of the fibre more than adhesion (Bolton, 1994). If adhesion and dispersion are adequate, wood fibres can work as reinforcement in the polymer matrix instead of being only fillers. In this way the mechanical properties of the composites can be improved by using wood fibres. The composite properties will depend on the fibre content. For example, the tensile strength of the material can be increased to a certain point after which it will begin to decrease. The stiffness of the composites will increase if the fibre content is increased but the strain and impact strength will decrease (Vikman, 1999). The most important factor to affect the properties of the composites is the adhesion between fibres and matrix. In paper making, cellulose fibres are better to handle because of their reactivity. On the other hand, the lignin forms better adhesion between hydrophobic and non-polar polymers such as polyolefins if used in wood–plastic composites. If adhesion is poor, wood fibres work mainly as filler in the matrix and therefore the tensile strength can even decrease. When the adhesion is adequate, the wood fibres will work as reinforcement in the matrix. Adhesion can be improved by coupling agents such as silanes. The natural ability to form adhesion between matrix and fibre is also dependent on the fibre. In paper fibres the cellulose fibres (chemical fibres) give poorer adhesion than the mechanically made fibres. Chemical treatment dissolves out the lignin from the fibre and therefore there are fewer hydroxyl groups in the structure to form bonds. Mechanically made fibres, on the other hand, have the lignin still in the structure and the hydroxyl groups are able to form bonds. Mechanically made fibres can be used without expensive coupling agents but the properties can be improved with good adhesion. Water absorption is also quite an important problem with all wood materials. Water absorption and decaying are characteristic of wood materials. In composites the water absorption of the matrix material increases due to the hydrophilic wood fibres. It has been shown that water absorption can be decreased by using fillers that react with the hydroxyl groups in the fibre (Raj and Kokta, 1989). With injection moulded parts the skin in the surface will affect the water absorption of the composite. The composites made of polyolefin and wood fibre will absorb less moisture than commercial hardboards in which wood flour is used as a filler (More, 1997).

5.5.1

The properties of the wood fibre and plastic composites

Natural fibres such as wood fibres have been investigated as reinforcing agents in plastic composites. One of the reasons for the interest is that cellulose

94

Green composites

fibres have a higher specific strength than glass fibres and a similar specific modulus. Given these properties, combined with an inexpensive source of fibres, wood fibre–plastic composites could theoretically offer desirable specific strengths and modulus at a low cost. Wood fibres including paper fibre, wood flour and wood are a good natural choice for synthetic fibres. Environmental issues are one of the basic concerns in wood fibre–plastic composites. Wood fibre is a biodegradable natural material from a renewable natural source. If we examine the situation from the point of view of the waste in the world, one very interesting fact is that wood fibre can be recycled. One way of doing this is to compound, for example, wood flour or recycled paper fibre with thermoplastics. The recycling of paper and plastic is desirable for social reasons and many others. Paper and plastic are available in abundance and are two of the largest landfill contributors in the Western world. Laws around the world prescribe that plastic packages have to be regenerated. Paper fibres contribute favourably to many of the composite properties and the advantageous properties of paper fibres are renewability, biodegradability, low price, abundance and strength. Recycled fibre and the regenerated plastic binder are advantageous and environmentally friendly. Finally at the end of the cycle the products can be burnt for energy recovery. In ANTEC-95 Dale (1995) presented results of tensile tests from compounds made of old newsprint and reground PE–HD. Some of the results are shown in Fig. 5.3. As can be seen, the tensile modulus is increased compared to the Tensile strength 100 = 52·30 MPa 100 80 60

Fibre content 100 = 50%

Modulus of elasticity 100 = 4.42 GPa

40 20 0

Impact strength 100 = 100 kJ/m2

Melt flow rate 100 = 10 g/10 min

Notched impact strength 100 = 100 kJ/m2 PE-HD PE-HD+onp 30%

PE-HD+onp 10% PE-HD+onp 40%

PE-HD+onp 20%

5.3 Test results from compounds made of reground PE–HD and old newsprint (onp) with different paper content.

Alternative fibre sources

95

Tensile strength 100 = 52.30 MPa 100 80 Fibre size 100 = 60 mesh

Modulus of elasticity 100 = 4.42 GPa

60 40 20 0

Fibre content 100 = 50%

Melt flow rate 100 = 10 g/10 min PP + np 10% + PVA 5%

Impact strength 100 = 100 kJ/m2

Notched impact strength 100 = 100 kJ/m2 PP + np 25% + PVA 10%

PP + np 25% + PVA 10%

5.4 Properties of PVA-treated newsprint (np).

pure PE–HD when newspaper is added. Tensile strength, on the other hand, remains almost the same. There has been a lot of research into improving the properties and the adhesion of the wood–plastic composites. There have been many studies to determine the best coupling agents, fibre content and processing methods for best adhesion and wetting, optimal properties and to avoid thermal degradation of fibres. Chemical coupling agents have been proved to be best for increasing adhesion. The other reason for their wide use is their ease of use and effectiveness. Most studies are made with silanes and maleic anhydride modified polymers. An investigation of the interfacial adhesion between reclaimed newspaper and recycled polypropylene composites was reported by Maldas and Kokta (1994). In Fig. 5.4 the tensile properties of PVA-treated newspaper are shown. PVA has been added for its coupling action. Tensile and impact properties and different coupling agents of polyolefin– wood fibre composites have been the subject of much study. In Fig. 5.5 the results of one study of the effect of two different coupling agents compared with a compound with no added coupling agent are shown (Sanadi et al., 1994). As can be seen, the coupling agent has the biggest influence on the tensile strength and impact strength but the effect on stiffness and notched impact strength is minimal. The most usable matrix materials in wood fibre composites are those with low melting points such as polyolefins, polystyrenes, thermosets, etc. Some studies have been done with a rubber matrix and a biopolymer matrix (Sameni et al., 2002). High temperature thermoplastic polymers (polyphenylene ether

96

Green composites Tensile strength 100 = 52.30 MPa 100 80 60

Fibre content 100 = 50%

Modulus of elasticity 100 = 4.42 GPa

40 20 0

Impact strength 100 = 190 J/m

Melt Flow rate 100 = 10 g/10 min

Notched impact strength 100 = 20.8 J/m None

E-43

G-3002

5.5 The effect of two different coupling agents compared with a compound with no added coupling agent.

PPE) reinforced with wood flour have also been studied. The problem with the high temperature thermoplastics is their high processing temperature compared with the decomposition temperature (ca. 200 ∞C) of wood fibres. Low weight thermoset liquid epoxy can be used as the reactive solvent of PPE to reduce the thermal gap between wood flour and PPE. The composite will then consist of a thermoplastic continuous phase and two dispersed phases: polymerised epoxy and wood flour particles coated with epoxy. The glass transition temperature of the PPE could be decreased to make the processability easier (Jana and Prieto 2002). The most commonly used polymer materials in wood fibre composites are still the polyolefins. They are widely available and their processing temperatures are low enough for wood fibre not to degrade. Sometimes waste polymers already have some paper combined with them. In these cases cleaning of the waste would be laborious. This is one reason why many studies have been done on compounding recycled paper and recycled polyolefins. Recent studies have produced some interesting results. The properties of the recycled plastics improved when recycled paper (usually newsprint) was added. In these composites paper fibres were able to disperse evenly on the matrix and tensile strength and stiffness were almost doubled. Figure 5.6 shows an SEM picture of composite made of polyethylene and newsprint. The adhesion in the sample has not been very good and a fibre pull-out phenomenon can be seen, althrough dispersion of the fibres is quite good.

Alternative fibre sources

97

5.6 SEM picture from the fracture surface of a PE-LD newsprint composite. The fibre content is 30 wt% and the sample has been mixed by a Brabender batch-type mixer and injection moulded to standard test specimens.

Despite the size of the paper fed into the compounding machine, dispersion was noted to be good and fibres were separate. If the fibre content rises over 50 wt%, processing of the composite becomes difficult and the properties begin to decrease. In cellulose fibres coupling agents would improve the adhesion between fibres and matrix and cause the properties to improve. On the other hand, when mechanically made fibres are used, it has been observed that the coupling agents have little or no effect on the properties. This is considered to be a consequence of the presence of hydroxyl groups (OHgroups) in the lignin. In Fig. 5.7 there are results that prove it is not necessary to use coupling agents to improve properties. Processing of thermoplastic composites can be, for example, by injection moulding, extrusion, compression moulding or transfer moulding methods. Processing temperatures should be as low as possible to avoid thermal degradation of the fibres and the odour of burnt paper. Through a combination of wood fibres, thermoplastics, modifiers and compounding methods it is possible to create new composite materials that have a desirable combination of properties. These composites can be used, for example, in building, engineering and in the automotive industry. They can also be used for processing furniture, storage bins, recreation boats, toys, games, etc. (Bledzki et al., 1998).

98

Green composites 50 44.8 45

46

45.5

46.3

47.6

42.4

Tensile strength (MPa)

40 35 30

28

25 20 15 10 5 0 PP

PP+onp 30%

PP+onp 30% PP+onp 30% PP+onp 30% PP+onp 30% MAHPP 0.5% MAHPP 1% + MAHPP 2% + MAHPP 3%

PP+onp 30% + MAHPP 5%

5.7 The effect of coupling agent on tensile strength in case of a PPbased composite (MAHPP=malgic anhydride modified polymers).

Acknowledgements First I would like to thank the editor and publisher for the opportunity to write this chapter. It has been a very educational experience. Secondly, I would like to thank all my co-workers who have been studying paper plastic composites here in Tampere University of Technology, e.g. Jouni Vikman, for his Master’s thesis, Owe Ulfsted, Liisa Kukko and all the others involved, for their efforts in this study. I also want to thank all the authors around the world who have published their studies, which I have found very helpful in writing this chapter.

References Airasmaa, I., Kokko, J., Komppa, V. and Saarela, O. (1991). Muovikomposiitit, Jyväskylä, Muoviyhdistys ry (in Finnish). Bledzki, A.K., Reihmane, S. and Gassan, J. (1998). Thermoplastics reinforced with wood fillers: a literature review. Polymer–Plastics Technology Engineering, 37/4, 451–68. Bolton, A.J. (1994). Natural fibres for plastic reinforcement, Materials Technology, 9, 12–20. Dale, B.T. (1995). Compounding processed old news print with recycled high density polyethylene to make a substitute lumber product. ANTEC. Häggblom-Ahnger, U. and Komulainen, P. (2000). Kemiallinen metsäteollisuus II Paperin ja kartongin valmistus. Opetushallitus, Helsinki (in Finnish). Jana, S.C. and Prieto, A. (2002). On the development of natural fiber composites of hightemperature thermoplastic polymers, Journal of Applied Polymer Science, 86, 2159– 67. Maldas, D. and Kokta, B.V. (1994). An investigation of the interfacial adhesion between reclaimed newspaper and recycled polypropylene composites through the investigation of their mechanical properties, Journal of Adhesion Science Technology, 8/12, 1439– 51.

Alternative fibre sources

99

Metsäteollisuus ry (2002). Euroopan paperin keräyksessä iso harppaus, Release 17.10.2002 (in Finnish). More, S. (1997). Process makes ‘wood’ from scrap film and waste paper. Modern Plastics International, October, 41–2. Peltola, P. (2002). Different kind of technology, Proceedings of the 23th Risø International Symposium on Materials Science, Sustainable Natural and Polymeric Composites – Science and Technology. Roskilde, Denmark. Raj, R.G. and Kokta, B.V. (1989). Compounding of cellulose fibres with the polypropylene: effect of fiber treatment on dispersion in the polymer matrix. Journal of Applied Polymer Science, 38, 1987–96. Sameni, J.K., Ahmad, S.H. and Zakaria, S. (2002). Effect of processing parameters and graft-copoly (propylene/maleic anhydride) on mechanical properties thermoplastic natural rubber composites reinforced with wood fibres. Plastics, Rubber and Composites (UK), 31/4, 162–6. Sanadi, A.R., Young, R.A., Clemons, C. and Rowell, R.M. (1994). Recycled newspaper fibres as reinforcing filler in thermoplastics: Part I – Analysis of tensile and impact properties in polypropylene. Journal of Reinforced Plastics and Composites, 13, 54– 66. Vikman, J. (1999). Paper fibre reinforced polyolefins. Master’s Thesis, Tampere University of Technology (in Finnish). www.paperinkerays.fi. www.paperrecovery.org.

6 Alternative solutions: recyclable synthetic fibre–thermoplastic composites R. A. S H A N K S RMIT University, Australia

6.1

Introduction and definitions

The intention in this chapter is to compare various composites from renewable and recyclable resources in the broader sense in terms of the property requirements, need for recycling or biodegradation. The focus is on recyclable composites of thermoplastics, and in particular of polypropylene. The chapter begins with definitions so that the scope for green composites can be discussed. Composite recycling is limited by the need for a thermoplastic matrix and a disperse phase that will be desired in a recycled product, since outside of factory waste recycling, the waste stream for a particular matrix polymer may be contaminated by many types of filler and other additives. A solution is to make the fibres from the same thermoplastic as the matrix, so that the entire mass can be melted to form new products without conserving the original fibres. Processing and characterisation are considered as necessary parts of the technology. A composite is a material consisting of two distinct phases, generally a matrix or continuous phase and a disperse phase. The disperse phase may consist of synthetic materials such as fibres, minerals that can be fibres, but are generally platelets or particles, or natural materials that are generally natural fibres. The matrix phase can be a synthetic or a natural polymer. The matrix can also be classified as thermoplastic or thermoset. If recycling is desired, then a thermoplastic is essential, whereas for biodegradation either can be used, although for similar chemical composition, a thermoplastic will biodegrade faster. A third part of a composite is the interface between disperse phase and matrix. The interface determines the reinforcement efficiency of a composite since strong bonding between matrix and disperse phase is essential. Often dissimilar materials are used in a composite. Minerals and natural fibres are generally hydrophilic, whereas the synthetic polymer matrix will most likely be hydrophobic. Matching the polarity of the matrix and disperse phase will assist interfacial adhesion, but specific interactions or chemical bonding will 100

Alternative solutions

101

be preferable. Often a third component, a coupling agent must be added to modify the surface of the matrix or disperse phase, and typically provide chemical bonding between the phases. For example, a coupling agent will be required for good performance properties in a cellulosic fibre–polypropylene composite. The chemistry of the matrix, disperse phase and any coupling agent additive is important in determining the ability of the composite to be recycled or biodegraded. The design of the original composite is essential in determining its future life cycle. In developing new composites, consideration must be given to whether the components will enable recycling. In this context long fibres, glass fibres and thermosetting polymers will not enable recycling. Any material can be shredded and used as a filler in another material, but the properties will deteriorate. Biodegradable materials do not need to be processed further but the choice of components must include only biodegradable components, so glass fibre cannot be used in a biodegradable matrix. A problem with composites intended for biodegradation is that the time at which the required useful lifetime must end should coincide with the onset of degradation. A continuous degradation mechanism will mean that the application of the composite will be terminated at the time when the properties fall below specification. The degradation will often depend on uncontrollable factors such as humidity, temperature and other environmental variables that will differ for any particular composite in use in a geographic, seasonal or specific locality. Recycling too is constrained by the variability of the composite when it is returned for recycling. Internal factory waste is the only controllable recyclable material.

6.2

Green composites and the structure and function of composites

6.2.1

Minimum environmental impact

Green composites, in the context of discussion in this chapter, are composites with minimum environmental impact. The raw materials may be sourced from renewable resources, but this is not essential. Most synthetic polymers are derived currently from petrochemical resources. Alternative chemistries are often available but are uneconomic under present circumstances. This situation will change with raw material availability, so it can be expected that new processes will be developed that will provide the raw materials most demanded. Many raw materials can be semi-synthetic; that is, they are a combination of natural or renewable materials. Some agriculturally derived raw materials require considerable purification and processing before they become suitable for use in products. The composites may be biodegradable, but biodegradability is dependent

102

Green composites

on conditions of time, temperature and the local chemical conditions. Biodegradability is desired when the composite is finally disposed of but, prior to this, biodegradability will harm the properties of the composite while in use. Biodegradable materials are most suitable for packaging where a single short-term use is prevalent, but not for more permanent structural applications in harsh environments (Ram, 1997). Composites will often fall into the category requiring longer term stability and retention of properties. A significant source of waste materials is factory waste. Products not meeting specification, edge trims, machine start-up materials, and many other materials not suitable for customers, are retained at a factory for disposal. These materials will have an identifiable composition, but be of variable quality. Disposal is expensive and recycling is clearly the best choice because there will be more control over this waste stream than over any other. Other waste streams must be collected from a diverse source of locations and contaminations. The waste must be separated into identifiable streams so that recyclable materials are obtained in a relatively pure form. Even then, the level of contamination, degree of degradation and the ability for efficient processing will be likely to vary considerably. Table 6.1 shows the options for re-use and disposal of waste composites. The table is arranged with the most desirable outcomes in ascending order. Recycling into the same product is the highest category. This is true recycling because the product can approach sustainability if it can be completely recycled. This is never going to be possible since loss in the waste stream and degradation will decrease the yield. Products with critical performance requirements may be too sensitive for the variable properties of recycled materials. The recycled materials can next be downgraded into a less critical product. In each level of Table 6.1 the economic value of the composite decreases. Desirably, the highest economic value will be extracted from the composite over the longest time. Biodegradation is of the lowest value; it will have a negative value due to disposal costs. Application as fertiliser may alleviate some disposal costs.

6.2.2

The structure and function of composites

Composites are materials adapted from nature. The structure of most natural structural materials from trees to bones and teeth are composites. In design and application of composites, there is much to be learnt from nature. Applications consist of structural (support loads), protective (covering, resistance to liquids, heat or fire) and coating. Composites are usually considered a coherent mass containing the disperse phase and matrix. In the broader sense, composites may be coated fabrics, carpets, paper and other structures where the fibre phase provides the main function and protrudes from the matrix. Composites need not contain fibres. The disperse phase

Alternative solutions

103

Table 6.1 Some options for the re-use and disposal of composites Post-consumer application

Composite

Comment

Recycle into the same product.

Thermoplastic matrix, short fibre composite with fibres of any composition

Able to be reprocessed although with some loss of fibre length due to shear.

Recycle into lower performance composite product.

Thermoplastic matrix, short fibre composite with fibres of any composition

Decrease in properties makes a lower performance application necessary, but still valuable as a composite.

Melt and recycle as a thermoplastic.

Thermoplastic matrix and thermoplastic fibres; fibres many be short or long, non-woven or woven

Matrix and fibres can both be melted and used in non-critical thermoplastic applications, if they are compatible and preferably miscible.

Recycle as filler in other polymer compositions.

Thermoplastic or thermoset matrix composite with fibres of any composition

Composite must be chopped or ground into very small particle size, large particles will cause mechanical weakness.

Shred composite and use as fuel.

Thermoplastic or thermoset matrix composite with fibres of any composition

No further value as a composite but heat energy value can be extracted by use as a fuel.

Dispose of composite as waste.

Natural fibre–biopolymer composites or natural fibre composites with other polymers

Natural fibres and biopolymers can biodegrade, natural fibre can degrade and render synthetic polymer more exposed to degradation. Material value is lost.

may be platelets like talc, which provides rigidity but not high tensile strength. Where the disperse phase is particulate-like clay or silica, hardness is increased but not necessarily stiffness. Bio-fillers are similarly diverse in that though most are fibrous, the long fibres of flax and hemp perform differently to the short fibres of rice husks and wood flour. Table 6.2 shows some typical fibres that can be used in composites. The table is arranged from the natural fibres through to synthetic and inorganic fibres, that is, from the most to least biocompatible. The fibres vary in density, modulus and strength. The specific modulus and specific strength are often used to compare properties since these values are based on unit density.

104

Green composites

Table 6.2 Selected fibres for composite applications and their properties Fibre type

Example

Description

Cellulose bast

Flax, hemp, ramie, kenaf, jute

High modulus, high strength, hydrophilic, moisture absorbent, biodegradable

Cellolose hair

Cotton, kapok

Chopped cotton waste and other cellulosic waste can be used as a filler.

Cellulose hard

Coir, sisal, wood flour, bagasse

Short fibres give low strength, more a filler than a reinforcement, able to be recycled, biodegradable.

Biopolymer, semisynthetic

Poly(lactate), poly (hydroxybutyrate)

Biodegradable or recyclable

Thermoplastic synthetic

Polypropylene, polyester, nylon, acrylic, polyethylene

Properties can be preselected, recycled by melting, mainly hydrophobic.

Thermoplastic synthetic bicomponent fibres

Polypropylene core with coAs for other synthetic fibres but polypropylene or polyethylene the sheath provides fibre with sheath. Nylon-6,6 core its own bonding facility. with nylon-6 sheath

High performance synthetic organic fibres

Poly(p -phenylene terephthalamide) (Kevlar), Poly(m-phenylene isophthalamide) (Nomex)

High modulus and strength, generally only used with thermoset polymers such as epoxy

Inorganic

Glass, alumina, boron

High modulus, high strength, brittle, hydrophilic, not biodegradable or recyclable

6.3

Natural material sources: reconstitution of thermoplastic polymers and the effect of water

6.3.1

Reconstituting traditional materials

Reconstitution of natural materials such as wood, stone/ceramic, minerals, chitin and collagen involves processing to enable the materials to be converted into the desired shape, then allowing the molecular structure to form. In the case of wood, the structure can be reconstituted as particleboard, plywood or wood flour (Nunez et al., 2003). Stone can be reconstituted as a ceramic or as a composite with concrete. Cellulose can be derivatised and/or dissolved directly to form new fibres and films. Cellulosic fibres are the most prevalent components of natural composites and they have been used in many semisynthetic composites (Bledzki and Gassan, 1999). Materials generally must be melted or dissolved in order to mould them into the desired shape or product. Waste sugar cane bagasse can be formed into moderately strong panels with good insulation properties by binding with a synthetic polymer

Alternative solutions

105

(Stael et al., 2001). Most natural materials are unable to be processed and must be chemically or physically modified.

6.3.2

Thermoplastic polymers formed by natural systems

Thermoplastic polymers formed by natural systems, in whole or in part, are rare and expensive. Typical examples are poly(hydroxybutyrate) and its copolymers with hydroxyvalerate (Wong et al., 2002a; Hodzic et al., 2002). Lactic acid is synthesised by a fermentation process and then synthetically polymerised into poly(lactate) by a typical step-growth process. These polymers are similar to synthetic thermoplastics, but are readily biodegradable. Many thermosetting compositions have been developed, but these cannot be recycled, so they need to be biodegradable (Wong et al., 2002b; Hodzic and Shanks, 2002). Thermoset materials have been formed from casein and collagen.

6.3.3

Water and humidity

Most natural materials are hydrophilic. Composites made from natural materials will be susceptible to changes in strength, chemistry or dimensions with water. Loss of water through excessive heat will cause friability of cellulosic fibres. Degradation will be a continuing and accelerating process. Desirably, a composite will have an application period during which no change occurs, followed by onset of degradation after the required lifetime. Absorption of water makes the material more susceptible to microbial attack, so toxic additives must be included to protect the material during its desired lifetime. This problem is the factor that makes natural composites less successful than synthetic ones in many applications.

6.4

Synthetic recyclable composites

6.4.1

Synthetic fibre in a thermoplastic matrix

Table 6.3 shows some typical thermoplastics that can be used in fibre composites. Each of the polymers in the Table represents a family of chemically related polymers with different structure and properties. Currently, polyethylene (PE) fibres are available in a PE matrix, all high density PE (HDPE) although other PE types could be used. Similar composites could be formed from poly(ethylene terephthalate) (PET) fibre in a PET matrix, or nylon fibre in a nylon matrix. In each case, the matrix needs to be a form of the polymer that has a lower melting temperature. In the case of ultra-high molar mass HDPE, the melting temperature of gel spun fibres is higher than that of the HDPE because the fibres have a chain-extended structure while the matrix will have a chain-folded structure.

106

Green composites

Table 6.3 Thermoplastic polymers suitable for fibre composite matrix application Polymer

Properties

Comment

Polyethylene

Flexible, hydrophobic, low melting temperature

Low temperature compaction and melting, useful for packaging and protective covers

Polypropylene

Flexible to semi-rigid, moderate melting temperature

Low temperature compaction and melting, useful for protective covers and load support

Nylons

High modulus and strength, high melting temperature, moisture absorptive

Strength is suitable for seating.

Poly(ethylene terephthalate) or related polyester

High modulus and strength, high melting temperature, relatively moisture resistant

Strength is suitable for seating.

In the case of polypropylene (PP), a random PP–PE copolymer can be used with advantage due to its lower melting temperature and its more desirable flexibility and toughness. PP is preferred because of it being common to many automotive components and to other widely available moulded products since such composites can readily be recycled, though the composite nature will be destroyed by recycling. The weave of continuous fibre composites is important in determining properties. Typically, non-woven structures will be used due to lower cost and convenient randomisation of the fibres for sheet-type composites. Additional strength can be obtained with woven fibres. The weave pattern is a determinant of the properties (Jordan et al., 2003). Figure 6.1 shows some typical weave patterns being used in our research. A simple weave will have maximum mechanical properties in the two perpendicular directions of the fibres. In the diagonal and other directions the properties will decrease. Weave patterns that decrease their symmetry in the diagonal directions will maintain strength in these directions as well. Other weave patterns can focus the strength in a single direction. The weaving of the fibres provides an interlocking that increases strength better than can be achieved by fibre matrix adhesion. Failure of the composite will require fibre breakage, since fibre pullout is not possible with tightly woven fibres.

6.4.2

Applications

Table 6.4 shows a classification of fibre composites where the term composite is extended to include thin fibre reinforced films, through to multilayer structural composites, carpets and webbing. Other possibilities include semirigid panelling, flexible sheets and layered constructions where all or some

Alternative solutions

107

6.1 Macro-photographs of woven polypropylene fibre mats, with various weave characteristics, for application in composites.

of the layers may be composites. A simple example is a reinforced film, where a PP woven scrim is compression moulded with a PP film. Multilayered structures increase rigidity and strength making them increasingly useful for structural materials. The theme binding the classification in Table 6.4 is that with suitable choice of materials each of the composites can be completed melted and recycled into a new matrix or backing material.

108

Green composites

Table 6.4 Classification of fibre composites with the same polymer in matrix and fibres Structure

Example

Comment

Single layer of bicomponent non-woven or woven mat

Polypropylene core with coHeat and compression used to polypropylene or polyethylene melt the sheath so as to bond sheath the fibres forming a discontinuous layer.

Single layer of random or non-woven fibres laminated with a film of thermoplastic

Polypropylene fibres with polypropylene film, or polyethylene fibres with polyethylene film

Lower melting temperature film used to bond fibres and form a continuous layer.

Multiple layers of fibres and thermoplastic films

Polypropylene fibres with polypropylene film in alternating layers

Melt bonding of the film layers provides a rigid structural layer or sheet that is thermoformable.

Non-woven or woven mat with thermoplastic powder impregnation

Polypropylene fibre with polypropylene or polyethylene powder scattered between the fibres and mixed by vibration

Melting of powder provides bonding, powder may give better penetration between the fibres than diffusion of a melted film.

Non-woven fibre mat with thermoplastic layer – film or powder on one side

Polypropylene mat with polyethylene or elastomer on one side, usually including much filler

Melting of the thermoplastic on one side provides a backing for the fibres suitable for carpet. The carpet is thermoformable.

6.5

Processing innovations and mineral-filled composites

6.5.1

Extrusion alignment

When fibres are extruded with a thermoplastic, the fibres and polymer molecules are orientated as they pass through the die. Orientation of the fibres is of advantage in providing maximum properties in the direction of orientation. This orientation is a normal consequence of the extrusion process. Extrusion of temperature sensitive composites is difficult due to additional shear heating experienced by the extrudate. Natural fibres may lose water and become brittle, while synthetic fibres may be induced to melt or lose crystalline orientation. Intensive shear can damage the fibres causing fibre breakage that will decrease the aspect ratio that is an important determinant of mechanical properties. Screws and dies with specialised custom design may be required for a particular composite.

6.5.2

Extrusion to provide special textures

The structure and properties of a composite depend on the morphology or texture of the structure. In wood, the oriented fibres are grouped giving a grain effect with varying density and porosity. This contributes to the techniques for the fabrication of wooden structures and their texture. Natural fibre

Alternative solutions

109

composites of wood flour and polypropylene have been extruded to provide the texture of wood using a special die to form fibril-like structures within the composite and at the surface (Laver, 1996). Other wood substitute products have been developed through application of special processing innovations (Deaner et al., 1998; Dahl et al., 2000). These wood substitutes typically use wood flour filler with a synthetic polymer matrix. Wood is a renewable resource, though with a very long renewal time, so substitution with a semisynthetic product is a positive conservation approach.

6.5.3

Pultrusion

Continuous fibres are pulled through liquid matrix polymer. Usually, but not necessarily, the matrix polymer is a thermosetting liquid resin. A polymer melt can also be used. The process for a thermoplastic is similar to coating wires with an insulation of polymer, except that bundles of fibres are used instead of a single wire. The process is suitable for two-dimensional profiles such as rods, beams and cladding boards. Pultrusion integrates compaction of orientated continuous fibres and matrix within the die. These composites will have maximum strength in the longitudinal direction. The surface can be embossed with patterns or made highly glossy by the design and pressure within the die.

6.5.4

Felting or needle punch

Woven fibres are expected to provide the best strength properties but they are more expensive than non-woven. A felted mat of fibres can be formed by several methods, of which needle punching is common. The non-woven mat is a tightly entangled web of fibres that lacks strength unless its integrity is increased with an adhesive. Entangled combinations of fibres can be formed by weaving or felting where one of the fibre types can melt to form the matrix while the other fibre remains as the reinforcement. An example is a non-woven mat of natural (typically flax in current commercial products) and polypropylene fibres. This process is suitable where all of the fibres are of the same polymer type but with some having lower melting temperature or bicomponent structure. Compression moulding of the mat will melt the polypropylene fibres and compact the composite to form a continuous sheet. Partial compaction can provide low-density panels. The panels can be thermoformed into self-supporting shapes as required (Goto and Kasahara, 1988; Beer and Brosch, 2000). The process can be used with mixed recycled fibres (Fig. 6.2). The fibres must first undergo carding (Fig. 6.3) to provide alignment of the fibres and increase the fibre mixing. The needle punch machine (Fig. 6.4) then provides intensive mixing and entanglement across the fibre web. The needles provide

110

Green composites

6.2 Recycled and new PP fibres entering a carding machine after being mixed and conveyed in an air stream.

6.3 PP fibre alignment in a carding machine.

a more even structure if they protrude from both above and below the fibre web. The shape and barbs on the needles and the pattern of needles in the holding panel determine the structure of the finished non-woven fabric. The

Alternative solutions

111

(b) (a)

(c)

6.4 Needle punch machine for manufacture of non-woven textiles; the fibre web passes between two punch units to mix fibres from above and below. (a) shows a freeze-frame of the high-speed motion, (b) shows the punch head with needles and (c) shows the double upper and lower needle punch arrangement.

type and diameter of fibre are also important in the choice of needle design and pattern. After needle punching the highly entangled fabric web (Fig. 6.5) is ready for bonding by fusion or coating with an emulsion polymer. The various fusion processes are described in the next sections.

6.5.5

Compression moulding and calendering

Impregnation of woven or non-woven fibres with polymer powder or film is followed by compression moulding to allow the polymer to diffuse into a fibre mat and provide continuous wetting. Fibre–fibre mats are compression moulded to form fibre–matrix composites, where the lower melting temperature fibres form the matrix. The composites for compression moulding can be of

112

Green composites

6.5 Non-woven fabric web after needle punching ready for bonding by fusion or coated adhesive.

any form (Table 6.4) where the matrix is a thermoplastic. Typically a polyolefin matrix is used (Blackmore and Spanton, 2000). The composite may be a carpet where the carpet can form a self-supporting shape, thereby providing comfort, function, appearance and structure. The composite usually consists of a woven or non-woven mat of the reinforcing fibres. The matrix polymer may be introduced as fibres during the mat-forming process as described above. This gives intimate mixing; the thin matrix polymer fibres melt rapidly and the composite is formed in one mixing process. Powdered matrix polymer may be scattered onto the mat and interpenetration assisted by vibration. This is useful for carpets since a penetration gradient from the base side can be obtained. Emulsion or plastisol dispersions can alternatively be applied by roller or blade coating for carpets. Lamination with films of the matrix polymer can be used on either or both surfaces and in intermediate layers. Lamination requires polymers that, when melted, can flow sufficiently easily and rapidly to interpenetrate the fibres. In these systems, the thermal binding treatment is provided by heated rollers. Recyclability requires that the components are miscible and that the total system can form a thermoplastic blend with useful properties for a second generation of products. Preferably, the recycled composition should be suitable as a matrix for the next generation of the same product. This is only likely to be achieved when both polymers are the same. Figure 6.6 shows a fusion bonded PP–polyester composite. The PP is concentrated on one side so that bonding will occur preferentially on that side. The side with no bonding retains a carpet-like or fibrous texture. The mat is thermoformable due to the lower melting temperature of the PP matrix. The mat is able to retain its shape and be self-supporting once moulded. The mat is suitable for lining the luggage compartment of a car.

Alternative solutions

(a)

113

(b)

6.6 PP–polyester melt bonded non-woven mat; (a) non-bonded surface, (b) bonded surface.

6.5.6

Point bonding of fibre mats

For many applications, partially bonded composites will provide suitable properties. The fibres can be bicomponent or made from a single composition. Bicomponent fibres are bonded by melting the sheath while retaining the orientation and strength of the core. The fibres generally only make point contact with the diameters of adjacent fibres so that bonding is not extensive, but it will be sufficient for many applications. Webbing for straps, automotive seat belts and seating are examples where biocomponent fibre bonding is applied. The strength is provided almost entirely by the fibres, while the bonding is used to keep the fibre mass intact. The materials look so similar to textiles that they may be considered not to be composites. They are twophase materials where the core and sheath are distinct phases, one phase providing strength, while the other provides bonding. Single component fibres can also provide composite-like properties. Here there is only one phase, but partial melting provides bonded regions while the interiors of fibres maintain their orientation and high strength properties. An example of a PP reusable-recyclable supermarket shopping bag is shown in Fig. 6.7. The fibre structure is shown in Fig. 6.7(b). The fibres are formed into a non-woven mat, which is then fusion bonded in an array of points giving a weave type appearance. The fused points are the lighter areas. The unmelted fibres provide strength for the bag in orthogonal directions and the fused areas maintain the integrity. Since the bag is made of only PP, it can easily be recycled, but its quality is such that it can be used many times before recycling is necessary. Hot compaction of PE and PP woven fibre composites has been performed on single component systems (Jordan et al., 2003); Hine et al., 1998, 2003). In these systems the melting is extensive, though the interior of the fibres

114

Green composites

(a)

(b)

6.7 A re-usable and recyclable shopping bag (a) made of bonded polypropylene fibres and (b) the fibre mesh and fusion pattern.

must be retained to retain orientation and hence fibre-like properties. The composites were formed from multilayers of PP fabrics so that interlayer adhesion was a concern.

6.5.7

Mineral-filled composites

Undoubtedly, fibre composites have the highest performance because of the high aspect ratio of the fibres. Platelet-shaped minerals such as talc can also provide strength and stiffness to polymers. Talc is often in polypropylene to enhance its properties in specific applications where elevated temperatures are experienced and in automotive components to provide stiffness (Long et al., 1995). Mica also has platelets with a high aspect ratio and has proved suitable for reinforcement (Shucai et al., 1996). These mineral-filled composites are easier to mix and process than fibre-filled composites. They are suitable for recycling and this is currently done with automotive bumper bar and other components. The minerals are natural materials that will cause no problem on disposal if the matrix polymer can be biodegraded. In this regard they are preferable to glass fibres, but they cannot provide the same level of reinforcement as glass fibres.

6.6

Properties of single polymer fibre–matrix composites

Strong interfacial adhesion is expected in composites with the same polymer in both fibre and matrix. This type of composite is entirely recyclable and the

Alternative solutions

115

6.8 Optical microscopy of a PP fibre in coPP matrix showing transcrystallinity emanating from the fibre (¥ 200).

ultimate green composite. The fibres and matrix have the same crystalline form and the fibre crystals can act as nuclei for the lower temperature crystallising matrix. This is apparent as transcrystallinity in the matrix near fibres (Vaisman et al., 2003). An example of transcrystallinity is shown in Fig. 6.8 where crystals from coPP grow within the matrix perpendicular to the PP fibre axis. This type of morphology increases the fibre–matrix interfacial strength. In addition, it is expected that some of the fibre PP will melt or dissolve in the molten matrix, near the fibre surface. This will allow an interdiffusion at the interface increasing the interfacial strength. Similar phenomena are expected with other composites of like polymers. The interface of cellulosic fibres with thermoplastics cannot provide such adhesion. Figure 6.9 shows differential scanning calorimetry (DSC) curves for a PP–coPP composite. The top curve is melting and the lower curve is crystallisation. The melting curve for the composite shows a small lower temperature peak representing melting of the coPP matrix. The higher temperature peak represents melting of the fibres, compared with the curve for the fibres only. The peaks due to the matrix and fibres in the composite overlap so there will be some fibre melting before the matrix is completely melted. The matrix will be completely melted since superheating cannot occur, though the composite needs to be heated uniformly which is difficult because of its low thermal conductivity. After completely melting both fibre and matrix, the material was cooled to provide the lower crystallisation

116

Green composites

Specific heat (J/g K)

20 16 12 8 4 0 30

60

90 120 Temperature (∞C)

150

180

150

180

(a)

Specific heat (J/g K)

25 20 15 10 5 0 30

60

90 120 Temperature (∞C) (b)

6.9 DSC curves showing (a) separate melting and (b) crystallisation of matrix and fibres in PP–coPP composite.

curve. The fibre crystallised at higher temperatures, then the matrix crystallised as shown by the two peaks. The two peaks are coincident with the crystallisation temperatures of the pure matrix and pure fibres shown on the same lower graph. The matrix and fibres are miscible and if they are held in the melt for about 1 hour the two peaks merge and one crystallisation peak is observed. In the example shown in Fig. 6.9, the material was held in the melt for 1 minute before cooling. There was insufficient interdiffusion for the fibres and matrix to mix completely so two peaks characteristic of the original components were observed. On recycling, intensive mixing would be provided by an extruder and the result would be a homogeneous melt. Figure 6.10 shows tensile stress–strain curves for low strain. The modulus, or slope of the curves demonstrate that the modulus of a composite with 50% fibres has been enhanced compared with that of the matrix polymer. Dynamic

Alternative solutions

117

16

Stress (MPa)

12

8

4

0 0

0.5

1 Strain (%)

1.5

2

6.10 Stress–strain curves of tensile properties of polypropylene matrix– fibre composites compared with co-polypropylene.

mechanical analysis was performed to measure the storage and loss modulus with change in temperature as shown in Figs. 6.11 and 6.12. Figure 6.11 shows the storage modulus of a composite with 50% of fibres. The storage 5

Storage modulus (GPa)

4

Composite PP

3

2

1

0 –40

–20

0

20 Temperature (∞C)

40

60

80

6.11 Storage modulus of coPP matrix and PP fibre–matrix composite showing decrease in stiffness over the glass transition and crystal rearrangement regions.

118

Green composites 450 400

Loss modulus (MPa)

350 300 250 200

PP

150 100 50 0 –45

Composite –25

–5

15 35 Temperature (∞C)

55

75

6.12 Loss modulus of coPP matrix and PP fibre–matrix composite showing damping over the glass transition and crystal rearrangement regions.

modulus is enhanced most in the normal operating temperature range of 0– 60 ∞C. At lower temperatures, where both the matrix and fibres are below their glass transition temperature, the enhancement is reduced. The difference is also less apparent above 100 ∞C where the melting temperature of the matrix (145 ∞C) is approached. The loss modulus curve (Fig. 6.12) shows a maximum at about 0 ∞C due to the glass transition of PP. This maximum is placed conveniently since the composite will have maximum damping or energy absorbing characteristics over this range, thereby increasing the toughness. On the high temperature side of the main peak in the loss modulus is a smaller peak at about 35 ∞C. This peak is due to crystal–crystal rearrangement of the PP. Both peaks in the curves are due to both the fibres and the matrix. These synergistic properties are not provided by natural fibre composites. Modulus and strength are not the only important mechanical properties of a composite. An often more important mechanical property is creep. If a film or sheet of polypropylene is held under a constant force (stress) it will gradually extend (strain). Creep is defined as the strain in a material when under constant stress. Creep may occur over several years and will be undesirable for any load or pressure-bearing product. Figure 6.13 shows the results of a creep test where a stress was immediately applied for a discrete time, and then the stress was immediately removed. The change in strain of the specimen was measured during and after the application of stress. It can be seen that the composite is more resistant to creep than a similar sheet of

Alternative solutions

119

Relative creep (%)

0.8

0.6

coPP

0.4

0.2

0

Composite

0

5

10 Time (min)

15

20

6.13 Creep recovery curves for coPP and PP fibre–matrix composite showing a reduction in creep for the composite.

the matrix coPP. The orientated PP fibres resist creep, although when the stress is first applied they undergo a small elastic response. The viscoelastic approach to equilibrium of both materials is similar. When the stress was removed, the recovery of the composite was less since again the fibres are less able to flow due to high orientation of the crystals. The polypropylene fibres give a composite with good elasticity, whereas natural fibres such as flax will give a more creep resistant composite. The mechanical properties of composites with differing composition and fibre geometries have been modelled using finite element analysis (FEA) (Zako et al., 2003). FEA has been found to be useful in predicting the properties of composites of many types. In particular, FEA is a useful modelling tool for the design of thermoplastic fibre composites such as have been described in this chapter. The many variables, such as fibre packing, fibre orientation and the type of weave to be chosen for a composite, create a large array of variables. The FEA simulation tool can predict the expected properties and it can be used to select an optimum composite structure for a particular application. The range of properties of these recyclable composites is extensive, more so than of natural fibre composites, though their ultimate properties are lower than those of glass fibre composites.

6.7

Future trends

A trend in the automotive industry is to use standardised materials. PP can be used in fibre composites, mineral-filled composites, and toughened with elastomers, so it can be used for many components in automobiles. Standardisation of PP facilitates separation and recycling. The recycled material will often be a combination of PP with several other fillers, but it can still be

120

Green composites

processed under the same conditions as any other PP, and the PP of each matrix will be completely miscible. PP is also used in many other consumer appliances. PP has been chosen for these applications because it has a high enough melting temperature to resist most expected temperatures, particularly boiling water. Polyethylenes of all types melt at temperatures too low for continuous resistance to boiling water. PP can be enhanced by many fillers from fibres to minerals and toughened with elastomers. Its versatility has made it the choice as a standardised material. Biopolymer thermoplastics are currently too expensive for most applications. Polylactate is a biopolymer with significant commercial production expansions, driving availability and cost. It can form both fibres and matrix components. Similarly, other biopolymers such as poly(hydroxybutyrate) and its copolymers will become more widely used when production can provide price reduction. Each of these biopolymers needs to find speciality applications so that production of them can increase.

Sources of further information and advice Many new composites are being developed at an exponentially increasing rate. A search of Chemical Abstracts revealed the exponential increase in publications and patents relating to thermoplastics and natural fibres. Patents provide information on many of these new composites as the primary literature. New processing machinery has opened up possibilities for natural materials and new structures. Many patents and supplier information show the possibilities of innovative processing equipment forming new composites. Many of the most useful composite structures can only be obtained with the appropriate processing equipment, so there is an expensive entry point to the technology. Waste disposal has become a concern for municipalities everywhere. Waste materials are too expensive to dispose of and so they are being separated for recycling. Waste material sources are expanding, offering new opportunities for creation of new products. There are many authorities, government reports and offices that co-ordinate the application of waste materials. These groups are available to provide advice and to facilitate the collaboration of groups with complementary requirements or technologies.

Acknowledgements Figures for thermal and mechanical properties of PP and its composites were measured by Shadi Houshyar. Macro-digital photography was provided by Dr Antonietta Genovese.

Alternative solutions

121

References Beer, B. and Brosch, J. (2000). Production of interior paneling for motor vehicles from natural fibers. Ger. Offen. De: Ibs Brocke G.m.b.H. & Co. K.-G., Germany, 4 pp. Blackmore, P.W. and Spanton, D.L. (2000). Composite fabrics for roofings and carpet backings comprising a layer of nonwoven webs sandwiched between a layer of warp strands with low heat shrinkage and a layer of filling strands with low heat shrinkage and reinforced products therefrom. PCT Int. Appl. Wo: Bay Mills Ltd., Can.; Certainteed Corporation, 18 pp. Bledzki, A.K. and Gassan, J. (1999). Composites reinforced with cellulose based fibres. Prog. Polym. Sci., 24, 221–74. Dahl, M.E., Rottinghaus, R.G. and Stephens, A.H. (2000). Extruded wood polymer composite and method of manufacture. US Patent 6 153 293. Deaner, M.J., Puppin, G. and Heikkila, K.E. (1998). Advanced polymer wood composite. Anderson Corporation, US Patent 5 827 607. Goto, F. and Kasahara, Y. (1998). Thermoplastic composites reinforced with hemp fibers. PCT Int. Appl. Wo: Namba Press Works Co., Ltd., Japan, 15 pp. Hine, P.J., Ward, I.M. and Teckoe, J. (1998). The hot compaction of woven polypropylene tapes, J. Mater. Sci., 33, 2725–33. Hine, P.J., Ward, I.M., Jordan, N.D., Olley, R.H. and Bassett, D.C. (2003). The hot compaction behaviour of woven oriented polypropylene fibres and tapes. I. Mechanical properties. Polymer, 44, 1117–31. Hodzic, A. and Shanks, R.A. (2002). Thermoplastic and biopolyester flax fibre composites. Composite Systems: Macrocomposites, Microcomposites, Nanocomposites, Proceedings of ACUN-4, International Composites Conference, 4th, Sydney, Australia, July 21– 25, 351–6. Hodzic, A., Shanks, R.A. and Leorke, M. (2002). Polypropylene and aliphatic polyester flax fibre composites. Polym. Polym. Compos., 10, 281–90. Jordan, N.D., Bassett, D.C., Olley, R.H., Hine, P.J. and Ward, I.M. (2003). The hot compaction behaviour of woven oriented polypropylene fibres and tapes. II. Morphology of cloths before and after compaction. Polymer, 44, 1133–43. Laver, T.C. (1996). Extruded synthetic wood composition and method for making same, Strandex Corporation, US Patent 5 516 472. Long, Y., Tiganis, B.E. and Shanks, R.A. (1995). Evaluation of recycled PP/rubber/talc hybrids. J. Appl. Polym. Sci., 58, 527–35. Nunez, A.J., Sturm, P.C., Kenny, J.M., Aranguren, M.I., Marcovich, N.E. and Reboredo, M.M. (2003). Mechanical characterisation of polypropylene–wood flour composites. J. Appl. Polym. Sci., 88, 1420–28. Ram, A. (1997). Fundamentals of Polymer Engineering. New York: Plenum Press, Chapter 7, pp. 216–29. Shucai, L., Jarvela, P.K. and Jarvela, P.A. (1996). Properties of polypropylene copolymer– mica composites. Plast., Rubber Compos. Process. Appl., 25, 441–7. Stael, G.C., Tavares, M.I.B. and d’Almeida, J.R.M. (2001). Impact behaviour of sugarcane bagasse waste–EVA composites, Polym. Test., 20, 869–72. Vaisman, L, Gonzalez, M.F. and Marom, G. (2003). Transcrystallinity in brominated UHMWPE fiber reinforced HDPE composites: morphology and dielectric properties. Polymer, 44, 1229–35. Wong, S., Hodzic, A. and Shanks, R.A. (2002a). Morphology and mechanical properties of modified natural fibre biopolyester composites. Composite Systems: Macrocomposites,

122

Green composites

Microcomposites, Nanocomposites. Proceedings of the ACUN-4, International Composites Conference, 4th, Sydney, Australia, July 21–25, 357–62. Wong, S., Shanks, R.A. and Hodzic, A. (2002b). Properties of poly(3-hydroxybutyric acid) composites with flax fibres modified by plasticiser absorption. Macromol. Mater. Eng., 287, 647–55. Zako, M., Uetsuji, Y. and Kurashiki, T. (2003). Finite element analysis of damaged woven fabric composite materials, Compos. Sci. Technol., 63, 507–16.

7 Natural polymer sources D. P L A C K E T T A N D A. V Á Z Q U E Z * Risø National Laboratory, Denmark *Universidad Nacional de Mar del Plata, Argentina

7.1

Introduction: biocomposites and biodegradable polymers

This chapter is focused on polymers derived from natural sources and their use in composites reinforced with natural fibres. For the purposes of this chapter these materials are referred to as biocomposites, a term frequently used in the literature (Mohanty et al., 2000a, 2002), but also now commonly extended to composites in which biopolymers include inorganic fillers (e.g. nanoclays). Naturally derived polymers are usually biodegradable, and therefore biocomposites reinforced with natural fibres should also be fully biodegradable. In addition to this characteristic, biocomposites containing natural fibres generally exhibit enhanced mechanical properties and, because of the partial replacement of the polymer matrix with less costly reinforcing material, are less expensive than the starting polymer in overall material costs. Biocomposites have been the subject of international research since at least the mid-1990s and a number of practical applications are now emerging, including interior automotive components and housings for notebook computers. Commercial interest in manufacturing these products is driven by the derivation of the polymers from renewable sources as well as by their specific properties including biodegradability. The chapter is divided into two main parts; the first begins with an initial discussion on biodegradability and then reviews selected natural polymers and their key characteristics. The polymers included in this section are those that are commercially available and have been the focus of significant past research activity aimed at their use in biocomposites. Literature on natural fibre biocomposites is reviewed in the second part of the chapter and biopolymer nanocomposites are also briefly discussed. The chapter concludes with sections on possible future trends in biocomposites and recommended further reading.

123

124

7.1.1

Green composites

Biodegradable polymers

Biodegradation has been defined as ‘an event that takes place through the action of enzymes or through chemical decomposition associated with living organisms’ (e.g. bacteria, fungi) (Albertsson and Karlsson, 1994). Another proposed definition is ‘the gradual breakdown of material mediated by specific biological activity’ (Ali et al., 1994). Abiotic reactions involving photodegradation, oxidation and hydrolysis may also occur before, during or instead of biodegradation. Biodegradable polymers are classed as biosynthetic, semi-biosynthetic or chemosynthetic depending on their manner of preparation. Steinbüchel (1995) studied the use of biosynthetic, biodegradable thermoplastics and elastomers from renewable resources and Mohanty et al. (2000b) prepared a recent summary in the context of biocomposite applications. At present, the most important biosynthetic polymers from an overall market perspective are the polylactides, polyhydroxyalkanoates (PHAs), starch and starch blends and cellulose derivatives. Synthetically derived biodegradable polymers include polyesters (e.g. polycaprolactone), copolyesters, modified polyethylene terephthalates and polyvinyl alcohol. Biopolymers that have received particular attention for use in biocomposites include polylactides, PHAs, starch and starch blends and a summary of the present commercially available polymers of these types is presented in Table 7.1.

7.2

Polylactides: polylactic acid (PLA) synthesis, properties, biodegradation, processing and applications

7.2.1

Background

Polylactide or poly(lactic acid), otherwise known as PLA, is a biodegradable thermoplastic polyester that is manufactured by biotechnological processes from renewable resources (e.g. corn). Although other sources of biomass can be used, corn has the advantage of providing the required high-purity lactic acid. The use of alternative starting materials (e.g. woody biomass) is being pursued in order to reduce process costs; however, the number of steps involved in deriving pure lactic acid from such raw materials means that their use remains much less cost effective at present.

7.2.2

PLA synthesis

PLA can be synthesised from lactic acid by two main routes, a direct polycondensation reaction or ring-opening polymerisation of a lactide monomer in the presence of a Lewis acid catalyst (Fig. 7.1). Cargill Dow LLC, the largest current producer of PLA, has developed and patented a low-cost

Natural polymer sources

125

Table 7.1 Summary of commercially available polylactides, polyhydroxyalkanoates and starches or starch blends Polymer type

Manufacturer

Polylactides

Biomer Birmingham Polymers, Inc. Boehringer Ingelheim Cargill-Dow LLC Galactic SA Hycail Mitsubishi Plastics, Inc. PURAC Shimadzu Corporation

Polyhydroxyalkanoates

Starch

Biomatera Inc. Biomer Metabolix, Inc. Procter & Gamble PHB Industrial S/A Avebe BioPlastic (Michigan) BIOTEC GmbH Earth Shell Groen Granulaat Hayashibara Biochemical Labs Midwest Grain Products National Starch Novamont Rodenburg Biopolymers Starch Tech Supol Vegemat

Product name

Lactel, Absorbable Resomer® NatureWorks™ Galactic Ecoloju Purasorb® Lacty

PHA, Biopol® Nodax™ Paragon Envar Bioplast®, Bioflex®, Biopur® Starch-based composite Ecoplast Pullulan Polytriticum® 2000 Eco-Foam® Mater-Bi Solanyl® RenEW,ST1,ST2,ST3 Supol Vegemat®

Sources include: Mohanty et al. (2000b), Johnson et al. (2003a), www.biopolymer.net.

continuous process for production of lactic acid-based polymers (Gruber and O’Brien, 2002). The ring-opening polymerisation route has the advantage that polymers of high molecular weight can be obtained. In practice, the molecular weight distribution of the final product is determined by factors such as catalyst concentration, reaction temperature and reaction time (Sødergaard and Stolt, 2002). In the industrial manufacturing of PLA, starch from the starting raw material is converted to dextrose, which is then fermented to produce lactic acid. The lactide monomer is prepared from lactic acid by a condensation process. A key feature of the fermentation step in the process is the need for suitable nutrients such as soluble proteins, phosphates and ammonium salts. As with lactic acid, lactide monomers have optically active D- and Lforms and consequently polymers of different optical activity can be obtained

126

Green composites H2O

O

O O

HO OH H L-lactic

H CH3

CH3

L-PLA

acid O

H2O

H O

CH3 O

H3C

L-lactide

H O

7.1 Polymerisation of L-lactic acid to L-PLA by direct condensation or by ring opening via the L-lactide.

(e.g. isotactic poly (D-) and (L-) lactides, a syndiotactically alternating D-, Lcopolymer and statistical copolymers with L- and D-units). Enantiomerically pure PLA is semi-crystalline while polymers prepared from meso- or racemiclactide are usually amorphous, although crystalline polymers have been obtained through the use of stereoselective catalysts.

7.2.3

Properties of PLA

PLA is commercially interesting because of its good strength properties, film transparency, biodegradability, biocompatibility and availability from renewable resources. In addition to property adjustments through variations in polymer morphology and crystallinity, copolymers of PLA can range from glass-like through to rubbery materials with a correspondingly large variation in mechanical properties. In general, these properties of PLA (e.g. tensile strength) are highly dependent on molecular weight. Studies show that lactide/ glycolide polymers and copolymers can retain their stability under storage conditions for at least a year. However, in distilled water at ambient temperature the mechanical properties decrease within a few months. Polymer strength properties also decrease over a period of 3 to 6 weeks in vivo depending on crystallinity and molecular weight (Lipinsky and Sinclair, 1986). The degree of crystallinity of polylactides depends to some extent on previous thermal history. For example, moulded L-PLA articles initially retain their amorphous character during extrusion or injection moulding but crystallise during tempering. The brittle nature of PLA can be modified by the use of plasticisers

Natural polymer sources

127

or by forming copolymers with more flexible biodegradable polymers (e.g. polycaprolactone).

7.2.4

Biodegradation of PLA

PLA and its copolymers degrade to non-toxic breakdown products under certain minimum conditions of temperature and moisture content and degradation occurs initially by a non-enzymatic hydrolytic process. Since PLA and its copolymers are soluble in water only at very low molecular weights, the molecular weight may decrease rapidly after water penetration but the mass as well as the shape of a PLA article can be preserved until extensive degradation has taken place. In general, the decomposition rate depends not only on the characteristics of the polymer at the molecular level but also on bulk factors such as surface properties, porosity and the presence of process additives.

7.2.5

Processing of PLA

Polymers based on lactic acid can be handled by conventional thermoplastic processing techniques; however, care has to be taken to avoid chemical, thermal and mechanically induced degradation during processing. Processing temperatures much in excess of 200 ∞C are generally to be avoided. The presence of residual monomer in processing equipment can present problems; however, such difficulties can be overcome by cleaning equipment during long breaks in processing. In practice, additives are frequently used; for example, nucleation agents are added in order to injection mould parts with acceptable cycle times. Other additives such as plasticisers, lubricants, impact modifiers and pigments may also be employed and some research has been devoted to the selection and use of biodegradable additives (Ljungberg and Wesslén, 2002).

7.2.6

Applications

Recent developments in PLA technology have brought costs down to a range in which competition with more expensive commodity thermoplastics such as polyethylene terephthalate (PET) is being targeted (e.g. in packaging). The relatively high cost of PLA arises from a number of factors, especially the need for very pure lactic acid as a starting material. However, as with all new polymers, initially low production capacity and consumer demand also influence the economics. Supermarket chains in Europe are now becoming more interested in food biopackaging based on PLA and a recent study in Europe has highlighted consumer support for such developments (Lichtl, 2003). PLA is of interest as a packaging film for certain foodstuffs because

128

Green composites

of its transparency and its barrier properties. Research on modifications of PLA to allow its use in a wider range of food packaging has been pursued through activities such as the European Union 5th Framework BIOPACK research project (Haugaard, 2003). Notwithstanding recent developments, the price of PLA has tended to favour high-value applications such as in medicine. The properties of PLA are suitable for medical applications and biocompatible PLA composites can now replace steel nails, plates and other objects as surgical implants (Ikada and Tsuji, 2002). In addition, since the bioresorbability of a PLA composite can be controlled, it is possible, for example, to match this with the rate at which a fractured bone repairs. A further attraction is that secondary operations to replace the implant are generally unnecessary when bioresorbable PLA-based materials are used. PLA has also long been used in medical sutures. Cargill Dow LLC presently manufactures an estimated 95% of the world’s production of PLA. There are other manufacturers based in the USA and Japan. In Europe, Purac (the Netherlands), a major supplier of lactic acid and lactide, produces some PLA for medical applications and Galactic (Belgium) also manufactures lactic acid, lactic acid esters and PLA. Hycail b.v. can produce PLA at a factory in the Netherlands.

7.3

Polyhydroxyalkanoates: polyhydroxyalkanoate (PHA) synthesis, properties, biodegradation, processing and applications

7.3.1

Background

The simplest of the family of polyhydroxyalkanoate (PHA) biopolymers is poly-R-3-hydroxybutyrate or PHB. This polymer was first discovered in 1925 by Lemoigne and was initially described as a lipid inclusion in the bacterium Bacillus megaterium (Gilmore et al., 1990). Later research has demonstrated that PHB is a high molecular weight polymer used in carbon and energy storage by a variety of microorganisms (Fig. 7.2) (Luzier, 1992). PHAs (Fig. 7.3) have received much research attention in recent years with a large number of publications ranging in topic from biosynthesis, microstructure, thermal and mechanical properties through to studies on biodegradation (Hankermeyer and Tjeerdema, 1999). Much of the research has been driven by the availability of PHAs from renewable resources and the similarity of PHA physical properties to those of conventional plastics (Evans and Sikdar, 1990; Williams and Peoples, 1996).

Natural polymer sources

129

7.2 Photomicrograph showing growth of polyhydroxyalkanoate (PHA) particles (white) inside cells of Ralstonia eutropha (source: Professor Jian Yu, Hawaii Natural Energy Institute, University of Hawaii at Manoa). R

O

H O

x

OH n

7.3 Generic structure of polyhydroxyalkanoates (PHAs) where R = hydrogen or hydrocarbon chain and x can range from 1 to 3 or more.

7.3.2

Synthesis of PHAs

As an example of PHA synthesis, Alcaligenes eutrophus has been used to produce the commercial product Biopol“, which is a polyhydroxybutyrateco-valerate (P(3HB-co-3HV)) originally developed by Imperial Chemical Industries (ICI) using a process based on extraction of the polymer from whole bacterial cells (Asrar and Gruys, 2002). The technology for producing Biopol“ is now owned by Metabolix Inc. The conditions under which PHAs are produced vary according to the selected microorganism. The necessary enzymes, together with coenzymes, that catalyse synthesis are produced under nutrient-limited conditions in which the carbon source is in excess. A wide range of microorganisms can accumulate PHB but genetic manipulation is usually required in order to generate the biopolymers in suitable dry mass amounts for large-scale production. A large number of PHBV random copolymers can be produced from A. eutrophus depending upon the carbon substrate. Examples of carbon sources have included propionic acid, pentanoic acid, 4-hydroxybutyric acid,

130

Green composites

1,4-butanediol and a number of other diols and substituted propionic acids. Following the fermentation process, the dilute aqueous broth is extracted to obtain biopolymers that must then be isolated and purified. PHB of very high purity can be produced by continuous fermentation in combination with advanced isolation procedures. Future developments in PHA production could include the use of new bacterial substrates as well as ways to achieve better quality control. The use of alternative energy sources for PHA production (e.g. food wastes) is also an area of considerable potential (Yu, 2001; Yu et al., 1999). The production of PHB in transgenic plants has often been proposed as a promising way of manufacturing PHB in large quantities at low cost and a recent grant has been made to Metabolix Inc. by the US Department of Agriculture for a 3year project aimed at combining advanced plant gene expression technology and high-throughput metabolic profiling to develop robust plants capable of producing PHAs. However, because of the wide range of available PHAproducing bacteria plus the ability to manipulate the process through adjustment of feeding substrate and the time required to genetically modify plant metabolism, it may still be some time before plant-derived PHAs will compete commercially with PHAs derived from bacteria.

7.3.3

Properties of PHAs

The properties of PHB and PHBV are sometimes compared with those of polypropylene (Evans and Sikdar, 1990) and some researchers have suggested that the properties of these polymers are similar to those of high molecular weight polypropylene because of their similar morphologies. There are numerous references to studies on the mechanical properties of PHB-based polymers, especially P(3HB-co-3HV) (Strasser and Owen, 1991; Marchessault and Bluhm, 1988). For instance, the tensile strength of copolymers containing up to 28 mol % HV cast from chloroform was measured (Mitomo et al., 1988). The tensile strength at 25 ∞C was reported to be about 45 MPa, which decreased with increasing 3HV content up to 20 mol %. Elongation at break increased from 3 to 27% as the 3HV content increased from 0 to 20 mol %, whereas the sample with 28 mol % HV reached 750% elongation but showed reduced tensile strength. PHB is a brittle polymer with a white colour, while the copolymer with HV becomes more flexible and transparent with increasing HV content. This behaviour is thought to be due to the growing spherulite size with increasing 3HV content. The addition of nucleating agents such as boron nitride is reported to improve the mechanical properties of PHB by reducing the average spherulite size (Marchessault and Bluhm, 1988). Melt-processed copolyesters based on PHB, without added nucleating agents, can be flexible and transparent when cooled at low temperatures (e.g. 5 ∞C) but these polymers stiffen and turn opaque at room temperature.

Natural polymer sources

7.3.4

131

Biodegradation of PHAs

Various researchers have examined the biodegradation of PHAs in terms of enzymology as well as polymer composition. Certain bacteria can utilise PHB as their sole carbon source and these same organisms can metabolise PHB when nutrient-limited conditions are removed. However, the ability to store PHB does not necessarily mean that the same bacteria can degrade this polymer in the environment. Enzymes capable of depolymerising PHB and oligomers derived from this polymer have been found in a few organisms but much of the past research has focused on depolymerase produced by Alcaligenes faecalis. PHAs are not affected by moisture alone and are indefinitely stable in air (Luzier, 1992). The biodegradation rate is influenced by polymer surface area, microbial activity in the disposal environment, pH, temperature, moisture level and nutrient supply. Since amorphous PHAs are hydrolysed more easily than crystalline PHAs, the rate of polymer erosion quickly increases with decreasing film crystallinity (Doi et al., 1992).

7.3.5

Processing of PHAs

PHB is susceptible to thermal degradation at temperatures not far above its melting point and, as with PLA, satisfactory processing requires careful control of processing temperatures. Care has to be taken to clean equipment with low-density polyethylene (LDPE) during breaks in processing, even if stopping for only a few minutes, so as to avoid thermal degradation. Nucleating agents are commonly used as process additives in order to optimise crystallisation rate. The addition of nucleating agents can lower cycle times and may also increase the rate of biodegradation. Both PHB and PHBV exhibit a high dependence of shear viscosity on temperature, and penetration of mould gaps as small as 0.01 mm is possible without good quality control during processing.

7.3.6

Applications

As indicated earlier, one of the main commercial developments to date in PHA technology has been the production of PHBV in the form of Biopol“. Some of the very first products from this polymer were shampoo bottles and cosmetic containers. Future applications for PHB-based polymers could be in disposable products such as diapers, garbage bags and fast-food utensils. Hocking and Marchessault (1994) reviewed the use of PHB or PHBV in a variety of products such as films, bottles and containers. Companies presently manufacturing PHB include Metabolix (USA), Biomer (Germany), Biomatera (Canada) and PHB Industrial S/A (Brazil). Procter & Gamble has also developed a line of PHAs under the Nodax‘ trade name.

132

Green composites

7.4

Starch-based polymers: properties, biodegradation, processing and applications

7.4.1

Background

Starch is an energy storage material occurring as granules in some plants and microorganisms. The size and shape of starch granules depends upon the source. On a molecular level, starch contains glucosidic units and consists of about 70–80% amylopectin, a highly branched polymer with a weight-average molecular weight (Mw) of 107–109, and 20–30% amylose, a linear polymer with Mw of 105–106 (Fig. 7.4). The relative proportions of amylopectin and amylose in starch are determined by genetic and environmental control during biosynthesis. H CH2OH H

O H

HO H

OH O

H

n

(a) H CH2OH H

O H H

HO H H

CH2OH H

OH

O

O HO

H H H

OH O H

H2C H

(b)

O H

HO H H

OH O n

7.4 Structure of the polysaccharide components of starch: (a) amylose and (b) amylopectin.

Natural polymer sources

7.4.2

133

Properties of starch

Starch is an interesting alternative to synthetic polymers in situations where long-term durability is not a requirement. The properties of starch have been discussed in a recent review article (Shogren, 1998). Starch is a semi-crystalline material and amylopectin content is a major factor influencing the degree of crystallinity in most starches. The commercially important properties of starch are strongly influenced by the strength and character of the crystalline region, which is in turn determined by the ratio of amylopectin to amylose, the type of plant, the molecular weight distribution, the degree of branching and the conformation of each of the polymer components. All types of starch may be subjected to so-called destructurisation processes leading to thermoplastic materials. Destructurisation consists of the conversion of semi-crystalline starch granules into a homogeneous amorphous polymer matrix with extensive destruction of intermolecular hydrogen bonding combined with partial depolymerisation of starch macromolecules and can be achieved through the application of mechanical, thermal or thermomechanical energy (e.g. extrusion cooking). The most commonly employed starch types are those derived from maize, wheat, potato and tapioca. Special amyloses derived from high amylose content starches can also be processed into biodegradable thermoplastic films. The energy input in extrusion cooking can come from several different sources including dissipation of mechanical energy by the screw, heat transfer through the extruder barrel wall and the latent heat of condensed steam injected directly into the ingredients in a preconditioner or through the extruder barrel wall. Thermoplastic starch (TPS) can been defined as the product that is formed by a destructurisation of the native and semi-crystalline structures in a molecular-disperse homogeneous mixture consisting of poly-a-anhydroglucose and additives. The mechanical properties of destructurised starch depend upon the degree of destructurisation that is attained. As destructurisation increases, tensile strength and elongation also increase but the elastic modulus is reduced. This means that the material becomes increasingly flexible. Additives used to improve the appearance and properties of TPS include plasticisers, lubricants, fillers and dyes. Effective plasticisers reduce intermolecular hydrogen bond forces and resist migration within the biopolymer. Plasticisers that are suitable for TPS include water, urea, ammonia, diethylene glycol, triethylene glycol, polyvinyl alcohol and citric acid. Suitable fillers are various proteins, water-soluble polysaccharides and water-soluble polymers. TPS products with different viscosity, water solubility and water absorption properties have been prepared by altering the moisture content, the amylose/amylopectin ratio of the raw material and the temperature and pressure in the extruder.

134

7.4.3

Green composites

Biodegradation of starch

The ability to produce starch-degrading enzymes is widespread in microorganisms. Reports on the biodegradability of starch suggest that amorphous TPS may offer greater biological access for water and enzymes than more crystalline starch-based materials. In general, the degradation of starch is thought to take place in three steps, namely phosphorolysis, hydrolysis and transglycolysis. In the first of these steps, phosphorylases act as catalysts to convert starch to glucose-1-phosphate. These enzymes play an important role in renewed utilisation and mobilisation of the polysaccharides stored within the plant cells. In the second step, starch is hydrolysed outside the cell by amylases. These enzymes contribute to rapid breakdown of starch because of simultaneous attack at many a-linkages. Only b-amylases are found in plants and these enzymes separate maltoses with free, non-reducing ends, resulting in rapid release of sugars. In transglycolysis, the action of transglycolases produces a-, b- or g-cyclodextrins (Fritz et al., 1994).

7.4.4

Processing of thermoplastic starch

TPS can be processed as a conventional plastic but lubricants are essential in order to improve flow properties and also to lower the melting temperature so that processing becomes possible below the decomposition temperature range of starch. Lubricants that have been used include lipids, vegetablederived fatty acids, talc, silicone (although not biodegradable) and others. Colouring agents have included azo dyes, organic and inorganic pigments and dyes of natural origin. Among the non-biodegradable inorganic pigments, ferric and titanium oxides are suitable due to their low toxicity. Inorganic non-biodegradable fillers have also been used to improve material texture.

7.4.5

Applications

TPS has an affinity for water and is therefore unsuitable for most food packaging applications for example. Acetylation of starch can confer some degree of moisture resistance. Blends of starch with other polymers (e.g. polycaprolactone) allow starch-based polymers with greater flexibility to be produced. The hydrophilicity and poor mechanical properties of starch have so far resulted in few applications in which starch is used alone and not as part of a blend. However, current applications of TPS do include expanded loose-fill materials and liners inside some types of packaging. An interesting application for biodegradable starch films is in agricultural mulch films where successful trials have already been completed (Bastioli, 2003). This application could be particularly important in the developing world where intrinsically non-biodegradable polyethylene mulch film is now seen as a

Natural polymer sources

135

significant environmental pollutant. Applications of extruded starch foams in loose fills or in trays are also developing and may become much more significant in future because of the sustainable nature of these materials. Inevitably, in such applications where product requirements are low, manufacturers have to keep processes as simple and as cheap as possible. In the case of packaging materials, this may in part be achieved through the use of plant fibres as reinforcements or by the addition of foaming agents. As with cellulose, the free hydroxyl groups in starch can be chemically modified to produce a number of derivatives (e.g. starch esters, starch ethers). The hydroxyl groups on the 2-, 3- and 6-carbon atoms can be modified but attack at the C-6 primary hydroxyl is generally preferred owing to its exposed position within the molecule. Starch can also be cross-linked and examples include starch phosphates or diphosphates and starch modified by epoxidation (e.g. with ethylene oxide or epichlorohydrin). Through modification reactions, starch-based polymers can be made anionic or cationic in character and grafted polymers can be generated through polymerisation in the presence of unsaturated monomeric acids. However, the modification processes are usually time consuming and labour intensive and yields are often low. There have therefore so far been few commercial applications of chemically modified starches. Little information is available concerning the biodegradability of modified starches and the information that is available usually refers to the digestability of these derivatives in food applications. In general, the susceptibility to microbiological attack is thought to decrease in chemically modified starches, while water solubility goes up as the degree of substitution increases. Novamont, based in Italy, is a prominent European company in the business of manufacturing and supplying starch-based products. The company markets its products under the name Mater-Bi. Biotec GmbH in Germany has three product lines manufacturing Bioplast“ granules for injection moulding, Bioflex“ film and Biopur“ foamed starch. Other manufacturers of starchbased polymers presently include Avebe, Earth-Shell, Groen Granulaat, Hayashibara Chemical Labs, Midwest Grain Products, National Starch, Rodenburg Biopolymers, Starch Tech, Supol and Vegemat.

7.5

Bio-based composites: mechanical properties, processing, characterisation, modification, water absorption, biodegradation and reinforcement

In general, fibre composites are used where there is the need for structural materials with good durability, good mechanical properties and low weight. In this respect, natural fibres have a number of advantages as a polymer reinforcing phase, including low weight, low density, high specific properties,

136

Green composites

low cost, non-abrasive processing characteristics, lack of residues upon incineration and availability from renewable resources. The disadvantages of natural fibres are moisture absorption leading to fibre swelling, low thermal resistance, local or seasonal variations in quality and a tendency to agglomerate during processing. Natural fibres have the potential to substitute for glass fibres, particularly when specific mechanical properties are considered, as recently outlined by Wambua et al. (2003). Poor natural fibre–matrix adhesion occurs with hydrophobic polymers such as polyethylene or polypropylene; however, in practice, this problem can be addressed by the use of compatibilisers (e.g. maleated polypropylene) as process additives. Mechanisms based on the covalent attachment of such additives to hydrophilic surfaces have been proposed to explain their effectiveness (Lu et al., 2000). Since biodegradable polymers are usually more expensive than commodity thermoplastics, a high percentage of reinforcing fibre is desirable to improve mechanical performance while reducing overall material costs.

7.5.1

Mechanical properties of biocomposites

Much of the past research on biocomposites has included studies on mechanical characteristics (e.g. Wollerdorfer and Bader, 1998; Hermann et al., 1998). A number of researchers have investigated the reinforcement of PHAs with natural fibres such as cellulose (Gatenholm et al., 1992; Gatenholm and Mathiasson, 1994) and pineapple fibre (Luo and Netravali, 1999). A general finding from tensile tests is that the modulus increases but the elongation at break decreases as a result of fibre reinforcement. Lanzillota et al. (2002) obtained such results in their research with flax fibre-reinforced PLA composites and these researchers also found no increase in properties at fibre percentages greater than 30 wt %. This finding may be due to the increase in voids within the composite at higher fibre loadings. In another study, PLA–jute composites were produced by compression moulding under vacuum (Plackett et al., 2003). These composites had a significantly higher tensile modulus and strength than unreinforced polylactide and showed little sign of fibre pullout but, as expected, the samples were brittle (Fig. 7.5). Various researchers have prepared and characterised biocomposites from other natural polymers such as starch, soy protein and gelatin (Averous et al., 2001; Chiellini et al., 2001a; Carvalho et al., 2000; Dufresné et al., 1996). In the case of TPS composites prepared with glycerine as plasticiser and bleached eucalyptus grandis pulp fibres as reinforcement, the tensile modulus of the composites increased from 125 MPa with no reinforcement to 320 MPa with 16 wt % fibre reinforcement, elongation decreased from 31 to 11% and the tensile strength increased from 5 to 11 MPa. Water absorption was reduced when fibres were incorporated, which might be due to the use of cellulose fibres with less hydrophilic character than the matrix. The

Natural polymer sources

137

120 Jute/PLA 210 ∞C Jute/PLA 190 ∞C

Stress (MPa)

100 80

PLA 190 ∞C

60 40 20 0

0

1

2

3

4 5 Strain (%)

6

7

8

7.5 Tensile strength of compression-moulded jute–PLA composites, showing the enhanced strength and brittleness of the composites compared with the unreinforced polymer. (Source: Plackett et al., 2003.)

tensile properties of a variety of different biocomposites are summarised in Table 7.2. Various models have been used to simulate the mechanical properties of biocomposites but there are a number of factors that generally result in nonideal modelling processes. These factors include: ∑ the high variability of natural fibre mechanical properties ∑ non-uniform fibre cross-sections (models may assume that fibres are circular in cross-section) ∑ the fibrillar nature of natural fibres and the possibility of altering their morphology by pre-treatment and/or processing ∑ non-homogeneous distribution and orientation of fibres in composites ∑ imperfect fibre–matrix interfaces ∑ variable fibre aspect ratios with shorter fibres providing less effective reinforcement. However, despite non-ideal assumptions, the application of models to mechanical properties can help in the interpretation of experimental results. For example, strength increased with fibre content in the case of cellulose derivative–starch and sisal fibre composites (Alvarez et al., 2003; Ali et al., 2003) and the Bowyer–Bader model (Bowyer and Bader, 1972) was successfully applied to compute the strength of these composites. In addition to standard tensile and flexural test methods, cyclic-dynamical tests (DMTA) or fracture mechanics measurements have been applied to biocomposites. Creep, an important property in composite materials, is influenced by a number of factors including fibre content, temperature and applied stress. Fibre reinforcement generally improves the creep resistance of polymers but this improvement is very dependent on fibre percentage,

138

Green composites

Table 7.2 Tensile properties of various biodegradable composites and matrices Composites

Tensile modulus (MPa)

Processing

Reference

PHBV/20 wt % 0/90/0 pineapple fibres TPS/16 wt % wood pulp TPS/5 wt % clay hybrids TPS/20 wt % flax Mater-BiY/20 wt % flax Mater-BiY/20 wt % sisal Mater-BiZ/20 wt % sisal Mater-BiZ/20 wt % sisal PLA/20 wt % flax Soy protein isolate/20 wt % ramie fibre (l = 10 mm)

2158

3.0

46.0

Compression

Luo and Netravali, 1999

320

11.0

11.0

–

57.2

3.3

Curvelo et al., 2001 Park et al., 2002

–

–

36.4

3900

–

47.0

1183

10.0

18.6

257

8.0

12.7

117

–

6.7

5700

–

66.0

550

10.0

17.0

Intensive mixer-compression Intensive mixer-injection Twin-screw extruder-injection Twin-screw extruder-injection Intensive mixer-calender Intensive mixer-calender Intensive mixer-calender Twin-screw extruder-injection Mixingcompression

10.2

26.2

31.0

5.0

420

–

8.9

Mater-Bi-Y

1000

–

25.0

Mater Bi-Y

705

–

17.6

Mater-Bi-Z

37

859

7.3

MaterBi-Z

27

>30

4.0

3100

–

67.0

200

80

5.0

Matrices PHBV 1086 (Tm=162 ∞C) TPS 125 TPS

PLA Soy protein isolate

Elongation Tensile (%) strength (MPa)

Compression

Wollerdorfer and Bader, 1998 Lanzillota et al., 2002 Ali et al., 2003 Cyras et al., 2001 Ali et al., 2003 Lanzillota et al., 2002 Lohda and Netravali, 2002

Luo and Netravali, 1999 Intensive Curvelo et al., mixer-compression 2001 Twin-screw Wollerdorfer and extruder-injection Bader, 1998 Twin-screw Lanzillota et al., extruder2002 injection Intensive Ali et al., 2003 mixer-calender Intensive Cyras et al., 2001 mixer-calender Intensive Ali et al., 2003 mixer-calender Twin-screw Lanzillota et al., extruder-injection 2002 MixingLohda and compression Netravali, 2002

Natural polymer sources

139

fibre dimensions, fibre orientation, fibre distribution and fibre–matrix adhesion (Alvarez et al., 2004b; Ali et al., 2003). Composite creep behaviour has also been modelled and a good fit of experimental results to a four-parameter model has been reported, providing a relationship between the observed creep behaviour and composite morphology (Alvarez et al., 2004b). More research is necessary in order fully to evaluate long-term creep and the impact properties of biocomposites. A database of model parameters for natural fibre composites would be helpful for the future selection and use of these materials in practical applications.

7.5.2

Influence of processing on the mechanical properties of biocomposites

The identification of optimum processing methods (e.g. injection moulding, compression moulding) for biocomposites is clearly an essential step in the development of these products. Lanzilotta et al. (2002) produced trial automotive components by compression moulding of hybrid PLA–flax fleeces. This reinforced PLA material showed higher strength than short-fibre reinforced injection-moulded samples partly because of the longer fibre (14 mm) and the break-up of fibres in injection moulding to lengths of 0.3–0.4 mm. However, the modulus was lower than in the case of injection-moulded samples, which was possibly a result of voids at the fibre–matrix interface. Characterisation of the biopolymer matrix after processing has received relatively little attention. Plackett et al. (2003) converted L-PLA to film and used the film in combination with jute fibre mats to generate composites by a film stacking technique. Degradation of the biopolymer matrix was investigated using size exclusion chromatography, which revealed that only minor changes in the molecular weight distribution of the PLA occurred during the process, possibly because a vacuum was applied during the heating stage which reduced the opportunity for oxidative or hydrolytic degradation of the polymer. The thermal degradation of PHB after processing with natural fibres has also been studied. Unless stabilised, PHB can start to degrade at temperatures just above the melting point producing crotonic acid and other volatiles. In previous studies, when PHB was blended with natural fibres in a Brabender kneader, the fibre length distribution changed and PHB molecular weight decreased as a consequence of degradation by crotonic acid during processing (Gatenholm et al., 1992; Gatenholm and Mathiasson, 1994). Carvalho et al. (2000, 2003) studied the effect of glycerol and fibre content on the degradation of starch polymer. The addition of glycerol and fibres to starch was found to influence the extent of degradation during processing in opposite ways. The addition of glycerol decreased the degradation while the degradation increased with increasing fibre content. A reduction of molecular weight in starch was more significant for the high molecular weight fraction,

140

Green composites

suggesting that the molecular weight reduction during extrusion occurred mainly in the amylopectin fraction. Processing can influence polymer degradation and may also alter the length and diameter of natural fibres (Fig. 7.6). As a consequence, some attention has been paid to the adjustment of conditions in order to optimise biocomposite mechanical properties. For example, Wollerdorfer and Bader (1998) used a twin-screw extruder to compound fibre with biopolymers for injection moulding. In their research, machine operation variables such as temperature profile, injection and cooling time, mould temperature and pressure in the injection machine were selected for increased composite creep resistance. The shear stress that develops during mixing and extrusion processes can cause damage to the fibres and it should be possible to correlate these effects to the rheological and mechanical properties of the thermoplastic matrices. Operational conditions such as screw rotational speed and barrel temperature profile, as well as fibre volume fraction and fibre size, have been varied in the compounding of cellulose derivative/starch blends and miscanthus grass fibre. The results showed that the most important factors influencing the

Natural fibres

Biodegradable polymer granules

Drying

Variables = screw speed, feed rate, torque, temperatures

Twin-screw extruder Biocomposite

7.6 Schematic of biocomposite extrusion processing showing process variables and fibre degradation.

Natural polymer sources

141

Table 7.3 Mechanical properties of sisal fibre–starch biocomposites as a function of fibre aspect ratio after processing Sisal–reinforced PCL–starch composites

Sisal fibre-reinforced cellulose derivative–starch composites

Length/ diameter

Tensile strength (MPa)

Tensile modulus (MPa)

Length/ diameter

Tensile strength (MPa)

– 72.4 97.2 120.4 126.2

4.03 6.67 9.03 10.39 10.09

27.5 116.7 196.6 222.0 275.8

– 50.4 60.6 66.4 73.4

17.6 12.3 14.1 12.0 12.4

Tensile modulus (MPa) 704.6 958.0 1032.2 1081.6 1174.1

Source: Ali et al. (2003)

impact performance of the material were screw rotational speed and temperature (Johnson et al., 2003b). The effect of shear stress on the aspect ratio (length/diameter = l/d) of natural fibre was studied for starch blends with sisal (Vázquez et al., 1999; Iannace et al., 1999, 2001). The tensile modulus and the tensile strength of the composites generally increased as the fibre aspect ratio increased. In order to facilitate mixing, the temperature should be high; however, a high processing temperature can result in polymer degradation and lignin extraction from the fibres as well as fibre defibrillation. These effects produce changes in the surface properties of the fibres and also in the matrix with consequent changes in fibre–matrix adhesion. Table 7.3 shows results obtained by Ali et al. (2003) in terms of fibre aspect ratio and its influence on biocomposite tensile properties.

7.5.3

Characterisation of the interfacial properties of biocomposites

The mechanical properties of biocomposites depend on the properties of the constituent fibre and the polymer matrix, as well as the fibre–matrix interfacial shear strength (IFSS). Improvement of interfacial adhesion increases the tensile and flexural strength of composites but usually at the cost of lower impact strength and toughness. Fibre–matrix adhesion is influenced by the critical fibre length and is a result of the balance between interfacial shear force and normal force in the fibre. Fibres shorter than the critical length will not carry their maximum load and are thus unable to function effectively. Beyond the critical length the fibres will carry an increasing fraction of the applied load and may fracture before the matrix, especially if the matrix material has some ductility. Sufficient adhesion, low fibre diameter and high tensile strength allow for

142

Green composites

short critical fibre lengths. The maximum tensile strength of fibres can only be exploited if they can be stressed until they break without being pulled out. The critical fibre length can be estimated by the surface of fracture since the average pull-out length of a fibre cannot be longer than half of the critical fibre length. Microbond testing is a useful method for evaluating IFSS. In one study, the average value of the IFSS in pineapple fibre–PHB composites was 8.23 MPa, which compares with 30 MPa to 60 MPa for the IFSS between fibres such as Kevlar, glass and graphite and epoxy resins. The low IFSS value in the natural fibre composite was attributed to the high surface irregularity of pineapple fibres (Luo and Netravali, 1999). Lodha and Netravali (2002) measured the fracture stress in biocomposites based on ramie fibres and soy protein isolate and found it increased with increasing fibre length and fibre content. The average IFSS determined using the microbond technique was 29.8 MPa and the critical fibre length was calculated to be 2.54 mm. Composites containing 10 wt % fibre and 5 mm fibres were not significantly reinforced even though the critical length was exceeded. Composites with low fibre content had reduced tensile properties and it appeared that short fibres at low weight contents could act as imperfections or flaws and thereby reduce composite strength. It also seems that percolation behaviour occurred at this fibre content, because longer fibre will have fewer fibre ends within a given composite volume and hence there will be fewer flaws or low stress-bearing points. An optimum in fibre weight content exists, because at 30 wt % a large number of specimens showed delamination, possibly because of poor fibre wetting. Since PHAs are quite brittle, there have been a number of studies aimed at the use of plasticisers in PHA biocomposites (He et al., 2000; Cyras et al., 1999, 2000, 2001; Rozsa et al., 2002; Wong et al., 2002). Wong et al. (2002) prepared composites based on flax and PHB with added plasticisers. As revealed by examination of fracture surfaces using scanning electron microscopy (SEM), adequate interfacial adhesion was observed although this was influenced by migration of plasticiser to the fibre–matrix interface. Good fibre–matrix adhesion in flax–PHB–plasticiser combinations was also demonstrated by extensive transcrystallinity along the fibre surfaces. SEM can also be used to make observations about the coverage of fibres by the matrix and whether or not there are voids at the fibre–matrix interface. Lodha and Netravali (2002) showed traces of polymer matrix adhering to ramie fibres indicating some adhesion between soy protein–modified matrix and ramie fibre. This adhesion may be explained by hydrogen bonding between the ramie fibres and amine, amide, carboxyl and hydroxyl sites on the protein molecules. However, in contrast, Luo and Netravali (1999) used SEM and found pineapple fibre surfaces without PHB attached after debonding. Plackett and Andersen (2002) used environmetal scanning electron microscopy to

Natural polymer sources

143

7.7 Tensile fracture surface of a compression-moulded PHB–jute composite showing jute fibre bundles and void spaces at the fibre– polymer interface (¥ 800). (Source: Plackett and Anderson, 2002.)

examine tensile fracture surfaces and found evidence of voids at the fibre– polymer interface in compression moulded PHB–jute composites (Fig. 7.7).

7.5.4

Modification of the fibre–matrix interface and its influence on composite performance

Improved interfacial adhesion can potentially be achieved either by better fibre wetting during processing or by chemical bonding between fibre and matrix. Natural fibre surfaces are rich in hydroxyl groups and therefore have poor compatibility with hydrophobic polymer matrices. However, the presence of hydroxyl groups also ensures a reactive fibre surface that is highly suited to chemical modification. Surface modification of natural fibres can also potentially influence moisture uptake from the environment. Alkaline treatment of fibres has been widely studied (Kokot and Stewart, 1995; Joseph et al., 1996; Paul et al., 1997). High fibre fibrillation can occur if the treatment is carried out at high temperature or with mechanical stirring (i.e. breaking down of the fibre bundles into smaller fibres). Although such treatments weaken the fibres and do not necessarily improve composite mechanical properties, the collapse of the fibre cellular structure can produce a higher fibre density allowing an increase in fibre content without therma1 and mechanical degradation of the composite. Alkaline treatment also results in

144

Green composites

a higher fibre aspect ratio because of a reduction in fibre diameter. Consequently, samples with treated fibres having the highest l/d values can produce composites with higher tensile strength and modulus values (Ali et al., 2003). Another effect of alkaline treatment is to remove lignin and, since the fibres become more hydrophilic, one result is increased water absorption. Diffusion coefficients for composites containing treated fibres can show a slight decrease, possibly as a consequence of network formation by fibrillated fibres. Alkalinetreated fibre composites have been found to have a higher critical strain energy release rate (GIC), which could be a consequence of toughening mechanisms activated by fibre treatment (Alvarez et al., 2003). Mohanty et al. (2000a) studied the modification of two varieties of jute fabric (hessian cloth and carpet backing cloth) using dewaxing, alkali treatment, cyanoethylation and grafting. The results showed that the alkali-treated fabrics had better tensile and bending strengths than dewaxed samples. In defatted samples, dewaxing of the jute occurred which also helped to improve fabric– matrix interaction. The superior tensile strength of alkali-treated fabrics may result because alkali treatment improves the adhesive characteristics of the fibre surface through the removal of natural and artificial impurities and the production of rough fibre surface topography. In addition, alkali treatment leads to fibre fibrillation with an increase in the effective surface area available for contact with the polymer matrix. Thus, the development of rough surface topography and the enhancement of the fibre aspect ratio offer a better fibre– matrix interface adhesion and an increase in mechanical properties. Mohanty et al. (2000a) showed that reinforcement of a PHB matrix with untreated jute doubled the impact strength relative to the pure matrix. However, the highest interaction between fibre and matrix was obtained with alkali treatment and acrylonitrile treatment at low grafting percentages.

7.5.5

Water absorption in biocomposites

Since water can act as a plasticiser, absorbed moisture in composites can influence both the dimensional stability and the mechanical properties. Mechanisms of water uptake in a composite include diffusion through the matrix, capillarity through natural fibres or movement via porosities in the matrix or at the fibre–matrix interface. Consequently, water absorption depends not only on the relative hydrophilic character of the fibre and the matrix but also on the fibre–matrix interphase and the morphology of the composites. Interphase morphology can be influenced by voids, fibrillation and diffusion of additives to the fibre–matrix interface. Composite morphology can vary according to processing conditions; for example, fibres are generally more aligned in direction in injection-moulded composites than in compressionmoulded products.

Natural polymer sources

145

Water uptake in composites can be measured at different relative humidities or during water immersion. In the case of water immersion, some additives or other soluble materials can be extracted from the specimen. There are a wide number of variables that can influence water absorption and the resulting mechanical properties in biocomposites (Alvarez and Vázquez, 2004; Alvarez et al., 2003, 2004a and b). In general, biocomposite mechanical properties should decrease as water uptake increases. However, the loss of modulus can be less pronounced at high fibre contents due to the opposing effects of matrix plasticisation by water and fibre reinforcement.

7.5.6

Biodegradation of biocomposites

The biodegradation rate in biocomposites depends on a number of factors including fibre content, the biodegradability of each component and the quality of the interface. Wood fibre-reinforced PHA composites have been found to be highly biodegradable and to degrade faster than the pure matrix during incubation in activated sludge soil for five weeks at 40 ∞C. Wood fibres are thought to act as conduits for water, thus allowing greater access for microorganisms and faster degradation than in the pure matrix (Peterson et al., 2002). Poor fibre–matrix adhesion and the consequent void spaces can also result in enhanced water penetration and increased rates of biodegradation. Mohanty et al. (2000a) studied the degradation of composites based on PHAs and treated jute fibres. The highest weight loss during degradation in contact with compost was obtained with dewaxed jute composites and this was explained by weak fibre–matrix adhesion. A low degradation rate was found with polyacrylonitrile (PAN)-grafted jute fibre because PAN is a nonbiodegradable polymer. Alkaline treatment of fibres also produced a slightly higher degradation rate than that found in the pure matrix. Other research has shown that composite biodegradability depends on the relative biodegradation of the polymer and the reinforcing phase. Mohanty et al. (2000b) suggested that fibre addition generally produced an increase in the degradation rate of composites. Chiellini et al. (2001a) studied the properties of cast films based on waste gelatine and examined the influence of sugar cane bagasse on biodegradation. The addition of fibres reduced the durability in water as the fibre–matrix adhesion was easily destroyed causing physical disintegration of the composite. However, films containing variable amounts of bagasse showed fairly limited biodegradation.

7.5.7

Micro- and nano-scale reinforcement of biodegradable polymers

Some researchers have investigated reinforcement of biopolymers with cellulose microfibrils derived from natural fibres (Dufresné et al., 1997;

146

Green composites

Dufresné and Vignon, 1998; Mathew and Dufresné, 2002). Dufresné et al. (1997) extracted cellulose microfibrils from the sugar beet cell wall and obtained a microfibril suspension by a mechanical treatment with partial individualisation of microfibrils. The mechanical behaviour of films cast from sugar beet cellulose microfibrils was investigated by tensile testing. The tensile modulus of film samples was 2.3 GPa at 25% relative humidity (RH) but was reduced to 0.8 GPa at 75% RH. In another study, composite materials were processed from potato cellulose microfibrils, potato starch as matrix, and glycerol as plasticiser (Dufresné and Vignon, 1998). The cellulose microfibril content was varied from 0 to 40 wt %. The addition of glycerol to the starch produced a sharp decrease in the storage modulus in the rubbery state. This modulus measured at 30 ∞C increased from 10 MPa to 5 GPa when cellulose contents were 10 wt % or less. Water uptake increased with the addition of plasticiser and decreased as the cellulose content increased. The use of inorganic nanoclays in biopolymer reinforcement has received considerable attention and montmorillonite (MMT) is a type of clay nanofiller that has been widely examined for this application (Paul et al., 2003; Sinha Ray et al., 2002; Chen et al., 2002). The efficiency of MMT clay in modifying the properties of a polymer is determined particularly by the degree of its dispersion in the polymeric matrix. The hydrophilic nature of the mineral hinders homogeneous dispersion of the montmorillonite in certain polymers; however, in practice, due to their rich intercalation chemistry, clays can be organically modified to achieve greater compatibility with polymer matrices. In this way, nanocomposites are obtained in which the nanoparticles are either intercalated by the polymer or exfoliated into individual silicate layers of 1 nm thickness. In intercalation, polymer enters between individual nanoclay platelets but the platelets remain in a stacked form. When exfoliated the individual platelets are fully separated and the full potential enhancement of polymer properties can be achieved. The morphology of nanoclay platelets is illustrated in Fig. 7.8. Park et al. (2002) prepared a blend of thermoplastic starch and clay and investigated the influence of the nanostructure on the properties of the 8–10 nm

1–10 mm

1 nm

100–1000 nm nanoclay platelet

Primary nanoclay particle

7.8 Nanoclay morphology.

Aggregate

Natural polymer sources

147

composites. A TPS composite with an unmodified MMT gave a higher storage modulus and a larger shift in the relaxation temperature compared to other TPS–nanoclay combinations and this was attributed to better matching of surface polarities and improved interaction between filler and matrix. The nanocomposite had reduced water vapour permeability while the elongation at break in tensile tests increased from 47 to 57% and the tensile strength also increased from 2.61 to 3.32 MPa when an unmodified MMT clay was used.

7.6

Future trends

The biopolymers discussed in this chapter are at varying stages of development but as discussed and as shown in Table 7.1 there are a significant number of companies worldwide in a position to supply commercial quantities or trial shipments from pilot plant facilities. Future uses for biopolymers will naturally depend upon the possible evolution to less expensive products, as happened with the current non-biodegradable commodity thermoplastics some decades ago, as well as the continued development of new processes and new biopolymers for high-value applications. From a research perspective, the full potential of biocomposites may be realised in the future through studies on a number of key topics including: ∑ continued investigation of inexpensive natural fibre reinforcement of a wide range of biopolymers ∑ synthesis or modification of natural polymers in order to improve their stability for specific applications ∑ improved processing through design of new elements for twin-screw extrusion of natural fibre–biopolymer combinations ∑ processes to improve the impact properties of biocomposites ∑ improved understanding of biocomposite fracture mechanics ∑ characterisation of long-term properties such as creep ∑ studies of biocomposite ageing behaviour ∑ studies on thermal degradation of biocomposites ∑ investigation of nanocomposites combining natural fibres with inorganic fillers (e.g. nanoclays) ∑ continued investigation of methods for characterisation of nanocomposites ∑ examination of recyclability and life cycle aspects. In terms of applications, it can be anticipated that the reinforcement of biopolymers with natural fibres could be of particular relevance in the packaging field where biopolymer films are already attracting interest. On the other hand, the biocompatibility of biopolymer surfaces means that we will likely see further development of reinforced biopolymers for application in the medical field. Although natural fibres are perhaps unlikely to appear in the

148

Green composites

latter application, organic/inorganic hybrid materials (e.g. nanocomposites) with their combination of enhanced properties may become increasingly important in this and other areas.

Sources of further information There are numerous current sources of information on naturally derived polymers. One of the most useful on-line sources is the website for the biopolymer network (www.biopolymer.net). This website provides many links to other sites with information on commercial biopolymers listed by name, recent press releases, research organisations, certification schemes and other related topics. The link to books on biopolymers is also very helpful. Other interesting biopolymer-related sites on the internet include those for the US National Bio-based Product and Energy Initiative, the Bioenvironmental Polymer Society (BEPS) and the International Biodegradable Products Institute. The website for the German company, Biomer (www.biomer.de) has good links and can be a useful reference to a crosssection of the biopolymer research literature. A particularly comprehensive report prepared for the Agro-Industrial Division of the European Commission entitled ‘Production of Thermobioplastics and Fibres based mainly on Biological Materials’ (Fritz et al., 1994) is a good source of background information on biopolymers, especially when combined with more recent reports on specific developments in the scientific literature. A book edited by Chiellini (2001b) provides a summary of papers given at international conferences on biopolymer technology in 1999 and 2000. Another useful and more recent document is RAPRA Report 159 entitled ‘Biopolymers’ (Johnson et al., 2003a) in which a general overview of biopolymers is presented in combination with an extensive series of abstracts from the RAPRA polymer library database. For those with interests in life cycle analysis, the article by Gerngross that featured in the Scientific American (Gerngross and Slater, 2000) provides food for thought on biopolymers in the context of sustainability and has recently stimulated further debate on this topic (Gerngross et al., 2003). For space reasons, this chapter could not deal with all possibilities in the context of biopolymers and biocomposites and a notable exception has been the extensive research and commercialisation of soybean-derived polymers and their biocomposites. For further reading on this topic, Wool et al. (2002) can provide a useful starting point. An interesting source of further information on natural fibre-reinforced biopolymers is the website for the UK-based Sustainable Composites network (www.suscomp.net). This network was established in the UK in 2001 to advance the development and commercialisation of sustainable composite materials. In this context, sustainable composites are deemed to be materials

Natural polymer sources

149

that have minimal impact upon the environment and include materials that are readily recyclable. The Sustainable Composites network website provides links to the Alternative Crops Technology Interaction Network (ACTIN), the Biomimetics Network for Industrial Sustainability (BIONIS), the Centre for Advanced and Renewable Resources (CARM) and the European Renewable Resources and Materials Association (ERRMA).

References Albertsson, A.-C. and Karlsson, S. (1994). Chemistry and biochemistry of polymer biodegradation. In Griffin G J L, Chemistry and Technology of Biodegradable Polymers, Glasgow: Blackie, pp. 7–17. Ali, S.A.M., Doherty, P.J. and Williams, D.F. (1994). The mechanism of oxidative degradation of biomedical polymers by free radicals. J. Appl. Polym. Sci., 51 (8), 1389–98. Ali, R., Iannace, S. and Nicolais, L. (2003). Effect of processing conditions on mechanical and viscoelastic properties of biocomposites. J. Appl. Polym. Sci., 88 (7), 1637–42. Alvarez, V.A., Ruseckaite, R.A. and Vázquez, A. (2003). Mechanical properties and water absorption behaviour of composites made from a biodegradable matrix and alkaline-treated sisal fibers. J Compos. Mater., 37 (17), 1575–88. Alvarez, V.A. and Vázquez, A. (2004). Effect of water sorption on the flexural properties of a fully biodegradable biocomposite. J. Compos. Mater., in press. Alvarez, V.A. Fraga, A.N. and Vázquez, A. (2004a). Effect of the moisture and fiber content on the mechanical properties of biodegradable polymer and sisal fiber biocomposites. J. Appl. Polym. Sci., 91 (6), 4007–16. Alvarez, V.A., Kenny, J.M. and Vázquez, A. (2004b). Creep behavior of biocomposites based on sisal fiber reinforced cellulose derivatives/starch blends, Polym. Compos, 25 (3), 280–8. Asrar, J. and Gruys, K. (2002). Biodegradable polymer (Biopol“). In Doi, Y. and Steinbüchel, A., Biopolymers, Volume 4, Weinheim: Wiley–VCH Verlag, pp. 53–90. Averous, L., Frigant, C. and Moro, L. (2001). Plasticized starch–cellulose interactions in polysaccharide composites. Polymers, 42 (15), 6565–72. Bastioli, C. (2003). Starch-based biopolymers – applications and environmental benefits. In Proceedings of the International Symposium on Advanced Bioplastics, Nürnberg, Germany, 12–13 February, 2003. Bowyer, W.H. and Bader, M.G. (1972). Reinforcement of thermoplastics by imperfectly aligned discontinuous fibers. J. Mater. Sci., 7 (11), 1315–21. Carvalho, A.J.F., Curvelo, A.A.S. and Agnelli, J.A.M. (2000). Thermoplastic starch composites. In Proceedings from the Third International Symposium on Natural Polymers and Composites and the Workshop on Progress in Production and Processing of Cellulosic Fibres and Natural Polymers, Sao Pedro, Brazil, May 14–17, 2000, pp. 212–217. Carvalho, A.J.F., Zambon, M.D., Curvelo, A.A.S. and Gandini, A. (2003). Size exclusion chromatography characterization of thermoplastic starch composites 1. Influence of plasticizer and fibre content. Polym. Degrad. Stabil., 79 (1), 133–8. Chen, G.X., Hao, G.J., Guo, T.Y., Song, M.D. and Zhang, B.H. (2002). Structure and mechanical properties of poly (3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV)/clay nanocomposites. J. Mater. Sci. Lett., 21 (20), 1587–9.

150

Green composites

Chiellini, E., Cinelli, P., Corti, A. and Kenawy, E.R. (2001a). Composite films based on waste gelatin: thermal–mechanical properties and biodegradation testing. Polym. Degrad. Stabil., 73 (3), 549–55. Chiellini, E., Gil, H.; Braunegg, G.; Burchert, J.; Gatenholm, P. and van der Zee, M. (2001b). Biorelated Polymers: Sustainable Polymer Science and Technology, Dordrecht, The Netherlands: Kluwer Academic/Plenum Publishers. Curvelo, A.A.S., Carvalho, A.J.F. and Agnelli, J.A.M. (2001). Thermoplastic starch cellulosic fibers composites: preliminary results. Carbohyd. Polym., 45 (2), 183–8. Cyras, V.P., Galego Fernández, N. and Vázquez, A. (1999). Biodegradable films from PHB-8HV copolymers and polyalcohols blends: crystallinity, dynamic mechanical analysis and tensile properties. Polym. Int., 48 (8), 705–12. Cyras, V.P., Vázquez, A., Rozsa, Ch., Galego, N., Torre, L. and Kenny, J.M. (2000). Thermal stability of P(HB-co-HV) and its blends with polyalcohols: crystallinity, mechanical properties and kinetic of degradation. J. Appl. Polym. Sci., 77 (13), 2889– 900. Cyras, V.P., Iannace, S., Kenny, J.M. and Vázquez, A. (2001). Relationship between processing conditions and properties of a biodegradable composite based on PCL/ starch and sisal fibers. Polym. Compos., 22 (1) 104–10. Doi, Y., Kumagai, Y., Tanahashi, N. and Mukai, K. (1992). Structural effects on biodegradation of microbial and synthetic poly(hydroxyalkanoates). In Vert, M., Feijen, J., Albertsson, A.-C., Scott, G. and Chiellini, E., Biodegradable Polymers and Plastics, Cambridge: The Royal Society of Chemistry. Dufresné, A. and Vignon, M.R. (1998). Improvement of starch film performance using cellulose microfibrils. Macromolecules, 31 (8), 2693–6. Dufresné, A., Cavaillé, J.-Y. and Helbert, W. (1996). New nanocomposite materials: microcrystalline starch reinforced thermoplastic. Macromolecules, 29 (23), 7624–6. Dufresné, A., Cavaillé, J.-Y. and Vignon, M.R. (1997). Mechanical behavior of sheets prepared from sugar beet cellulose microfibrils. J. Appl. Polym. Sci., 64 (6), 1185–94. Evans, J.D. and Sikdar, S.K. (1990). Biodegradable plastics: an idea whose time has come? Chemtech, 20, 38–42. Fritz, H.-G., Seidenstücker, T., Bölz, U., Juza, M., Schroeter, E. and Endres, H.-J. (1994). Production of thermo-bioplastics and fibres based mainly on biological materials. Report for the Agro-Industrial Division (E2) of EC Directorate General XII for Scientific Research and Development. Gatenholm, P. and Mathiasson, A. (1994). Biodegradable natural composites. II. Synergistic effects of processing cellulose with PHB. J. Appl. Polym. Sci., 51 (7), 1231–7. Gatenholm, P., Kubat, J. and Mathiasson, A. (1992). Biodegradable natural composites I. Processing and properties. J. Appl. Polym. Sci., 45 (9), 1667–77. Gerngross, T. and Slater, S. (2000). How green are green plastics? Sci. Am., 283, 36–41. Gerngross, T., Slater, S., Gross, A.R. and Kalra, B. (2003). Biopolymers and the environment. Science, 299, 822–3. Gilmore, D.F., Fuller, R.C. and Lenz, R. (1990). Biodegradation of poly (betahydroxyalkanoates). In Barenberg, S.A., Brash, J.L., Narayan, R. and Redpath, A.E., Degradable Materials: Perspectives, Issues and Opportunities, Boca Raton, Fl: CRC Press. Gruber, P.R. and O’Brien, M. (2002). Polylactides ‘NatureWorks PLA’. In Doi, Y. and Steinbüchel, A., Biopolymers, Volume 4, Weinheim: Wiley–VCH Verlag, pp. 235–50. Hankermeyer, C.R. and Tjeerdema, R.S. (1999). Polyhydroxybutyrate: Plastic made from and degraded by microorganisms. Rev. Environm. Contam. Toxicol., 159, 1–24.

Natural polymer sources

151

Haugaard, V.K. (2003). Pak maden i maden, Alimenta., May, 4–5. He, Y., Asakawa, N. and Inoue, Y. (2000). Biodegradable blends of high molecular weight poly (ethylene oxide) with poly(3-hydroxypropionic acid) and poly(3hydroxybutyric acid): a miscibility study by DSC, DMTA and NMR spectroscopy. Polym. Int., 49 (6), 609–17. Hermann, A.S., Nickel, J. and Riedel, U. (1998). Construction materials based upon biologically renewable resources-from components to finished parts. Polym. Degrad. Stabil., 59 (1–3), 251–61. Hocking, P.J. and Marchessault, R.H. (1994). Biopolyesters. In Griffin, G.J.L., Chemistry and Technology of Biodegradable Polymers, Glasgow: Blackie, pp. 48–96. Iannace, S., Nocilla, G. and Nicolais, L. (1999). Biocomposites based on sea algae fibers and biodegradable thermoplastic matrices. J. Appl. Polym. Sci., 73 (4), 583–92. Iannace, S., Ali, R. and Nicolais, L. (2001). Effect of processing conditions on dimensions of sisal fibers in thermoplastic biodegradable composites. J. Appl. Polym. Sci., 79 (6), 1084–91. Ikada, Y. and Tsuji, H. (2002). Biodegradable polyesters for medical and ecological applications. Macromol. Rapid Commun., 21 (3), 117–32. Johnson, R.M., Mwaikambo, L.Y. and Tucker, N. (2003a). Biopolymers, RAPRA Report 159, 14 (3). Johnson, R.M., Tucker, N. and Barnes, S. (2003b). Impact performance of Miscanthus Novamont Mater-Bi(R) biocomposites. Polym. Test, 22 (2), 209–15. Joseph, K.M., Thomas, S. and Pavithran, C. (1996). Effect of chemical treatment on tensile properties of short sisal fibre-reinforced polyethylene composiltes. Polymers, 37 (23), 5139–49. Kokot, S. and Stewart, S. (1995). An exploratory study of mercerized cotton fabrics by DRIFT spectroscopy and chemometrics. Textile Res. J., 65 (11), 643–51. Lanzillotta, C., Pipino, A. and Lips, D. (2002). New functional biopolymer natural fiber composites from agricultural resources. In Proceedings of the Annual Technical Conference – Society of Plastics Engineers, San Francisco, California, Vol. 2, pp. 2185–9. Lichtl, M. (2003). The Kassel project – a test market for compostable packaging. In Proceedings of the International Symposium on Advanced Bioplastics, Nürnberg, Germany, 12–13 February, 2003. Lipinsky, E.S. and Sinclair, R.G. (1986). Is lactic acid a commodity chemical?. Chem. Eng. Proc., 82 (8), 26–32. Ljungberg, N. and Wesslén, B. (2002). The effect of plasticizers on the dynamic mechanical and thermal properties of poly(lactic acid). J. Appl. Polym. Sci., 86 (5), 1227–34. Lodha, P. and Netravali, A.N. (2002). Characterization of interfacial and mechanical properties of ‘green’ composites with soy protein isolate and ramie fiber. J. Mater. Sci., 37 (17), 3657–65. Lu, J.Z., Wu, Q. and McNabb, H.S. Jr. (2000). Chemical coupling in wood fiber and polymer composites: a review of coupling agents and treatments. Wood Fibre Sci., 32 (1), 88–104. Luo, S. and Netravali, A.N. (1999). Interfacial and mechanical properties of environment friendly ‘green’ composites made from pineapple fibers and poly(hydroxybutyratecovalerate) resin. J. Mater. Sci., 34 (15), 3709–19. Luzier, W.D. (1992). Materials derived from biomass/biodegradable materials. Proc. Natl Acad. Sci. USA, 89 (3), 839–42.

152

Green composites

Marchessault, R. and Bluhm, T.L. (1988). Poly(b-HA)s: biorefinery polymers in search of applications. Makromol. Chem. Macromol. Symp., 19, 235–54. Mathew, A.P. and Dufresné, A. (2002). Morphological investigation of nanocomposites from sorbitol plasticized starch and tunicin whiskers. Biomacromolecules, 3 (3), 609– 17. Mitomo, H., Barham, P.J. and Keller, A. (1988). Temperature-dependence of mechanicalproperties of poly(beta-hydroxybutyrate-beta-hydroxyvalerate). Polym. Commun., 29 (4), 112–15. Mohanty, A.K., Khan, M.A. and Hinrichsen, G. (2000a). Surface modification of jute and its influence on performance of biodegradable jute-fabric/Biopol composites. Comp. Sci. Tech., 60, 1115–24. Mohanty, A.K., Misra, M. and Hinrichsen, G. (2000b). Biofibres, biodegradable polymers and biocomposites: an overview. Macromol. Mater. Eng., 276/277, 1–24. Mohanty, A.K., Misra, M. and Drzal, L.T. (2002). Sustainable bio-composites from renewable resources: opportunities and challenges in the green materials world. J. Polym. Environ., 10 (1/2), 19–26. Park, H.-M., Li, X., Jin, Ch.-Z., Park, Ch.-Y., Cho, W.J. and Ha, Ch.-S. (2002). Preparation and properties of biodegradable thermoplastic starch clay hybrids. Macromol. Mater. Eng., 287, 553–8. Paul, A., Joseph, K. and Thomas, S. (1997). Effect of surface treatments on the electrical properties of low-density polyethylene composites reinforced with short sisal fibers. Comp. Sci. Tech., 57 (1), 67–79. Paul, M.-A., Alexandre, M., Degee, P., Calberg, C., Jerome, R. and Dubois, P. (2003). Exfoliated polylactide/clay nanocomposites by in-situ coordination-insertion polymerization. Macromol. Rapid Commun., 24 (9), 561–6. Peterson, S., Jayaraman, K. and Bhattacharyya, D. (2002). Forming performance and biodegradability of wood fibre-Biopol“ composites. Comp. Part A: Appl. Sci., 33 (8), 1123–34. Plackett, D.V. and Andersen, T.L. (2002) Biocomposites from natural fibres and biodegradable polymers: processing, properties and future prospects. In Proceedings of the 23rd Risø International Symposium on Materials Science: Sustainable Natural and Polymeric Composites – Science and Technology, Risø, Denmark, 2–5 September, 2002. Plackett, D., Andersen, T.L., Pedersen, W.B. and Nielsen L. (2003). Biodegradable composites based on L-polylactide and jute fibres. Comp. Sci .Tech., 63 (9), 1287–96. Rozsa, Ch., Ortiz, P., Cyras, V. P., Vázquez, A. and Galego, N. (2002). Poly(hydroxybutyrateco-hydroxyvalerate)-polyadipate blends. Int. J. Polym. Mater., 51 (7), 619–31. Shogren, R.L. (1998). Starch: properties and material applications. In Kaplan, D.L., Biopolymers from Renewable Resources, Macromolecular Systems-Materials Approach, Berlin: Springer-Verlag, pp. 30–46. Sinha Ray, S., Yamada, K., Okamoto, M. and Ueda, K. (2002). Polylactide-layered silicate nanocomposite: a novel biodegradable material. Nano Lett., 2 (10), 1093–6. Sødergaard, A. and Stolt, M. (2002). Properties of lactic acid based polymers and their correlation with composition. Progr. Polym. Sci., 27 (6), 1123–63. Steinbüchel, A. (1995). Use of synthetic, biodegradable thermoplastics and elastomers from renewable resources – the pros and cons. J. Macromol. Sci. Pure Appl. Chem., A32 (4), 653–60. Strasser, H. and Owen, A.J. (1991). Yield behaviour of bacterially produced poly(bhydroxybutyrate-co-b-hydroxyvalerate). Trends Polym. Sci., 1 (1), 63–71.

Natural polymer sources

153

Vázquez, A., Cyras, V.P., Kenny, J.M. and Iannace, S. (1999). In Proceedings of the 12th International Conference on Composite Materials, Paris, France, Cambridge: Woodhead. Wambua, P., Ivens, J. and Verpoest, I. (2003). Natural fibres: can they replace glass in fibre reinforced plastics? Comp. Sci. Tech., 63 (9), 1259–64. Williams, S.F. and Peoples, O.P. (1996). Biodegradable plastics from plants. Chemtech, 38, 38–44. Wollerdorfer, M. and Bader, H. (1998). Influence of natural fibres on the mechanical properties of biodegradable polymers. Ind. Crop. Prod., 8 (2), 105–12. Wong, S., Shanks, R. and Hodzic, A. (2002). Properties of poly(3-hydroxybutyric acid) composites with flax fibres modified by plasticiser absorption, Macromol. Mater. Eng., 287 (10), 647–55. Wool, R.P., Khot, S.N., LaScala, J.J., Bunker, S.P., Lu, J., Thielemans, W., Can, E., Morge, S.S. and Williams, G.I. (2002). Affordable composites and plastics from renewable resources. Part 1. Synthesis of monomers and polymers’. In Advancing Sustainability through Green Chemistry and Engineering, ACS Symposium Series, 823, 177–204. Yu, J. (2001). Production of PHA from starchy wastes via organic acids. J. Biotech., 86 (2), 105–12. Yu, P.H.F., Chua, H., Huang, A.L., Lo, W.H. and Ho, K.P. (1999). Transformation of industrial food wastes into polyhydroxyalkanoates. Water Sci. Technol., 40 (1), 365– 70.

8 Optimising the properties of green composites S. H. A Z I Z A N D M. P. A N S E L L University of Bath, UK

8.1

Introduction

Natural fibre polymer composites offer a sustainable alternative to commercial composites. The production of synthetic fibres such as glass, aramid and carbon fibres is energy intensive, and synthetic polymer matrices, such as epoxy thermosets and polypropylene thermoplastics are by-products of the oil industry. However, the choice of natural fibre and polymer alternatives for applications in the automotive, construction and other industries is constrained by the performance specification for the composites and the method of manufacture. There may also be constraints based on a future need to recycle composites, the environmental stability of composites and the level of ‘greenness’ of composites. In this chapter the factors that determine the mechanical performance and thermal stability of natural fibre composites are investigated and a broad examination of thermoplastic polymer matrices versus thermosetting polymer matrices is made in the context of manufacturing methods. In order to exploit the ultimate mechanical properties of natural fibres, they should be combined with thermosetting matrices to form composites. Furthermore, a truly green composite requires a green matrix. Accordingly, the mechanical and thermal performance of thermosetting polymer composites with partially green and completely green credentials are examined and compared. The composites are based on hemp or kenaf fibres, organised both unidirectionally and as a fibre mat, in a thermosetting matrix of polyester resin (partially green) or cashew nut shell liquid (CNSL) resin (completely green). Factors which influence properties, such as fibre treatment and resin modification, are examined.

154

Optimising the properties of green composites

155

8.2

Thermosetting matrices versus thermoplastic matrices: a comparison

8.2.1

Differences in chemistry and mechanical behaviour of polymeric matrices

Cured thermosetting polymers, such as epoxy, polyester and phenolic resins, possess chemically cross-linked structures with the following thermomechanical properties when fully cross-linked: ∑ ∑ ∑ ∑ ∑ ∑ ∑

essentially brittle behaviour over a wide range of temperatures inability to deform viscoelastically no rubbery behaviour non-crystalline structure in the form of a cross-linked network polymer will degrade or burn rather than melt may be resin transfer moulded through one cycle only cannot be recycled by melting and reforming.

On the other hand, thermoplastic polymers such as polypropylene, polyethylene and polytetrafluoroethylene, contain no cross-links and exhibit the following thermomechanical properties: ∑ brittle behaviour below the glass transition temperature, Tg ∑ viscoelastic behaviour immediately above Tg followed by rubbery and then flow behaviour at higher temperatures ∑ generally semi-crystalline or completely crystalline with a spherulitic structure ∑ ability to deform viscoelastically and plastically ∑ may be melted and reformed through a limited number of cycles before degrading significantly ∑ may be injection moulded through several cycles, especially polypro– pylene, ∑ may be tough in impact if room temperature is above Tg ∑ maximum working temperature is generally lower than for thermosets. In practice partially cross-linked thermosets will possess a measurable Tg and some of the viscoelastic response exhibited by thermoplastics. The elastic modulus versus temperature characteristics for a fully cross-linked thermoset, an amorphous thermoplastic and a semi-crystalline thermoplastic are sketched in Fig. 8.1. Typical elastic moduli (strictly stress relaxation moduli) of fully crosslinked thermosets are usually higher than for thermoplastics and neither a Tg or a crystalline melting point, T m, are measurable. Semi-crystalline thermoplastics possess both a Tg and a Tm, whereas amorphous thermoplastics obviously lack a Tm.

156

Green composites

Modulus of elasticity

Thermosetting polymer

Semi-crystalline polymer

Elastic Amorphous thermoplastic Viscoelastic

Rubbery

Tg

Tm

Temperature

8.1 Elastic modulus versus temperature characteristics for polymers.

Natural fibres held in a polymer matrix at room temperature therefore find themselves in different chemical environments depending on whether they are surrounded by a thermoplastic or thermosetting polymer. The level of adhesion between fibre and matrix will depend on: ∑ the surface energies of the fibre and matrix phases ∑ the nature of fibre chemical pretreatment, e.g. as-received condition, alkalisation, acetylation, silane treatment ∑ the addition of adhesion promoters to the resin, e.g. maleic anhydridemodified polypropylene (MAPP) ∑ the degree of cross-linking of thermosetting matrices ∑ the Tg of thermoplastic matrices in relation to room temperature ∑ the degree of shrinkage of the matrix onto the fibres during manufacture of composites. These factors influence strength, stiffness and toughness of the natural fibre composite.

8.2.2

Enhancement of the natural fibre to matrix interfaces

Impurities such as wax and natural oils covering the external surface of the cell wall of plant fibres may be removed by chemical treatment to improve adhesion between fibre and resin matrix. Sodium hydroxide (NaOH) in solution is the most commonly used chemical for bleaching or cleaning the surface of plant fibres in the process of alkalisation. This process exposes the microfibrils and gives a rough surface topography to the fibre. Cellulose microfibrils are modified, developing changes in morphology and increase in the number of reactive hydroxyl groups. These changes result in improved surface tension, wetting ability, swelling, adhesion and compatibility with polymeric materials (Felix et al., 1994). Several researchers have reported an improvement in the

Optimising the properties of green composites

157

mechanical properties of cellulose fibres when alkalised at different NaOH concentrations. Bisanda and Ansell (1992) treated sisal fibres with a 0.05 N NaOH solution while Sreekala et al. (1997) and Geethamma et al. (1995) used the same concentration of NaOH to remove surface impurities on oil palm fibres and short coir fibres, respectively. Mwaikambo and Ansell (2002) found a rapid change in crystallinity index (CI) for NaOH concentration between 0.8% and 8% after alkalisation beyond which the change in CI was found to be marginal. Alternative treatments include impregnation of fibres with LDPE–xylene solution (Herrera-Franco and Aguilar-Vega, 1997), silane treatment of fibres (Devi et al., 1997), benzoylation, polystyrene maleic anhydride coating and acetylation of fibre (Nair et al., 2001) and cyanoethylation of fibres (Saha et al., 1999). These techniques have been reported to improve fibre–matrix adhesion and hence the mechanical properties of natural fibre composites.

8.2.3

Short fibre versus long fibre reinforcements

Bader (2000) examines the influence of the orientation of fibre reinforcement on the stiffness index of fibre composites. The stiffness index is the product of the maximum fibre volume fraction and the orientation factor and both of these parameters depend on the manufacturing process for the composite. Table 8.1 is reproduced from Bader’s excellent paper and the manufacturing processes include filament winding, pultrusion, hot pressing, autoclaving, Table 8.1 Implications of the choice of reinforcement format for fibre composites Reinforcement format

Maximum fibre volume fraction

Orientation factor

Stiffness index

Manufacturing processes

UD tow

0.80

1.00

0.80

UD prepreg Multi-axial prepreg* 2D non-crimp fabric Woven 2D fabric*^ Orthogonal 3D fabric Random planar

0.65 0.60

1.00 0.31

0.65 0.19

Filament wind, pultrusion, hot press Autoclave, RFI Autoclave, RFI

0.55

0.30

0.17

RFI, RTM

0.50 0.40

0.27 0.30

0.14 0.12

RTM, wet lay-up RTM

0.30

0.30

0.09

0.20

0.12

0.02

SMC, RTM, wet lay-up BMC, IM (thermoplastic)

Random 3D short fibre

*Quasi-isotropic lay-up, ^allows for the effect of crimp. Source: Bader (2000).

158

Green composites

resin film infusion (RFI), resin transfer moulding (RTM), wet lay-up, sheet moulding from sheet moulding compound (SMC), bulk moulding from bulk moulding compound (BMC) and injection moulding (IM). It can be observed that the stiffness index ranges from 0.8 for pultruded composites to 0.02 for injection mouldings, a factor of 40 difference. The stiffness of the composite, a major factor in design, depends on the closeness of fibre packing. For cylindrical fibres aligned parallel to each other this corresponds to a theoretical maximum volume fraction of 0.907 when the fibres are organised unidirectionally (orientation factor equal to 1). In practice, the maximum fibre volume fraction is of the order of 0.8, because packing is never perfect and good fibre wetting by the resin matrix must be achieved. Natural fibres comprise either single hollow ultimate fibres (e.g. cotton and kapok) or more usually fibre bundles (e.g. jute, sisal, hemp and kenaf). It is possible to compress natural fibres because of their porosity so the concept of volume fraction becomes more flexible. At the other extreme to composites manufactured with unidirectional fibres, injection moulding of short-fibre filled thermoplastics results in poor fibre packing (volume fraction ~0.2) and fibre randomisation (orientation factor ~0.12) giving a stiffness index of approximately 0.02. In practice the flow characteristics of short fibres in the mould may develop regions of fibre alignment, e.g. near mould surfaces.

8.2.4

Manufacture of thermoplastic matrix versus thermosetting matrix–natural fibre composites

Injection moulding is an ideal route for producing thermoplastic composite components with complex shapes but is limited to utilising short fibres because of the size of the injection port and the requirement for unimpeded flow of the melted polymer. The downside is the poor stiffness index of the composites, the low maximum working temperature of the thermoplastic matrix and the dubious degree of bonding between the natural fibres and the polymer matrix. For effective reinforcement of plastics, fibres must reach a minimum critical fibre length, such that on the application of external load to the composite, shear interaction between fibre and matrix allows the axial stress along the fibre to build up from each fibre end until it reaches the average composite stress. Short fibres may not reach the critical length and are very likely to deviate from the principal axis of the composite. Their role may therefore be that of a filler rather than a reinforcement, cutting the cost of the oil-based polymer matrix but having a minimal effect on composite strength and stiffness. Indeed, thermoplastic matrices are observed to yield around short fibres (Spear, 2002). Pre-polymerised thermosets may also be used for injection moulding, e.g. phenol formaldehyde, cellulose propionate and polyurethane rubber, however, cycle times are higher.

Optimising the properties of green composites

159

The stiffness index of composites may be raised by laying fibres straight or cross-laminating them to produce composites with excellent unidirectional or bidirectional mechanical properties, respectively. These fibres may be infiltrated with low viscosity thermosetting resin and hardener and consolidated by processes such as hot pressing, pultrusion or RTM. Room temperature cured thermosetting resins are less thermally stable as they are generally only partially cross-linked. Hence processes such as hot pressing drive the resin cross-linking to completion, often in conjunction with a post-cure in an oven. The interfacial bond between fibre and matrix is usually sufficiently coherent to allow efficient stress transfer between resin and fibre, and fibre surface preparation and adhesion promoters such as silanes improve bonding and stress transfer. In order to produce optimised properties from a natural fibre composite the following characteristics are required: ∑ long, straight fibres to allow efficient stress transfer and to optimise mechanical properties along the fibre axis ∑ a thermosetting matrix which will not yield significantly ∑ optimisation of the fibre to resin interface by chemical treatment of the fibres or modification of the resin matrix. There are clear environmental advantages in selecting a natural thermosetting matrix. This chapter compares the performance of polyester resins and cashew nut shell liquid resins for thermosetting polymer matrices.

8.2.5

Green thermosetting polymers – cashew nut shell liquid resin

Natural resin precursors are mostly obtained from plants. The most common natural resin precursors are phenol based, namely tannins and cashew nut shell liquid (CNSL). Cashew nut shell liquid oil is a blend of naturally occurring phenol-based monomers. The shell of the cashew nut contains alkylphenolic oil amounting to nearly 25% of the total weight of the nut (Tyman, 1979). It is extracted as a reddish brown viscous liquid from the shell of the cashew nut, from the cashew tree, Anacardium occidentale L. The oil is highly flammable and toxic. The cashew tree, originally from Brazil is now extensively grown in India, Mozambique, the Malagasy Republic, Tanzania, Philippines, Panama and other tropical countries. The world production of cashew nuts is nearly 500 000 tonnes per year with Brazil as the largest producer (Mwaikambo, 2002). CNSL is traditionally obtained as a by-product during the process of removing the cashew kernel from the nut. The kernel is located in the centre of the nut, surrounded by the shell. The shell weighs 60–70% of the nut of which 30–35% is oil. Three principal processes are commercially used for

160

Green composites

the extraction of CNSL: hot oil bath, roasting and solvent extraction process. In the most common method of commercial extraction, the hot oil process, CNSL is extracted by passing the raw nuts through a bath of hot CNSL heated at 180–200 ∞C (Menon et al., 1985). The oil in the shell oozes out into the bath. Fifty per cent of the oil can be extracted by using this method, which can be further improved by first subjecting the dry nuts to surface wetting and then passing them through the hot oil. The excess moisture content of about 7–10% of the weight of the nuts causes the cells to burst, with the result that the oil oozes into the bath. The roasting process involves subjecting the nuts to a sudden change in temperature, from room temperature to that necessary to cause charring. This creates an explosive pressure within the cellular structure of the shell, which forces the oil to ooze out. In the solvent extraction process the nut is sliced and the kernel is removed. The shell fragments are crushed and passed through organic solvents such as benzene, dissolving the CNSL into solution. This method is highly efficient and up to 90% of the CNSL in the shell can be extracted (Bisanda, 1991). CNSL has found numerous applications, and hundreds of patents have been published. Their main applications include the manufacture of friction linings, varnishes and paints, laminating and casting resins, surfactants, etc. Because of their phenolic character and the long alkyl chain, which varies in its degree of unsaturation attached to the benzene nucleus of their molecules, the chief constituents of CNSL find many applications in several other fields (Pansare and Kulkarni, 1964) including natural fibre composites. The main constituents of CNSL are anarcadic acid, cardanol and cardol. Other constituents include 2-methyl cardol and a small amount of polymeric materials. CNSL extracted by cold-solvent method is called natural CNSL, whereas hot oiland roasting-processed CNSL is called technical CNSL (Mwaikambo, 2002) (Table 8.2). In the extraction processes involving heat, anarcardic acid is usually decarboxylated into cardanol (Pansare and Kulkarni, 1964) and hence cardanol is the main constituent of technical CNSL. All the constituents of CNSL are phenolic compounds and the presence of hydroxyl (-OH) group, carboxyl (-COOH) group, and variable aliphatic Table 8.2 Phenol components (%) in natural and technical CNSL Components

Natural CNSL (%)

Technical CNSL (%)

Anacardic acid Cardanol Cardol 2-Methyl cardol Polymers by difference

77.0 2.37 16.8 2.8 –

– 82.2 13.7 4.1

Sources: Menon et al. (1985), Gedam and Sampathkumaran (1986).

Optimising the properties of green composites

161

unsaturation in its side chain makes CNSL able to take part in several chemical reactions. CNSL can be polymerized in a variety of ways (Pillai et al., 1980) including addition and condensation polymerisation. In the condensation polymerisation process, CNSL can be condensed with active hydrogencontaining compounds such as formaldehyde at the ortho and para positions of the phenolic ring under acidic or alkaline conditions to yield a series of polymers of ‘novolac’ or ‘resole’ type (Menon et al., 1985). This method is used in the polymerisation process of CNSL in this work. Novolacs are prepared by reacting an excess of phenol with formaldehyde with a typical ratio of phenol to formaldehyde of 1.25 : 1. Novolacs are twostep resins, in that they require the addition of a cross-linking agent and heat to achieve curing. Resoles are referred to as one-step resin because they can be cured by heat alone and do not require any cross-linking agent. Phenols react with excess formaldehyde in an alkaline medium in the preparation of resole resins and typical ratios of phenol to formaldehyde are 1 : 1.5 and 1 : 3. An extensive study was carried out by Misra and Pandey (1984) to investigate the effect of various parameters on the polymerisation of cardanol, e.g. cardanol–formaldehyde molal ratio, catalyst (NaOH), concentration and temperature.

8.3

Selecting natural fibres for composites: stress transfer and physical characteristics

Composites reinforced with synthetic fibres have been thoroughly described in the literature, e.g. books by Harris (1999) and Hull and Clyne (1996). Predictive equations for strength, stiffness and density have been developed in the form of rules of mixtures and shear lag theory for stress transfer between matrix and solid fibre has highlighted the key role of the bond between fibre and matrix. In general a strong interfacial bond improves the composite strength and stiffness but reduces the work of fracture in impact.

8.3.1

Stress transfer by hollow fibre bundles versus conventional solid fibres

Natural plant fibres are known to comprise either single hollow ultimate cells (e.g. cotton, kapok) or, more commonly, multicellular, parallel arrays of ultimate fibres in bundles (e.g. jute, hemp and kenaf) (Fig. 8.2). The major chemical constituents of these plant fibres are cellulose (‘fibre’ component of the cell wall), hemicellulose and lignin (‘matrix’ components of the cell wall). The hemicellulose is a mostly amorphous polymer whilst the lignin is completely amorphous. The outer surface of plant fibre bundles will also contain waxes, fats and pectins which do not bond well to synthetic polymer matrices. Cellulose has a Young’s modulus ~130 GPa (Bodig and

162

Green composites

8.2 Section through kenaf fibre bundles (¥ 350) (image supplied by Peter Rendle, University of Bath).

Jayne, 1982) and in order to maximise the stiffness of the natural fibre composite it is important to select fibres where the spirally disposed cellulose in the cell wall has a low winding angle q with respect to the cell axis (Fig. 8.3). Not only is the stiffness of the cellulose in the plant cell wall diluted by the off-axis disposition of the fibres but it is also diluted by the ‘matrix’ components of the cell wall and, when incorporated into a plant fibre composite, by the alignment of fibres in the composite and the resin matrix content of the composite. In order to maximise the composite strength and stiffness and in order to transfer stress effectively between natural fibres and resin matrix the following factors should be considered. Lumen

S3 S2 S1

q

P

Pectin, waxes, fats

8.3 General features of morphological structure of the cell wall of an ultimate fibre showing the microfibril angle q in the dominant S2 layer.

Optimising the properties of green composites

163

∑ Winding angle q of cellulose in ultimate plant cell with respect to fibre axis should be small. ∑ Fibres should contain a high cellulose content. ∑ The plant fibre should be surface treated to ensure compatibility between fibre and matrix by developing a good interfacial bond. ∑ Matrix should be a thermosetting polymer to prevent plastic deformation in the matrix at low stresses. ∑ Fibres should be disposed in a direction parallel to an applied, uniaxial stress but should be cross-laminated, rather than randomised, if the stresses are bidirectional. Whilst the hollow nature of natural fibres reduces their apparent density, it is clear that ultimate fibres with large lumens may experience distortion or even collapse when they are processed into plant fibre composites, especially by processes such as hot pressing. The exact nature of the interaction between plant fibre and polymer matrix is therefore complex and all stages of the manufacture of plant fibre composites will influence the resulting properties of the composites.

8.3.2

Physical characteristics of natural fibres

The characteristics of selected plant fibres are presented in Table 8.3. In practice, all properties vary and will depend on the source of fibre, the position of the fibre in the plant and conditions of growth. Specific properties are the property (strength or modulus) divided by specific gravity. It can observed that the mechanical properties of the cotton and coir seed fibres are disadvantaged by their high microfibril angles. All of the bast fibres are, in fact, multicellular fibre bundles with low microfibril angles. Table 8.3 Physical and mechanical properties of bast (b), leaf (l) and seed (s) fibre Fibre type

Apparent density (kg/m3)

Tensile strength (MPa)

Specific strength (MPa)

Young’s modulus (GPa)

Specific modulus (GPa)

Microfibril angle (q)

Flax (b) Hemp (b) Jute (b) Kenaf (b) Banana (l) Pineapple (l) Sisal (l) Cotton (s) Coir (s)

1500 1500 1500 1200 1350 1440 1450 1550 1150

500–900 310–750 200–450 295–1191 529–914 413–1627 80–840 300–700 106–175

345–620 210–510 140–320 246–993 392–677 287–1130 55–580 194–452 92–152

50–70 30–60 20–55 22–60 27–32 60–82 9–22 6–10 6

34–48 20–41 14–39 18–50 20–24 42–57 6–15 4–6.5 5.2

5 6.2 8.1 – 11–12 6–14 10–22 20–30 39–49

Adapted from Mwaikambo (2002).

164

8.4

Green composites

Case study: natural fibre composites with thermosetting resin matrices

In order to demonstrate the various factors described above, which influence the composite properties, natural fibre composites were manufactured and tested with matrices comprising either a standard laminating polyester resin or CNSL resin. Hemp and kenaf bast fibres were utilised as they represent fibres grown in the UK and the Far East, respectively, each possessing very good mechanical properties.

8.4.1

Fibres and polymer resin matrices

The Hemcore Company Limited of the UK supplied hemp fibre used in this work and the Araco Corporation of Japan supplied Chinese kenaf fibre. Fibre was provided as unidirectional tow and random mat. Some short fibres were also cut from tow for the manufacture of randomly orientated fibre reinforcement. The Tanzanian Italian Company Limited of Tanzania provided cashew nut shell liquid (CNSL) and the Scott Bader Company Limited (Wollaston, UK) supplied the four types of polyester resins used in this work. Sodium hydroxide pellets of 98% strength and formaldehyde solution of about 40% weight of solid per unit volume were supplied as general laboratory reagents for the surface treatment of fibres. The Chemical Release Company Limited (Harrogate, UK) supplied PAT 607/PCM mould release agent. Resins selected in the first stage of this study were a conventional unsaturated polyester resin in styrene monomer, produced by Scott Bader Ltd with the product name Crystic 2-406PA, and resin formulated from cashew nut shell liquid. Polyester resins were prepared by combining 100 ml of the resin (unsaturated polyester resin in monomer) with 1 ml of methyl ethyl ketone peroxide in dimethyl phthalate (catalyst). The mixture was stirred and left to stand until a colour change was observed. The mixture was then ready to be used for composite manufacture. CNSL resins were prepared by combining 75 ml of CNSL with 100 ml of formaldehyde (40%) and 10 ml of 2M NaOH solution (alkaline catalyst). The mole ratio of CNSL and formaldehyde used was 1 : 1.33 (CNSL:H2CO). The mixture was stirred and left to settle for a few minutes. The excess water at the bottom of the beaker was removed by using a pipette.

8.4.2

Chemical treatment of fibres

Hemp and kenaf fibres were either washed and dried and otherwise left untreated or alkalised. In the latter case fibres were soaked in 6% (0.06 M) NaOH solution in a water bath where the temperature was maintained

Optimising the properties of green composites

165

throughout at 19 ∞C ± 2 ∞C for 48 h. The fibres were rinsed and left to dry at room temperature before being placed in an oven for 5 h at 110 ∞C. The treated short fibres were chopped into 15 mm lengths after alkali treatment.

8.4.3

Manufacture of composites

Composites were made using a stainless steel mould measuring 240 ¥ 60 ¥ 40 mm (length, width and depth, respectively). The PAT 607/PCM release agent was sprayed onto a laboratory tissue and smeared evenly onto the surface of the mould. Resin was poured onto each layer of fibre in a zig-zag configuration to ensure even delivery of the resin and the procedure was repeated for each layer of fibre. About six to seven layers of fibre were aligned in the mould and resin was poured onto each layer except for the uppermost layer. The layers of the wetted fibres in the mould were then placed between the electrically heated platens of a hot press at the desired temperature. The mould was heated initially for 5 minutes without applying pressure (preheating) and then pressure was gradually increased for the time and pressure required before the composites were taken out of the mould. The pre-heating was carried out to allow the resin to become more viscous before the full pressure was applied. For the fibre–polyester composite, post-curing of the composites was carried out in an oven at 80 ∞C overnight whereas, for the fibre–CNSL composites, post-curing was performed at 100 ∞C for 24 h. Pressing conditions for polyester resin matrix composites were a hot press temperature of 50 ∞C, a pressing time of 25 minutes (pre-heating and heating) and pressures of 60 bar (6 MPa) and 80 bar (8 MPa). Pressing time and conditions for fibre–CNSL resin composites were a hot press temperature of 180 ∞C, a pressing time of 25 minutes (pre-heating and heating) and a pressure of 60 bar (6 MPa). Different pressures were used to evaluate the effect of pressure on the properties of hemp–polyester composites. It should be noted that the aim of the manufacturing process was to maximise fibre content without ‘drying out’ the composite through insufficient wetting of fibres.

8.5

Mechanical properties of composites as a function of design

The coding system assigned to composites was as follows: H = hemp, K = kenaf, U = untreated, T = treated, L = long fibres, S = short fibres, RM = random mat fibres, P = polyester resin, CNSL = cashew nut shell liquid resin.

166

Green composites

Hence, for example: ∑ UL–KP = natural fibre composite reinforced with untreated long kenaf fibres in a polyester resin matrix ∑ TRM–HCNSL = natural fibre composite reinforced with treated random mat hemp fibres in a cashew nut shell liquid resin matrix. The strength, stiffness and toughness of composites with long fibre and random mat reinforcement, with treated (alkalised) and untreated fibres, are reported below. Flexural strength and flexural stiffness were measured in three-point bending and toughness (or work of fracture) was measured by Charpy impact of unnotched material.

8.5.1

Effect of fibre treatment on properties of long fibre composites

Figure 8.4 presents the flexural moduli of treated and untreated long fibre HP, HCNSL, KP and KCNSL composites (corresponding to codes UL-HP, TL-HP, UL-HCNSL, TL-HCNSL, UL-KP, TL-KP, UL-KCNSL and TLKCNSL). From the histogram it can be seen that the treated fibres for all of the composite types possess superior flexural modulus of elasticity values compared to the untreated fibre composites. The flexural strengths of the treated and untreated long fibre HP, HCNSL, KP and KCNSL composites are presented in Fig. 8.5. Again, the treated 25

2 standard deviations

Flexural modulus (GPa)

20

15 64%

10 65% 5

64%

61%

66% 60%

0

66% HP

67%

HCNSL KP Composite types Treated

KCNSL

Untreated

8.4 Flexural moduli of treated and untreated long fibre hemp and kenaf composites. The values in each bar correspond to the volume fraction of fibre.

Optimising the properties of green composites 250

167

2 standard deviations

Flexural strength (MPa)

200

150 64%

100 65% 50

61%

64% 66%

60% 66%

0 HP

67%

HCNSL KP Composite types Treated

KCNSL

Untreated

8.5 Flexural strengths of treated and untreated long fibre hemp and kenaf composites. The values in each bar correspond to the volume fraction of fibre.

fibres gave superior results compared to the untreated fibre composites. KCNSL composites possessed the highest flexural strength compared to the other treated fibre composites. The work of fracture values for long fibre composites are presented in Fig. 8.6. Here the trends are not quite so obvious but there is a tendency for the untreated fibre composites to be tougher in impact than the caustic soda treated fibre composites. This observation has often been reported in the literature and confirms the trend that poorly bonded interfaces act as crack stoppers but reduce the ability of load to be transferred by shear between the fibre and matrix. In a treated fibre composite, the interfacial friction stress between the matrix and fibre is generally higher when compared to the untreated ones and this causes a drop in toughness. Impact strength can be improved by reducing the friction stress between the fibre and the matrix in a controlled manner. According to the following equation (Kelly and Tyson, 1965) the work of fracture of fibre composites:

gf =

Vf s f u lc 24

where

s fur t g f = work of fracture lc =

168

Green composites

120

2 standard deviations

Work of fracture (kJ/m2)

100

80

60 64% 60%

40 61% 20

0

66% 67%

64%

65% 66%

HP

HCNSL KP Composite types Treated

KCNSL

Untreated

8.6 Work of fracture of treated and untreated long fibre hemp and kenaf composites. The values in each bar correspond to the volume fraction of fibre.

Vf = volume fraction of fibre sfu = fibre strength lc = fibre critical length r = fibre radius t = interfacial friction stress. Essentially a low interfacial friction stress (weak binding between fibre and matrix) results in a high critical length and high work of fracture. For example, taking values of lc = 2 ¥ 10–3 m, Vf = 0.6, sfu = 1 ¥ 109 N/m2, the work of fracture, gf = 50 kJ/m2. The lc value is based on an assumption that fibre radius = 0.2 ¥ 10–3 m and interfacial stress, t = 10 MPa. This theoretical value agrees well with values for the treated HP and KCNSL composites. It can therefore be seen that high Vf, high sfu, high lc and low t improve toughness. The lc term is increased by reducing the frictional stress t. The toughening effect is considerably greater when the fibre surfaces are clean than when they are coated or treated, which is a normal practice for producing a good fibre–matrix bond (Harris, 1999). Oever et al. (1999) reported that the Charpy impact strength decreased with increasing fibre internal bonding and enhanced fibre–matrix adhesion. This agrees with many of the results reported in this chapter. Oever et al. (1999) also suggested that the decrease in impact strength could be explained by assuming that a high level of fibre–matrix adhesion results in shorter average pull-out lengths and therefore causes lower impact strengths. A precondition for this

Optimising the properties of green composites

169

hypothesis is that the fibres are longer than the critical fibre length lc. From the histograms presented above, the small changes in Vf do not seem to have much effect on toughness.

8.5.2

Effect of fibre treatment on properties of short fibre composites

Mechanical test results for short and random-mat fibre composites (corresponding to codes US-HP, TS-HP, URM-HCNSL, TRM-HCNSL, URMKP, TRM-KP, URM-KCNSL and TRM-KCNSL) are presented below. Flexural modulus results are compared in Fig. 8.7 and the same pattern is observed as in Fig. 8.4 where the treated short fibre composites show superior flexural modulus values compared to the untreated composites. However, the flexural modulus of long fibre composites is much higher than for short fibre composites due to the random orientation of the fibres corresponding to the trends in Table 8.1. Short fibre flexural modulus results are closer to the theoretical Rule of Mixtures values when compared to the long fibres. The Rule of Mixtures for the modulus of elasticity Ec of composites predicts that: E c = h q Ef V f + E m V m where Ef and Em are the fibre and matrix moduli and Vf and Vm are the fibre 9

Flexural modulus (GPa)

8

2 standard deviations

7 6 5 4 3

63%

51% 70%

2

55%

63%

66%

55%

1 56% 0 HP

HCNSL KP Composite types Treated

KCNSL

Untreated

8.7 Flexural moduli of treated and untreated short and random-mat fibre hemp and kenaf composites. The values in each bar correspond to the volume fraction of fibre.

170

Green composites

100 90

2 standard deviations

Flexural strength (MPa)

80 70 60 50

30 20

51% 55%

70%

66% 56%

63%

10 0

63%

55%

40

HP

HCNSL KP Composite types Treated

KCNSL

Untreated

8.8 Flexural strengths of treated and untreated short and random-mat fibre hemp and kenaf composites. The values in each bar correspond to the volume fraction of fibre.

and matrix volume fractions, respectively. The orientation factor (or Krenchel coefficient) hq is equal to 1 for unidirectionally oriented fibres. For short fibre composites, taking values of hq = 0.25 (for random oriented fibre), Vf = 0.6 and Ef = 45 GPa (modulus of hemp fibre according to Mwaikambo, 2002), the predicted flexural modulus of the composite, Ec = 6.8 GPa. The modulus values obtained for the short fibre hemp composites are within 4 to 5 GPa at Vf values of between 0.5 to 0.7 (Fig. 8.7). The measured values obtained are quite close to the theoretical values in this case. Also, for the short fibre composites, the fibre volume fraction does not significantly influence the flexural modulus as it does for the long fibre composites. In Fig. 8.8, flexural strengths are displayed for the kenaf- and hempreinforced composites and the property trends are very similar to those in Fig. 8.7 for the flexural modulus. Work of fracture values for the short and random-mat fibre composites are seen in Fig. 8.9. For HP and HCNSL composites the untreated fibre composites are tougher than treated fibre composites but the KP and KCNSL treated fibre composites are tougher than their untreated counterparts. There is not a big difference in the work of fracture values between the short and random-mat fibre composites (Fig. 8.9) and the long fibre composites (Fig. 8.6). However, the flexural modulus and flexural strength values are significantly different. For all of the mechanical properties displayed here in Fig. 8.4 to 8.9, the magnitude of the standard deviation bars should be noted. The scatter is

Optimising the properties of green composites

171

80

Work of fracture (kJ/m2)

70

2 standard deviations

60 50 40 30

55% 55%

20 10 0

51%

HP

63%

56% 63%

70% HCNSL KP Composite types Treated

66%

KCNSL

Untreated

8.9 Work of fracture of treated and untreated short and random-mat fibre hemp and kenaf composites. The values in each bar correspond to the volume fraction of fibre.

particularly high for work of fracture results, as expected, but the static moduli and strengths also vary significantly and this variability poses problems for the acceptability of natural fibre composites in engineering applications.

8.5.3

Effect on properties of composites of chemical modifications to polyester resin

The sustainability of conventional glass reinforced plastics (GRPs) may be improved by replacing glass fibres with natural fibres. The performance of natural fibre composites with a polyester resin matrix (resin A) was examined in Section 8.5.1. The effects on composite performance of chemically treating fibres with a caustic soda solution was demonstrated. A further route to improving composite performance is to modify the chemistry of the polymer. Three additional unsaturated polyester resins (B, C and D) were specifically formulated by Scott Bader Ltd for use with natural fibres. Resin B has a polymer structure modified to make it more polar in nature, making it more hydrophilic, so that it can interact or bond better with the OH-groups on the surface of the natural fibres. Resin C contains an additional monomer besides styrene. This monomer contains polar groups designed to interact with, or possibly to bond to, the OH-groups on the surface of the natural fibres. Resin D also contains an additional monomer designed to improve interaction with natural fibres. The flexural modulus, flexural strength and work of fracture of treated

172

Green composites

long kenaf fibre composites were measured for A, B, C and D polyester matrices, corresponding to codes TL–KP(A), TL–KP(B), TL-KP(C) and TL– KP(D). Treated long kenaf fibre was selected for this work because alignment of kenaf fibre during the manufacture of composites was better compared to hemp fibre and treated fibres gave superior flexural modulus and strength results in the work reported above. From the overall results for flexural modulus (Fig. 8.10) and flexural strength (Fig. 8.11), for the long kenaf–polyester composites, it can be seen 45 40

2 standard deviations

Flexural modulus (GPa)

35 30 56% 25 20

58%

15 10

62% 63%

5 0 A

B

C

D

Polyester resin

8.10 Flexural moduli of treated long kenaf–polyester A, B, C and D composites. The values in each bar correspond to the volume fraction of fibre. 500 450

2 standard deviations

Flexural strength (MPa)

400 350 300

56%

250 62% 200 58%

150 100 50

63%

0 A

B

C

D

Polyester resin

8.11 Flexural strengths of treated long kenaf–polyester A, B, C and D composites. The values in each bar correspond to the volume fraction of fibre.

Optimising the properties of green composites 120

173

2 standard deviations

Work of fracture (kJ/m2)

100

80

60

63% 56%

62%

40

58%

20

0 A

B

C

D

Polyester resin

8.12 Work of fracture of treated long kenaf fibre–polyester A, B, C and D composites. The values in each bar correspond to the volume fraction of fibre.

that the modification of the polymer structure of polyester B and the additional monomer added to polyester C and D made a positive impact on the strength of the composites. Matrix B composites achieved an excellent mean flexural stiffness of 35 GPa, close to that of GFRP, and a mean flexural strength of 380 MPa. Treated fibres were used in this experiment and the bonding and mechanical adhesion between the B, C and D matrices and the kenaf fibres were superior to polyester A. It is no surprise to note that the work of fracture (Fig. 8.12) of the composite with matrix A has the highest mean value, approaching 90 kJ/m2, compared with about 60 kJ/m2 for matrix B. The interfacial bond between fibre and matrix is therefore a deciding factor in determining stiffness, strength and work of fracture. The most compelling outcome of modifying the resin chemistry is the big improvements in stiffness and strength in flexure. Attention should be focused on resin modification as well as fibre treatment for optimising the performance of natural fibre composites.

8.6

Dynamic mechanical thermal analysis (DMTA) of long fibre composites

DMTA is a sensitive technique that characterises the mechanical responses of a material by monitoring dynamic property changes over a range of temperatures at a fixed frequency or over a range of frequencies at a fixed

174

Green composites

temperature. The technique separates the dynamic modulus ΩEΩ of materials into two distinct parts, namely an elastic (storage) part (E¢) and a viscous (loss) component (E≤). The elastic storage modulus E¢ is the component of the dynamic modulus ΩEΩ where the strain is in phase with the applied stress and the loss modulus E ¢¢ is the component of the dynamic modulus ΩEΩ where the strain is 90∞ out of phase with the applied stress. The ratio of E≤ to E¢ gives the tangent of the phase angle d and tan d is known as the damping and is a measure of energy dissipation. Such parameters provide quantitative and qualitative information about the behaviour of materials. Dynamic mechanical methods are the most sensitive way of measuring the glass transition temperature (Tg), which is one of the key properties of a polymer from both structural and processing viewpoints. The construction for determining T g from the storage modulus E¢ versus temperature characteristic (intersecting tangents to elastic and viscoelastic parts of the characteristic) is illustrated in Fig. 8.1. It has also been reported that values for the glass transition temperature Tg measured by differential scanning calorimetry (DSC) and DMTA generally agree within ±4 ∞C for a wide variety of commercially available polymers (Ferrillo and Achorn, 1997). The dynamic properties of composites were measured using a Tritec 2000 DMTA. The samples were cut to size using a diamond cutter. A single cantilever clamping fixture was used for testing with a span of 16 mm and the sample width and depth were approximately 9 mm and 3 mm, respectively. Dynamic properties were measured at a fixed frequency of 1 Hz in the temperature range from 30 ∞C to 180 ∞C at a heating rate of 2 ∞C/min–1. The storage modulus (E¢), loss modulus (E≤) and tan d of the samples were recorded during the run. The effect of fibre treatment, with 6% NaOH solution, on the dynamic properties of kenaf–polyester A composites is examined here. Storage modulus values for untreated and treated long kenaf–polyester A composites are compared in Fig. 8.13. The E¢ values of the composites fall steeply around the glass transition temperature. The treated fibre composite E¢ value is higher than the untreated value and the degree of modulus loss at temperatures greater than Tg is also less than the untreated fibre composite. The higher E¢ value of the treated long kenaf–polyester composite is due to greater interfacial adhesion and bond strength between matrix and fibre as reported by several authors (Nair et al., 2001; Ray et al., 2002; Saha et al., 1999). The tan d peak can be related to impact properties of a material (Saha et al., 1999). From Fig. 8.14 it can be observed that the untreated long kenaf– polyester composite gives a higher tan d peak compared to the treated fibre composite. The magnitude of the tan d peak corresponds to the interfacial adhesion between the treated fibres and matrix. Good interfacial bonding corresponds to inferior impact properties and reduced damping.

Optimising the properties of green composites

175

1.8 ¥ 109 1.6 ¥ 109

TLKP

Storage modulus (Pa)

1.4 ¥ 109 1.2 ¥ 109

ULKP

1.0 ¥ 109 8 ¥ 108 6 ¥ 108 4 ¥ 108 2 ¥ 108 0 30.0

50.0

70.0

90.0

110.0

130.0

150.0

Temperature (∞C)

8.13 Variation of storage modulus of untreated (ULKP) and treated (TLKP) long kenaf–polyester composites as a function of temperature. 0.200 ULKP

0.180 0.160 0.140

TLKP

Tan d

0.120 0.100 0.080 0.060 0.040 0.020 0.000 30.0

50.0

70.0

90.0 110.0 Temperature (∞C)

130.0

150.0

8.14 Variation of tan d of untreated (ULKP) and treated (TLKP) long kenaf–polyester composites as a function of temperature.

The mechanical properties of these composites are presented in Figs. 8.4, 8.5 and 8.6. The mean flexural modulus (Fig. 8.4) and strength (Fig. 8.5) of the treated fibre composites is much higher than for the untreated fibre composites yet the work of fracture (Fig. 8.6) is similar. For other fibre– matrix combinations (e.g. HP, KCNCL) the work of fracture of the untreated

176

Green composites

fibre composites is higher than for the treated fibre composites. In conclusion, it is clear that by chemically fine tuning the interface the spectrum of mechanical properties can be optimised to produce stiff, strong and tough natural fibre composites.

8.7

Environmental stability of natural fibre composites

Mechanical tests on polyester A, B, C and D composites showed that polyester B matrix composites were stiffest and strongest (Figs. 8.10 and 8.11) and polyester A matrix composites were the least stiff and strong. The polymer structure of polyester B contains more polar groups and was able to interact better with the OH groups present on the surface of natural fibres. DMTA analysis of the polyester composites, not reported here, also demonstrated that the interfacial adhesion between polyester B and the treated long kenaf fibre was superior to the other polyester resin–matrix composites. The environmental stability of natural fibre composites is a key issue of concern to industry and Zadorecki and Flodin (1985) state that strong interfacial adhesion, low water absorption and good dimensional stability are desirable properties for cellulose–polyester composites. For this reason a severe boiling water test was devised and applied to treated long fibre kenaf composites with polyester A and B matrices in order to observe and compare the physical effects of immersion in boiling water on these composites. Figure 8.15 shows the average percentage weight increase of the composites as a function of time. 80

Weight increase (%)

70

TLK-polyester A

60 50 40 30 TLK-polyester B 20 10 0 0

5

10

15

20

25

Time (h)

8.15 Average percentage weight increase (water absorption) of TLK– polyester A and TLK–polyester B composites against time following immersion in boiling water.

30

Optimising the properties of green composites

177

The increase in weight due to water absorption in the TLK–polyester A composite was very significant. After one hour in boiling water, the percentage weight gained was 58%. The highest percentage weight gain of 73.3% was recorded after six hours after which specimens began to break up. The initial average percentage weight gain for the TLK–polyester B composites was observed at 7.3% after one hour and a steady highest level of 20.1% was recorded after 14 hours of refluxing in boiling water. Specimens remained largely dimensionally stable. The percentage weight gain due to water absorption was small but significant when compared to the TLK–polyester A composites in this short-term test. The low average weight gain in polyester B matrix demonstrates the superior interfacial adhesion in these composites.

8.8

Discussion and conclusions

A number of factors emerge from this paper which are summarised in the conclusions below. As well as mechanical properties, thermal stability and environmental performance there are additional factors which affect the degree to which natural fibres are likely to be taken up by industry. Manufacturing of long fibre composites is a key issue. GFRPs represent by far the largest proportion of fibre composites manufactured industrially (Bader, 2000). The textile industry is geared up to produce single glass filaments in multikilometre lengths, to spin glass yarns, to weave and knit glass fabrics and to produce dry stitch-bonded preforms and resin-impregnated preforms or prepregs. Similar facilities are necessary for the whole spectrum of plant fibres, in order to achieve the highest performance levels from plant fibre composites. Natural fibres such as cotton and flax are routinely woven by the textile industry but further advances must be made to improve the availability of plant fibres in forms that are easy to process into composites. Experimental work reported above is based on hot-pressed composites but industry requires low pressure processes such as vacuum forming and resin transfer moulding where the performance of natural fibre composites is not expected to be so good because of the problems associated with densifying high volume, low density plant fibres. Processing temperatures for CNSL are particularly high and reducing polymerisation temperatures for green thermosetting resins is a major challenge for polymer chemists. In order to break into markets dominated by GFRP the environmental stability of natural fibre composites must be improved. Whilst the properties of modified polyester resin B composites are impressive in a boiling water test, they are not sufficiently good for extreme outdoor exposure applications such as moulded boat hulls. Where natural fibre composites are used in indoor or in-vehicle applications, their environmental stability is less critical. However for future acceptability by industry, environmental performance must be improved.

178

Green composites

∑ Natural fibres are generally composed of clusters of hollow ultimate cells arranged as fibre bundles. ∑ Strong, stiff ultimate fibres contain cellulose microfibrils oriented at a small angle to the fibre axis in the cell wall, e.g. kenaf and hemp fibres. ∑ The stiffest, strongest composites are based on unidirectionally aligned natural fibre bundles impregnated with synthetic or natural resins. ∑ Thermally stable, non-yielding composites require thermosetting rather than thermoplastic matrices. ∑ Modification of the fibre to matrix interface to improve interfacial bonding enhances the stiffness and strength of composites to the detriment of toughness. ∑ Cashew nut shell liquid resin is a sustainable matrix for natural fibre composites but processing temperatures are high. ∑ Mechanical properties of some plant fibre composites approach those of GFRPs. ∑ The mechanical properties and the environmental stability of natural fibre composites may be improved by introducing more polar groups into the resin.

Sources of further information and advice Readers are referred to papers presented at ECOCOMP-1 (September 2001) and ECOCOMP-2 (September 2003) at Queen Mary University of London. See: www.materials.qmw.ac.uk/ecocomp/ The journals Composites Science and Technology and The Journal of Applied Polymer Science publish many papers in the field of natural fibre composites.

Acknowledgements S.H. Aziz is grateful to the Malaysian Government and the Malaysian Institute of Nuclear Technology Research (MINT) for sponsoring her research at the University of Bath. Scott Bader Ltd, Wollaston, UK helped to fund the research and the authors thank Dr Simon J. Clarke for his enthusiastic contribution to the work. We acknowledge our productive associations with Dr Jalaluddin Harun (Universiti Putra Malaysia), Dr Khairul Zaman Hj Mohd Dahlan (MINT) and Dr Takuya Nishimura (Araco Corp).

References Bader, M.G. (2000). Polymer composites in 2000: structure, performance, cost and compromise. Presented at Microstructure of Composite Materials – V, St John’s College, Oxford, 3rd and 4th April 2000, Royal Microscopical Society.

Optimising the properties of green composites

179

Bisanda, E.T.N. (1991). Sisal fibre reinforced composites. PhD thesis, Department of Engineering and Applied Science, University of Bath, UK. Bisanda, E.T.N. and Ansell, M.P. (1992). Properties of sisal–CNSL composites. Journal of Materials Science, 27, 1690–700. Bodig, J. and Jayne, B.A. (1982). Mechanics of Wood and Wood Composites. New York: Van Nostrand Reinhold. Devi, L.U., Bhagawan, S.S. and Thomas, S. (1997). Mechanical properties of pineapple leaf fibre-reinforced polyester composites. Journal of Applied Polymer Science, 64, 1739–48. Felix, J.M., Gatenholm, P. and Schreiber, H.P. (1994). Plasma modification of cellulose fibres: effect on some polymer composite properties. Journal of Applied Polymer Science, 51, 285–95. Ferrillo, R.G. and Achorn, P.J. (1997). Comparison of thermal techniques for glass transition assignment. II. Commercial polymers. Journal of Applied Polymer Science, 64, 191–5. Gedam, P.H. and Sampathkumaran, P.S. (1986). Cashew nut shell liquid; extraction, chemistry and applications. Progress in Organic Coatings, 14, 115–57. Geethamma, V.G., Joseph, R. and Thomas, S. (1995). Short coir fibre-reinforced natural rubber composites: effects of fibre length, orientation and alkali treatment. Journal of Applied Polymer Science, 55, 583–94. Harris, B. (1999). Engineering Composite Materials. 2nd edn., Cambridge, UK: Cambridge University Press. Herrera-Franco, P.J. and Aguilar-Vega, M.D.J. (1997). Effect of fibre treatment on the mechanical properties of LDPE–henequen cellulosic fibre composites. Journal of Applied Polymer Science, 10, 197–207. Hull, D. and Clyne, T.W. (1996). An Introduction to Composite Materials. 2nd edn., Cambridge Solid State Science Series. Kelly, A. and Tyson, W.R. (1965). Tensile properties of fibre-reinforced metals – copper/ tungsten and copper/molybdenum. Journal of the Mechanics and Physics of Solids, 13, 329–50. Menon, A.R.R., Pillai, C.K.S., Sudha, J.D. and Mathew, A.G. (1985). Cashew nut shell liquid – its polymeric and other industrial products. Journal of Scientific and Industrial Research, 44, 324–38. Misra, A.K. and Pandey, G.N. (1984). Kinetics of alkaline-catalyzed cardanol – formaldehyde reaction 1. Journal of Applied Polymer Science, 29, 361–72. Mwaikambo, L.Y. (2002). Plant-based resources for sustainable composites. PhD thesis, Department of Engineering and Applied Science, University of Bath, UK. Mwaikambo, L.Y. and Ansell, M.P. (2002). Chemical modification of hemp, sisal, jute and kapok fibres by alkalization. Journal of Applied Polymer Science, 84 (12), 2222– 34. Nair, K.C.M., Thomas, S. and Groeninckx, G. (2001). Thermal and dynamic mechanical analysis of polystyrene composites reinforced with short sisal fibres. Composites Science and Technology, 61, 2519–29. Oever van den, M.J.A., Bos, H.L. and Molenveld, K. (1999). Flax fibre physical structure and its effect on composite properties: impact strength and thermo-mechanical properties. Die Angewandte Makromolekulare Chemie, 272, 71–6. Pansare, V.S. and Kulkarni, A.B. (1964). Azo dyes from cashew nut shell liquid derivatives. Journal of the Indian Chemical Society, 41 (4), 251–5. Pillai, C.K.S., Prasad, S.V. and Rohatgi, P.K. (1980). To evaluate the polymerization characteristics of cashew nut shell liquid obtained from the expeller, Report No. RRL/ M/27/79, Materials Division, Regional Research Laboratory, Trivandrum.

180

Green composites

Ray, D., Sarkar, B.K., Das, S. and Rana, A.K. (2002). Dynamic mechanical and thermal analysis of vinylester-resin-matrix composites reinforced with untreated and alkalitreated jute fibres. Composites Science and Technology, 62, 911–17. Saha, A.K., Das, S., Bhatta, D. and Mitra, B.C. (1999). Study of jute fibre reinforced polyester composites by dynamic mechanical analysis. Journal of Applied Polymer Science, 71, 1505–13. Spear, M.J. (2002). Transcrystalline interphases in polypropylene composites. PhD thesis, University of North Wales, Bangor. Sreekala, M.S., Kumaran, M.G. and Thomas, S. (1997). Oil palm fibres: morphology, chemical composition, surface modification and mechanical properties. Journal of Applied Polymer Science, 66, 821–35. Tyman, J.H.P. (1979). Non-isoprenoid long chain phenols. Chemical Review Society, 8, 499–537. Zadorecki, P. and Flodin, P. (1985). Surface modification of cellulose fibres: 11. The effect of cellulose fibre treatment on the performance of cellulose – polyester composites. Journal of Applied Polymer Science, 30, 3971–83.

9 Green fibre thermoplastic composites M. SAIN AND S. PANTHAPULAKKAL University of Toronto, Canada

9.1

Introduction: biofibre production

Ecological concerns such as recyclability and environmental safety have resulted in a renewed interest in natural fibre composites. Natural fibre composites are undergoing a high tech revolution and are replacing conventional composites in high performance applications due to their advantages over conventional reinforcements. The annual global disposal of millions of tonnes of plastics, especially from packaging, has raised the demand for means of managing this non-biodegradable waste. The use of biofibres in a thermoplastic matrix provides positive environmental benefits with respect to disposability and raw material utilisation. Since biofibres are relatively less expensive and biodegradable, biocomposites from biodegradable polymers will render a contribution in the twenty-first century. Based on the material resource, natural fibre-thermoplastic composites can be classified into two types. First, composites with non-degradable synthetic thermoplastics: although these cannot undergo biodegradation, they can be recycled easily compared with conventional thermoset composites. Secondly, composites with a biodegradable polymer matrix, the so called green composites: these can undergo complete biodegradation. Biofibres that are useful for manufacturing composites are obtained from two major sources: forest floor and agro residues. Unlike wood fibres, agro fibres undergo several transformation processes before they are ready for use in composite manufacturing. The following processes are discussed in the literature to develop a good quality fibre.

9.1.1

Retting

Retting is a preferential rotting process, which separates the fibres from the woody stem by removing the non-fibrous cementing materials that hold the fibres together (mainly pectin and hemicelluloses), without damaging the fibre cellulose. Depending upon the plant type, retting can take from 2 days 181

182

Green composites

Table 9.1 Characteristics of water and dew retting1 Water retting

Dew retting

Advantages More uniform and efficent retting Better fibre quality Possible to control the retting condition Disadvantages Takes place in stagnant water Too labour expensive Too costly and causes water pollution due to the release of galacturonic acid into the stagnant water Retting is by the action of anaerobic bacteria

Advantages In the field Easy process No environmental harm Retting is done by the action of fungi Disadvantages Not possible to control the process as it depends on the climatic condition Not efficient retting Poor fibre quality

to 2 weeks. Fibre quality depends on the retting period, and by controlling the retting process, ultimate fibre quality and consistency can be improved. Over-retting causes fibre degradation whereas under-retting results in the incomplete removal of the gummy pectinic substances and both retting conditions result in inferior fibre quality. Some of the characteristics of water retting and dew retting are shown in the Table 9.1. Developments in technology and handling equipment led to the enzyme retting process, in which pectinolytic enzymes (pectin depolymerising) are used to remove the non-cellulosic parts of the fibre.2 Fibre quality depends on enzyme concentration, temperature, pH value and duration of treatment. For example, too great a concentration of enzyme or too lengthy a treatment may dissolve the cellulose and weaken the fibre. Some advantages of enzyme retting are: (i) it does not pollute the environment as the used solution is biodegradable; (ii) the process can be carried out anywhere where there is running water; and (iii) retting can be performed as a local cottage industry, saving transport costs and fuel. To improve the fibre quality, mixtures of enzymes such as pectinases, xylinases and cellulases can be used.

9.1.2

Steam explosion

This technique was first introduced by Mayson in 19273 to produce wood fibre from wood for board manufacture. It is currently being used as an efficient and economic pretreatment to enhance the enzymatic hydrolysis of cellulose and hence to separate major components of lignocellulosic biomass.4 In this technique lignocellulosics are subjected to steam treatment at high pressure and temperature in a vertically mounted cylinder and at the end of the treatment, the biomass which is at a high pressure is discharged from the cylinder. Saturated steam at high temperature (about 160 ∞C) penetrates and softens the middle lamella of the fibre bundles and due to the sudden release

Green fibre thermoplastic composites

183

of pressure (explosive decompression), the material is mechanically separated into single fibres. This technique can provide fine fibres.

9.2

Green fibres for composite production

Cellulose-based fibres, the so-called green fibres, are used as reinforcements in natural fibre–thermoplastic composites enhancing the strength and stiffness of the resulting composites. Unlike conventional fibres like glass, carbon and aramid, properties of natural fibres vary considerably with the quality of the plant locations, age of the plant, mode of extraction of the fibres from the plant and the part of the plant from where it is taken, e.g. leaf, bast, fruit, etc. Natural fibres are grouped based on their origins; plant, animal or mineral. Generally, fibres from plants are used to reinforce plastics in the composite industry. Depending on their origin, the plant fibres are mainly grouped into: ∑ leaf fibres which run the length of leaves, for example: sisal, abacca, pineapple, banana fibres ∑ bast or stem fibres which are fibrous bundles in the inner bark of the plant stem and run the length of the stem, for example, flax, jute hemp, ramie, kenaf ∑ seed hair fibres, for example, cotton and kapok ∑ fibres from fruits: coir. Other groups include core, pith or stick fibres from the centre of the plant, or pith fibres of plants such as jute and kenaf, fibres from root crops, seed hulls, leaf segments and flower heads. Of these fibres the most widely used are flax, jute, sisal, ramie and hemp. All these natural fibres are lignocellulosic in nature and are the most abundant renewable biomaterials distributed in the form of trees (wood), plants and crops. Typical world production values of these commercially important fibrous resources are given in Table 9.2 and the characteristics of these fibres and their growing areas are given in Table 9.3.

9.2.1

Chemical composition and structural aspects of natural fibres for thermoplastic composites

The age of the plant, climatic conditions and fibre processing techniques influence the structure of fibres as well as their chemical composition. The primary constituents of green fibres are cellulose, hemicellulose, liginin and pectin. The minor components are waxes, silica and other water-soluble substances. The percentages and properties of the constituents contribute to the overall properties of the fibres. Cellulose, a hydrophilic linear polymer consisting of 1,4-b-bonded

184

Green composites

Table 9.2 World production and origin of commercially available fibres Fibre source

World production 1995a (103 tonnes)

World production, 1999b (103 tonnes)

World production, 2001c (103 tonnes)

Origin

Wood Bamboo Cotton Jute Kenaf Flax Sisal Hemp Coir Ramie

1 750 10 18 2

– – – 2 562 – 636 315 79 – –

– – – 3 105 – 635 335 55 654 178

Stem Stem Seed Stem Stem Stem Leaf Stem Fruit Stem

000 000 450 300 970 830 370 214 100 100

Source: aTaken from Eichhorn et al. (2001),5 bhttp://www.nova-institute.de/pdf/nova-studyfull.pdf, chttp://www.utexas. edu/centers/nfic/natstat/data 1 ¥ 001.pdf.

anhydroglucose units containing alcoholic hydroxyl groups, is the essential component of all green fibres. Because of these hydroxyl groups, all the plant fibres are hydrophilic in nature. The basic chemical structure of cellulose in all fibres is the same (albeit with different degrees of polymerisation) whereas the cell geometry of each type of cellulose varies with the fibre: this is one of the factors which contribute to the mechanical properties of the green fibre. Cellulose is not crystalline. The microcrystalline structure of cellulose includes crystalline regions of high order, which are extensively distributed throughout the material, and amorphous regions of low order. The structure of cellulose units is shown in Fig. 9.1.8,9 Hemicellulose is another component of plant fibres. This is not a form of cellulose, rather a group of polysaccharides (excluding pectin) attached to the cellulose after the removal of pectin. Unlike cellulose, which contains only a 1,4-b-glucopyranose ring, hemicellulose contains different types of sugar units. It is also a highly branched polymer (contrasting with the linear cellulose) and has a degree of polymerisation 10–1000 times lower than that of cellulose. Lignin is a hydrocarbon polymer containing phenolic groups, whose exact chemical structure still remains obscure.10 Lignin acts as a structural support material in plants by filling the spaces between the polysaccharide fibres, which hold the natural structure of the plant cell walls together. Lignin stiffens the cell walls and acts as a protective barrier for the cellulose. The properties of lignin vary with the fibre type, but it always has the same basic composition. Mechanical properties of lignin are lower than that of cellulose. Pectin is a collective name for heteropolysacharides, which consist

Green fibre thermoplastic composites

185

Table 9.3 Fibre characteristics and growing area of commercially available fibres6,7 Fibre type

Origin

Species

Length (mm)

Width (mm)

Growing area

Cotton

Seed hair

12–64

20

Flax

Bast

10–36

10–25

Jute

Bast

3–5

Hemp

Bast

Gossypium sp. Linum usitatissimum L. Corchorus capsularis L Cannabis Sativa L.

0.017– 0.023 mm 0.015– 0.46 mm

Ramie

Bast

Boehmeria nivea Gad

15–25

0.02– 0.08

Southern parts of North America EU, Canada, Argentina, USSR, India India, China, Bangladesh EU, USSR, Philippines, Central Asia, China China, Brazil, Thailand, Japan, USA, Malaysia

Kenaf

Bast

Hibiscus cannabinus L

2–6

0.014– 0.033

2.2–8.1

0.010– 0.033 0.011– 0.30

Abaca

Leaf

Sisal

Leaf

Esparto

Stem

Wheat

Stem

Bamboo

Stem

Begasse

Stem

Musa textilis Louis Nee Agave sisalina Perr

6.5–37.2

1.2–5.8

Stipa 0.25–2.0 tenacissima L. Triticum 0.5–3.1 aestivum L. 0.21–37 Saccharum officinarum L.

0.8–2.8

0.010– 0.015 0.008– 0.030 0.006– 0.035 0.010– 0.034

Thailand, India North and South America, Iran Southern USSR Philippines and Equador Central and South America, Africa, Indonesia, West Indies Spanish plateau Western Asia, USA Japan, India India

essentially of 1,4-linked galacturonic acid units, especially methyl esters of various sugar units.8 These are water soluble when they partially neutralise with alkali or ammonium hydroxide. Waxes constitute a small percentage of the structure and contain different types of alcohols that are soluble in water as well as in acids, such as phenolic, oleaginous and stearic acid. Mechanical properties and structural parameters of some commercially important fibres are given in Table 9.4.11–21 The mechanical properties of the fibres depend on the chemical composition of the constituents in a fibre.

186

Green composites H

CH2OH

O

H

H O

H

H

OH H

H

H

CH2OH

O

O

HO

H

OH

HO

H

CH2OH

O

O O

HO

H

H

OH H

H

H

9.1 Structure of cellulose unit.

Each of these constituents contributes to a different extent to the properties and the effects of these constituents in accordance with their importance are shown in Fig. 9.2.8 The price of biofibres varies a lot depending on the economy of the countries where such fibres are widely available. The price of the biofibre depends on availability, the fibre preparation process and the pre-treatment of the fibre (sizing). In recent years, prices for natural fibres have been fluctuating, especially for flax fibres: these are the highest strength fibres and are marginally more expensive than glass fibres. The price value of Table 9.4 Mechanical properties and structural aspects of available fibres11–21 Fibre type

Properties

Moisture Elongation content at break (%) (%)

Microfibril angle (degree)

5.5– 12.6 27.6

7–8

–

33–34

2.7–3.2

10

6–7

– 13– 26.5 61.4– 128 2.86

1.6 1.16–1.5

10.8 12.6

6.2 7–9

1.2–3.8

8.0

7–8

3.5

–

–

106 psi 9.4–22

– 3–7

– 11.0

– 10–12

4–6

15–40

8.0

30–45

70

2.5

–

–

63–67

3.3–3.7

–

–

230–240

1.4–1.8

–

–

Density (g/cm3)

Diameter (mm)

Tensile strength (MPa)

Tensile modulus (GPa)

Cotton

1.5–1.6

–

Flax

1–5

–

Hemp Jute

– 25.2

Ramie

– 1.3– 1.45 1.5

287– 800 345– 1100 690 393– 773 400– 938

Kenaf

–

–

Abaca Sisal

1.35 1.45

– 50–200

Coir

1.15

E-glass

2.5

100– 450 –

Aramid

1.4

–

Carbon

1.7

–

–

980 468– 640 131– 175 2000– 35 000 3000– 3150 4000

Green fibre thermoplastic composites Strength

Moisture absorption Crystalline cellulose Non-crystalline cellulose Hemicellulose + lignin Lignin

Thermal degradation

187

Hemicellulose Non-crystalline cellulose Lignin Crystalline cellulose UV degradation

Hemicellulose Cellulose Lignin

Lignin Hemicellulose Non-crystalline cellulose Crystalline cellulose

Biological Hemicellulose degradation Non-crystalline cellulose Crystalline cellulose Lignin

9.2 Properties and their dependence on chemical constituents.8

Table 9.5 Price of natural fibres in comparison with glass fibres Fibre

*Price ($/ lb)

E-glass Jute Hemp Flax Sisal Coir

0.60–1.75 0.16–0.91 0.27–1.82 0.23–1.82 0.27–0.32 0.11–0.23

*Varies depending on source, quality and fibre form. Source: Taken from Beekwith (2003)11.

some of the fibres in comparison with glass fibres is shown in Table 9.5.11 Though they are more expensive than glass fibres, the advantages of natural fibres make them attractive to the composite industry.

9.3

Thermoplastics for natural fibre composites

9.3.1

Synthetic thermoplastics

Synthetic thermoplastics commonly used for natural fibre composites are polyethylene (PE), polypropylene (PP), polystyrene (PS) and polyamides (nylon 6 and 6,6). PP is the most widely used thermoplastic material in the natural fibre composite industry owing to its low density, excellent processibility, good mechanical properties, high temperature resistance, excellent electrical properties, good dimensional stability and good impact strength. The exciting growth of PP in the composite industry is demonstrated by the large quantity of research work and numerous publications devoted to this thermoplastic material. High impact grade polymers of polyolefins and

188

Green composites

Table 9.6 Properties of commercially important thermoplastic polymers22,23 Property

PP

LDPE

HDPE

PS

Density (g/cm3)

0.899– 0.920

0.910– 0.925

0.941–1

1.04–1.09 1.09–1.14

1.090–1.19

0.01–0.02 < 0.015

0.01–0.2

0.03–0.10 1.3–1.8

1.0–1.6

–10 to –23 –125

–

48

80

110–135

215–216

250–269

Water absorption after 24 hours (%) Tg(∞C)

Tm (∞C) Heat deflection temp. Td, at 1.8 MPa (∞C) Coefficient of linear thermal expansion, a T (mm/mm/∞C ¥ 105) Tensile strength (MPa) Young’s modulus (GPa) Elongation (%) Izod impact strength (J/m)

Nylon 6

Nylon 6,6

160–176

105–116

–133 to –100 120–140

50–63

32–50

43–60

max. 220

56–80

75–90

6.8–13.5

10

12–13.0

6–8

8–8.6

7.2–9.0

26–41.4

4–78.6

14.5–38

–

43–79

12.4–94

0.95– 1.776 15–700

0.055– 0.38 90–800

0.413– 1.490 12–1000

4–5

2.9

2.5–3.9

1–2.5

20–150

35 – >300

21.4–267

>854

26.7– 1068

0.05– 0.55*

42.7–160

16.0–654

*Units ft lb/inch.

polystyrene (blended with ethylene–propylene rubber, nitrile rubber) are also available to improve the impact strength of the corresponding polymers. Physical, mechanical and thermal properties of these polymers are given in Table 9.6.22,23

9.3.2

Biodegradable polymers

These are polymers that can undergo degradation processes under the action of enzymes and/or chemical dissociation with living organisms such as fungi and bacteria into harmless secretion products. The use of these polymers in natural fibre-filled composites leads to the development of economically and ecologically attractive technology. Advantages of biodegradable composites include: complete biological degradation; reduction in the volume of garbage; compostability in the natural cycle; preservation of fossil-based raw materials and protection of climate through the reduction of carbon dioxide emission. Commonly used biodegradable polymer matrices are polylactides (PLA), polyglycolic acid (PGA), poly-b-hydroxyalkanoates (PHA), co-polymer of

Green fibre thermoplastic composites

189

Table 9.7 Physical properties of PLA, PGA, PCL and PHB23 Type of polymer

Density (g/cm3)

Tensile strength (MPa)

Tensile modulus (GPa)

PLA PGA PCL PHB

1.21–1.25 1.5–1.707 1.11–1.146 1.18–1.262

21–60 60–99.7 20.7–42 40

0.35–3.5 6–7 0.21–0.44 3.5–4

Elongation (%)

Tg (∞C)

Tm (∞C)

2.5–6 45–60 150–162 1.5–20 35–45 220–233 300–1000 –60 to –65 58–65 5–8 5–15 168–182

3-hydroxy butyrate (HB) and 3-hydroxyl valearate (HV) (PHBV) and poly(whydroxy alkanoate) mainly poly(e-caprolactone) (PCL).23–30 Physical properties of commercially important biopolymers are given in Table 9.7.23

9.4

High performance fibres: thermal, chemical and mechanical treatments

Researchers are trying to develop high performance fibres (fibres with a high modulus) from plant fibres that can improve the properties of the resulting composites and can be used in high tech applications. As discussed before, natural fibres have a quite complicated structure of crystalline and noncrystalline cellulose fibrils, hierarchically organized with a high degree of order on different scales from micro- and nano- to angstrom levels.31–33 Elementary fibres, having a diameter range of 10–20 mm, of all plant-based natural fibres are formed from bundles of single fibres, which are basically crystalline cellulose microfibrils with a diameter of 5–50 mm, connected to the amorphous lignin and hemicellulose. These elementary fibres bound together by pectin and lignin form a technical fibre of diameter 50–100 mm. Natural fibres contain multiple layers of cellulose-lignin/hemicellulose in one primary and three secondary cell walls stuck together. The composition of each part of the cell wall differs in the orientation of the cellulose microfibrils. The mechanical properties of the fibre generally depend on the content of the cellulose and the spiral angle of the fibrils. Owing to the presence of defects, the crystal modulus of the cellulose is never achieved and this is the main drawback of the cellulose composites. These microfibrils contain monocrystalline cellulose domains with the microfibril axis parallel to the cellulose chains. Each microfibril can be thought of as a polymer chain containing only a small number of defects and chain folding and hence they can provide a modulus close to that of the perfect crystal of native cellulose (estimated to be 250 GPa). Moreover, these microfibrils have a width of 5– 60 nm and much higher length resulting in an infinite aspect ratio that can provide good reinforcement at lower volume fraction of the fibres when incorporated into the matrix. Production of high performance fibres includes

190

Green composites

Untreated hemp fibre

Heat-treated hemp fibre

9.3 SEM photographs of treated and untreated hemp fibre (¥ 500).

the removal of the binding materials such as lignin, pectin and hemicellulose and the techniques used are described below.

9.4.1

Thermal treatment

The basic principle involved in this technique is the removal of lignin by subjecting the fibre to thermal treatment at the glass transition temperature of lignin. The temperature required for the removal of lignin will vary with the type of fibre as the nature of lignin varies with the fibre type. As a result of the thermal treatment, the lignin and some of the hemicellulose undergo depolymerisation, resulting in the lower molecular aldehyde and phenolic functional compounds leaving behind the cellulose microfibrils. The presence of lignin and other extractives on the fibre surface may also improve the compatibility with thermoplastic matrix.34 Heating in an inert atmosphere seems to be a better route than heat treatment under normal conditions for providing high performance fibres. The SEM photomicrographs of heat treated hemp fibres in Fig. 9.3 show the opening up of the fibre bundle.

9.4.2

Nano-biofibrils

Cellulose microfibrils can be separated from agricultural biomass by pulping, acid and alkaline treatment followed by mechanical treatment to obtain an aqueous homogeneous suspension of the microfibrils. 9.4.2.1

Chemical treatment

The fibres are allowed to swell in alkali in order to render the fibre surface more accessible to the following chemical treatments and make the biomass

Green fibre thermoplastic composites

191

pulp. The basic idea of the chemical treatment is to remove pectinic and hemicelluosic polysaccharides. Acid treatment removes acid soluble pectins while alkaline treatment is expected to hydrolyse pectins by a b-elimination process and solubilise them; after the treatments, the solubilised polysaccharides can be removed by filtration and washing with water.35–37 9.4.2.2

Mechanical treatment

The fibre suspension is propelled under the influence of a high pressure gradient through the valves. Within a very short time, the potential energy is converted into kinetic energy which, in turn, changes into heat. As a result of this energy transfer, high shear and normal stresses and high energy, particle collisions occur in the medium leading to microfibril individualisation. Separation of microfibrils from sugar beet, potato pulp, wheat straw and tunicin using this technique has already been reported.35, 38–41 A typical transmission electron microscopy (TEM) photograph of the microfibrils separated from flax fibres is shown in Fig. 9.4. 9.4.2.3

Mechano-physical separation of microfibrils

The underlying principle of this technique is to provide enough shear to the fibre to separate microfibrils of diameter 0.1 to 1 mm. The pulp from the biomass is disintegrated in a fibre disintegrator and this is followed by a thorough shearing. The refined fibres collected are frozen by immersion in liquid nitrogen, which leads to the formation of ice crystals within the cells. Under high mechanical impact, the crystals slash the cellular wall and release wall fragments. The processed fibrils can be further processed by two

9.4 TEM photograph of nanofibrils produced from flax (diameter 15–60 nm).36

192

Green composites

9.5 AFM photograph of microfibril from bleached kraft pulp.42

techniques. First, the fibres can be freeze dried and ground again in a Wilely mill, with a sieve size of 60 mesh, to provide further shear, and the fibrils coming through the sieve are collected. In the other process, microfibrils are separated by being passed through a high pressure disintegrator. The atomic force microscopy (AFM) photograph of the microfibrils is shown in Fig. 9.5.42

9.5

Processing of natural fibre-filled composites

The major problems associated with the processing of natural fibre-filled systems include variation in the quality of the raw material, poor compatibility between the hydrophilic natural fibre and the hydrophobic matrix and the poor thermal stability of these fibres at temperatures above 230 oC. Unlike conventional glass fibres, the poor thermal stability of natural fibres limits viable thermoplastics to those which have a processing temperature of less than 230 oC and these include PP, PS, PE (LDPE as well as HDPE), PVC and polyamide 6 and 6,6. In the case of natural fibres, since less damage is done to the processing equipment, a higher loading (50–65 wt %) of the fibre can be processed compared to that of glass fibres. Size of the natural fibre may vary from 20-mesh particle size to a few mm of fibre size. The greatest concern in compounding or blending of natural fibre with thermoplastics seems to be:43 ∑ the feeding difficulty due to the difference in the bulk density of the two components ∑ poor dispersion of the fibres in the matrix due to the intramolecular as well as intermolecular bonding in the lignocellulosic fibres, which leads

Green fibre thermoplastic composites

193

to agglomeration. This in turn results in the poor reinforcing ability of the natural fibres ∑ the degree of shear of the compounding equipment that is related to the fibre length retention or reduction. The ultimate goal of compounding is the production of a compounded pelletised feed stock that can be processed further just like any other thermoplastic processing techniques, such as injection molding, extrusion and thermoforming.

9.5.1

Batch style, kneading-type compounding equipment

Batch style, kneading-type compounding equipment can be used to avoid feeding difficulties and consists of a mixing chamber that contains two low speed high torque kneading rotors. Operation of this equipment is as follows: (i) pre-weighed amounts of materials are loaded into the mixing chamber with the help of a ram feeder to facilitate feeding of low bulk density lignocellulosics; (ii) the chamber is closed after feeding and the motors are run to enable mixing of the components; (iii) the material is discharged as a whole after mixing finishing and fed through a single screw extruder and pelletiser line for the production of pellets. Mixing time, batch temperature and energy consumption are the three controllable variables in this type of compounding equipment.

9.5.2

Continuous kneading mixers

Continuous kneading mixers43,44 consist of two long intermeshing rotors in a heated barrel. Gravity or ram feeders are used to feed the material into one end of the mixing barrel. The action of the forces of the rotors leads the material through the mixing chamber, where it is blended by the shear force resulting from the rotating action of intermeshing rotors and subsequently discharged once the mixing is complete through a sheet or strand die, usually directly into a pelletiser. The advantages and disadvantages of the two machines are shown in Table 9.8. To make a product, both batch and continuous kneading compounding machines require downstream processing equipment. A more expensive, but at the same time more sophisticated, compounding equipment is a twin screw extruder, which is similar in operation to a continuous kneader. The twin screw extruder consists of two screws running in a heated barrel. Ram feeders or crammers are used to feed the material into the extruder at various points to accommodate various compounding schemes. As the twin screws act as pumps, they can be used to make products by profile extrusion.

194

Green composites

Table 9.8 Comparison between batch-type and continuous-type compounding equipment Batch-style kneading

Continuous kneading mixers

Advantages

Advantages

Accommodate a wide range of feedstocks Work with extremely high viscous materials.

Higher output rate compared to batchstyle mixers at the same power rating.

Very low bulk density materials can feed by special ram feeders. Relatively low shear forces help retain fibre length. Excellent quality control can be attained as the formulation components of each batch are weighed individually.

Disadvantages Lower output compared to continuous kneading mixers at the same power rating. Discharge of the material as a lot requires expensive downstream processing equipment for practical material utilisation by pelletisation.

Discharged material can directly feed into the pelletiser. Time-sensitive additives can be added at various stages of mixing. Relatively low moderate shear, hence the retention of fibre length is good.

Disadvantages Loss of formulation control, as the formulation is dependent on the ability to feed the components consistently. Do not function as pumps, so they cannot be used to extrude or mould finished products.

Addition of time-sensitive additives are not feasible due to the batch nature of the equipment.

9.5.3

High shear mixers

High shear mixers43 can be used if the fibre length is not a concern. One such type of mixer is the thermokinetic batch-style machine, known as a K-Mixer. This machine comprises two high speed rotors in a heated chamber and operation is similar to batch-style kneading equipment. Melt blending is effected by the kinetic energy developed by the high speed rotor enclosed. Once the materials have been compounded, they can be processed into products. Some of the methods for the production of composite products include extrusion, injection molding, thermoforming, foaming and film casting.

9.5.4

Extrusion

During extrusion44 the compounded composite material is fed into the heated barrel of the extruder using feeders and is heated so that the thermoplastic component can flow. The rotation of the screws inside the heating chamber plasticises the material, which is then continually pumped and forced through a die of given cross-section configuration. The extrudate is supported and then cooled in a water bath, after which the profile is cut into length.

Green fibre thermoplastic composites

195

9.6 Typical products produced by injection, extrusion and compression moulding.

Co-rotating and counter rotating twin screw extruders as described above can be used for the profile extrusion. They are now very widely used for long natural fibre dispersion with special feeding technology. These premixed compounds are then processed by further extrusion or injection molding. Examples of products made by extrusion are pipes and tubings, furniture edgings, mouldings and sheet goods.

9.5.5

Injection moulding

Injection moulding44,45 is the most widely used processing technique for making composite products especially for automotive applications due to the ease of production of intricately shaped articles in a cyclic manner. This process differs from extrusion in that the compounded material is heated and pumped into a permanent mould of the desired shape where it takes shape while cooling. The mould is then opened and the finished product is discharged. Examples of products are interior door panels, automotive products and door stems. Some of the products produced by extrusion, injection moulding and compression moulding are shown in Fig. 9.6.

9.5.6

Thermoforming

The thermoforming46,47 technique is mainly used to produce natural fibre mat thermoplastic (NMT) composites. The process is as follows: pre-cut

196

Green composites

car sun visor car speaker 9.7 Typical NMT products for automobile applications.

layers of fibre mat and polymer sheet are placed in a heated mould of desired shape and placed in a hot press at ambient temperature. Pressure applied to the mould consolidates the material as the heat is transferred through conduction to melt the thermoplastic, which flows around and penetrates into the lignocellulosic component. The pressure is kept constant during the heating and cooling phase. The processing time can be reduced by shifting the heating and cooling operations to two phases. After reaching the melt temperature in a hot press, the molten hybrid material is consolidated into a composite in a cold press within less than two minutes and then cooled down to room temperature so that the finished product can be removed from the mould. Some of the NMT- based products are shown in Fig. 9.7.

9.5.7

Foaming

The foaming48 technique is used to produce foamed products that can be used in upholsteries and in insulation applications. After blending the components (fibre, thermoplastic and blowing agents), the material is fed into a single screw extruder using a special force fed hopper. The extrudate coming out of the other end of the extruder is then passed through a static mixer (diffusion enhancing device) to ensure the complete dissolution of the gas generated from the blowing agent into the plastic matrix. The maximum temperature in the barrel and diffusion section is about 205 ∞C, ensuring the decomposition of the blowing agent. The extrudate is allowed to cool uniformly to 150 ∞C by passing it through a heat exchanger and then finally the extrudate is passed through a nozzle die to the product.

Green fibre thermoplastic composites

9.5.8

197

Film casting

Film casting39,49 is used whenever the processing components are in solution, dispersion or suspension form and it is mainly used to produce nano-composite films. In this process, aqueous suspensions of nanosized micro fibrils and water-soluble biopolymers are mixed and homogenised. After homogenising, the air in the suspension is removed by pumping the suspension under vacuum in order to avoid bubble formation during casting. Casting is done by pouring the suspension into a plastic mould at 35–37 ∞C to allow water evaporation for 48 hrs so that the polymer coalesces and encloses the fibre.

9.6

The performance and durability of natural fibres

A wide variety of natural fibres/lignocellulosics have been used as reinforcements for thermoplastics such as PE, PP, PS and nylon 6. Depending on their performance when used in a plastic matrix, cellulosic fibres can be classified under three categories: (i) Wood flour and other low cost agricultural-based flours: particulate filler that enhances the tensile and flexural moduli of the composite with little effect on the composite strength. (ii) Fibres with a higher aspect ratio: these can contribute to an increase in modulus as well as an improvement in strength when suitable additives are used to enhance the matrix-to-fibre stress transfer, e.g. wood fibres and recycled newspaper fibres. (iii) The most efficient cellulosic fibres with a high cellulose content and low microfibril angle: these can provide better properties compared with wood flour and wood fibres, e.g. kenaf, jute, flax, etc. In general, natural fibres have a higher Young’s modulus compared to thermoplastics; they make the composite stiffer. The performance properties of these natural fibre composites largely depend on the ratio of the modulus of the fibre to the matrix, the dispersion of the fibres in the matrix, fibre length, fibre length distribution and the interaction between the fibre and matrix50–66. Use of coupling agents or dispersing agents is a cheaper way to improve properties of composites for high volume, low cost applications, rather than grafting the polymer with functional molecules. Coupling agents can improve the interfacial interaction between the fibre and the matrix and therefore improve the compatibility in several ways:12 ∑ elimination of weak boundary layers, formation of a tough, flexible interphase layer and development of a highly cross-linked region with a modulus intermediate between those of the fibre and the matrix ∑ formation of covalent bonds with both materials

198

Green composites

∑ occurrence of the acid–base effect, altering the acidity of the substrates. The most widely used coupling agents are maleated polyolefins, silanes, titanates, isocyanates, maleimides and triazine derivatives.50,53–58 Most of the studies of natural fibre composites involve the study of mechanical properties as a function of fibre content, effects of various treatments on fibres and the use of external coupling agents.59–70 Both the matrix and fibre properties play an important role in improving the mechanical properties of the composite. The strength of the composite is more sensitive to matrix properties, while the modulus of the composite depends more on the fibre properties. The aspect ratio is the main parameter governing fracture properties. In any short fibre composite, a critical fibre length is necessary for the fibre to develop its full stressed condition in the polymer matrix for efficient stress transfer between the fibre and matrix. Fibre lengths shorter than this critical length lead to failure owing to debonding at lower loads. On the other hand, for fibre lengths greater than the critical length, fibre is stressed under applied load, leading to efficient stress transfer between the matrix and the fibre and thus resulting in higher strength composites. For good impact strength, an optimum bonding level is necessary. The degree of adhesion, fibre pull out and energy absorption mechanisms are some of the parameters that can influence the strength of the short fibre composites. The mechanical properties vary with composites according to the rule of mixtures and increase linearly with composition. However, at higher fibre loading, this linear dependence of the fibre content does not exist: this may be due to the lack of wetting of the fibre by the polymer. Mechanical properties of some of the natural fibre–PP composites and glassfilled PP prepared by injection moulding are shown in Table 9.9.45,71 Mechanical properties of typical extruded, and compression and sheet moulded natural fibre–polyolefine composites are given in Table 9.10.45,47,66

9.6.1

Creep behaviour

Alhough many studies have reported on various aspects of natural fibre reinforced composites, creep behaviour as a performance criterion for these composites is still under-researched.72–74 The change in strain with respect to time under constant load measures the creep. The creep properties of the composites depend upon the crystallinity of the plastics, fibre content, fibre type, temperature, humidity, fibre–matrix interaction, stress level and fibre geometry. Bledzki studied the effect of coupling agent and fibre content on the creep behaviour of jute–PP composites up to a maximum of 300 minutes and reported that after 300 minutes, a strain on the outer fibre of around 5% was produced in composites without coupling agent and with a fibre content of 23 vol%. Increasing the fibre content and interfacial interaction between the fibre and matrix reduced the creep rate.12

Table 9.9 Physical properties of natural fibre-filled composites45,71 Filler/reinforcement

Tensile modulus (GPa) Flexural strength (MPa) Flexural modulus (GPa) Notched impact strength (J/m)

– 30.4 1.15 42.1 1.10 16

* Charpy impact strength in kJ/m2.

Wood

Old newsprint

TMP

Jute

Kenaf

Kenaf with PP impact copolymer

Glass

30

40

30

30

50

50

40

25.9

32.3

34.59

36.1

65

53

39

– 48.6

3.22 57.4

2.59

3.53

*1.87

3.3

2.45 65.6 3.45

4.6 56.48 4.62 *269

8.3 98 7.3 32

7.5 – – 74

7.6 62 6.9 27

Green fibre thermoplastic composites

% filler (wt) Tensile strength (MPa)

None

199

200

Green composites

Table 9.10 Properties of composites prepared by different processing techniques45,47,66 Filler and process

Wood flour/ extrusion

Wood flour compression moulding

% filler (wt) Tensile strength (MPa) Tensile modulus, (GPa) Flexural strength (MPa) Flexural modulus, (GPa) Impact strength (J/m) Notched Un-notched

50 29.6 1.52 45.6 2.08

35 22.1 1.64 – – – 11.2 –

67 131

Hemp/NMT

64 41 – 63 4.9 – – 84

*Charpy impact strength in kJ/m2.

Temperature and moisture also complicate the creep. At high temperatures the polymer chain relaxes and the modulus decreases: creep would increase at higher temperature. Moisture absorbed by the lignocellulosics would also increase the creep. Creep behaviour of 50% wood flour-filled HDPE composite was studied using a flexural creep with a four point loading configuration, with respect to temperature, moisture content and stress level.75 The experimental set-up and the corresponding creep behaviour at different stress levels are shown in Figs. 9.8 and 9.9, respectively. Values for g1 (the transient compliance which measures stress and temperature effects on creep

9.8 Indoor weathering set-up for natural fibre composites.

Green fibre thermoplastic composites

201

Strain ¥ 102

6

4

2

0 0

50

50% max stress

100 150 Time (min) 45% max stress

200

250

35% max stress

9.9 Creep behaviour of natural fibre composites at different stress levels.

compliance) and g2 (which measures the influence of loading rate on creep, depends on stress and temperature and indicates the amount of recovery) are found to be 2.5595 and 0.7685 (35% maximum stress), 2.2146 and 1.0400 (45% maximum stress) and 2.0867 and 1.2293 (50% maximum stress). With stress, the non-recoverable damage (permanent set) to the composite material increases.

9.7

Environmental benefits of using natural fibre-reinforced thermoplastics

The primary environmental advantages of using natural fibre-reinforced thermoplastic are as follows: ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑

biodegradability reduction of greenhouse gas emission enormous variety of structural fibres available throughout the world creation of job opportunities in rural areas development of non-food agricultural/farm-based economy low energy consumption low cost low energy utilisation.

Using agricultural materials as raw materials for making composite products provides a renewable resource as well as generating a non-food source of economic development from farming and rural areas. Also, use of renewable fibres in the composites produces an overall CO2 balance, as the amount of CO2 taken up during their growth is matched (apart from the efforts necessary to grow and harvest the fibres) by the CO2 released during their disposal, i.e. either by burning or by rotting. Replacing conventional fibres based on

202

Green composites

petroleum resources with natural fibres reduces the greenhouse gas emissions considerably. The amount of energy required for the production of natural fibres is less than that of glass fibres. Moreover, their lower density (>40%) compared to glass fibres leads to fuel-efficient production of composite products, especially in automotive applications: this, in turn, leads to a reduction in greenhouse gases.50 The carbon sequestration and storage potential of hemp-based natural fibre mat thermoplastic composites has been estimated to be about 325 kg carbon per metric tonne during their useful lifetime.76 A net carbon sequestration of 0.67 tonne/ha/yr was estimated for a composite containing 65 wt % of hemp fibre. It has been found that replacing 30% glass fibre with 65% hemp fibre in thermoplastic composites produces a net saving of energy consumption of 50 000 MJ (about 3 tonne CO2 emission) per tonne of thermoplastic. Also, by substituting 50% of the glass fibre by natural fibre in automotive applications, 3.07 million tonnes of carbon dioxide emissions and 1.9 million m3 of crude oil can be saved.

9.7.1

Recycling aspect of composites

Natural fibre-filled thermoplastic composites are easier to recycle than the conventional mineral-based fibre filled thermoset composites. This is due to the less brittle nature and softer texture of the fibre and the processibility of the thermoplastic. Unfortunately, not much literature is available regarding the recycling of post-consumer products. The repeated process of injection molding and granulation and the influence of this process on the mechanical properties of wood fibre-filled composites have been studied by Sain and Balatinecz. The properties of the composites after reprocessing three, six and eight times are given in Table 9.11.77 The deterioration in the properties is due to fibre attrition and oxidative degradation of the PP matrix during the repeated grinding and injection molding processes.

9.8

Future trends

Owing to their renewability, worldwide distribution and recyclability, the market for these composites will be able to expand. It will be possible for them to be used in a wide range of products, from those where very inexpensive low performance composites are suitable, to those where expensive high performance structural components are required. Enormous amounts of natural fibre-filled thermoplastic composites have been used in the building, automotive and packaging industries and also in other low cost, high volume applications. Examples of various applications are as follows: ∑ Automobile: upholstery, rear shelves, door-trim panels, head liners ∑ Building: decking, railing, flooring, wall frame

Green fibre thermoplastic composites

203

Table 9.11 Properties of recycled wood fibre polyolefin composites Number of recycling

Property

Polypropylene

30% wood fibrefilled PP

45% wood fibrefilled PP

0 3 6 8

Tensile strength (MPa)

28.76 26.31 27.00 26.00

38.00 33.08 31.38 30.46

36.77 29.38 31.08 –

0 3 6 8

Flexural strength (MPa)

40.5 34.67 40.67 38.33

60.67 54.33 60.33 53

65 59.67 53 –

0 3 6 8

Flexural modulus (GPa)

1.7 1.1 1.8 1.5

4.3 4.4 5.1 4.0

8.0 11.3 6.9 –

0 3 6

Melt flow index (g/10min)

11.04 16.25 24.80

1.46 1.88 3.13

10.00 12.71 7.50

∑ Consumer: furniture, toys, gardening equipment ∑ Packaging: bio-packaging. Developments in processing technology enabling the production of complexshaped products and the use of proper fibre treatments and compatibilisers and/or coupling agents will lead to the production of composites with optimum properties to meet end use requirements. It has been reported that composites based on polyolefins offer advantages of a 30% weight reduction in addition to a 20% reduction in processing temperature and a 25% reduction in cycle time. Researchers are also trying to produce hybrid composites containing different types of fibres for high performance applications. The creation of high modulus biofibres and the production of biopolymers will also expand the horizon of green thermoplastic composites from packaging to structural applications.

References 1 W.H. Morrison III, D.D. Archibald, H.S.S. Sharma and D.E. Akin (2000). Industrial Crops and Products, 12, 39–46. 2 J.A. Foulk, D.E. Akin and R.B. Dold (2001). Industrial Crops and Products, 13, 234– 48. 3 W.H. Mayson (1927). US Patent No. 16155618. 4 Z. Koran, B.V. Kokta, J.L. Valda and K.N. Law (1978). Pulp and Paper Canada, T107–13. 5 S.J. Eichhorn, C.A. Baillie, N. Zafeiropoulos et al. (2001). J. Mater. Sci., 36, 2107– 31.

204

Green composites

6 M. Morimiotto (1989). Biomass Handbook, eds. O. Kitani and C.W. Wall, Gordon and Breach Science, New York, pp. 642–6. 7 A.K. Mohanty, M. Misra and G. Hinrichson (2000). Macromol. Mater. Eng., 276/ 277, 1–24. 8 T.P. Nevell and S.H. Zeronian (1985). Cellulose Chemistry and Its Applications, Wiley, New York. 9 R.M. Rowell (1992). ACS Symposium Series; Emerging Technologies for Materials and Chemicals from Biomass, eds. R.M. Rowell, Tor P. Schultz and R. Narayan, ACS, Washington. 10 W.G. Glasser (1980). Pulp and Paper Chemistry and Chemical Technology, Wiley International Science, NewYork, pp. 39. 11 S.W. Beekwith (2003). Compos. Fabrication, January, 50–2. 12 A.K. Bledzki and J. Gassan (1999). Prog. Polym. Sci., 24, 221–74. 13 A.N. Shan and S.C. Lakkard. (1981). Fibre Sci. Technol., 15, 41–6. 14 M.K. Sridhar and G. Baravarajappa (1982). Text. Res. J., 87–92. 15 P.J. Roe and M.P. Ansell (1985). J. Mater. Sci., 20, 4015–20. 16 A.K. Bledski and J. Gassan (1990). Handbook of Engineering Polymeric Materials, ed. N.P. Cheremisinoff, Marcel Dekker, New York. 17 H. Saechtling (1987). International Plastic Handbook, Hanser Publishers, Munich. 18 K.G. Satyanarayana, B.C. Pai, K. Sukumaran and S.G.K. Pillai (1990). Handbook of Ceramics and Composites, Vol.1, ed. N.P. Cheremisinoff, Marcel Dekker. 19 S.C.O. Ugbolue (1990). SCO Textile Institute, 20 (4), 1–43. 20 S.H. Zeronian (1991). J. Appl. Polym. Sci., 47, 445–61. 21 E.T.N. Bisanda and M.P. Ansell (1992). J. Mater. Sci., 27, 1690–700. 22 S.S. Schwartz and S.H. Goodman (1982). Plastic Materials and Processing, Van Nostrand, New York. 23 K. Van de Velde and P. Kiekens (2001). Polym. Testing, 20, 885–93. 24 M. Mochizuki and M. Hirami (1997). Polym. Adv. Technol., 8, 203. 25 D. Marshal (1998). Eur. Plast. News, March, pp. 23. 26 R. Leaversueh (2002). Plast. Technol., March, pp. 50. 27 K. Van de Velde and P. Kiekens (2002). Polym. Testing, 21, 433–42. 28 N. Grassie, E.T. Murray and P.A. Holmes (1984). Polym. Degrad. Stabil., 6, 47; 6, 95; 6, 127. 29 R.S. Lebrk and R.J. Williams (1994). Macromolecules, 27, 3782. 30 M.C. Flemings, R.J. Brook and S. Mahajan (1994). Encyclopaedia of Advanced Materials, Vol.1, Pergamon, p. 238. 31 M.C. McCann, B. Wells and K. Robarts (1990). J. Cell. Sci., 96, 323–34. 32 M.C. McCann, B. Wells and K. Robarts (1993). J. Microscopy, 166, 123–36. 33 R.A. Young (1994). Kirk–Othmer Encyclopedia of Chemical Technology, 4th edn. Wiley and Sons, New York. 34 B. Prasad and M. Sain (2003). Mater. Res. Innovation, 7 (4), 231–8. 35 A. Dufresne, J.Y. Camille and M.R. Vignon (1993). J. Appl. Polym. Sci., 64, 1185– 94. 36 A. Bhatnagar and M. Sain (2003). Proceedings of 53rd Canadian Chemical Engineering Conference and 6th Conference on Process Integration and Optimization for Energy Saving and Pollution Production, Oct 26–29. 37 A. Bhatnagar and M. Sain (2003). Patent Pending, August. 38 A. Dufresne, D. Dupeyre and M.R. Vigon (2000). J. Appl. Polym. Sci., 76, 2080–92. 39 D. Dubief, E. Samain and A. Dufresne (1996). Polym. Compos., 17, 604.

Green fibre thermoplastic composites

205

40 A. Dufresne, J.Y. Camilla and W. Helbert (1997). Polym. Compos, 18, 198. 41 A. Dufresne (2000). Compos. Interfaces, 7, 53. 42 A. Chakraborty, M. Sain and M. Kortschot (2003). Proceedings of 53rd Canadian Chemical Engineering Conference and 6th Conference on Process Integration and Optimization for Energy Saving and Pollution Production, Oct 26–29. 43 B. English, P. Chow and D.S. Bajwa (1996). Paper and Composites from Agrobased Resources, eds. R.M. Rowell, R.A. Young and J.K. Rowell, CRC Press, Lewis Publishers, New York. 44 J.L. Thorne (1998). Handbook of Composites, 2nd edn, ed. S.T. Peter, Chapman and Hall, London. 45 S. Law, S. Panthapulakkal and M. Sain (2003). Physical properties of injection molded wood fibre–plastics composites, Proc. Injection Molding Minitech, Toronto, ON. (CD), March 20. 46 M. Sain and M. Pervaiz (2002). High impact natural fibre composites. Canadian Patent No.2,407,880 Filing date: October 26, 2002. 47 M. Pervaiz and M. Sain (2003). Macromol. Mater. Eng., 288 (7), 553–7. 48 G. Rizvi, L.M. Matuana and C.B. Park. (2000). Polym. Eng. Sci., 40, 2124–32. 49 P.D. Hoagland and N. Parris (1996). J. Agric. Food Chem., 44, 1915–19. 50 D. Nabi Saheb and J.P. Jog (1999). Adv. Polym. Technol., 18, 351–63. 51 I. Grof, M.M. Sain and V. Sunova (1992). J. Appl. Polym. Sci., 44, 1061–8. 52 I. Leduce, M.M. Sain and J. Kozankova (1991). J. Polym. Testing, 10, 378–97. 53 D. Maldas and B.V. Kokta (1989). J. Vinyl Tech., 11, 751–55. 54 D. Maldas, B.V. Kokta and C. Daenault (1989). J. Appl. Polym. Sci., 37, 90–9. 55 R. Karani, M. Krishnan and R. Narayan (1997). Polym. Eng. Sci., 37, 476–83. 56 B. Liao, Y. Huang and G. Long (1997). J. Appl. Polym. Sci., 66, 1561–8. 57 P. Zadorecki and T. Ronnhult (1986), J. Polym. Sci., Part A. Polym. Chem., 24, 737– 45. 58 P. Zadorecki and P. Foldin (1986). J. Appl. Polym. Sci., 31, 1699–707. 59 M. Sain, S. Law S. Panthapulakkal, and A. Boullioux, (2003). Stiffness-correlation of polyolefin–woodfibre composites. Proc. International Woodfibre Plastics Conference, Bordeaux, France, March 27. 60 V. Khunova, C. M. Liauw, P. Alexy and M. Sain (1999). Die Angew. Makromol. Chem., 269, 78–83. 61 M. Sain, C. Daneault and C. Lavoy (1996). Acta Polymerica, 4 (22), 178–80. 62 V. Khunova and M.M. Sain (1995). Die Angew Makromol. Chem., 72 (4), 77. 63 M.M. Sain and B.V. Kokta (1995). J. Appl. Polym. Sci., 54, 1545–59. 64 M.M. Sain and B.V. Kokta (1993). J. Appl. Polym. Sci., 12 (2), 167–83. 65 M.M. Sain and B.V. Kokta (1993). Die Angew Makromol. Chem., 210, 33–46. 66 M. Sain and B.V. Kokta (1994). Polym.-Plast. Technol., 33(1), 89–104. 67 M.M. Sain and B.V. Kokta (1994). J. Reinforced Plast. Compos., 13(1), 38–53. 68 M.M. Sain B.V. Kokta and D. Maldas (1993). J. Adhesion Sci. Technol., 7, 49–61. 69 V. Khunova, M.M. Sain and I. Simke (1993). Polym.-Plast. Technol., 32, 299–309. 70 V. Khunova, M.M. Sain, I. Simke and Z. Brunovska (1993). Polym-Plast. Technol., 32, 311–20. 71 A.R. Sanadi, D.F. Caulfield and R.E. Jacobson (1996). Paper and Composites from Agrobased Resources, eds. R.M. Rowell, R.A. Young and J.K. Rowell, CRC Press Lewis Publishers, New York. 72 B. Xu, J. Simonsen and W.E. Rochefort (2001). J. Appl. Polym. Sci., 79, 418–25. 73 M. Sain, J. Balatinecz and S. Law (2000). J. Appl. Polym. Sci., 77, 260–8.

206

Green composites

74 W.S. Lin and M. Sain (2002). Effects of long-term stress and aging on mechanical performance of wood fiber filled polyolefin decking composites. Proceedings 7th Toronto Conference on Progress in Woodfiber–Plastic Composites, Toronto, ON., May 23–24. 75 W.S. Lin, A.K. Pramanick and M. Sain (2003). J. Comp. Mater., (in press). 76 M. Pervaiz and M. Sain (2003). Resources, Conser. Recyc., 16, 1–16. 77 M. Sain and J. Balatinecz (1997). Proceedings of IUPAC Conference, July, Prague.

10 Clean production N. TUCKER University of Warwick, UK

10.1

Introduction: clean processing

Companies are beginning to recognise the value of the good corporate citizenship and intelligent stewardship of finite resources. The reasons for this are a mixture of enlightened self-interest and response to legislation. The generic term used to describe this is clean processing. The term was popularised by Thorpe (1999), who defined it as ‘. . . a way to reverse our current non-sustainable use of materials and energy’. The essence of clean production is an attempt to move manufacturing and use of articles from the linear use of resources to make an article that is then used and thrown away, into the cyclical use of resources that do not produce waste products that cannot be used as the feedstock for some other process or manufacture. Clean processing is the descendant of the sprawling clan of last centuries manufacturing nostrums such as lean manufacturing (reduction of waste in all aspects of the manufacturing processes), (just-in-time JIT – building a relationship with preferred suppliers to minimise stock levels) – see Womack et al. (1990), and Kaizen (continuous improvement of all aspects of the manufacturing process – see Imai (1986)). These and other methods promised economic survival to the companies who bought into them, during the whittling to the bone of the infrastructure of manufacturing. In the UK, survivors of this 66–75% reduction in manufacturing capacity over the last quarter of a century are lean, agile, innovative and well aware of the forces of global competition. There is now very little technological competitive edge left in volume manufacturing. Factories in low wage areas no longer lag behind in technology or ability, nor are they short of investment capital or educated work forces. The stage is set for a new revolution in manufacturing processes, where the driving forces will not be those of simple bottom line measurements. Clean technologies, coupled with a more complete consideration of the product life cycle have the potential to apply the jump leads to manufacturing industries in the old manufacturing countries. Cleaner production methods must satisfy the triple bottom line.

207

208

10.1.1

Green composites

Environmental

Does the method have minimal effect on our environment? Factors such as inprocess waste, e.g. bleed fabrics used in the pre-pregging process, in-process scrap – cut-offs from fibre reinforcement packs or spilt resin, energy use – either to run the process or to manufacture the raw materials must all be considered.

10.1.2

Social

Exposure of the workforce to toxic materials must be kept to a safe minimum. This is particularly important when working with unreacted thermoset resin systems. These materials have an unattractive combination of high reactivity (may be carcinogens or mutagens) and high mobility – consider the sharp penetrating smell of the styrene solvents associated with thermosetting polyester resin systems.

10.1.3

Economic

Methods must be profitable if they are to be taken up and developed by industry. It is arguable that following on from Crosby’s dictum that quality is free (Crosby, 1979), that sustainability in manufacturing processes is not only right, it is free. It is not only free, it is the most profitable product line we have. In other words, low environmental impact is another facet of good housekeeping or lean manufacturing. However, it should be noted that the UN environment programme (UNEP) (Pearce, 2002) suggests that market forces are unlikely to cause the spontaneous adoption of sustainable methods at present. At this point it is necessary to examine in greater detail what is meant by ‘clean’. Composite manufacturing processes are frequently described as ‘clean’. For example, when manufacturing aeroplane parts by the autoclave route, great care is taken to prevent contamination of the components. However in addition to high quality (i.e. consistent and void free) end products, the manufacturing route produces a considerable quantity of waste materials. Similarly in low volume yacht manufacture, about 63 m3 of waste are produced for every tonne of finished boat (Fox, 2002). However, if we consider the life of the manufacture beyond the factory gate, it is likely that a helicopter, for example, will consume more energy in its life than a sailing boat. Surfboard manufacture in the South West of England is typically carried out by small to medium sized enterprises. The surfboard enjoys the reputation of being a low environmental impact pastime (http://www.sas.org.uk/). The companies employ skilled artisans who start with an imported pre-cast polyurethane foam cast blank, finish it to their preferred shape, and then hand lay a glass polyester composite onto it. The board is then sanded and polished to a smooth finish. It is estimated that about one-third of the raw materials used to make a board end up on the workshop floor as in-process scrap (Henty, 2002). With regard to endof-life disposal, in common with many glass-reinforced articles, the likely route

Clean production

209

is through landfill. The surfboard is an obvious candidate for re-engineering of both manufacturing methods and material choice. However, as with boat building, this method provides jobs in an area of high unemployment, it is also a process route that does not require large amounts of capital to start up. Injection moulding (used at the moment to make short fibre composites) also enjoys a clean reputation – machines are operable in literal clean room conditions, however depending on the moulding, 20% of the raw materials used may end up as in-process scrap, and the manufacturing history of the ancillaries such as the mould tools should be included in the process assessment. The Chrysler Composite Vehicle, described as a low environmental impact vehicle, needs a set of mould tools weighing 450 tonnes each to manufacture the body shell. It is therefore evident that considerations of environmental cleanliness extend back down the supply chain, and forward into the proposed use of the manufactured article. In summary, Thorpe (1999) suggests that clean production: ∑ questions the need, as opposed to desire, for products ∑ produces for durability and reuse rather than recycling ∑ aims to reduce consumption in affluent economies while maintaining quality of life ∑ implements the precautionary approach (the proponents of an activity should prove there is no safer way of proceeding, rather than the victims of the activity proving that it is harmful) to material selection, process and product design. Hence, the use of renewable resource materials is encouraged and the use of energy is minimised. The use of less toxic and safer inputs in production processes is required. Less direct benefits to the stakeholders in manufacturing should include the assurance of sustainable work for the workforce and protection of biological and social diversity for the community at large. These aims are laudable, but need to be approached with care. The rapid introduction of novelties into the market (‘product churn’) is highly significant in maintaining the levels of apparent prosperity in the richer countries. Whilst it is to be hoped that we can change our lifestyles to minimise undesirable impacts on the planet by raising the standards of living of the poor to match the rich rather than vice-versa (see Von Weizäcker et al. (1998) for an optimistic statement of this view), the alternate view of Pearce (2002) that the existing status quo will prevail is perhaps more likely. Thorpe’s view of ‘need’ rather than ‘desire’ as the driver for new products has the unfortunate appearance of pushing us towards some sort of centrally planned economy, with external assessment of consumer needs. A happier thought is that education of us, as consumers, could result in a fashion for low environmental impact manufactures. This latter route holds out the possibility of providing a smooth evolutionary transition to sustainable practices via established mechanisms of product introduction. A fuller discussion of this is beyond the scope of this chapter. The discussion will concentrate on the technical aspects of the application of clean production to composites manufacture.

210

10.2

Green composites

Energy saving in the manufacture and production of composites

This section assumes a general knowledge of the techniques used for manufacture of composite articles. For a full description of composite manufacturing processes, the reader is referred to Tucker and Lindsey (2002), Kelly (1994) or Rudd et al. (1997). The section concentrates on the most energy intensive methods of composite manufacture. In the UK the government has proposed considerable reductions in energy used. As part of support for the Kyoto agreement, the UK is expected to reduce its energy consumption by 20% by 2010 (these values take the energy used in 1990 as a baseline). Looking further into the future (http://www.dti.gov.uk/ energy/whitepaper/index.shtml), the UK government has set four goals for the measurement of successful energy policies: ∑ to put the UK on a path to cut carbon dioxide emissions – the main contributor to global warming – by some 60% by about 2050 with real progress by 2020 ∑ to maintain the reliability of energy supplies ∑ to promote competitive markets in the UK and beyond, helping to raise the rate of sustainable economic growth and to improve our productivity; and ∑ to ensure that every home is adequately and affordably heated. These goals do not lack ambition, and it would perhaps be cynical to observe that the short-term results must be seen against a background of declining manufacturing industry, and the long term goals are safely in the distant future. It is also worth noting that the deregulation of electricity supplies in the UK has lead to reductions in costs, and it is in this light that the cost increases associated with the climate change levy should be seen (Table 10.1). The levy came into effect in the UK in April 2001 and the UK Government expected it to generate £1 billion in its first year, all of was to be returned to business through tax cuts and additional support for energy efficiency measures. However, of current energy use, and the belief that the decline of manufacture leads us into a low energy economy, Oswald observes that we should ‘forget software versus steel ingots. The new economy runs on petrol and aviation fuel’ (Oswald, 2000). DTI figures for the year 2000 show that industry uses 21% of the UK’s energy, compared with 38% for transport. It is therefore important that composites manufacture should seek to minimise the transportation of raw materials and finished products. For example, if we are Table 10.1 Predicted increase in fuel costs due to the UK climate change levy Fuel

Increase in cost per kWh(p)

% increase

Gas Electricity Coal

0.15 0.43 0.15

20 12 16

Source: http://www.entech.co.uk/

Clean production

211

to make a thermoset composite in Europe, then rapeseed or Euphorbia can be used to provide raw materials for a polyurethane or epoxy matrix, binding hemp or flax reinforcement. If we are to make a similar product in Malaysia, then palm nut oil is more readily available for the resin precursor and palm oil bunch fibre for the reinforcement, this second material being particularly appropriate because it is currently a waste product from the first.

10.2.1

Energy tariffs

It should be noted that some processes are necessarily used together. For example, the raw materials for thermoplastic injection moulding must be compounded, although this may be done by a raw material supplier rather than on-site at the manufacturer. Table 10.2 shows estimates of the intensity of energy use for various composite manufacturing methods. The energy costs most relevant to composite manufacturers are electricity costs. Even though services such as compressed air and vacuum are often employed in composites manufacture, these services are ultimately dependant on electricity. Suppliers set up tariffs for electricity supply to encourage the use of electricity when the greatest spare capacity is available. For the commercial user there will be a number of factors to be considered as well as the usual cost per kWh. The Energy Efficiency Best Practice Programme Good Practice Guide No. 292 Energy in Plastics Processing – a Practical Guide lists four main factors: ∑ Maximum power requirement (MPR) in kVA – this is the maximum current that the site can draw. This is a fixed charge and the opportunity for matching it to the likely peak demand when taking up new premises is an important one. If energy intensive plant is used, starting the machines in sequence rather than all at once may avoid tripping the main circuit breaker, if working near the MPR. ∑ Maximum demand (MD) in kVA or kVAh – this is a measurement of the current actually drawn at the supply voltage, averaged over half an hour. Again, in the ideal situation, energy intensive plant start-ups should be staggered, allowing time for the process to settle down before starting the next machine. ∑ Power factor (PF) Alternating current distribution systems are affected by the nature of the load connected to them. In an ideal situation the supply voltage and the current through the load are synchronised. This ideal situation occurs if the load attached is a simple resistor. The only plant of this type in wide industrial use is the heater element. Other typical loads such as motors have a reactive element to them – in other words, are used to create magnetic fields. A reactive load will cause the supply voltage and the load current to move out of phase. The effect of this as far as the supplier is concerned is that more generating capacity is required to supply the load. A PF of 1 means that the supply voltages and currents are synchronised, as the PF declines towards 0, the lack of synchronisation increases, and at a certain trigger level, a

Long or short fibre reinforcement

Blend/ compound/ mix

Compounding (extrusion)

TP

S

Thermoplastic injection moulding Compression moulding (GMT) Thermoset injection moulding Contact moulding Dough moulding SRIM/RRIM

TP

S

TP

L

4 (greatest – energy input is mechanical) 6 (greatest – energy input is mechanical) – 1

TS

S

2

TS

L

TS

Resin transfer moulding (RTM) Vacuum infusion

Mould closure

Process heating

Process cooling

Post-process heating/ cooling

In-process waste*

Total

1

1

2(cooling strand)

–

8

2 (includes drying polymer) –

1

–

10

2

–

–

3

–

–

1

–

–

–

–

S

3

2

3

–

–

1 (reusable scrap from sprue, etc) 2 (trimmings from stock materials) 2 (non-reusable scrap from sprue, etc) 1 (brushes, solvents, etc) 1

TS

L/S

2

–

3

–

TS

L

1

–

3

–

TS

L

1

–

2

–

3 (optional: for best properties) 3 (optional: for best properties) 3 (optional: for best properties)

1 (reinfor6–9 cement trimmings) 2 (reinforcement 6–9 trimmings and purging solvents 4 (diffusion 8–11 materials are one trip)

6

7

3 9

Scale is arbitrary – the larger the number, the more energy used. SRIM = structural reaction injection moulding. KRIM = reinfored reaction injection moulding. GMT = glass mat thermoplastic.

Green composites

Matrix (thermoplastic and thermoset)

212

Table 10.2 Estimates of energy intensity for composite manufacturing processes (*in-process waste is energy related, and must be considered in terms of environmental impact)

Clean production

213

premium charge will be levied. If the applied load has a capacitive element, the phase difference will be affected in a reverse direction, and there exists the possibility of correcting the PF by the addition of banks of capacitors across the supply. However, for this technique to work effectively it does rely on a certain consistency of load. It should be noted that lightly loaded motors have the most effect on PF, and therefore motors should be specified to match the expected loads. ∑ Load factor (LF) The most desirable situation for the electricity generators is a constant demand for power. This supplier must have capacity to meet peak demand. If there are wide variations demand, this capacity will be standing idle. Also, the costs associated with maintaining a network are to a degree fixed. Therefore, it is desirable to maximise the use of the distribution network in order to offset these costs with revenue. LF is the fraction of hours in the day that power is used on site and allows the generator to charge a premium for inconsistent use of power. Other methods worth considering are the installation of insulating jackets on the heated elements of moulding machines and the upgrading of motors to energy efficient types when replacement is due.

10.2.2

Materials

The obvious choices for materials are recent biological origin fibres and polymers, because the ultimate aim of clean manufacturing is to emulate the natural cycles of material use. However, some researchers (Gerngross and Slater, 2000) have estimated energy used in the manufacture of plant origin polymers (in this case polyhydroxyalkanoate (PHA) and polylactic acid (PLA)) compared with commodity polymers such as polyethylene (PE), polyethylene terephthalate (PET), and concluded that these fossil origin polymers require less energy to manufacture than PHA or PLA. However, it should be noted that fossil origin polymers are made on a much larger scale and with technologies that have been developed over the past half-century.

10.2.3

Production processes

10.2.3.1

Hydraulics versus electrics in injection moulding

The injection-moulding machine is the workhorse of short fibre composite manufacture. Manufacturers are offering moulding machines powered directly by electricity. The advantages claimed by the manufacturers are as follows.

10.2.3.2

Reduction of energy use

Conventional machines use hydraulic power. To obtain accurate control with hydraulic fluid, the temperature of the fluid must be kept within prescribed

214

Green composites

limits. This means a non-productive ‘warming up’ stage upon start-up, and the use of cooling water during the productive running of the machine. Machine manufacturer Fanuc claims a reduction of 90% with electric machines in the requirement for chilled water for this reason. It is usual to keep the hydraulic prime mover running for the duration of the working day. If electrical motive power is used, it is possible to switch the power on and off within a single injection cycle and to recover energy during the deceleration of the motor. Independent energy measurements (Dawson et al., 2002) support these claims. Two machines of 100 tonne capacity were run back-to-back using the same mould. The servo-electric machine used one-eighth of the energy per kilogram of product produced compared to the servo-hydraulic machine. The power factor for the electrical machine was measured at 0.985 compared with 0.489 for the hydraulic machine.

10.2.3.3

Process control improvements

With an electrical machine it is possible to achieve more consistent control of the pressure profile (measured within the injection barrel) during injection and the packing phase. This allows the machine to produce more consistent products and to cope with a wider range of raw materials including those with uncertain processing properties such as recyclate.

10.2.3.4

Cleanliness (no leakage of hydraulic fluid)

With regard to cleanliness, hydraulic machines still score when it comes to larger mouldings and that development of hydraulic machines still continues.

10.2.3.5

Supercritical CO2 (SC-CO2)

If liquid CO2 is heated at atmospheric pressure, it will boil at –60 ∞C. However, if the liquid gas is constrained and the pressure is allowed to rise, the CO2 can be heated far above its normal boiling point, whilst still remaining liquid. If the temperature is raised above the critical point (defined as the temperature at which a vapour cannot be liquefied by increase of pressure alone) SC-CO2 is formed. SC-CO2 exhibits properties of both a liquid and a gas. It has gas-like properties of high diffusivity and solubility, yet the density of a liquid. If SC-CO2 is mixed into the polymer melt, it insinuates into the interstices between the long chain polymer molecules and provides internal lubrication resulting in a lowering of the apparent viscosity of the melt, with an associated lowering of melt temperatures (see Table 10.3). These results were presented by Brooks et al. (2002) and were obtained using a MCP minimoulder (from MCP Equipment, a division of Mining Chemical Products Ltd, UK) modified to accept a cavity transfer mixer. These results show how the melt temperature can be lowered by the addition of SC-CO2 without significant raising of the melt viscosity. This results in a

Clean production

215

Table 10.3 Temperature profiles for a general purpose polypropylene (GP-PP) moulding with varying amounts of supercritical CO2 Increasing amounts of CO2 Æ

No CO2

Screw Cavity transfer mixer Static mixer Block Plunger Hot-tip

T = To

T = To

T = To – 20

T = To – 30

T = To – 40

210 210

210 210

210 210

210 210

210 210

210 210 210 225

210 210 210 225

190 190 190 205

180 180 180 195

170 170 170 185

lowering of the net energy input to the process. The mouldings produced by this method are either foamed, or if the packing pressure is raised, solid polymer mouldings. The CO2 entrained in the mouldings diffuses out over a period of a few hours after demoulding. It is likely that the lowering of viscosity could be exploited to extend the range of materials processed by thermal injection moulding to include injection over long fibre reinforcements in a fashion similar to SRIM or RTM.

10.3

Limiting the environmental impact of processing

In the previous section the use of energy in composites manufacture was examined. The discussion of energy saving was directed at the most energy-intensive high volume processes. In this section the processes that are broadly low users of energy are examined. There is a trade-off between high capital cost, high volume manufacturing methods and low capital cost, low volume methods. The high volume methods are necessarily economical in terms of in-process waste, but the low volume methods, although mostly low energy users, tend to produce more in-process waste. This section will examine some of these methods and suggest possible methods of reducing the waste associated with them. These methods are exclusively thermoset resin-based, presumably because the low viscosity of thermoset resins suits the impregnation of fibrous reinforcement without the need for the high pressures and temperatures associated with thermoplastic materials. It is particularly the high pressures associated with processing high viscosity thermoplastic resins that lead to the high capital costs of the manufacturing plant. Thermosets are supplied to the manufacturer in an unpolymerised form. The consequence of this is that thermoset resins are characterised by having low molecular weights (and are therefore volatile) and are formulated to react readily under processing conditions. The workforce is hence required to work with highly mobile and chemically active materials. Note that this holds true for natural resins as well as for fossil origin materials. For example, cashew nut

216

Green composites

shell liquid is classed as an ‘irritant’ (a non-corrosive substance or preparation which, through immediate, prolonged or repeated contact with the skin or mucous membrane, may cause inflammation) and its most frequently used natural origin cross-linker (formaldehyde) is a Class 3 carcinogen (a substance that causes concern owing to possible carcinogenic effects but for which available information is not adequate to make satisfactory assessments) under Schedule 1 of the Chemicals (Hazard Information and Packaging for Supply)(CHIP) Regulations 1994 classifications. Strategies for safe use of such materials already have well-established legal frameworks such as the Health and Safety at Work Act (1974) and The Control of Substance Hazardous to Health (COSHH) Regulations (2002). However, the precautionary principle, and the desire to minimise waste, encourages the manufacturer to take further steps to isolate the workforce from contact with these materials. Manufacturing techniques suited to this will be examined in this section. In-process waste will also be examined. Examples of waste streams are: ∑ solid, e.g. autoclaving: bleeder fabrics, release films and vacuum film ∑ liquid, e.g. resin transfer moulding: solvents from purging processes ∑ vapour, e.g. contact moulding: vapour from open mould processes and spraying. The survey of processes will start with the open mould process of contact moulding, as this is the process that popularised the use of composites, and then explore a number of closed mould processes that may have better clean credentials. These include soft-top tooling methods such as autoclaving and vacuum infusion, and the resin injection techniques of resin transfer moulding and reaction injection moulding.

10.3.1

Contact moulding

The low capital costs of start-up (bucket, brush, roller, shed) lead to the rapid growth of low volume kit car and boat manufacture and the widespread acceptance of composite materials. Contact moulding consists of the following stages: ∑ Buck manufacture – the buck is a master pattern of the article. This master can be a wooden frame, covered in wire mesh, with the final surface finish being plaster. The surface is then treated to make it impermeable (cellulose acetate being one option) and coated with a release agent. ∑ Mould manufacture – a gel coat of neat (usually polyester) resin is then painted on to a thickness of about 0.5 mm. As a rule of thumb, the next layers of fibre and resin should be laid up about twice as thick as the finished mounding is expected to be. ∑ Article manufacture – the mould is then gel coated (slightly thinner coat than for the mould). After the gel coat has dried to a tacky finish, the reinforcement is laid up on top of it. Resin is brushed in and stippled and rolled as before to produce a well-consolidated composite (say 30 vol % reinforcement).

Clean production

217

This method clearly suits the artisan and is very suitable for short production runs. Craft level wet hand lay-up methods were the mainstay of composite production throughout the 1960s, favouring low volume high value applications such as specialist sports car manufacture (Köster, 1990). The method is quite economic as far as materials’ use is concerned, as resins are mixed in quantities to suit the working time (pot life) of the resin and the reinforcements are placed in the mould in relatively small pieces, meaning that not much material is wasted in the form of off-cuts. Semi-automatic techniques using spray-gun technology have developed, where the simultaneous application of resin and reinforcement speeds up the manufacturing process. However, if the exposure of the operator to potentially harmful materials is considered, this process is not so satisfactory. Consider the large surface area of the open mould and the need for the operator to work in close proximity to uncured resins particularly during mixing. In addition, the manual application of the resin and consolidation of the composite means that the risk of exposure to volatile organic compounds (VOC) is high. Increasingly, strict legislation to limit styrene emissions (in this case from polyester resin systems) has been the key factor in promoting the increased use of techniques that avoid the need for such concentrated exposure of the operator to VOC from the resin system. The obvious route to improve this situation is to cover the mould, either by a flexible impermeable membrane, or a matching solid mould top (often called ‘matched tooling’).

10.3.2

RIFT (resin infusion under flexible tooling)

The easiest upgrade from simple contact moulding is RIFT (resin infusion under flexible tooling). The technique has a number of varieties, include Lotus cars’ vacuum assisted resin injection (VARI) and Seemans composites resin infusion process (SCRIMP). RIFT can use tooling based on that employed in contact moulding. As with contact moulding, the mould is release treated and gel coated, and dry reinforcing fibre is laid up in the mould. A flexible membrane is then laid in the mould and sealed at the edges. The mould cavity is then evacuated and the pressure of the atmosphere is used to infiltrate the resin through the reinforcement. However, RIFT processing requires a low viscosity (possibly high solvent level) resin system with a long time to gelation or ‘pot life’. Therefore, RIFT is limited to lower production volume mouldings such as niche cars and boats.

10.3.3 ∑ ∑ ∑ ∑

RIFT summary

VOC emissions are estimated at 10% of open mould processes. Existing moulds can be adapted. Large mouldings can be produced (30 metre boat hulls). Heating and ventilation costs are reduced.

218

Green composites

∑ Labour savings of over 50% over hand lay-up are claimed. ∑ The quality of mouldings is enhanced (higher fibre volume means increased strength). ∑ The reproducibility of moulding is improved because of the uniformity of consolidation of the mouldings. ∑ The laminating process can be integrated with the addition of cores, stiffening, inserts, etc. ∑ Most resins and reinforcements can be used. ∑ There is an add-on cost of moulding disposables, e.g. bleed fabrics and membranes, which must be balanced out against labour savings, reduced styrene levels and improved working environment.

10.3.4

Pre-pregging (autoclaving)

Prepregs are reinforcement materials prepared for the manufacturer by the PRE imPREGnation of the reinforcement by partially polymerised resin. The builder then has only to cut the desired shape from the prepreg, drape it over a form and apply heat and pressure to consolidate and cure the finished article. This method produces high quality void free components and is popular with the aerospace industry who have learnt to cope with its slow cycle times (for example, 80 hours for a helicopter rotor blade). The workforce at the manufacturing site is not exposed to unreacted liquid resins or VOC because resin mixing is done at the prepreg manufacturing site. However, the list of consumable items in autoclave processing is considerable: ∑ Release agent – allows the release of the cured prepreg component from the tool. ∑ Peel ply – optional layer allowing free passage of volatiles and excess matrix during the cure. Can be removed easily after cure to provide a bondable or paintable surface. ∑ Bleeder fabric – usually made of glass fabric felt and absorbs the excess matrix. The matrix flow can be regulated by the quantity of bleeder to produce composites of known fibre volume. ∑ Release film – this layer prevents further flow of matrix and can be slightly porous (with pin-pricks) to allow only the passage of air and volatiles into the breather layer above. ∑ Breather fabric – provides the means to apply vacuum and assists removal of air and volatiles from the whole assembly. Thicker breathers are needed when high autoclave pressures are used. ∑ Vacuum bag and sealant – provides a sealed bag to allow removal of air and consequent formation of a vacuum.

10.3.5

Prepregging/autoclave summary

∑ Good control of fibre orientation using unidirectional prepreg.

Clean production

219

∑ High temperature properties available from thermoset resins. ∑ However, it should be noted that, if ultimate mechanical properties are required, then plant origin fibres are not the obvious place to start due to limitations, as yet not overcome in fibre strength. ∑ Inexpensive versatile tooling concepts. ∑ Resin formulation and impregnation is the prepreg manufacturers’ responsibility – limits VOC exposure to the workforce at the article manufacturing plant. ∑ Expensive in raw materials both prepreg and throw-away breather, vacuum bagging, etc. ∑ Prepreg has limited shelf life. Refrigeration of raw materials is required. ∑ Expensive and time consuming hand lay-up is required. ∑ Long energy intensive cure cycles (may be multiple cycles to allow the build of different functional layers in the component, e.g. heater elements).

10.3.6

Double rift diaphragm forming (DRDF)

The relatively new method of DRDF strives to integrate the advantages of RIFT and diaphragm forming into one low cost continuous operation. The environmental benefits of this are that the low levels of VOC exposure are retained without the necessity of using a large amount of one-trip disposable material. DRDF can use any type of fabric or mat (unidirectional fabric, felt mat): the reinforcement is cut to the two-dimensional (2D) drape predicted shape. The cut shapes are placed flat (no three-dimensional (3D) lay-up) between two elastomeric membranes. A vacuum is created and resin is then infused into the fibres using the pressure gradient, resulting in a simple 2D flow front. Once infused, the fabric can be vacuum formed over a ‘low cost’ mould tool. Wrinkling is eliminated because the fabric material is held between the elastomeric membranes that dissipate stress concentrations during the forming process by applying a uniform pressure over the entire surface, which eliminates folds, tears and wrinkles. The mould can be made from any low cost material, such as wood or plaster. This is because the composite does not come in contact with the mould surface. The mould surface also does not have to be ‘smooth’, obviating the need for skilled polishing and mould release compounds. Being a one-step injection and moulding process, a preforming mould and the labour to stack and trim are not required.

10.3.7 ∑ ∑ ∑ ∑ ∑ ∑

DRDF summary

Minimal hand lay-up required. Fibre placement is predictable and repeatable. Less material waste. Low labour content. Low VOC exposure. No complex 3D flow modelling or vent placement required.

220 ∑ ∑ ∑ ∑ ∑

Green composites

Infusion delivers results comparable to prepreg quality. 1% void content with high fibre to resin ratio. ‘Rapid prototype’ low cost mould tools can be used (or conventional moulds). Mould can be easily modified. No mould release or polishing required.

10.3.8

RTM/RIM

Investment in matched tooling – a top and a bottom, male and female tool set is another possible route to the reduction of the environmental impact of composites manufacture. These closed mould processes using preplaced reinforcements (resin transfer moulding or RTM and structural reaction injection moulding or SRIM) use layers of reinforcement, for example, hemp mat, preplaced in the mould before injection of the resin matrix. Since the mixed thermoset reagents are initially of low viscosity, injecting to impregnate through the mat fills the mould and produces a composite article. Using this technique, reinforcement loadings of near the theoretical maximum can be obtained. This may be an advantage in clean composite production because the reinforcement is the cheapest and least dense component of the composite.

10.3.9

Resin transfer moulding (RTM)

The RTM method was devised for the manufacture of aircraft radomes in the late 1940s. It has since proved an attractive first step into the production of closed mould composite articles. RTM machines usually inject long pot life resins at low pressures (less than 10 bar) and slow speeds, producing high quality mouldings. The low injection pressures mean that tooling can be composite and consequently the whole process is of low capital cost. Injectors for such systems need be little more than heated pressure pots. The resin is introduced to the chamber and injected into the mould by means of air pressure. The mould cavity is often evacuated before injection to reduce the chance of bubbles being captured in the finished article. This method does require the external mixing of the resin and curing agent again VOC exposure may be a problem. More sophisticated multistream systems are available whereby the resin is mixed with its catalyst just before it enters the mould. Unfortunately, the low injection rates mean that the resin/catalyst stream must pass through a static mixer (a tube with a convoluted internal geometry) in order to make mixing complete. It is necessary to flush this chamber with air and solvent (usually acetone) at frequent intervals to avoid blockages with cured resin. Disposal of the waste solvent may be a problem, although if the volume is large enough it is possible and economical to recover the solvent for further use by distillation. Filling may take from 5 minutes to up to an hour depending on the fibre loading and part geometry. It may also be necessary to control mould venting to

Clean production

221

make sure that all parts of the moulding are filled with resin. The purging from these vents will also need to disposed of in a safe manner: this material can be allowed to react to produce a relatively inert solid before disposal.

10.3.10 RTM summary ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑

Short simple cure cycles. Limited exposure of workforce to VOC and uncured resin. Precise control of fibre volume. Complex components easily manufactured. High quality low voidage parts. Minimal amount of waste (solvent flushing and vent purged resin). Good dimensional control of finished product. Large parts may need multiple injection ports. Resins may need to be mixed on site, therefore making it more difficult to guarantee correct mixing of components. ∑ Tooling (double-sided, solid) may be expensive compared to that required for contact moulding or RIFT. ∑ Reinforcement may be shaped to fit in the mould as a separate process (preforming).

10.3.11 Structural reaction injection moulding (SRIM) Reaction injection moulding is a faster, more sophisticated, version of RTM. Reaction injection moulding systems have two components that are mixed and then co-react, usually with the addition of heat, to form a solid cured resin. RIM technology was developed for the polyurethane industry. Some polyurethane formulations have very fast cure times (approximately 30 seconds) that necessitate the two components being kept separate until just before injection. Mixing of the ingredients is carried out by squirting impinging jets at each other within a valve mounted directly onto the mould (known as the mixing head). The design of the mixing head does not require a solvent/air purge to clean it. Currently, the polyol component of polyurethane is available from plant oil raw materials, but the isocyanate curing agent is only made from fossil origin precursors. Low (80– 150 bar) injection pressures allow lower tooling costs compared to thermal injection moulding. Long and short fibre composites can be made by RIM. The short fibre method is known as reinforced reaction injection moulding (RRIM). In RRIM short (typically up to 10 mm) fibres are mixed with the resin. Natural fibres may well prove less abrasive than the currently favoured glass fibres, resulting in less wear to the mixing chamber, pumping equipment and moulds. For long fibre work, RIM technology can be used to inject over preplaced reinforcement (the so-called ‘preform’). This technique is known as structural reaction injection moulding (SRIM).

222

Green composites

Strict control of the mixing ratios is required. Flory (as described by Macosko, 1989) notes that if a polyurethane polymerisation reaction of a diisocyanate and a diol goes to 98% conversion when the volumes dispensed are held exactly at the stoichiometric ratio, the molecular weight of the reaction product will be 50 ¥ the number average molecular weight of the unreacted mixture. However, if the dispensed volumes give an error of 2% from the stoichiometric ratio, then the number average molecular weight of the reaction product at the same conversion value for the diisocyanate drops to ª66% of the theoretical value at the stoichiometric value. This has clear implications for operator exposure to unreacted components upon opening of the mould. The control of the dispensing ratios is usually by real-time computer or programmable logic controller (PLC).

10.3.12 RRIM/SRIM summary ∑ SRIM technology allows high fibre reinforcement content. ∑ RRIM/SRIM is not limited to polyurethanes, other thermoset resins can be used. Examples that can be made from biological precursors are epoxies and polyamides. ∑ The design of RIM machines does away with the need to solvent flush between shots (a drawback of RTM). ∑ A precision engineered mixing head allows jets of the component materials to be fired into one another at high speed (Reynolds number of greater than 300). Coupled with precision control of dispense ratios, this high quality mixing minimises residual unreacted component materials in the finished mouldings. ∑ Residence time within the mixing head is of the order of 20 ms, therefore very fast reacting chemical systems can be used. ∑ RRIM/SRIM has a high capital cost of plant (equivalent to thermal injection moulders) and higher cost of tooling – higher than RTM, but not as high as thermal injection moulding. ∑ RRIM/SRIM machines minimise exposure of the workforce to unreacted chemical components.

10.4

The use of additives

The resin systems used as the matrices in the manufacture of composites are rarely the unadorned formulation. Additives to the matrix mixture are used to assist in processing the composite into a finished article and to tailor the properties of the finished article. Rudd et al. (1997) list the following as examples: ∑ shrinkage control additives (low profile additives) to improve cosmetic surface finish ∑ fillers to reduce article cost, reduce flammability of the finished product and to reduce the heat build-up during the curing process.

Clean production

223

Other additives of relevance to green composite manufacturers are: ∑ lubricants – lubricating the flow of thermoplastic polymers during processing ∑ plasticisers – increasing the toughness of the manufactured article at the expense of rigidity ∑ colourants ∑ biocides and antimicrobial agents. These additives are usually added in small quantities (fillers being an exception) and, with the exception of the biocides, are unlikely to have any effect on the gross biodegradation of the composite. However, the selection of additives for biodegradable polymer systems should be undertaken with care to avoid compromising the biodegradability certification of the finished article. An accepted standard by which biodegradability is assessed for biodegradable polymer matrix composite articles is the ‘DIN CERTCO Certification scheme: products made of compostable materials’, 3rd revision, Berlin, 2001.

10.4.1

Shrinkage control additives

Thermoset resins used for resin transfer moulding can have high levels of diluent to lower the viscosity of the resin system and hence improve the penetrative ability of the liquid. As a result of this, the shrinkage upon solidification can be up to 8% (Rudd et al., 1997). The addition of reinforcement reduces the apparent shrinkage, but a number of thermoplastic admixtures are also used to control the amount of shrinkage. The most acceptable for green composite manufacture is polyvinyl alcohol (PVA). PVA is also used as a cloth size and is removed from the cloth before dying by bacterial action. PVA is a fossil origin material, but it is biodegradable.

10.4.2

Plasticisers and lubricants

Plasticisers increase the flexibility of the polymer product and also decrease the viscosity of the polymer melt, a role also played by lubricants. In terms of biological origin materials, epoxidised soyabean oil is used as a plasticiser and heat stabiliser in PVC production. However, concerns have been voiced over the migration of the epoxidised material into the environment when used in contact with foodstuffs. Epoxidised linseed oil and tall oil are also used as PVC plasticisers. It may be that these materials will also find application in biopolymers. Internal lubricants perform a similar role to that of plasticisers in terms of easing the flow of material during processing. External lubricants are also added to the formulation, but function as release agents. Richter (2000) classifies the majority of lubricants as waxes or fat derivatives, and notes that the fatty acids are mostly obtained from beef tallow now that the use of whale origin oils (spermaceti) has been banned. Natural origin waxes are said currently to be of little importance to the polymer industry in this function.

224

Green composites

The obvious sustainable origin mould release agents are waxes such as bees’ wax, carnauba wax and rape seed oil (for example, as marketed by Leahy Limited as BioForm). These materials are effective mould release agents but care must be taken to avoid overdosing the mould surface. This will have the consequence of producing greasy bloom on the surface of the moulding. DIN CERTCO (2001) list the following processing auxiliaries as certified compostable up to levels of 10%: ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑

benzoic acid/sodium benzoate erucic acid amide glycerol monostearate glycerol monooleat natural waxes paraffins, paraffin waxes (natural) polyethylene glycol (up to molecular weight 2000) stearates

whilst the following can be included at levels of up to 49% (DIN CERTCO, 2001): ∑ ∑ ∑ ∑ ∑

glycerol sorbite citric acid ester glycerol acetates xylite.

10.4.3

Colourants

Some colourants may be objected to due to inherent toxicity, but environmentally acceptable substitutes of either mineral or vegetable origin are increasingly available. DIN CERTCO (2001) lists the following mineral colourants as certified compostable up to levels of 49%: ∑ ∑ ∑ ∑

carbon black iron oxide graphite titanium dioxide.

10.4.4

Flame retardants

Commonly, the chemicals used to reduce the flammability of polymers are chlorinated and brominated compounds (sometimes with antimony-based synergists) used as flame quenchers, and phosphorous compounds to improve char strength. These are regarded as undesirable due to perceptions of their toxicity. Alumina trihydrate (Al2O3.5H2O) releases its water of hydration when subjected to heat and hence limits the propagation of combustion. However, it is

Clean production

225

added in large amounts to the polymer (50–60%), and may at the upper limits of addition, compromise the biodegradability of the article.

10.4.5

Fillers

Fillers are the most commonly used additives. Fine mineral powder fillers are added as nucleating agents in small (~1%) quantities to limit the size of crystalline structures. This is of limited application with biopolymers as the complexity of the molecules in most biopolymers limits the degree of crystalline structure formation. DIN CERTCO (2001) lists the following fillers as certified compostable up to levels of 49%: ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑

aluminium silicates ammonium carbonate calcium carbonate calcium chloride dolomite gypsum mica kaolin chalk sodium carbonate natural silicates silicon dioxide; quartz talc wollastonite vegetable fibres wood flour/wood fibres cork bark starch rye flour and other flours starch acetate (up to a substitution level of 1).

10.4.6

Biocides and antimicrobials

Biocides and antimicrobials kill or limit the growth of viruses, bacteria and fungi. They may be useful in extending the life of a biodegradable composite, but it should be noted that the employment of such materials in the environment is likely to lead to the evolution of resistant strains of microorganism. Ochs (2000) provides a comprehensive list of available antimicrobials and lists the following properties as desirable: ∑ low toxicity to humans, animals and the environment ∑ easy application

226

Green composites

∑ no negative impact on the properties and appearance of the article ∑ storage stability and long lasting efficacy. The method of application of antimicrobials is to mix them physically into the resin formulation. Over the life of the product, the agent diffuses out on to the surface of the moulding.

10.5

End-of-life disposal strategies

This section uses the automotive industry as an example for examining the current position on management of waste streams, and how the use of green composite materials may be integrated with current waste management practice. Problems facing the introduction of green composites include the demonstration of biodegradability and the separation of biodegradable materials from mixed waste streams. In the former case, standards and product marking systems are required and, in the latter, economic production and recovery routes. The recovery of a product is dealt with in Directive 75/442/EEC recycling (Council Directive, Official Journal of the European Communities, 1975) and says that it is the treatment of waste for re-use. Recovery is also classified as ‘energy recovery’ where, at the end of life, the packaging is combusted to produce energy. McCrum et al. (1997) argue that taking into account the total life cycle of polymers versus steels in the production of automobiles, plastic materials require three times less energy to produce than steel. If the plastics are combusted at the end of life and the energy recovered, then the energy required to produce the steel is five times greater. Recycling purports to be the reprocessing of waste materials in a production process, but also includes ‘organic recycling’, which is the aerobic and anaerobic treatment of waste to produce stabilised organic residues or methane. Landfill is not considered to be a form of organic recycling (Council Directive, Official Journal of the European Communities, 1994). In tandem, the evolution of new composting technologies, such as ‘in-vessel’ techniques – leading to a cycle time of about 9–14 weeks – has made this a more favourable end-of-life disposal route for both ecological and economic reasons (De Wilde and Boelens, 1998). The reasoning behind the move away from the landfilling of biodegradable waste is that the degradation of organic matter produces methane gas, which is classified as a greenhouse gas and thus contributes to global warming. The strategy for the future disposal of organic waste centres depends on the creation of dedicated composting facilities. These facilities will allow the waste to be composted in a controlled environment and the methane captured and used as a ‘biogas’ fuel (Council Directive, Official Journal of the European Communities, 1999). There are a number of problems with the current end-of-life disposal routes for products manufactured from thermoset and thermoplastic polymers. Thermoset polymers, once cured, are notoriously difficult to recycle. This is due to the curing mechanism being chemical and generally irreversible in any economic

Clean production

227

Table 10.4 Cost of waste management options in the Netherlands, Germany and Belgium (in US$ per tonne) Option

The Netherlands

Germany

Belgium

Composting Incineration Landfilling

60 135 105

151 486 402

80 110 75

Adapted from: De Wilde and Boelens (1998).

fashion. Thermoset composites such as sheet moulding compound (SMC) and bulk moulding compound (BMC) can be recycled at end of life, and the process for performing this is to grind the scrap and then add to virgin material at levels of up to 20% (Maxwell, 1994). The options that are available for the disposal of both thermoplastic and thermoset components are currently the same options that are available for the disposal of most discarded items: incineration and landfilling, and unfortunately the majority of polymeric waste is currently being landfilled (McCrum et al., 1997). The advantages of biodegradable composites in terms of recycling over the mechanical recovery of fossil origins composites are clear. In the composting– regrowth route, the composite constituents are broken down to a fundamental level and then largely rebuilt by the biological synthesis of plant growth. In a mechanically recovered thermoplastic polymer–matrix composite, the shear forces and high temperature act upon the long chain molecules during recompounding, shortening some of the chains. Hence a mechanically recycled polymer will have reduced mechanical properties and a larger degree of uncertainty in processing properties. Mechanical recyclate is also somewhat limited in choice of colours, with black predominating. This leads to a significant tailing off in the possible markets for such material. A comparison of three different waste management options in three European countries is shown in Table 10.4. The table shows how Germany is attempting to use economic instruments to force end users to recycle and compost waste instead of opting for the more traditional routes of disposal. In considering the market for biocomposites, it is the opinion of the author that, in preparing business strategies for enterprises using biopolymers, entrepreneurs should base predictions upon the effect of legislation, rather than on the more uncertain projected effects of environmental considerations. Further governmental legislation that has an impact on the disposal routes available for materials is the European Council Directive 1999/31/EC on the landfill of waste (Council Directive, Official Journal of the European Communities, 1999). This brought into enforcement the ‘polluter pays’ principle; this is detailed in directive 75/442/EEC (Council Directive, Official Journal of the European Communities, 1975). In brief, the ‘polluter pays’ principle says that ‘the cost of disposing of waste, less any proceeds derived from treating the

228

Green composites Table 10.5 Strategy for the reduction of the amount of biodegradable waste that is landfilled Year

Amount (vs. 1995 levels) (%)

2006 2009 2016

75 50 35

Adapted from: Council Directive, Official Journal of the European Communities (1999).

waste, shall be borne by the producer or previous handler of the waste’. This could have a number of repercussions for industries that generate a large quantity of waste such as the tobacco and food industries. Another facet of the landfill of waste directive is that the amount of biodegradable refuse that is currently disposed of via landfilling is to be reduced over time. Table 10.5 shows the targets that are required for reductions in the landfilling of biodegradable waste and this is expressed as a proportion of the landfill waste that was produced in 1995.

10.5.1

Automotive waste streams

The end-of-life vehicle directive entered into force on 1 July 2002 and governs the disposal of vehicles at the end of their useful life. One of the fundamental aspects of the directive is that preference should be paid to re-use, recycling and recovery instead of disposal. In particular, the directive states that the recycling of all plastics from vehicles should be improved and the development of markets for recycled materials should be encouraged. The onus of these retrieval and recycling programmes and facilities is placed upon the ‘economic operators’, who include vehicle producers, distributors and insurers. The directive details targets for the re-usability, recovery and recyclability and these are shown in Table 10.6. Recycling can be interpreted to include ‘organic’ recycling, and recovery can include the recovery of energy (Council Directive, Official Journal of the European Communities, 2000). The reason why recycling is favoured for polymeric materials is that 80% of the energy required to produce polypropylene is expended on the Table 10.6 EC directive on end-of-vehicle-life recovery and recycling weights Date

Re-use and recovery (weight %)

Re-use and recycling (weight %)

1 January, 2006 1 January, 2015

85 95

80 85

Adapted from: Council Directive, Official Journal of the European Communities (2000).

Clean production

229

polymerisation and associated processes. If the polypropylene can be (economically) recycled to produce other parts at the end-of-life, then this is deemed to be environmentally better than chemical or energy recycling as the energy does not need to be expended to polymerise the feedstock (McCrum et al., 1997). However, the high shear and temperature conditions required to recover mechanically a thermoplastic inevitably lead to increasing uncertainty in the processing and mechanical properties of the recyclate. This makes mechanically recovered polymer unattractive to large parts of the market. Other problems relating to the recycling of thermoplastics from vehicles, besides separation, are concerns relating to disassembly. A study carried out by the Ford Motor Company (Maxwell, 1994) on the dismantling of the Escort car confirmed that there were diminishing returns in the recycling of plastics stripped from automobiles. In the first 10 minutes, 20 kg of recyclable plastics were recovered and in the next 10 minutes only 10 kg of plastics were recovered. The dismantling costs were calculated to be 50 p/kg after 20 minutes, rising to £12/ kg after 50 minutes. The latter figure is far in excess of the prices of many virgin polymers. Until manufacturers standardise on type and improve ‘design for disassembly’, the recycling of polymers from automobiles will remain an uneconomical proposition. The current expectation (Taylor, 2001) of the Ford Motor Company is that it will be an economic necessity to move away from dismantling towards shredding, in order to turn scrap cars into single material particles. These particles are then separable into recoverable waste streams. Further issues appear when an article that comprises a number of polymeric materials comes to the end of life, and a case in point is the recycling of the plastics from automobiles. McCrum et al. (1997) state that there are 16 different types of polymers used in a BMW 5 series and that, within these groups, there are different grades relating to parameters such as molecular weight that are optimised for performance. The solution to the sorting problem would be to standardise plastic types ‘across the board’, although this may present other problems such as increased weight and decreased functionality of the components. The separation of polymer waste streams is currently accomplished by hand – a situation that reflects the practice in the minerals industry of a century and a half ago. Pascoe (2000) surveys the possibility of automating separation, by the application of mineral processing technologies to the problem – it is likely that the application of automation to this problem will dramatically change the economics of separation. Automatic methods of separation, such as froth flotation and dense media separation can cope with waste streams containing metal– polymer mixtures and polymer–polymer mixtures.

10.5.2

Summary

∑ The main difficulties associated with increasing the use of biocomposites are associated with the development of an infrastructure spanning design to endof-life disposal, to allow the exploitation of their unique properties.

230

Green composites

∑ In particular, biocomposites must be separated from waste (especially fossil origin polymer matrix composites) if the advantages of these materials are to be exploited. The separation of large composite articles into different grades would be aided by adequate identification marks on the article such as the IBAW mark. However, much of the waste will be small-sized particles and therefore technology transfer of automated separation techniques must be implemented.

10.6

Future trends

This section is mostly a wish list of things desired and required to advance the use of green composites. It is to be hoped that the uptake will be an evolutionary process, rather than a panic-stricken revolution. Stages along the route will include end-of-pipe measures such as recycling of post-consumer waste and the use of partially green materials such as the polypropylene flax composites currently being used as car door liners.

10.6.1

Materials

10.6.1.1

Fibres

∑ Technical grades of fibres and reinforcements. These materials will be priced somewhere between fine linens and soil-stabilising fabrics. ∑ Methods of fibre preparation that are consistent and do not damage the fibres. ∑ Methods of non-woven mat preparation that do not damage the fibres. ∑ Revisiting the material science of fibre reinforcement to develop natural fibre reinforcements that are of equal sophistication to current artificial fibre reinforcements.

10.6.1.2

Matrices

∑ Development of easier to handle curing agents for thermoset systems. ∑ The development of high volume manufacturing capacity for bioresins, with the consequent lowering of material costs.

10.6.1.3

Methods

∑ Reduction in use of organic solvents. ∑ Move to net shape processing – reducing fibre waste. ∑ Improving isolation of work force from reactive chemicals.

10.6.1.4

Other factors

∑ Education of designers about the possibilities and limitations of green composites.

Clean production

231

∑ Controllable and programmable triggers for the process of biodegradation. ∑ Pricing structures based on the whole life cycle of green composite articles.

References Brooks, N., Willoughby, B., Dawson, A.J. and Tucker, N. (2002). Development of a minimoulding machine with liquid CO2 injection. Micro-moulding 2002 – advances and commercial opportunities in micro and miniature moulding, University of Warwick, 4th July 2002. Council Directive (1975). 75/442/EEC of 15 July 1975 on waste, Official Journal of European Communities L 194, 25/07/1975 P. 0039–0041. Council Directive (1994). 94/62/EC of 20 December 1994 on packaging and packaging waste, Official Journal of European Communities L 365, 31/12/1994 P. 0010–0023. Council Directive (1999). 1999/31/EC of 26 April 1999 on the landfill of waste, Official Journal of European Communities L 182, 16/07/1999 P. 0001–0019. Council Directive (2000). 2000/53/EC of 18 September 2000 on end-of life vehicles, Official Journal of European Communities L 269, 21/10/2000 P. 0034–0043. Crosby, P.B. (1979). Quality is Free – The Art of Making Quality Certain. New American Library, New York. Dawson, A.J., Rajamani, H.S., Collis, R., Owen, L.D. and Coates, P.D. (2002). Detailed energy measurements in injection moulding. Annual Technical Conference (ANTEC), Conference Proceedings. De Wilde, B. and Boelens, J. (1998). Prerequisites for biodegradable plastic materials for acceptance in real-life composting plants and technical aspects. Polymer Degradation and Stability, 59, 7–12, Elsevier. DIN CERTCO (2001). Certification scheme for compostable plastics, July 2001 (3rd revision). Available at http://www.ibaw.org/eng/downloads/baw_certification_engl.doc Fox, A. (2002). Ecocats – environment friendly marine transport, personal communication, 19th December 2002. Gerngross, T.U. and Slater, S.C. (2000). How green are green plastics? Scientific American, August, 24–29. Henty, R. (2002). MD - Henty Surfboards, personal communication, 19th December 2002. Imai, M. (1986). Kaizen – The Key to Japan’s Competitive Success. New York, USA: McGraw Hill. Kelly, A. (ed.) (1994). Concise Encyclopedia of Composite Materials. Oxford, UK: Pergamon. Köster, J. (1990). Advanced composite engineering: emerging large volume concepts and processes. Composites Manufacturing, 1, (2), June. McCrum, N.G., Buckley, C.P. and Bucknall, C.B. (1997). Principles of Polymer Engineering, 2nd edn, Oxford: Oxford University Press. Macosko, C.W. (1989). RIM: Fundamentals of Reaction Injection Moulding, SPE/Carl Hanser Verlag, Munich. Maxwell, J. (1994). Plastics in the Automotive Industry, Cambridge: Woodhead Publishing and the Society of Automotive Engineers. Ochs, D. (2000). Antimicrobials. In Plastics Additives Handbook, ed. Zweiful, H., Munich: Hanser. Oswald, A. (2000). Oil and the real economy: interview with Andrew Oswald, March 17 2000, www.oswald.co.uk.

232

Green composites

Pascoe, R.D. (2000). Sorting of waste plastics for recycling. Rapra Review Reports, 11, (4). Pearce, F. (2002). Despite all the talk, real change is as elusive as ever. New Scientist, 176, No. 237/5, 21/28 Dec., 18–19. Richter, E. (2000). Lubricants. In Plastics Additives Handbook, ed. Zweiful, H., Munich: Hanser. Rudd, C.D., Long, A.C., Kendall, K.N. and Mangin, C.G.E. (1997). Liquid Moulding Technologies, Cambridge, UK: Woodhead Publishing. Taylor, A.M.S. (2001). Director, Corporate Citizenship, Ford of Europe Inc, personal communication, 24th August. Thorpe, B. (1999). Citizen’s guide to clean production. Clean Production Action, Montreal, Canada. Tucker, N. and Lindsey, K.A. (2002). A Handbook of Automotive Composites, Shawbury: Rapra Technology. Von Weizäcker, E., Lovins, A.B. and Lovins, L.H. (1998). Factor four – doubling the wealth, halving resource use – the new report to the club of Rome. London: Earthscan Publications. Womack, J.P, Jones, T.J. and Roos, D. (1990). The Machine That Changed The World, New York, USA: Rawson Associates.

11 Applications M. HUGHES University of Wales, UK

11.1

Introduction and definitions

For millennia, humans have used materials readily available to them – stone, clay, mud, wood, bone and hide, for example – and have fashioned these to suit their purposes. With the coming of the industrial revolution in the eighteenth century and importantly, with the development of synthetic resins and plastics in the late nineteenth and early twentieth centuries, skill in the use of these natural resources has gradually declined. We now live in an age where we are generally unfamiliar with many of these traditional materials, to the extent that ‘natural’ materials are sometimes regarded with some misgiving and may indeed be viewed as being, in some way, inferior to the synthetic materials available to us today. Increasingly, however, we are becoming aware of the need to reduce the adverse effects of our activities upon the environment and, as a result, renewed interest is being shown in ‘traditional’ or ‘natural’ materials, particularly those that are derived from renewable resources. Composites are an important class of engineering materials that are finding increasing use in applications ranging from leisure goods to construction. Their excellent specific properties make them particularly attractive for applications in which weight saving is advantageous, such as transportation and aerospace. Nevertheless, despite their undoubted ubiquity, they exact a heavy price in environmental terms. Problems with end-of-life disposal and non-renewable, fossil-based raw materials are considered to be particularly challenging and are leading to resurgent interest in composite materials based upon natural, renewable resources. This chapter sets out to review and discuss applications for composites based upon renewable resources, so-called ‘green’ composites. To this end, the historical applications for these materials will highlighted and some examples of the early uses of green composites, at the dawn of the synthetic polymer age in the early twentieth century, will be given. The focus, however, will be on current and future potential applications for these materials and 233

234

Green composites

will include likely trends in usage. As with all materials there are limitations which will effectively govern the range of potential end uses; these limitations will be highlighted in the context of intended applications. Whilst there are certain technical barriers hindering the widespread adoption of these materials, there are also a number of market and consumer issues which will, in the short to medium term, affect the uptake of these materials. These issues will also be highlighted and discussed. Myths: over the past few years there has been much written about the potential for high performance composites based on natural fibres. Much of the early optimism about composite performance has now been replaced with a more realistic view of what can be achieved in performance terms. Nevertheless, there is still significant scope for performance improvement and whilst green composites may not find use in aircraft primary structures (although they nearly did…!), there is a raft of structural or semi-structural applications for which they may well be entirely suitable.

11.1.1

Definitions

Before discussing applications, it is worthwhile to describe first what is meant by green composites. Although elsewhere in this text alternative approaches to green, or ‘eco’, composites may have been put forward, for the purposes of the following discussion, green composites are defined as materials composed wholly, or in part, of constituents which come, ultimately, from a renewable resource. This definition applies to both the reinforcement and matrix phases of the composite. In terms of the reinforcement this could include plant fibres such as cotton, flax, hemp and the like, or fibres from recycled wood or waste paper, or even by-products from food crops. Within this definition of ‘fibres from renewable resources’ are regenerated cellulose fibres – rayon or viscose (as ultimately these too come from biomass), as well as natural ‘nano-fibrils’ of cellulose and chitin. Matrices could be of either an organic or inorganic nature. Organic matrices may be polymers themselves derived from renewable resources, such as vegetable oil or starch and may or may not be biodegradable, depending upon the type and chemical structure of the polymer. Alternatively, the organic matrix may be a synthetic, fossil-derived polymer and may be either virgin or recycled. For certain applications, it may be desirable for inorganic matrices to be used in preference – say, for improved fire performance or for resistance to biodegradation. Although not discussed specifically, these materials are also included within the definition of green composites. Wood fibre–plastic composites (WPCs), a particular subset of green composites, are currently receiving a significant amount of attention and are in widespread commercial production, particularly in North America. WPCs generally consist of a cellulose-based fibre in combination with a thermoplastic

Applications

235

polymer matrix. In view of their current importance as green composite materials, the term ‘wood fibre–plastic composite’ will also be used throughout this chapter. Although the distinction is to an extent arbitrary, this definition of green composite does not extend to include flat pressed wood-based panel products such as particleboard, fibreboard or other forms of engineered wood panels commonly used in construction and furniture making, for example.

11.2

Historical applications of green composites

There are numerous examples of green composites used in early times, from the straw-reinforced mud bricks of the ancients to the composite bows of medieval times.1 The history of ‘modern’ natural fibre-reinforced polymers can, however, be traced back to the advent of synthetic polymers in the early part of the twentieth century. Before this even, examples of the use of natural fibres with natural or semi-synthetic polymers exist. For instance, in the USA in the 1850s, shellac was being compounded with wood flour to mould union cases to display early photographs, whilst Lepage worked in France with albumen and wood flour to produce his decorative Bois Durci plaques.2,3 With the invention of bakelite phenolic moulding resin in 1909, it was not long before natural fibre, in the form of wood flour or waste string and rags, was added to form the earliest form of synthetic composite.4 These ‘new’ composite materials found application in consumer goods such as radio and speaker cases. Indeed, it was not until the commercial introduction of glass fibre and the concurrent development of ‘cold curing’ synthetic resins such as unsaturated polyesters and epoxies during and just after the Second World War that what we know today as synthetic composites became viable. Up to this time, the only practicable reinforcement was natural fibre, either organic plant fibre or inorganic asbestos. Even as late as 1947, Brown5 stated (with reference to research being carried out at the time into natural fibre-reinforced high pressure laminates) that, ‘the beneficial results of work . . . will result in a great extension in the application of this material’. It is interesting to speculate just where we would be today in terms of natural fibre-reinforced composite technology and applications had plentiful quantities of glass fibre not become available! During this intervening period, covering the 1920s, 1930s and early 1940s, a good deal of research was carried out into these early green composites. A key feature of these materials was, as was alluded to in the previous paragraph, that they were generally manufactured using high temperatures and pressures – so-called ‘high pressure laminates’. Manufacture by this method was necessary since the early resins had poor flow properties and/or required elevated temperatures to cure them. Nevertheless, the properties of some of

236

Green composites

these materials were, even by today’s standards, excellent and in view of this, it is worthwhile devoting some time and space to give brief mention to these early composites. Pioneering work was carried out in 1924 by Messrs Caldwell and Clay, into the use of fabric reinforced synthetic resins for airscrews,6 however, it was not until the 1930s that any significant interest was shown in the potential of these ‘synthetic’ composites as structural materials.4 Much of this early work on natural fibre reinforcement for synthetic resins was spurred on by the search for lighter materials for use in aircraft primary structures.4 One of the first, true synthetic composites, potentially capable of being used in structural applications, was ‘Gordon-Aerolite’. This was a composite consisting of unidirectionally aligned unbleached flax thread impregnated with phenolic resin.4 The development of this material began in 1936 with work undertaken by De Bruyne to utilise cotton fabric as reinforcement in phenolic mouldings.7 In 1936 De Bruyne in conjunction with The De Havilland Aircraft Company Ltd, with whom De Bruyne had a consultancy,7 were granted a patent entitled ‘Improvements relating to the manufacture of material and articles from resinous substances’.8 A number of prototype aircraft structural components were produced from Gordon-Aerolite. One of the first of these was a wing spar for the Bristol Blenheim.7 As well as this, there was the production of an experimental fuselage for the Supermarine Spitfire fighter.9 This development was instigated by a threatened shortage of bauxite for the production of duralumin. In the event this threat did not materialise and so this line of research was eventually discontinued.9 Other cellulose-based composites, employing paper impregnated with adhesives were, however, used successfully in a number of wartime applications. The most notable of these was probably a composite pilot seat for the Spitfire.4 It is interesting to note, however, that apart from this and fuel drop-tanks, no use was made of cellulose-based composites for aircraft primary structures.4 By the mid-1940s, the use of cellulosic fibre (either as fabric or in paper form) reinforced polymers was well established and much interest was being shown in the use of these materials for structural parts.10 However, with the advent of strong and stable synthetic fibres and liquid polymers such as unsaturated polyesters and epoxies, the use of cellulose fibre-reinforced composites in structural applications was superseded by wholly synthetic composites. The properties of these early cellulose-based composites, nevertheless, remain extremely impressive and are worthy of note. Gordon-Aerolite was produced by laying up skeins of resin impregnated unbleached flax yarn to form a cross-ply laminate structure, or a unidirectional bar or strip of material. Several skeins of material were placed upon one another until the required thickness was obtained. This preformed resin impregnated yarn was then hot pressed to form the consolidated laminate.9

Applications

237

The resultant laminate had a fibre volume fraction of around 75%. The ultimate tensile strength and Young’s modulus of a longitudinally loaded unidirectional skein of this material were around 480 MPa and 48 GPa, respectively, with a density of 1363 kg/m3.9 It was stated9 that for a cross-ply laminate, ‘The material had approximately equal strength and stiffness along and across the sheet, but the strength and stiffness at 45∞ to the grain was only one-half that along the fibres’. Furthermore, ‘The specific tensile strength at 0∞ and 90∞ was approximately the same as that of duralumin, while the specific stiffness (tensile at 0∞ and 90∞ and shear at 45∞) was about threequarters that of duralumin’. The compressive strength of Gordon-Aerolite was found to be 200 MPa parallel to the fibres and 95 MPa perpendicular to them. The shear strength parallel to the fibres was 38 MPa.10 Unfortunately, no figures were quoted for the toughness properties of this material. At that time, much of the reinforcement was of cotton fabric, either in sheet form or as diced, chopped or shredded material.5 In 1937, De Bruyne gave a lecture to the Royal Aeronautical Society entitled ‘Plastic materials for aircraft construction’. In this he presented details of the properties of a material referred to as ‘Cord-Aerolite’. This was a woven cotton fabric in which the number of ‘cords’ in the warp direction formed around 90% of the total. This fabric was embedded in a bakelite matrix. The cord reinforcement was noted to have ‘a remarkable effect on the properties in tension’. The tensile strength was reported to be 180 MPa and Young’s modulus around 13.8 GPa. Early uses for green composites were not restricted to aircraft structures; in the late 1930s Henry Ford promulgated the use of green composite materials in automotive applications. Fifty years or so later the automotive industry again took an interest in green composite materials and nowadays a significant number of car manufacturers incorporate components manufactured from green composites into their vehicles.

11.3

Contemporary applications of green composites

Over the past two decades or so, renewed interest has been shown in the use of green composite materials in applications ranging from aerospace to consumer goods. Many drivers may be seen to have contributed to this revival. Two principal drivers, nevertheless, appear to have come to the fore – environment and cost. Increasing concern over the influence that our activities are having upon the environment is stimulating the search for new and more ‘environmentally friendly’ materials and products. Natural fibre reinforced composite materials – green composites – offer the attractive prospect of our being able to grow our own materials to suit our needs. By combining natural fibres with resins produced from renewable resources, such as vegetable

238

Green composites

oils or other plant derivatives, the opportunity exists to produce wholly biobased materials. These materials should, in theory, have impeccable ‘green’ credentials. Not only would they be derived from plant biomass, which actively sequesters atmospheric carbon dioxide (CO2) during growth, but also when they reach the end of their useful lives, they may be disposed of through biodegradation or composting. Whilst returning the CO2 captured during growth to the atmosphere, the ability to biodegrade alleviates the problem of synthetic polymeric items which do not biodegrade – or at least not over a time scale which we would readily associate with biodegradation. The situation is complicated somewhat, in that to convert biomass into a form readily recognisable as a synthetic composite material, energy input is required. So, whilst the raw materials may effectively be CO2 neutral, the energy for conversion (most probably coming from fossil reserves) may be appreciable! Nevertheless, the combination of a natural fibre together with a bio-based resin or polymer is, potentially, very attractive from an environmental point of view. In terms of the environmental profile of green composite materials, an intermediate position is the use of waste, recycled or reclaimed materials. Combining waste fibre from recycled wood or newsprint, with post-industrial or post-consumer plastics like polyethylene or polypropylene to produce a ‘hybrid’ natural-synthetic composite offers many advantages too. This approach has led to the development of a burgeoning industry in North America. In the UK and Europe too, significant growth is to be expected in the near future.11 These so-called wood fibre–plastic composites help in environmental terms by removing from the waste stream materials that would otherwise be difficult to dispose of. The properties of these materials are often unique, combining the attributes of the polymer (e.g. non-biodegradable) with some of the desirable characteristics of the fibre to produce a material with reasonable mechanical properties and aesthetic appeal. For many applications, the unique combination of properties makes these materials well suited. Decking is one application area in which WPCs have found extensive use. Exterior use and consequent exposure to changing climatic conditions can result in microbial as well as insect attack to naturally non-durable wood species. To ensure an adequate lifespan, either more durable wood species or suitably treated nondurable species would be required. There are issues with both options, but particularly with the preservative treatment of wood, which is increasingly becoming the subject of restriction and legislation because of the hazardous substances, such as arsenic, which many preservative treatments contain. For these applications, WPCs are an attractive option. The second major driver is cost. The development of WPCs has, for instance, generally focused on low (or no) cost feedstock such as recycled plastics and waste fibre. WPCs have tended to replace sawn timber products, rather than compete with structural composite materials such as glass fibre

Applications

239

reinforced plastics (GRP), although there are now signs that WPC products are finding roles in structural or semi-structural applications.12 The raw materials themselves, particularly natural fibres, have tended to be viewed as a low cost option. Recycled wood fibre, waste agricultural fibres or byproducts from textile manufacture have all been considered as reinforcement in green composite materials. Now, however, there is evidence that natural fibres are starting to be grown specifically for industrial end uses, with a number of surveys having been conducted into the potential impact that industrial crops might have.13 Recently, the role that green composite materials could play in sustainable development has been highlighted. 14 Sustainable development, that is ‘development that meets the needs of the present without compromising the ability of future generations to meet their own needs’,15 is increasingly becoming a priority for businesses and governments alike. In a recent survey,16 over 50% of businesses in the UK that responded to the poll stated that sustainable development, whether formally or not, was part of their long-term business strategy. The importance of green composites and more generally materials and products derived from renewable resources, in sustainable development, was highlighted recently by the specific inclusion of renewable resources in the first call under the European Research Area Framework Programme (2002– 2006). The role of green composites in sustainable development and likely future trends in this area will be discussed in greater detail in the next section.

11.3.1

Material and process considerations

Before discussing in detail the applications for ‘green’ composites, it is necessary to highlight briefly some of the important characteristics of both reinforcing fibre and polymer matrix and the processing routes available, since these will ultimately have an impact upon the applications for these materials. Both fibres and matrices have already been dealt with in detail elsewhere in this volume. 11.3.1.1

Reinforcement

Generally speaking reinforcement can be classified as either wood or nonwood fibres. In the case of wood fibre, it is generally necessary to break down the solid wood, if this has not already been done, into a fibrous or ‘flour’ form. Individual wood fibres are generally of the order of a few millimetres in length and the only practicable way of obtaining these fibres is through refining or pulping. Thermomechanical pulping is used, for example, to produce fibre for newsprint and for medium density fibreboard (MDF). Whilst the quality of the fibre thus produced should be of good quality, it is both expensive and energy intensive to produce. Much of the ‘fibre’ produced

240

Green composites

for WPCs is termed ‘flour’ and is produced by breaking down wood chips mechanically into a fine, dust-like material, typically by hammermilling. In this form, the flour acts more in the capacity of a filler or extender, rather than as a true reinforcement. Wood fibre, on the other hand, could provide marginally more reinforcement, provided the fibre length can be maintained during the extrusion process.* Non-wood fibres are generally obtained from either industrial crops (e.g. flax, hemp) or as by-products of agricultural practice (e.g. wheat straw). The length of the useful ‘technical’ non-wood fibre can vary from more than a metre for flax, hemp and the other so-called ‘bast’ fibres down to little more than a few millimetres for wheat straw.17,18 The length of the fibre invariably dictates the processes of composite manufacture and ultimately the application. Long fibres, for example, can be manipulated by weaving to form useful reinforcement fabrics for use in thermosetting applications. 11.3.1.2

Matrices

At the present time, it is probably fair to state that the majority of matrices used in green composite applications are derived from petrochemical sources. These may be either recycled plastic resins, such a polypropylene or polyethylene, used in some WPCs, or thermosetting unsaturated polyesters and epoxies. As noted earlier, the existence of bio-based resins and plastics will make the wholly renewable composite material a reality. At the time of writing, the commercial supply of bio-based thermoplastics is limited to a few speciality biodegradable plastics, based on starches or polylactic acid (PLA). On the thermosetting side, the choice is even more restricted, with few bio-based resin systems being readily available commercially in any quantity. The systems that are available are generally based upon epoxidised vegetable oils such as linseed or soy. The processes used to form green composites are generally similar to those used to process ‘synthetic’ composites. Techniques such as extrusion and compression moulding are employed with the thermoplastic matrix materials, whilst thermosetting resins can be processed using techniques such as hand lay-up or resin transfer moulding (RTM). Generally, thermosetting resins are processed using long, non-wood fibres, such as flax or hemp, although compression moulding of long fibre-thermoplastic green composites is also commonplace. Short fibres are generally compounded with thermoplastic matrices. * Wood fibre and flour are used extensively with thermoplastics in WPCs, where the short fibre length makes it relatively easy to compound in an extruder. Long vegetable fibres such as flax or hemp are often employed as reinforcement in thermosetting matrices or in combination with thermoplastics as thermoformable media.

Applications

11.3.2

241

Applications

Broadly speaking, ‘green’ composites may find use in applications where timber, synthetic polymers or synthetic composites are currently used. Whilst this is a generalisation, it is useful broadly to place bio-composites in terms of technical performance requirements and hence cost. In North America, wood fibre–plastic composites frequently occupy the position of replacement timber products (or at least timber used in mainly non-structural applications). WPC manufacture is suited to high production rates and since the raw materials are generally inexpensive, the products manufactured can be sold competitively in applications where wood is currently used, although still carrying a premium. It has been estimated that, in 2001 in the USA, the total market for WPC materials was in excess of US$ 350 million.11 At the time of writing, however, the market in Europe has been described as ‘embryonic’.11 As we have seen, the main constituents of WPCs are a natural fibre ‘reinforcement’, most probably in the form of a short wood fibre or ‘flour’, which forms up to around 70% of the volume of the material, in combination with a thermoplastic polymer matrix. This matrix is generally a commodity thermoplastic such as polypropylene (PP), high-density polyethylene (HDPE) or polyvinyl chloride (PVC) and may be derived from either a recycled source, or may indeed be a virgin polymer. The reinforcement and matrix are combined, along with various additives such as adhesion promoters or process aids, in an extruder. The materials may be either formed into pellets to be used as feedstock for other processes, or formed directly into extruded profiles (Fig. 11.1). Decking is currently a major market for WPCs in North America but has, as yet, not taken off to any great extent in Europe. Several advantages over wood are claimed for WPCs. These include easy processing and treatment, good appearance, no splintering, improved resistance to biodegradation and insect (termite) attack as well as low maintenance. Other building products such as railings, fencing, window/door profiles, siding and shingles are, however, starting to be manufactured from this material. Good dimensional stability is a further attractive feature of WPCs, especially in applications where tight control over tolerances is a necessity, such as in window or door profiles. In this latter category, wood fibre–plastic composites provide direct competition, not only for wood products, where low maintenance, good rot resistance and dimensional stability offer technical improvements over wood products, but for polymers too. The replacement of polymers such as PVC with wood fibre–plastic composites offers certain advantages, in particular an improved environmental profile. Particularly in Europe, this could lead to expanding markets in this field of application, due to mounting pressure to reduce the use of PVC.

242

Green composites

11.1 Extruded profiles. (Source: Entek Extruders.)

With an established foothold and rapidly growing markets for WPCs, further technical developments will undoubtedly lead to improved performance and reliability. These, in turn, may well open up new application areas for WPCs in the buildings sector. One of the major potential drivers that may increase the use of WPCs in many diverse applications are the restrictions being placed upon preservative treated timber (European Commission Directives). Legislation is presently being introduced which will, in time, severely restrict the use of existing preservatives such as chromated copper arsenate (CCA) and creosote.19, 20 Environmental pressure on the use of naturally durable wood species (particularly tropical species) will, undoubtedly, open up new opportunities for WPCs in applications where exposure to rapid biodeterioration may take place. This pressure is redoubled if the area of application is itself environmentally sensitive, such as protected natural habitats. A particular current example of work being undertaken to develop WPCs specifically for ‘sensitive’ areas, is the US Navy’s ‘Engineered Wood Composites for Naval Waterfront Facilities’ research programme.21 This programme is seeking to develop suitable wood fibre-reinforced composites

Applications

243

to replace the preservative treated timbers that have traditionally been used in waterfront structures. This section has thus far described applications for extruded wood fibre– plastic composite products in the construction and building sectors. These products principally replace wood or timber products such as decking or window profiles. In addition to these, however, there are a number of further application areas for which these materials are potentially suitable. Of particular note is packaging. For example, with increased demands for hygiene, safety and longevity, pallets, which have traditionally been fashioned from lowgrade timber, are now being manufactured from WPCs. Crates and boxes are other examples of applications for WPCs in the packaging sector. The end uses discussed thus far have focussed on wood replacements. WPCs are, nevertheless, being used extensively to replace ‘traditional’ synthetic composite materials such as glass fibre reinforced plastic as well as engineering thermoplastics. These ‘green’ composites, whilst often generically referred to as wood fibre–plastic composites, more likely than not employ a long natural fibre such as flax, hemp or jute as reinforcement. These may be combined with either a thermoplastic polymer such as PP or HDPE or a thermosetting unsaturated polyester (UP), or epoxy resin (EP) as the matrix. At the present time in Europe, a large and still expanding market for ‘wood fibre–plastic composites’, or more accurately natural fibre-reinforced composites plastics, is in automotive applications. In 1999, it was estimated that more than 20 000 tonnes of natural fibre, mainly flax and hemp were used in automotive components.22 To date, most of the applications have been non-structural, but many of these green composite materials are now being introduced into more technically demanding roles.12 The long-term future of natural fibre-reinforced composites in these applications, however, looks questionable as legislation is driving towards greater and greater recyclability of automobiles, and natural fibre-reinforced composites, particularly those based upon thermosetting resin systems like polyurethanes, UP or EP resins, are difficult to recycle. Even those based on thermoplastics pose more difficulties than the pure polymer. Nevertheless, at the present time automotive components are an area of application in which green composites have found a niche and it is worthwhile examining this sector further. One of the first production cars to employ green composite materials extensively in its construction was the Eastern European Trabant. The body of this car was composed of panels of a natural fibre-reinforced plastic composite called Duroplast,23 which were screwed to the galvanised steel frame of the sub-structure. Surely this must be one of the first examples of mass production using green composites! The general trend to incorporate natural fibre-reinforced plastic components in automotive applications, however, started in earnest in the mid-1990s. A

244

Green composites

number of advantages were seen to using natural fibres in place of the traditionally used glass fibre. These included advantages in terms of weight saving (the density of natural fibres is about 60% that of glass fibre), reduced raw material cost, the ability to ‘thermally recycle’ the material and the marketing advantage of utilising an ‘environmentally friendly’ material. Other advantages include improved health and safety and reduced tool wear. It has been estimated that between 5 and 10 kg of fibre could be used in composite applications in each car produced.24 Although natural fibres have been used for ‘low performance’ applications in interior automotive situations, such as thermal and acoustic insulation, the introduction of natural fibre composites for door panels in the Mercedes– Benz E-Class provided a step towards higher performance applications. In this particular application, a flax/sisal mat was used as reinforcement in an epoxy matrix. A weight reduction of some 20% was claimed over the existing wood fibre material.25 At the present time, the number of commercial applications for green composite materials is relatively limited. Wood–plastic composites used either for construction applications or natural fibre-based synthetic polymers for automotive applications are the main success stories. Research and development, nevertheless, continues in this area and it is important to briefly touch upon the factors that are likely to influence future trends as well as to identify applications in which green composites could find function.

11.4

Future trends

11.4.1

Background factors

More and more, sustainable development is becoming a priority of businesses and governments alike. We frequently hear about the effects of global warming and the depletion of fossil reserves, and it would seem to be only a matter of time before we are forced to look at renewable resources to fulfil our need for materials in a sustainable fashion. In the near term, it seems likely that environmental concerns will continue to be a major driving factor affecting our use of materials and products. Pressure to reduce the amount of material going to landfill is stimulating initiatives such as the Waste and Resources Action Programme (WRAP)26 in the UK set up to create new and sustainable markets for waste materials. Coupled with this, there are other factors, such as the desire to promote the use of crops grown for industrial non-food purposes, which in the UK has led to the setting up of the National Non-Food Crops Centre (NNFCC)27 to promote the industrial use of crops. In addition to providing a renewable raw material for a range of industrial applications, the development of a sustainable industrial crops industry would help rejuvenate beleaguered rural communities, particularly if local processing were to be reintroduced.

Applications

245

Table 11.1 Selected physical and mechanical properties of some synthetic and natural (plant) fibres28, 29 Fibre type

Density (¥103 kg/m3)

Young’s modulus (GPa)

Tensile strength (MPa)

Failure strain (%)

Synthetic fibres E-glass High strength carbon Kevlar™ (aramid) Boron

2.56 1.75 1.45 2.6

76 230 130 400

2000 3400 3000 4000

2.6 3.4 2.3 1.0

Natural fibres Flax Hemp Jute Sisal Cotton

1.4–1.5 1.48 1.4 1.45 1.5

50–70 30–60 20–55 9–22 6–10

500–900 310–750 200–450 80–840 300–600

1.3–3.3 2–4 2–3 3–14 6–8

Against this backdrop of increasing awareness of sustainability issues and environmental pressures, there are exciting opportunities for ‘green’ materials. Nevertheless, there are certain technical, commercial and consumer barriers that will need to be addressed if these forms of material are to enter into the mainstream market. Understanding the nature and technical limitations of ‘green’ composites is an important consideration. It is of no benefit to attempt to enter markets where the technical requirements are too demanding of the material in question. For instance, whilst much has been written about the excellent tensile properties of natural fibre, particularly flax and hemp, much of this has been based upon the strength and stiffness of the fibre under ideal laboratory conditions. Realising this potential in an industrial process is a different matter. A comparison between the properties of a number of synthetic fibres and a range of different natural fibres is presented in Table 11.1. Whilst the figures quoted probably represent a ‘best case scenario’, it is interesting to note that not only is the density of natural fibre about 60% that of its main competitor, glass fibre, but that the stiffness, particularly that of flax, is approximately the same. What is also apparent is that the strength of the fibre is significantly less and also that there is wide variation in the mechanical properties. Nevertheless, the prospect of combining light weight with good stiffness is extremely attractive for many products and applications. The variation shown in the properties alludes to another potential barrier to using natural materials – quality and consistency of the resources and reliability of supply. Particularly within the non-food crops sector, the supply of industrial fibres is currently limited. As applications become more widespread the conditions for the development of a sustainable supply chain will certainly

246

Green composites

become more favourable. To a certain extent there will be similar problems in the short term with the use of recycled materials. Within the UK at least, the recycled polymer and recycled fibre (mainly wood fibre) infrastructure is underdeveloped; however, new composite materials and products manufactured from these materials would undoubtedly help develop the supply chain. 11.4.1.1

‘Greenness’

If ‘green’ composites are to be marketed on this basis, then it is vital that they can substantiate their environmental credentials. Life cycle assessment (LCA) is a tool that can be applied to assess the environmental impact of a particular product on a ‘cradle to grave’ basis. The results of LCAs can be revealing and it is by no means a given that if, for example, glass fibre reinforcement is replaced directly with a natural fibre alternative the product will be ‘greener’.30,31 It has been demonstrated that the greatest impact in environmental terms often arises from the polymer matrix, usually derived from petrochemical resources, rather than from the reinforcement fibre.32 It is partly for this reason that there is a significant amount of research interest being directed towards the development of bio-based thermosetting resins and of renewable resource-based biodegradable thermoplastics. Thermoplastics such as the Cargill Dow LLC ‘NatureWorks™ PLA’, a cornstarch-based polylactic acid thermoplastic or Novamont’s ‘Mater-Bi’, a starch-based thermoplastic are examples of renewable resource-based polymers currently in commercial production. A number of bio-based thermosetting resins are under development. These include materials based on various vegetable oils such as soy, linseed, cashew nut shell liquid and oilseed rape. One of the most notable of these is Cara Plastic’s thermosetting resin based on soy oil.33 The development of polymer resins and plastics from renewable resources offers the potential for producing true green composite materials, which could carry real environmental advantages over the current range of synthetic composites and it is likely that these will feature at the forefront of green composite technology in the future.

11.4.2

Future developments and applications

It seems likely that further technical developments with existing wood– plastic composites may well take place in parallel with market growth and offer up new opportunities for these materials. Presently, the main market for WPCs is in the construction sector (e.g. decking, window profiles, etc.) but with possible further developments in the type of fibre used, improvements in additives to promote adhesion between fibre and matrix and processing techniques, new opportunities and applications may well arise. Since many current WPC materials utilise durable recycled polymers such as polypropylene

Applications

247

and polyethylene it would be preferable to focus attention on applications where the good durability of the material can be exploited. A particular area, noted previously, where there would appear to be significant potential for growth is in the replacement of preservative treated wood in high biological hazard applications. In addition to this, improvements in the mechanical performance of existing WPCs, through the introduction of new fibre types, processing and additives may well result in an expansion in their use into more diverse and technically demanding application areas. Where product service lives are short and where durability is not a significant factor, but where end-of-life disposal is, WPCs based on biodegradable polymers such as PLA or starch have excellent potential. The introduction of directives such as the end-of-life vehicle (ELV) and Waste Electrical and Electronic Equipment Directive (WEEE), based on the ‘polluter pays’ ethos, may not only stimulate improvements in the ‘recycleability’ of products but may also create opportunities for biodegradable materials in these products. Here, WPCs based upon natural fibre and biodegradable polymers could play a significant role. Components such as the casings for computers and monitors, mobile phone covers could all, in theory, be produced from biodegradable composite materials. Green composite materials based upon thermosetting resins in combination with long natural fibres such as flax and hemp, offer potential in true structural applications. With few exceptions, however, there has been little in the way of commercialisation of such materials. Nevertheless, significant research efforts are being directed towards the development of fully bio-based composite materials suitable for structural uses, in applications ranging from leisure goods to construction components. Unlike biodegradable polymers, however, there are few thermosetting resins based upon renewable resources currently available commercially. This has tended to limit composites reinforced with natural fibre to those incorporating petrochemical-based resins such as unsaturated polyesters and epoxies. In time, it is to be expected that biobased thermosetting resin systems, competitive in terms of cost and performance may well become available. This would open up new and exciting possibilities for true structural ‘green’ composites.

11.5

Conclusions

Although the history of the application of green composites can be traced back to the mid-nineteenth century, it is only in the last decade or so that renewed interest has been shown in these materials. This interest has been spurred on by a number of factors, but potentially the most significant of these is the desire to lessen the effects of mankind’s activities upon the environment – to ‘green’ our materials. In this respect, there is significant potential for composites based upon renewable resources. Although at present

248

Green composites

the commercial applications for green composites are limited, principally to WPCs for some construction and automotive applications, ongoing research and development programmes into bio-polymers and natural fibre-reinforced composites is likely to lead to further advances and new opportunities in this sector. Underpinning this potential growth is significant political impetus to reduce or recover waste and to increase the use of renewable materials in place of fossil reserves.

Sources of further information and advice Over the years, there has been a significant volume of material published in the area of ‘green’ composites. Some of the technical information on automotive applications may well be proprietary and thus not in the public domain; however, there are a number of sources to which the interested reader can refer for further information. In the scientific literature, there are many publications relating to specific relevant research work. This covers the production, processing and technology of fibre, the development of bio-based resins and plastics, materials science and applications. Several reviews of this work have been undertaken and have been published. Rapra Technology Ltd’s review report, ‘Natural and wood fibre reinforcement in polymers’ is one such example.17 There are several conference series, which are devoted to green composite materials and much useful information is available in the proceedings of these conferences. The USDA Forest Service, Forest Products Laboratory’s ‘Woodfiber–plastics conference’34 has been running now for a number of years and a significant amount of market, as well as scientific information is available in the proceedings accompanying the conferences. In Europe too, there are now a number of ‘green’ composites-related conferences. Of particular note is the ‘EcoComp’ series.35 For market information, particularly on WPCs, a number of consulting firms such as Principia Partners LLC36 and Kline & Company37 in the US provide detailed market reports. Whilst such specialist market reports can be purchased directly, much detail is often to be found in the public domain, and recently WRAP published the results of a technology and market study on WPCs in the UK.11 A number of reports relating to the production of natural fibres for specific end uses have been published by, for example, the Nova Institute GmbH in Germany22 and Ministry for Agriculture Fisheries and Food (MAFF), now Defra, in the UK.38 In the UK and Europe, many of these reports are the result of various R&D projects sponsored either by national bodies or by the European Commission. A search of relevant databases such a Cordis39 can also yield detailed information on the applications for green composites. In addition to these, there are a number of specific interest organisations

Applications

249

and networks that have been set up to promote and disseminate information on green composite materials as well as, more generally, materials based on renewable resources. Amongst these are the aforementioned National Non-Food Crops Centre27 in the UK and the Sustainable Composites Network.40

References 1 Gordon, J.E. (1976). The New Science of Strong Materials. Penguin Books, London. 2 British Plastics Federation (Viewed at: http://www.bpf.co.uk Accessed on 27 January 2004). 3 Plastics Historical Society (Viewed at: http://www.plastiquarian.com/ Accessed on 27 January 2004). 4 McMullen, P. (1984). Fibre/resin composites for aircraft primary structures: a short history, 1936–1984. Composites, 15(3), 222–30. 5 Brown, W.J. (1947). Fabric Reinforced Plastics. Cleaver-Hume Press, London. 6 De Bruyne, N.A. (1937). Plastic Materials for Aircraft Construction. The 615th Lecture read before the Royal Aeronautical Society Lecture, 28 January, 1937, Royal Society of Arts, John Street, Adelphi, London. 7 Bishopp, J.A. (1997). The history of redux® and the redux bonding process. International Journal of Adhesion and Adhesives, 17, 287–301. 8 Patent Office (1936). Patent specification: ‘Improvements relating to the manufacture of material and articles from resinous substances’. (No. 3040/36. 470,331). 9 Aero Research Limited (1945). A fighter fuselage in synthetic material. Aero Research Technical Notes. 10 Livingstone Smith, S. (1945). A survey of plastics from the viewpoint of the mechanical engineer. Institute of Mechanical Engineering, 29–43. 11 Optimat Ltd and MERL Ltd. (2003). Wood plastic composites study: technologies and UK market opportunities, The Waste Action and Resources Action Programme, The Old Academy, 21 Horsefair, Banbury, Oxon., UK. (Viewed at: http:// www.wrap.org.uk Accessed on 26 January 2004). 12 Nickel, J. and Riedel, U. (2003). Activities in biocomposites. Materials Today, 6(4), 44–8. 13 ACTIN (2001). Realising the economic potential of UK-grown industrial crops, ACTIN, Pira House, Randalls Road, LEATHERHEAD Surrey UK (http://www.nnfcc.co.uk/). 14 Tucker, N. and Hughes, M. (2002). SusCompNet – the sustainable composites network. In Proceedings of the 23rd Risø International Symposium on Materials Science: Sustainable Natural and Polymeric Composites – Science and Technology, eds. H. Lilholt, B. Madsen, H.L. Toftegaard, E Cendre, M. Megnis, L.P. Mikkelsen and B.F. Sørensen, Risø National Laboratory, Roskilde, Denmark. 15 WCED (1987). The Brundtland Report, 43. 16 CARM (2002). Renewable Feedstock for Sustainable Materials – BIOPRODUCTS – Their Importance to Wales: A Scoping Study. Centre for Advanced and Renewable Materials (http://www.carmtechnology.com/) 17 Bledzki, A.K., Sperber, V.E. and Faruk, O. (2002). Natural and wood fibre reinforcement in polymers. Rapra Review Report No. 152, 13, No. 8. RAPRA Technology Ltd, Shawbury, Shrewsbury, Shropshire SY4 4NR, UK.

250

Green composites

18 Weindling, L. (1947). Long Vegetable Fibers: Manila, Sisal, Jute, Flax and Related Fibers of Commerce. Columbia University Press, New York. 19 Commission Directive 2003/02/EC of 6 January 2003, relating to restrictions on the marketing and use of arsenic (tenth adaptation to technical progress to Council Directive 76/769/EEC). 20 Commission Directive 2001/90/EC of 26 October 2001, adapting to technical progress for the seventh time Annex I to Commission Directive 76/769/EEC on the approximation of the laws, regulations and administrative provisions of the Member States relating to restrictions on the marketing and use of certain dangerous substances and preparations (creosote). 21 Washington State University Engineered Wood Composites for Naval Waterfront Facilities (Viewed at: http://composites.wsu.edu/navy/ Accessed on 27 January 2004). 22 Karus, M., Kaup, M. and Lohmeyer, D. (2000). Study on markets and prices for natural fibres (Germany and EU). Nova Institute GmbH (Viewed at: http://www.novainstitut.de Accessed on 26 January 2004). 23 Netcomposites: Newsroom: ‘Eastern European icon set for Africa’ (Viewed at: http:/ /www.netcomposites.com/news.asp?1680 Accessed on 26 January 2004). 24 Brouwer, W.D. (2000). Natural fibre composites in structural components: alternative applications for sisal? In Alternative Applications for Sisal and Henequen’. A seminar, organised jointly by the Food and Agriculture Organization of the United Nations (FAO) and the Common Fund for Commodities (CFC), held in Rome on Wednesday 13, December 2000. (Viewed at: http://www.fao.org/DOCREP/004/Y1873E/ y1873e0a.htm#bm10. Accessed on 26 January 2004). 25 Schuh, T.G. (1999). Renewable materials for automotive applications. Natural Fibres Performance Forum, Copenhagen 27–28 May 1999. 26 Waste and Resources Action Programme – WRAP (www.wrap.org.uk). 27 National Non-Food Crops Centre – NNFCC (http://www.nnfcc.co.uk/). 28 Hull, D. and Clyne, T.W. (1996). An Introduction to Composite Materials. Cambridge University Press, Cambridge, UK 29 Ivens, J., Bos, H. and Verpoest, I. (1997). The applicability of natural fibres as reinforcement for polymer composites. In: Renewable Byproducts: Industrial Outlets and Research for the 21st Century. June 24–25, 1997, EC-symposium at the International Agricultural Center (IAC), Wageningen, The Netherlands. 30 Black, A., Anderson, J. and Steele, K. (2003). A simplified guide to assessing environmental, social and economic performance for the composite industry. In Proceedings of ‘EcoComp 2003’, 1 & 2 September 2003, Queen Mary, University of London, UK. 31 Peijs, T. (2002). Composites turn green! e-Polymers 2002, no.T_002. (Viewed at: http://www.e-polymers.org/papers/peijs_110202.pdf Accessed on 27 January 2004). 32 Positioning the UK as a world leader in innovative plant fibre processing. Technology Foresight Challenge Programme. Project final report. The BioComposites Centre, University of Wales, Bangor. 33 Cara Plastics Inc. (viewed at: http://www.caraplastics.com/default.htm Accessed on 26 January 2004). 34 7th International Conference on Woodfiber–Plastic Composites (and other natural fibers), Monona Terrace Community and Convention Center, Madison, Wisconsin, USA, May 19–20, 2003 (Viewed at: http://www.forestprod.org/wpc03powerpoints.html Accessed on 26th January 2004). 35 EcoComp (2003). 2nd International Conference on Eco-Composites, 1–2 September

Applications

36 37 38

39 40

251

2003, Queen Mary, University of London, London, UK (Viewed at: http:// www.materials.qmul.ac.uk/ecocomp Accessed on 26 January 2004). Principia Partners LLC (Viewed at: http://www.principiaconsulting.com Accessed on 26 January 2004). Kline and Company (Viewed at: http://www.klinegroup.com Accessed on 26 January 2004). Ellison, G.C. and McNaught, R. (2000). The use of natural fibres in non-woven structures for applications in automotive component substrates. MAFF Project Reference No. NF0309. European Commission, Community Research and Development Information Service ‘CORDIS’ (Viewed at: http://www.cordis.lu/ Accessed on 26 January 2004). The Sustainable Composites Network (Viewed at: http://www.bc.bangor.ac.uk/suscomp/ Accessed on 27 January 2004).

12 Re-use, recycling and degradation of composites A. HODZIC James Cook University, Australia The atmosphere doesn’t recognize company expansion or company balance sheets, it only recognizes CO2 molecules. Roger Higman, Chemistry and Industry

12.1

Introduction

In 1906 Dr Baekeland discovered that phenol and formaldehyde react to make a resin which can be hardened by heat to form an infusible, insoluble solid.1 Today, the annual world production of plastic materials including composites reaches around 130 million tonnes, and the predictions expect around 5% increase in production of these materials every year. Although 80% of this astounding figure are thermoplastic materials suitable for recycling,2 only 1–2% of all plastic materials reach one of recycling processes described in this chapter. The remaining plastic waste is either incinerated to obtain heating energy or deposited onto landfills. The costs of disposing of plastics are high and the impact on the environment has taken many forms of damage: greenhouse emissions and toxic gases as the result of incineration, damage to wildlife and marine species, permanent landfill accumulation. The options for plastic waste reduction are: (i) to extract monomers from used plastic material and re-create virgin materials suitable for further production (ii) to develop cost-effective recycling processes that will justify re-use of plastic products (iii) to develop biodegradable, naturally sourced polymers to replace the conventional family of synthetic polymers and reduce impact on the environment (iv) to introduce biochemical processes which will eventually degrade synthetic polymers down to environmentally degradable short polymer chains. Extraction of polymers is defined as a process in which a polymeric material, consisting of macromolecules differing in some characteristic affecting their solubility, is separated from a polymer-rich phase into fractions by successively increasing the solution power of the solvent, resulting in the repeated formation of a two-phase system in which the more soluble components concentrate in 252

Re-use, recycling and degradation of composites

253

the polymer-poor phase. The process becomes increasingly complex when different constituents of composite materials are involved. Extraction of plastic waste to obtain a pure monomer phase that can be used for the best quality recyclate is economically justified for polymers and polymer blends. In the case of advanced composite materials, with an increasing number of additives and reinforcement materials, the only solution to reducing waste is to crush the compound and to recycle or downcycle. To recycle means to recover a product at the end of its useful life, break it down into its constituent components and re-incorporate it into new product that has an inherent value equal to the original product. Downcycling implies the identical process, with the resulting material having an inherent value less than the original product. Polymer composites possess a number of advantageous properties over metallic materials: light weight with comparable strength and cost, better durability and resistance to corrosion, easier manufacturing processes and more convenient acoustic properties. The research conducted in the area of composites has been mainly focused on the manufacturing and design of durable, high-strength, high temperature-resistant composites to replace metallic components utilised in the industry. Thus, polymer-based materials are known for their permanence and durability, which can be viewed from different angles. A short-term perspective emphasises that durable composites, in addition to their excellent mechanical properties, are suitable for re-use and are not directly hazardous to wildlife and marine species due to their lack of degradability. Another perspective, which takes into consideration a complete lifecycle of these materials, recognises the problem of disposal and transformation of durable plastics into other forms of energy. Plastics contribute significantly to the heating value of municipal solid waste, with a heating value of three times that of typical municipal waste.3 Plastic additives containing heavy metals such as lead and cadmium contribute to the metal content and possibly to the toxicity of incinerator ash. It is difficult to consider source reduction of plastic waste or any single component of the waste stream in isolation because the goals of source reduction are to reduce the amount of toxicity of the whole waste stream, not just of a single component. Thus, it has been suggested that, among other ways of sourcereduction in waste stream, materials should me made more durable so that they may be re-used and a lifecycle evaluation should be included in polymer composite design. In other words, any attempt to produce synthetic material will eventually lead to either immediate incineration of disposed plastic product, or the material will be subjected to a number of recycling processes, and then it will eventually reach incinerator. Recycling of plastics, although extremely useful in terms of cost and raw plastic reduction, sometimes offers a perfect excuse to overlook the inevitable negative result of synthetic polymers production.

254

12.2

Green composites

Recycling of polymers and composites

Recycling of plastics consists of four phases of activity: collection, separation, processing/manufacturing and marketing.4 Effective separation of mixed plastic waste is necessary because only clean, homogeneous resins can produce the highest quality recycled plastic products.3 Unlike post-consumer waste plastic such as polythene, which can be melted and is relatively easily reprocessed, fibre-reinforced polymers contain a considerable fraction of glass fibre and filler such as calcium carbonate and talc. Polymer composite products are often highly engineered. On the other hand, the value of the material constituents and hence any waste or recovered demolition material is low. Composite materials have high strength and stiffness, which is a disadvantage during reprocessing, since it implies that heavy machinery is required for shredding and grinding. Polymer composite products are also often bulky and lightweight engineered sections or profiles, which cause transport of non-ground waste to be uneconomic. The most common disposal method is landfill. There are many waste legislations that could influence composite industry: in Europe, there are EU waste management directives on landfill, incineration, construction and demolition waste, end-of-life vehicles, electrical and electronic equipment; the UK has government policy such as the Waste Strategy 2000, the sustainable construction strategy, the landfill tax, and local government policy. These waste legislations focus on dealing with waste through the waste hierarchy and will therefore put more pressure on solving fibre-reinforced polymer composites waste management through recycling and re-use. The end-of-life vehicle directive is the most significant policy change relating to the use of fibre-reinforced polymers, and has some effect on the composite industry concerned with vehicle component manufacture. Although it has no immediate effect on the construction sector, it could influence attitudes towards composites in purchasing policies and in future legislation. Recycling of composites opens two important technical issues. The first technical challenge is the development of a grinding process that will separate the fibres from the matrix polymer. The second challenge is incorporating the recycled fibres into a resin system without adversely affecting the properties of the resin. The viscosity of the resin is increased with the addition of the recycled fibres and common processing techniques become unsuitable at this stage. The fibres are completely surrounded by the polymer resin in the composite, which renders the mechanical separation process economically unjustifiable. Henshaw et al.5 noted that the composite community favours secondary recycling, i.e. granulation followed by injection or compression moulding, over tertiary or primary recycling options such as extraction, because the secondary processes use existing technology, and extensive markets already

Re-use, recycling and degradation of composites

255

exist, whereas extraction and manufacture into new high performance composites are rarely economical.

12.3

Recycling of thermoplastic composites

Thermoplastic resins have viscosities ranging from 500–5000 Pas at temperatures within their processing range, whereas uncured thermosetting resins have a viscosity of about 100 Pas. Other major difficulties with thermoplastic resins are poor surface adhesion and fibre wetting. On the other hand, thermoplastic materials offer more recycling options including remoulding and regrinding. The main issues being addressed in thermoplastic recyclate are thermally induced chemical degradation of polymers and the physical attrition of glass fibres that occur during repeated cycles of extrusion and injection moulding.6 In recycling thermoplastics a mixed waste stream unavoidably occurs due to difficulties in collection and sorting, associated with the visual similarity of commonly used polymers as well as their similar physical properties. Many polymers are not compatible and yield low properties when blended and moulded directly. Chemical compatibilisers are being developed to overcome this problem. It has been reported that the addition of glass fibres acts to compatibilise ‘mechanically’ normally incompatible phases.7 Papaspyrides et al.8 employed a solvent-based technique to recycle thermoplastic glass fibre composites. A polymer solvent was used to dissolve the polymer enabling removal of glass fibres for re-use. Toluene was used as a solvent for the LDPE matrix. An interesting result is that glass fibres from the recycled route formed stronger second generation composites because residual polymer on the fibre surfaces aided in matrix–fibre bonding in the recycled composite. Information about the economic feasibility of the solventbased aspect of the treatment was not reported. Henshaw et al. reported studies on the recycling of polycarbonate matrix with short glass fibre composites.6 Recycled composites of this material manufactured by the injection moulding method showed excellent properties, whilst those that were compression moulded showed poor tensile but acceptable impact properties. The compression moulded material exhibited microstructural inhomogeneity and micro-cracks caused by the processing technique. Processing methods of recycled composites were reported to have a large effect on the final product quality. Another novel method: low temperature, high energy ball milling of mixed polymers was used to co-pulverise polymer powder particles or shredded polymer films in order to promote a substantial size reduction and to create mechanical and chemical effects on the milled material.9–11 The energy transfer promoted by the milling process causes impact, compressive and shear forces on the different polymer powder particles, and the combination of such

256

Green composites

effects induces chain scission with free radical formation.12 Reaction of free radicals from different chain species of intrinsically incompatible polymers can easily couple and produce a suitable blend.13 This method has been further improved, resulting in a new, near room temperature, high energy ball milling technique.14,15 The mechanochemical effects of ball-milling are enhanced through the insertion of a quantity of liquid CO2 in the milling vial. In this condition the energy transfer from the milling device to the milled material promotes repeated microexplosive evaporation of liquid CO2, which is trapped between the ball and the vial wall. In this process, the mechanochemical compatibilisation of the polymer blend occurs in a short milling time of 10–60 min. Composites were made from polymer blend thus obtained and short polyester fibres of volume fraction 30%, and melted between 110 and 140 ∞C. The best mechanical properties were reported for the composite consolidated at 130 ∞C, with the tensile yield strength of the compatibilised material 70% higher than the reference material. Properties can be improved further by changing the fibre length and type. It is suggested by the authors that the material should be utilised for non-critical structural applications. A number of articles have appeared recently, which report radiation treatment of recovered polymer scrap. Gamma and e-beam irradiation have been applied to recovered LDPE samples in order to improve properties of recycled material.16,17 Mechanical properties were improved with e-beam treatment, but deteriorated with g-irradiation. Czvikovszky et al. have reported a number of studies where recycled reinforced polymer systems were prepared using polypropylene from reprocessed car bumpers.18–20 A variety of fibres were used, including wood, glass, viscose and waste cord-yarns from the tyre industry. A small amount of reactive additives was used to enhance the adhesion between the constituents. A significant improvement in mechanical properties was observed upon low e-beam irradiation for the recycled composite with wood and viscose fibres. Ionising radiation offers unique possibilities for application in recycling of polymers, due to its dual ability to initiate two opposite physical processes in polymers, cross-linking and scission. Adhesion between particulates of different material types, or with reinforcing additives, can be promoted by coating the particles with a radiation-activated cross-linking monomer prior to remoulding of the recovered material.21 A broad range of possibilities exist for using various combinations of recovered scrap of one or more materials, reactive additive and inexpensive fillers in order to create structural engineering materials.

Re-use, recycling and degradation of composites

12.4

257

Recycling of thermosetting composites

The majority of studies on the recycling and automotive thermosetting composites were started off by German legislation in 1995 which established that 80% of composites should be recycled.22 Development of uses for scrap advanced composite laminates and pre-preg materials is important because of the potential for recovering the reinforcing fibres, such as Kevlar, graphite or glass fibres, from the expensive composite materials and because of the almost infinite life of the scrap material when disposed of in landfills. The use of ground composite material as an additive in the manufacturing of other materials can have two purposes: filler and reinforcement. The cured composite material is relatively inert and finely ground material could possibly be added as a substitute for common fillers such as calcium carbonate or silica. These filler materials are low cost and the cost to process composite material adds to the cost. The four main processes in recycling of thermosetting composites are: (i) grinding, (ii) chemical degradation and fibres recovery, (iii) pyrolysis and (iv) incineration. Recycling of thermosetting composites by grinding enables re-use of glass fibres, calcium carbonate and polymeric matrix without separation of the components. The composite is shredded, granulated to small fibre and used as a filler in a new process of manufacture. These products have the same, or even better, mechanical properties as the initial first-generation composite material.23,24 In addition, grinded recyclate has a lower specific gravity and contributes to reduction in weight when compared to conventional fibres.

12.4.1

Grinding

The grinding process consist of the following stages: the composite material is shredded to a convenient size using shredders designed for high torque and low speed. Hammer mills are used to reduce the size of recyclate further. The major challenge is the damage posed to cutting tools and blades, which are subjected to a high degree of abrasive wear. A typical high quality recyclate produced by the ERCOM composite recycling process is shown in Fig. 12.1. This process produces a controlled length of glass fibre-rich composite recyclate. ERCOM Composites Recycling GmbH based in Germany is a consortium of companies which began recycling of composites in 1992 and include BASF, BWR, Cray Valley, Dow Chemical, DSM, Duroform, Elastogran, Menzolit, Mitras, OCF and Vetrotex. A major feature of this recycling process is the use of a mobile shredder, which reduces the size of the bulk composite parts down to chips and removes metallic inserts for better purity of the recyclate. More details about this successful commercial process are available elsewhere.25

258

Green composites

12.1 Example of fibre-rich sheet moulding compound recyclate produced by the ERCOM composite recycling process.25

It has been reported26 that the performance of ground composite materials, when added as a reinforcement, improved the strength in epoxy-moulding compounds. In the first phase, it was determined that to process the waste composite material into fibres required a two-step operation consisting of pre-shredding and then processing the shredded material through a hammer mill. The production of the shorter fibres required multiple passes through the hammer mill, which reduced the throughput. The cost to process this material was estimated to be less than $2 per pound. In the second phase of the research, it was determined that a maximum of 1% addition of fibres could be made to the epoxy-resin before the viscosity increased to unacceptable levels. The length of the fibre also affected the viscosity, with longer fibres increasing the viscosity of the resin. In phase three the results showed the strength of the epoxy resin was increased by 16% with the addition of 1% fibres with fibre lengths less than 0.5 mm. The fibres with length less than 0.5 mm produced not only higher strengths when compared to fibres greater than 0.5 mm, but also produced a lower viscosity resin with the same percentage addition. Even with relatively small percentage additions, significant strength improvements were achieved at a relatively low cost compared with virgin fibres.

Re-use, recycling and degradation of composites

259

Scientists at Brunel University26 focused on strategies and technology for the re-use of suitably ground uncontaminated thermosetting recyclate as a functional filler for polymers leading to products with added value. Important factors include reducing the size of the thermoset scrap to a form suitable for incorporation into the host polymer matrix, the economics of the grinding procedure and identification of filler characteristics, which may add value to the host matrix by influencing its physical or chemical properties. An integrated process technology was developed to combine the necessary functional steps within a unified continuous conversion procedure. Moisture and volatiles were removed during the grinding stage aided by vacuum extraction. The recyclate was combined with the polymer matrix ensuring that effective dispersion and wetting occurred. A range of polymer compositions was created containing fibre-reinforced polyester and phenolic recyclate derived from industrial scrap. Thermosetting fillers have been successfully incorporated into polypropylene and it was possible to enhance mechanical properties relative to unfilled polymer, especially in the case of phenolic material that had a higher glass content and greater fibre integrity than the polyester waste used. Reinforced phenolic recyclate has been successfully incorporated into polyester resin and the presence of this recyclate increased fire performance (reduction of smoke emission) relative to unfilled polyester resin.

12.4.2

Chemical degradation and fibre recovery

Chemical degradation of composites involves partial or selective degradation of polyester/styrene polymer network in the presence of water, ethanol, potassium hydroxide and various amides.27 This process is rather inferior when compared with the quality of grinding recyclate, as potassium hydroxide causes detrimental effects on the recovered glass fibres. Also, a neutralisation step is required, which generates large quantities of waste water and adds to the cost. Instead of aggressive substances, ethanolamines have been evaluated for fibre recovery and have resulted in improved quality of recycled glass fibres. The disadvantage of this process is that it yields fibres whose volume per weight is greater than that of the virgin constituents. Scientists at the University of Nottingham28 have developed fluidisedbed thermal processing techniques to recover energy and fibres in a form suitable for recycling into high value products. These techniques are suitable for contaminated and mixed scrap material from end-of-life applications, especially in the automotive industry. Initial investigations were conducted using a typical industrial sheet moulding compound based on polyester resin. The optimum process temperature was 450 ∞C, as glass fibres suffer a reduction in strength during processing at higher temperatures. At 450 ∞C the recovered fibres (in the form of short individual filaments) had the same stiffness, but however, only 50% of the strength of virgin glass fibres. The purity of fibres

260

Green composites

was 80% with little surface contamination. The fluidised bed process effectively cleaned the fibres of the polymer matrix. In addition, a painted car boot lid made from a double skin glass-reinforced plastic, based on a polyester resin with a polyurethane foam core and metal inserts was processed in the fluidised bed at 450 ∞C. This mixed and contaminated feed was processed in the same way as the other glass-reinforced polymer materials. The recovered glass fibres were of the same quality as the fibres recovered from the pure composites. The fluidised bed process has also proved useful in processing scrap carbon fibre composites and could yield good quality carbon fibres that are potentially of high value. The problem in recovery of glass and carbon fibres lies in the economy of the process. An economic analysis of the fluidised bed recycling process showed that an operation recycling in excess of 10 000 tonnes per year of glass fibre composite material is needed to ensure the commercial viability of the process and these quantities do not exist in UK at present, or in any other single country within the EU. A carbon fibre recycling plant has the potential to be viable at much lower annual throughputs and may show a more favourable prospect of being viable in the short term. Pyrolysis is a technologically simple and well-controlled process that recovers a good part of glass fibres and calcium carbonate, which can then be re-used as filler and reinforcement. The process separates organics from inorganics, due to a significant difference between the temperatures of their thermal decompositions. High-temperature pyrolysis (~750 ∞C) leads to a considerable reduction in the strength of glass fibres, which prevents their re-use as a high quality reinforcement. Low-temperature pyrolysis (below 200 ∞C) is applicable to thermoplastic composites and yields excellent recovery of glass fibres and inorganic fillers.22

12.5

Degradation of polymers: UV light and biodegradation

12.5.1

Degradation by UV light

Photolysis with UV light of polymers generates radicals and ions that can initiate cleavage and cross-linking. Oxidation also occurs, complicating the situation, since exposure to light is seldom in the absence of oxygen.29 Generally this changes the material’s susceptibility to biodegradation. It is expected that the observed rate of degradation should increase until most of the fragmented polymer is consumed and a slower rate of degradation should follow for the cross-linked portion of the polymer. A study of the effects of UV irradiation on hydrolysable polymers confirmed this expectation.30 Photooxidation of polyalkenes promotes the biodegradation where the formation of carbonyl and ester groups is responsible for this change.31,32 Processes

Re-use, recycling and degradation of composites

261

have been developed to prepare co-polymers of alkenes containing carbonyl groups so they will be more susceptible to photolytic cleavage prior to degradation. The problem with this approach is that negligible degradation was observed over a 2-year period for the buried specimens. Unless a prephotolysis arrangement can be made, the problem of plastic waste disposal remains serious, as it is undesirable to have open disposal, even with constant sunlight exposure.

12.5.2

Biodegradation of synthetic polymers

Biodegradation is a natural process by which organic chemicals in the environment are converted to simpler compounds, mineralised and redistributed through elemental cycles such as the carbon, nitrogen and sulphur cycles. Biodegradation can only occur within the biosphere as microorganisms play a central role in the biodegradation process. There are four biodegradation environments for polymers and plastic products: soil, aquatic, landfill and compost. Each environment contains different microorganisms and has different conditions for degradation.33 In soil, fungi are mostly responsible for the degradation of organic matter including polymers. The aquatic environment is dominated by two types of bacteria, on the surface and in the sediment, with bacterial concentration in water decreasing with increasing depth. Microorganisms biodegrade organic materials by the use of their enzymatic apparatus; however, microorganisms have not yet had time to adapt and to synthesise polymer-specific enzymes capable of degrading and consuming synthetic polymers of recent origin. Enzyme activity is inhibited by the hydrophobic character of the plastics and high molecular weight. The amount of non-degradable plastic waste can be greatly reduced by proper development of biodegradable polymers and composites for short-term products. Biodegradation can occur by two different mechanisms: namely, hydrobiodegradation (hydrolysis followed by oxidation) and oxo-biodegradation. The former is much more important in the case of hydrophilic natural polymers such as cellulose, starch and polyesters, whereas the latter predominates in the case of other natural polymers such as rubber and lignin. Synthetic polymers do not hydrolyse under normal environmental conditions but it was shown in the early studies discussed previously that, after transition metal-catalysed thermal peroxidation, they biodegrade readily in the presence of a variety of thermophilic microorganisms in the surface layers of the polymer (Fig. 12.2). An SEM image confirms that microorganisms congregate on the polymer surface and after their removal, the surface becomes physically pitted and eroded. The surface of the polymer after biological attack is physically weak and readily disintegrates under mild pressure.34 For a synthetic polymer to be degradable by enzyme catalysis, the polymer

262

Green composites

¥100

¥500

¥1500

¥ 5000

12.2 SEM micrograph of compacted films oven aged for 300 h at 60 ∞C and incubated for two months with R. rhodochrous. From S. Bonhommea et al., in press.34

chain must be flexible enough to fit into the active site of the enzyme. This most probably accounts for the fact that the more rigid aromatic poly(ethylene terephthalate) is generally considered to be bio-inert, whereas the flexible aliphatic polyesters are readily degraded by biological systems. Polymers with main chains containing mostly covalent bonds show little or no susceptibility to enzyme-catalysed degradation reactions, especially those with higher molecular weights. Several approaches have been used to insert ‘weak links’ within, or immediately attached to, the backbones of such polymers. These ‘weak links’ are designed to permit the controlled degradation of an initially high molecular weight, hydrophobic polymer into a lower molecular weight oligomer, which can then be utilised and consumed by microorganisms through biodegradation processes. Particular emphasis in this approach to create useful biodegradable polymers has been placed on two types of polymer modifications: insertion of functional groups in the main chain, especially ester groups, which can be cleaved by chemical hydrolysis, and insertion of functional groups in, or on, the main chain that can undergo photochemical chain-cleavage reactions, typically of carbonyl groups. Environmental Products, Inc. (EPI) has recently developed a proprietary technology, Totally Degradable Plastic Additive (TDPA™), which is able to

Re-use, recycling and degradation of composites

263

degrade common plastics in a controlled, ecologically responsible manner. The technology has been recently tested with common plastics and reported in the literature.34,35 TDPA™ was used with polyethylene and low density polyethylene on the understanding that thermal and/or photolytic pre-abiotic treatment constitutes a major route for promoting the eventual biodegradation. The biodegradation was measured by monitoring the initial variation of molecular weight and other structural parameters (tensile strength, degree of crystallinity, spectroscopic characteristics). Biodegradation was observed when the degraded polymers are exposed to biotic environment.36 LDPE–TDPA™ oxidised samples endured significant biodegradation (50– 60% by carbon dioxide evolution) over a period of 18 months as mediated by soil microorganisms in closed respirometric vessels (biometer flasks). At this stage of ongoing research activity, it is clear that LDPE–TDPA™ formulations are effective in promoting the oxidation and subsequent biodegradation of polyethylene in soil environments. Still, several issues remain to be investigated, including control of the rate and completeness of biodegradation and the cumulative time for oxidation and biodegradation under different environmental conditions.

12.6

Recycling of composites in the automotive industry

The automotive industry has taken a strong initiative to include more recyclable parts in automotive structures and to use more recycled materials in automobile manufacture. With the increasing use of composites and plastics in automobiles (see Fig. 12.3), even separating functional units such as bumpers and dashboards will result in the mixture of polymer materials.36 Recycling of advanced composite materials, including those used for bumpers and dashboards, poses a serious problem in the industry. These materials are famous for being lightweight and durable, and therefore difficult to recycle. A new process of recycling developed by the University of Leeds will be able to transform composites to their original recyclable constituents, oil and fibre. The process is not only useful for the automotive industry, but also opens up possibilities for new applications for advanced composites by making recycling possible. The new recycling process combines pyrolysis and physical separation. Composites are crushed into gas, oil, a small amount of carbon and fibres. Recycled oil and fibre are reprocessed into composite plastics. Pyrolysis involves heating the plastic in the absence of oxygen, at temperatures below burning, and the fibres retain much of their original strength instead of becoming brittle. The research showed that the oil and the fibres can be processed back into composite plastic or can be re-used in other ways. Recycled nylon has been used in air-cleaner housing and in fan assemblies.37

264

Green composites Automobiles Aeronautics Civil engineering Sports and recreation Electricity and electronics Industrial engineering Shipbuilding Medical equipment Railways Miscellaneous 0

10

20

30

40

Composite user industries (in value, in %)

12.3 Market share of composite materials utilised in major industries. Source: Study on composite materials – nodal consultants estimates – 2000

Recycled ABS and polyester/polycarbonate alloys are being used for brackets to hold radio antennae, splash shields and small under-the hood-parts.38 Recycled polycarbonate/polybutylene terephthalate bumpers are being used in new bumpers. Thermoplastic olefin bumpers are being recycled into bumper fascias, splash shields, air dams and claddings.39 Recycled polypropylene is used for power-train applications, fender liners, air conditioning evaporative housings, vents and in other miscellaneous applications.38–40 Recycled PET is used in headliners and in engine covers.40,41 Among all industries that utilise composite materials, the automotive has the strongest impetus for development of recyclable and degradable materials. Owing to their excellent economical resources and strong legislative initiatives, automotive companies work closely with research institutes in Europe and USA on development of recycling processes for all composite and plastic parts currently used in automotive production. The results of ongoing projects have not been reported in the scientific literature at this stage; however, major developments are expected in the next 3–5 years.

12.7

Utilising green composites and incinerating polymers

12.7.1

Green composites

Green composites constitute natural fibres embedded in a matrix made of a plant-based or other resin.42 Natural macromolecules such as protein, cellulose

Re-use, recycling and degradation of composites

265

and starch are generally degraded in biological systems by hydrolysis followed by oxidation. The blending of biodegradable polymers, such as starch, with inert polymers, such as polyethylene, with natural fibre reinforcement, has received a considerable amount of attention for possible applications in the waste disposal of plastics.43–48 There is a large range of natural fibres that has been successfully used in many composites in recent years. These include jute, hemp, kenaf, ramie, sisal, flax and sugar cane bagasse fibre. These natural fibres are of low density, have high strength and stiffness relative to their density, low cost, good thermal and acoustic insulating properties and are friendly to processing equipment.49 Also they are fully recyclable and combustible without any noxious gases.50 However, when used as reinforcement in polymers, a major problem is their incompatibility with hydrophobic polymers such as polypropylene due to their hydrophilic nature. Insufficient wetting of the fibres by the polymer matrix has been proved to lower the tensile strength51 and stiffness of a composite, as a poor interface cannot effectively transfer the stress from the polymer matrix to the fibres. Moisture absorption is another problem with natural fibres, as the moisture present is known to cause voids, reducing the strength of the composite. The moisture content will be variable, depending on relative humidity or wetting of the composite. Moisture will interfere with melt dispersion and processing of the composites since processing temperatures of the order of 180–200 ∞C are necessary. When the moisture is removed from natural fibres, they become brittle, thereby losing their effectiveness as reinforcements. Utilisation of fully biodegradable composites has more than one positive side: the biopolymer matrix and the cellulose fibres are fully degradable and all the costs associated with the waste management are reduced to a minimum. Naturally sourced polymer materials do not endanger wildlife during utilisation and decomposing, and do not contribute to greenhouse emissions. There is the possibility of recycling biodegradable thermoplastic composites, within a relatively short period of their lives, due to environmental effects (moisture, UV, natural decomposition). Recyclability and mechanical properties of biodegradable composites can be optimised with additives such as plasticisers, silane coupling agents and hydrogen bonding agents.52,53 Naturally sourced biopolymers and green composites still need to pass certain requirements to reach a large-scale of utilisation:29 (i) Fully biodegradable polymers and their composites must have the ability to satisfy urgent market needs. Some 30% of the plastics in municipal waste originate from goods that have been less than 1 year in use and tend to be heavily soiled by food and organic residues. Biodegradable alternatives could replace a large part of this voluminous part of plastic waste, which is difficult to dispose of or to recycle. The waste could be

266

(ii)

(iii)

(iv)

(v)

Green composites

diverted from landfills and incineration to composting sites near the end user. Biocomposites and biopolymers are suitable for the packaging industry, especially for producing goods that are not subject to durability and are likely to end up soiled with organic matter. Such new materials require new and efficient composting systems, naturally present in rural areas, to be installed in urban areas. In urban areas new systems for collecting and composting ‘garden and kitchen waste’ are being installed for reducing landfill problems, especially in western European countries. Green composites must compete with the present plastics as far as quality and processing performance are concerned. Bacterial polyesters, for example, meet various quality and processing performance requirements. These materials can be processed by all types of thermal manufacturing. However, the esters are not flexible enough to form films or foils. They also tend to become brittle and to lose their vapour barrier properties. It is expected that these limits will be overcome by improving blend formulations. One specific obstacle is that they have to meet requirements for registration as food packaging materials, a process that requires long and costly tests. Bacterial polyesters are not yet allowed for use as food packaging material, because the esters represent a novel product. The crucial obstacle at present is that biodegradable materials have to meet competitive price limits. The present prices for bacterial polyesters are far too high to be accepted on a large scale by the processing and packaging industry, mostly due to the raw material prices and to the small-scale production units. They can be lowered to acceptable levels by investing in larger units in countries where inexpensive raw materials are available. Thus they will be able to meet the price limits. Since bacterial polyesters increasingly meet the requirements for penetration of the mass market, and since more and more consumers accept composting as an environmentally sound way of recycling organic materials, polyesters are expected to penetrate a significant part of the short-lived and contaminated plastic products’ markets in this decade.

12.7.2

Incineration of polymers

Incineration of plastics and composites is the opposite to the above, in the sense that it involves annihilation of the material, with inevitable air and land pollution, due to poor combustion and to gaseous/liquid/solid products of the process. Around 1 million tonnes of plastic materials are incinerated in the USA alone annually, and 11% of the present total is poly(vinyl chloride). Western Europe generates around 100 million tonnes of municipal solid waste annually, with 7% being plastics. Of that part, 30% is disposed of by

Re-use, recycling and degradation of composites

267

incineration with and without energy recovery and 70% by landfilling. The polymer matrix resins used in composites possess higher calorific values (30 000 kJ/kg) than coal (26 000–30 000 kJ/kg).25 The presence of noncombustible fillers and reinforcements reduces the calorific content down to 15 000 kJ/kg. The rate of combustion of plastics is determined by parameters such as surface to volume ratio, density, ignition point, combustion rate, etc. The flammability parameter, defined as the ratio of the heat of combustion divided by the heat required to produce volatiles, which in turn promotes ignition, is used for quantitative analysis of solid fuels. According to this characteristic, polypropylene and polystyrene have combustion performance comparable to wood. Some advantages of an increased content of plastic waste for incineration have been listed as: (i) higher temperature of incineration due to volatiles, which reduce leachability of the fly ash and (ii) shorter combustion zones and more intensely burning fire. Particular disadvantages from increased plastic waste are: (i) increased formation of NOx with an increase in incinerating temperatures, which automatically contradicts the above listed advantages, (ii) increased CO emissions, and (iii) other excessive emissions such as chlorine, dioxins and furans. In general, there is no policy to support increased implementation in incineration of synthetic materials and this process is widely used only because other avenues of re-use and re-processing have not yet been fully explored.

12.8

Conclusions and future trends

Innovations in composite recycling methods and processes are on the rise. Research projects are being established between universities worldwide and predominantly the automotive industries, with the aim of achieving reduction in composite waste, and an increase in the amount of re-used polymer and reinforcing materials. Thermoplastic polymers are easily recyclable and the research in this area focuses on obtaining a high degree of purity of recyclate (> 80%) and successful separation of matrix from reinforcing materials. The question regarding product application of recycled reinforced thermoplastics still remains open, as does whether the market holds sustainable avenues for increased amounts of recycled materials. Re-use of thermosetting polymers poses a different challenge, with thermosets being highly resistant to mechanical, thermal and environmental effects. The development of recycling processes is complicated by the presence of a high volume fraction of reinforcing fibres and is put under additional pressure by the economics of re-production compared with manufacture of virgin materials. The projects recently initiated in this area seem to be highly promising, though still behind a justified economical outcome. An exciting discovery has been revealed in the proprietary technology,

268

Green composites

totally degradable plastic additive (TDPA™), which is able to degrade common plastics in a controlled, ecologically responsible manner. The additive is being tested for various mechanical and environmental effects and has already found a number of successful applications on the market. There is an issue concerning the impact of such imposed synthetic polymer degradation on the aquatic environment, which has not yet been investigated. It is known that polymer degradation affects marine species and this issue needs to be covered before the technology is applied globally. The most recent advances in waste management and plastic waste reduction have taken the direction of leading green composites into large-scale production. Although the majority of naturally sourced and biodegradable composites cannot find their application in responsible mechanical structures, implementation of these materials in packaging and similar products could lead to positive changes in today’s environment. A reduction in CO2 emission could regenerate the atmosphere and turn this century’s negative chain of natural phenomena back to positive. If production of synthetic plastic materials is reduced to a minimum, it will be possible to manage the remaining plastic waste by re-use, degradation and incineration, without it having a highly detrimental impact on the environment. More work is needed in the optimisation of the properties of recycled composites, in the development of economical processing and in educating the market to accept re-used products. Scientists and legislators have put industry under pressure to implement novel technologies and old materials in their production. What is needed is a balance between production and disposal of synthetic polymers, satisfying both the market and the environment.

References 1 Gordon, J.E. (1976). The New Science of Strong Materials, Aylesbury, Bucks: Hazell Watson & Viney Ltd. 2 Rowell, R.M. (1998). Economic opportunities in natural fiber–thermoplastic composites. In Science and Technology of Polymers and Advance Materials, ed. Prasad, P.N. et al., New York: Plenum Press, 869. 3 Curlee, T.R. and Das, S. (1991). Plastic wastes. In Management, Control, Recycling and Disposal, New Jersey, NJ: Noyes Data Corporation. 4 Shent, H., Pugh, R.J. and Forssberg, E. (1999). A review of plastic waste recycling and the floatation of plastics. Resources, Conservation and Recycling, 25, 85–109. 5 Henshaw, J.M., Han, W. and Owens, A.D. (1996). An overview of recycling issues for composite materials. Technomic. Journal of Thermoplastic Composite Materials, 9, 4–20. 6 Henshaw, J.M., Owens, A.D., Houston, D.Q., Smith, I.T. and Cook, T. (1994). Recycling of a cyclic thermoplastic composite material by injection and compression molding. Technomic. Journal of Thermoplastic Composite Materials, 7, 14–29. 7 Scobbo, J.J. (1991). Polymer composites from mixed recycled plastics, International SAMPE Technical Conference.

Re-use, recycling and degradation of composites

269

8 Papaspyrides, C.D., Poulakis, J.G. and Arvanitopoulos, C.D. (1995). Recycling of glass fiber reinforced thermo-plastic composites. I. Ionomer and low density polyethylene based composites. Resources, Conservation and Recycling, 14, 91–101. 9 Shaw, W.J.D. (1998). Current understanding of mechanically alloyed polymers, Materials Science Forum, 19, 269–72. 10 Bai, C., Spontak, R.J., Koch, C.C., Saw, C.K. and Balik, C.M. (2000). Structural changes in poly(ethylene terephthalate) induced by mechanical milling. Polymer, 41, 7147–57. 11 Nesarikar, A.R., Khait, K. and Mirabella, F. (1997). Self-compatibilisation of polymer blends via novel solid-state shear extrusion pulverisation. Journal of Applied Polymer Science, 63, 1179–87. 12 Casale, A. and Porter, R.S. (1996). XVIII AIM School Meeting Proceeding, 9–14 June, 189. 13 Cavalieri, F. and Padella, F. (2002). Development of composite materials by mechanochemical treatment of post-consumer plastic waste. Waste Management, 22, 913–16. 14 Pedella, F., Magini, A. and Incocciati, E. (1998). Patent No. RM98000372. 15 Padella, F., Filacchioni, G., La Barbera, A., Magini, M. and Plescia, G.P. (1999). A way towards plastic recycling: the mechanochemical treatment of polymeric waste. In First International Conference on Solid Waste Proceedings, Rome, Italy 7–9 April, 138. 16 Adem, E., Avalos-Borja, M., Carrilo, D. et al. (1998). Crosslinking of recycled polyethylene by gamma and electron beam irradiation. Radiation Physics and Chemistry, 52, 171–3. 17 Adem, E., Sanchez, E., Aliev, R. and Burillo, B. (1999). Radiation crosslinking of the virgin and recycled low density polyethylene blend. Revista de la Sociedad de Quimica de Mexico, 43, 201–3. 18 Czvikovszky, T. (1995). Reactive recycling of multiphase polymer systems through electron beam. Nuclear Instruments and Methods in Physics Research B, 105, 233– 7. 19 Czvikovszky, T. and Hargitai, H. (1997). Electron beam surface modifications in reinforcing and recycling of polymers. Nuclear Instruments and Methods in Physics Research B, 131, 300–4. 20 Czvikovszky, T., Hargitai, H., Racz, I. and Csukat, G. (1999). Reactive compatibilisation in polymer alloys, recyclates and composites. Nuclear Instruments and Methods in Physics Research. B, 151, 190–1. 21 Burillo, G., Clough, R.L., Czvikovszky, T. et al. (2002). Polymer recycling: potential application of radiation technology. Radiation Physics and Chemistry, 64, 41–51. 22 Allred, R.E. (1996). SAMPE Journal, 32, 46. 23 Petterson, J. and Nilsson P.J. (1994). Recycling of SMC and BMC in standard processing equipment. Journal of Thermoplastic Composite Materials, 7(1), 56–63. 24 Inoh, T., Yokoi, T., Sekiyama, K.I., Kawamura, N. and Mishima, Y. (1994). SMC. recycling technology. Journal of Thermoplastic Composite Materials, 7(1), 42–55. 25 Scheirs, J. (1998). Recycling of polymer composites. In Polymer Recycling, Science, Technology and Applications. Chichester, UK: John Wiley & Sons, 372. 26 Clean Washington Centre, Recycling Advanced Composites, Report No. IBP-95-3. 27 Winter, H., Mostert, H.A.M., Smeets, P.J.H.M. and Paas, G. (1995). Recycling of sheet-molding compounds by chemical routes. Journal of Applied Polymer Science, 57, 1409.

270

Green composites

28 Kennerley, J., Fenwick, N.J., Pickering, S.J. and Rudd, C.D. (1996). The properties of glass fibres recycled from the thermal processing of scrap thermoset composites. Proceedings ANTEC, 890–894. 29 Chandra, R. and Renu R. (1998). Biodegradable polymers, Progress in Polymer Science, 23, 1273–335. 30 Huang, S.J., Bryne, C. and Palisko, J.A. (1980). ACS Symp. Ser., 121, 299. 31 Albertsson, A.C. (1985). Preprints of the International Symposium on Characterization and Analysis of Polymers, 477. 32 Albertsson, A.C., Anderson, S.O. and Karlsson, S. (1987). Polymer Degradation Stability, 18, 73. 33 Wypych, G. (2003). Biodegradation. In Handbook of Material Weathering, 3rd edn. ChemTech Publishing, Chapter 19, 523–33. 34 Bonhomme, S., Cuer, A., Delort, A.M., Lemaire, J., Sancelme, M. and Scott, G. (2003). Environmental biodegradation of polyethylene, Polymer Degradation and Stability, 81(3), 441–52. 35 Chiellini, E., Corti, A. and Swift, G, (2003). Biodegradation of thermally-oxidized, fragmented low-density polyethylenes. Polymer Degradation and Stability, 81, 341– 51. 36 Modern Plastics and Charles A. Harper, Technology Seminars, Inc (2000). Plastics recycling and biodegradable plastics. In Modern Plastic Handbook, Lutherville, MD. McGraw-Hill. 37 Pryweller, J. (1997). Ford driving recycled nylon applications. Plastic News, Feb 24, 7–9. 38 Pryweller, J. (1997). Cost is king in auto-related recycling. Plastic News, Sep 8, 1– 8. 39 Sherman, L.M. (1996). Compounders take the lead in post-use bumper recycling. Plastics Technology, Mar, 27–9. 40 Grande, J.A. (1996). Ford is targeting 50% use of recycle-content resin by 2002. Modern Plastics, Jul, 32–3. 41 Pryweller, J. (1998). Projects could turn plastics into a recycling headliner, Plastic News, Feb 23, 2–9. 42 Mohanty, A.K., Misra, M. and Hinrichsen, G. (2000). Biofibres, biodegradable polymers and biocomposites: an overview, Macromolecular Materials and Engineering, 276/ 277, 1–24. 43 Joseph, P.V., Kuruvilla, J. and Thomas, S. (1999). Effect of processing variables on the mechanical properties of sisal-fiber-reinforced polypropylene composites. Composites Science and Technology, 59, 1625–40. 44 He, Y., Asakawa, N., Masuda, T., Cao, A., Yoshie, N. and Inoue, Y. (2000). The Miscibility and biodegradability of poly(3-hydroxybutyrate) blends with poly(butylenes succinate-co-butylene adipate) and poly(butylenes succinate-co-e-caprolactone). European Polymer Journal, 36, 2221–9. 45 Wielage, B., Lampke, T., Marx, G., Nestler, K. and Starke, D. (1999). Thermogravimetric and differential analysis of natural fibres and polypropylene. Thermochimica Acta, 337, 169–77. 46 Joly, C., Gauthier, R. and Chabert, B. (1996). Physical chemistry of the interface in polypropylene/cellulosic-fibre composites. Composite Science and Technology, 56, 761–5. 47 Li, Y., Mai, Y.-W. and Ye. L. (2000). Sisal fibre and its composites: a review of recent developments. Composite Science and Technology, 60, 2037–55.

Re-use, recycling and degradation of composites

271

48 Eichhorn, S.J., Sirichaist, J. and Young, R.J. (2001). Deformation mechanisms in cellulose fibres, paper and wood. Journal of Materials Science, 36, 3129–35. 49 Brouwer, W.D. (2000). Natural fibre composites: where can flax compete with glass? Centre for Lightweight Structure. Delft Technical University (TUD)–TNO. 50 Van de Velde, K. and Baetens, E. (2001). Thermal and mechanical properties of flax fibres as potential composite reinforcement. Macromolecular Material in Engineering, 286, 342. 51 Tjong, S.C., Xu, Y. and Meng, Y.Z. (1999). Composites based on maleated polypropylene and methyl cellulosic fiber: mechanical and thermal properties. Journal of Applied Polymer Science, 72, 1647. 52 Hodzic, A., Shanks, R.A. and Leorke, M. (2002). Polypropylene and aliphatic polyester natural fibre composites. Polymers and Polymer Composites, 10(4), 281–90. 53 Wong, S., Shanks, R.A. and Hodzic, A. (2002). Properties of poly(3-hydroxybutyric acid) composites with flax fibres modified by plasticiser absorption. Macromolecular Material in Engineering, 287, 647–55.

13 Reprocessing J.C. ARNOLD Swansea University, UK

13.1

Introduction

The worldwide production of plastics has grown from less than 5 million tonnes per year in the late 1950s to nearly 100 million tonnes per year today (Environment Agency, 2001; APME, 2001a). In addition to primary production, the use of plastics to make goods has increased dramatically (from about 155 000 tonnes in the UK in 1950 to about 4.6 million tonnes in 1999), and has more than doubled since 1973, as shown in Fig. 13.1 (Environment Agency, 2001). About 70% of plastics used worldwide are the ‘bulk’ polymers of polyvinyl chloride (PVC), polypropylene (PP), polystyrene (PS) and polyethylene (PE), as seen in Fig. 13.2. In Western Europe the consumption of plastics and polymer composites is predicted to grow by 12 million tonnes, from 24.9 million tonnes to 36.9 million tonnes between 1995 and 2006; this is equivalent to a growth of 4% per year (APME, 1998). However, over the last few years, the annual percentage growth has exceeded the forecasted 4%. For example, a 4.8% growth in consumption was recorded in 1998, while a further rise in consumption of

Consumption (ktonnes)

5000 4000 3000 2000 1000 0 1973 1976 1979

1982

1985 1988 Year

1991

1994 1997

13.1 Annual plastic consumption in the UK from 1973 to 1997.

272

Reprocessing

273

Polyester resin Phenolic resins ABS/SAN PUR PET PS PP PVC PE 0

0.5 1 Consumption (million tonnes)

1.5

13.2 Consumption of major plastic types in the UK in 1998 (ABS/SAN = acrylonitrile butadiene styrene/styrene acrylonitrile; PUR = polyurethane; PET = polyethylene terephthalate).

5.4% was recorded in 1999 (APME, 2000, 2001a,b), as such consumption is already 33.6 million tonnes. The consumption of plastics is of a similar size in the USA and growing at about 1–2% per year due mainly to population increase (EPA, 2000). There are many significant problems when dealing with waste plastics and composites. Mechanical recycling is possible, but is generally only successful where relatively clean single-source materials are available. In cases where a mixed waste stream is present, the recycled material will be of inferior quality unless significant sorting and cleaning are employed. Certain materials pose a particular problem in this respect; for instance the presence of small amounts of PVC can cause significant degradation of other materials on reprocessing (Arnold and Maund, 1999a). The wide range of material grades, presence of potentially hazardous additives and relatively low cost of virgin materials also add to the problems. Recycling of polymer composites presents further challenges. In many cases composites are produced using thermosetting matrix materials that do not allow further melt processing. In addition, fillers and fibres are hard to separate from the matrix material. Methods of dealing with waste composite materials are described below, but in many cases these involve either using the material as a low grade filler for further composites or using costly chemical separation methods.

13.2

Management of waste plastics and composites

Plastics and composites nowadays are used within all major sectors of industry, with the predominant sectors in terms of tonnage being packaging, building

274

Green composites Agriculture 3%

Other household/ domestic 19%

Automotive 8% Large industry 5%

Electrical and electronic 8%

Building and construction 18%

Packaging 39%

13.3 Plastic consumption by industry sector across Western Europe in 1997.

and construction, electrical and automotive respectively, as seen in Fig. 13.3. Plastic packaging is used for almost half of all packaged goods, consuming 11.6 million tonnes in Western Europe in 1997 (APME, 1999). However, as plastics are often significantly lighter than other materials, they account for only 17% of total packaging by weight in Western Europe. During 1995, 4.89 million tonnes of plastic was used by the European building and construction industry; however, this figure is expected to rise to 8 million tonnes by the year 2010. One of the major problems with dealing with waste plastic is the large number of different materials and grades available. It is thought that there are as many as 20 000 different varieties currently in use, which can be divided into around 50 major types, or families, of polymer (Lundquist et al., 2000). Plastics are very versatile and have a range of properties; for example, they are relatively cheap, light and easily processed, which has led to their widespread use. In many cases they can be tailored to have good durability and Fig. 13.4 shows the average lifespan of plastic products (Environment Agency, 2001). It can be seen that many have lifetimes upwards of 10 years. However, the qualities of plastics can also be disadvantageous. Their good durability, often used as a sales argument, can become a problem in applications where the service life is short, such as packaging. Those sectors with the shortest lifetimes will be the most problematic for waste management and, not surprisingly, the packaging sector was the first to fall under recycling legislation.

13.2.1

Waste plastic arisings

The UK produces around 3 million tonnes of plastic waste each year, the major source of post-consumer plastic waste being households, which contribute 71%. There has been considerable growth in plastics in household waste in the UK since 1945 (Environment Agency, 2001), with the breakdown by

Reprocessing

275

Packaging Computers Telecommunications Automotive Large domestic appliances Profiles, e.g. window frames Pipes/ducts 0

20 <2

2 to 5

40

60 % of products

5 to 10

10 to 20

80 20 to 40

100 >40

13.4 The lifespan of plastic products by sector (years).

material shown in Fig. 13.5. As a result, plastics now make up about 11% of household waste by weight (24 million tonnes in the USA, 2.8 million tonnes in England and Wales), of which 6% is hard plastic and 5% is film, mainly of packaging origin (Bickerstaffe, 1996). Packaging is the largest source of plastic waste (54–60%), three-quarters of which is used to package foodstuffs while the last quarter is used as distribution packaging. The UK produced just less than 10 million tonnes of waste packaging in 1999, of which 1.6–1.7 million tonnes were plastics (Environment Agency, 2001). The sources of post-consumer plastic waste are shown in Fig. 13.6 (Environment Agency, 2001; APME, 2001b). This shows that domestic waste PET 6%

PS + EPS 12%

Others 13% PP 17%

PVC 13%

LDPE 25%

HDPE 14%

13.5 The relative proportions of different plastics in the domestic waste stream.

276

Green composites Distribution and industry 21%

Electrical and electronic 4% Building and construction 3%

Municipal solid waste 67%

Automotive 4% Agriculture 1%

13.6 Sources of plastic waste in Western Europe in 1999.

accounts for the largest bulk and unfortunately also presents the greatest challenge for recycling due to a very widely distributed source of highly mixed and contaminated material. The other sources generally pose fewer problems for recycling as the material is normally available in fewer locations, with less contamination and fewer materials types. As a consequence, the recycling rates for these types of waste are higher than for municipal domestic waste.

13.2.2

Waste management options

The main options for dealing with waste plastics are to landfill, incinerate or recycle. The options employed by various countries are shown in Fig. 13.7 (Environment Agency, 2001). The landfilling of plastics is a waste of resources, and there are also concerns over the long-term leaching of chemicals from plastics in landfill, but current information is not comprehensive. With the increase in municipal solid waste (2% per annum typically), increasing landfill 100 80

%

60 40 20 0 d rlan itze Sw

ark nm De

Landfill

Jap

an Country

Incineration

A US

UK

Recycle/re-use

13.7 Waste management options employed in various countries.

Reprocessing Reduction

Mechanical recycling

Re-use

Elimination of excess packaging

In-house

Feedstock recycling

Incineration

277

Landfill

PCW Energy recovery

Revitalisation

Return and refurbish

Reduction in resource use

Chemical production

Re-usable materials

Fuel production

Energy

Resource loss

13.8 The options available for dealing with waste plastics.

taxation, special waste regulations for landfill, and in order for governments to meet set targets, landfilling of all waste is not acceptable or desirable. Waste management techniques are required in order to meet these targets. In order to manage and reduce the amount of waste plastics that are being landfilled, there are a series of waste management options available (Lundquist, 2000; Bickerstaffe, 1996). These are shown schematically in Fig. 13.8 and include routes such as landfill, where there is a loss of resource, and alternative routes that either result in new materials and products or energy recovery.

13.3

Methods of sorting and separating plastics and polymers

In contrast to virgin plastics, material for recycling hails from several sources but falls into three broad classes: post-processing or industrial waste; commercial scrap; and post-consumer waste (Belofsky, 1995). Post-processing and commercial scrap is relatively clean, easily collected and identified, and can be recycled after undergoing some separation and purification (Rebeiz and Craft, 1993). However, post-consumer waste (PCW), which constitutes the lion’s share of plastic waste, is also the most problematic material stream to deal with.

13.3.1

Collection

The collection of plastics from PCW has been most successfully developed with plastic bottles as they are limited to three distinct materials and can

278

Green composites

relatively easily be identified and collected (Environment Agency, 2001; Rebeiz and Craft, 1993). The USA has extensive recycling schemes for plastic bottles and currently has a recycling rate of over 30%. The first plastic bottle schemes in the UK were launched in 1989/90 and were heavily sponsored; however, their early success has led to about 45 local authorities now operating plastic bottle collection schemes in their areas (Recoup, 2001; McDonald and Ball, 1998). There is a wide variety of techniques used to collect plastic bottles with a series of factors determining the system applied. The most successful techniques for collection are bring schemes, i.e. bottle banks and kerbside collection, with both techniques requiring an initial dependency on the public to sort the materials correctly before collection (Lundquist et al., 2000). There is also a number of variables to relate the effectiveness of the system. These range from collection frequency and cost, socio-economic factors and type of container used, to more specific variables such as site density. For example, a low-density site can achieve material recovery rates of less than 5%, but a high-density site can achieve recovery rates of around 40% (Recoup, 2001). Nevertheless, recovery of plastic bottles from PCW is increasing continually both in the UK and in Europe (Lundquist et al., 2000). Recycling rates are also increasing in the USA, but at a slower rate. Without public contributions, recycling from domestic waste would not be possible. The public enjoys a pivotal role in determining the route taken by domestic waste. Without the public’s conscious decision to support an alternative route for their waste, there will be no feedstock for the PCW industry (McDonald and Ball, 1998). This puts these industries in the unique position of not being able to purchase their raw material; they must rely on the goodwill of the general public (Belofsky, 1995).

13.3.2

Separation methods

Most collection schemes generate heterogeneous material streams, which must be efficiently sorted to provide generic material streams. Frequently used sorting methods are generalised as manual sorting and automated sorting, both of which are described below. 13.3.2.1

Manual sorting

Manual sorting can be performed at the source of waste by giving directives to the consumers, as in kerbside collection schemes. It is also carried out on an industrial scale at material recovery facilities and by reprocessors. Virtually all manual sorting and identification methods require trained staff that sort the plastic into polymer type and colour at rates of around 60–100 kg/person/ hour (Recoup, 2001). Large objects identifiable with a particular application,

Reprocessing

2

1

5

1. 2. 3. 4.

3

6

Polyethylene terephthalate (PET) High density polyethylene (HDPE) Vinyl/PVC Low density polyethylene (LDPE)

279

4

7

5. Polypropylene (PP) 6. Polystyrene (PS) 7. Other

13.9 The SPI coding system for plastics identification.

such as PET for carbonated bottles, are the simplest to separate in this way. Other methods of sorting include material characteristics, ‘burn and sniff’ methodology (MRW, 2001) and above all else, experience. Identification and separation is made even easier by the use of logos or bar coding, the best known of which is the Voluntary Plastic Container Coding System of the Society of the Plastics Industry, USA, shown in Fig. 13.9. Established in 1988 the six specified generic materials cover 95% of the current domestic plastic waste produced (Lundquist et al., 2000; Belofsky, 1995). However, the above codes only specify a plastic family, not detailed material grades, and as such create difficulty in maintaining the quality of the feedstock needed for credible recyclate quality. 13.3.2.2

Automated identification and separation

Commercially available sorting equipment is able to sort plastics rapidly by polymer type and colour. Automated identification can offer potential advantages, such as reduction of separation costs by as much as 24% compared to manual sorting, and separation accuracy of better than 90% appears to be routinely achievable (MRW, 1998). Automated identification also allows for greater volumes of waste to be processed. The main techniques for identification and separation are discussed below. 13.3.2.3

Density methods

The density of commonly used plastics varies between 0.9 g/cm3 and 1.7 g/cm3, so density-related methods can be used for both purifying and sorting plastic waste. The most commonly used techniques are air classifiers, hydrocyclones, and float/sink baths (Ferrara and Malloy, 1999; Ferrara et al., 2000; Shen et al., 2002).

280

Green composites

Table 13.1 Feedstock that cannot be separated by density methods Feedstock

Problem

Highly contaminated with dirt/ oil

Agglomerations of polymers with different densities, dirt and entrapped air make sorting inefficient.

Plastics of similar density

Immiscible materials may be present in the same fraction.

Polymer blends

Density does not correspond to a single polymer.

Composites/ foams

Density does not correspond to the matrix polymer.

The methods provide a high throughput and sorting efficiency if the feedstock contains only a few materials with reasonable differences in density. They are most frequently used for commingled polyolefins; however, precautions must be taken to avoid the feedstocks described in Table 13.1. More efficient separation is possible by introducing plastic granules into a pressurised vessel containing a supercritical fluid. By altering the pressure and composition, the density of such fluids can be controlled to within ± 0.001 g/cm3. This allows the selective separation of all polymers with the exception of those with overlapping densities such as PVC and PET (Shen et al., 2001). 13.3.2.4

Infra-red and X-ray spectroscopy

This method is frequently used as a way of identifying plastics (Warmington, 2001). A specimen is exposed to infra-red radiation or X-rays (Drelich et al., 1999) and the difference between the source and reflected spectra are used to identify the molecular structure of the material. The use of filters can improve the speed of the analysis, usually limited by the complexity of the software used to analyse the spectra (Lundquist et al., 2000). The most frequently used spectra, near infrared range (NIR), cannot be applied to dark coloured objects such as those common within the automotive sector and thus the mid infrared range (MIR) was developed. IR spectroscopy methods can be classed in two categories: those that are fully automated (Warmington, 2001; Kosier et al., 1999) and those that are manually operated (MRW, 2001). 13.3.2.5

Thermal separation

Differences in the thermal properties of plastics, such as their softening and melting points, can be exploited to separate mixed post-consumer plastic flake. This form of separation is most effective if differences in melting or softening points of the two plastics are considerably large. Generally, a mixture of different plastic particles comes into contact with heated rolls or

Reprocessing

281

Rotating charging drum Charged plate

Vibratory feed

Charged roller

Positive Negative Charged product

13.10 A schematic diagram of an electrostatic separator.

belts, which sorts the plastics by selective thermoadhesion of the softened particles to the rolls or belts. Similar methods use heated drums instead of the heated rolls or belts; however, the flakes must form a monolayer to be effectively heated or sorting cannot occur (Dvorak, 2001). 13.3.2.6

Electrostatic separation

This method relies on the fact that different polymers have different electrostatic properties such that, if particles of two dissimilar polymers are rubbed against each other, one will gain a negative charge with the other gaining a positive charge. This can then be used to separate the materials, as shown in Fig. 13.10. The drawback of this technique, as with thermal separation, is that it is limited to separation of a mixture of just two dissimilar materials. 13.3.2.7

Selective dissolution

In a given solvent, plastics will dissolve at different temperatures; a typical process involves the plastics being washed, milled and then dissolved in a heated solvent. By raising the temperature step-wise and using a controlled temperature–solvent extraction, individual polymers can be separated and

282

Green composites

then precipitated by adding a non-solvent or through centrifuge methods. Frequently used solvents are ketones and acetic acids, which are low in toxicity and degrade quickly. All solvents used in the process can be recovered and reused; however, this is possibly due to their volatile and expensive natures (Drelich et al., 1999). It should be noted that this technique is classed as a separation/mechanical recycling method because, although the plastic is dissolved down to its individual polymer chains, it is not separated to the original feedstock as in chemical recycling. Selective dissolution may also be used in the recycling of fibre-reinforced composites, where several consecutive washings with a solvent allow the separate recovery of the matrix and reinforcements. While all of the above techniques have specific disadvantages, there are some common problems. Each one is capital intensive and although savings are possible compared to manual sorting, unless high throughputs of material are possible, the economic viability is limited (Warmington, 2001). Most of the techniques are aimed at the waste packaging sector and as such in 1999 only three automated separators were in operation in the UK, operating floatation and hydrocyclone separators usually in combination to achieve as high a quality as possible (Drelich et al., 1999). While separation accuracy of 90% seems possible, studies have shown that the increase of throughput rate tends to decrease sorting accuracy by a considerable percentage (Kosier et al., 1999). At the present time, the plastics recycling industry requires relatively pure feed materials as accomplished by expensive sorting operations. The markets for clear materials are large; however, outlets for tinted materials are limited, while multi-layer bottles which usually have a barrier coating sandwiched between two layers of PET are usually impossible to recycle. These barrier coatings are designed to extend the shelf-life of the products. These come in a number of forms such as polyvinyldichloride (PVdC), plasma coatings such as carbon or silicon or reactive barriers (Bucklow, 2000). However, the main recycling concerns seem to be with ethylene vinyl alcohol (EVOH) and nylon MXD6. It has been found that a PET bottle may be made up of at least five different materials, making accurate sorting nearly impossible (Recoup, 2001). This causes problems with the contamination of feedstock; for example, PVC contamination of levels over 200 ppm can significantly degrade PET during extrusion processes. The PET pellets become significantly discoloured and the intrinsic viscosity of the PET lowered, caused by the evolution of hydrochloric acid gas bubbles from the PVC, promoting chain scission of the PET (Dvorak, 2001).

Reprocessing

13.4

Methods of recycling plastics

13.4.1

Mechanical recycling

283

Mechanical recycling can be defined as processing of waste plastics by a physical means into new plastic products. Mechanical recycling is the European plastics industry’s preferred recovery technique (APME, 1999). Because of this fact the recovery of post-use plastics waste is increasing as can be seen in Figure 13.11 (APME, 2001a). Mechanical recycling of post-use plastic waste in Western Europe has the potential to double, from 1.2 million tonnes in 1995 to 2.7 million tonnes in 2006. This is equivalent to a growth rate of 8.4% per year. Generally, however, the industry believes it will be difficult significantly to exceed the forecast recovery rate of around 11% of post-use plastic waste. There are a number of reasons for this. There is an imbalance between the waste collectable and the potential end markets. Plastic waste supply is not usually uniform over time (Rebeiz and Craft, 1993) and there is also a clear lack of end markets for recycled products (Fletcher and Mackay, 1996). What markets there are usually only accommodate HDPE, PET and PVC, their ease of collection and separation making them the dominant resins. Pre-consumer recycling recovery rates are static. In-house recycling will always occur due to cost issues; however, pre-consumer recycling has not increased and shortfalls have even been reported, although it easier to recycle compared with post-consumer waste. 3000

Amount recycled (t/year)

2500

2000

1500

1000

500

0 1989

1991

1993

1995

Mechanical recycling

1997

1999

2001

2003

Forecast potential mechanical recycling

13.11 Plastic recycling rates for Western Europe.

2005

284

Green composites

There are present large quantities of mixed plastic waste, where the difficulties and energy consumed in separating and cleaning outweigh the environmental gain of mechanical recycling (APME, 2001a; Subramanian, 2000). This is especially problematic in the UK where the capacity to treat mixed plastic waste has declined, possibly due to issues of cost (Environment Agency, 2001; Cooper, 1998; Patel et al., 2000). Development in the technologies required to treat these wastes successfully is likely to improve this situation in the future; however, it is currently uneconomical, even if mechanical recycling is reported to be the best energy saving option for plastic waste recovery. The export and import of feedstock and recyclate occurs because there are wide variations in mechanical recycling between countries (APME, 1999). For example, high performing countries such as Germany achieve levels of 18% recovery; however, low performing countries such as the UK achieve levels as low as 7.4%. Countries currently achieving high mechanical recycling levels can export plastic recyclate to neighbouring countries. However, once these importing countries begin processing and selling their own recyclate in greater quantities, saturation of the market may be quickly reached. In the UK 18% of packaging recycled was sold to overseas recyclers in 2000; however, this has a detrimental effect on the industry. Overseas buyers, especially those from the far east, can offer higher material prices because labour and reprocessing costs are a fraction of the UK’s (Recoup, 2001). Currently as things stand, many UK recyclers are better off importing their waste at lower prices from other European countries as commercially it is difficult to commit capacity to the recycling of UK plastic waste. Mechanical recycling has two forms of feedstock: in-house and postconsumer plastic waste. In-house recycling of industrial scrap has long been practised at a high percentage by the plastics industry (Fletcher and Mackay, 1996). Runners, sprues and off-specification products are easily identified and are of a high quality. As such, instead of being rejected as waste, they are granulated and blended in with virgin material to give acceptable products (Lundquist et al., 2000). Post-consumer recycling, operating at much lower levels, is the greatest challenge for the plastics industry. In comparison with in-house recyclate, post-consumer waste (PCW) undergoes degradation of the polymer during service and is likely to be mixed and contaminated. The PCW requires intermediate stages of decontamination from metal residues and other contaminants and the waste products are reduced in size by primary grinding. Following contaminant separation, a secondary grinding step further reduces the size of the particles before cleaning and drying, as shown in Figure 13.12. Homogeneous PCW can be reprocessed in much the same way as in-house recycling, but commingled PCW requires special processing techniques (Lundquist et al., 2000). The final steps in the process can involve extrusion with melt filtration for

Reprocessing

285

Paper

Air classifier Granulator

Washer Water in Water out

Hydrocyclone

Metals

Dryer

Recyclate Melt extruder

13.12 A schematic diagram of a typical mechanical recycling process.

purification and extrusion compounding to pellets. Plastic recycling uses modified extrusion and moulding equipment. In the extruder, the main modifications are in the feeding area where low bulk density feedstock may have to be stuffed or crammed mechanically into the feed zone of the screw to obtain steady input (Belofsky, 1995). The extruders for PCW are also modified. Single screw types have dualdiameter screws; a larger diameter front feed screw to densify the input, and a smaller homogenising screw zone for the discharge, seen in Fig. 13.13. Mechanical rams stuff feedstock into the extruder. Vented screws are common to degas dirty or wet stock and various filtering and screening devices have been made to try to avoid shutdowns for screen maintenance (Belofsky, 1995).

286

Green composites

Vibrated hopper Vent Screen pack

Chipper Water bath

13.13 A schematic diagram of a dual diameter extruder used to recycle low density plastic waste.

Twin-screw extruders are useful for reclaim as they generate high shear rates for good mixing with low energy consumption and can be efficiently vented. A co-rotating twin screw can be run at high screw speed with low wear, but a counter-rotating twin provides better venting or degassing. The final product from these steps is granules and is termed recyclate.

13.4.2

Revitalisation and restabilisation

Processing operations and the conditions encountered during service life tend to cause irreversible chemical changes and as such, degrade the properties of the materials contained in the product (Lundquist et al., 2000; Rebeiz and Craft, 1993; Arnold and Maund, 1999b). Additives and stabilisers are added when plastics are first manufactured, but these are consumed during processing and use, so when the product is discarded the plastic will not always be reusable (Rebeiz and Craft, 1993). Some of the common problems encountered during recycling and the mechanisms responsible are shown in Table 13.2. This shows that combinations of chain degradation, additive loss, impurities and immiscible plastics can lead to a variety of problems that may be manifest during processing or later service. For appropriate revitalisation to occur, it is first important to know the effects of the previous life cycles on the material properties, secondly, the material characterisation methods, thirdly, the performance requirements for and expected life of the secondary application, and finally, the processing

Reprocessing

287

Table 13.2 Problems encountered when reprocessing plastics Problem

Source

Occurrence

Low melt viscosity: difficult to extrude, brittleness, demoulding becomes difficult

Polymer chain shortening

Processing

Thermal degradation during processing

No process stabilisers

Processing

Degradation during processing if temperature too high, unmelted regions during processing if temperature too low

Immiscibility due to different melt rheology and melting point

Processing

Difficult to produce thin-walled parts

Impurities

Processing

Brittleness, high melt shrinkage, low dimensional stability

Polymer chain shortening

Service

Rapid yellowing, low chemical resistance, surface embrittlement, poor electrical properties

Insufficient antioxidants

Service

Accelerated deterioration of physical properties, yellowing

Insufficient heat stabilisers

Service

Poor surface appearance, colour variations

Immiscibility

Service

Poor surface appearance, voids, and embrittlement

Impurities

Service

and compounding technique. The most frequently used revitalisation technologies are blending a limited amount of recyclate into virgin material and restabilisation (Lundquist et al., 2000), which is described below. 13.4.2.1

Restabilisation

Restabilisation, the addition of processing, thermal and light stabilisers in the post-used material, aims to minimise degradation effects during recycling and subsequent service. The objective is to limit the degradation of the polymer chain and side reactions, such as cross-linking, in order to maintain the molecular weight and rheological behaviour of the material, while also improving the optical and mechanical properties (Kartalis et al., 2000). To achieve restabilisation of a material, it is first necessary to establish the content and nature of any residual short and long-term stabilisers present in the material, as well as the effects of processing and use on these stabilisers. The type and quantity of stabiliser will then be added depending on the application requirements. Stabilisers, however, increase the cost of the material

288

Green composites

and as such it is usual to use regenerative or slowly consumed stabilisers to minimise costs. 13.4.2.2

Other additives

As seen in Table 13.2, a frequent problem in repeated recycling is chain scission, and several additives have been developed as remedies for this. Some have been developed to adjust the melt viscosity to a desired level by vis-breaking the polymer chain. Vis-breaking, the controlled degradation of a polymer to produce a narrower molecular weight distribution, improves the melt flow properties. Others have been developed to increase the tensile strength and elongation of recycled grades by cross-linking without creating gels. This method also allows the grafting of monomers to the polymer backbone during reactive processing (Lundquist et al., 2000).

13.4.3

Recycling of mixed plastic waste

A significant fraction of plastics and products cannot be identified and recovered in a cost-effective way as individual polymers. Recycling of commingled PCW presents a number of additional problems in melt processing such as contamination, materials with different melt indexes, varying additive and colourant levels and otherwise immiscible materials, all of which can lead to almost impossible processing conditions. For example, the presence of material with high melt index in commingled feedstock would require mixing at high shear rates and high temperatures. Any heat-sensitive polymers would be degraded under such conditions. For this reason, when extruding PCW there is a need to include as high a percentage of polyolefinic waste as possible in order to achieve any practical material use (Jansen et al., 2001), while excluding some other plastic wastes such as PVC. These problems however, are becoming more complex with the introduction of complex co-injection processing routes. In this type of processing, mouldings are manufactured in a single shot from a number of different plastics or from the same plastic but in different colours (Tomlins, 1999; Czerski and Griffiths, 2002). Ultimately, the quality of the end product, its end use and ability to substitute virgin material, is dependent on the quality of the feedstock. ‘High-quality’ recyclates (in-house or homogeneous waste) can be reused in the same applications, whereas ‘low-quality’ recyclates (PCW) are ‘down-cycled’ into products usually made from other material (Belofsky, 1995). This is also coupled with the fact that the infinite recycling of plastic is technically impossible since the molecules degrade during repeated processing placing a finite lifetime on plastics (Fletcher and Mackay, 1996). A variety of products can be made from mixed plastics using an extrusion moulding technique. This uses an extruder to force material into a large simple mould at much

Reprocessing

289

slower rates than would be found with injection moulding. This technique is generally limited to materials with high levels of polyethylene and polypropylene, and to relatively low specification products such as pallets, fence posts, outdoor furniture and decking. The advantages of this ‘Plaswood’ over its timber and concrete counterparts are clear; it does not rot or splinter and is near zero maintenance, unlike wood. It is very tough, light and there is no on-site wastage, unlike concrete where wastage can often be as high as 20% (Cooper, 1997; Mason, 2001). 13.4.3.1

Blending

Blending and compatibilisation are traditional methods for modifying the properties of virgin plastics to obtain specific properties and these methods can also be applied to recycling. There are two main categories of polymer blends: miscible and immiscible. Miscible blends are homogeneous to the molecular level and properties scale with the ratios of the constituent polymers. Immiscible blends are frequently heterogeneous and their mechanical performance is often inferior to that of individual constituents in terms of impact strength, elongation at break and resistance to stress cracking. Immiscible blends, however, can often be compatibilised by several means, of which the addition of graft and block copolymers in small quantities is most relevant. These compatibilisers are used to strengthen the physical and chemical bonding between the different phases or to decrease the surface energy of the interface. The general effects of a compatibiliser are increased impact strength, elongation at break and stress cracking resistance, with decreased stiffness, hardness and heat resistance. However, immiscibility is not a phenomenon exclusive to heterogeneous mixtures of different plastics; plastics of the same base polymer may be immiscible with each other, depending on the molecular structure and size as well as the presence of additives. One significant disadvantage of the use of compatibilisers is the relatively high cost. 13.4.3.2

Pulverisation

The pulverisation of plastics gives advantageous properties as well as a uniform coloured powder. It has been found that the pulverised particles have a unique elongated and smooth shape attributed to the high shear conditions (Tatemichi and Tomizawa, 1997). The particles also display mechanical and physical properties comparable or better than the properties resulting from conventional processing of PCW (Schoeke et al., 1999). Another process for plastic recycling is via solid-state shear pulverisation (S3P). In this process unsorted commingled PCW is subjected to high shear and pressure in the solid state. By rapidly removing frictional heat, the

290

Green composites

polymer mix is transformed to a uniform powder of controlled variable particle size suitable for processing by all conventional fabrication techniques (Nesarikar et al., 1997; Bilgili et al., 2001a, b). The process was developed by Northwestern University in the USA, after the basic principle of disintegration of plastics, called elastic strain-assisted grinding, was developed at the Academy of Science in Moscow, Russia. For pulverisation to occur, the plastics are processed below their melting points, in the case of crystalline or semi crystalline polymers, or glass transition temperatures in the case of amorphous polymers. By controlling the temperature and process parameters, optimal fragmentation of the feedstock occurs, resulting in a powder production of a given size range. S 3P is governed by mechanochemistry, where the use of mechanical energy causes a chemical reaction in the commingled polymers. When sufficient mechanical force is applied, chain scission occurs and macroradicals are formed, which then act as initiators for further reactions, i.e. the creation of block or graft copolymers as in situ compatibilising agents. Powder formation occurs due to the high shear and pressure that leads to fragmentation of the feedstock. 13.4.3.3

Composite production

The production of composite materials from the waste stream allows for the production of recycled materials by mechanical means, thereby keeping reprocessing costs relatively low while maintaining or improving the materials properties. Composites from recycled materials are commonly made up of either blended waste, waste filler or fibres in a cement, thermosetting resin or thermoplastic matrix. There are many wood fibre or sawdust/resin composites in use (Breslin et al., 1998; Chow et al., 1998; Deaner et al., 1996; Xie et al., 1997), as well as paper/resin composites and foam/cement composites (Boser et al., 2002). Plastic/resin, textile/resin and slate/resin composites are also being developed. One such composite, ‘Plasmega’, transforms mixed plastic and residual waste such as ash into a construction material. The product is manufactured entirely from mixed waste plastic in combination with many other waste materials such as incinerator ash, pulverised fly ash, foundry sand, contaminated soil and many forms of waste matter. The Plasmega process involves a system of regulated high rate size reduction, combined with controlled parameters of low temperature mixing and blending that alleviate noxious emissions. The process requires no sorting, separation or cleaning of mixed waste products, and transforms contaminated wastes into an inert matrix that is lighter but much stronger than concrete, with near zero porosity. The manufacture of products from Plasmega requires no more than standard technology, as Plasmega can be extruded, intruded or simply moulded to virtually any desired shape. The product can be manufactured for use as sea

Reprocessing

291

wall defences, drains, trench fill, concrete and timber replacement and as a scalping base for road surfacing, along with many others at a rate of 20 tonnes per hour. Studies of synthetic lightweight aggregate (SLA) have shown that mixed plastic waste may be used as binder material for other wastes in SLA as replacement for typical inorganic aggregates (Malloy, 2001). The studies highlighted the need for high temperatures in order for all plastic components to melt on production of the binder and the usual intermediate stages of decontamination, etc. are less important in this application. Similarly, studies on the substitution of Portland cement by polyester resins to produce a polyester concrete (PC) material have been performed (Rebeiz and Craft, 1993; Rebeiz and Fowler, 1996). Here, a polyester resin (10%) was supplemented with initiators and promoters, gravel (45%), sand (30%) and fly ash (15%) to produce a PC material. A further use of waste plastics is covered within the patented Strumat technique, which involves incorporating waste plastics into a polyester resin. The key factor with this is that the curing of the polyester generates sufficient heat partially to melt the waste plastic particles and provide better bonding. The waste material used may come from industrial or domestic sources, and needs to be pulverised to a particle size of less than 10 mm (typical particle size is 1–3 mm). One advantage of this method is that it is particularly suitable for incorporating waste PVC that is normally hard to deal with.

13.4.4

Chemical recycling

Chemical recycling is a relatively new route for converting plastic waste by returning it back to its original constituents, that is, monomers or petrochemical feedstock (Lundquist et al., 2000; Belofsky, 1995). Chemical recycling can be split into chemical processes, where the polymers are broken down to constituent chemicals similar to the initial monomers and thermal processes, where the polymers are broken down to fuels and simple hydrocarbons (Milne et al., 1999). Chemical recycling or solvolytic processes are essentially depolymerisation processes involving reactions with water, alcohol or other solvents. Condensation polymer scrap such as PET and nylon (Takuma et al., 2000) is heated in presence of a liquid and the polmerisation process is reversed. The material most widely treated in this way is PET, as mechanical recycling is very sensitive to impurities. The three major processes for the depolymerisation of PET are hydrolysis, methanolysis and glycolysis. PET may be hydrolysed by treatment with water, acids or caustic soda to give terephthalic acid (TA) and ethylene glycol, requiring purification before re-use (Masuda et al., 2001). Methanolysis involves the reatment of PET with methanol, under pressure at around 200 ∞C and results in depolymerisation giving dimethyl terephthalate

292

Green composites

(DMT) and ethylene glycol (EG). The DMT is then purified by distillation and crystallisation to give a high quality intermediate, which may be used to make new PET. Once refined, the EG may be used for a variety of applications, including antifreeze and PET production (Belofsky, 1995). Glycolysis is the reaction of recovered PET with excess ethylene glycol under pressure at about 200 ∞C. It reverses the polymerisation reaction to give bishydroxyethylterephthalate (BHET) and short chain polymers of just a few repeat units (oligomers). The BHET formed is purified by melt filtration under pressure to remove physical impurities and treated with carbon to remove chemical impurities (Rebeiz and Craft, 1993; Belofsky, 1995). However, such processes are also sensitive to impurities. Addition polymers cannot be recycled this way and must be carefully extracted beforehand (Masuda et al., 2001). The above processes are best suited to easily defined large volume sources of relatively pure material but they are relatively capital intensive with slow production rates (Czernik et al., 1998) and only commercially viable where comparatively large throughputs are possible.

13.4.5

Recycling of composites

Polymer composites present additional problems for recycling, due to the combination of large amounts of fillers or fibres and the widespread use of thermosetting matrices that cannot be melt processed. Possible recycling options are shown in Fig. 13.14. One of these options is simply to grind the composite into granules or powder that can then be incorporated into a thermosetting or cement-based matrix to produce a further composite material. Such materials find applications in the construction industry. Studies have shown that, if the composites are ground to a small particle size of less than All fibre composites

Granulation / pulverisation

Use as filler

Short fibre thermoplastics

Granulation and remoulding

Re-use as composite

Long fibre thermoplastics

Dissolution / separation

Re-use of fibres and polymer

Fibre thermosets

Thermal processing

Re-use of fibres fuel/energy

13.14 The options available for recycling waste composite materials.

Reprocessing

293

20 mm, they prove to be very efficient fillers for high strength mortars, concretes and even paint films (Kojima and Furukawa, 1997). 13.4.5.1

Recycling of short fibre thermoplastic composites

Short fibres are often incorporated into thermoplastics to improve strength and stiffness, whilst retaining the ability to use conventional processing such as extrusion or injection moulding. As such, they are then amenable to general mechanical recycling techniques. The materials can be granulated and then remoulded (Henshaw et al., 1996). It has been found that the fibres are not significantly reduced in length during recycling but that a key issue was the bonding between the glass fibres and the polymer. This can be improved by incorporating recycled material into virgin material, or by using coupling agents to improve the bond (VanLochem et al., 1996; Chu and Sullivan, 1996). Further issues with this method include faster wear of processing equipment and slightly greater energy requirements. 13.4.5.2

Recovery of fibres by thermal processing

With either long fibre composites and/or with thermosetting matrices, mechanical recycling is impossible apart from pulverisation to a powder. In these cases it is possible to recover some of the material by various means. Thermal processing is possible and utilises the different temperature resistance of fibres and matrix. The thermal processing may involve simple combustion of the polymer, with energy generation, leaving glass or carbon fibres relatively intact. The heat treatment does affect the properties of the fibres (Kennerley et al., 1997) and lower temperatures of heat treatment are required to maximise the strength of recovered fibres. Pyrolysis, with insufficient oxygen for combustion can break the polymer matrix down to smaller organic molecules that can be used as feedstock or fuels. Again, this leaves the fibres relatively intact and these can then be reused for further composite production (Kouparitsas et al., 2002; DeMarco et al., 1997). This method has been successfully applied to both carbon fibre composites and glass fibre reinforced sheet moulding compounds (DeMarco et al., 1997), where the liquid fraction produced from the polymer provides the energy for the pyrolysis plant. 13.4.5.3

Recycling by dissolution

This is the only method that has been established to separate and re-use both components of the composite. It involves the dissolution of the polymer matrix into a suitable solvent, often at elevated temperatures, followed by filtration of the fibres and precipitation of the polymer. This has been

294

Green composites

shown to be an effective method for certain types of composites, including those with carbon and aramid fibres, which would be adversely affected by higher temperatures (Ramakrishna et al., 1998). There are, however, two distinct drawbacks to this method; first the method is limited to thermoplastic materials as thermosetting materials are virtually insoluble. Also, handling of the large amounts of solvents can cause health, safety and environmental concerns.

13.5

Future trends

In the current economic climate increases in recycling rates for plastics and composite will not occur simply by market forces. Recycling tends to be economically viable only where relatively clean single-source material is available. In other cases, the costs of sorting and treatment, or reductions in material quality do not allow recycled material to compete with virgin material. In order to increase recycling rates, some form of legislation or market intervention is required. The European Union has been very active in drawing up legislation with the aim of increasing recycling. The first major area to come under such legislation was the packaging sector, where recycling targets have been agreed by each member state. These targets have increased gradually over recent years. The UK’s method of ensuring these targets are met is via the ‘packaging recovery notice’ scheme, which effectively forces packaging producers and users to subsidise the cost of recycling. In the near future, similar legislation will come into force for the electrical and electronic industry and for end-of-life vehicles. Both of these areas pose greater challenges, as the materials combinations used are more complex, the lifetimes of products are much longer and there are more potentially hazardous materials to deal with. As an example of the problems that need to be addressed to comply with such legislation, the issue of flame retarded plastics in the electronic sector is significant. Certain brominated flame retardants have been identified as hazardous and, if components contain these, they cannot be recycled and must be treated separately. It is estimated that currently between 20 and 30% of waste plastics from IT equipment contains such additives (Brennan et al., 2002). At present, there is no completely accurate method of identifying these additives, and so the regulations would dictate that all such components must be assumed to be hazardous and recycling would be impossible. The consequences of driving recycling rates higher and targeting sectors where the materials combinations are more complex will mean that more of the material we need to recycle will pose challenges for separation, sorting and cleaning. This may well mean that we will need to find more applications for mixed plastic waste. In many cases incorporating them into a binder to produce a relatively low-grade composite is seen as a viable

Reprocessing

295

route. The problem with this is that there are limited markets for such materials and many other waste materials are seeking to exploit this area as well. It may be that certain methods of processing mixed plastics for use in more demanding applications will become more attractive, such as solid state shear pulverisation. There may also be an increase in the use of chemical recycling methods. In addition, the concept of ‘design for recyclability’ is becoming much more widespread, especially where legislation is impending, such as the automotive industry. In such cases there is a trend towards the use of fewer different materials, short fibre composites rather than continuous ones, thermoplastic matrices rather than thermosets and thermoplastic fibres. Of especial interest is the growing interest in all-polymer and natural fibrebased composites. The all-polymer composites produced from woven polymer fibres or tapes (often PP or PE) with a thermoplastic matrix allow relatively easy recycling as the entire composite can be melt processed. Such materials are discussed in more detail elsewhere in this volume. There is a wide range of natural fibres that can be used for composites and it could be argued that these should produce ‘greener’ materials; however some caution needs to be raised. Certainly, if useful composites can be produced from natural fibres and a natural matrix, where the entire composite is biodegradable, then it would certainly justify such a label. The combination of natural fibres in a thermosetting matrix, however, poses as many difficulties for recycling as a conventional composite and the benefits of a natural material are often lost if intensive processing is required to convert it to a useful material.

Sources of further information In addition to the specific references listed below, there are several important sources of current information within this area. Details of the latest legislation relating to waste plastics is available from the European Union, UK Government or USA Government Departments. Their websites have full versions of the relevant legislation as well as summaries and implications. A significant amount of information is available via the trade organisations relating to plastics usage. The most useful one specifically relating to plastics is the Association of Plastics Manufacturers in Europe (APME), which has large amounts of information either directly on their website or as available reports. The British Plastics Federation is a good source of further information relating to the position in the UK. The automotive industry is also keen to promote their recycling activities and many automotive companies have information available. In addition to these industry-specific sources, there are several organisations dedicated to increasing the amount of recycling activity. WRAP is a

296

Green composites

government-funded body looking at best practice for recycling and has considerable information available. Recoup is the primary organisation dealing with plastic bottle recycling in the UK and is another good source of information. Finally, the trade publications Recycling World and Materials Recycling Weekly are good sources of information on latest trends and developments.

References APME (1998). Assessing the Potential for Post-use Plastics Waste Recycling – Predicting Recovery in 2001 and 2006. Brussels: Association of Plastic Manufacturers in Europe. APME (1999). Plastics A Material of Choice for Packaging: Insight into Consumption and Recovery in Western Europe. Brussels: Association of Plastic Manufacturers in Europe. APME (2000). 2000 Annual Report. Brussels: Association of Plastic Manufacturers in Europe. APME (2001a). Consumption Data by Plastics Type. Brussels: Association of Plastic Manufacturers in Europe. APME (2001b). Plastics: An Analysis of Plastics Consumption and Recovery in Western Europe 1999. Brussels: Association of Plastic Manufacturers in Europe. Arnold, J.C. and Maund, B. (1999a). The properties of recycled PVC bottle compounds: 1-mechanical performance. Polymer Engineering and Science, 39, 1234–41. Arnold, J.C. and Maund, B. (1999b). The properties of recycled PVC bottle compounds: 2-reprocessing stability. Polymer Engineering and Science, 39, 1242–50. Belofsky, H. (1995). Plastics: Product Design and Process Engineering. Munich: Hanser/ Gardner. Bickerstaffe, J. (1996). Environmental Impact of Packaging in the UK Food Supply System. London: The Industry Council for Packaging and the Environment (INCPEN). Bilgili, E., Arastoopour, H. and Bernstein, B. (2001a). Pulverisation of rubber granulates using the solid-state shear extrusion (SSSE) process: Part I. Process concepts and characteristics. Powder Technology, 115, 265–76. Bilgili, E., Arastoopour, H. and Bernstein, B. (2001b). Pulverisation of rubber granulates using the solid-state shear extrusion process Part II. Powder characterisation. Powder Technology, 115, 277–89. Boser, R., Ragsdale, T. and Duval, C. (2002). Recycled foam and cement composites in insulating concrete forms. Journal of Industrial Technology, 18, 23–27. Brennan, L.B., Isaac, D.H. and Arnold, J.C. (2002). Recycling of ABS and HIPS from waste computer equipment. Journal of Applied Polymer Science, 86, 572–8. Breslin, V.T., Senturk, U. and Berndt, C.C. (1998). Long-term engineering properties of recycled plastic lumber used in pier construction. Resources, Conservation and Recycling, 23, 243–58. Bucklow, I. (2000). Plastic proves it can hold its beer. Materials World the Journal of the Institute of Materials, 8, 14–16. Chow, P., Bajwa, D.S., Lu, W.D. et al. (1998). Injection-molded composites from Kenaf and recycled plastic. Proceedings of 1st Annual American Kenaf Society Meeting, San Antonio, TX, February 1998. Chu, J. and Sullivan, J.L. (1996). Recyclability of a glass fiber PBT composite. Polymer Composition, 17, 523–31.

Reprocessing

297

Cooper, J. (1997). Plastic transformations. Materials Recycling Week, 170, 9–10. Cooper, J. (1998). Fulfilling a niche for PET. Materials Recycling Week, 171, 11–12. Czernik, S., Elam, C.C., Evans, R.J., Meglen, R.R., Moens, L. and Tatsumoto, K. (1998). Catalytic pyrolysis of nylon-6 to recover caprolactam. Journal of Analytical and Applied Pyrolysis, 46, 51–64. Czerski, J. and Griffiths, A. (2002). Show–stoppers: pick of the plastics from K and Euromold. Materials World the Journal of the Institute of Materials, 10, 32–4. Deaner, M.J., Puppin, G. and Heikkila, K.E. (1996). Advanced polymer/wood composite structural member. Journal of Cleaner Products, 4, 263–7. DeMarco, I., Legarretta, J.A., Laregsgoiti, M.F. et al. (1997). Recycling of the products obtained in the pyrolysis of fibre-glass SMC, Journal of Chemical Technology and Biotechnology, 69, 187–92. Drelich, J., Kim, J.H., Payne, T., Miller, J.D. and Kobler, R.W. (1999). Purification of polyethylene terephthalate from polyvinyl chloride by froth floatation for the plastics (soft-drink bottle) recycling industry. Separation and Purification Technology, 15, 9– 17. Dvorak, R. (2001). Development of a continuous thermal separation system for the removal of PVC contamination in post-consumer PET flake. ANTEC 2001, Dallas, Texas: Society of Plastic Engineers. Environment Agency (2001). Plastics in the Environment. Bristol: Environment Agency, Bristol. EPA (2000). Municipal Solid Waste in the US 2000, Washington DC: Environmental Protection Agency. Ferrara, G. and Malloy, T.P. (1999). Low dense media process: a new process for lowdensity solid separation. Powder Technology, 103, 151–5. Ferrara, G., Bevilacqua, P., DeLorenzi, L. and Zanin, M. (2000). The influence of particle shape on the dynamic dense medium separation of plastics. International Journal of Mineral Processes, 59, 225–35. Fletcher, B.L. and Mackay, M.E. (1996). A model of plastics recycling: does recycling reduce the amount of waste?. Resources, Conservation and Recycling, 17, 141–51. Henshaw, J.M., Han, W.J. and Owens, A.D. (1996). An overview of recycling issues for composite materials. Journal of Thermoplastic Composite Materials, 9, 4–20. Jansen, D.C., Kiggins, M.L., Swan, C.W., et al. (2001). Lightweight fly-ash/plastic aggregates in concrete. Proceedings of Concrete 2001 – Materials and Construction, Washington: Transport Research Board. Kartalis, C.N., Papaspyrides, C.D. and Pfaendner, R. (2000). Recycling of post-used PE packaging film using the restabilisation technique. Polymer Degradation and Stability, 70, 189–97. Kennerley, J.R., Fenwick, N.J., Pickering, S.J. and Rudd, C.D. (1997). The properties of glass fibers from the thermal processing of scrap thermoset composites. Journal of Vinyl Additive Technology, 3, 58–63. Kojima, A. and Furukawa, S. (1997). Recycling of resin matrix composite materials 7. Advances in Composite Materials, 6, 215–25. Kosier, E., Dvorak, R., Iovenitti, P. and Masood, S. (1999). Enhanced automatic sorting of post-consumer bottles. ANTEC 1999 Conference Proceedings, New York: Society of Plastic Engineers. Kouparitsas, C.E., Kartalis, C.N., Varelidis, P.C., Tsenoglou, C.J. and Papaspyrides, C.D. (2002). Recycling of the fibrous fraction of reinforced thermoset composites. Polymer Composites, 23, 682–9.

298

Green composites

Lundquist, L., Leterrier, Y., Sunderland, P. and Manson, J.E. (2000). Life Cycle Engineering of Plastics Technology, Economy and the Environment, Oxford, UK: Elsevier. McDonald, S. and Ball, R. (1998). Public participation in plastics recycling schemes. Resources, Conservation and Recycling, 22, 123–41. Malloy, R. (2001). High carbon fly ash/mixed thermoplastic aggregate for use in lightweight concrete. In ANTEC 2001 Conference Proceedings, Dallas, Texas, May 6–10. Society of Plastic Engineers. Mason, S. (2001). Plaswood plans to grow sales in the garden. Materials Recycling Week, 178, 8. Masuda, T., Kushino, T., Matsuda, T., Mukai, S.R., Hashimoto, K. and Yoshida, S. (2001). Chemical recycling of mixture of waste plastics using a new reactor system with stirred heat medium particles in steam atmosphere. Chemical Engineering Journal, 82, 173–81. Milne, B.J., Behie, L.A. and Berruti, F. (1999). Recycling of waste plastics by ultrapyrolysis using an internally circulating fluidized bed reactor. Journal of Analytical and Applied Pyrolysis. 51, 157–66. MRW (1998). Easing the separation of plastic. Materials Recycling Week, 171, 13. MRW (2001). Technology reduces risk of contaminated recyclate. Materials Recycling Week, 178, 15–17. Nesarikar, A.R., Carr, S.H., Khait, K. and Mirabella, F.M. (1997). Self compatibilisation of polymer blends via novel solid-state shear extrusion pulverisation. Journal of Applied Polymer Physics, 63, 1179–87. Patel, M., Von Thienen, N., Jochem, E. and Worrell, E. (2000). Recycling of plastics in Germany. Resources, Conservation and Recycling, 29, 65–90. Ramakrishna, S., Tan, W.K., Teoh, S.H. and Lai, M.O. (1998). Recycling of carbon fiber PEEK composites. Polymer Blends Polymer Composite, 137, 1–8. Rebeiz, K.S. and Craft, A.P. (1993). Plastic waste management in construction: technological and institutional issues. Resources, Conservation and Recycling, 15, 245–57. Rebeiz, K.S. and Fowler, D.W. (1996). Flexural strength of reinforced polymer concrete made with recycled plastic waste. ACI Structure Journal, 5, 524–30. Recoup (2001). Survey 2000 Report. Peterborough: Recoup. Schoeke, D., Arastoopour, H. and Bernstein, B. (1999). Pulverisation of rubber under high compression and shear. Powder Technology, 102, 207–14. Shen, H., Forssberg, E. and Pugh, R.J. (2001). Selective floatation of plastics by particle control. Resources, Conservation and Recycling, 33, 37–50. Shen, H., Pugh, R.J. and Forssberg, E. (2002). Floatability, selectivity and floatation separation of plastics by using a surfactant. Colloids and Surfaces A: Physiochemical and Engineering Aspects, 196, 63–70. Subramanian, P.M. (2000). Plastics recycling and waste management in the US. Resources, Conservation and Recycling, 28, 253–63. Takuma, K., Uemichi, Y. and Ayame, A. (2000). Product distribution from catalytic degradation of polyethylene over H-gallosilicate. Applied Catalysis A: General, 192, 273–80. Tatemichi, Y. and Tomizawa, T. (1997). Development of recycling method by application of fine pulverisation. Technical Notes/JSAE Review, 18, 79–82. Tomlins, P. (1999). Saving more waste plastics from the scrap heap, Materials World, The Journal of the Institute of Materials, 7, 137–8. VanLochem, J.H., Henriksen C. and Lund, H.H. (1996). Recycling concepts for thermoplastic composites. Journal of Reinforced Plastic Composites, 15, 864–76.

Reprocessing

299

Warmington, A. (2001). Overcoming barriers in sorting mixed polymers for recycling. Materials Recycling Week, 177, 8. Xie, K.Y., Locke, D.C., Habib, D., Judge, M. and Kriss, C. (1997). Environmental chemical impact of recycled plastic timbers used in the Tiffany Street pier, South Bronx, New York. Resources, Conservation and Resources, 21, 199–211.

300

Index

Index

additives 222–6 antimicrobials 225–6 biocides 225–6 colourants 224 fillers 103, 225 flame retardants 224–5, 294 lubricants 223–4 plasticisers 223–4 in recycling 288 shrinkage control 223 aesthetic language 9, 15 agricultural fibres 34 algal cellulose 75–6 alkaline treatments 143–4, 156–7 annual tree rings 83–4 antimicrobial additives 225–6 applications 233–49 contemporary 237–44 historical 235–7 of polyhydroxyalkanoate (PHA) 131 of polylactic acid (PLA) 127–8 of starch-based polymers 134–5 of wood-based composites 234–5, 238–43 aspect ratios 198 autoclaving 218–19 automated sorting equipment 279 automotive products 4, 42, 74, 119–20 end-of-life disposal strategies 226– 30 recycling 4, 263–4 bacteria cellulose gel 75 Bakelite 235 balsa 92 bast fibres 50

batch style equipment 193 bio-fillers 103, 225 biocides 225–6 biocomposites 33–5, 123–4, 135–47 alkaline treatments 143–4, 156–7 biodegradation 145 creep resistance 137, 139 fibre–matrix adhesion 141–2 fibre–matrix modification 143–4 injection moulding 140 interfacial properties 141–3 matrix degradation 139 mechanical properties 136–9, 141 processing 139–41 reinforcement 145–7 sheer stress effects 141 water absorption 136–7, 144–5 biodegradable polymers 123–4, 188–9 biodegradation 101–2, 145 of polyhydroxyalkanoate (PHA) 131 of polylactic acid (PLA) 127 of starch–based polymers 134 of synthetic polymers 261–3 UV light biodegradation 260–1 see also recycling; waste management biofibre production 181–3 retting 181–2 steam explosion 182–3 biomass 49 biopolymer thermoplastics 120 bleaching plant fibres 156 blending and recycling 289 building products 241 bundle forms 21 calendering 111–12

301

302

Index

Canada 3–4 Cara Plastic 246 carbon dioxide 214–15 carbon fibres 68 Cargill Dow 124–5, 128, 246 cashew nut shell liquid resin 159–61, 215–16 Castiglione, Achille 9 cellulose 49, 84, 92, 236 algal cellulose 75–6 bacteria cellulose gel 75 chemical composition 184 crystal structure 54–6 hemicellulose 85, 161, 184 macrofibrils 66–8 mechanical properties 65–8 microfibrils 65–6, 190–2 reconstituting 104 tensile strength 62 Young’s modulus 161–2 chemical composition of fibres 183–7 chemical degradation 259–60 chemical pulp 85, 87 chemical recycling 291–2 chemical treatments 190–1 China reed 42–3 chitin 64 chitosan 64 Chrysler Composite Vehicle 209 clay nanofillers 146–7 clean production 7, 207–31 additives 222–6 autoclaving 218–19 choice of materials 213, 230–1 contract moulding 216–17 DRDF (double rift diaphragm forming) 219–20 economic aspects 208–9 end-of-life disposal strategies 226–30 energy saving 210–15 environmental aspects 208, 215–22 prepegging 218–19 RIFT (resin infusion under flexible tooling) 217–18 RTM (resin transfer moulding) 220–1 social aspects 208 SRIM (structural reaction injection moulding) 220, 221–2 see also processing

climate change levy 210 cold curing 235 collection of waste 277–8 colourants 224 compatibilisers 289 compression moulding 111–12 concrete 291 coniferous trees 83–4, 85, 86 consumption 14–20, 81 contemporary applications 237–44 continuous kneading mixers 193–4 contract moulding 216–17 Control of Substances Hazardous to Health 216 Cord-Aerolite 237 coupling agents 95, 101, 197 CRAFT project 2 creep behaviour 118–19, 137, 139, 198– 201 critical reviews 28–9 crystal modulus 56–65 crystal structure of celluloses 54–6 data collection 29–30 quality 28, 32 validation 30 deciduous trees 83, 85, 86 decking 238, 241 degradation see biodegradation; recycling density sorting equipment 279–80 design 9–22 aesthetic language 9, 15 challenges 14–20 design thinking 9–11 forms 21 life cycle design 37–40 natural material structure 17–20 and obsolete value systems 12–14 principles of development 11 for recyclability 295 visual language 21 destructurisation 133 disperse phase 100–1, 102–3 dissolution recycling 293–4 downcycling 253 DRDF (double rift diaphragm forming) 219–20 drivers for change 3–5

Index durability 197–201 dynamic mechanical thermal analysis (DMTA) 173–6 e-beam irradiation 256 eco-labelling 45 eco-marketing 12 electrostatic separation 281 end-of-life disposal strategies 34–5, 226–30, 254 energy saving 210–15 climate change levy 210 tariffs 211–13 and transportation 210–11 environmental audits 25 environmental footprint 1–3 environmental impact 101–2 environmental product declarations (EPDs) 45 Environmental Products Inc. 262–3 environmental stability 176–7 EU Directives 4 extrusion 108, 194–5, 284–6 felting 109–11 fibre–matrix adhesion 93, 141–2 fibre–matrix interface 156–7 fibre–matrix modification 143–4 fillers 103, 225 film casting 197 finite element analysis (FEA) 119 flake orientation 39–40 flame retardants 224–5, 294 flexural strengths 170 flowforms 21 foaming 196 food packaging 4, 127–8, 274, 275 Ford, Henry 237 forest management 33–4 fracture values 167 friction stress 168 function of composites see structure and function of composites gamma irradiation 256 garage door systems 40 glass fibres 40, 68 glass reinforced plastics (GRPs) 171, 243 GMT materials 2–3

303

Gordon-Aerolite 236 grinding 257–9 hair fibres 50 Health and Safety at Work Act (1974) 216 hemicellulose 85, 161, 184 hemp 40, 154, 164 high performance fibres 189–92 chemical treatments 190–1 mechanical treatments 191 mechano-physical separation of fibres 191-2 nano-biofibrils 190–2 thermal treatments 190 high pressure laminates 235 high shear mixers 194 historical applications 235–7 hollow fibre bundles 161–3 humidity effects 105 hydraulic power 213–14 hydrophilicity of starch-based polymers 134 impact strength 31–2, 198 incineration 266–7 incremental life cycle assessment 44–5 information requirements 43–4 infra-red sorting equipment 280 injection moulding 140, 158, 195, 209, 213 SRIM (structural reaction injection moulding) 220, 221–2 innovation 14–20, 108–14 interfacial properties 141–3 inventory analysis 29–31 ionising radiation 256 ISO 14040 series of standards 25–7 jute 136, 144 kenaf 50–4, 68–73, 154, 164, 172 Kevlar 62 kneading mixers 193–4 kraft pulp 74–5 Kyoto Protocol 3 lactic acid 124–5 lactide monomers 125–6

304

Index

laminates 112, 236–7 high pressure laminates 235 landfill 4, 254, 276–7 length of fibres 86, 87 life cycle assessment 1–3, 23–46, 246 allocation procedures 30–1 of bio-based composites 33–5 category selection 31, 36 characterisation 31–2 classification 31 critical reviews 28–9 data collection 29–30 data quality 28, 32 data validation 30 development of 24–5 and environmental product declarations (EPDs) 45 fields used in 23–4 future trends 43–6 goals 27 impact assessment 31–2 incremental 44–5 information requirements 43–4 interpretation of results 32–3 inventory analysis 29–31 ISO 14040 series of standards 25–7 limitations 33 methodology 24–35 multiple processes 30–1 normalisation 32 peer reviews 29 plant fibre-based composites 40–3 process flow diagram 29 in product development and design 44–5 reference flows 30 scope definition 27–8 software packages 24, 30 system boundaries 28, 30 transparency 28–9 units of study 28 weighted results 32 wood-based composites 35–40 life cycle design 37–40 lignin 81, 85, 184 load factors 213 long fibre composites 166–9, 173–6 long fibre reinforcements 157–8 lubricants 223–4

machinery and energy use 213–14 macrofibrils 66–8 manual sorting of waste 278–9 marketing, eco-marketing 12 matrices 230, 240 degradation 139 fibre–matrix adhesion 93, 141–2 fibre–matrix interface 156–7 fibre–matrix modification 143–4 materials 95–6 polymer resin matrices 164 polymeric matrices 155–6 powdered matrix polymers 112 synthetic fibres in thermoplastic matrix 105–6 thermoplastic 155–61 thermosetting 155–61, 164–5 matrix phase 100–1 maximum energy demand 211 maximum power requirements 211 mechanical properties 118, 163, 165–73, 185–6 of biocomposites 136–9, 141 of cellulose 65–8 flexural strengths 170 fracture values 167 friction stress 168 of lignin 184 of long fibre composites 166–9, 173– 6 of natural fibres 65–8 of polyester resin 171–3 of short fibre composites 169–71 see also optimisation of properties mechanical pulp 85–6, 87 mechanical recycling 283–6 mechanical treatments 191 mechano-physical separation of fibres 191–2 medium density fibreboard (MDF) 239 melting curves 115–16 mica 114 microfibrils 65–6, 189, 190–2 microstructure of natural fibres 49–54 mineral filled composites 108–14 calendering 111–12 compression moulding 111–12 extrusion alignment 108

Index felting 109–11 needle punch 109–11 point bonding of fibre mats 113– 14 pultrusion 109 special textures 108–9 Miscanthus sinensis see China reed montmorillonite (MMT) 146–7 nano-biofibrils 190–2 nanocomposites 146 natural fibres 6, 49–76, 264–6 biofibre production 181–3 chemical composition 183–7 crystal modulus 56–65 durability 197–201 environmental benefits 201–2 environmental stability 176–7 future trends 74–6, 202–3 grouping by origin 183 kenaf 50–4, 68–73 mechanical properties 65–8 microstructure 49–54 performance 197–201 physical characteristics 163 prices 186–7 processing 192–7 as reinforcements 76 stress transfer 161–3 structural aspects 183–7 sustainable polymer composites 68–73 types of 49–50 natural material structure 17–20 natural polymer sources 123–49 biocomposites 123–4, 135–47 future trends 147–8 information sources 148–9 polyhydroxyalkanoate (PHA) 128– 31, 142 polylactic acid (PLA) 124–8 starch-based polymers 132–5 needle punch 109–11 NMT materials 2–3 Novamont 135 novolacs 161 nucleating agents 131 nylon 264 obsolete value systems 12–14

305

optimisation of properties 154–8 dynamic mechanical thermal analysis (DMTA) 173–6 fibre–matrix interface 156–7 physical characteristics 163 reinforcements 157–8 resin precursors 159–61 stress transfer 161–3 thermoplastic matrices 155–61 thermosetting matrices 155–61, 164– 5 oriented strandboard (OSB) 37–40 packaging 4, 127–8, 274, 275 paper and wood fibres 6, 81–98 adhesion between fibre and matrix 93 consumption of paper 81 from coniferous trees 83–4, 85, 86 from deciduous trees 83, 85, 86 length of fibres 86, 87 lignin content 81 matrix materials 95–6 polarity 82 pulp-making processes 85–6 recycled paper 82–3, 86, 87–90 reinforcement theory 90–1 water absorption 93 wood fibre–plastic composites 93–7 see also wood-based composites pectin 185 peer reviews 29 performance of natural fibres 197–201 PET fibres 105 petroleum-derived plastics 49 phenol formaldehyde (PF) bonded OSB 38–9 physical characteristics 163 pine trees 83–4 PLA-jute composites 136 plant fibre-based composites 40–3, 183 Plasmega 290–1 plasticisers 223–4 plastics collection of waste 277–8 consumption statistics 272–3 glass reinforced (GRPs) 171, 243 incineration 266–7 plastic composites 90–1, 93–7 range of types and qualities 274

306

Index

plastics (continued) recycling 82–3, 96, 252–3, 273, 283– 94 sorting and separating 277–82 Totally Degradable Plastic Additive (TDPA™) 263 uses of waste plastic 290–1 waste management 82, 273–7 worldwide production 272 Plaswood 289 PLLA composites 68–73 point bonding of fibre mats 113–14 polarity of paper and wood fibres 82 polyamides 187 polyester concrete (PC) 291 polyester resin 171–3 polyethylene 105, 187 polyhydroxyalkanoate (PHA) 128–31, 142 applications 131 biodegradation 131 nucleating agents 131 processing 131 properties 130 synthesis 129–30 thermal degradation 131 polylactic acid (PLA) 124–8 applications 127–8 biodegradation 127 PLA-jute composites 136 processing 127 properties 126–7 synthesis 124–6 polymer resin matrices 164 polymeric matrices 155–6 polymers 56–7 biodegradable 123–4, 188–9 green thermosetting 159–61 incineration 266–7 recycling 254–5, 273, 292–4 sorting and separating 277–82 sustainable polymer composites 68– 73 see also natural polymer sources polynosic fibre 67–8 polypropylene 100, 106, 119–20, 187, 241 polystyrene 187 polyvinyl alcohol (PVA) 223 polyvinyl chloride (PVC) 241, 288

post-consumer wood 35–7 powdered matrix polymers 112 power factors 211–13 power requirements 211 PP composites 6 prepegging 218–19 preservative treated timber 242 pressing conditions 165 prices 186–7 processing 192–7 batch style equipment 193 biocomposites 139–41 biofibre production 181–3 blending with thermoplastics 192–3 continuous kneading mixers 193–4 extrusion 194–5 film casting 197 flow diagram 29 foaming 196 high shear mixers 194 injection moulding 158, 195, 209, 213 innovations 108–14 kneading type compounding equipment 193 polyhydroxyalkanoate (PHA) 131 polylactic acid (PLA) 127 pulp-making 85–6 starch-based polymers 134 thermoforming 195–6 see also clean production product churn 209 pulp-making processes 85–6 pultrusion 109 pulverisation 289–90 pyrolysis 260, 263, 293 ramie 57–9 reconstituted materials 104–5 recovery of wood materials 35–7 recycling 100–1, 202, 252–68 additives 288 automotive parts 4, 263–4 blending 289 chemical degradation 259–60 chemical recycling 291–2 compatibilisers 289 design for recyclability 295 dissolution recycling 293–4

Index downcycling 253 end-of-life disposal strategies 34–5, 226–30, 254 future trends 294–5 grinding 257–9 ionising radiation 256 mechanical recycling 283–6 nylon 264 paper 82–3, 86, 87–90 plastics 82–3, 96, 252–3, 273, 283– 94 polymers 254–5, 273, 292–4 pulverisation 289–90 pyrolysis 260, 263, 293 quality of end products 288 restabilisation 287–8 revitalisation 286–7 sorting and separating 277–82 synthetic recyclable composites 105– 8 thermal recycling 293 thermoplastic composites 255–6, 293 thermosetting composites 257–60 waste composites 102 see also biodegradation; waste management reference flows 30 reinforcements 76, 90–1, 145–7, 157–8, 239–40 long fibre 157–8 short fibre 157–8 reprocessing see biodegradation; recycling resins 159–61, 164, 222–3 resin infusion under flexible tooling 217–18 resin transfer moulding 220–1 resoles 161 restabilisation 287–8 retting 181–2 revitalisation 286–7 RIFT (resin infusion under flexible tooling) 217–18 RIM (reaction injection moulding) 220, 221–2 RTM (resin transfer moulding) 220–1 scope definition 27–8 selective dissolution 281–2

307

separating and sorting waste 277–82 sheer stress effects 141 short fibre composites 169–71 short fibre reinforcements 157–8 shrinkage control additives 223 silk 64, 76 single polymer fibre–matrix composites 114–19 sisal fibres 157 social aspects of production 208 software packages 24, 30 solid fibre bundles 161–3 sorting and separating waste 277–82 soy oil 246 special textures 108–9 SRIM (structural reaction injection moulding) 220, 221–2 starch-based polymers 132–5 applications 134–5 biodegradation 134 destructurisation 133 hydrophilicity 134 processing 134 properties 133 thermoplastic starch (TPS) 133, 134 steam explosion 182–3 stem fibres 50 stiffness index 158–9 storage modulus 118 strength, impact strength 198 stress transfer 161–3 structural aspects of natural fibres 183– 7 structure and function of composites 101–4 Strumat technique 291 supercritical carbon dioxide 214–15 surfboard manufacture 208 sustainable plastics 49 sustainable polymer composites 68–73 synthesis of polyhydroxyalkanoate (PHA) 129– 30 of polylactic acid (PLA) 124–6 synthetic composites 236 synthetic fibres in thermoplastic matrix 105–6 synthetic lightweight concrete (SLA) 291 synthetic polymers 261–3

308

Index

synthetic recyclable composites 105–8 synthetic thermoplastics 187–8 system boundaries 28, 30 tariffs, on energy 211–13 thermal degradation of polyhydroxyalkanoate (PHA) 131 thermal recycling 293 thermal separation 280–1 thermal treatments 190 thermoforming 195–6 thermoplastic starch (TPS) 133, 134 thermoplastics 5, 7–8, 92, 95–6, 100, 181–203, 192–3 biofibre production 181–3 biopolymer thermoplastics 120 clean production 215–16 formed by natural systems 105 high performance fibres 189–92 matrices 155–61 recycling 255–6, 293 retting 181–2 synthetic 187–8 synthetic fibres in matrix 105–6 viscosity 92 thermosetting composites 5, 7, 8, 257– 60 thermosetting matrices 155–61, 164–5 thermosetting polymers 159–61 Totally Degradable Plastic Additive (TDPA™) 263 toxic materials 208 Trabant 243 transcrystallinity 115 transportation 210–11 tree rings 83–4

utilisation 14–20 UV light biodegradation 260–1 value systems 12–14 viscosity of thermoplastics 92 visual language 21 volatile organic compounds (VOC) 217 waste management 82, 102, 273–7 collection schemes 277–8 incineration 266–7 landfill 4, 254, 276–7 sorting and separating 277–82 see also biodegradation; recycling Waste Strategy 4 water absorption 93, 136–7, 144–5, 265 effects of 105 and weight gain 177 water flowforms 21 waxes 185, 223–4 weave patterns 106 webbing 113 weight gain 177 wood-based composites 35–40, 90–1, 92–7 applications 234–5, 238–43 life cycle design 37–40 oriented strandboard (OSB) 37–40 post-consumer wood 35–7 see also paper and wood fibres wood-fibre plastic composites 93–7 woven fibres 109 weave patterns 106 X-ray spectroscopy 280 yacht manufacture 208