Management, Recycling and Reuse of Waste Composites

Management, recycling and reuse of waste composites Related titles: Wood–polymer composites (ISBN 978-1-84569-272-8) W...

Author: V. Goodship

248 downloads 3111 Views 13MB Size Report

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

DOWNLOAD PDF

Management, recycling and reuse of waste composites

Related titles: Wood–polymer composites (ISBN 978-1-84569-272-8) Wood–polymer composites are becoming more accepted in advanced engineering applications. The latest generation of wood–polymer composites are being used in automotive, civil and marine engineering. Advances in durability, mechanical properties and materials used in their production have allowed a significant increase in their use in outdoor applications such as decking, fencing, utility poles and exterior woodwork on buildings. Wood–polymer composites also benefit from being more sustainable than other, traditional, composites and possess a consistency in quality that cannot be achieved by wood alone. This book provides a comprehensive survey on major new developments in wood–polymer composites and presents current research from leading innovators around the world. Recycling in textiles (ISBN 978-1-85573-952-9) An increasing amount of waste is generated each year from textiles (including carpets and clothing) and their production. For economic and environmental reasons it is necessary that as much as possible of this waste is recycled instead of being disposed of in landfill sites. On average approximately ten million tonnes of textile waste is currently dumped in Europe and America each year. Recycling in textiles is the first book to bring together textile recycling issues, technology, products, processes and applications for all of those in the industry who are now looking for ways to recycle their textile waste. Design and manufacture of textile composites (ISBN 978-1-85573-744-0) The term ‘textile composites’ is often used to describe a rather narrow range of materials, based on three-dimensional reinforcements produced using specialist equipment. The intention in this book is to describe the broad range of polymer composite materials with textile reinforcements, from woven and non-crimp commodity fabrics to 3D textiles. Whilst attention is given to modelling of textile structures, composites manufacturing methods and subsequent component performance, it is substantially a practical book intended to help all those developing new products with textile composites. Details of these and other Woodhead Publishing materials books can be obtained by: • •

visiting our web site at www.woodheadpublishing.com contacting Customer Services (e-mail: [email protected]; fax: +44 (0) 1223 893694; tel.: +44 (0) 1223 891358 ext. 130; address: Woodhead Publishing Limited, Abington Hall, Granta Park, Great Abington, Cambridge CB21 6AH, UK)

If you would like to receive information on forthcoming titles, please send your address details to: Francis Dodds (address, tel. and fax as above; e-mail: francis. [email protected]). Please confirm which subject areas you are interested in.

Management, recycling and reuse of waste composites Edited by Vannessa Goodship

Oxford

Cambridge

New Delhi

Published by Woodhead Publishing Limited, Abington Hall, Granta Park, Great Abington, Cambridge CB21 6AH, UK www.woodheadpublishing.com Woodhead Publishing India Private Limited, G-2, Vardaan House, 7/28 Ansari Road, Daryaganj, New Delhi – 110002, India www.woodheadpublishingindia.com Published in North America by CRC Press LLC, 6000 Broken Sound Parkway, NW, Suite 300, Boca Raton, FL 33487, USA First published 2010, Woodhead Publishing Limited and CRC Press LLC © Woodhead Publishing Limited, 2010 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 Woodhead Publishing Limited. The consent of Woodhead Publishing Limited 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 Limited 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 978-1-84569-462-3 (book) Woodhead Publishing ISBN 978-1-84569-766-2 (e-book) CRC Press ISBN 978-1-4398-0104-8 CRC Press order number: N10005 The publishers’ policy is to use permanent paper from mills that operate a sustainable forestry policy, and which has been manufactured from pulp which is processed using acid-free and elemental chlorine-free practices. Furthermore, the publishers ensure that the text paper and cover board used have met acceptable environmental accreditation standards. Typeset by Toppan Best-set Premedia Limited, Hong Kong Printed by TJ International Limited, Padstow, Cornwall, UK

Contents

Contributor contact details Preface

xiii xvii

Part I Management of waste composites

1

1

3

1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10

An introduction to composites recycling N. Reynolds and M. Pharaoh, University of Warwick, UK Introduction Composite material types Physical properties Current composite market What is ‘suitability for recycling’? Recycling methods Conclusions References Legislation for recycling waste composites R. Stewart, c/o University of Warwick, UK Legislation Brief history of the European Union Environment Action Programme 6 and the shape of future legislation Next steps for waste management The Waste Framework Directive Environmental Permitting Regulations The Landfill Directive Integrated product policy United Kingdom legislation: The Climate Change Act 2008 End-of-Life Vehicle Directive

3 5 8 12 13 17 18 19 20 20 21 26 28 29 30 30 31 32 33 v

vi

Contents

2.11 2.12 2.13

Waste Electric and Electronic Equipment Regulations Conclusions References

34 35 36

3

Waste management R. Stewart, c/o University of Warwick, UK Introduction The Waste Framework Directive Environmental Permitting Regulations End-of-Life Vehicle Directive Waste Electric and Electronic Equipment Regulations Classification and labelling of waste Conclusions References

39

3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8

39 40 44 50 51 55 59 60

Part II Thermal technologies for recycling waste composites

63

4

65

4.1 4.2 4.3 4.4 4.5 4.6 4.7 5 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8

Thermal methods for recycling waste composites S. J. Pickering, University of Nottingham, UK Introduction The fluidised bed recycling process Properties of the recycled fibre Applications for reuse of the fibre recycled from the fluidised bed process Prospects for commercial operation Current research, future trends and sources of further information and advice References Pyrolysis for recycling waste composites M. Blazsó, Hungarian Academy of Sciences, Hungary Introduction Pyrolysis reactions and products of thermoplastics Pyrolysis reactions and products of thermosets Pyrolyser reactors for polymer recycling Pyrolysis of polymer composites Pyrolysis conditions for polymer composites recycling Environmental concern about pyrolysis products of composites Summarizing comments on recycling polymer composites by pyrolysis

65 67 74 83 98 98 100 102 102 104 109 110 112 114 115 116

Contents

vii

5.9 5.10

Acknowledgements References

117 118

6

Catalytic processing of waste polymer composites M. A. Keane, Heriot-Watt University, UK Introduction Waste polymer recycle: motivation Thermal decomposition of waste plastics Catalytic approach to polymer recycling Catalytic treatment of non-halogen containing polymer waste Catalytic treatment of halogenated polymers: focus on polyvinyl chloride Future trends and conclusions Sources of further information and advice References

122

6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 7

7.1 7.2 7.3 7.4 7.5 7.6 7.7 8

8.1 8.2 8.3 8.4 8.5 8.6 8.7

Advanced thermal treatment of composite wastes for energy recovery P. Lettieri, L. Yassin and S. J. R. Simons, University College London, UK Introduction Introduction to waste management Techno-economic analysis of energy from waste advanced thermal processes Sample calculations Conclusions Notation References Fluidized bed pyrolysis of waste polymer composites for oil and gas recovery W. Kaminsky, University of Hamburg, Germany Introduction Pyrolysis Fluidized bed pyrolysis Polymer composite materials Gas and oil recovery Possibilities and limits: future trends References

122 123 127 127 139 141 143 143 144

152

152 156 171 183 187 188 189

192 192 193 194 200 202 209 212

viii

Contents

Part III Mechanical technologies for recycling waste composites 9

9.1 9.2 9.3 9.4 9.5

10

10.1 10.2 10.3 10.4 10.5 10.6 10.7

11

11.1 11.2 11.3 11.4 11.5 11.6 11.7 11.8 11.9 11.10

Mechanical methods for recycling waste composites K. Makenji, University of Warwick, UK Introduction and background Identification of waste plastic materials and fillers Waste preparation, sorting and readiness for end application use Conclusions References

Additives to upgrade mechanically recycled plastic composites R. Pfaendner, Ciba Lampertheim GmbH, Germany Introduction Properties of recycled plastics and recycled composites Additives to upgrade recycled plastics Specific examples of additives for recycled plastic composites Conclusions and future trends List of additives mentioned in Sections 10.3 and 10.4 References

Improving the mechanical recycling and reuse of mixed plastics and polymer composites K. Tarverdi, Brunel University, UK Introduction Thermoplastic and thermosetting polymers Polymer composites Materials recycling Consumer protection The powder impression moulding process Future technologies for converting mixed waste plastic (composites) into products Case studies: recycling archives Sources of further information and advice References

215

217 217 224 225 246 250

253

253 254 258 267 273 273 276

281 281 282 284 286 290 292 296 297 300 301

Contents 12

12.1 12.2 12.3 12.4 12.5

Quality and durability of recycled composite materials K. L. Pickering, University of Waikato, New Zealand, and M. D. H. Beg, University of Malaysia Pahang, Malaysia Introduction Recycling thermoset matrix composites Recycling thermoplastic matrix composites Conclusions References

Part IV Improving sustainable manufacture of composites 13

13.1 13.2 13.3 13.4 13.5 13.6 13.7 13.8 13.9 13.10 14

14.1 14.2 14.3 14.4 14.5 14.6 14.7

Clean and environmentally friendly wet-filament winding N. Shotton-Gale, D. Harris, S. D. Pandita, M. A. Paget, J. A. Allen and G. F. Fernando, University of Birmingham, UK Introduction Resin impregnation modelling The application of selected impregnation models to clean filament winding Clean filament winding: resin impregnation unit Experimental Results and discussion Conclusions Acknowledgements Notation References Process monitoring and damage detection using optical fibre sensors D. Harris, V. R. Machavaram and G. F. Fernando, University of Birmingham, UK Introduction to optical fibres Introduction to chemical process monitoring Hybrid fibre sensors (small-diameter optical fibres) Damage detection using self-sensing composites Conventional optical fibre sensors Sensing strategies using conventional optical fibres Sensors for monitoring strain and temperature

ix

303

303 304 312 323 324

329

331

331 341 349 352 355 357 363 364 364 365

369

369 374 384 385 387 387 407

x

Contents

14.8

Applications of fibre Bragg grating and extrinsic fibre Fabry–Perot interferometric sensors in composites Multi-measurand sensor design Conclusions Acknowledgements References

14.9 14.10 14.11 14.12 15

15.1 15.2 15.3 15.4 15.5 15.6 15.7 15.8 15.9 15.10 15.11 15.12

New developments in producing more functional and sustainable composites G. F. Smith, University of Warwick, UK Introduction Glass fibre composites Carbon fibre composites Natural fibres Multi-layer, multi-functional composites Sustainability Biomimetics Self-reinforced composites Nanoparticulate composites Hybrid structures Conclusions References

Part V Case studies 16

16.1 16.2 16.3 16.4 16.5 16.6 17

17.1 17.2

Designing composite wind turbine blades for disposal, recycling or reuse N. Papadakis, Technological Educational Institution (TEI) of Crete, Greece, C. Ramírez, Centro de Ingeniería Avanzada en Turbomáquinas S.de R.L. de C.V., México and N. Reynolds, University of Warwick, UK Current wind energy market and trends Turbine design and manufacture Usage End of life Conclusions References In-process composite recycling in the aerospace industry K. Potter and C. Ward, University of Bristol, UK Introduction Composite consumption in the aerospace industry

409 414 415 417 417

425 425 426 428 429 433 434 436 436 436 437 438 438

441 443

443 444 451 453 456 456

458 458 459

Contents 17.3 17.4 17.5 17.6 17.7 17.8 17.9 17.10 17.11

18

18.1 18.2 18.3 18.4 18.5 18.6 18.7 18.8 18.9 18.10

19

19.1 19.2 19.3 19.4 19.5 19.6 19.7 19.8 19.9

Scrap in the aerospace composites industry Composite design and manufacture and their influence on scrap generation Composite design choices and their effect on manufacture Further composite design effects Uncured scrap material and reuse potential Future trends in composites manufacture; impacts on waste generation Conclusions Sources of further information and advice References

Disposal of composite boats and other marine composites M. M. Singh, J. Summerscales, University of Plymouth, UK and K. Wittamore, Triskel Consultants Limited, UK Introduction Market size The design phase The manufacture and marketing phase The use phase End of life Vive la différence? Conclusions Acknowledgements References

Sustainable fibre-reinforced polymer composites in construction M. Fan, Brunel University, UK Basic concept and history Polymer composites in building construction Composites in bridge construction Composites in other constructions Performance in use Construction wastes, reclaim and recycling New development and challenge of construction composites Acknowledgements References

xi 459 464 468 473 477 483 487 490 490

495

495 496 499 500 500 501 514 514 515 515

520 520 525 535 541 543 549 556 561 561

xii

Contents

20

Recycling of concrete P. Purnell, University of Leeds, UK and A. Dunster, BRE, UK Introduction Concrete as a composite Sustainability: incentives for recycling Future trends and drivers Recycling of concrete Recycling of concrete from pre-cast operations End uses (recycled concrete aggregate in readymixed concrete) End uses of recycled concrete aggregate in other construction applications Overall view Sources of further information and advice References

569

Index

593

20.1 20.2 20.3 20.4 20.5 20.6 20.7 20.8 20.9 20.10 20.11

569 570 575 576 577 583 584 586 589 589 589

Contributor contact details

(* = main contact)

Editor

Chapter 4

Dr Vannessa Goodship WMG Department of Engineering University of Warwick Coventry CV4 7AL UK Email: [email protected]

Dr Stephen Pickering Department of Mechanical, Materials and Manufacturing Engineering University of Nottingham University Park Nottingham NG7 2RD UK Email: Stephen.pickering@ nottingham.ac.uk

Chapter 1 Neil Reynolds* and Dr Mark Pharaoh WMG Department of Engineering University of Warwick Coventry CV4 7AL UK Email: Neil.reynolds@warwick. ac.uk [email protected]

Chapter 5 Dr Marianne Blazsó Institute of Materials and Environmental Chemistry Chemical Research Center Hungarian Academy of Sciences Pusztaszeri út 59-67 1025 Budapest Hungary Email: [email protected]

Chapters 2 and 3 Dr Rebecca Stewart, Ph.D, C.Phys. M.Inst.P c/o V. Goodship IARC School of Engineering University of Warwick Coventry CV4 7AL UK Email: [email protected]

Chapter 6 Professor Mark A. Keane Chemical Engineering School of Engineering & Physical Sciences Heriot-Watt University Edinburgh EH14 4AS UK Email: [email protected] xiii

xiv


Chapter 7

Chapter 10

Dr Paola Lettieri,* Dr Liban Yassin and Professor Stefaan J. R. Simons Centre for CO2 Technology Department of Chemical Engineering University College London Torrington Place London WC1E 7JE UK Email: [email protected]

Dr Rudolf Pfaendner Ciba Lampertheim GmbH DE-68623 Lampertheim Germany Email: [email protected]

Chapter 8 Professor W. Kaminsky Institute of Technical and Macromolecular Chemistry University of Hamburg Bundesstraβe 45 20146 Hamburg Germany Email: [email protected]

Chapter 9 Kylash Makenji, M.Sc. WMG Department of Engineering University of Warwick Gibbet Hill Road Coventry CV4 7AL UK Email: [email protected]

Chapter 11 Dr Karnik Tarverdi Reader/Director of Extrusion Technology Wolfson Centre for Materials Processing Brunel University Kingston Lane Uxbridge UB8 3PH UK Email: Karnik.Tarverdi@brunel. ac.uk

Chapter 12 Dr Kim L. Pickering* Department of Engineering University of Waikato Private Bag 3105 Hamilton New Zealand Email: [email protected] Dr Mohammad Dalour Hossen Beg Faculty of Chemical and Natural Resources Engineering University of Malaysia Pahang Lebuhraya Tun Razak 26300 Kuantan, Pahang Darul Makmur Malaysia Email: [email protected]


xv

Chapter 13

Chapter 16

Nick Shotton-Gale, Dee Harris, Surya D. Pandita, Mark Paget, John A. Allen, and Professor Gerard F. Fernando* School of Metallurgy and Materials University of Birmingham Edgbaston Birmingham B15 2TT UK Email: [email protected]

Dr Nikolaos Papadakis Wind Energy and Power Synthesis Laboratory Technological Educational Institution (TEI) of Crete Estavromenos Heraklion Crete GR 71004 Greece Email: [email protected] [email protected]

Chapter 14 Dee Harris, Dr Venkata R. Machavaram and Professor Gerard F. Fernando* School of Metallurgy and Materials University of Birmingham Edgbaston Birmingham B15 2TT UK Email: [email protected]

Chapter 15 Professor G. F. Smith WMG Department of Engineering University of Warwick Coventry CV4 7AL UK Email: [email protected]

Dr Carlos Ramírez Av. Constituyentes 120 Pte 2do Piso Col. El Carrizal. Cp 76030 Queretaro, Qro. México Centro de Ingeniería Avanzada en Turbomáquinas S.de R.L. de C.V. Email: carlos.ramirez@ warwickgrad.net Neil Reynolds* WMG Department of Engineering University of Warwick Coventry CV4 7AL UK Email: Neil.reynolds@warwick. ac.uk

xvi


Chapter 17

Chapter 19

Dr Kevin Potter* and Carwyn Ward Department of Aerospace Engineering Queens Building University Walk University of Bristol Bristol BS8 1TR UK Email: [email protected]

Dr Mizi Fan Head of Research Civil Engineering School of Engineering and Design Brunel University Uxbridge UB8 3PH UK Email: [email protected]

Chapter 18

Phil Purnell, Ph.D., B.Eng. Reader (Civil Engineering Materials) School of Civil Engineering University of Leeds Leeds LS2 9JT UK Email [email protected]

Dr Miggy M. Singh and Dr John Summerscales* Advanced Composites Manufacturing Centre School of Marine Science and Engineering University of Plymouth Plymouth Devon PL4 8AA UK Email: [email protected] [email protected] Ken Wittamore Triskel Consultants Limited 12 St Fimbarrus Road Fowey Cornwall PL23 1JJ UK Email: [email protected]

Chapter 20

Andrew Dunster, Ph.D., B.Sc., C.Chem., MRSC Principal Consultant (Materials) BRE Bucknalls Lane Watford Hertfordshire WD25 9XX UK Email [email protected]

Preface

Research and development into materials and materials science underpin the modern world we live in. However it is a world in which our increasing knowledge and use of advanced materials have enabled us to see the damage being inflicted on the fabric of those materials that make up our natural environment. Material manufacture accounts for a high level of toxic manufacturing emissions, even before the effects of the consumer are felt in the waste stream. Therefore considerable emphasis has been placed on environmental policy makers, materials manufacturers, designers and consumers to consider the effects of their materials, products and lifestyles. It is therefore timely to consider research and current industrial practices which challenge the negative effects of materials in use. Composite materials have a smaller presence in public perception than other waste material types, considered generically as just plastics, ceramics, paper, metal, etc. This is because the quantities of use are smaller, the life cycles much longer, and their general visibility is lower. They have therefore received less attention from those seeking to meet environmental quotas than more noticeable and more easily dealt with material types. However, it is often these ‘invisible’ mixed materials in the waste stream, the left-over residue materials and the difficult to handle wastes that present the greatest challenge and pose the greater environmental risk than those materials more easily reused, recycled or recovered. The area of composite recycling and recovery continues to provide a substantial challenge to the academic community in developing new methods to alleviate environmental effects. Problems, potential solutions and current state of the art are all emphasised in this book. It should provide a good general coverage across the scope of the composite waste field. The legislative guide, in conjunction with the industrial case studies (each operating under their own different legislative pressures and framework), allows a comparison to be made of current and future trends. xvii

xviii

Preface

Insights into current university research by leaders in the composite field, in areas such as cleaner processes, more durable materials, and state of the art disposal methods, all contribute to provide the academic community with a comprehensive guide in which to continue breaking the boundaries in this important area of research. Vannessa Goodship

1 An introduction to composites recycling

Saturday, August 06, 2011 3:21:49 PM

N. R E Y N O L D S and M. P H A R AO H, University of Warwick, UK

Abstract: This chapter discusses the broad spectrum of physical attributes and mechanical properties of the entire family of polymer composites, which make some an inherently viable recycling proposition whilst leaving others consigned to landfill. All common material types are covered, from the high value speciality polymer matrices such as polyetheretherketone (PEEK) and carbon reinforcements, through to commodity materials such as polypropylene (PP) and glass. The question of what makes a material ‘recyclable’ is considered in terms of ease of recovery, practicalities of the recycling process itself and, importantly, the demand for the resultant recyclate material – effectively, the economics behind recycling these materials. This question inevitably covers some of the existing legislation that drives the economic argument. The effect of parameters such as raw material format, subsequent material processing, resultant reinforcement architecture and in-service usage all have on the economic viability are considered. Key words: polymer composites, recycling.

1.1

Introduction

1.1.1 Why would we want to recycle composites? Composites are generally considered high value, high performance materials that are employed in producing high net worth end products. When considering a typical end-of-life product made using composite materials, if the cost of raw material, the production tooling and the associated manufacturing equipment (including both moulding and finishing processes) are taken into account, it is obvious that such a component represents significant prior investment and embodied energy. Products made using materials of such high intrinsic value can be a wise target for the recycling industry when compared with lower net value materials as the corresponding recyclate material could have a similar high value, depending on the effort required to retrieve and reprocess the material. 3

4


1.1.2 Scope


The term composite can be used to describe a large number of multi-phase materials, consisting of a wide variety of matrix materials along with a correspondingly large array of different fillers and reinforcements. This book refines that description to only consider polymeric matrixes reinforced with various commonly used fibres. Figure 1.1 shows a scanning electron micrograph of a fractured polypropylene (PP) glass composite test specimen that clearly demonstrates the bi-phase nature of composite materials. In the micrograph, the glass fibres are clearly separately visible (diameter approximately 17 μm) to the PP matrix. This refined description still covers a broad scope of materials with a wide range of material properties, both physical and mechanical. Commonly used examples of polymeric matrix materials include: ‘short’ and ‘long’ fibre reinforced thermoplastic and thermoset compounds (processed using injection or compression moulding); long/continuous fibre reinforced laminates (typically processed using hand lay-up/vacuum bagging or resin transfer moulding (RTM)). The scope also includes newer fibre/resin combinations such as naturally sourced ‘bio-resins’ and natural fibres.

1.1.3 Thermoplastic versus thermosetting The resin chemistry employed in polymer matrix composite materials can be divided into two types, thermoplastic and thermosetting. This distinction has a very large impact on the inherent recyclability of a composite material.

˝50 μm

1.1 Scanning electron microscope image of polypropylene–glass composite fracture surface.

An introduction to composites recycling

5


Thermoplastic polymer matrices soften and melt with the application of heat. Any process step throughout a thermoplastic composite’s life cycle, from the initial introduction of reinforcement fibres to the final moulding of a component, takes place with sufficient heating to melt the polymer. Although this ability to melt can limit the application of such composites due to comparatively low maximum in-service temperatures, it does mean that end-of-life thermoplastic composite components can be shredded/ ground and readily re-processed via heating and moulding. The penalty for repeated processing in this manner is only limited degradation in matrix properties and reinforcement fibre damage. However, thermosetting systems undergo a permanent cross-linking reaction when curing that, although resulting in a stiffer (and more brittle) matrix material, cannot be reversed with the application of heat. The application of heat after curing only degrades the cross-linked polymer matrix and will not melt it. This means that practical end-of-life recycling options are limited, and could more properly be defined as ‘reuse’, such as in the case of incineration with energy recovery and also the reuse of thermosetting composite (via regrinding) as a low value filler material.

1.2

Composite material types

The following section covers the more commonly encountered composite materials listed by process method, along with their usual name and acronym. Resin type and fibre type are given, along with the approximate fibre volume fraction most commonly encountered.

1.2.1 Injection moulding Injection moulding is a high pressure (20+ MPa), flow-forming process whereby both the polymer and fibres exhibit bulk flow. Generally used for mass-produced components, the injection moulding process allows for fast cycle times, offering low cost at high volume. Such materials are short fibre reinforced, with predominantly glass fibre, although some carbon fibre materials are used. The fibre length is less than 5 mm once the material is processed, as shear forces in the injection moulding barrel damage the fibres and reduce the overall length. Common thermoplastic resins are polypropylene, polyamides; polyester and vinylester are typically used thermosetting resins. •

Short fibre reinforced thermoplastic injection moulding grades. Fibre volume fraction (v.f.) 20–40%. Despite the shorter fibre lengths (5 mm or less), some grades are referred to as LFT (long fibre technology) due to the preservation of fibre diameter to fibre length ratios of over

6

•

Management, recycling and reuse of waste composites 100× when using specific pellets and low shear moulding equipment. Typical applications include automotive (highly integrated front-end carrier). Bulk moulding compound (BMC) or dough moulding compound (DMC). These are thermosetting polyester matrix materials that are loaded with both glass fibres and fillers. Fibre v.f. 12–25%, typical filler fraction by weight (w.f.) 50%. Typical applications include automotive (headlamp reflector), white goods (oven/grill handle) and electrical (insulator).


1.2.2 Compression moulding Compression moulding is a high pressure (10+ MPa), flow-forming process whereby the polymer exhibits bulk flow, with limited movement of the fibres as well. The end result is a pseudo-random 3D fibre architecture, with the possibility of resin-rich regions towards component edges and in smaller details (e.g. ribs and bosses). Owing to this risk, these components are generally simpler in geometry when compared with injection moulded parts, although often larger in size and with correspondingly longer cycle times. As the moulding cycle does less damage to fibres, the fibre length can be much greater. Glass is the primary reinforcement to be used, and fewer resins are commonly used than with injection moulding, with the predominant choices being polypropylene (thermoplastic) and polyester/ vinylester (thermoset). •

Sheet moulding compound (SMC). This is a very similar material to BMC type materials utilising highly filled polyester matrices, but with longer fibres (needled, continuous fibre mat). Uncured sheets of prewetted composite material are transferred to a heated matched press tool. Fibre v.f. 15–60%, filler w.f. up to 40%. Large volumes used in automotive (car/truck bonnets, exterior panels, bumper beams, underbonnet components), and other applications include domestic (shower tray, sinks), marine (jet skis, boat parts), electrical (housings, insulator blocks) and transport (rail carriage interior panels). • Glass mat thermoplastic (GMT). This process utilises glass fibres in needled-mat sheets (chopped and continuous) within a thermoplastic matrix (usually PP). The fully consolidated sheets are pre-heated in process ovens, and then transferred (often manually) to a matched press tool. Fibre v.f. 20–40%. Applications include automotive (bumper beams, tailgates), construction and furniture manufacture. • Long fibre technology (LFT). LFT is a successful and more recently used modification of GMT technology. In LFT compression moulding, glass rovings are compounded with PP in an extruder and either imme-


7

diately transferred to a compression moulding press as a bulk material (direct extrusion/compression) or granulated and supplied as a pelletised material for subsequent extrusion/compression moulding. Reinforcement v.f. 20–40%. This is typically used for complex parts with high levels of integration such as automotive front end carrier modules and as a replacement for GMT material.


1.2.3 Laminate processing These materials with very long (effectively continuous) aligned fibre reinforcements provide the highest available mechanical properties, but at a cost of longer cycle times. Both carbon and glass fibres are regularly used, with the most common resin systems used being thermosetting polyesters and epoxies. All of the production routes used for laminate processing have the common feature of maintaining initial reinforcement fibre placement whilst the resin system is either introduced or consolidated throughout manufacture. This means that the reinforcement fibres, which are typically continuous, can be placed exactly in line with the expected in-service loading, tailoring the material directly to the application. However, it must be remembered that loads acting away from the axes of reinforcement fibres are not supported, and failure can occur. Laminate fibre architecture varies from unidirectional (UD), where all of the fibres are aligned in a single direction, to multiply (layered) arrangements such as cross-ply (plies oriented at 0° and 90°) and on to quasiisotropic (a ply stacking sequence of 0°, 45°, 90°, −45°). •

•

Hand lay-up. The simplicity of this technique allows the manufacture of very large structures utilising continuous fibres, usually glass but sometimes carbon. Achievable glass volume fraction is low compared with both RTM and pre-preg processes, with higher porosity and lower quality surface finish. Vacuum-bagging is a common variant of this process whereby the pre-wetted, hand-laid material is enclosed in an evacuated airtight bag during resin cure to aid consolidation and improve part quality. Maximum fibre v.f. 40%. Owing to the low investment costs arising from requiring no capital expenditure on press moulding equipment, and simple single-sided tooling made from cheap materials (e.g. wood, foam), this approach is often used for very large structures such as boat hulls (with or without vacuum bag) and also prototyping/short production runs. RTM. A pressure-assisted resin infusion route that is capable of producing higher quality components than hand lay-up, with higher volume fractions of fibre. A fibre preform is placed in a sealed, matched

8

•


•

Management, recycling and reuse of waste composites (often steel) mould and the resin is introduced under pressure. The higher part quality and hence part performance combined with the applied resin pressure means that carbon fibre (or other high performance reinforcement) is more regularly used. Maximum fibre v.f. 60%. This is used in niche market automotive for body panels and structural members (e.g. Lotus Elise body and structural front-end, Aston Martin Vanquish A-pillar), light aircraft manufacture and transport (train seats). Vacuum assisted RTM (VARTM). A manufacturing route typically used for smaller volumes or very large parts (e.g. wind turbine blades and boat hulls) that combines the simplicity of hand lay-up techniques with the achievable part quality towards that of RTM. Fibre preforms are laid dry into a single-sided tool with an upper vacuum bag, or a sealed matched tool, and resin is then introduced under vacuum. Maximum fibre v.f. 55–60%. Pre-preg material vacuum bagging or autoclave. Pre-preg (pre-impregnated) material consists of a reinforcement textile (typically woven) fully wetted with a partially (B-stage) cured thermoset resin. When used with the vacuum bag (1 bar, 100 kPa) or autoclave (5+ bar, 0.5 MPa), this technique is capable of producing components of very high fibre volume fraction, so often uses ‘high performance’ fibres such as carbon. Owing to the need for higher temperature and pressures within the processing route, large component size can be difficult to achieve. Maximum fibre v.f. 55–60%. Used for high end applications where specific properties (weight) is an issue such as motorsport and aerospace.

1.2.4 Natural fibre and natural resin materials (biocomposites) Although natural fibre is one of the oldest types of reinforcement used, it is a relatively novel component in terms of use in modern materials. Hemp, kenaf and jute are typical examples and are often added as reinforcements to injection moulding grade thermoplastics (such as PP) as used in non-structural automotive applications such as door inner modules and undershields. Natural resins (biopolymers) represent a new and rapidly developing area of composites research and application, production volumes are very small and these materials are typically employed in non-reinforced applications such as biodegradable packaging (Carus & Gahle, 2008).

1.3

Physical properties

The mechanical properties of composite materials can vary widely depending on the matrix-reinforcement fibre combination, reinforcement architec-


9

ture, fillers and chosen manufacturing process. Consequently a table of outline material properties must have a large variation in quoted values. Such a table is still of use as it allows initial comparison of material properties as well as providing a benchmark against which to measure the success of the recycling process in achieving acceptable material properties.


1.3.1 Polymer resin matrices As covered in the introduction, the main distinguishing feature of a polymer matrix material relates to the cure chemistry, i.e. thermoplastic and thermosetting. Typical values of selected mechanical properties for some (unreinforced) resins that are commonly used within composite production are given in Table 1.1. The selection is divided into thermoplastic and thermosetting resin types. The modulus and strength data from Table 1.1 are plotted together for each resin in Fig. 1.2. It is obvious from this graph that there is a much wider variation of properties available for the thermoplastics, and particularly that only the very high performance thermoplastic resins such as PEI (polyether imide) and PEEK (polyetheretherketone) compete with the thermosetting polyester and epoxy matrix materials.

1.3.2 Reinforcements Typical properties of individual fibre filaments are given in Table 1.2. It is important to note that these properties are somewhat idealised, and the

Table 1.1 Typical properties of selected commonly used polymer matrices

Material

Tensile modulus (GPa)

Tensile strength (MPa)

Specific gravity

Thermoplastic PP ABS PEI Nylon PET PEEK

1.1 2.3 2.9 2.8 3 3.9

30 45 85 60 70 90

0.90 1.04 1.27 1.13 1.35 1.28

Thermosetting Vinylester Polyester Epoxy

2.9 3.6 3.7

55 68 75

1.15 1.2 1.16

10

Management, recycling and reuse of waste composites 100


90 80 70 60 50 40 30 20

Thermoplastic resins

10 0

Thermosetting resins 0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5


1.2 Tensile strength versus modulus for selected polymer matrices.


Table 1.2 Typical reinforcement fibre filament properties

Fibre type

Modulus (GPa)

Strength (GPa)

Strain to failure (%)

Specific gravity

Diameter (μm)

E-glass S-glass HM carbon HS carbon Aramid

70 85 500 290 110

2.4 4 4 5.5 3

4.5 5.5 1 2 2.4

2.5 2.5 1.8 1.8 1.5

15 9 4.5 7 12

properties of fibres within a given bundle or tow will vary considerably, giving a more distributed mechanical response when used in bulk within a composite material. A comparison of the commonly used E-glass against the premium high modulus (HM) carbon fibre shows almost an order of magnitude of variation in modulus in favour of the carbon, but at the expense of a correspondingly low strain to failure, which results in a brittle failure mode within a composite material. Figure 1.3 plots filament strength against the strain to failure; the approximate fibre modulii can be observed as the gradients of the individual slopes.

1.3.3 Composite properties Table 1.3 presents average mechanical properties for some of the more commonly encountered composite material types, indicating the chosen fibre type and fibre volume fraction (v.f.). The mechanical performance of


11

Filament strength (GPa)

6 5 4 3 2

E-glass S-glass HM carbon HS carbon Aramid

1 0 0

1

2

3

4

5

6

Elongation to failure (%)

1.3 Reinforcement fibre filament strength versus elongation to failure.


Table 1.3 Mechanical properties of production composites

Processing

Material

Fibre

v.f. (%)

Modulus (GPa)

Strength (MPa)

Specific gravity

Flow (injection)

BMC PP-LFT

Glass Glass

20 30

13 6.5

31 100

1.8 1.12

Flow (compression)

SMC PP-GMT

Glass Glass

30 40

11 5.3

80 100

1.9 1.2

Hand lay

Polyester PP

Glass Glass

40 35

20 15

320 310

1.25 1.2

RTM

Polyester

Glass

55

22

360

1.9

VARTM VARTM (UD)

Epoxy Epoxy

Glass Carbon

55 55

24 130

550 1200

1.9 1.6

Prepreg

Epoxy

Carbon

50

60

890

1.5

the selected composites varies from being only slightly above that expected of unfilled polymer for the low (∼20%) volume fraction, randomly oriented, short fibre reinforced flow forming materials, towards metal-like properties for high v.f. continuous aligned fibre reinforced laminates.

1.3.4 Fillers It is not always possible to make a distinction between fillers and other reinforcements – indeed in some cases fillers are used to directly alter (improve) the performance of the chosen polymer. In other cases, the filler is added as an economic consideration to reduce the volume of polymer needed in a given material system.

12


Common filler materials are calcium carbonate, talcum powder and mica (Murphy, 2001). Inorganic particulate fillers can have a dramatic effect on the recyclability of composite materials, as they have low value, are hard to separate from the polymer and moreover will not burn. Heavily filled materials such as SMC/BMC (over 40% v.f. filler) are therefore not attractive for some of the recycling/reuse scenarios discussed here.


1.4

Current composite market

The most recent figures for glass reinforced plastic (GRP) production in Europe show an overall production volume of ∼1 200 000 t in 2007 (Witten, 2008). This represents a yearly growth of over 5% since 2005 (1 060 000 t). Europe produces 31% of the worldwide supply of GRP, and GRP comprises 90% of the world composites market (Bunsell and Renard, 2005). We could therefore estimate that the worldwide production of GRP in 2007 would be approximately 3 600 000 t, and that the global production volume for all polymer matrix composites is nearer to 4 000 000 t. Figure 1.4 shows European GRP production by processing method. It can be seen from this breakdown that the primary thermosetting composite production methods account for more than 70% of total GRP output: compression/injection moulding (SMC/BMC), 26%; open mould methods (hand-lay, spray-up), 30%, RTM and continuous processing methods, ∼20%.

GMT/LFT, 8%

Others, 2% SMC, 19%

Centrifugal casting, 6% Filament winding, 7%

BMC, 7% Pultrusion, 4%

Sheets, 7% Hand lay-up, 20%

RTM, 10% Spray-up, 10%

1.4 European GRP production by process route, 2005–2007 (Witten, 2008).


13

Recent figures for the total annual biomaterials usage in Europe amount to approximately 350 000 t, with ∼200 000 t of this total representing natural fibre (NF) reinforced composites (Carus and Gahle, 2008). Over two-thirds of this NF composite usage is composed of wood reinforced plastic compression- and extrusion/injection moulding for the automotive, construction and furniture industries.

1.5

What is ‘suitability for recycling’?

There are many factors which can affect suitability for recycling, but the driving arguments must be ‘is there a market for the recyclate?’ and ‘is the market economically viable?’ Without these drivers in place a recycling plan is almost certain to fail and a material must be considered unsuitable for recycling. Many technical factors affecting the recyclate such as collection, separation and reprocessing, can be overcome, but lack of market demand will stop any further development.


1.5.1 Economics The basic economic model must consider a number of factors. Many of these factors are closely interlinked, as shown in Fig. 1.5 and described

Financial

Investment Profit Prohibition Subsidy Penalty

Legislation

Local Governmental

Level

Federal (EU) Logistical Economic suitability

Transport infrastructure Buildings and plant

Technical

Recoverability Separation method Virgin material Supply

Confidence Demand

Recyclate

Post-industrial Post-consumer

Virgin material Recyclate

1.5 Schematic showing economic drivers for recycling suitability.

14


here. When embarking upon a recycling plan all of the factors need to be considered and some given action whilst others may be ignored. These factors are: technical, logistic, financial, confidence, supply and demand, and legislative.

1.5.2 Technical suitability In many cases it is often down to the simple practicality of the process that can drive the suitability for recycling. This practicality is affected by: waste stream type, recoverability, contamination and reliability of feedstock.


Waste type There are broadly two types of waste material to be dealt with – post-industrial waste and materials arising from end-of-life components (post-consumer waste). Post-industrial waste can be an attractive area for recycling schemes, as the waste stream has a clear history and contamination can be controlled and minimised, and the waste material is otherwise unaged and pristine. Typically post-industrial recycling can be done in-line with an existing ongoing production process – an example would be the introduction of thermoplastic LFT moulded component waste off-cuts back into a continuous direct extrusion/compression process for re-extrusion. End-of-life waste is obviously difficult to assure and control in terms of age, usage history and contamination. This is one of the biggest problems facing a recycling scheme.

Recoverability This can cover a number of aspects, from the basic recoverability of the composite material within a given component, to resin recovery and fibre recovery. There is a conflict between the economic considerations between producing a component and dealing with the same part once it has reached end-of-life. The drive to cut costs encourages designers to increasingly employ an approach of parts integration whereby a component becomes a multi-material system often assembled in a single step. If this parts integration path is followed without consideration towards design for disassembly, it can result in multi-material components that are very difficult to process at end-of-life. Composite components can be produced with items such as metallic fixing inserts, aesthetic coverings such as paint films, carpet or leather all included in-mould; these present a significant challenge to a prospective dismantler. With one-way assembly methods such as adhesive


15

bonding or painting, if the non-composite parts cannot be removed, the resultant recyclate will then be contaminated. Constituent material recoverability can affect the recycling viability of composite materials in different ways. The material architecture as well as material formulation can affect this. Resin recovery from heavily glass filled materials such as SMC is very difficult. Fibre recovery can be much simpler, particularly for glass and carbon fibres, but here the limit can be recyclate market demand. Contamination Although this is a major issue for general plastic recycling, for composite materials this is less so, owing to the fewer material options and grades available, making any sorting easier. Also in the case of end-of-life large scale constructions, such as wind turbine rotor, a large volume is available at a single time (this is also a consideration for feedstock, below).


Reliable feedstock Although volumes of composites used are increasing, owing to a wide variety of factors, overall they are still relatively niche compared with mainstream materials. This is exacerbated by the long lifetime of these materials in operation. This could create an unreliable feedstock for longterm planning, but this should change in the future as composites used in long term structures come to their end of life.

1.5.3 Financial suitability: legislative, investment and material costs •

Legislative financial instruments: subsidies/penalties. Governments are able to stimulate markets by legislative means (e.g. Council Directive 1999/31/EC). They can legislate to influence the final recyclate cost and likely demand in two ways: subsidies and penalties (commonly phrased ‘carrot versus stick’). Subsidies are regularly used to stimulate demand, but will often be short-lived, and their subsequent removal can easily destroy demand that has come to rely on the effect of subsidies too much. These subsidies can take two forms, either assistance with plant investment or payments upon recyclate produced. Penalties take the form of financial disincentives such as increased landfill taxes that actively encourage reuse and recycling thus ultimately making disposal economically unviable. Within the economic argument, legislation can have a large effect, as it is used to drive the existence of a market until

16

•

•

Management, recycling and reuse of waste composites the infrastructure is in place to create usable volumes at a competitive market price. Additionally, penalties are likely to be much longer lived – indeed indefinite – and so have a more permanent effect on the market than the more transient subsidies. Capital investment. Recycling plant, building and infrastructure all require a large amount of initial investment and securing this can prove very difficult. This can take the form of assistance from local or national government or direct investment from industry. Material costs. Often driving the value of the resultant recyclate is the initial value of the material. Materials with a high initial raw material value such as carbon fibre will often have a strong economic driver for recycling as long as there is potential for a high value recyclate product, even if re-processing costs are high.


1.5.4 Confidence End user confidence plays a vital part in a recycling programme, particularly with respect to the supply side. This can be seen in the need to guarantee volume of supply to large users, who cannot afford to stop production, and a lack of confidence in ability to supply sufficient reprocessed material can provide problems, particularly for small recyclers.

1.5.5 Supply and demand Demand and confidence are closely interlinked, as one will tend to create the other. It is still important that a need is developed as this demand will create a market and obviously dictates the prices. The market demand for resultant recyclate is of huge importance. Unlike the initial cost, it is often the lower value materials with a high usage volume where the greatest demand exists. Typically here glass reinforced PP has a relatively low initial value but the volumes used make it easy to feed recycled material into the virgin feedstock with no problems.

1.5.6 Legislation Governmental legislation can be employed, not just to provide financial incentives/penalties, but to prohibit current practices (e.g. disposal) thus making reuse and recycling the only option (Waste Electrical and Electronic Equipment, Council Directive 2002/96/EC; End-of-Life Vehicles, Council Directive 2000/53/EC). This direct action approach can create supply without necessarily stimulating demand and presents a significant challenge to the recycling industry.


1.6

17

Recycling methods

As previously demonstrated, composite materials can offer greatly enhanced performance over commodity materials. They are of relatively high value and their usage pattern as engineering materials is generally in long-lived applications. This longer life cycle means that potential end-of-life scenarios must be judged differently from those commodity materials that are used in higher volume/lower cost applications.

1.6.1 Landfill disposal


Once a material is destined for landfill it is considered to be of no value; indeed costs will be then incurred at point of disposal; furthermore this option is growing politically less favourable. Although disposal is a simple process, the inherent environmental durability of composite materials means that they undergo only slow degradation in landfill conditions. On the positive side, the low volumes of composite materials currently used when compared with other material markets means that they make up only very small volumes of landfill waste.

1.6.2 Incineration When combined with energy recovery, incineration can be viewed as a recycling technique as the material has effectively been reused. The potential effectiveness of this as a technique is very dependent on the composition of the composite. High glass filled materials with large inorganic filler content, such as SMC, contain very little inflammable material and will burn poorly and will be therefore of little value. Other materials, such as those with a flammable thermoplastic matrix will burn readily and have a large calorific value.

1.6.3 Chemical techniques These techniques can involve either chemical de-polymerisation of the matrix into oils which also frees the fibres for further recycling, or chemical removal of the matrix which also free up high value fibres, such as carbon. These techniques are still in the developmental stage and hardly employed. They could potentially be of value, but the currently low recycling volumes of composite materials means that to be cost effective, the technique will have to be flexible enough to chemically re-process other materials.

18


1.6.4 Thermomechanical processes These processes are the most widely used to re-process both composites and other materials. The techniques are generally confined to thermoplastic resins, as fully cross-linked thermosetting resins are virtually impossible to re-process. Plants performing thermomechanical reprocessing of thermoplastic composites are also able to re-process other thermoplastic (unreinforced) materials, which increases throughput volumes and therefore significantly improves the economic viability of the business.

1.6.5 Mechanical processes


Mechanical processing is a low-value option that can be employed in the case of materials where no recycling/reuse approaches exist other than disposal. Materials such as highly filled thermosetting composites can be ground and used as filler in virgin materials. One study found that up to 25% SMC regrind could be introduced into new SMC/BMC formulations without any performance penalty (Murphy, 1998).

1.6.6 Biomaterials Composites made with natural fibres offer similar recycling routes to their inorganically reinforced counterparts. Thermoplastic natural fibre composites can be thermomechanically reprocessed in the same way as other thermoplastic composites. Both thermoplastic and thermosetting natural fibre composites are attractive for incineration with energy recovery due to their lack of inorganic content – they are combustible in their entirety. Using certain biopolymers in conjunction with natural fibres can result in a completely biodegradable composite material system. These biomaterials would be suitable for end-of-life composting, given the correct environmental conditions. This composting can be carried out aerobically as a simple disposal solution or anaerobically in a bio-gasification process – which amounts to a ‘reuse’ scenario, producing useful combustible gases (European Bioplastics Organisation, 2008).

1.7

Conclusions

It is clear that composites offer the design engineer an attractive range of material solutions with a correspondingly wide range of production routes. Material properties can be accurately tailored to design specifications, resulting in parts that are correctly engineered and fit for purpose with minimum material usage (via avoiding the all-too-common approach of



19

over-engineering). By careful selection of material type and corresponding production route, solutions can be formulated that satisfy an extremely broad range of applications and production volumes, from one-off 30 metre long boat hulls to 20 g injection mouldings produced at rates of over 100 000 parts per annum. This design freedom and flexibility has led to an ever-increasing market for composite materials, which now makes the issue of how to deal with end-of-life composites an important consideration. There are a number of recycling routes available now, with many more variants currently under research. As the waste composite material market expands due to increasing production and the advent of legislation prohibiting disposal, economies of scale will come into play that will make hitherto unprofitable recycling schemes viable. However, many recycling schemes are focused on dealing with thermoplastic composite waste. One major hurdle that remains is the lack of ‘true’ recycling routes for waste thermosetting composite materials. Efforts have instead examined ‘reusing’ thermosetting composites due to the difficulties associated with recycling, using routes such as incineration with energy recovery, and shredding and grinding to make low-value filler to be used in the production of other materials. The current optimum solution for thermoset recycling is reinforcement fibre recovery used in conjunction with these thermal and mechanical ‘reuse’ methods.

1.8

References

bunsell a r and renard j (2005), Fundamentals of Fibre Reinforced Composite Materials, CRC Press carus m and gahle c (2008), ‘Natural fibre reinforced plastics – material with future’, Presentation at International Congress – Raw Material Shift & Biomaterials, nova-Institut GmbH, Huerth, available from http://www.eucia.org/ publications/publications council directive 1999/31/EC of 26 April 1999 on the Landfill of Waste council directive 2000/53/EC of the European Parliament and of the Council of 18 September 2000 on End-of-Life Vehicles council directive 2002/96/EC of the European Parliament and of the Council of 27 January 2003 on Waste Electrical and Electronic Equipment (WEEE) european bioplastics organisation (2008), ‘FAQ paper on bioplastics’, http:// www.european-bioplastics.org murphy j (1998), The Reinforced Plastics Handbook, Elsevier murphy j (2001), Additives for Plastics Handbook, Edition 2, Elsevier witten e (2008), ‘The composites market in Europe: market developments, challenges, and opportunities’, AVK (Industrievereinigung Verstärkte Kunststoffe), Online report (http://www.eucia.org/publications/news)

2 Legislation for recycling waste composites R. S T E WA RT, c/o University of Warwick, UK


Abstract: This chapter first looks at how legislation is passed in the European Union and the relationship between EU and domestic UK laws and policies. EU environmental policy and legislation have been developed and shaped by six environment action programmes for over 30 years. The first batch of environmental legislation put in place the framework for dealing with specific problems such as waste oil. This was built upon by legislation that developed standards for landfills and incinerators for example whilst the third batch of legislations introduced recycling directives which put in place the necessary organisation and finances to assist the recycling of key waste flows such as end-of-life vehicles. Some lack of clarity in the wording of the directives meant that implementation among member states was variable and the current legislation seeks to harmonise and streamline environmental law making it simpler and more enforceable. The chapter then goes on to discuss some of the key pieces of legislation that are likely to be of most relevance to composite materials. A number of key changes have been proposed in some areas, for example, the reduction of carbon dioxide emissions from passenger vehicles and potential areas for further legislation are highlighted. Key words: recycling legislation, environmental legislation, climate change, landfill, EU directives, UK regulations, environmental action programmes, waste management.

2.1

Legislation

Virtually all of the legislation covering the control and management of waste, not just of composite materials, is enshrined in European Union (EU) law and policy which has become transposed into domestic law. It is therefore worth beginning this chapter with a recap of the history and structure of the EU and its relationships with the UK governance with regards to environmental issues. The workings of the EU are complex and this chapter only intends to give a broad-brush outline of how laws and policies are made and how these become transposed into UK laws. The chapter will then go on to discuss the main EU regulations which apply to composite materials and then discuss what is likely to apply to such materials in the future. 20

Legislation for recycling waste composites


2.2

21

Brief history of the European Union

The European Union was established in 1993 and was borne out of the 1957 European Economic Community (EEC) which the UK joined in 1973. At the heart of both the EEC and the EU are the founding treaties, beginning with the Treaty of Rome which the UK signed up to in 1973 when it joined the EEC and the Treaty of the European Union (often known as the Maastricht Treaty) which established the EU. Each member state has to ratify the treaty. There have been a number of amending treaties but the founding treaties set out the basis for law in Europe, and this will be discussed further later in the chapter. The Treaty of the European Union divided responsibilities into three ‘pillars’ or areas of policy. Pillar one deals with matters which affect the EC and legislative matters which effectively bind all EU citizens, and pillars two and three are based on intergovernmental cooperation with respect to foreign and security policy and on police, criminal and judicial policies respectively. After significant expansion of the EU, most recently by Bulgaria and Romania in 2007, there are now 27 member states. In response to the enlargement, the EU treaties have most recently been amended by the EU Reform Treaty signed in Lisbon in 2007. The Lisbon treaty provides for a number of reforms but one of those will be to effectively dispense with the pillar system (http://europa.eu/abc/history/index_en.htm). The administration of the European Union is rather complex but the important bodies for the purposes of this section are shown in Fig. 2.1. •

The European Parliament is the only body in the EU which has elected members and together with the Council of the European Union (often just referred to as the Council) they are the main legislators of the EU Council of the European Union

Council of Ministers

Commission

2.1 Main bodies within the EU.

EU Parliament


22


and are responsible for passing legislation. There are two levels of the council, the Council of the European Union is a fixed composition and is made up of heads of state or government and the president of the EU commission. The composition of the Council of Ministers, however, will vary depending on the issue being discussed. The Council holds the executive power in the EU which it confers to the European Commission. The strength of the Union was increased by the ratification of the Amsterdam Treaty in 1999, which had the effect of increasing the legislative power of the European Parliament and also for overseeing the work of the European Commission. • The European Commission (the Commission) is the executive of the EU and is responsible for the day-to-day running of the EU. It has two main areas of responsibility. Firstly, the Commission drafts legislation and proposes it to the Council or the Parliament who can amend it or pass it as legislation. Secondly, it ensures that the regulations and directives passed by the Council and Parliament are being implemented by the member states and if not, the Commission has the ability to take the non-compliant party to the European Court of Justice for enforcement. Each member state of the Union is represented by a commissioner but their role is not to represent the member state from which they originate but the Union’s interests. The Commission has a great degree of independence in carrying out its functions. As the EU’s executive, the Commission has broad powers to manage the EU’s common policies, such as research and technology, overseas aid, regional development and so on. It also manages the budget for these policies. The Commission is assisted by a civil service made up of 46 directorate generals. • The European Court of Justice (ECJ) is the highest court in the EU and has final say over all other courts on matters to do with EU law. It is completely independent of the bodies shown in Fig. 2.1. In terms of environmental policy and law, it is clear from the explanation of the Commission’s (and its relationship with the Directorate Generals) roles and responsibilities that it is through this arm of the EU that the current and future law and policy for the UK will flow.

2.2.1 The relationship between European Union and United Kingdom law The primary source of European laws are the founding and amending treaties already discussed. However, in and of itself, a ratified treaty does not have a legal standing in UK law and for it to have any legal force it has to be passed by an act of government. This was effectively achieved by the UK’s passing of the European Communities Act (1972), as amended. The


23

act also instructs courts that priority is given to EC law over UK law which means that UK legislators have to modify domestic laws to align with EC laws even if there is a conflict between the two. The secondary sources of EU law transposed from the treaties consist of regulations and directives which are characterised as follows: •

•

Regulations – these are binding and are directly incorporated into member state law without that member state having to use national legislation to incorporate them into law. If there is a conflict between a regulation and an existing national law, the regulation takes precedence. Directives – a directive sets out the requirements that the legislation should achieve. Whilst there is an obligation on the member state to change its national laws, a directive then leaves it to the member states to implement as it sees fit to meet the requirements. The directive will provide for a set period of time within which implementation has to be achieved via national legislative processes.


2.2.2 The European Union and the environment The origins of environmental law and policy in Europe lie in the European Community dating back to 1972. The European Community founding treaty, the Treaty of Rome, does not contain any reference to environmental legislation and a number of further amendments, including the Maastricht Treaty (1992) and the Amsterdam Treaty (1999), have been necessary to bring forth the legislation: more than 80% of the UK’s environmental legislation today is drawn from the EU. However, it was in 1972 that the European Community determined that a community environmental policy was needed. The framework which developed and shaped environmental legislation was provided by the Environment Action Programmes (EAPs), of which there are six to date (EAP1 1973, EAP2 1977, EAP3 1983, EAP4 1987, EAP5 1993, EAP6 2002). Each programme is specific to a particular environmental issue and spans a 5–10 year time frame. The EAPs are the main driving force behind today’s and tomorrow’s future environmental policy in the EU and the UK as well as the global agenda. Whilst the environment action programmes are not legally binding, they do clearly set out the aspirations of the community.

2.2.3 Past and present Environment Action Programmes In 1972 the European Community adopted its first EAP covering the period 1972 to 1976 (EAP1 1973). The first EAP set out the principles and priorities that would guide the community’s policies for the future. This was followed by a second five year programme from 1978 to 1982.

24



The first and second EAPs set out in some detail a series of actions that should be taken to control a range of pollution problems (EAP1 1973, EAP2 1977). There were 11 principles listed in the first programme in total and it is worth listing them in full, as many of these have been carried forward to subsequent programmes: 1. Pollution is better prevented at source. 2. In decision making processes, environmental effects should be taken into account as soon as possible. 3. Exploitation of natural resources to be avoided if they cause significant harm to the ecological balance. 4. Scientific knowledge should be advanced to help in the protection of the environment. 5. The polluter pays principle. 6. A country’s activities should not degrade the environments of others. 7. The environmental policy of member states must take into account the interests of developing countries. 8. The community should participate in international organisations to promote global environmental policy. 9. Environmental education should be promoted throughout the community. 10. Pollution control should be established at all levels (local, regional, national, community, international). 11. National environmental policy to be harmonised within the community. The third and fourth EAPs running from 1983 to 1993 tried to provide an overall strategy for protecting the environment and natural resources (EAP3 1983, EAP4 1987). The emphasis shifted from pollution control to pollution prevention and the concept of environmental protection was broadened to include land use policy, identifying agriculture as an area in need of more environmental awareness and the integration of environmental concerns into other EU policies. Each programme was a building block on the previous one but much of the focus of the first three programmes was devoted to developing the policies and little attention seems to have been actually given to whether or not the policies were effective or indeed being implemented at all (Hawke 2002). It was not until the fourth EAP that meaningful consideration was given to implementation. The fifth EAP from 1993 to 2000 (EAP5 1993) began to set longer-term objectives and the concept of sustainable development was introduced as well as managing waste not just on a community level but on a global level. The prevailing sustainable objectives from the programme were:

Legislation for recycling waste composites • • • •

25

to maintain the overall quality of life; to maintain continuing access to natural resources; to avoid lasting environmental damage; to consider as sustainable a development which meets the needs of the present without compromising the ability of future generations to meet their own needs.

As with earlier programmes, the fifth programme identified sectors where further work on environmental progress was necessary. These were industry, the energy sector, agriculture, tourism and, for the first time, transport. The EAP identified seven themes of work and began to identify targets that should be achieved in each theme. The guiding principles that came out of the fifth EAP were:


• environmental protection targets can only be achieved by involving those policy areas causing environmental deterioration; • replace the command-and-control approach with shared responsibility between the various parties (governments, industry and the public), to obtain commitment to agreed measures. The fifth programme identified that pure reliance on environmental legislation was not a means to an end and a number of policy regulations would be more appropriate: • • • •

Legislation to set environmental standards. Economic instruments to encourage the production and use of environmentally friendly products and processes. Horizontal support measures (information, education, research). Financial support measures (funds).

In effect what this EAP did was to replace the original top-down approach seen in the previous EAPs with a bottom-up one and recognition that in order to increase the level of implementation it would be necessary to get stakeholders to cooperate with one another. Naturally, the outcomes and recommendations of the fifth EAP fed into the sixth (current) generation of action programmes which runs from 2002 until 2012 (EAP6 2002). The sixth EAP entitled ‘Environment 2010: Our Future, Our Choice’ was adopted by the European Parliament and the council of the European Union in 2002. The treaty of Amsterdam (1997) signed previously fed into the sixth EAP with its requirement to give consideration to sustainability. Article 6 of the treaty states that ‘environmental protection requirements must be integrated into the definition and implementation of the Community policies . . . in particular with a view to promoting sustainable development’. There is consequently a greater impetus towards achieving sustainability and an emphasis on extending the

26


bottom-up approach and less of a focus on relying on legislation to tackle the environmental issues targeted. The sixth programme proposes a focus on four priority areas: 1. Climate change – cuts to global emissions of 20–40% by 2020. 2. Biodiversity – reduce threats to habitats and the survival of number of species. 3. Environment and health – water, noise and air strategies. 4. Sustainable management of resources and wastes – improve recycling rates and waste prevention strategies.


The sixth EAP has seven themes running through it, namely: air pollution, the marine environment, sustainable use of resources, prevention and recycling of waste, sustainable use of pesticides, soil protection and urban environment. In previous EAPs there has been a focus on specific types of pollutants such as sulphur dioxide or specific types of activities such as transport or tourism. In this case the strategies are global in nature and arranged into themes instead.

2.3

Environment Action Programme 6 and the shape of future legislation

The EAPs have been setting the EU’s agenda for environmental protection for over 30 years and much legislation has subsequently been created. Regulation is just one of a number of methods that have been employed in trying to implement environmental policies and is generally a means of last resort when voluntary agreements, taxes and subsidies fail to have the desired effect. Whilst regulation is an important tool in bringing about environmental change it also has its limitations as it can be a lengthy process to implement and amend with low enforcement rates. For example, the End-of-Life Vehicle (ELV) Directive (Directive 2000/53/EC) was first conceived in the 1960s with the recovery of valuable metals and the re-sale of component parts. During the mid-1990s, when the fifth EAP was in place, the EU and national leaders put pressure on the automotive industry to sign up to voluntary agreements which would see increases in the rates of recycling and recovery. Such voluntary agreements would also force the automotive industry to accept responsibility for the treatment of ELVs. By this time, priorities had changed towards waste prevention and this led to discussions on an ELV Directive in 1996 but this was not adopted until the year 2000. Ambitious targets in the Directive meant that member states failed to meet the target date for transposition into domestic laws and it was not


27

until 2003 that the regulations were implemented in the UK, but not in full (ELV Regulations 2003). The remaining provisions of the directive were transposed into UK law in 2005 by the End-of-Life Vehicle (Producer Responsibility) Regulations. More detail on the ELV Directive is given later in the chapter. In light of such slow progress in some areas, the sixth EAP seeks to toughen the EU’s approach on how it regulates and enforces environmental regulation but puts a great deal of effort into pursuing other non-legislative means of achieving the programmes objectives. Article 3 for example, sets out some strategic objectives which are both stick and carrot approaches to address the issues.


2.3.1 Legislative approaches • Develop and amend existing legislation. • Develop new legislation. • Encourage more effective implementation and enforcement, for example by better inspection procedures. • A systematic review of how legislation is being applied across member states and better exchanges between member states on best practice for implementation.

2.3.2 Non-legislative approaches • •

•

•

More integration of environmental protection requirements into the EU policies. Promotion of sustainable production and consumption by, for example, reforming subsidies, promoting of emission trading, reviewing environmentally related taxes and incentives. Improve the environmental performance of enterprises with a view to achieving sustainable production patterns. It is suggested that this may be assisted by an integrated product policy approach which takes into account product life cycle, encouragement for companies to publish independently verified environmental or sustainable development reports and a greater take up of voluntary agreements to achieve environmental objectives. Integrated product policy (IPP) and sustainable development (SD) will be discussed in more detail later in the chapter. Put more effort into informing consumers and purchasers (individuals, enterprises and public bodies) about the environmental impact of the processes and products they come into contact with. Examples of this may be through encouraging the use of eco-labels and the promotion of green public policy.

28


• Integration of environmental policy in the financial sector, including for instance agreements on including data on environmental costs in company annual financial reports and using environmental consideration in making lending decisions.


Article 8 of EAP 6 echoes Article 3’s strategic approach and deals specifically with the priority areas for sustainability and waste management which states the following objectives: – achieving a significant overall reduction in the volumes of waste generated through waste prevention initiatives, better resource efficiency and a shift towards more sustainable production and consumption patterns; – a significant reduction in the quantity of waste going to disposal and the volumes of hazardous waste produced while avoiding an increase of emissions to air, water and soil; – encouraging reuse and for wastes that are still generated: the level of their hazardousness should be reduced and they should present as little risk as possible; preference should be given to recovery and especially to recycling; the quantity of waste for disposal should be minimised and should be safely disposed of; waste intended for disposal should be treated as closely as possible to the place of its generation, to the extent that this does not lead to a decrease in the efficiency in waste treatment operations.

The objectives are to be achieved by; • Development and implementation of a broad range of instruments including research, technology transfer, market-based and economic instruments, programmes of best practice and indicators of resource efficiency; • Developing and implementing measures on waste prevention and management by, inter alia: (a) Measures aimed at ensuring source separation, the collection and recycling of priority waste streams; (b) further development of producer responsibility; (c) development and transfer of environmentally sound waste recycling and treatment technology; • Developing or revising the legislation on wastes, including, inter alia, construction and demolition waste, sewage sludge, biodegradable wastes, packaging, batteries and waste shipments.

2.4

Next steps for waste management

It is clear, therefore, that although further legislation to deal with waste management is inevitable, it is only one of a number of methods that will be employed to implement the EU environmental policy. In April 2006 the Commission produced a mid-term report on the sixth EAP (COM (2007)


29

225) which outlined what progress had been made since its inception in 2002 and prior to a planned full review in 2010. The mid-term report concluded that ‘the EU was still not yet on the path towards genuine sustainable development’ but had brought about significant improvement in a number of areas. The report outlined areas of the programme where further work was needed and in terms of legislation and policy to deal with the environment it was proposed that more needed to be put in place to deal with waste management. With most relevance to composite materials, the future areas to consider are:


1. 2. 3. 4. 5. 6. 7. 8.

The Framework Directive on Waste Environmental Permitting Regulations The Landfill Directive IPP The UK Climate Change Act Carbon dioxide emissions from passenger cars End-of-life vehicles Waste electrical and electronic equipment (WEEE)

2.5

The Waste Framework Directive

The Waste Directive has been amended a number of times since its introduction in 1975 (Directive 75/442 EEC). However, in response to the thematic strategies adopted in the sixth EAP, a review of the Waste Directive commenced and a proposal was put forward by the Commission in 2005 (COM (2005) 667). After public consultation, the proposal was adopted by the European Council as a directive in October 2008 and member states have until 2010 to transpose it into domestic law (Directive 2008/98/EC). The effect of the new directive will be to repeal the existing waste directive but also the directives on hazardous waste (Directive 91/689 EEC) and waste oils (Directive 75/439 EEC) thereby simplifying the legislation. This directive is the window through which all EU legislation on waste should be read. A significant driver behind updating the directive was the lack of clarity and precision in the definitions used in the 1975 directive. This led to variable implementation and differing interpretation between member states that had to be settled by the European Court of Justice mainly around the definition of waste and the distinction between the terms recovery and disposal (COM (2005) 667). The ethos behind the directive is to promote waste as a secondary resource in order to reduce landfill levels even further and thereby additionally bring about a reduction in greenhouse gas emissions. There is a

30


strong emphasis on waste prevention and the new directive sets even more ambitious recycling targets of 50% of household waste and 70% of construction waste by 2020. The inclusion of construction waste will have a significant impact on composite materials, with 11% of the UK’s fibre reinforced plastic production being used in the construction industry (Conroy et al. 2006). The language used in the directive has been harmonised with other directives and it harnesses product life-cycle thinking by encouraging producers to prevent waste generation in the first instance, for example. The 1975 directive set out a waste management hierarchy with the first priority being to prevent waste in the first instance, then to recycle and reuse. This hierarchy is the bedrock of sustainable waste management and has been embedded into legislation. The new proposals extend this thinking further and introduce further measures into the hierarchy based on their environmental impact.


2.6

Environmental Permitting Regulations

The Environmental Permitting Regulations 2007 (SI 3538) came into force in April 2008 and replace around 40 pieces of environmental legislation. A series of guidance documents has also been issued to accompany the regulations (http://www.defra.gov.uk/environment/epp/). The Environmental Permitting Regulations bring together a number of pollution control legislations into a single set of regulations. A number of environmental directives such as the Waste Electrical and Electronic Equipment Directive (Directive 2002/96/EC) and the End-of-Life Vehicle Directive (Directive 2000/53/EC) require waste operators to hold an authorised permit which cover their facilities, processes and compliance procedures. The Environmental Permitting Regulations set out the generic and specific terms that must be adhered to along with the reporting, inspection and enforcement procedures. The effect of the regulations is to harmonise environmental permits and reduce bureaucracy especially where more than one waste facility operates from a single site. Subject to public consultation, phase 2 will be rolled out between 2008 and 2011 (http://www.defra.gov.uk/environment/epp/deliver.htm#2).

2.7

The Landfill Directive

The 1999 EU Landfill Directive (Directive 1999/31/EC) was transposed into English law through the Landfill (England and Wales) Regulations 2002. It was subsequently amended in 2004 and 2005, which incorporated the council decision in 2003 (Council Decision 2003/33/EC). It is now implemented in the UK through the Environmental Permitting Regulations (SI 3538).


31

The overall objective of the regulations is to supplement the requirements of the Waste Framework Directive (Directive 2006/12/EC) discussed in the preceding section and prevent, or reduce, the negative effects of landfilling on the environment as well as any risk to human health as a result. The regulations attempt to achieve this through specifying technical standards of landfill sites and operations and setting out requirements for the location, conditioning, management, control, closure and preventative and protective measures. A strategy on biodegradable waste must also be implemented as part of the regulations that provides for the progressive diversion of biodegradable municipal waste from landfill in order for the UK to meet challenging targets: • •


•

By 2010 to reduce biodegradable municipal waste landfilled to 75% of that produced in 1995. By 2013 to reduce biodegradable municipal waste landfilled to 50% of that produced in 1995. By 2020 to reduce biodegradable municipal waste landfilled to 35% of that produced in 1995.

The target years for the UK set out above take advantage of the offer in the Directive to member states who landfilled more than 80% of their municipal waste in 1995 to have the flexibility to postpone the deadlines by up to four years.

2.8

Integrated product policy

IPP is a public policy initiative which has been on the EU agenda since roughly the late 1990s and is concerned with the reduction of environmental impact associated with products and services. The purpose of the EU initiative was to harmonise varying environmental product policy strategies that were developing within the Community to minimise the environmental impact of their products at varying stages of the product life cycle: for example, take-back schemes, product labelling, taxes or other economic initiatives. These differing approaches among member states led to a call for harmonisation of such measures across the EU in order to avoid problems that might arise from different policy approaches. Consequently, an IPP green paper was published, promoting common framework within the EU (COM (2001) 68). The European Commission published a green paper on IPP in 2001 (COM (2001) 68) which set out the rationale behind developing the IPP approach to environmental policies and set out possible mechanisms that might be used to achieve ‘greener’ products. Two pilot projects were

32


concluded in 2006 which focused on a mobile telephone and a tropical garden chair as example products (IP/06/1233). The work of the Commission has gone on to identify further products that can benefit from environmental improvement which includes passenger cars (Nemry et al. 2008). There is currently no specific legislation governing IPP.

2.9

United Kingdom legislation: The Climate Change Act 2008


In addition to complying with EU laws on the environment, the UK has its own domestic laws and in November 2008 it became the first country to introduce a climate change act (The Climate Change Act 2008). The act places legally binding obligations on polluters and sets challenging targets for the reduction in emission of greenhouse gases, meaning this act will have long-term implications for all UK businesses and producers. Key features of the act include: • in comparison to the level of emissions of greenhouse gases in 1990, a reduction of 26% by 2020 and at least 80% by 2050; • introduction of legally binding carbon budgets which cap emissions over five year periods, with three budgets set at a time; • creation of an independent Committee on Climate Change (CCC) to advise on the level of budgets and report to Parliament on progress on an annual basis – the CCC will publish its first progress report in September 2009; • the Government must report on a five yearly basis on the risks to the UK from climate change, and publish a programme setting out how these impacts will be tackled; • Government powers that oblige public bodies to conduct their own risk assessments and formulate action plans that address the risks identified.

2.9.1 Future passenger vehicle emissions The Climate Change Act should ensure that the UK easily meets its EU obligations to comply with vehicle emission standards that will be implemented in 2009 as part of Euro 5 and 2014 under Euro 6 (Regulation 715/2007). In December 2007, the Commission published a proposal to harmonise and reduce carbon dioxide emissions from passenger cars (COM (2007) 856). The Commission proposed a cut to the current 160 g CO2/km level of carbon dioxide emissions to 120 g CO2/km for the whole car industry by 2012.


33

In December 2008, Members of the European Parliament (MEPs) and the presidency of the European Council came to an agreement on new carbon dioxide emission levels (IPR 43441 2008). The agreement that was finally reached was 130 g CO2/km arising from advances in motor vehicle technology with a further 10% saving from other means such as tyre technology and use of biofuels. The Commission has a long-term objective to reduce carbon dioxide emissions further to an average of 95 g CO2/km across a car fleet by 2020 and has agreed to the industry phasing in steadily reducing values between now and 2015. It has set interim figures of 65% of the fleet by 2012, 75% by 2013, 80% by 2014 with full compliance with 95 g CO2/km by 2015. Any manufacturer who fails to meet these levels will be charged with an excess emission premium of: •


•

between 2012 and 2018 • 15 for the first gram • 115 for the second • 125 for the third • 195 for the fourth after 2019 • 195 for every gram over

These proposals will now be put through the legislative process but it is not yet known when a final agreement will be announced. A review of the legislative framework for the automotive industry in Europe has been ongoing since 2005 through the Competitive Automotive Regulatory System for the 21st century (CARS21). It is a high level group set up in 2005 which has objectives of generating recommendations for the short-, medium- and the long-term public policy and regulatory framework for the European automotive industry.

2.10

End-of-Life Vehicle Directive

The fifth EAP (EAP 5 1993), previously discussed in this chapter, began to set longer-term environmental objectives and introduced the concept of sustainable development as well as managing waste not just on a community level but on a global level. The programme stated that the attitude to waste management needed to change and suggested two strategies, one of avoiding waste by improving product design in the first instance but, where waste arises, a second strategy of recycling and reusing the waste. This European Community strategy led almost seamlessly into new legislation on managing of waste streams and from this flowed the End-of-Life Vehicle (ELV) Directive (Directive 2000/53/EC) and which was transposed into UK law on 3 November 2003 (End-of-Life Vehicle Regulations).


34


The regulations were subsequently amended in 2005 by the End-of-Life Vehicles (Producer Responsibility) Regulations 2005 with further minor revisions in 2008 (Environmental Permitting Regulations 2007, SI 3538). The first focus of the ELV legislation is to prevent waste in the first place and then to consider reuse and recycling as well as other forms of energy recovery. The scope of the legislation is not just contained to the vehicle itself but extends to their components and materials, irrespective of whether or not they were factory fitted to the vehicle. One of the first obligations placed on manufacturers was to stop the use of hazardous waste such as lead and mercury in vehicles made after July 2003. For the first time, the regulations place a requirement on vehicle manufacturers to take back vehicles at the end of their life via authorised treatment facilities (ATF), which would be charged with its environmentally sound recovery. The cost of treatment by the ATFs is to be borne for the large part by the producers rather than by the last keeper or owner and since the regulations were amended in 2005 there is a requirement that authorised treatment facilities are required to provide the service free of charge, effective from 1 January 2007 (where vehicles are largely complete and have not had extra waste added). The original regulations (End-of-Life Vehicle Regulations 2003) set out national targets for reuse and recycling as follows: • No later than 1 January 2006 the reuse and recovery rate for all ELVs will be at least 85% (average weight per vehicle and year). However, the rate for vehicles produced before 1 January 1980 was set at 75%. • By no later than 1 January 2015, all ELVs will need to have a reuse and recovery rate of no less than 95%.

2.11

Waste Electric and Electronic Equipment Regulations

The European Directive on WEEE (Directive 2002/96/EC) was implemented in the UK through the Waste Electrical and Electronic Equipment Regulations 2006 (SI 3289). The regulation finally became effective in August 2007. The regulations are implemented through the Environmental Permitting Regulations 2007 (SI 3538). The aim of the directive is to prevent and minimise WEEE and it puts a responsibility on producers and distributors to pay for the costs associated with the collection, treatment, recycling and recovery of WEEE. To ensure this takes place, the regulations require approved compliance schemes to be established which will finance the collection and treatment of WEEE. The WEEE regulations require all WEEE arising from the business sector to be collected for recycling, reuse and safe disposal. For domestic



35

household WEEE, the regulations set a collection target of 4 kg per head of the population per year. Since August 2005 producers of electrical and electronic goods have had an obligation imposed on them by Regulation 15 to mark their products with a crossed out wheelie bin symbol and a registration number from which the producer can be identified. Further obligations are placed on distributors to provide information to consumers on the environmental impact of the products they purchase. Regulation 10 requires producers to join an approved producer compliance scheme (PCS) which is overseen by the Department for Business, Enterprise and Regulatory Reform (BERR). No business is exempt from this, irrespective of its size. The scheme is financed by the producers who are members of the scheme and the levy is dependent on how much of a market share they have, i.e. the more electrical and electronic equipment they produce, the more they pay into the scheme. Distributors must also belong to and fund either a take-back scheme or must offer in-store take-back facilities for domestic consumers. An in-store scheme must allow consumers who are replacing equipment to bring back, free of charge, an old equivalent item even if it was not originally purchased from that distributor. If the distributor has joined a distributor take-back scheme then the consumer can take the WEEE to a designated collection facility (DCF) free of charge. Typically, DCFs will be located at locally run council waste sites. Amendments to the WEEE regulations came into force on 1 January 2008 at the end of the first compliance period which was 1 July to 31 December 2007 (WEEE Amendment Regulations 2007). The amendments include counting of whole appliances for reuse and providing for private households to return WEEE to the system free of charge even where they do not replace it with an equivalent. The Commission proposes to revise the WEEE Directive (URN 08/1516, 2008 and COM (2005) 667) again and overhaul the regulatory framework so that it is easier and cheaper to administer and enforce. In the UK the Environmental Permitting Regulations will be used to achieve this.

2.12

Conclusions

EU environmental policy and legislation have been developed and shaped by six environment action programmes for over more than 30 years. The first batch of environmental legislation put in place the framework for dealing with specific problems such as waste oil. This was built upon by legislation that developed standards for landfills and incinerators, for example, whilst the third batch of legislations introduced recycling directives which put in place the necessary organisation and finances to assist


36


the recycling of key waste flows such as end of life vehicles. Some lack of clarity in the wording of the directives meant that implementation among member states was variable and the current legislation seeks to harmonise and streamline environmental law making it standardised and more enforceable. The Commission is setting more and more ambitious environmental targets which will see impacts on the composites industry particularly as the mid-term review of the EAP 6 has singled out housing and transport as priority sectors. Both of these sectors use composite materials heavily and it is clear that future management of composite waste is set to become even more regulated. Whilst legislation is an important tool in achieving environmental targets, it is only part of the toolkit for achieving implementation alongside voluntary agreements and economic tools such as taxes for instance. At the core of all environmental legislation surrounding waste is the allimportant waste hierarchy which sets out the preferred routes for dealing with waste. In furtherance of this, all new waste legislation places the onus on producers to apply life-cycle thinking at the conception stage with a view to preventing waste in the first instance. Sitting alongside the legislation to achieve effective waste management are policy initiatives such as the integrated product policy which aim to reduce the environmental impact of products. Ultimately, however, if environmental degradation is to be halted, better use of resources and more efficient waste management will be necessary. The EU is addressing this through its sustainable development strategy and hence the priorities set here are likely to flow down into legislation in the future.

2.13

References

1st environment action programme 1973–1976. Official Journal of the European Communities, C 112, 20.12.1973 2nd environment action programme 1977–1981. Official Journal of the European Communities, C 139, 13.6.1977 3rd environment action programme 1982–1986. Official Journal of the European Union, C 46, 17.2.1983 4th environment action programme 1987–1992. Official Journal of the European Union, C 328, 7.12.87 5th environment action programme 1993–2000. Official Journal of the European Union, C 138, 17.5.1993 6th environment action programme 2001–2010. Official Journal of the European Union, L 242, 10.9.2002 com (2001) 68 – Green paper on Integrated Product Policy com (2005) 667 – Proposal for a directive of the European parliament and of the council on waste



37

com (2007) 225 – Communication from the Commission to the European Parliament, the Council and the European Economic and Social Committee and the Committee of the Regions on the Mid-term Review of the sixth Community Environment Action Programme com (2007) 856 – Proposal for regulation setting emission performance standards for passenger cars as part of the Community’s integrated approach to reduce CO2 emissions from light duty vehicles conroy a, halliwell s, reynolds t, ‘Composite recycling in the construction industry’, Composites: Part A, 37, 2006 council decision 2003/33/EC establishing criteria and procedures for the acceptance of waste into landfills. Official Journal of the European Communities, L 11/27, 16.1.2003 directive 1999/31/EC on the Landfill of Waste. Official Journal of the European Union, L 182/1, 16.7.1999 directive 2000/53/EC on End-of-Life Vehicles. Official Journal of the European Union, L 269/34, 21.10.2000 directive 2002/96/EC on Waste Electrical and Electronic Equipment 27 January 2003. Official Journal of the European Union, L 37/24, 13.2.2003 directive 2008/35/EC of the European Parliament and of the Council of 11 March 2008 amending Directive 2002/95/EC on the restriction of the use of certain hazardous substances in electrical and electronic equipment as regards the implementing powers conferred on the Commission. Official Journal of the European Union, L 81/67, 20.3.2008 directive 2008/98/EC on waste. Official Journal of the European Union, L 312, 22.11.2008 directive 75/439/EEC on waste oils. Official Journal, L 194, 25.7.1975 directive 75/442 EEC on waste. Official Journal, L 194, 25.7.1975 directive 91/689/EEC on hazardous waste. Official Journal, L 377, 31.12.1991 european communities act, 1972. http://www.opsi.gov.uk/acts/acts1972/ukpga_ 19720068_en_1 hawke n, Environmental Policy: Implementation and Enforcement, Ashgate 2002 ip/06/1233 – Environment: Commission pilots projects resulting in industry commitments to make greener products. http://europa.eu/rapid/pressReleasesAction.do? reference=IP/06/1233&format=HTML&aged=1&language=EN&guiLanguage =en ipr 43441 – European Parliament press release – MEPs and presidency reach deal on CO2 emissions from cars. http://www.europarl.europa.eu/news/expert/ infopress_page/064-43442-336-12-49-911-20081202IPR43441-01-12-2008-2008false/default_en.htm nemry f, leduc g, mongelli i and uihlein a, Environmental improvement of passenger cars (IMPRO – Car). EU Joint Research Centre, Scientific and Technical Report, JRC 40598, EUR 23038 EN 2008 regulation (EC) No 715/2007 of the European Parliament and of the Council of 20 June 2007 on type approval of motor vehicles with respect to emissions from light passenger and commercial vehicles (Euro 5 and Euro 6) and on access to vehicle repair and maintenance information. Official Journal of the European Union, L 171/1, 29.6.2007 statutory instrument 2006, No. 3289. The Waste Electrical and Electronic Equipment Regulations 2006. http://www.berr.gov.uk/files/file35992.pdf

38



statutory instrument No. 3538. The Environmental Permitting (England and Wales) Regulations 2007. http://www.opsi.gov.uk/si/si2007/uksi_20073538_en_1 the climate change act 2008. http://www.opsi.gov.uk/acts/acts2008/ukpga_ 20080027_en_1 treaty of amsterdam. Official Journal of the European Union, C 340, 10.11.1997 urn 08/1516, 2008. BERR report. WEEE Regulations – Government consultation on new regulations and further development of the supporting infrastructure to take effect from the 4th compliance period, Jan–31 December 2010. http://www. berr.gov.uk/files/file49460.pdf

3 Waste management R. S T E WA RT, c/o University of Warwick, UK


Abstract: EU Directive 75/442 EEC set out the early stages of the current waste management practices by introducing a waste hierarchy (reduce, reuse, recover) and a system of permits for those involved in waste treatment with the conditions for obtaining permits and the requirements placed on waste disposal facilities being strictly regulated (e.g. Directive 96/61/EC, SI 1056). Member states have a further requirement to produce a waste management plan and introduce waste prevention programmes and the 1975 directive also established the ‘polluter pays’ principle. In 1994 this was strengthened through legislation (Directive 94/62/EC) which placed a responsibility on the producer of the waste to bear the costs of collection, sorting or treatment and recycling or recovery. Economic instruments were employed in 1996 through a landfill tax (SI 1527) which was designed to divert waste going to landfill and promote the options set out in the waste hierarchy. Later recycling directives dealing with end-of-life vehicles (Directive 2000/53/EC) and waste electrical and electronic equipment (Directive 2002/96/EC), for example, set out more detail about how waste management facilities should operate and report on their activities. A revised Waste Framework Directive (COM (2005) 667) was proposed in 2005 to harmonise the legislation and in the UK it is being implemented through the Environmental Permitting Regulations (SI No. 3538) which also brings together the various conditions placed on waste operators. In addition to the legislative approaches to waste management already outlined, there are EU (EAP 6) and UK strategies on waste which detail plans on waste management and how the member state can reach targets set out in legislation such as levels of recycling or waste diverted from landfill. There are also public policy tools like the integrated product policy (IPP) which encourages producers to take a life-cycle approach to waste. Key words: waste, waste management, WEEE, ELV, labels, environmental permits, waste hierarchy.

3.1

Introduction

In 1972 the European Community determined that a community environmental policy was needed and waste management would be central to it. The framework which developed and shaped environmental legislation and policy came in the form of a series of Environment Action Programmes (EAP), of which there are six to date (EAP1 1973, EAP2 1977, EAP3 1983, EAP4 1987, EAP5 1993, EAP6 2002). Each programme is specific to a particular environmental issue and spans a 5–10 year time frame. The 39

40


current programme in operation is EAP 6 which identifies waste as one of its priority areas. The programme is thematically based with one specific theme being devoted to waste prevention and recycling. The year on year increase in waste generation has resulted in the European Union (EU) consuming natural resources at an unsustainable rate and therefore managing the remaining resources is paramount. Managing waste is a key feature in the EU strategy for sustainable development (COM (2001) 264) and this chapter will look at the requirements and obligations placed on those involved in generating, collecting, treating and disposing of waste. The waste management procedures most relevant to composites will be explored in more detail.


3.2

The Waste Framework Directive

The Waste Directive has been amended a number of times since its introduction in 1975 (Directive 75/442 EEC) and most recently in 2008 (Directive 2008/98/EC) which gives member states until 2010 to transpose it into domestic law. The founding directive from 1975 established some of the key criteria on waste management and these have been built on progressively. Article 3 of the directive established the ‘waste hierarchy’ which is a list of options for dealing with waste in order of preference. This hierarchy has been pivotal in shaping waste management strategy, practice and environmental law for the last three decades. The waste hierarchy consists of: • • • •

prevent waste from being generated in the first place; reduce the quantities of waste; recycle and reuse waste; energy recovery.

The directive also required member states to set up ‘competent authorities’ to be responsible for the planning, organisation, authorisation and supervision of waste disposal operations. In the UK this function is now performed by the Environment Agency. Once established, the authorities were required to make waste management plans which set out the type and quantity of waste for disposal as well as associated costs and technical requirements for doing so. The competent authorities were given responsibility for granting permits to any undertaking or holder processing waste. To obtain a permit, the waste operator had to demonstrate how the waste would be disposed of without causing harm to human life or the environment. This obligation meant providing technical details and safety precautions taken to protect human health and the environment but also a requirement to provide the competent authority with information on the origin, destination, treatment, type and quantity of the waste. To ensure

Waste management

41

waste operators complied with the criteria detailed in the permit they would be subject to inspection by the authority. In turn, the competent authority would have to report back to the EU Commission on how it was implementing the directive. The burden of costs associated with disposing of waste were placed upon the waste producer and the ‘polluter pays’ principle was firmly established, the more waste a producer created, the more they would pay to have it disposed of. An example of this is the Landfill Tax Regulations (SI 1527) which will be discussed later in the chapter. In 2006 the founding directive was amended (Directive 2006/12/EC), putting in place the essential requirements for the management of waste and adding a definition of waste as well as adding an aspiration for the EU as a whole to become self sufficient in waste disposal, but creating a specific aim for member states to achieve this. The latest amendment to the Waste Framework Directive took place in 2008 (Directive 2008/98/EC) and the following areas were addressed:


•

• • • • •

The introduction of an environmental objective, focusing specifically on reducing the environmental impact of waste and the life cycle of waste; Clarification on recovery and disposal; Clarification of the conditions for mixing of hazardous waste; Introduction of a procedure which clarifies when waste ceases to be waste for a particular waste stream; Introduction of minimum standards for a number of waste management operations; Introduction of a requirement to develop national waste prevention programmes.

The language used in the directive has been harmonised with other directives and it harnesses product life-cycle thinking by encouraging producers to prevent waste generation in the first instance for example. The 1975 directive set out a waste management hierarchy with the first priority being to prevent waste in the first instance, then to recycle and reuse. This hierarchy is the bedrock of sustainable waste management and has been embedded into legislation. The new proposals reinforce the waste hierarchy of prevention, reuse, recycling, recovery (including by energy recovery) and safe disposal. Not only is there a requirement for member states to produce waste management plans but a new requirement for waste prevention plans has been introduced. A report by the Department for the Environment, Food and Rural Affairs (Defra) sets out when each obligation has to be met by the UK (http://www.defra.gov.uk/environment/waste/thematicstrat/ revised-wfd-transposition-implementation-dates.pdf). Both the Disposal of Waste Oil Directive (Directive 75/439/EEC) and the Hazardous Waste


42


Directive (Directive 91/689/EEC) have been subsumed into the Waste Framework Directive (Directive 2008/98/EC) in order to simplify the legislation. In terms of the requirement for a waste operator to have a permit issued by the competent authority, each revision of the Waste Framework Directive has added details to this requirement and this has been supplemented by other supporting legislation such as the Waste Management Licensing Regulations (SI 1056). The 2008 amendment to the directive goes further than previous revisions by introducing minimum standards for permits designed to ensure that waste is treated in an environmentally sound manner. The Waste Framework Directive (Directive 2008/98/EC) defines waste as ‘any substance or object which the holder discards or intends or is required to discard’ and management as ‘the collection, transport, recovery and disposal of waste, including the supervision of such operations and after-care of disposal sites’. These definitions mean that environmental rules for dealing with waste apply not only to waste disposal facilities themselves but also to recovery and recycling operations as well as to waste brokers, dealers and carriers. A waste broker is a third party who makes arrangements for others to handle, transport, dispose of or recover waste rather than doing it themselves and a waste dealer is a type of broker who sells on the waste. The Environmental Protection Act (1990) also imposes a Duty of Care on the handling of waste, which affects anyone who: • • • • • • •

produces; imports; stores; transports; treats; recycles; or disposes of controlled waste.

Under the Duty of Care, businesses producing, carrying, storing or treating waste must meet guidelines to ensure that waste management facilities do not pollute the environment or cause harm to human health. The primary onus of responsibility for the handling of waste is on the waste producer, who must ensure that waste is stored securely and transferred to an ‘authorised’ person only.

3.2.1 Licensing of waste operations To comply with the permitting requirements of the Waste Framework Directive previously discussed, the UK implemented the Waste

Waste management

43

Management Licensing Regulations (WMLR) in 1994 (SI 1056 as amended). The Waste Management Licensing Regulations set out the procedure for obtaining a licence and also identified a number of activities that are excluded from requiring a licence or that are exempt from licensing. If an activity is exempt it probably needs to be registered by the competent authority, i.e. the Environment Agency in the UK. The types of waste treatment activities that required a licence include: landfilling; disposal of waste using plant equipment (incineration, shredding, sorting); • treatment, keeping or disposal of controlled waste on land; • treatment, keeping or disposal of controlled waste by mobile plant; • treatment, keeping or disposal of controlled wastes in a manner likely to cause detriment to the environment or human health.


• •

In order to be granted a licence certain criteria have to be met such as demonstrating technical competence underpinned by a formal training certificate in waste management. Applicants also have to demonstrate evidence of financial provisions for the activities covered by the permit and adequate security, typically in the form of insurance in case of seepages into the water course, for example, but also a requirement for the waste site itself to be secure from the public. However, the Waste Management and Licensing Regulations have now been superseded by the Environmental Permitting Regulations (SI No. 3538). It is not only waste management operations that require a permit to operate but also the installations themselves and this is detailed in the Directive on Integrated Pollution Prevention and Control (IPPC), (Directive 96/61/EC). The directive has been amended a number of times and as a result was codified in 2008 (Directive 2008/1/EC). Codification means bringing all amendments to a given law adopted at different times into one law. The IPPC Directive establishes an approach for pollution prevention from stationary ‘installations’ involved in industrial and agricultural processes. The purpose of the directive is to achieve a high level of environmental protection using measures to prevent or, where that is not practicable, to reduce emissions to air, water and land from industrial processes. An installation is classed as a stationary technical unit. The IPPC Directive is based on a number of core principles which are an integrated approach, use of best available techniques, flexibility and public participation. The directive requires industrial and agricultural activities to have a permit which can be granted only if certain environmental conditions are met, meaning that the industries themselves take responsibility for

44


preventing and reducing any pollution they cause. The types of industry covered include the waste management industry. In order to receive a permit, an industrial or agricultural installation must comply with certain obligations such as: • use the best available techniques (BAT) as pollution-prevention measures; • prevent all large-scale pollution; • prevent, recycle or dispose of waste in the least polluting way possible; • use energy efficiently; • prevent accidents and limit the consequences if they do happen; • restore sites to their original state when the activity is over.


Those applying for a permit have to give information on how they will implement the obligations outlined above such as the proposed waste management techniques and how emissions will be monitored. Like the Waste Management and Licensing Regulations, IPPC is implemented in the UK under the Environmental Permitting Regulations 2007.

3.3

Environmental Permitting Regulations

The Environmental Permitting (EP) Regulations 2007 came into force in April 2008. A series of guidance documents has also been issued to accompany the regulations along with a core guidance document (http://www.defra.gov.uk/environment/epp/). This regime streamlines and combines waste management licensing (WML) and pollution prevention and control (PPC) to create a single environmental permit with a common approach to permit applications, maintenance, surrender and enforcement. The EP Regulations bring together a number of pollution control legislations into a single set of regulations and replaces around 40 pieces of individual environmental legislation. A number of environmental directives such as the Waste Electrical and Electronic Equipment Directive (Directive 2002/96/EC) and the End-of-Life Vehicle Directive (Directive 2000/53/ EC) are set out in separate guidance documents which outline measures that are specific to these waste streams. These will be considered in more detail later in the chapter. The EP Regulations set out the generic and specific terms that must be adhered to, along with the reporting, inspection and enforcement procedures. The effect of the regulations is to harmonise environmental permits and reduce bureaucracy especially where more than one waste facility operates from a single site.


Waste management

45

Waste activities are classed as low, medium and high risk and where the risks are low the EP Regulations commit the operator to complying with a standard package of rules. These standard rules define how a waste operator must carry out particular activities such as limiting the types of waste that can be brought on to a site. Low to medium risk activities would include waste transfer stations and civic amenity sites. Activities involving waste can be undertaken only by competent persons and under the previous Waste Management and Licensing Regulations (SI 1056) operators had to demonstrate this by obtaining a Certificate of Technical Competence. However, the EP Regulations have identified that different types of waste facility present different levels of environmental risk and therefore allow for different levels of competency demonstrations through an industry-approved scheme, i.e. proportional to the level of risk presented. There is the requirement that whatever the level of competency attained, it must be continual in nature and therefore there is an ongoing two yearly assessment. The Waste Management Licensing Regulations (SI 1056) set out a number of exemptions for the requirement to have a waste management licence. The EP Regulations have made changes to the list of exemptions and a number of operations that did not previously require a permit will now do so under the new scheme; these are detailed in schedule 3 of the regulations. An exemption allows certain waste management operations to be carried out without needing an environmental permit and exemptions are designed to provide a lower level of regulation for low risk activities. Under the new proposals businesses may need to renew and re-register exemptions more frequently to ensure waste registers are up to date. It is proposed that there will be a charge for the registration of all exempt waste operations and a requirement for exemptions to be re-registered every three years. This is to cover the cost to the Environment Agency of regulating the exemptions. The public consultation closed in October 2008 and any proposed changes will be brought on stream in 2009 as set out in a review document issued by the Environment Agency (http://www. environment-agency.gov.uk/business/topics/waste/32158.aspx). Part 4 of the EP Regulations deals with enforcement and this can consist of the issuing of an enforcement notice to those who contravene the conditions set out in their permit or a suspension notice if there is a risk of serious pollution. Suspension and enforcement notices set out the reasons behind issuing the notice, what steps must be taken to remedy the breach of the environmental permit and dealing with the pollution and a timeframe for complying. Breaches of the environmental permit can lead to criminal convictions and depending on the offence, a person can be fined up to £50 000 or imprisoned for up to 12 months.

46


The second phase of the environmental permitting programme will extend the regulations to include: • • • • •

discharge consenting; groundwater authorisations; water abstraction and impoundment; radioactive substances regulation; licensing of some waste carriers and brokers.

Subject to public consultation, phase 2 will be rolled out between 2008 and 2011 (http://www.defra.gov.uk/environment/epp/deliver.htm#2).

3.3.1 Activities requiring an environmental permit


In line with the previous legislation the following activities require an environmental permit: • Installations carrying out activities as set out in schedule 1 of the EP Regulations. These will include activities from the energy sector such as refining, production and processing of metals, surface treatment of metals and plastics, production of cement and lime, mineral fibre production, ceramics, and chemicals. Waste management is also covered by the regulations and EP permits are also required by these activities, and include incineration (and co-incineration), disposal of waste by landfill, recovery of waste, fuel from waste. An installation is any stationary technical unit which carries out one or more of the activities listed in schedule 1. • Waste operations. Any disposal or recovery of waste which is not exempt in the EP Regulations. The list of exempt operations has changed under the EP Regulations compared to the WMLR. • Mobile plant carrying out the activities or waste operations as above. By definition a mobile plant is not stationary and therefore can not be described as a facility.

3.3.2 Landfill Directive The EP Regulations provide separate guidance on specific waste streams which detail additional measures that must be complied with. The Landfill Directive (Directive 1999/31/EC) is an example of this; it was transposed into English law through the Landfill Regulations 2002 (SI 1559). Landfill is defined as ‘waste disposal site for the deposit of waste onto or into land’ and the overall objective of the regulations is to prevent or reduce the negative effects of landfilling on the environment as well as any risk to

Waste management

47

human health posed as a result. The directive also requires member states to implement a strategy on biodegradable waste that provides for the progressive diversion of biodegradable municipal waste from landfill in order for the UK to meet challenging targets: By 2010 to reduce biodegradable municipal waste landfilled to 75% of that produced in 1995; • By 2013 to reduce biodegradable municipal waste landfilled to 50% of that produced in 1995; • By 2020 to reduce biodegradable municipal waste landfilled to 35% of that produced in 1995.


•

The regulations attempt to achieve this through specifying technical standards of landfill sites and operations and setting out requirements for the location, conditioning, management, control, closure and preventative and protective measures. A timetable has been set for existing sites to be brought up to the standards or they must be closed. UK landfill sites must be licensed by the Environment Agency and require a permit to be able to deal with waste of a particular type. The Landfill Directive supplements the requirements of the IPPC Directive which lists non-inert landfill as an activity that requires sites above a certain size to be permitted (Directive 96/61/EC). This is now achieved through the Environmental Permitting Regulations in the UK (SI 3538). There are three categories of landfill sites and they may only be licensed for a particular type of disposal: • hazardous waste as set out in the Hazardous Waste Directive (Directive 91/689/EEC); • non-hazardous waste; • inert waste (waste that does not undergo significant change). Landfill operators must demonstrate that measures are in place to prevent cross-contamination between differently classified sites where they are near to each other. A ban on the practice of co-disposal of hazardous waste and nonhazardous waste has been in place since 2004. From October 2007 there has been a requirement to treat all non-hazardous waste before it is landfill except if treatment is not technically feasible or where treatment would not reduce the quantity of the waste or its environmental impact. The regulations define treatment as ‘ “treatment” means physical, thermal, chemical or biological processes (including sorting) that change the characteristics of waste in order to reduce its volume or hazardous nature, facilitate its handling or enhance recovery.’ The proposed treatment must conform to three set criteria:

48



• it must be a physical, thermal or chemical treatment of a biological process which includes sorting; • it changes the waste’s characteristics; and • it does so to reduce its volume or hazardous nature, facilitates handling or enhances recovery. In order to be granted and maintain a permit, a landfill operator must comply with strict reporting and inspection procedures. The permit will set down conditions that have to be met by the landfill operator such as steps that must be taken by that operator to ensure they only accept waste that has been treated or does not need to be treated. The operator is restricted in what types of waste it may accept into the landfill, for example, liquid, tyres, explosive materials and hospital waste are banned but it is the responsibility of the waste holder to ensure it is correctly identified using the List of Wastes Regulations (SI 895) codes. Any waste going to a landfill site must comply with the waste acceptance criteria set out in the directive and the Council Decision of 2003 (CD 2003/33/EC). These are procedures that operators must follow and cover basic characterisation of the waste (source, origin, appearance, list of waste code), compliance testing and on-site verification of the original characterisation. The waste acceptance criteria are set out for each class of landfill. The operator must also show evidence of how it intends to comply with all the sections of the directive including the waste acceptance criteria but also the obligations on record keeping and monitoring and a detailed environmental risk assessment. The cost that the landfill operator charges must reflect the costs of operating the site but also need to include an element of the costs likely to be incurred by closure of the site after 30 years in service.

Landfill Allowances and Trading Scheme So that the UK could meet the targets for diverting municipal waste away from landfills set out in the Landfill Directive, the Waste Emissions and Trading Act was passed by Parliament in 2003. The act puts an obligation on Waste Disposal Authorities (WDA) to reduce the amount of biodegradable municipal waste (BMW) it landfills. Each WDA is given a tradable allowance and they are allowed to landfill up to the number of allowances they have. One allowance is equal to one tonne. The criteria for trading are set out in the Landfill Allowances and Trading Scheme Regulations 2004 (SI 3212). This was the world’s first trading scheme for municipal waste allowances and the scheme will operate between 2005/6 and 2019/20. Each WDA is given an annual figure for the amount of BMW it is allowed to landfill for that year. Under the Landfill Allowances and Trading

Waste management

49

Scheme (LATS) the WDA is able to use the allowance flexibly and may: • • •

save or bank allowances to use in future years; buy and sell them; borrow up to 5% from the next year’s allocation.

A searchable register of LATS activity has been set up to facilitate trade and there are clear reporting and monitoring requirements.


Landfill tax Part of the toolkit of waste management and environmental legislation is the use of economic instruments and the landfill tax is an example of this which is implemented by the Landfill Tax Regulations (SI 1527) and the Landfill Tax (Qualifying Material) Order (SI 1528). The tax is regulated by HM Revenue and Customs. The Landfill Tax Regulations were amended in 2008 (SI 770). Landfill tax is a tax on the disposal of waste and it aims to encourage waste producers to produce less waste, recover more value from waste, for example through recycling or composting, and to use more environmentally friendly methods of waste disposal. Landfill tax applies to all waste: • • • •

disposed of by way of landfill; at a licensed landfill site; on or after 1 October 1996; unless the waste is specifically exempt.

The site falls within the scope of the tax if there is a licence or permit authorising disposals. Liability for tax on disposal rests with the licence holder of a landfill site and any waste deposited in it, including waste that the site itself produced, will be liable to tax. The tax is chargeable by weight and there are two rates, a standard rate for active waste and a lower rate for inactive waste (rocks and soil for example). Rates for 2008/09 are: • •

Active waste – £32/tonne (+ VAT) Inactive waste – £2.50/tonne (+ VAT).

The rate for active waste increased by £8/tonne per annum from 1 April 2008 and will remain at this level until April 2010 when it will increase to £48 + VAT per tonne. The rate for inactive waste increased to £2.50 + VAT per tonne on 1 April 2008. The Government has announced that the rate will be frozen at £2.50 per tonne in 2009–10. The rates were set out in the UK’s 2008 Finance Bill. Landfill operators can apply to have tax-free status for any part of a site that is not used for landfill, but is instead used for;

50 • • • • •

Management, recycling and reuse of waste composites recycling waste; incinerating waste; sorting waste pending its disposal or transfer; storing inactive waste for use in landfill restoration; sorting material to obtain the above.

Tax credits are available for waste that landfill operators send: • •

for recycling, incineration or reuse; to another landfill site.

The rules for doing so have been changed with the implementation of the Landfill Tax (Amendment) Regulations (SI 770) which limits the maximum credit a landfill site operator can claim, in respect of qualifying contributions made against their annual landfill tax liability to 6.0% from the previously higher level of 6.6%.


3.4

End-of-Life Vehicle Directive

The End-of-Life Vehicle (ELV) Directive (Directive 2000/53/EC) was transposed into UK law in 2003 by the End-of-Life Vehicle Regulations (SI 2635) and is now regulated through the EP Regulations 2007 (SI 3538). The EP Regulations widen out the scope of the ELV Directive by applying to all end-of-life vehicles rather than just to those listed in Annex IIA of the directive. Hence, in the UK the regulations apply to commercial vehicles such as buses and coaches. The first focus of the ELV legislation is to prevent waste in the first place and then to consider reuse and recycling as well as other forms of energy recovery. The scope of the legislation is not just contained to the vehicle itself but extends to its components and materials, irrespective of whether or not they were factory fitted to the vehicle. If the regulations can discourage waste being created, it is anticipated that they will soon force manufacturers to design not just for disassembly but with the whole life cycle in mind. In terms of preventing waste, one of the first obligations placed on manufacturers was to stop the use of hazardous waste such as lead and mercury in vehicles made after July 2003. Under the reporting requirements of the regulations manufacturers must publish reports to consumers on: •

the design of vehicles and their components with a view to their recoverability and recyclability; • the environmentally sound treatment of end-of-life vehicles in particular the removal of all fluids and dismantling; • the development and optimisation of ways to reuse, recycle and recover end-of-life vehicles and their components; • the progress achieved with regard to recovery and recycling to reduce the waste to be disposed of and to increase the recovery and recycling rates.

Waste management

51

For the first time, the regulations place a requirement on vehicle manufacturers to take back vehicles at the end of their life via authorised treatment facilities (ATF) who would be charged with its environmentally sound recovery. Typically it is the ATF that would issue the last keeper with a certificate of destruction allowing it to be deregistered but there is also scope within the regulations for dealers or producers to be authorised, providing they could prove the vehicle was indeed transferred to the ATF. The cost of treatment by the ATFs is to be borne for the large part by the producers rather than by the last keeper or owner and since the regulations were amended in 2005 there is a requirement that ATFs are required to provide the service free of charge, effective from 1 January 2007 (where vehicles are largely complete and have not had extra waste added). The original Regulations (End-of-Life Vehicle Regulations 2003 SI 2635) set out national targets for reuse and recycling as follows:


• No later than 1 January 2006 the reuse and recovery rate for all ELVs will be at least 85% (average weight per vehicle and year). However, the rate for vehicles produced before 1 January 1980 was set at 75%. • By no later than 1 January 2015, all ELVs will need to have a reuse and recovery rate of no less than 95%. In order to assist ATFs and to meet national targets for reusing and recovering components, producers are charged under the regulations to label each component according to a set material and component coding standard so that it can be readily identified. Further, manufacturers must provide sufficient instructions or guidance so that assemblies and subassemblies can be efficiently dismantled.

3.5

Waste Electric and Electronic Equipment Regulations

The European Directive on Waste Electric and Electronic Equipment (WEEE) (Directive 2002/96/EC) was implemented in the UK through the Waste Electrical and Electronic Equipment Regulations 2006 (SI 3289) although it should have been implemented earlier in 2005. The regulation finally became effective in August 2007. The regulations are now implemented in the UK through the Environmental Permitting Regulations (SI 3538). The aim of the directive is to prevent and minimise WEEE and it puts a responsibility on producers and distributors to pay for the costs associated with the collection, treatment, recycling and recovery of waste electric and electronic equipment. To ensure this takes place, the regulations require approved compliance schemes to be established which will finance the collection and treatment of WEEE.

52


The WEEE regulations require all WEEE arising from the business sector to be collected for recycling, reuse and safe disposal. For domestic household WEEE, the regulations set a collection target of 4 kg per head of the population per year. There are 10 categories of WEEE covered by the regulations along with set national targets for each category;


• • • • • • • • • •

large household appliances; small household appliances; IT and telecommunications equipment; consumer equipment; lighting equipment; electrical and electronic tools; toys, leisure and sports equipment; medical devices; monitoring and control instruments; automatic dispensers.

Schedule 2 of the directive gives many specific examples of the types of products included and these range from toasters to dialysis machines. Since August 2005 producers of these goods have had an obligation imposed on them by Regulation 15 to mark their products with a crossed out wheelie bin symbol and a registration number from which the producer can be identified. Further obligations are placed on distributors to provide information to consumers on the environmental impact of the products they purchase.

3.5.1 Approved schemes Regulation 10 requires producers to join an approved producer compliance scheme (PCS) which is overseen by the Department for Business, Enterprise and Regulatory Reform (BERR). No business is exempt from this, irrespective of its size. The scheme is financed by the producers who are members of the scheme and the levy is dependent on how much of a market share they have, i.e. the more electrical and electronic equipment they produce, the more they pay into the scheme. The operators of the scheme are responsible for implementing systems that enable reuse of parts as well as the best methods of treatment, recycling and recovery. The recording and reporting of these activities are stringent but are intended as a means of monitoring standards and ensuring the UK achieves the targets set out in the regulations. Reports need to be made on an ongoing basis for each compliance period, of which there are currently three between 1 July 2007 and 31 December 2009. As with producers, distributors must also belong to and fund either a distributor take-back scheme or must offer in-store take-back facilities for


Waste management

53

domestic consumers. An in-store scheme must allow consumers who are replacing equipment to bring back, free of charge, an old equivalent item even if it was not originally purchased from that distributor. If the distributor has joined a distributor take-back scheme then the consumer can take the WEEE to a designated collection facility (DCF) free of charge. Typically, DCFs will be located at local run council waste sites. Even those who import electrical and electronic equipment (EEE) into the UK are affected by the regulations as they are covered by the definition of a producer. It is for importers and manufacturers to determine which will hold responsibility as the registered producer for the purposes of the regulations, but according to Flynn (2007), it is likely that the Environment Agency will deem the importer responsible where the manufacturer is not UK based. WEEE can only be processed by an approved authorised treatment facility (AATF) that has the correct type of waste management licence. The processes and procedures used by the ATF to sort, dismantle and treat WEEE are also closely regulated and are administered by the Environment Agency. The regulations set out best practice that must be followed in processing techniques. Post-processing materials such as plastic and metal can then be transferred to a reprocessor and, in turn, be put back into the recycling chain as new products.

3.5.2 Integrated product policy Integrated product policy (IPP) is a public policy initiative which has been on the EU agenda since roughly the late 1990s and is concerned with the reduction of environmental impact associated with products and services. The purpose of the EC initiative was to harmonise varying environmental product policy strategies that were developing within the Community to minimise the environmental impact of their products at varying stages of the product life cycle: for example, take-back schemes, product labelling, taxes or other economic initiatives. These differing approaches among member states led to a call for harmonisation of such measures across the EU in order to avoid problems that might arise from different policy approaches. Consequently, an IPP green paper was published promoting a common framework within the EU (COM (2001) 68). IPP requires all, from government and other major stakeholders through to the supply chain and the consumer, to be mindful of every stage of the products life cycle in order to promote sustainability. All products cause some form of environmental degradation, be that during manufacturing, in use or during their disposal. When considering the life cycle of a product, account needs to be taken of potential environmental harm through all the areas of the cycle right from the extraction of


54


natural resources, through the design process, manufacturing and assembly stages and on to marketing, distribution, sale and use and finally to their eventual disposal as waste. Clearly, depending on the product there may be a highly complex mix of contributors through the supply and demand chain during the life-cycle process and IPP seeks to place an onus on each to think about and improve their individual environmental performance. The IPP applies a number of techniques to encourage compliance and monitor the objectives set out in the policy, for example economic instruments such as taxes and subsidies, substance bans, voluntary agreements, environmental labelling and product design guidelines. The European Commission published a green paper on IPP in 2001 (COM (2001) 68) which set out the rationale behind developing the IPP approach to environmental policies and set out possible mechanisms that might be used to achieve ‘greener’ products. Two pilot projects were concluded in 2006 which focused on a mobile telephone and a tropical garden chair as example products (IP/06/1233). The work of the Commission has gone on to identify further products that can benefit from environmental improvement which includes passenger cars (Nemry et al. 2008). There is currently no specific legislation governing IPP but it is undoubtedly going to feature much more prominently in the future which was echoed in the mid-term report of the sixth EAP. IPP is, however, already being incorporated into other legislation and it is certain to do so further as regulations are amended and further regulation should not be ruled out. For example, the End-of-Life Vehicle Directive (Directive 2000/53/EC) and the Waste Electrical and Electronic Equipment Directive (Directive 2002/96/EC) can be considered to be examples of IPP. Both directives place a requirement on producers to prevent waste in the first instance through the design of their products and with a view to their recyclability.

3.5.3 Sustainable development One of the objectives of the sixth EAP which implements the EU waste strategy was to decouple economic growth from environmental degradation and to bring about a situation where resources were used more efficiently and waste management was improved so that more sustainable patterns in production and consumption were established. This philosophy is aligned with the EU strategy on sustainable development which was published in 2001 (COM (2001) 264) and updated in 2005 (COM (2005) 658). The strategy is based on seven key challenges, and most relevant to this discussion are: • •

to limit climate change; to limit the adverse effects of transport;

Waste management

55

• to promote more sustainable modes of production and consumption and breaking the link between economic growth and environmental degradation; • more responsible management of natural resources. A progress report on the sustainable development strategy in 2007 (COM (2007) 642) indicated that positive progress had been made towards reducing greenhouse gas emission, for example, through the reduction in vehicle emissions but that the UK was still not on a path towards sustainable transport, with demand for transport growing at 1.3% per year. The rate of sustainable products and services is on the rise but the Commission intends to take further action including adopting an IPP approach and supporting research. Member states are reported to be updating national strategies to align with the EU sustainable development strategy and hence the priorities set here are likely to flow down into legislation in the future.


3.6

Classification and labelling of waste

The Environmental Protection Act (1990) sets out the duty of care obligations (DoC) for those who handle waste and it requires that any waste be properly described. Accurate descriptions of waste are at the core of waste management and any movement of waste needs to be accompanied by a waste transfer note that provides details about the waste according to a standardised code. Other types of labelling are used to assist in recycling and disassembly operations or to provide the consumer with information about the environmental impact of the products they purchase or use.

3.6.1 European Waste Catalogue The descriptions of waste are standardised by the European Waste Catalogue (COM 94/3/EC and COM 2000/532/EC) which enables waste to be recorded, monitored and controlled across Europe. The Catalogue is a standard set of descriptions and 6 digit codes for waste streams from each industry sector encompassing hazardous and non-hazardous waste. The European Waste Catalogue (EWC) has been transposed into domestic law by the List of Wastes Regulations (SI 895) and the Hazardous Waste Regulations (SI 894). The regulations are set out in chapters 1 to 20 depending on the industry or process that generated the waste. For example, chapter 19 deals with wastes from waste management facilities, off-site waste water treatment plants and the preparation of water intended for human consumption and water for industrial use and chapter 20 deals with municipal wastes. Each chapter has a sub-chapter code and within each

56


sub-chapter is a code for a specific waste stream. If the waste is hazardous it is marked with an asterisk. To assign the correct code to the waste it will be necessary to know: • • • • • •

the type of industry that produced the waste; what activity resulted in the waste being produced; a description of the waste; the constituents of the waste; the concentration of any dangerous substances; if there are any hazards associated with the waste.


The chapters of the list are set out according to business types but certain industries such as car manufacturing use a number of different processes and will need to code their waste based on multiple chapter headings. As an example, waste carbon fibres would have a waste code of 04 02 21 which is derived from chapter 04 wastes from the leather, fur and textile industries, sub-chapter 02 wastes from the textile industry, followed by 21 which is the code for the specific waste stream, i.e. wastes from unprocessed textile fibres.

3.6.2 Recycling labels Both legislation and increasing environmental awareness have forced through the requirement for increased waste prevention and minimisation as well as life-cycle thinking. In order to improve the processes of recycling and reuse it is crucial to clearly identify waste and waste streams. In many instances producers are responsible for taking back products at the end of their lives through legislative tools such as the Producer Responsibility Obligations (Packaging Waste) Regulations 1997 (SI 648), so it is in industry’s interests to implement easier methods of identification and sorting to meet ever more ambitious targets. For example, the 2008 amending regulations (SI 413) require 81% of glass packaging to be recycled by 2010. As a result, the use of recycling symbols is common but there is not a systematic regulated approach bar a few exceptions such as the wheelie bin sign required by the WEEE directive (SI 3289). The symbols typically fall into three categories that will signify if the material can be recycled, show what the material is and show whether or not it is compliant with a recognised compliance scheme. Environmental claims about products are addressed through ISO 14021 which attempts to harmonise information on self-declared environmental or ‘green’ claims. The standard outlines conditions for using certain selfdeclared claims such as whether or not an item is recyclable and the use of symbols and logos such as the Mobius loop.

Waste management

57

Recyclability symbols Probably the most familiar symbol that denotes if an item can be recycled is the Mobius loop shown in Fig. 3.1. The symbol is intended for use only on goods that are capable of being recycled, if a facility exists, but it has no precise meaning and is often misused. The Mobius loop with a percentage sign indicates that the product contains the given percentage of recycled content which is more meaningful than the empty loop. Material-specific symbols


There are a number of labels in common use to identify what a product is made of and some also denote if it is recyclable. Examples of this are given in Fig. 3.2 for glass, steel and aluminium. The composition of plastic materials are described by a series of numbered Mobius loop symbols developed by the Society of the Plastics Industry (SPI) and are known as resin identification codes. Usually below the symbol is a shorthand version of the name of the resin the item was made from as shown in Fig. 3.3. Eco-label scheme The eco-label was introduced to the European Union in 1992 (Regulation EEC 889/92) and seeks to encourage businesses to provide goods and services that have a reduced environmental impact. It is a market-based initiative and its function is to stimulate the supply and demand of green and

X%

3.1 The Mobius loop.

alu RECYCLABLE STEEL

3.2 Recycling and material identification symbols for glass, steel and aluminium.

58


1 PETE

Polyethylene terephthalate

2 HDPE

High density polyethylene

3 V

Polyvinyl chloride


4 LDPE

Low density polyethylene

5 PP

Polypropylene

6 PS

Polystyrene

7 OTHER Any other plastic 3.3 Resin identification codes. Note: PETE, V and OTHER can also be known as PET, PVC and O, respectively.

Waste management

59


3.4 European eco-label symbol.

environmentally friendly products that are licensed by the scheme. The scheme also allows the consumer to take into consideration the environmental impact a product has before they choose to buy it. The eco-label scheme is therefore putting into practice the IPP approach and using lifecycle thinking to reduce the environmental impact of goods. Licensed products are permitted to use the flower symbol shown in Fig. 3.4. The product types and specific criteria are set out in the regulation and the amending regulation (Regulation EC 1980/2000). The scheme is open to any product and service, except food, drink, pharmaceuticals and medical devices. Furthermore, the flower cannot be awarded to substances that are toxic, carcinogenic or dangerous to the environment or to goods that could be harmful to the consumer in their normal use.

3.7

Conclusions

Waste management in the UK is framed by European Union and domestic environmental strategies, direct regulation, environmental taxes, trading schemes as well as self regulation and voluntary agreements all of which are necessary to bring about sustainability. The waste hierarchy set out in the Waste Framework Directive and EU environmental strategy ultimately determines the economics and practices in waste management through a number of routes including: • • • • •

preventing waste being produced; decoupling economic growth from waste generation; recovering resources from waste; only landfilling waste for which there is no market; setting of progressively stricter emissions levels;

60 • • •

Management, recycling and reuse of waste composites setting up networks of appropriate waste management facilities; generating markets for reused, recycled and recovered products; waste resource trading and treatment internationally.

Current and future environmental policy is likely to put ever-increasing targets and challenges on waste management and in response better and more efficient regulation has come on stream to assist the waste management industry. There is also an onus on those dealing with waste at whatever point in its life cycle to use the best available techniques.


3.8

References

1st environment action programme 1973–1976. Official Journal of the European Communities, C 112, 20.12.1973 2nd environment action programme 1977–1981. Official Journal of the European Communities, C 139, 13.6.1977 3rd environment action programme 1982–1986. Official Journal of the European Union, C 46, 17.2.1983 4th environment action programme 1987–1992. Official Journal of the European Union, C 328, 7.12.87 5th environment action programme 1993–2000. Official Journal of the European Union, C 138, 17.5.1993 6th environment action programme 2001–2010. Official Journal of the European Union, L 242, 10.9.2002 com (2001) 68 – Green paper on Integrated Product Policy com (2001) 264 – A Sustainable Europe for a Better World – A European Union Strategy for Sustainable Development com (2005) 658 – On the review of the Sustainable Development Strategy. A platform for action com (2007) 642 – Communication from the Commission to the Council and the European Parliament Progress report on the sustainable development strategy 2007 com 2000/532/EC Commission Decision of 3 May 2000 replacing Decision 94/3/EC establishing a list of waste com 94/3/EC Commission Decision establishing a list of wastes. Official Journal, L 5, 7.1.1994 council decision 2003/33/EC establishing criteria and procedures for the acceptance of waste into landfills. Official Journal of the European Communities, L 11/27, 16.1.2003 directive 1999/31/EC on the Landfill of Waste. Official Journal of the European Union, L 182/1, 16.7.1999 directive 2000/53/EC on End-of-Life Vehicles. Official Journal of the European Union, L 269/34, 21.10.2000 directive 2002/96/EC on Waste Electrical and Electronic Equipment 27 January 2003. Official Journal of the European Union, L 37/24, 13.2.2003 directive 2006/12/EC on waste. Official Journal of the European Union, L 114/9, 27.4.2006


Waste management

61

directive 2008/1/EC concerning integrated pollution prevention and control (Codified version) 2008. Official Journal, L 24, 29.1.2008 directive 2008/98/EC on waste. Official Journal of the European Union, L 312, 22.11.2008 directive 75/439/EEC, Disposal of Waste Oil, Official Journal, L 194/23, 1975 directive 75/442 EEC on waste. Official Journal, L 194, 25.7.1975 directive 91/689/EEC on hazardous waste. Official Journal, L 377, 31.12.1991 directive 94/62/EC on packaging and packaging waste, Official Journal, L 365, 31.12.1994 directive 96/61/EC. Integrated Pollution and Prevention Control. Official Journal, L 257, 10.10.1996 environmental protection act 1990. http://www.opsi.gov.uk/acts/acts1990/Ukpga_ 19900043_en_1.htm flynn b, ‘WEEE, the way to go’, The New Law Journal, 157, issue 7258, 2007 ip/06/1233 – Environment: Commission pilots projects resulting in industry commitments to make greener products. http://europa.eu/rapid/pressReleasesAction.do? reference=IP/06/1233&format=HTML&aged=1&language=EN&guiLanguage =en nemry f, leduc g, mongelli i and uihlein a. Environmental Improvement of Passenger Cars (IMPRO – Car). EU Joint Research Centre, Scientific and Technical Report, JRC 40598, EUR 23038 EN 2008 regulation (eec) No 1980/2000 on a Revised Community eco-label award scheme. Official Journal, L 237, 21.09.2000 regulation (eec) No 880/92 on a Community eco-label award scheme. Official Journal, L 099, 11.04.1992 statutory instrument No. 1528, The Landfill Tax (Qualifying Material) Order 1996 statutory instrument 2002 No. 1559. The Landfill (England and Wales) Regulations 2002. http://www.opsi.gov.uk/si/si2002/20021559.htm statutory instrument 2003 No. 2635. End-of-Life Vehicle Regulations 2003. http:// www.opsi.gov.uk/si/si2003/20032635.htm statutory instrument 2006, No. 3289. The Waste Electrical and Electronic Equipment Regulations 2006. http://www.berr.gov.uk/files/file35992.pdf statutory instrument 895. List of Wastes Regulations. http://www.opsi.gov.uk/si/ si2005/20050895.htm statutory instrument No. 1056. The Waste Management Licensing Regulations, 1994. http://www.opsi.gov.uk/si/si1994/Uksi_19941056_en_2.htm#fnf010 statutory instrument No. 1527. The Landfill Tax Regulations, 1996. http://www. opsi.gov.uk/si/si1996/Uksi_19961527_en_1.htm statutory instrument No. 3212. The Landfill Allowances and Trading Scheme Regulations 2004. http://www.opsi.gov.uk/si/si2004/20043212.htm statutory instrument No. 413. The Producer Responsibility Obligations (Packaging Waste) (Amendment) Regulations 2008 statutory instrument No. 648. The Producer Responsibility Obligations (Packaging Waste) Regulations 1997 statutory instrument No. 770. The Landfill Tax (Amendment) Regulations 2008 statutory instrument No. 894. The Hazardous Waste Regulations 2005. http:// www.opsi.gov.uk/si/si2005/20050894.htm statutory instrument No. 3538. The Environmental Permitting (England and Wales) Regulations 2007. http://www.opsi.gov.uk/si/si2007/uksi_20073538_en_1

4 Thermal methods for recycling waste composites S. J. P I C K E R I N G, University of Nottingham, UK

Abstract: This chapter describes a thermal process based on a fluidised bed for the recycling of glass and carbon fibre from thermoset composites. The process is particularly suitable for producing high quality fibre from contaminated or mixed materials. The chapter will describe the fluidised bed process, and detail the quality of the fibres that are recovered. Applications for the reuse of the fibre that have been investigated will then be explained and the prospects for commercial scale operation considered. Current and future research directions will be outlined; other relevant research activities will also be mentioned.


Key words: recycling, composite materials, carbon fibre, glass fibre, fluidised bed.

4.1

Introduction

The aim of recycling processes is to reduce environmental impact by reusing materials in a more sustainable way. In general, it makes sense to try to recover as much economic value from a material in a recycling operation since the value of a material represents to a large extent the input of resources needed to produce the material or the scarcity of the material. Recycling processes in which a more valuable recyclate is produced are therefore likely to reduce environmental impact the most and they are also more likely to be cost effective. Polymer composites made from thermoplastic matrices can be recycled directly by remelting and remoulding into high value materials (Goodship 2007). However, recycling polymer composites based on thermoset matrices is much more difficult as the polymer cannot be remoulded (Pickering 2006). Recycling processes have therefore been developed to separate and recover valuable materials from thermoset composites that have the potential to be reused. A number of recycling technologies have been proposed and developed for thermoset composite materials and these are summarised in Fig. 4.1. There are fundamentally two categories of process: those that involve mechanical comminution techniques to reduce the size of the scrap to 65

66

Management, recycling and reuse of waste composites Recycling routes for thermoset composites

Thermal processes

Mechanical recycling

Powdered fillers

Fibrous products

Combustion with energy recovery (and material utilisation)

Fluidised bed thermal process

Clean fibres and fillers with energy recovery

Pyrolysis and other thermo/chemical processes

Fibres and chemical products


4.1 Potential recycling processes for thermoset composite materials.

produce recyclates; and those that use thermal and/or chemical processes to break the scrap down into materials and energy. Mechanical recycling techniques can recycle all of the composite material, but the recyclate does not usually provide the same function as the original material and so there is a reduction in value. The simplest thermal process is combustion with energy recovery. Whilst this is a robust process capable of dealing with a wide range of different materials, only the energy value in the scrap composite is recovered. Pyrolysis and other chemical processes have the advantage of being able to recover material from the polymer separated from the inorganic components of the composites, such as glass and carbon fibre and fillers. In some pyrolysis processes the fibres recovered are contaminated with char and in general the processes are less tolerant of mixed and contaminated materials. In a thermoset composite the most valuable constituent within the material is usually the reinforcement fibre, particularly in the case of carbon fibre composites, where the value of the carbon fibre is typically an order of magnitude greater than that of the polymer. In the fluidised bed thermal process the aim is to recover high grade fibres from scrap glass fibre or carbon fibre reinforced thermoset composites. This is done by removing the polymer from the fibre by an oxidative thermal process to yield clean fibres. The process also has the advantage of being very tolerant to mixed and contaminated materials and so it is particularly suitable for recycling waste composites from end-of-life products. This chapter will describe the fluidised bed process, and detail the quality of the fibres that are recovered. Applications for the reuse of the fibre that have been investigated will then be explained and the prospects for commercial scale operation considered.

Thermal methods for recycling waste composites

4.2

67

The fluidised bed recycling process

4.2.1 Description of the process Fluidised bed technology has a wide range of applications in combustion and process engineering. It is particularly suitable for solids processing where good mixing and close temperature control are required. It is used widely for the combustion of solid fuels and in the chemical industry for gas–solid reactions (Kunii and Levenspiel 1993). In research conducted at the University of Nottingham, a fluidised bed process has been developed for recycling thermoset composite material. The process aims to remove the thermoset polymer from the reinforcement fibres by a thermal oxidative process, leaving clean fibres which are then suitable for recycling back into composite material. The general principles of fluidisation are described below, followed by a detailed description of the process.


4.2.2 Principles of a fluidised bed Fluidisation is a technology in which a bed of solid particles is transformed into a fluid-like state through suspension in a fluid, usually a gas. The solid particles rest on a porous air distributor plate inside a containment vessel. The distributor plate allows the gas to pass into the bed from a plenum chamber below. It may be made from a sintered porous material but more usually a solid plate with a number of small holes or air nozzles is used. At low air velocities through the bed the particles remain stationary as the aerodynamic drag on the particles is low and air only percolates through the inter-particle voids. With increasing velocity, the particles start moving apart and vibrate owing to increase in drag. This leads to bed expansion and an increase in pressure drop. Increasing the velocity above a critical value, which is generally known as the minimum fluidisation velocity, results in the particles being held in suspension in the gas flow and the weight of the entire bed of particles is balanced by the upward drag. As the air velocity increases further, no increase in pressure drop is observed as the bed expands continuously, allowing more air to pass through. A typical transition from a packed bed to fluidisation with increasing air velocity is depicted in Fig. 4.2. The suspended particles behave as fluid. At higher velocity, bubbles start forming at the distributor, then rise through the bed and burst at the upper sand surfaces. Good mixing is attained in this stage, which promotes uniformity in the bed temperature and composition. Higher velocities cause bubble growth and fluctuations in bed pressure drop. When the air velocity exceeds the particles’ terminal velocity, the bed upper surface disappears and entrainment becomes appreciable. At a still higher velocity, all the

68


Pressure drop across bed

Minimum fluidising velocity

Gas velocity

4.2 Relationship between air flow velocity and pressure drop in a packed bed and fluidised bed.


particles are carried out of the bed with the air, which is similar to pneumatic transport of fine particles.

4.2.3 Fluidised bed recycling process The basic principle of the fluidised bed recycling process is shown in Fig. 4.3. The fibres are recovered from the composite by removing the polymer matrix by heat in a bed fluidised with air. Once the polymer has been removed from the fibres and any mineral fillers that may also be present in the composite, they are carried out of the fluidised bed (elutriated) in the gas stream and can be separated from the gas in a cyclone or other gas–solid separation device. At the temperature at which the fluidised bed operates, the polymer does not fully oxidise and so, after separation of the fibres and fillers, the gas stream passes through to a secondary combustion chamber where high temperature combustion achieves full oxidation. Energy in the form of heat can then be recovered from the high temperature exhaust gases and, if necessary, any pollutants can be scrubbed before being emitted to atmosphere. In general, thermoset polymers contain mainly carbon, hydrogen and oxygen and so the products of full oxidation are carbon dioxide and water, which can be emitted directly to the atmosphere. Some composites may contain halogenated resins for fire retardancy and, in this case, the flue gas would require scrubbing to remove the resulting acid gases before being emitted to atmosphere. The process therefore recovers reinforcement fibres and fillers in a clean form for reuse. The polymer is fully oxidised and its energy value is recovered.

Thermal methods for recycling waste composites Scrap FRP

Fluidised bed

69

Clean flue gas

Fibres and fillers carried in gas flow

Separation of fibres and fillers

Recovered fibres

Secondary combustion chamber

Recovered fillers

Heat recovery

Recovered energy

4.3 Diagram showing principle of fluidised bed recycling process (FRP = fibre reinforced polymer).


Clean flue gas

Scrap FRP

Afterburner

Cyclone (to separate fibres)

Induced draught fan Recovered fibres Air preheating elements

Fluidised bed Air distributor plate

Air inlet

4.4 Diagram of the fluidised bed recycling process test facility.

4.2.4 Process operation Figure 4.4 shows a schematic representation of the fluidised bed recycling process. Fluidising air, preheated typically to 450–550 °C, is introduced into a bed of silica sand, with particle diameter typically 0.85 mm and with a fluidising velocity in the region of 1 m/s. At this condition the fluidised bed is in the bubbling regime where there is very good mixing and uniformity of temperature in the bed. The process has been developed on a laboratory pilot scale in which the fluidised bed is 300 mm in diameter (Pickering et al. 2000). In this test rig, electric heaters are used to preheat the air.


70


However, in a commercial scale process heat could be recovered from the gases downstream of the secondary combustion chamber. The temperature of the fluidised bed is chosen to be high enough to give rapid decomposition of the polymer but not too high that the fibres are degraded significantly. For polyester resins a temperature of 450 °C has been found to be suitable, whereas epoxy resins are more stable thermally and generally require a higher temperature of 550 °C for processing at a practical rate. The process is suitable for the processing of both glass fibre and carbon fibre reinforced composites. Carbon fibres are made of graphite and oxidise at high temperature in air. It has been found that at a fluidised bed temperature of 550 °C epoxy resin decomposes rapidly, but given the residence time of the fibres in the fluidised bed, of about 20 minutes, there is little oxidation of the carbon fibre as indicated by a reduction in fibre diameter. Scrap composite material is fed into the bubbling fluidised bed and at high temperature the thermoset resin rapidly decomposes. From thermogravimetric analysis (TGA) of the decomposition of epoxy resin, it is understood that there are two stages in the decomposition of the resin (Jiang et al. 2005). Initially a fast devolatilisation stage takes place in which the resin loses about 50% of its mass and the remaining polymer is in the form of char. This is followed by a slower char oxidation stage. Once the char residue has reduced to about 3% of the weight of the original polymer, the fibres become released from each other due to the mechanical agitation in the fluidised bed and they can then be elutriated. The rate of elutriation depends on the concentration of fibres in the fluidised bed as well as the operating parameters of the bed. It is found that if the concentration of fibre in the bed is too high that fibre agglomeration can take place and this results in large clumps of agglomerated fibre remaining in the bed. It is also observed that longer fibres agglomerate much more easily than shorter fibres (Jiang et al. 2005). Research is currently in progress to obtain a better understanding of this phenomenon and of the parameters controlling the rate at which material can be processed. The clean fibres are released and elutriated into the air stream. The fibres are then separated from the air in a cyclone and collected for recycling. If the composite contains a mineral filler in addition to glass or carbon fibre, the fibres can be separated from the fillers by means of a rotating screen positioned above the fluidised bed as shown in Fig. 4.5. The filler particles pass through the screen and can be separated from the gas stream with a cyclone. The fibres are collected on the screen and are removed continuously by a counter-current of air as shown in Fig. 4.5.

4.2.5 Waste preparation The composite material being recycled must be reduced in size so that it can be fed into the process. The fluidised bed process recovers the rein-


71

Gases + fillers Secondary air inlet Hot gases + fibres + fillers

Rotating screen

Secondary air outlet

Fibre collection bin

Secondary air extraction


4.5 Rotating screen to separate fibres from fillers.

forcement fibres in the form of short individual filaments. The length of these must be limited to prevent agglomeration in the fluidised bed and also materials handling and processing problems during reuse. This is done by ensuring that the fragments of composite fed into the process do not contain fibres longer than about 30 mm. When scrap composite contains continuous reinforcement this means that the dimensions of the fragments fed into the fluidised bed should have a maximum dimension less than 30 mm. Several methods can be used to prepare the waste. For scrap from manufacturing operations it is usually necessary to cure the waste if required as uncured composite material is very difficult to reduce in size. Waste in the form of thin sheets, for example prepreg, can easily be chopped up in a conventional granulator of the type commonly used in the plastics industry. Figure 4.6 shows a laboratory granulator manufactured by the Zerma company. Larger granulators are available for processing large quantities of waste. The waste circulates within the granulator and is repeatedly cut by the knives until it is small enough to pass through the holes in the screen. The size that the waste is reduced to is generally determined by the size of the holes in the screen. But composite in which the fibres are highly aligned tend to shred into fragments of high aspect ratio. These may pass end-on through the screen and so the length of the fibres in the granulate may be significantly longer than the size of the holes in the screen. Scrap end-of-life components may be large in size and have thick crosssections and larger machines may be needed to process this material. The

72

Management, recycling and reuse of waste composites Scrap feed

Rotor knife – 30x

150 rpm Stationary blades Perforated mesh


4.6 Laboratory granulator.

initial size reduction may be done in large, slow-speed shredding machines. These often use two counter-rotating shafts with knives mounted to grip and cut the material as shown in Fig. 4.7. These machines are usually single pass and there is no screen to control the output size, the main purpose being to give an initial size reduction before further processing. Very large shredders which are capable of taking components of 1–2 m in width in which the thickness of the composite could be in excess of 25 mm are available. Following the initial size reduction, a further reduction could be done in a granulator. Alternatively, hammer mills, have been shown to be very effective in reducing the size of composites. Here the action is by impact rather than cutting and it has been found that whilst the polymer is pulverised into a powder, the fibre fragments usually retain some integrity and a useful fibre length. In a hammer mill the output size is also determined by a mesh screen.

4.2.6 Tolerance to contamination Components made using composites often involve a number of different materials bonded or moulded together. For instance a composite component may be made of two skins of a glass reinforced thermoset, one or both with a painted surface, moulded or bonded to a foam or aluminium honeycomb core with metal inserts at fixing points. Such a mixture of


73


4.7 Twin screw shredder with counter-rotating shafts.

materials is difficult to separate and one of the advantages of the fluidised bed process is that it is tolerant of mixtures of different material and contamination in the composite feed. Any organic material in the feed is fully decomposed in the fluidised bed. This means that contamination from paints, foam cores or thermoplastics is not a problem and tests have shown that in small quantities these do not affect the process. So for instance, the thermoplastic backing film on a prepreg does not have to be removed before the material is shredded and fed into the fluidised bed. In general thermoplastics can be recycled by remoulding and this is generally the most suitable recycling method for these materials. So it would be expected that significant quantities of thermoplastic would have been separated from the carbon fibre composite before feeding into the fluidised bed. It is also found that any metals in the fluidised bed fall to the bottom of the bed during processing and can be recovered. Contamination from metals could result from metal inserts for attachment points, aluminium honeycomb cores, wiring and even metals electroplated onto the composite; all of these have been tested in the fluidised bed. So whilst it would be envisaged that large metal pieces would be removed from the waste, metals that are intimately mixed with the composite can be separated in the fluidised bed process. In a commercial scale process the bed material could be continuously removed and regraded to remove these metal pieces.

74


The fluidised bed process is therefore particularly well suited to dealing with waste composites from end-of-life parts where the waste may not only be a mixture of different materials, but there may also be contamination that has arisen during service. Tests have shown that clean, good quality glass and carbon fibres can be recovered from end-of-life components, in which there were mixtures of materials, in both the aerospace and automotive industries.

4.3

Properties of the recycled fibre

4.3.1 Physical form The fibres are elutriated from the fluidised bed in the form of short individual filaments. These are separated from the gas stream in a cyclone and collected in a bin. As the fibres are in a fluffy form the bulk density is very low, typically in the order of 50 kg/m3. However, the fibres can easily be compressed for packaging for further processing. Some fluffy carbon fibres are shown in Fig. 4.8.


4.3.2 Fibre length distribution The fibre length distribution of the recycled fibres depends on the fibre length distribution of the fibres in the waste feed and also on degradation in fibre length that takes place in the fluidised bed. Measuring the input fibre length in feed material is not easy as the size-reduced composite feed will contain irregularly shaped fragments in which there will be a distribution of fibre lengths. The most practical way is to separate the fibres from a sample of waste by burning off the polymer and then using an image

4.8 Recycled carbon fibre in a fluffy form.


75

30

25

Mass (%)

20

15

10

5

0


0

2

4

6 8 Fibre length (mm)

10

12

14

4.9 Typical length distribution for recycled glass fibres recovered from an SMC. The scrap fed into the fluidised bed had a maximum particle size of about 15 mm (Pickering et al. 2000).

analysis technique to measure fibre lengths. Figure 4.9 shows a typical fibre length distribution of the recycled glass fibre from a polyester/glass sheet moulding compound (SMC) feed. The scrap fed into the fluidised bed had a maximum particle size of about 15 mm. An investigation of carbon fibre length degradation in the fluidised bed process has been undertaken in which pieces of composite with fibres of uniform known length were fed into the process. The fibre length distribution and mean fibre length of the recycled fibres was then measured. Figure 4.10 shows the fibre length distribution in the recycled fibre, when fragments of carbon fibre composite with a uniform fibre length of 10 and 20 mm were fed into the fluidised bed. It is found that the amount of fibre degradation in the fluidised bed depends on the initial length of the fibres. Longer fibres have a greater tendency to break and hence have a greater length degradation. This is illustrated in Fig. 4.11, which shows the degradation in mean fibre length as a function of input fibre length. The figure expresses the length degradation in terms of mean fibre length based on fibre weight and also on fibre number. Mean fibre length by weight is considered to be a more meaningful quantity than mean fibre length based on fibre number as it represents the fibre length with most of the mass. In contrast the fibre length distribution by fibre number can be distorted by a large number of very short fibres, which have little mass.

76

Management, recycling and reuse of waste composites Initial fibre length = 20 mm

16 14 12 10 8 6 4 2 0

Mass fraction (%)

Mass fraction (%)

Initial fibre length = 10 mm

16 14 12 10 8 6 4 2 0

0 2 4 6 8 10 Recovered fibre length (mm)

0

5

10

15

20

Recovered fibre length (mm)

Length degradation (%)


4.10 Recycled fibre length distributions for two different initial fibre lengths (Yip et al. 2002). 90 80 70 60 50 40 30 20 10 0

Ln Lw

5

10

15

20

25

30

Initial fibre length (mm)

4.11 Average recovered fibre length degradation (by number, Ln, and by weight, Lw) versus initial fibre length (Yip et al. 2002).

4.3.3 Properties of recycled glass fibre Visualisation Glass fibres recycled from polyester or vinylester composites are found to be clean with very little residual polymer remaining. Figure 4.12 shows a scanning electron micrograph of some recycled glass fibres. Mechanical properties The mechanical properties of the recycled fibres are measured by single fibre tensile tests according to BS ISO 11566. The tests are done by mounting individual fibres with adhesive across a hole in a cardboard frame. After mounting the frame in a tensile testing machine the cardboard is cut so that fibres can be pulled to failure. Glass and carbon fibres are brittle and the tensile strength is affected by the number of microscopic flaws in the surface.


77

JK 08

100 μm × 100

10 KV

22 mm

4.12 Scanning electron micrograph of recycled glass fibres. 3.0 2.5 2.0 1.5

Virgin fibre

Weibull scale parameter (GPa)


With longer gauge lengths the fibres will be more likely to have large flaws and so lower tensile strengths are obtained. Gauge lengths of 6 and 10 mm are typically used for the recycled fibres. For brittle materials there is a distribution in tensile strength and so a minimum of 20 fibres are tested. The results are analysed using Weibull statistical analysis and the tensile strength reported as the Weibull modulus. It is found that the modulus of recycled glass fibres is not affected by the recycling process and, for E-glass, the modulus is approximately 70 GPa. However, the tensile strength decreases significantly and the strength degradation is greater at higher processing temperatures. Figure 4.13 shows the tensile strength of recycled glass fibres at different temperatures in the

450 °C

1.0 550 °C 0.5

650 °C

0.0

4.13 Tensile strength of virgin and recycled glass fibre. The recycled fibre was recovered from the fluidised bed process operating at temperatures of 450, 550 and 650 °C (Pickering et al. 2000).

78


fluidised bed. Polyester and vinylester polymers decompose rapidly at a temperature of 450 °C. Consequently, operating the fluidised bed at this temperature will result in the recycled fibres having a tensile strength about half that of virgin fibre. The reasons for the reduction in strength are not certain (Pickering et al. 2000), but are believed to be due to a combination of mechanical damage to the fibre surface due to the agitation in the fluidised bed, annealing of the fibre causing stress relaxation at high temperature and also chemical diffusion of sodium ions from the surface of the glass with replacement with smaller hydrogen ions, resulting in tensile stresses in the fibre surface.

4.3.4 Properties of recycled carbon fibre Visualisation The recycled carbon fibres are also found to have a clean fibre surface with only a few traces of polymer residue, as shown in Fig. 4.14.


5 μm

T600s

5 μm

×10 000

5 μm

T600s

MR60H

×10 000

5 μm

×10 000

MR60H

×10 000

4.14 Scanning electron micrographs of virgin (upper) and recycled (lower) carbon fibres.


79

Oxidation and fibre diameter As the fluidised bed process takes place at temperature in air there is likely to be some oxidation of the carbon fibres. It is known, however, that graphite oxidises relatively slowly in air at temperatures of 500–600 °C (Yin et al. 1994) and it is found that the diameter of the fibres is not significantly reduced during processing in the fluidised bed. Measurements of fibre diameter show reductions of a few tenths of a micron. Tests undertaken by TGA show virgin carbon fibres exposed to air for 20 minutes at 550 °C experience a weight loss in the region of 2% (Wong et al. 2006) as shown in Fig. 4.15. The residence time of the fibres in the fluidised bed is in the order of 20 minutes, but the fibres are not exposed to air throughout that time as they initially have a protective covering of polymer, so less oxidation would be expected. Exposure of carbon fibre to a temperature of 550 °C in air for periods of up to an hour or more show noticeable pitting of the fibre surface as shown in Fig. 4.16 (Wong et al. 2006). However, this is not observed in any fibres recycled from the fluidised bed process.

The tensile strength and modulus of the recycled carbon fibre vary depending on the grade, and several different grades of fibre have been processed in the fluidised bed. The mechanical properties are measured by the same method as glass fibres by single fibre tensile tests according to BS ISO 11566. Tests were undertaken with a gauge length of 6 mm. A measurement of fibre diameter is also necessary to determine the tensile strength and

100 98 96 Weight (%)


Mechanical properties

94 92 90 88 86

T600s T600s

MR60H MR60H

84 0

20

40

60

80

Heat treatment duration (min)

4.15 Loss of weight of virgin carbon fibre exposed to air at 550 °C (taken from thermogravimetric analysis) (Pickering et al. 2006).

80

Management, recycling and reuse of waste composites 5 μm

5 μm

4.16 Pitting of the surface of carbon after 60 minutes heating in air (Wong et al. 2006).


Table 4.1 Measured tensile properties of carbon fibre (single fibre tensile tests) recycled in the fluidised bed process

Fibre type

Tensile strength (GPa)

% tensile strength retention in recycled fibre


Toray T600s virgin Recycled at 550 °C

4.84 3.18

66

208 218

Toray T700 virgin Recycled at 550 °C

6.24 2.87

46

219 205

Hexcel AS4 virgin Recycled at 550 °C

4.48 2.78

62

231 242

Grafil MR60H virgin Recycled at 550 °C

5.32 2.63

49

227 235

Grafil 34–700 virgin Recycled at 450 °C

4.09 3.05

75

242 243

modulus of the fibre from the tensile tests and scanning electron microscopy (SEM) was used to do this. Table 4.1 shows tensile test results for a range of different fibre types. In each case a sample of virgin fibre was tested for comparison. The results show that the modulus of the recycled fibres is similar to that of the virgin fibre, but there is a significant decrease in tensile strength. The tensile strength reduction appears to be less at lower temperatures in the fluidised bed. But for carbon fibre processed at 550 °C, the practical processing temperature, there is a strength reduction of a third to half the virgin tensile strength. It is perhaps not surprising that the modulus of the fibre does not

Thermal methods for recycling waste composites 250

6.0 5.5

200

5.0 4.5

150

4.0 100

3.5 3.0

50

2.5 2.0 –10

Tensile strength


Tensile strength (Weibull scale) (GPa)

81

Tensile modulus 0

0

10 20 30 40 50 Heat treatment duration (min)

60

70


4.17 Effect of duration of heat treatment (in air at 550 °C) on tensile strength and tensile modulus of carbon fibre (Pickering et al. 2006).

change during recycling as the processing temperatures in the fluidised bed are well below the temperature at which the fibre was graphitised during manufacture. The reduction in tensile strength may be due to a number of factors including the effect of oxidation and mechanical damage due to abrasion with sand particles in the fluidised bed. It is clear from Fig. 4.16 that the effect of prolonged oxidation in air produces noticeable pits in the surface of the carbon fibre. These are not visible after exposure for 10 minutes but oxidation will nevertheless be taking place and these flaws produced will act as crack initiators and reduce the tensile strength. Figure 4.17 shows the effect of exposure to air at 550 °C on the mechanical properties of virgin carbon fibre. Whilst the modulus remains constant, there is a steady reduction in tensile strength and after about 60 minutes the strength of the fibre is similar to that recycled in the fluidised bed. The residence time of the carbon fibre in the fluidised bed is in the order of 20 minutes and the reduction in strength of the virgin fibre after 20 minutes exposure to air is not as much as that recycled in the fluidised bed. It is therefore suggested that the additional loss in strength observed in the recycled fibre is due to mechanical damage to the fibre surface caused by abrasion in the fluidised bed. Surface chemistry The mechanical properties of a carbon fibre reinforced composite are strongly controlled by the interfacial adhesion between the fibre and the matrix, which is determined to a large extent by the surface chemistry of


82


the fibre. X-ray photoelectron spectroscopy (XPS) has been used to study the surface of the carbon fibre, and it identifies and quantifies the elemental composition and functional groups present. The interfacial adhesion to a resin can be measured by a fibre pull-out test in which a small bead of resin (about 50–80 μm in diameter) is cast onto a fibre. A tensile test is then undertaken to measure the force needed to pull the fibre through the bead from which the interfacial shear stress (IFSS) can be calculated. The oxygen to carbon (O/C) ratio on the surface of recycled carbon fibre and IFSS of the virgin and recycled fibres are tabulated in Table 4.2. In both cases the virgin fibres have a higher O/C ratio than the recycled fibres and there is a decrease of 36% in the O/C ratio on the surface of the recycled T600s fibre. This is probably due to the removal of the oxygen-containing groups present in the surface sizing. However, there is only a reduction in the IFSS of 13% compared with the virgin T600s fibre. For the MR60H fibre, although a 22% drop in O/C ratio is measured, the recycled fibre shows a 28% increase in IFSS. The exact reason for this phenomenon is unknown although the scatter in the results may indicate that it is of less significance. Table 4.3 shows the functional groups present on the surface of the virgin and recycled T600s carbon fibre as derived from the XPS data. A significant increase in carboxyl group is found on the recycled fibre. However, there is a 22% and 36% reduction in the hydroxyl and carbonyl functional groups respectively and this might account for the 13% reduction in IFSS. There is no clear correlation between the functional groups and heating duration, but overall, the surface chemistry on the virgin and heat-treated fibres is quite similar. Table 4.2 Oxygen/carbon ratio and interfacial shear strength of carbon fibres Virgin T600s O/C ratio Interfacial shear strength (MPa)

Recycled T600s

Virgin MR60H

Recycled MR60H 0.195

0.25

0.16

0.25

52.7 (±11.6)

46.1 (±10.0)

46.5 (±8.2)

59.8 (±8.4)

Table 4.3 Proportions of functional groups on surface of carbon fibre (Wong et al. 2006) Functional group

Recycled

Virgin

10 min

60 min

Total carbon (graphitic + aliphatic) C–OH C=O COOH Carbonate

69.5 21.7 3.53 4.54 1.4

64.6 27.7 5.48 2.16 0

68.6 20.5 3.87 4.54 1.40

60.1 23.3 6.36 6.04 1.08


83

Table 4.4 Electrical resistivity of carbon fibres Electrical resistivity, ×10−3 Ω cm Fibre type

Virgin fibre

Recycled fibre

MR60H T700

1.68 ± 0.17 1.51 ± 0.15

1.46 ± 0.17 1.50 ± 0.12


Electrical resistivity The electrical resistivity of recycled carbon fibre has been measured to establish whether any changes take place during the fluidised bed processing. Resistivity was determined using a single filament resistance method in which the voltage drop along a fibre passing a known current is measured. The results of measurements on two types of fibre are shown in Table 4.4. The T700s fibre showed no change in resistivity but a 13% reduction was observed from the recycled MR60H. It is noted that the resistivity of the virgin MR60H fibre is slightly higher than that of the recycled fibres of virgin T700 fibre. It is possible that this is due to a surface treatment on the fibre that was removed during the recycling process. So the apparent decrease in resistivity or increase in conductivity may not be real. Nevertheless the conclusion of this test is that the recycled fibres have electrical conductivity at least as good as virgin fibre.

4.4

Applications for the reuse of fibre recycled from the fluidised bed process

4.4.1 The form of the recyclate The physical form of the recycled fibre from the fluidised bed process is a fluffy mass of individual short glass or carbon filaments. This is generally unlike any form of commercially available virgin fibre product. Glass and carbon fibres are manufactured as continuous fibre and they are produced in the form of continuous roving, each roving containing several thousand filaments, to which a size and/or binder has been applied to hold the fibres together to assist handling and processing. Manufacturing processes using glass and carbon fibre are thus designed for continuous rovings or, in some cases, chopped rovings. In reusing recycled fibre, an important issue is thus to identify ways in which the fluffy fibres can be processed. The reuse applications described in this chapter were investigated for recycled glass and carbon fibre from the fluidised bed recycling process. However, they are also applicable to recycled fibres produced from other composites

84


recycling processes in which the fibres are in the form of short individual filaments, such as described in George and Carberry (2007). There are a number of existing commercial applications in which fibres are processed as short fibres in a generally dispersed form and these may be appropriate for recycled fibre. Two examples are short fibre moulding compounds (Monk, 1997), such as bulk moulding compounds (BMC) or SMC, and non-woven fabrics or mat which in low areal density form are often known as veil or tissue. In the former, the fibres are used for their structural reinforcement properties. Non-woven fabrics may also be used to provide reinforcement, but they also have other applications, such as surfacing veil, to provide improved surface finish.


4.4.2 Reuse as a reinforcement Glass and carbon fibre are principally used as structural reinforcement in polymer composites and potential routes for reusing the recycled fibre in thermoset composites are shown in Fig. 4.18. At low volume fractions, BMCs have fibre volume fractions in the order of 10%. These are manufactured by blending chopped virgin rovings into a resin and filler mix to give a product with a 3D fibre dispersion. BMCs do not have high strength or stiffness and are used for semi-structural applications, such as equipment enclosures. Recycled fibre could be used directly in a BMC by blending the fluffy fibre with the resin and filler in the same way that virgin fibre is used. SMCs have higher fibre volume fractions in the region of 20–30%. They are manufactured in sheet form and have a predominantly 2D random fibre architecture and are used for semi-structural applications where better stiffness and strength are needed. They are produced by chopping continuous

Random

Recycled fibres

Non-woven mat BMC/ low value Low volume fraction ~10%

SMC/ prepreg Intermediate volume fraction 10– 40%

Thermoplastic injection moulding Intermediate volume fraction 20– 40%

Aligned

High volume fraction 30– 60% Intermediate aligned material

Wet processing techniques

Dry processing techniques

4.18 Potential routes for processing recycled carbon fibre as a structural reinforcement.



85

rovings and placing them randomly on a film of resin and filler. Typical applications are in the automotive industry for body panels. To use recycled fibre in an SMC, it is unlikely that the fluffy material could easily be processed directly and laid down in a measured way on to the resin film. An appropriate method of manufacture could be to process the recycled fibre into a non-woven mat first and then use this mat to manufacture an SMC. Higher strength and stiffness composites required higher fibre volume fractions and these are usually achieved by aligning the fibres in the composite so that a higher packing density can be achieved. Manufacturing methods for achieving this include the use of prepreg and filament winding techniques where fibre volume fractions of up to 60% can be achieved. Prepreg is manufactured by laying down continuous rovings, either directly or as a woven fabric, onto a resin film. To use recycled fibre in a prepreg such as this would require the fluffy discontinuous fibres to be processed into an aligned form. Glass and carbon fibre are also used as reinforcement in thermoplastic composites and there are numerous different processing techniques. Short fibres can be compounded with a thermoplastic to produce pellets suitable for injection moulding. Continuous fibre can be processed as a woven textile impregnated with a thermoplastic to produce a high fibre volume fraction composite.

4.4.3 Processing recycled fibre into a non-woven mat Wet processing techniques, based on paper making technology exist for the conversion of chopped continuous rovings into a non-woven fabric in which the fibres are dispersed into a random, predominantly 2D structure of short individual filaments. The technique involves dispersing the fibre in a large volume of water in which dispersion agents and viscosity modifiers have been added. The fibre concentration in the resultant slurry is typically less than 1% by volume. The slurry is then passed through a mesh where the liquid drains through and the fibres are laid down to form the non-woven fabric. A binder is usually added to hold the fibres together after drying. A commercial pilot plant was used to process recycled carbon fibre and high quality non-woven mats were produced with areal densities ranging from 10 to 100 g/m2 (Pickering et al. 2006). This demonstrated that the recycled fibres from the fluidised bed could be processed using an existing commercial process into a high quality product.

4.4.4 Reuse of recycled glass fibre Two applications of the use of recycled glass fibre from the fluidised bed process have been demonstrated: substituting recycled glass fibre for virgin

86


glass fibre in a BMC and for the manufacture of glass fibre veil. In both these demonstrations the recycled glass fibre had a mean fibre length of 5.6 mm (by weight) and the size distribution is shown in Fig. 4.9.


BMC Recycled glass fibre was substituted for virgin fibre in a BMC (Kennerley et al. 1998). The BMC was based on polyester resin with a fibre volume fraction of 13%, and the formulation is given in Table 4.5. The virgin glass fibre had a uniform length of 6 mm. A series of mixes were produced in which the virgin fibre was substituted with proportions: 0%, 25%, 50%, 75% and 100% of recycled fibre. The recycled fibre was E-glass processed in the fluidised bed at 450 °C. This was found to have the same tensile modulus as virgin glass fibre but the tensile strength was 49% of that of virgin fibre. The fibre was recycled from an SMC in which there was over twice as much filler as glass fibre by weight. After separation of the fibre from the filler, the fibre product used in the BMC had a purity of 92%, although the ratios of recycled fibre substituted for virgin in the BMC did not take account of fibre purity. Test plaques of the BMC were moulded and the mechanical properties tested using standard techniques. The results in Fig. 4.19 show no significant reduction in the tensile stiffness with substitution of up to 100% of the virgin glass fibre with recycled fibre. Tensile strength shows no significant change with substitution of up to 50% recycled fibre, although there is a reduction of about 50% in strength with 100% recycled fibre substitution. Impact strength shows steady reduction with increasing recycled fibre content, a reduction of about 30% is shown with substitution of 50% recycled fibre. A batch of BMC with 50% substitution of recycled fibre was made in which the recycled fibre had been treated with a silane solution in an attempt to improve the fibre to resin bonding by replacing the sizing present on the virgin fibre. However, no significant change was found in any of the properties measured. Fibre to matrix bonding would be expected Table 4.5 BMC formulation Parts by weight Polyester resin Shrink control agent Calcium carbonate filler Aluminium trihydrate Mould release agent

66.7 33.3 136.2 109.0 6.5

Glass fibre

13% volume fraction

30 20 10 0

0

20

40

60

80

Charpy impact strength (kl/m2)

15

10

5

0

100

Non-virgin reinforcement (%)


87

20

40 Tensile modulus (GPa)

Ultimate tensile strength (MPa)


0

20

40

60

80

100


20

15

10

5

0

0

20

40

60

80

100


4.19 Tensile strength, tensile stiffness and impact strength of BMC made with replace of virgin reinforcement with recycled glass fibre (solid markers represent samples in which the recycled glass fibre had been treated with a silane sizing agent) (Kennerley et al. 1998).

to influence the strength of the BMC, but no effect was observed, even on impact strength which was found to be most sensitive to substitution of recycled fibre. It is thus concluded that the reduction in tensile strength is more significant than the removal of the size during the recycling process in influencing the properties of BMC containing recycled glass fibre. A demonstrator component was made with a BMC containing recycled glass fibre. The component was an automotive headlamp moulding and 50% of the reinforcement was substituted with recycled fibre; the total glass content of the BMC was 18% by weight. Apart from the use of recycled fibre no changes were made to the BMC formulation, the compounding or the moulding procedures. The mechanical properties of the component were not significantly different to the control, although impact strength was slightly reduced. The electrical properties were unchanged.

88


10

No treatment Silane treated

8 6 4 2 0 0

20

40

60

80

100

Proportion of reclaimed fibres (%)

Breaking length (km)

A glass fibre veil, or tissue, was also made using recycled fibre (Pickering et al. 2000). Veil with an areal density in the range 60–80 g/m2 was manufactured using a wet technique in the laboratory. Quantities of glass fibre were dispersed in water containing dispersion aids and viscosity modifiers to give a slurry containing 0.2% fibre by weight. The slurry was then passed through a mesh on which the fibres were collected to produce the veil. After draining the liquid from the veil, an acrylic binder was applied by dipping in a liquid emulsion and the veil was then dried and the binder cured in an oven. The veil had a final binder content of 18–24% by weight. Veils were made in which the virgin glass fibre was substituted with proportions of recycled fibre of 50% and 100%. The virgin fibre used was uniformly chopped to 18 mm in length as this is the length used for a commercial grade of veil used as a benchmark. Two batches of recycled glass fibre were used which had been processed in the fluidised bed at 450 and 550 °C. The strength of the veil was measured and expressed in terms of breaking length – the vertical length of fibre which would break under its own weight. The results are shown in Fig. 4.20. Increasing the proportion of recycled fibres decreased the strength of the veil. Veil with a complete replacement of the virgin fibres by fibres processed at 450 °C had a breaking length that was approximately 50% that of virgin fibre veil. This agrees with the results of single fibre testing in which fibres processed at 450 °C retained about 50% of the virgin tensile strength. For the sample of veil made with recycled fibres processed at 550 °C the veil strength was 40% of the virgin value, although the fibres at 550 °C retained only 20% of the strength of virgin fibres. It appears therefore that veil strength does not depend on fibre strength alone, but is also a function of fibre length and binder content.

Breaking length (km)


Veil

10

No treatment Silane treated

8 6 4 2 0 0

20

40

60

80

100

Proportion of reclaimed fibres (%)

4.20 Effect of proportion of recycled glass fibre on the breaking strength of glass fibre veil (LH diagram for fibres recycled at a processing temperature of 450 °C and RH for fibres processed at 550 °C) (Pickering et al. 2000).


89

4.4.5 Reuse of recycled carbon fibre Reuse applications for recycled carbon fibre from the fluidised bed process include conversion to a veil for use in electromagnetic shielding, compounding with a thermoplastic for use in injection moulding and the manufacture of thermoset moulding compounds based on epoxy resin and these will be described. To achieve the best structural properties from a carbon fibre composite, high fibre volume fractions are required and these can be achieved with aligned fibre. The problems in aligning carbon fibre will be discussed and some initial investigations of a technique for aligning recycled carbon fibre presented.


Electromagnetic shielding As carbon fibre is electrically conductive, a veil applied to the surface of an otherwise non-conducting material, such as a glass fibre reinforced composite, can be used to provide shielding protection for electromagnetic radiation. Carbon fibre is commonly used as a conducting filler in thermoplastics for the same purpose. As recycled carbon fibre can easily be converted into a non-woven fabric or veil, the potential for using a surfacing veil on a glass reinforced thermoset made from recycled carbon fibre to provide electromagnetic shielding has been investigated (Wong et al. 2005). In the investigation a glass reinforced polyester resin composite was manufactured with a layer of carbon fibre veil moulded onto one surface. The veil was made from recycled and virgin carbon fibre (Toray T600SC) of varying fibre length and with areal densities ranging from 20 to 100 g/m2. The attenuation of electromagnetic radiation at frequencies from 30 MHz to 1.5 GHz through the composite was tested according to standard ASTM 4395-99. The recycled carbon fibre had a mean fibre length of 16.4 mm by weight and had an electrical conductivity measured to be 0.5% lower than virgin fibre. Figure 4.21 shows the shielding effectiveness of the different carbon fibre veils. The attenuation increases with increasing areal density of the veil from about 30 dB for a 20 g/m2 veil to 50 dB for a 100 g/m2 veil made from virgin fibre with a uniform fibre length of 14.4 mm, close to the mean length for the recycled fibre veil. The veils made from recycled fibre are shown to have a slightly lower shielding effectiveness that peaks at 40 dB for a 80 g/m2 veil. Figure 4.22 shows the shielding effectiveness for veils with different lengths of virgin fibre. It can be seen that fibre lengths less than about 14 mm give similar shielding but veils with longer fibre give slightly lower shielding. This is understood to be due to the longer fibres being more difficult to disperse, which gives the veil less homogeneity. This is also the reason for the recycled fibre veil having slightly lower

90


Average shielding effectiveness (dB)

60 50 40 30 20 Virgin fibre Recovered fibre

10 0 0

20

40 60 80 Veil areal density (g/m2)

100

120

Average shielding effectiveness (dB)


4.21 Average shielding effectiveness of glass reinforced composite with a layer of recycled or virgin carbon fibre veil with different fibre loadings (Wong et al. 2005).

50 45 40

6.4 mm 9.6 mm

35

14.4 mm 19.2 mm

30

28.8 mm

25 0

20

40

60

80

100

120

Veil areal density (g/m2)

4.22 Average shielding effectiveness of glass reinforced composite with a layer of virgin carbon fibre veil made with different fibre lengths and at different areal density (Wong et al. 2005).

shielding effectiveness as the recycled fibre contains a proportion of longer fibres up to 30 mm in length. The results of the tests show that recycled fibre has good potential for use as a veil for electromagnetic shielding applications but that it is desirable not to have any fibres longer than about 10–15 mm as these may hinder fibre dispersion.


91


Compounding in thermoplastics Short carbon fibres are compounded with thermoplastics to make fibre reinforced polymers suitable for injection moulding. If milled carbon fibre is used, where the fibre length is a few tenths of a millimetre, the main benefit is to make the thermoplastic electrically conducting and this has application to reduce static or provide electromagnetic screening. Longer fibre lengths of a few millimetres provide better mechanical properties where the carbon fibre also acts as a reinforcement. Recycled carbon fibre is suitable for this application as the carbon fibre in the polymer is in the form of dispersed short fibres and an investigation to compound recycled fibre with polypropylene for injection moulding has been undertaken (Wong et al. 2007a). There were two stages in the investigation: firstly the recycled carbon fibre was compounded with the polypropylene in a twin screw compounder to produce pellets for injection moulding with a carbon fibre volume content of 30%. Then the material was injection moulded into test specimens and the mechanical properties measured. Polypropylene is a non-polar polymer and does not bond well to carbon fibre. So the investigation considered the use of maleic anhydride grafted polypropylene (MAPP) as a coupling agent. This was compounded to the polypropylene before the addition of the recycled carbon fibre. One of the problems faced was how to introduce the recycled carbon fibre into the compounding machine. Virgin carbon fibre can easily be fed in either as continuous tow, or discrete carbon fibre pellets that have been manufactured by chopping virgin tow and adding a binder. The fluffy recycled fibre could not be used in this way and the method adopted for the trial was to convert the carbon fibre into a non-woven mat and then cut the mat into small pellets of about 5 mm in size and feed these directly into the compounder. Some of the key results obtained from the investigation are given in Table 4.6. A range of different coupling agents were added in different proportions and the results shown for the G3003 MAPP at 5% addition to the polypropylene gave some of the best properties. Carbon fibre clearly has a strong reinforcing effect in polypropylene but the addition of a coupling agent further significantly increases the tensile and impact strength and also improves the tensile stiffness by a smaller amount. The effect of the coupling agent is shown in Fig. 4.23, in which SEM images of the fracture surfaces are shown. The polypropylene with the coupling agent can be seen to adhere much better to the surface of the carbon fibre than that without. The material with the coupling agent gave properties that compared very favourably with those of commercially available grades of polypropylene reinforced with 30% volume fraction of carbon fibre. This

92


Table 4.6 Mechanical properties of polypropylene reinforced with recycled carbon fibre

Tensile strength (MPa) Tensile stiffness (GPa) Impact strength (kJ/m2)


5 μm

Polypropylene (PP)

PP + 30% recycled carbon fibre

PP with 5% G3003 + 30% recycled carbon fibre

29 1.7 No break

51 14.7 12.7

125 16 26

10 μm

4.23 Scanning electron micrographs of recycled carbon fibre from a fracture surface for a polypropylene/carbon fibre composite with no coupling agent (left hand image) and with the addition of 5% (wt) of a maleic anhydride–polypropylene coupling agent (right hand image) (Wong et al. 2007a).

demonstrates the potential for the use of recycled fibre compounded with thermoplastics.

Thermoset composites The most common applications for carbon fibre as a structural reinforcement involve its use at high fibre volume fraction in thermoset polymer composites and epoxy resin is the most common polymer used. As described earlier, the highest fibre volume fractions (up to 60%) are achieved by using highly aligned continuous fibre and when producing components using prepreg these volume fractions can be achieved at relatively low moulding pressures of less than 10 bar.

93

100 80 60 40 20 0 0.1

0.2 0.3 0.4 Fibre volume fraction LRCF--16.4 mm

0.5


Recycled carbon fibre cannot yet be processed in an aligned form and so current research for the reuse of recycled carbon fibre to make high volume fraction structural composites has focused on the use of the fibre in the form of a non-woven mat. An investigation to determine the best properties that could be achieved with a non-woven mat made from recycled carbon fibre is reported in Wong et al. (2007b). The aim of this study was to attempt to produce a composite which could outperform aluminium as a structural material that had the potential to be used in the aircraft industry for semistructural components in non-safety critical applications. The composite was made by compression moulding layers of non-woven recycled carbon fibre mat interleaved with a fire-retardant epoxy resin film. Plaques 2.5 mm thick were moulded with fibre volume fractions from 20 to 40%. Two grades of non-woven mat were used; both were made from recycled carbon fibre but one had a mean fibre length of 8.7 mm and the other a mean fibre length of 16.4 mm. The tensile strength and stiffness of the composites produced are shown in Fig. 4.24. It can be seen that the tensile properties increased significantly as the fibre volume fraction increased from 20 to 30% but the rate of increase in stiffness is reduced from 30 to 40% and strength decreases. This is understood to be due to increased fibre damage associated with higher moulding pressure at 40% fibre volume fraction and also the difficulty of achieving proper wet-out of the fibre with resin at higher fibre volume fractions due to the decreased permeability of the mat. Figure 4.24 also shows that better properties were achieved for the composite made from the mat with shorter carbon fibre. Analysis showed that both mats had some degree of fibre alignment; this was greater in the mat with shorter fibres and would explain the better properties

Tensile stiffness (GPa)



350 300 250 200 150 100 0.1

0.2

0.3

0.4

0.5

Fibre volume fraction SRCF--8.7 mm

Aligned

Random

4.24 Tensile properties of an epoxy recycled carbon fibre composite (LRCF – long recycled carbon fibre, SRCF – short recycled carbon fibre). The solid and dotted lines represent the theoretical stiffness prediction for randomly distributed and perfectly aligned fibres (Wong et al. 2007b).

30

250 30 vol% fibre

Strength 25

200

Stiffness 30 vol% fibre

20

150 31 vol% fibre

15

100 10 50

5 0

0 Density kg/m3

Specific tensile stiffness (MPa m3/kg)

Management, recycling and reuse of waste composites Specific tensile strength (kPa m3/kg)

94

SFL 1.42

LFL 1.42

2024-T4 2.77

GRP 1.69


4.25 Specific tensile properties of recycled carbon fibre epoxy composites with long carbon fibre (LFL) and short carbon fibre (SFL) compared with aluminium (grade 2024-T4) and a glass reinforced epoxy composite (GRP) (Wong et al. 2007b).

achieved. Despite the problems in manufacturing with a non-woven mat, good mechanical properties were achieved at 30% volume fraction and Fig. 4.25 compares the specific tensile strength and stiffness for the carbon fibre composites with a typical grade of aluminium and a glass fibre composite. Allowing for the reduced density of the carbon fibre composite then, with the properties achieved, it can be shown that for an application in which bending stiffness is the design criterion, a 35% reduction in weight could be achieved compared with aluminium by using the recycled carbon fibre composite with 30% fibre volume fraction. This demonstrates the potential for the use of recycled carbon fibre processed into a non-woven mat. Thermoset moulding compounds using glass fibre with lower fibre volume fractions are used in large quantities in automotive and industrial products. Recycled carbon fibre could be used in similar bulk and sheet moulding compounds to give improved properties and research work is being undertaken to investigate the potential for these materials (Pickering et al. 2006). The aim of the research is to develop BMC and SMC moulding compounds using recycled carbon fibre and epoxy resin with mechanical properties typically double those that can be achieved with glass fibre at similar fibre volume fractions, as shown in Fig. 4.26. At the time of writing the results of this research have not been published but it is understood that the properties aimed for have been exceeded.


95

250 UTS (MPa) Modulus (GPa)

UTS/modulus

200

150

100

50

0 Existing BMC Existing SMC Carbon SMC Recycled CF- Recycled CFBMC SMC


4.26 Target properties for BMC and SMC made with recycled carbon fibre and epoxy resin (Pickering et al. 2006).

The BMCs have fibre volume fractions of 10% and typically 50% filler and can be processed in a similar way to glass fibre BMC. The SMCs are based on non-woven recycled carbon fibre mat and it is found that this does not allow the material to flow during compression moulding to the same extent that can be achieved with a conventional glass fibre SMC. Whilst controlling the viscosity and filler loading of the SMC can improve flow, the presence of a random array of individual fibres in the non-woven mat makes fibre flow much more difficult that in a conventional SMC where the fibre is in the form of discrete bundles that can move more easily over each other.

Achieving high fibre volume fraction – fibre alignment In the investigations in which composites were made from non-woven recycled carbon fibre mat (Wong et al. 2007b) moulding pressure was not measured directly but subsequent work has shown that the moulding pressures required to achieve particular fibre volume fractions when moulding a non-woven mat are given generally by the curve shown in Fig. 4.27. It shows that pressures in excess of 60 bar are needed to achieve fibre volume fractions of 30% and that fibre volume fractions of 10% require moulding pressures of about 5 bar. In contrast, when moulding conventional prepregs,

96

Management, recycling and reuse of waste composites 20 Compaction pressure (MPa)

18 16 14 12 10 8 6 4 2 0 0

10

20 30 Fibre volume (%)

40

50


4.27 The pressure required (determined by measurement) to compress a non-woven recycled carbon fibre mat with a random fibre orientation to achieve a composite with varying fibre volume fraction.

fibre volume fractions of 60% can be achieved with moulding pressures in the order of 5 bar. A composite with aligned fibres will give better mechanical properties in the alignment direction but an equally important effect is that alignment allows higher fibre volume fractions to be achieved and these also give better mechanical properties. Methods of aligning short fibres have been developed in the past (Bagg et al. 1969; Edwards and Evans 1980) and a feasibility study was undertaken to investigate whether recycled carbon fibre could also be aligned (Jiang et al. 2006). In this investigation a slurry of dispersed fibre was fed through a narrow convergent slot onto the inside of a rotating drum lined with a fine mesh. The fibres were aligned as the slurry passed through the convergent slot and they were then laid down tangentially on the inside of the rotating drum. The liquid was centrifuged out through the mesh leaving an aligned fibre mat on the inside of the drum. A series of trials were undertaken with fibres of different length and then the aligned mat was moulded into a composite. A diagram of the test facility is shown in Fig. 4.28. Figure 4.29 shows an image of the aligned mat and the results of the testing of the composites manufactured are given in Table 4.7. Predictions of the expected stiffness of the composites manufactured were made by modelling assuming that, at the fibre volume fraction achieved, all the fibres were perfectly aligned in one direction. These predictions were then compared with the stiffness measurements and it was found that tensile stiffnesses between 80 and 90% of the theoretical maximum had been achieved. It was also noted however, that at a moulding pressure of 5 bar a fibre volume fraction of only 14% was achieved, whereas a fibre volume

Thermal methods for recycling waste composites Containment Rotating drum with mesh screen

97

Carbon fibre slurry feed

50 cm 30 cm

Alignment box Convergent nozzle


4.28 Diagram of fibre alignment test rig (Jiang et al. 2006).

4.29 An image of the alignment of fibres in the mat produced in the fibre alignment test rig (Jiang et al. 2006).

fraction approaching 60% can be achieved with unidirectional prepreg at the same pressure. The image of the aligned mat in Fig. 4.29 shows that most of the fibres appear to be aligned, although a significant proportion are misaligned and it is apparent therefore that this misalignment has a bigger effect on fibre volume fraction than on tensile stiffness.

98


Table 4.7 Flexural properties of aligned recycled carbon fibre/epoxy resin composites. The volume fraction of the carbon fibre was 14% Flexural strength (MPa)

Flexural modulus (GPa)

Fibre length (mm)

L

T

L

T

5 20 Continuous fibre (calculated)

361 ± 20 273 ± 31

142 ± 4.5 128 ± 17

26.7 ± 0.8 23.7 ± 1.0 29.5

9.3 ± 0.5 9.2 ± 0.8 3.5

4.5

Prospects for commercial operation


An assessment of the likely commercial viability of a recycling operation for glass fibre reinforced composites based on the fluidised bed process is described in Pickering et al. (2000). The assessment was carried out for plants sized to process from 1000 to 20 000 tonnes of scrap composites and the following were among the assumptions made: • • •

Scrap producers paid the avoided landfill price to send the material to the recycling plant. The plant was a stand alone operation, equipped with the gas clean-up necessary to process halogenated polymers. The recycled glass fibre would be sold at 80% of the price of virgin fibre.

The analysis showed that a plant would need to process in the region of 10 000 tonnes per year to break even and it was considered that there was not sufficient scrap available within easy transport distance to make a viable operation in the UK. The value of carbon fibre is typically at least ten times that of glass fibre and so it is likely that a plant recycling carbon fibre could break even at a significantly lower throughput. Hence, current research is focused on recycling carbon fibre composites.

4.6

Current research, future trends and sources of further information and advice

Current research is focusing on the recycling of carbon fibre composites since the higher value of carbon fibre makes a commercially viable recycling operation a more easily achievable goal. Further development of the fluidised bed process is taking place to identify the optimum processing conditions that can achieve high recycled fibre quality whilst maximising the throughput of carbon fibre.



99

Of equal importance is research to develop the applications for recycled fibre. This is applicable not only to recycled fibre from the fluidised bed process but for recycled fibre from other composites recycling processes, many of which produce a fibre output that is best utilised as short individual fibre filaments. Applications for the recycled fibre processed directly in its raw form are limited; it has been shown that low fibre volume fraction moulding compounds can be made relatively easily, but, based on carbon fibre, these are new materials and new markets must be developed. Achieving fibre alignment is a key goal for two reasons: it will allow composites with high fibre volume to be manufactured from recycled carbon fibre giving high grade structural properties, and it will also allow recycled carbon fibre to be converted into prepreg materials for which there are existing markets and which is an easy to use and well-understood material. Developing fibre alignment processes is the subject of ongoing research led by the University of Nottingham (www.nottingham.ac.uk/afrecar). The aim is to achieve a degree of alignment that will allow high fibre volume fractions to be achieved in composites and to demonstrate applications in the automotive and aerospace industries. It is recognised that the fluidised bed process is limited in that there is no recovery of material from the polymer and research is currently in progress at the University of Nottingham to develop new recycling processes in which chemical products can be recovered from the polymer. Feasibility studies have already been carried out which demonstrate the potential of using supercritical fluids, and in particular supercritical propanol, as a means of recovering chemicals whilst also yielding very high quality recycled carbon fibre (Jiang et al. 2007, 2009). Research is currently in progress to develop this recycling process at a larger scale at the University of Nottingham (www.nottingham.ac.uk/afrecar). Other carbon fibre recycling activities As the market for carbon fibre increases, particularly for use in the aerospace industry, and as environmental concerns are increasing, the need for recycling, research and development into carbon fibre recycling is taking place in a number of places around the world. Key work is ongoing in three places and further information is available in the references and on the internet as described: •

Adherent Technologies Inc. in New Mexico, USA, have developed a tertiary process for recycling carbon fibre with recovery of chemical products from the polymer (Allred et al. 2002). www.adherenttech.com • Recycled Carbon Fibre Ltd in Birmingham, UK, are a company commercially developing carbon fibre recycling operations based on a proprietary pyrolysis process. www.recycledcarbonfibre.com

100 •

• •


Hadeg Recycling GmbH in Stade, near Hamburg in Germany, are involved in a range of different carbon fibre recycling activities including pyrolysis. www.hadeg-recycling.de Recycling operations are also being developed at CFK-Valley in Stade. www.cfk-valley.com Research in carbon fibre recycling is also being undertaken locally to Stade at the Technical University of Hamburg-Harburg (Schulte et al. 2007).

Whilst these activities are based on different recycling technologies the issues faced in using the recycled carbon fibre are broadly similar. The problems faced and potential applications for use of recycled carbon fibre described in Section 4.4 are not only applicable to the recycled carbon fibre from the fluidised bed process but more generally to the carbon fibre output from these and other recycling processes.


4.7

References

allred r e, gosau j m and shoemaker j m (2002), ‘Recycling process for carbon/ epoxy composites’, Composites in Manufacturing, 18(2): May 13. bagg g e g, evans m e n, et al. (1969), ‘The glycerine process for the alignment of fibres and whiskers’, Composites, 1(2): 97–100. edwards h and evans n p (1980), ‘A method for the production of high quality aligned short fibre mats and their composites’, Economic Computation and Economic Cybernetics Studies and Research, 2: 1620–1635. george p e and carberry w l (2007), ‘Recycled carbon fibre performance in epoxy and polycarbonate matrices’, Composites Innovation Conference 2007 – Improved Sustainability and Environmental Performance, organised by NetComposites, Barcelona, Spain, 4–5 October. goodship v (2007), An Introduction to Plastics Recycling, 2nd edition, Smithers Rapra. jiang g, wong k h, pickering s j, rudd c d and walker g j (2005), ‘Study of a fluidised bed process for recycling carbon fibre from polymer composites’, Proceedings of 7th World Congress of Chemical Engineering, Glasgow, 10–14 July. jiang g, wong k h, pickering s j, walker g s and rudd c d (2006), ‘Alignment of recycled carbon fibre and its application as a reinforcement’, Proceedings of SAMPE Fall Technical Conference, Dallas, USA, 6–9 November. jiang g, pickering s j, lester e, blood p and warrior n (2007), ‘Recycling carbon fibre/epoxy resin composites using supercritical propanol’, 16th Intl Conference on Composite Materials, Kyoto, Japan, 8–13 July. jiang g, pickering s j, lester e h, turner t a, wong k h and warrior n a (2009), ‘Characterisation of carbon fibres recycled from carbon fibre/epoxy resin composites using supercritical n-propanol’, Composites Science and Technology 69: 192–198. Doi:10.1016/j.compscitech.2008.10.007 kennerley j r, kelly r m, fenwick n j, pickering s j and rudd c d (1998), ‘The characterisation and reuse of glass fibres recycled from scrap composites by the action of a fluidised bed’, Composites Part A, 29A: 839–845.



101

kunii d and levenspiel o (1993), Fluidization Engineering, 2nd Edition, Butterworth-Heinemann. monk j f (1997), Thermosetting Plastics: Moulding Materials and Processes, 2nd Edition, Longman. pickering s j (2006), ‘Recycling technologies for thermoset composites – current status’, Composites: Part A, 37: 1206–1215. Doi: 10.1016/j.compositesa.2005. 05.030 pickering s j, kelly r m, kennerley j r, rudd c d and fenwick n j (2000), ‘A fluidised bed process for the recovery of glass fibres from scrap thermoset composites’, Composites Science and Technology, 60: 509–523. pickering s j, turner t a and warrior n a (2006), ‘Moulding compound development using recycled carbon fibres’, Proceedings of SAMPE Fall Technical Conference, Dallas, USA, 6–9 November. schulte k et al. (2007), ‘Optimisation of a pyrolysis process for recycling of CFRPs’, 16th Intl Conference on Composite Materials, Kyoto, Japan, 8–13 July. wong k h, jiang g z, pickering s j and rudd c d (2005), ‘Effects of fibre length and loading on electromagnetic shielding of thermoset composite reinforced with virgin or recovered carbon fibres’, 15th Intl Conference on Composite Materials, Durban, South Africa, 27 June to 1 July. wong k h, jiang g, pickering s j, rudd and walker g s (2006), ‘Characterisation of recycled carbon fibre: mechanical properties and surface chemistry’, Proceedings of SAMPE Fall Technical Conference, Dallas, USA, 6–9 November. wong k h, pickering s j and brooks r (2007a), ‘Recycled carbon fibre reinforced polypropylene composites: effect of coupling agents on mechanical properties’, Composites Innovation Conference 2007 – Improved Sustainability and Environmental Performance, organised by NetComposites, Barcelona, Spain, 4–5 October. wong k h, pickering s j, turner n a and warrior n a (2007b), ‘Preliminary feasibility study of reinforcing potential of recycled carbon fibre for flame-retardant grade epoxy composite’, Composites Innovation Conference 2007 – Improved Sustainability and Environmental Performance, organised by NetComposites, Barcelona, Spain, 4–5 October. yin y, binner j p b, cross t e and marshall s j (1994), ‘Oxidation behaviour of carbon fibres’, Journal of Materials Science, 29: 2250–2254. yip h l h, pickering s j and rudd c d (2002), ‘Characterisation of carbon fibre recycled from scrap composites using a fluidised bed process’, Plastics, Rubber and Composites, 31(6): 278–282. Doi: 10.1179/146580010225003047

5 Pyrolysis for recycling waste composites M. B L A Z S Ó, Hungarian Academy of Sciences, Hungary


Abstract: Pyrolysis is a suitable process for recycling polymer composites because the thermal decomposition products of the polymer matrix evaporate, and thus the reinforcement materials can be recovered and reused. The products of pyrolysis carried out at an appropriate temperature are monomers and other valuable chemicals. This chapter describes the pyrolysis reactions and products of frequently used thermoplastics and thermosets in polymer composites. Published results on pyrolysis of various polymer composites are discussed in order to understand the requirements of successful plastic composite recycling by pyrolysis. The environmental concern related to pyrolysis of flame retardants containing polymer composites is also touched upon and some methods are referred to for decreasing or eliminating toxic and harmful compounds from the pyrolysis products of halogenated flame retardants. Key words: plastics composites, recycling of composite reinforcement, pyrolysis of polymer matrix.

5.1

Introduction

Thermal methods are successfully applied for waste elimination in general. Among them, combustion has been the most commonly used traditional method to reduce the quantity of discarded objects and garbage. However, in the twenty-first century, uncontrolled combustion is strictly banned worldwide because of environmental pollution from the harmful combustion products and carbon dioxide emissions. In up-to-date incineration plants solid wastes are effectively converted to energy and potential harmful products are trapped in various cleaning units and filters. Nevertheless, most of the value of recyclable materials is lost under combustion in incinerators, and an important part of the polymers used in composites is built not only from hydrocarbons but also from amides and nitriles emitting huge amounts of nitrogen oxides (NOx) in addition to carbon dioxide. Therefore pyrolysis is a more adequate thermal treatment technology for waste polymer matrix composites in which the material is heated in a low oxygen environment. Pyrolysis is a thermally initiated chemical process that generally decomposes the organic molecules to smaller ones in an inert atmosphere. The 102


Pyrolysis for recycling waste composites

103

term ‘pyrolysis’ is often misunderstood as meaning a high temperature reaction resulting only in gases and char, but it also means thermolysis of organic macromolecules in the temperature range from 250 to 800 °C resulting in gaseous, liquid and solid products. Above 800 °C carbonization of the organic material occurs, which is applied as a utilization technology mostly for biomass waste. Composites with polymer matrix material are produced for various applications and thus a range of different resins can be found in waste composites. The reinforcement part is typically inorganic material, fibres or ground substance. In special cases, polymer is used as the strengthener as well such as Kevlar. The pyrolysis process is a particularly good thermal recycling method for waste polymer composites because the matrix resin is decomposed to valuable feedstock or fine chemicals which evaporate, while the reinforcement is obtained as the solid residue. The preservation of the original quality of the latter is a crucial point of the pyrolytic recycling technology of polymer matrix composites, the main purpose of which is often the recycling of the valuable fibre or powder strengthener. The thermal decomposition of the polymer matrix of the composite should not differ from that of the polymer alone because matrix and reinforcement materials are separate and distinct on a macroscopic level within the composite. Obviously heat transport may be altered by an inorganic material of quite different heat conductivity from that of the polymer, resulting in differences in the heating rate and temperature gradient in the pyrolyser. Pyrolysis of the organic macromolecules is initiated by a primary intramolecular endothermic chemical reaction of the following types: 1. rearrangement of chemical bonds followed by elimination of small molecules, and/or separation of the macromolecule in two parts; 2. scission of chemical bonds followed by the stabilization of the unstable fragments. The temperature necessary for the initiation reaction depends on the activation energy of the chemical change (reactions of type 1 occur at around 300 °C, and of type 2 at above 500 °C). Initiation of the thermal decomposition may occur due to defect points of the polymer (structural error, oxidized moieties, etc.) that can reduce thermal stability of the polymer. In polymers of flexible macromolecular chain (thermoplastics) the unstable species formed by initiation are generally macroradicals which are either depolymerized in a free radical chain reaction, producing monomers and oligomers, or transferred to another macromolecule or to another part of the same macromolecule. In the case of radical transfer the polymeric radical is divided into two parts: a smaller

104


molecule and a smaller radical, and at the end of the radical chain reaction a series of product of broad volatility range is obtained. In network or cross-linked polymers (thermosets) both types of pyrolysis reaction require higher temperatures because of limited flexibility of the macromolecular segments. Catalysts may facilitate thermal decomposition of polymers by conducting the pyrolysis process through reactions requiring less energy. Thus catalytic pyrolysis generally lowers the optimal temperature of polymer decomposition, and at the same time the products could be different from those obtained without catalyst. Technically catalytic pyrolysis is carried out either in one step or in a coupled arrangement of a pyrolyser and a catalytic converter. Although thermal catalysis proved to be a successful option for utilizing waste polyethylene, this method is not often used for more complicated wastes such as composites.


5.2

Pyrolysis reactions and products of thermoplastics

Polypropylene (PP) and polyamides (PA66, PA6, PA12, PA46) are the most used matrices for thermoplastic composites. Polyesters (polyethylene terephthalate (PET), polybutylene terephthalate (PBT), polyphenylene oxide (PPO)), polycarbonate (PC) and thermoplastic polyurethane (TPU) are also used as matrix material. These polymers liquefy on heating and freeze to a solid state when cooled; they can be remelted. The reinforcement is introduced under melting of the matrix material. Heat-resistant thermoplastics, such as polyetheretherketone (PEEK), polyether sulphone (PES), polyimide (PI), polyetherimide (PEI) and polyphenylene sulphide (PPS) furnish advanced properties to their composites. The basic thermal decomposition records of thermoplastic polymers applied frequently in polymer composites are shown in Table 5.1. The data have been collected from published works in which thermogravimetric measurements were performed in an inert atmosphere and at a heating rate of 10 °C/min.

5.2.1 Polypropylene PP starts to decompose at around 400 °C, when heated slowly (at a rate of 10 °C/min) in a thermogravimeter. The volatile formation is fastest at 470 °C and the total mass of the polymer is volatilized by 500 °C. This means that no solid residue is left after pyrolysis of PP, hence pyrolysis is a promising method for recycling either fibreglass or mineral reinforced PP composites. The surface of glass fibres or other strengtheners is cleaned by the pyrolysis process, which occurs at moderate temperature and does not seriously

Pyrolysis for recycling waste composites

105

Table 5.1 Thermal decomposition data published on some thermoplastics applied as polymer matrix in composites. Data are taken from thermogravimetric measurements in an inert atmosphere, generally at 10 °C/min heating rate

Thermoplast


PP PA6 PA66 PA12 PET PBT TPU PPO PES PEEK PPS PC Kevlar

Decomposition range (°C) 400–500 370–490 390–490 390–500 360–490 340–470 250–400

550–640 450–700 420–600 520–800

Mass loss rate (DTG) max. (°C)

Residual mass (wt %)

470 461 454 465 429 405 330 550 570 590 600 520 600

0 0 0 1 11 3 5 54 48 55 22 34

Reference Czégény et al. (2002) Czégény et al. (2002) Bozi et al. (2008) Czégény et al. (2002) Jakab et al. (2005) Balabanovich (2004) Font et al. (2001) Montaudo et al. (1994a) Montaudo et al. (1994a) Day et al. (1990) Montaudo et al. (1994b) Bozi et al. (2007) Czégény et al. (2002)

influence the original quality of the fibre. In this way pyrolysis may compete with other fibreglass recycling methods (Cunliffe et al., 2003). We may expect that the chemical process of PP thermal decomposition is not altered in the presence of the strengthener if it is sufficiently inert in the pyrolysis processes. Glass is a generally used vessel material in moderate temperature thermal studies of organic substances. Carbon has been observed to have a significant influence on PP decomposition only when the studied carbon black contained non-negligible amounts of functional groups and organic compounds (Jakab and Omastová, 2005). The volatilized pyrolysis products of PP are propylene oligomers produced through a series of free radical chain reactions. The pyrolysis conditions (heating rate, pyrolysis temperature, residence time of the evolved products) influence the product distribution, which is generally composed of some light hydrocarbon gases, a major part of gasoline volatility liquids (including propylene trimer, 2,4-dimethylheptene; tetramer, 2,4,6-trimethyldecene; and pentamer, 2,4,6,8-tetramethyltridecene) and an important amount of higher oligomers of lower volatility. From the point of view of recycling of the whole composite the formula with PP matrix and strong PP fibre strengthener is exceptionally beneficial, resulting in the same pyrolysis products from matrix and strengtheners. Catalytic thermal decomposition of PP may lead to more valuable products of narrower volatility range and of higher storage stability (Ali et al., 2002); however, the separation of the catalyst from the recycled

106


strengthener or a tar-like deposit on the surfaces often observed in catalytic processes of hydrocarbons should be taken into account. The use of AlCl3 as catalyst mixed with the polymer in a batch reactor or in a fluidized bed reduces the process temperature dramatically (Kaminsky and Nuñez Zorriqueta, 2007). The use of a small amount of AlCl3 (0.1%) produces products at 400 °C similar to those obtained at 500 °C in a non-catalytic process from PP, but it is possible to pyrolyse PP at 300 °C when the amount of catalyst is higher. The increase of the amounts of catalyst then leads to an increase of light oil fraction (20 15–20 10–15 5–10 0–5

waste into secondary plastic materials while the latter involves a chemical transformation to hydrocarbon chemicals, i.e. in essence a ‘desynthesis’. Mechanical recycling is multistepped with collection, sorting, heat treatment with reforming, recompounding with additives and extruding operations to produce a recyclate that can substitute for the virgin polymer. The percentage of plastic waste that is subjected to mechanical recycling in Western Europe has been estimated to equal c.22% (Plastics Europe, 2004). There is a significant variation in the extent of mechanical recycle in different countries, as is revealed in the entries to Table 6.1. Mechanically recycled plastics typically find use in low grade applications, e.g. in the production of plastic bags, pipes and as composites in construction (La Mantia, 2002). Chemical or feedstock recycling is based on the


126


decomposition of polymers by means of heat, chemical agents and catalysts to yield a variety of products ranging from starting monomers to constituent hydrocarbons. Incineration of plastic waste can be coupled with energy recovery but this approach continues to meet with strong societal opposition. Moreover, the Kyoto Protocol has set, as a target, a 20% reduction in carbon dioxide emissions by 2010 (Defra, 2000). Investment and operating costs associated with incineration are significant but economies of scale apply with modern municipal solid waste incinerators having a capacity typically in the range 200–1000 ktonnes/year (Brown et al., 2000). The high costs of incineration can be offset by sales of energy, recovered as heat and/ or electricity and the sale of ash and ferrous metal recovered from the combustion process. Chlorine containing waste is typically xenobiotic and having no analogous compounds in nature, there is no natural means of ameliorating the negative environmental impact. A major problem in the recycling of nonbiodegradable polyvinyl chloride (PVC) is the high Cl content, i.e. 56% of the total weight. Incineration of PVC is problematic as it falls into the category of Principal Organic Hazardous Constituents, compounds that are inherently difficult to combust. Complete combustion of such compounds occurs at such high temperatures (>1700 K) to be economically prohibitive while the formation of hazardous by-products (polychlorodibenzodioxins and polychlorodibenzofurans), known carcinogens included on the US Environmental Protection Agency (EPA) Persistent Bioaccumulative Toxics List, can result from incomplete incineration (Hagenmaier et al., 1991; Costner, 1998). Chlorinated dioxins and furans are regarded as the most severe environmental contaminants with toxicities that are orders of magnitude greater than strychnine and sodium cyanide. Immune system damage, reproductive effects/birth defects, cancer and neurological effects have been established for short-term exposure to low concentrations (Thornton, 2000; Pollock, 1989). A comprehensive legislation has been developed to address issues associated with the management of waste plastics. There are a number of European Directives in place (notably the Waste Packaging, End-of-Life Vehicle and Electrical/Electronic Equipment Directives) that set increasingly stringent standards to minimise environmental impact due to waste plastics (Aguado et al., 2006). These directives particularly target a decrease in plastic waste sent to landfill and encourage an integrated approach to encompass prevention of waste at the production stage, reuse to extend productive lifetime, recycling and energy recovery. A chemical processing leading to feedstock recycling as applied to polymer waste represents a positive alternative to landfill with the potential to recover energy and/or raw materials. This approach can serve as a progressive response to existing and impending legislation but viability remains an issue.

Catalytic processing of waste polymer composites


6.3

127

Thermal decomposition of waste plastics

Thermal degradation (pyrolysis and/or cracking) is accepted as a feasible means of waste plastic reuse (McCaffrey et al., 1998; Lingaiah et al., 2001b). Indeed, the viability of thermal cracking has been established (Sakata et al., 2003) and has found application in a number of processing plants (Brophy and Hardman, 1996; Kaminsky and Sinn, 1996). Plastics can be divided into two groups: (i) condensation polymers; and (ii) addition polymers (polyolefins). Condensation polymers, which include such materials as polyamides, polyesters and nylon, can be ‘depolymerised’ via reversible synthesis to the starting diacids and diols or diamines. The latter involves alcoholysis, glycolysis and hydrolysis and is known to deliver high yields of the starting monomers (Cornell, 1995). In contrast, addition polymerisation is not readily reversed as it is an activated process. The thermal cracking of common plastics such as polyethylene and polypropylene has been reported in the literature (Songip et al., 1994; Sharratt et al., 1997; Costner, 1998). Ucar et al. (2002) have evaluated a co-processing of waste plastic with a heavy petroleum fraction blend but a narrow product distribution has yet to be realised. A co-processing of waste plastics with coal has also been considered as a route to fuel production (Ding et al., 1999). The main drawback to thermal degradation is the requisite high temperatures (process is highly endothermic) that results in a very broad range of products (Babu and Chaurasia, 2003; Demirbas, 2004; Kaminsky et al., 2004).

6.4

Catalytic approach to polymer recycling

Catalytic degradation provides control over the product (composition/ distribution) and serves to lower significantly the degradation temperature (Garforth et al., 1998). While thermal degradation of polyolefins results in a random scissioning of long polymer chains, the use of a solid catalyst is known to break the chain into smaller units resulting, for instance, in a higher yield of a saturated liquid product (Sakata, 1998). Songip et al. (1994) have proposed an effective process whereby the waste plastic is first cracked thermally in pyrolysis plants and the oil that is produced is then transferred to a catalytic cracking plant and converted to gasoline. Sufficient progress has now been made in the catalytic processing of polyethylene/polypropylene over solid acids (notably silica–aluminas and zeolites) to enable some optimisation of liquid, solid or gaseous hydrocarbon product fractions through the judicious choice of operating conditions, i.e. reaction temperature and reactor configuration (Kaminsky, 1985; Uemechi et al., 1989; Mordi et al., 1994; McCaffrey et al., 1995). There is also evidence that the product composition resulting from polyethylene and polypropylene


128


degradation can be manipulated through modifications to the catalyst properties (Sharratt et al., 1997; Negelein et al., 1998; Hesse et al., 2001; Serrano et al., 2005; Aguado et al., 2007). If the polymeric waste contains a halogenated component, as in the case of PVC, thermal degradation generates products with widely varying molecular weights and uncontrolled Cl content (Mordi et al., 1994; Blazó et al., 1995; Williams and Williams, 1999). The tightening legislation makes it essential that the Cl component must be removed from any waste plastic derived gas or oil before it can be used (Lingaiah et al., 2001a). One possible means of imposing control over product distribution is through catalytic degradation, i.e. a catalytic dechlorination can be employed to efficiently remove the Cl component where the starting Cl is converted to HCl, which is easily separated from the target product(s). It was observed in a stepwise pyrolysis of a mixed plastic/PVC feed that the majority of the Cl content was released from PVC (at c.573 K) prior to the degradation of the remaining polymers in a second step (>673 K) (Yanik et al., 2001b). Bockhorn et al. (1998) achieved a 99.6% elimination of Cl in a mixed PVC/plastics feed in a cascade of well-stirred reactors at 603 K. In the pyrolysis of municipal waste plastics, thermal degradation leads to the formation of conjugated double bonds, while the HCl released from PVC attacks these double bonds resulting in the formation of toxic chloroorganic compounds (Martinson et al., 1988). Chlorine separation from household waste plastics by PVC decomposition has been attempted using extruders in oil reclamation plants in Japan (Kaminsky and Kim, 1999). The waste plastics are pressurised and HCl gas is generated but, at high pressure, the HCl can readily react with other organic/inorganic components in the feed and complete dechlorination is problematic. The use of metal oxides as HX adsorbents (Horikawa et al., 1999; Yanik et al., 2001a) and lime and KOH for HX fixation (Kaminsky and Kim, 1999; Brebu et al., 2006) has met with limited success. Separation of halogenated polymers from a nonhalogenated polymeric matrix is possible by supercritical fluid extraction but this approach is very energy intensive, which minimises the overall benefit in any recycle operation (Gamse et al., 2000). A PVC recycle strategy involving a one step catalytic processing with concomitant dechlorination certainly represents a progressive option. A catalytic approach to waste polymer reprocessing has the following positive features: (a) lower operating temperatures (relative to pyrolysis), non-oxidative process with lower energy requirements and no directly associated NOx/SOx emissions; (b) absence of thermally induced free radical reactions that can lead to toxic intermediates; (c) possibility of selective Cl removal (in the case of PVC) with concomitant polymer degradation to a target recyclable product; and (d) operability in a closed system with no toxic emissions.


129

Various definitions of catalysis have been proposed, but an early classification provided by Wilhelm Ostwald in 1895 is still widely in use: ‘Catalysts are substances which change the velocity of a reaction without modification of the energy factors of the reaction’. This definition excludes substances that accelerate the rate of reaction by entering into reaction with a resultant disruption of the reaction equilibrium. A catalyst works by forming chemical bonds to one or more reactants which facilitates their conversion (Butt, 2000). A more rigorous definition of a catalyst is ‘a substance that increases the rate of reaction without modifying the overall standard Gibbs energy change in the reaction’. This is illustrated schematically in Fig. 6.4. In an uncatalysed transformation collisions between participating molecules must possess sufficient (activation) energy to pass over the energy barrier that is characteristic for that reaction. The reactants form a short-lived transition state or activated complex which reacts to give product(s) (Gates, 1992). Catalyst/reactant interactions serve to lower the energy barrier (transition state is at a lower energy) and enable reaction under more moderate reaction conditions. The catalyst reduces the enthalpy of activation for the forward reaction by exactly the same amount as it reduces the enthalpy for the reverse reaction, there is no entropy change and the free energy change/position of equilibrium remains the same for the catalysed and uncatalysed processes.

Transition state

Activation energy of uncatalysed forward reaction

Activation energy of uncatalysed reverse reaction

Transition state

Energy


6.4.1 Fundamentals of catalysis

Reactant(s) DG Product(s)

Activation energy of catalysed forward reaction

Activation energy of catalysed reverse reaction Reaction coordinate

6.4 Energy diagram for a catalysed and uncatalysed chemical reaction.

130



6.4.2 Heterogeneous catalysis Catalysts can be divided into two broad categories: heterogeneous and homogeneous (Steinfeld et al., 1999). Homogeneous catalysis occurs when the catalyst is uniformly dispersed in the reaction mixture, either a gaseous or liquid solution. In a heterogeneous reaction, the catalyst is in a different phase from the reactants, where the reaction occurs at the surface of a solid (catalyst) particle in contact with the gaseous or liquid solution. The main disadvantage associated with heterogeneous when compared with homogeneous catalyst operation is the lower effective concentration of catalyst as the reaction occurs only on the exposed active surface. Moreover, physical transport constraints can result in lower effective reactant concentrations at the catalyst surface relative to the bulk fluid. This is illustrated in Fig. 6.5 for a reaction involving liquid and gaseous (H2) reactants promoted by a solid catalyst where the effective H2 concentration shows a decrease in moving from the gas phase (Cg) to the liquid phase (Cl) to the catalyst surface (Cs), where the reaction occurs. Catalyst recovery and reuse are, however, far more facile in the case of heterogeneous operation. Any reaction that is promoted by heterogeneous catalysis involves a number of steps: transport of reactant(s) to the catalyst surface; formation of reactant/catalyst complex; surface reaction leading to product/catalyst complex formation; transport of product from the catalyst surface. The surface reaction is facilitated by reactant/catalyst interaction(s) that generates a reactive reactant/catalyst complex. These interactions may be weak, of the ‘non-bonding’ type with the reactant staying intact while sticking to (or adsorbing on) the surface. Alternatively, the interactions may involve the formation of new chemical bonds between the surface atoms and the adsorbed molecule, which necessarily involves extensive

Liquid–solid interface Catalyst particle Gas phase Gas bubble H2

Bulk liquid Gas–liquid interface Solid external surface

Cg Cl

Surface reaction Cs

6.5 Mass transfer processes associated with the transport of H2 to react with a liquid phase reactant at a solid catalyst surface.


131


reorganisation of the bonding within the reactant. This reactive type of interaction is generally known as ‘chemisorption’ in contrast to ‘physisorption’, the term used to describe the weaker unreactive binding of molecules to surfaces (Twigg, 1989; King and Woodruff, 1990). Physisorption involves weak van der Waals interactions and the adsorption energy is typically in the range 5–15 kJ mol−1, which is much lower than that associated with chemical bonding. Moreover, the van der Waals interaction(s) between adsorbed molecules does not differ significantly from the van der Waals interaction(s) with the surface, with the result that many layers of adsorbed molecules may be formed, as shown in Fig. 6.6. In the case of chemisorption, the reactant may chemisorb intact or it may dissociate (Fig. 6.6); the chemisorption energy is 30–70 kJ mol−1 for molecules and 100–400 kJ mol−1 for atoms (King and Woodruff, 1990; White, 1990; Bowker, 1998). The magnitude of the adsorption coefficient depends on the nature of the surface and the chemical identity of the reacting species. Once the reactant is bound to the surface, it can readily undergo reactions which take place only with difficulty in the gas or liquid phases. This may result from the close proximity of reactant molecules on the surface and/or the changes in bonding consequent upon chemisorption; both are essential (a) O O O O O O O O O O O O O O O O O O O OO O O OO OO O O OO OO O O O O O O O O O O O O O O O O O O O

(b) O O O O O O

(i) O O

(ii) O O

O O

O O

6.6 Schematic showing (a) physisorption and (b) chemisorption involving (i) associative and (ii) dissociative interactions.


132


features of the catalytic properties of the solid. The choice of a suitable catalyst for a particular reaction depends on the stability of the complexes formed between reactant and catalyst and/or product and catalyst. These must be stable enough to form and provide an alternative pathway to the uncatalysed reaction but they must not be too stable as this would lead to an increase in the associated activation energy with a consequent lowering of reaction rate. A heterogeneous catalyst is present as separate particles or agglomerates of particles immersed in a fluid medium in motion. Reactants and products diffuse in the gas or liquid phases at the boundary of the solid and in the pore spaces of the aggregates. Catalyst efficiency is assessed in terms of three parameters: activity; selectivity; lifetime. The activity is the extent to which the catalyst influences the rate of change of the degree of advancement of the reaction, i.e. reactant conversion (per unit mass or per unit volume of catalyst) under specified conditions. The activity per unit volume is of practical importance in terms of process economics where a low catalyst bulk density reduces the necessary reactor volume and associated cost. The turnover frequency represents the specific rate and is defined as the number of molecules reacting per active site per unit time. The usefulness of turnover frequency values is dependent on the validity of the method used to measure (or estimate) the number of active sites (Gates, 1992; Somorjai, 1994). Very often a reactant or set of reactants may simultaneously undergo several parallel reactions, giving different products that can react further in consecutive reactions to yield secondary products. Selectivity is an important catalyst property, serving as a measure of the extent to which a particular catalyst promotes the formation of a ‘target’ product, i.e. the ability of the catalyst to direct conversion to a desired product. The productive lifetime of the catalyst is the period during which the catalyst delivers a product yield in excess or equal to that designated (Twigg, 1989).

6.4.3 Catalyst structure Catalytic efficiency is influenced by four principal catalyst structural features which are interrelated: (a) the exposed area in contact with the fluid; (b) the intrinsic surface chemical reactivity; (c) surface topography – geometric and electronic features; and (d) occurrence of defects – vacancies, interstitials and dislocations. Commercial catalysts must possess sufficient mechanical strength to resist losses as a result of crushing (in packed bed operation) or attrition (in reactors involving vigorous agitation). High surface areas can be attained either by fabricating small particles or clusters where the surface-to-volume ratio of each particle is high, or by creating materials where the void surface area (pores) is high compared with the amount of bulk material. Many catalysts are porous solids with a high



133

surface area that is both ‘external’ and ‘internal’, the former represented by the envelope surrounding discrete particles. The internal surface comprises the walls of the pores/channels/cavities and the total surface area equals the sum of the external and internal areas. Gas adsorption methods, notably the Brunauer–Emmett–Taller (BET) approach, are widely used to determine surface areas (Ponec et al., 1974; Anderson, 1990). Porosity is a measure of the fraction of the bulk volume that is occupied by pore or void space. Pore size distribution is an important characteristic of porous catalysts where pores of diameter in excess of 50 nm are considered macropores, those less than 2 nm are termed micropores and pores of intermediate size are denoted mesopores (Lecloux, 1981; Gates, 1992). It must be stressed that a wide range of pore sizes, spanning both micro- and macro-porosity, is characteristic of standard solid catalysts. The pore size distribution is an important factor in controlling diffusion of reactants/products within any catalyst pore network and is an essential characteristic property of the catalyst. A distinction must then be drawn between the true catalyst density (solid mass to volume ratio excluding all pores and voids) and bulk or packing density. The location of the catalytically active component within the porous structure and the manner in which pores interconnect can have a profound effect on the accessibility of reactants to the catalytically active site, and to the removal of products. The catalyst particle can be a complex entity composed of a porous solid serving as a support for one or more catalytically active phases. The latter may comprise clusters, thin surface mono- or multi-layers or small crystallites where interaction with the support can impact on surface reactivity. The major active component is typically expressed on a percentage weight basis (e.g. 10% w/w Pd/Al2O3) where the crystallographic form of the support (e.g. γ-Al2O3) should be given. Secondary components or additives can serve as promoters where this modification may be directed towards enhancing activity/selectivity, poison resistance or textural properties. The determination of the surface chemical composition and structural properties, as opposed to bulk characteristics, can call upon a range of complementary surface science techniques. The structures of catalyst surfaces are notoriously difficult to elucidate owing to the involvement of microscopic and even macroscopic regions with different compositions, phases and structures, each bearing a diversity of imperfections. Adsorption/desorption measurements provide indirect structural information while transmission electron microscopy is applicable for surface analysis at the nanoscale level. In terms of catalysis, the atomic scale structure is critical, i.e. the arrangement of atoms involved in chemical bonding with reactants. X-ray diffraction is applicable to measurement of crystallite sizes and identification of any crystalline phases that are present. Electron spectroscopies, notably Auger electron spectroscopy (AES), x-ray


134


photoelectron spectroscopy (XPS) and secondary ion mass spectroscopy (SIMS), facilitate measurement of the chemical composition of reactive surfaces. Instrumentation developments are now directed towards detection of finer detail, i.e. atomic spatial resolution, ever smaller energy resolution and shorter timescales. In those cases where experimental characterisation techniques are impractical, theoretical calculations (ab initio, semi-empirical and force field methods) can be instrumental in gaining a better understanding of catalyst structure and the feasible transition states/ reaction pathways. Certain catalytic reactions proceed at the same rate regardless of the nature of the reactive surface and are deemed to be structure insensitive, whereas other reactions exhibit an appreciable structure sensitivity where the rate can vary by orders of magnitude from one crystal face to another (Gates, 1992). Masel (1996) has noted that all catalytic reactions exhibit some degree of structure sensitivity under certain reaction conditions. Heterogeneous catalysts cannot be regarded as representing a ‘model’ uniform reacting surface but display a distribution of interaction energetics associated with the different exposed crystal faces, occurrence of dislocations, defects and other disturbances (Boudart and Djéga-Mariadassou, 1984). The simplest surfaces can be regarded as ‘flat’ and deviations from this ideal arrangement include ‘ledges’, ‘kinks’, ‘adatoms’ and ‘vacancies’ (Lang and Kohn, 1970). The relative concentration of atoms in the ordered domain (flat surface) and in defects depends on surface preparation/pretreatment. Variation in catalytic particle size can result in a change in the distribution of sites and the preponderance of a particular defect which can result in a structure-sensitive response.

6.4.4 Zeolites Zeolites represent a commercially important branch of advanced ceramic catalytic materials that have now found use in waste polymer recycling applications. Zeolites are aluminosilicates that are structurally unique in having cavities or pores with molecular dimensions as part of their crystalline structure (Breck, 1974; van Bekkum, 1991). Zeolites occur naturally as minerals and are extensively mined in many parts of the world. The zeolite materials that find widespread use in the chemical industry are synthesised where the synthesis is controlled to produce a specific zeolite structure tailored for a particular application. The zeolite aluminosilicate framework is composed of oxygen tetrahedrons, each encasing either a Si or Al atom. The oxygen atoms can be shared by only two tetrahedra, and no two Al atoms can share the same oxygen atom, with the resultant restriction that the Al/O ratio ≤ 1. The zeolites finding the greatest application on a commercial scale belong to the family of faujasites and include zeolite X


135


and zeolite Y. The framework structure of zeolites X and Y is shown in Fig. 6.7(a) and is based on a regular arrangement of truncated octahedral and sodalite cages to generate a high surface area microporous structure. Zeolite Y is synthesised by a gelling process, is characterized by a void volume fraction of 0.48, a Si/Al ratio of 2.43 and is thermally stable up to 1063 K (Bhatia, 1990). The geometrical crystalline features associated with zeolite Y are evident from the scanning electron micrograph (SEM) presented in Fig. 6.7(b). Zeolite crystal structures are complex threedimensional frameworks with long-range crystalline order and pore sizes of sub-nano dimensions. Access to the intracrystalline Y zeolite sites is via an interconnecting, three-dimensional network of cavities, i.e. the accessible supercages of internal diameter 1.3 nm that are linked by shared rings of 12 tetrahedra (free diameter = 0.7–0.8 nm) and the less accessible sodalite units that are linked through adjoining rings of six tetrahedra which form the hexagonal prisms (free diameter = 0.20–0.25 nm). The size of the zeolite window is determined by the number of oxygens in the ring, as revealed in Table 6.2. This makes for a molecular ‘sieving’ effect where molecules can pass freely through the zeolite matrix or transport can be severely restricted or blocked depending on the relative dimensions of the

(a)

(b)

1 μm

6.7 (a) Structure of faujasite and (b) SEM micrograph showing the topographical features of Na-Y zeolite.

136

Management, recycling and reuse of waste composites Table 6.2 Relationship between the number of ring oxygens and zeolite ‘window’ dimensions Number of ring oxygens

Window diameter (Å)

4 5 6 8 10 12

1.2 2.0 2.8 4.5 6.3 8.0

Na+ O

O

O

–

Si O

Na+ O

Al O

O

O

Si O

O

–

Si O

O

O

O

Al O

O

Si O

O

O


6.8 Neutral sodium (charge) balanced zeolite framework.

incoming molecule and the zeolite cavities. This size/sieving property is put to good effect in separation applications. The void spaces in the crystalline structure of zeolites present a high capacity for adsorbates and the internal surface area typically provides the predominant contribution to the overall uptake. Microporous materials that are based on the oxides of metals other than silicon and aluminium have stretched the range of materials that are ‘zeolitic’ in nature (Dwyer, 1988). The majority of oxide structures with a well-defined porous structure are now lumped together and classified as zeolites – the term ‘zeotypes’ has emerged as a generic description. Zeolites can operate over a range of acid/alkaline conditions. As silicon is tetravalent and aluminium is trivalent, the zeolite framework has a net negative charge that is balanced by an exchangeable cation (typically Na+), as shown in Fig. 6.8. Zeolites have found widespread use in ion exchange as the indigenous charge balancing cations are not fixed rigidly to the hydrated aluminosilicate framework and are readily exchanged with metal cations in solution (Breck, 1974). In environmental remediation applications, both synthetic and naturally occurring zeolites have been used to remove a range of toxic heavy metals from water (Ouki and Kavannagh, 1997; Ahmed et al., 1998; Kim and Keane, 2000). Solution pH has a significant impact on zeolite exchange properties where a sufficiently low pH can cause structural damage while metal hydroxide precipitation/ deposition may predominate at high pH (Kim et al., 2001). Zeolite addition to water is accompanied by an immediate solution pH increase as a result


137

of a hydrolysis of the zeolite (Lutz et al., 1990), which in the case of zeolite Na–Y can be shown as

Na−Y + (H 2 O)X H−Y + (H 2 O)X − 1 + Na + + OH − The ion exchange of divalent metal (M2+) ions with Na-Y can be represented by the equilibrium (Breck, 1974):

M s2+ + 2Na z+ ↔ 2 Na s+ + M 2z + where s and z represent the solution and zeolite phases, respectively. The degree of divalent ion exchange is dependent on the zeolite composition (Si/Al ratio), size of the exchanging hydrated metal ions, metal ion concentrations and temperature (Keane, 1994). Metal ion exchange also serves as a synthetic route to supported metal catalysts where a reduction (in hydrogen) of the divalent metal exchanged zeolite generates a supported zero valent metal phase according to the equilibrium:


M 2z + + H 2 ↔ M 0z + 2 H +z Zeolite supported transition metal catalysts have been used to promote a range of hydrogenation/dehydrogenation, hydroisomerization, dehydrocyclization and hydrogenolysis reactions (Coughlan and Keane, 1991; Stanislaus and Cooper, 1994; Meriaudeau and Naccache, 1997). Two surface hydroxyl groups (Brønsted acid sites) are generated for each reduced divalent metal and these impart a surface acidity that can be employed to promote catalytic transformations that require acid sites, e.g. alkylation and dehydration (Coughlan and Keane, 1990, 1992a; Park and Keane, 2001). Brønsted acidity can also be introduced through hydrolysis (as shown above) and by zeolite exchange with NH4+ followed by thermal treatment (Coughlan and Keane, 1992b; Bortnovsky et al., 2001). The ability of zeolites to preferentially sieve molecules can be put to good effect in catalytic applications in that the production of a chemical of particular size and/or shape may be preferentially promoted (Meisel et al., 1976). Where a reactant is sterically hindered in accessing the active sites located within the zeolite pore network, then the product resulting from that reactant will also be restricted. Alternatively, if a ‘bulky’ product is formed within the zeolitic cavities, its intracrystalline diffusional transport will also be restricted. The molecular sieving properties of zeolites, in consort with the dual functionality (metal and acid sites) of the zeolite surface place these materials in a unique category of highly efficient, selective and widely applicable catalysts.

138



6.4.5 Catalyst deactivation It is typical that a heterogeneous catalytic unit operation exhibits a progressive drop in conversion with catalyst use. Catalyst deactivation as a function of time is often unavoidable and the deactivated catalyst must either be regenerated or replaced (Bartholomew, 1984). The causes of catalyst deactivation are numerous but they can be conveniently grouped into three general categories (Butt, 2000): sintering, poisoning and coking. Sintering refers to a diminution of active site dispersion and can apply to all phases present in the catalyst, i.e. active phases, modifiers and support. The overall effect of sintering is a reduction in active surface area per unit volume of catalyst and is normally the result of excessively high reaction temperatures (Wanke and Flynn, 1975). The presence of deactivating species in the reactant feed (as impurities) or formed during reactant conversion (transformation of an intermediary and/or product) can induce a partial or total loss of activity. Catalyst poisoning can be irreversible (true poisoning), reversible or transient (inhibition) or may involve fouling agents which induce a mechanical inhibition. The latter refers to non-covalent bonding (van der Waals interaction, hydrogen bonding, ionic interactions, etc.) that serve to physically impede access of reactants to the active sites. True poisoning involves strong chemical interaction with the active sites where catalytic activity cannot be recovered without drastic change in the operating conditions (Fitzharris et al., 1982; Butt and Peterson, 1988). The time-dependent loss of activity can be linked to a migration of active species into the catalyst pellet, morphological changes of surface crystallites of a given phase, change in the number of steps, kinks and vacancies on the surface and modifications to the surface/bulk composition ratios. Activity loss due to coke formation is typical of reactions involving hydrocarbons and is due to reactant or product degradation that produces a carbonaceous residue on the surface. Coke that accumulates on a catalyst may cause deactivation either by covering active sites or by occluding the pores in the catalyst, as shown in Fig. 6.9. Coke deposits can amount to 15–20% w/w of the catalyst, depending on the operating conditions and the nature of the catalyst and reactant(s) (Forzatti and Lietti, 1999). Moreover, carbon deposition on reactor tubes and heat exchanger surfaces can adversely affect the performance of an array of unit operations associated with catalytic processing. Such deposits invariably contain both carbon and hydrogen with H/C ratios varying from almost zero up to 2 (Bond, 1997). The carbonaceous byproduct deposits are a complex mixture of amorphous and graphitic structures (Espinat et al., 1990; Shimada et al., 2000). Considerable effort (use of promoters/catalyst regeneration) is expended to minimise carbon deposition and extend the productive lifetime of the catalyst (van Santen and Jansen, 1991; Bond, 1997; Thomas and Thomas, 1997). The catalyst regeneration


139

Pores Catalyst pellet Metal

Support


Coke

Metal

6.9 Catalyst deactivation by coke deposition.

strategy depends on the cause(s) of deactivation (Peterson and Bell, 1987; Bhatia et al., 1989). Deactivation due to carbon deposition can be reversed by heating the spent catalyst in air/oxygen which serves to ‘burn off’ the carbon deposit. This oxidative (highly exothermic) treatment must be carefully controlled in order to avoid any possible sintering due to excessive high temperature fluctuations.

6.5

Catalytic treatment of non-halogen containing polymer waste

Feedstock recycling applied to waste plastics, sometimes known as chemical recycling or tertiary recycling has enormous potential to enhance waste recovery. This approach does not have the negative societal impact of incineration and the recovered materials may have a broader range of applications than mechanically recovered plastics (Hardman and Wilson, 1998). In the catalytic degradation of non-halogenated plastic waste, the majority of published studies have focused on polyolefins, notably high


140


density polyethylene (HDPE), low density polyethylene (LDPE), linear low density polyethylene (LLDPE) and polypropylene. This is largely due to the fact that these polymers make up the largest component (60–70%) of municipal solid waste plastics (Lin et al., 1997). There have been a limited number of studies (Nambu et al., 1987; de la Puente and Sedran, 1998; Serrano et al., 2000; Ukei et al., 2000) describing the catalytic cracking of polystyrene over a range of zeolite (HZSM-5, HMOR and zeolite Y), SiO2–Al2O3, BaO and sulphur promoted zirconia catalysts. In these studies, the operating temperature (623–773 K) employed during the catalytic processing was appreciably lower than that required for thermal cracking. In the catalytic degradation of HDPE (over the temperature range 563–703 K), the yield of volatile hydrocarbon product increased in the order: HZSM-5 > HY ≈ HMOR (Garforth et al., 1998). A two stage catalytic degradation process involving amorphous SiO2–Al2O3 and H-ZSM-5 zeolite has been used to convert polyethylene into a gasoline fraction with a high octane number (Uemichi et al., 1999). In a two-stage process involving degradation of HPDE and polypropylene over a solid acid followed by hydrogenation over a Pt catalyst, Walendziewski and Steininger (2001) obtained a 90% yield of gas and liquid fractions with boiling points less than 633 K. Aguado et al. (2000) have achieved good selectivity (40–60%) to C5–C12 hydrocarbons at 40–60% conversion of polypropylene and polyethylene over β– zeolite. Product distribution has been demonstrated to depend on zeolite pore size and surface acidity where HZSM-5 preferably generated C3–C5 products compared with HY, which delivered a higher yield of C3–C8 (Negelein et al., 1998; Hesse et al., 2001). The molecular sieving or shape selectivity properties of microporous zeolites can be used to exert some control over product composition. While the studies conducted to date have considered ‘fresh’ catalysts, there is some evidence (Ali et al., 2003) that ‘used’ zeolites are effective in the processing of HDPE, which is significant in terms of the cost of the overall feedstock recycle operation. As hydrocracking is an exothermic process (Scheirs, 1998) it presents advantages over the endothermic pyrolytic options. Moreover, hydroprocessing can facilitate a reduction in aromatic content and transform heteroatoms (e.g. O, N and S) that are present in the plastic waste. Hydrocracking operation involves a complex network of hydrogenolysis, hydrogenation, β-scission and isomerisation reactions (Weitkamp, 1975). High hydrogen pressures are typically employed to hydrogenate coke precursors and limit poisoning by sulphur-containing compounds (van Veen, 2002). Hydrocracking catalysts must be bi-functional, containing a metal function (Ni, Pd, Pt, Co/Mo, Ni/Mo and Ni/W) to promote hydrogen mediated steps and an acid function (see Section 6.4.4) to catalyse cracking (and isomerisation) steps. The balance of both functions is critical in that a strongly acidic catalyst with a relatively weak metal function will preferentially produce



141

naphtha (light product) with a high degree of isomerisation, resulting in a high octane number. On the other hand, a weak acidic function in combination with a very active metal phase will lead to a high quality middle distillate product. Hydrocracking has been applied in a two-step process to treat the product of a first stage thermal or catalytic pyrolysis (Ali and Siddiqui, 2005; Joo and Guin, 1997). The available literature deals with a range of feeds: polyethylene, polypropylene, polystyrene and mixed plastics (Ding et al., 1997; Walendziewski, 2002; Hesse and White, 2004); municipal plastic waste (Feng et al., 1996; Walendziewski and Steininger, 2001); co-processing of plastics with coal (Feng et al., 1996; Luo and Curtis, 1996; Rothenberger et al., 1997; Wang and Chen, 2004); co-mixing of polymers with different refinery oils such as vacuum gas–oil (Ucar et al., 2002; Karagoz et al., 2003b; Cakici et al., 2004). Processing to date has been limited to batch (autoclave, 30–120 bar) treatment with volumes ranging from 20 cm3 to 1 dm3 and temperatures in the range 623–723 K. Luo and Curtis (1996) have demonstrated an appreciable dependence of product yield and selectivity on the nature of the catalyst. Ding and co-workers (1997) have obtained a liquid fraction from comingled plastics that approaches a boiling point response close to that of commercial premium gasoline. More recently, Metecan et al. (2005) obtained a product with a yield >85% (from the hydrocracking of HDPE) that exhibited the characteristics of a commercial naphtha. However, in the majority of these studies the catalyst has been poorly characterised, if at all. Consequently there is a dearth of information that provides an explicit link between catalyst structure and hydroprocessing activity and selectivity.

6.6

Catalytic treatment of halogenated polymers: focus on polyvinyl chloride

The catalytic processing of halogenated plastics has been conducted in laboratory scale batch (Uddin et al., 1999, 2002), semi-batch (Brebu et al., 2002) and fixed bed (Sakata et al., 2003) reactors under atmospheric pressure. These unit operations involved the use of hydrogen. Catalytic hydrogen mediated dehalogenation is an emerging methodology for a low energy treatment of a range of halogenated waste streams involving the selective cleavage of one or more C–X bonds, lowering toxicity and generating reusable raw material (Keane, 2003). A distinction should be drawn between hydrodehalogenation and dehydrohalogenation, the former involving hydrogenolysis of C–X bonds. The latter describes the internal elimination of HX where an external hydrogen supply is not necessary but can serve to limit catalyst deactivation (Tavoularis and Keane, 1999). Urbano and Marinas (2001) have noted that the ease of C–X bond scission decreases in


142


the order R–I > R–Br > R–Cl >> R–F, which matches the sequence of decreasing bond dissociation energies. In the treatment of polychlorinated reactants, a range of partially chlorinated products have been obtained where product composition depends on the nature of the catalyst and process conditions, i.e. temperature, concentration, residence time, etc. (Shin and Keane, 2000). Hydrogen mediated dehalogenation has been successfully promoted using supported Pd (Coq et al., 1986; Bodnariuk et al., 1989), Pt (Creyghton et al., 1995), Rh (Coq et al., 1986; Ukisu and Miyadera, 1997) and Ni (Shin and Keane, 1999; Shin et al., 1999; de Jong and Louw, 2004) catalysts. Based on three comprehensive reviews of the available dehalogenation literature (Lunin and Lokteva, 1996; Urbano and Marinas, 2001; Alonso et al., 2002), it is clear that Pd is the most active metal. However, Pd catalysts suffer from appreciable deactivation with time-on-stream (Armendia et al., 1999). Catalyst deactivation has been attributed to coke deposition (Creyghton et al., 1995), the formation of surface metal halides (Park et al., 2002; Keane, 2004; Murthy et al., 2004) and metal sintering (Ohtsuka, 1989). Thermogravimetric analysis of PVC has established a two-stage decomposition: c.60% mass loss up to 693 K corresponding predominantly to the evolution of HCl and the release of low molecular weight hydrocarbons; a secondary higher temperature mass loss (c.20%) attributed to the release of more complex volatile hydrocarbon species, including aromatics (Jiminez et al., 1993; Chatterjee et al., 1994; Keane, 2007). Below c.623 K, dehydrochlorination involving free radical reactions results in the formation of C=C bonds accompanied by the near stoichiometric evolution of HCl (Murty et al., 1996), i.e.

(−CH − CHCL −)n → (− HC = CH − CH = CH −)n + nHCL Above 623 K the dechlorinated polymer undergoes further cracking and pyrolysis to linear and cyclic compounds while the aromatic to aliphatic hydrocarbon ratio is reversed above 803 K (Yassin and Sabaa, 1990; Marcella and Beltran, 1995). It has been demonstrated that the incorporation of an alumina-supported Pd catalyst (Keane and Patterson, 2005; Keane, 2007) resulted in a marked enhancement in degradation rate, an increase in the product liquid fraction, a decrease in the liquid phase Cl content (by a factor of over 500) and an increase in the gas phase C1–C4 hydrocarbon content with a higher overall alkane content (ethane/ethylene in excess of 20). Sakata and co-workers have conducted a series of thermal and catalytic degradation of polyethylene/PVC and polypropylene/PVC mixtures (Lingaiah et al., 2001a,b; Sakata et al., 2003; Bhaskar et al., 2006). They found that catalytic dechlorination over Fe-based catalysts is effective in removing the Cl content as HCl but long-term catalyst stability remains


143

an unresolved issue. The full extent of the catalytic impact in terms of product distribution has yet to be irrefutably established but the results suggest that catalytic dechlorination can be a feasible route to a Cl-free hydrocarbon product.


6.7

Future trends and conclusions

While the multiple applications of synthetic polymers are of considerable societal benefit, a build up of polymeric waste can have a deleterious effect on the environment. Although global environmental systems are extremely resilient, there is a limit to the pollution burden that can be sustained. Landfill or incineration do not represent best practicable environmental options. Research in catalytic feedstock recycle of waste polymers is still at the development stage. In the case of polyolefins, it has been established that the use of a catalyst serves to significantly lower the requisite operating temperature, which is an important consideration in terms of energy usage. Solid acids, notably zeolites, are now established as effective polymer degradation agents where preliminary results have demonstrated that some control is possible over product composition by varying zeolite acidity and pore structure. An explicit relationship between zeolite structure and performance will require further fundamental research but this is essential in order to facilitate process optimisation. In dealing with PVC waste, the Cl content must be removed in a controlled fashion as part of an overall recycle operation. Supported metal mediated dechlorination has a decided role to play in generating a Cl-free product, i.e. ultimate production of a fuel oil from a starting waste PVC. Future work must be directed at extending the limited database of catalytic waste plastic processing to consider a wider range of reaction conditions and alternative catalytic materials with a full assessment of the final product distribution. Control over product composition and catalyst reuse are critical in order to implement an economically sustainable recycling unit operation.

6.8

Sources of further information and advice

A wide range of academic journals are available that cover all areas of catalysis research. The Journal of Catalysis publishes original and scholarly contributions in the fields of heterogeneous and homogeneous catalysis with a particular emphasis on studies that relate catalytic function to fundamental chemical processes. Applied Catalysis publishes papers on all aspects of catalysis that is of practical interest. The Journal of Molecular Catalysis deals with research articles that consider molecular and atomic aspects of catalytic activation and reaction mechanisms. Other journals devoted to catalysis research include: Catalysis Letters; Catalysis


144


Communications; Catalysis Today; Topics in Catalysis. The North American Catalysis Society (NACS) has an official web site (http://www.nacatsoc. org) that provides a range of general and specific information on catalysis developments with useful catalysis-related links. The International Association of Catalysis Societies also has a web site (http://www.iacs-icc.org) where details of upcoming meetings/conferences are posted. The Royal Society of Chemistry maintains a site (http://www.rsc.org/chemistryworld) where the latest developments in chemical technologies, many involving catalysis, are presented in a very clear and informative manner. Useful contacts for further information on plastics production, consumption and recycling include: Association of Plastic Manufacturers in Europe (APME); British Plastics Federation; British Polyethylene Industries PLC; Department for the Environment, Food and Rural Affairs (Defra). Topical issues relating to the polymer industry receive a comprehensive treatment in the European Plastics News and Plastics & Rubber Weekly magazines. The ‘wasteonline’ web site (http://www.wasteonline.org.uk) provides a valuable plastics recycling information sheet that is regularly updated. A comprehensive treatment of waste plastic reuse is provided in ‘Plastics in the UK Economy’, which is available to download from: www.wasteonline. org.uk. The Recycled Products Guide (available at www.recycledproducts. org.uk) provides a listing of products made from recycled plastics. Legislation which is particularly relevant to waste plastic recycling is the 1994 European Union Directive on Packaging and Packaging Waste 94/62/EC (the Packaging Directive). Croner (see http://www.croner.co.uk) is a useful resource providing information on environmental management, waste minimisation and legislation compliance.

6.9

References

aguado j and serrano dp (1999), Feedstock Recycling of Plastic Wastes, Cambridge, Royal Society of Chemistry. aguado j, serrano dp, escola jm, garagorri e and fernandez ja (2000), ‘Catalytic conversion of polyolefins into fuels over zeolite beta’, Polym Degrad Stab, 69, 11–16. aguado j, serrano dp and san miguel g (2006), ‘European trends in the feedstock recycling of plastic wastes’, Global NEST J, 9, 12–19. aguado j, serrano dp, miguel gs, escola jm and rodriguez jm (2007), ‘Catalytic activity of zeolitic and mesostructured catalysts in the cracking of pure and waste polyolefins’, J Anal Appl Pyrol, 78, 153–161. ahmed s, chughtai s and keane ma (1998), ‘Removal of cadmium and lead from water by ion exchange with NaY’, Sep Technol, 13, 57–64. ali mf and siddiqui mn (2005), ‘Thermal and catalytic decomposition of plastic waste with petroleum residue’, J Anal Appl Pyrol, 74, 282–289. ali s, garforth a, harris dh and shigeishi ra (2003), ‘Polymer waste recycling over “used” catalysts’, in Keane MA, Interfacial Applications in Environmental Engineering, New York, Marcel Dekker, 295–328.



145

alonso f, beletskaya ip and yus m (2002), ‘Metal-mediated reductive hydrodehalogenation of organic halides’, Chem Rev, 102, 4009–4091. anastas pt and lankey rl (2000), ‘Life cycle assessment and green chemistry: the yin and yang of industrial ecology’, Green Chem, 2, 289–295. anderson aw (1990), Physical Chemistry of Surfaces, 5th Edition, New York, Wiley & Sons. armendia ma, boráu v, jiménez jm, marinas jm and urbano fj (1999), ‘Influence of the reaction conditions and catalytic properties on the liquid-phase hydrodebromination of bromobenzene over palladium supported catalysts: activity and deactivation’, Appl Catal B: Environ, 20, 101–110. babu bv and chaurasia a (2003), ‘Modeling for pyrolysis of solid particles: kinetics and heat transfer effects’, Energy Convers Manag, 44, 2251–2275. bartholomew ch (1984), ‘Catalyst deactivation’, Chem Eng, 12, 96–112. bhaskar t, negoro r, muto a and sakata y (2006), ‘Prevention of chlorinated hydrocarbons formation during pyrolysis of PVC or PVDC mixed plastics’, Green Chem, 8, 697–700. bhatia s (1990), Zeolite Catalysis: Principles and Applications, Boca Raton, FL, CRC Press. bhatia s, beltramini j and do dd (1989), ‘Deactivation of zeolite catalysts’, Catal Rev Sci Eng, 31, 431–480. blazsó m, zelei b and jakab e (1995), ‘Thermal decomposition of low density polyethylene in the presence of chlorine-containing polymers’, J Anal Appl Pyrol, 35, 221–235. bockhorn h, hornung a and hornung u (1998), ‘Stepwise pyrolysis for raw material recovery from plastic waste’, J Anal Appl Pyrol, 46, 1–13. bodnariuk p, coq b, ferrat g and figueras f (1989), ‘Carbon–chlorine hydrogenolysis over Pd/Rh and Pd/Sn bimetallic catalysts’, J Catal, 116, 459–466. bond gc (1997), ‘The role of carbon deposits in metal-catalyzed reactions of hydrocarbons’, Appl Catal A: General, 149, 3–25. bortnovsky o, sobalik z and wichterlova b (2001), ‘Exchange of Co(II) ions in H-BEA zeolites: identification of aluminum pairs in the zeolite framework’, Micropor Mesopor Mat, 46, 265–275. boudart m and djéga-mariadassou g (1984), Kinetics of Heterogeneous Catalytic Reactions, Princeton, Princeton University Press. bowker m (1998), The Basis and Applications of Heterogeneous Catalysis, Oxford Chemistry Primers No. 53, Oxford, Oxford University Press. brebu m, uddin ma, muto a, sakata y and vasile c (2002), ‘The role of temperature program and catalytic system on the quality of acetonitrile–butadiene–styrene degradation oil’, J Anal Appl Pyrol, 63, 43–57. brebu m, bhaskar t, muto a and sakata y (2006), ‘Alkaline hydrothermal treatment of brominated high impact polystyrene (HIPS-Br) for bromine-free plastic recovery’, Chemosphere, 64, 1021–1025. breck dw (1974), Zeolite Molecular Sieves, Structure, Chemistry and Use, New York, John Wiley & Sons. british plastics federation (2004), ‘Recycling of plastic 1984–2001’, e-Digest of Environmental Statistics, Department for Environment, Food and Rural Affairs, UK. brophy bjh and hardman s (1996), ‘Low temperature pyrolysis for feedstock preparation’, in Brandrup J, Bittner M, Michaeli W and Menges G, Recycling and Recovery of Plastics, Munich, Hanser, 420–434.


146


brown ka, holland mr, boyd ra, thresh s, jones h and ogilvie sm (2000), Economic Evaluation of PVC Waste Management, Report No. EPCS/20725, Abingdon, Oxford, AEA Technology. butt jb (2000), Reaction Kinetics and Reactor Design, 2nd Edition, New York, Marcel Dekker. butt jb and petersen ee (1988), Activation, Deactivation and Poisoning of Catalysts, New York, Academic Press. cakici ai, yanik j, ucar s, karayildirim t and anil h (2004), ‘Utilization of red mud as catalyst in the conversion of waste oil and waste plastics to fuel’, J Mater Cycle Waste Manag, 6, 20–26. chatterjee n, basu s, palit sk and mait mm (1994), ‘A reexamination of the degradation of PVC by thermal analysis’, J Polym Sci, 32, 1565–1573. coq b, ferrat g and figueras f (1986), ‘Conversion of chlorobenzene over palladium and rhodium catalysts of widely varying dispersion’, J Catal, 101, 434–445. cornell dd (1995), ‘Rubber and paper recycling – a pragmatic approach’, ACS Symp Ser 609, 72–79. costner pt (1998), ‘Correlation of chlorine input and dioxin output from combustors: A review and reanalysis’, J Organohalogen Compds, 32, 436– 440. coughlan b and keane ma (1990), ‘Reduction of Ni2+ cations in Y zeolites I: Effect of sample pre-treatment’, J Catal, 123, 364–374. coughlan b and keane ma (1991), ‘The hydrogenation of benzene over nickel supported Y zeolites I: a kinetic approach’, Zeolites, 11, 12–17. coughlan b and keane ma (1992a), ‘Reduction of Ni2+ cations in Y Zeolites II: effect of the environment’, J Catal, 136, 170–181. coughlan b and keane ma (1992b), ‘Benzene ethylation and cumene dealkylation over nickel loaded Y zeolites’, J Catal, 138, 164–178. creyghton ej, burgers mhw, jensen jc and van bekkum h (1995), ‘Vapour phase hydrodehalogenation of chlorobenzene over platinum/H-BEA zeolite’, Appl Catal A: General, 128, 275–288. defra (2000), Waste Strategy for England and Wales, London, Department for Environment, Food and Rural Affairs Publication. de jong v and louw r (2004), ‘Performance of supported nickel and other metal catalysts in the hydrodechlorination of chlorobenzene and 1-chlorohexane’, Appl Catal A: General, 271, 153–163. de la puente g and sedran u (1998), ‘Recycling polystyrene into fuels by means of FCC: performance of various acidic catalysts’, Appl Catal B: Environ, 19, 300–311. demirbas a (2004), ‘Pyrolysis of municipal plastic wastes for recovery of gasolinerange hydrocarbons’, J Anal Appl Pyrol, 77, 97–102. ding wb, liang j and anderson ll (1997), ‘Hydrocracking and hydroisomerization of high-density polyethylene and waste plastic over zeolite and silica–alumina supported Ni and Ni–Mo sulfides’, Energy & Fuels, 11, 1219–1224. ding wb, tuntawiroon w, liang j and anderson ll (1999), ‘Depolymerization of waste plastics with coal over metal-loaded silica-alumina catalysts’, Fuel Proc Technol, 49, 49–63. dwyer j (1988), An Introduction to Zeolite Molecular Sieves, New York, John Wiley & Sons.



147

espinat d, freund e, dexpert h and martino g (1990), ‘Localization and structure of carbonaceous deposits on reforming catalysts’, J Catal, 126, 496–518. feng z, zhao j, rockwell j, bailey d and huffman g (1996), ‘Direct liquefaction of waste plastics and coliqufaction of coal–plastic mixtures’, Fuel Process Technol, 49, 17–30. fitzharris wd, katzer wd and manogue wh (1982), ‘Sulfur deactivation of nickel methanation catalysts’, J Catal, 76, 369–384. forzatti p and lietti l (1999), ‘Catalyst deactivation’, Catal Today, 52, 165–181. gamse t, steinkellner f, marr r, alessi p and kikic i (2000), ‘Solubility studies of organic flame retardants in supercritical CO2’, Ind Eng Chem Res, 39, 4888–4890. garforth aa, lin yh, sharratt pn and dwyer j (1998), ‘Production of hydrocarbons by catalytic degradation of high density polyethylene in a laboratory fluidised-bed reactor’, Appl Catal A: General, 169, 331–342. gates bc (1992), Catalytic Chemistry, New York, John Wiley. hagenmaier h, horch k, fahlenkamp h and schetter g (1991), ‘Destruction of PCDD and PCDF in refuse and incineration plants by primary and secondary measures’, Chemosphere, 23, 1429–1437. hardman s and wilson dc (1998), ‘Polymer cracking – new hydrocarbons from old plastics’, Macromol Symp, 135, 115–120. hesse nd and white rl (2004), ‘Polyethylene catalytic hydrocracking by PtHZSM-5, PtHY and PtHMCM-41’, J Appl Polym Sci, 92, 1293. hesse nd, lin r, bonnet e, cooper iii e and white rl (2001), ‘In situ analysis of volatiles obtained from the catalytic cracking of polyethylene’, J Polym Sci, 82, 3118–3125. horikawa s, takai y, ukei h, azuma n and ueno a (1999), ‘Chlorine gas recovery from polyvinyl chloride’, J Anal Appl Pyrol, 51, 167–179. jiminez a, berenguer v, lopez j and sanchez a (1993), ‘Thermal degradation studies of poly(vinyl)chloride: kinetic analysis of thermogravimetric data’, J Appl Polym Sci, 50, 1565–1573. joo hs and guin ja (1997), ‘Hydrocracking of plastic pyrolysis gas oil to naphtha’, Energy & Fuels, 11, 586–592. kaminsky w (1985), ‘Thermal recycling of polymers’, J Anal Appl Pyrol, 8, 439–448. kaminsky w and kim js (1999), ‘Pyrolysis of mixed plastics into aromatics’, J Anal Appl Pyrol, 51, 127–134. kaminsky w and sinn h (1996), ‘Petrochemical processes for recycling plastics: pyrolytic techniques’, in Brandrup J, Bittner M, Michaeli W and Menges G, Recycling and Recovery of Plastics, Munich, Hanser, 434–443. kaminsky w, predel m and sadiki a (2004), ‘Feedstock recycling of polymers by pyrolysis in a fluidised bed’, Polym Degrad Stabil, 85, 1045–1050. karagoz s, karayildirim t, uncar s, yuksel m and yanik j (2003a), ‘Liquefaction of municipal waste plastics in VGO over acidic and non-acidic catalysts’, Fuel, 82, 415–423. karagoz s, yanik y, uncar s, saglam m and song c (2003b), ‘Catalytic and thermal degradation of high denisty polyethylene in vacuum gas oil over non-acidic and acidic catalysts’, Appl Catal A: General, 242, 51–62. keane ma (1994), ‘The role of the alkali metal co-cation in the ion exchange of Y zeolites I: Alkali metal and nickel ion exchange equilibria’, Micropor Mater, 3, 93–105.


148


keane ma (2003), Interfacial Applications in Environmental Engineering, New York, Marcel Dekker. keane ma (2004), ‘Hydrodehalogenation over supported Ni and Pd: a comparative study’, Appl Catal A: General, 271, 109–118. keane ma (2007), ‘Catalytic conversion of waste plastics: focus on waste PVC’, J Chem Technol Biotechnol, 82, 787–795. keane ma and patterson pm (2005), ‘Catalytic recycling of PVC waste’, in Muller-Hagedom M, Innovative Plastic Recycling Techniques, Karlsruhe, Karlsruhe University Press, 50–58. kim j-s and keane ma (2000), ‘Divalent cobalt and iron exchange with Y zeolites: binary and ternary exchange equilibria’, J Coll Interf Sci, 232, 126–132. kim j-s, zhang l and keane ma (2001), ‘The removal of iron from water by ion exchange with Y zeolites’, Sep Sci Technol, 36, 1509–1525. king da and woodruff dp (1990), The Chemical Physics of Solid Surfaces and Heterogeneous Catalysis, Volume 3A: Chemisorption Systems, Amsterdam, Elsevier. la mantia f (2002), Handbook of Plastics Recycling, Shropshire, Rapra Technology Ltd. lang nd and kohn w (1970), ‘Theory of metal surfaces: charge density and surface energy’, Phys Rev B, 1, 4555–4568. lecloux aj (1981), Catalysis Science and Technology, Volume 2, Berlin, Springer. lin yh, sharratt pn, garforth aa and dwyer j (1997), ‘Deactivation of US-Y zeolite by coke formation during the catalytic pyrolysis of high density polyethylene’, Thermochim Acta, 294, 45–52. lingaiah n, uddin ma, morikawa k, muto a, murata k and sakata y (2001a), ‘Catalytic dehydrochlorination of chloro-organic compounds from PVC containing waste plastics derived fuel oil over FeCl2/SiO2 catalyst’, Green Chem, 3, 74–75. lingaiah n, uddin ma, muto a, imai t and sakata y (2001b), ‘Removal of organic chlorine compounds by catalytic dehydrochlorination for the refinement of municipal waste plastic derived oil’, Fuel, 80, 1901–1905. lunin vv and lokteva es (1996), ‘Catalytic hydrodehalogenation of organic compounds’, Russ Chem Bull, 45, 1519–1534. luo m and curtis cw (1996), ‘Effect of reaction parameters and catalyst type on waste plastics liquefaction and coprocessing with coal’, Fuel Process Technol, 49, 177–196. lutz w, fichtner-schmittler h, bulow m, schierhorn e, van phat n, sonntag e, kosche i, amin s and dyer a (1990), ‘Occurrence and removal of hydroxide precipitates on ion exchanged faujasite type zeolites’, J Chem Soc Faraday Trans I, 86, 1899–1904. marcella a and beltran m (1995), ‘Kinetic study of thermal decomposition of polystyrene and polyethylene–vinyl acetate graft polymers by thermogravimetric analysis’, Polym Degrad Stabil, 50, 117–124. martinson e, hjertberg t and sorvik e (1988), ‘Catalytic effect of HCl on the dehydrochlorination of polvinyl chloride’, Macromolecules, 21, 136–141. masel ri (1996), Principles of Adsorption and Reaction on Solid Surfaces, New York, John Wiley and Sons. mccaffrey wc, kamal mr and cooper dg (1995), ‘Thermolysis of polyethylene’, Polym Degrad Stabil, 47, 133–139.



149

mccaffrey wc, cooper dg and kamal mr (1998), ‘Tertiary recycling of polyethylene: mechanism of liquid production from polyethylene by thermolysis reactive distillation’, Polym Degrad Stabil, 62, 513–521. mceldowney jf and mceldowney s (1996), Environment and the Law, Essex, Longman. meisel sl, mccullogh jp, lechthaler ch and weisz pb (1976), ‘Gasoline from methanol in one step’, Chem Tech, 6, 86–89. meriaudeau p and naccache c (1997), ‘Dehydrocyclization of alkanes over zeolitesupported metal catalysts: monofunctional or bifunctional route’, Catal Rev, 39, 5–48. metecan ih, ozkan ar, isler r, yanik j, saglam m and yuksel m (2005), ‘Naphtha derived from polyolefins’, Fuel, 84, 619–628. mordi rc, field r and dwyer j (1994), ‘Thermolysis of low density polyethylene catalyzed by zeolites’, J Anal Appl Pyrol, 29, 45–55. murthy kv, patterson pm, jacobs g, davis bh and keane ma (2004), ‘An exploration of activity loss during hydrodechlorination and hydrodebromination over Ni/ SiO2’, J Catal, 223, 74–85. murty mvs, rangarajan p, grulke ea and bhattacharyya d (1996), ‘Thermal degradation/hydrogenation of commodity plastics and characterisation of their liquefaction products’, Fuel Process Technol, 49, 75–90. nambu h, sakuma y, ishihara t, takesue t and ikemura t (1987), ‘Catalytic degradation of polystyrene’, Polym Degrad Stabil, 19, 61–76. negelein dl, lin r and white rl (1998), ‘Effects of catalyst acidity and HZSM-5 channel volume on polypropylene cracking’, J Appl Polym Sci, 67, 341–348. ohtsuka y (1989), ‘Influence of hydrogen chloride treatment on the dispersion of nickel particles on carbon’, J Mol Catal, 54, 225–235. ouki sk and kavannagh m (1997), ‘Performance of natural zeolites in the treatment of mixed metal contaminated effluent’, Waste Manage Res, 15, 383–394. park c and keane ma (2001), ‘Gas phase dehydration of C6 alcohols promoted by Y zeolite and supported nafion catalysts’, J Mol Catal A: Chemical, 166, 303–322. park c, menini c, valverde j-l and keane ma (2002), ‘Carbon–chlorine and carbon– bromine bond cleavage in the catalytic hydrodehalogenation of halogenated aromatics’, J Catal, 211, 451–463. petersen ee and bell at (1987), Catalyst Deactivation, New York, Marcel Dekker. plastics europe (2004), An Analysis of Plastics Consumption and Recovery in Europe 2002–2003, A report for Plastics Europe. Brussels, Association of Plastics Manufacturers. plastics europe (2007), The Compelling Facts about Plastics. An Analysis of Plastics Production, Demand and Recovery for 2005 in Europe. A report for Plastics Europe, Brussels, Association of Plastics Manufacturers. pollock t (1989), Dioxins and Furans: Questions and Answers, Philadelphia, PA, Academy of Natural Sciences. ponec v, knor z and cerny s (1974), Adsorption on Solids, London, Butterworth. rothenburger ks, cugini av, thomson rl and ciocco mv (1997), ‘Investigation of first stage liquefaction of coal with model plastic waste mixtures’, Energy & Fuels, 11, 849–855. sakata y (1998), ‘Catalytic degradation of polyethylene and polypropylene to fuel oil’, Macromol Symp, 135, 7–18.


150


sakata y, bhaskar t, uddin ma, muto a and matsui t (2003), ‘Development of a catalytic dehalogenation (Cl, Br) process for municipal waste plastic-derived oil’, J Mater Cycles Waste Manag, 5, 113–124. scheirs j (1998), Polymer Recycling. Science, Technology and Applications, Chichester, John Wiley & Sons. serrano dp, aguado j, escola jm and rodriguez jm (2000), ‘Catalytic conversion of polystyrene over HMCM-41, HZSM-5 and amorphous SiO2–Al2O3: comparison with thermal cracking’, Appl Catal B: Environ, 25, 181–189. serrano dp, aguado j, escola jm and rodriguez jm (2005), ‘Influence of nanocrystalline HZSM-5 external surface on the catalytic cracking of polyolefins’, J Anal Appl Pyrol, 74, 353–360. sharratt pn, lin yh, garforth aa and dwyer j (1997), ‘Investigation of the catalytic pyrolysis of high-density polyethylene over a HZSM-5 catalyst in a laboratory fluidized-bed reactor’, Ind Eng Chem Res, 36, 5118–5124. shimada h, imamura m, matsubayashi n, saito t, tanaka t, hayakawa t and kure s (2000), ‘Characterization of coke deposited on catalysts by carbon K-edge near edge X-ray absorption fine structure (NEXAFS) spectroscopy’, Topic Catal, 10, 265–271. shin ej and keane ma (1999), ‘Detoxification of dichlorophenols by catalytic hydrodechlorination of chlorophenols using a nickel/silica catalyst’, Chem Eng Sci, 54, 1109–1120. shin e-j and keane ma (2000), ‘Gas phase catalytic hydroprocessing of trichlorophenols’, J Chem Technol Biotechnol, 75, 159–167. shin e-j, spiller a, tavoularis g and keane ma (1999), ‘Chlorine–nickel interactions in gas phase catalytic hydrodechlorination: catalyst deactivation and the nature of reactive hydrogen’, Phys Chem Chem Phys, 1, 3173–3181. somorjai ga (1994), Introduction to Surface Chemistry and Catalysis, New York, Wiley Interscience. songip ar, masuda t, kuwahara h and hashimoto k (1994), ‘Kinetic studies for catalytic cracking of heavy oil from waste plastics over ReY zeolites’, Energy & Fuels, 8, 131–135. stanislaus a and cooper bh (1994), ‘Aromatic hydrogenation catalysis – a review’, Catal Rev, 36, 75–123. steinfeld ji, francisco js and hase wl (1999), Chemical Kinetics and Dynamics, 2nd Edition, Upper Saddle River, NJ, Prentice Hall. tavoularis g and keane ma (1999), ‘Direct formation of cyclohexene via the gas phase catalytic dehydrochlorination of cyclohexyl halides’, Appl Catal A: General, 182, 309–316. thomas jm and thomas wj (1997), Principles and Practice of Heterogeneous Catalysis, Weinheim, VCH. thornton j (2000), Pandora’s Poison: Chlorine Health and a New Environmental Strategy, Cambridge, MA, MIT Press. twigg mv (1989), Catalyst handbook, 2nd Edition, London, Wolfe Publishing Co. ucar s, karagoz s, karayildirim t and yanik j (2002), ‘Conversion of polymers to fuels in a refinery stream’, Polym Degrad Stabil, 75, 161–171. uddin ma, sakata y, shiraga y, muto a and murata k (1999), ‘Dechlorination of chlorine compounds in poly(vinyl chloride) mixed plastics derived oil by solid sorbents’, Ind & Eng Chem Res, 38, 1406–1410.



151

uddin ma, bhaskar t, kaneko j, muto a, sakata y and matsui t (2002), ‘Dehydrohalogenation during pyrolysis of brominated flame retardant containing high impact polystyrene (HIPS-Br) mixed with polyvinylchloride (PVC)’, Fuel, 81, 1819–1825. uemichi y, makino y and kanazuka t (1989), ‘Degradation of polypropylene to aromatic hydrocarbons over Pt containing and Fe containing activated carbon catalysts’, J Anal App Pyrol, 16, 229–238. uemichi y, nakura j, itoh t, sugioka m, garforth aa and dwyer j (1999), ‘Conversion of polyethylene into gasoline-range fuels by two-stage catalytic degradation using silica-alumina and HZSM-5 zeolite’, Ind Eng Chem Res, 38, 385–390. ukei h, hirose t, horikawa s, takai y, taka m, azuma n and ueno a (2000), ‘Catalytic degradation of polystyrene into styrene and a design of recyclable polystyrene with dispersed catalysts’, Catal Today, 62, 67–75. ukisu y and miyadera t (1997), ‘Hydrogen transfer hydrodechlorination of aromatic halides with alcohols in the presence of noble metal catalysts’, J Mol Catal A: Chemical, 125, 135–142. urbano fj and marinas jm (2001), ‘Hydrogenolysis of organohalogen compounds over palladium supported catalysts’, J Mol Catal A: Chemical, 173, 329–345. van bekkum h (1991), Introduction to Zeolite Science and Practice, Amsterdam, Elsevier. van santen ra and jansen apj (1991), ‘Cation induced changes in chemical reactivity of small metal particles’, in Grasselli RK and Sleight AW, Structure, Activity and Selectivity Relationships in Heterogeneous Catalysis, Amsterdam, Elsevier, 221–232. van veen jar (2002), ‘Hydrocracking’, in Guisnet M and Gilson JP, Zeolites for Cleaner Technologies, London, Imperial College Press, 131–152. walendziewski j (2002), ‘Engine fuel derived from waste plastic by thermal treatment’, Fuel, 81, 473–481. walendziewski j and steininger m (2001), ‘Thermal and catalytic conversion of waste polyolefines’, Catal Today, 65, 323–329. wang l and chen p (2004), ‘Development of first stage co-liquefaction of Chinese coal with waste plastics’, Chem Eng Process, 43, 145–148. wanke se and flynn pc (1975), ‘Sintering of supported metal catalysts’, Cat Rev-Sci Eng, 12, 93–135. weitkamp j (1975), ‘The influence of chain length in hydrocracking and hydroisomerization of n-alkanes’, in Ward JA and Qater SA, Hydrocracking and Hydrotreating, Washington DC, American Chemical Society, 1–27. white mg (1990), Heterogeneous Catalysis, Upper Saddle River, NJ, Prentice Hall. williams pt (1988), Waste Treatment and Disposal, Chichester, UK. williams pt and williams ea (1999), ‘Fluidised bed pyrolysis of low density polyethylene to produce petrochemical feedstock’, J Anal Appl Pyrol, 51, 107–126. yanik j, uddin ma, ikeuchi k and sakata y (2001a), ‘The catalytic effect of red mud on the degradation of poly(vinyl chloride) containing polymer mixture into fuel oil’, Polymer Degrad Stabil, 73, 335–346. yanik j, uddin ma and sakata y (2001b), ‘The effect of red mud on the liquefaction of waste plastics in heavy vacuum gas oil’, Energy & Fuel, 15, 163–169. yassin aa and sabaa mw (1990), ‘Degradation and stabilization of poly(vinyl chloride)’, J Macromol Sci Rev Chem Phys, 30, 491–502.

7 Advanced thermal treatment of composite wastes for energy recovery P. L E T T I E R I, L. YA S S I N and S. J. R. S I M O N S, University College London, UK

Abstract: The chapter begins by describing the evolution of the strategies and legislations that have been introduced to address waste management. It then reviews the different thermal treatment options available, namely combustion, gasification and pyrolysis, before discussing the techno-economic performance of advanced thermal processes that recover energy from municipal solid waste using fluidised-bed combustion and gasification technology at two different scales. The chapter includes also a comparison between these advanced treatment technologies and the traditional moving-grate process.


Key words: energy from waste, combustion, gasification, fluidised beds, technical and economic evaluation.

7.1

Introduction

From 1995 to 2000, the world’s urban population grew at a rate of 2.2% per year and by 2000 75% of the population in the developed world lived in urban areas. This figure is projected to rise to nearly 83% by 2030, while in the developing world, the rate of urbanisation is even faster (United Nations, 2002). Along with this comes increased demand for energy and natural resources, fuelled by increasing consumption levels per capita in rich countries and rapid rise in consumption in developing countries, in particular China and India. If we are to meet this increased demand and, at the same time, stabilise climate change, then we must triple the planet’s current energy-producing capacity by 2050, with all new additions being carbon neutral. Any solution will necessarily involve a mixture of technologies, coupled with regulatory and policy incentives, including a massive scaling-up of energy efficiency measures and the use of renewable sources of energy to reduce our dependence on fossil fuels. The burning of fossil fuels and other anthropogenic activities, such as changes in land use, are leading to increases in the concentrations of a number of greenhouse gases (GHG), such as carbon dioxide, methane, chlorofluorocarbons and ozone, in the atmosphere, all contributing to global warming and climate change, posing major risks both to the 152


Advanced thermal treatment of composite wastes

153

environment and to the global economy. In the Stern Review, Nicholas Stern, the former Chief Economist and Senior Vice-President of the World Bank, reported that climate change, if unabated, will have a serious impact on global economic growth. He stated that greenhouse gas emissions need to be stabilised in the next 20 years, then fall 1–3% after that, if we are to avoid a global recession worth up to 20% of global gross domestic product (GDP) (Stern, 2007). The cost of stabilisation would be something approaching 1% of global GDP per year by 2050 (global GDP currently stands at around $44 trillion). Although Stern’s report was criticised by some economists, his final message was loud and clear, ‘if we act now, we can avoid the very worst’. If we are to satisfy our basic needs and ‘enjoy a better quality of life without compromising the quality of life of future generations’ (Defra, 2005a), we must, amongst other things, improve our resource efficiency and reduce our impacts on the environment. This involves getting the most out of our finite resources and minimising waste. Ultimately, we need to shift processes from linear and ‘open loop’ systems, where natural resources and capital investments move through the system to become waste, to ‘closed loop’ systems, where waste outputs from one process may be used as resource inputs to another and where any residual wastes are returned to the environment in a way that enables them to be extracted and used again. This ‘resource cycle’ is depicted in Fig. 7.1.

Materials Materials

Nature Natural resource

Extraction and processing

Primary material

Reuse

Recycling Materials recovery

Manufacture Consumption Goods

Disposal Resource management

End-of-life goods Energy recovery Residuals Energy

Product waste Extraction and processing waste

7.1 The resource cycle.

154



7.1.1 Energy: the changing climate In 2000, the UK’s Royal Commission on Environmental Pollution (RCEP) presented its report, entitled Energy – The Changing Climate, detailing preventative measures to increasing carbon dioxide (CO2) emission levels and a completely different approach to the way we obtain and use energy in the UK. In order to ensure that the concentration of CO2 in the atmosphere does not exceed 550 ppmv (parts per million by volume), widely regarded as the ‘tipping point’ beyond which levels will rise uncontrollably (due to feedback effects), the RCEP called for a reduction of the UK’s CO2 emissions by 60% from their current level by 2050. The RCEP presented various energy scenarios to the UK’s energy supply and demand that would enable the UK to reach this target, while protecting the UK economy, its environment and quality of life (RCEP, 2000). The scenarios put forward by the RCEP included the reduction of energy use through smarter application of technology, especially in the heating and cooling of buildings, which accounts for approximately 8% of CO2 emissions (Stern, 2007), more efficient use of fossil fuels and large-scale deployment of alternative energy sources. The efficient use of fossil fuels entails the transition to a new energy economy by switching to gas, which has lower carbon content in relation to its energy content than compared with oil and coal (see Section 7.2.2). Combined heat and power (CHP) plants supplying heat for district heating systems were also recommended, to provide a growing market for renewable fuels, such as biomass. The move to alternative energy sources that are renewable and sustainable is required to lessen our dependence on fossil fuels. The potential for growth and expansion of renewable sources of energy, such as solar, hydropower, wind, biomass and geothermal, is great, especially as oil resources diminish and the price per barrel begins to increase. However, to fully realise this potential, there are numerous challenges that must be addressed, such as the development and deployment of technological innovations that will allow us to keep pace with the increasing global demand for energy whilst, at the same time, moving away from the use of fossil fuels. Market take-up of such technologies will need to be encouraged and enabled by regulatory and policy measures that will ease the transition to a low carbon future.

7.1.2 Alternative renewable energy The UK Government has committed itself to reducing carbon dioxide emissions by 60% by 2050, as recommended by the RCEP. However, in order for this target to be achieved, alternatives to fossil fuels must be found and

Advanced thermal treatment of composite wastes Coal 18%

155

Solar 1% Nuclear 15%

Others 6%

Wind 3% Geothermal 4% Hydro 27% Biomass 65%

Oil 38% Gas 23%


7.2 Breakdown of energy consumption in the EU in 2002 (adapted from European Commission, 2005).

brought into use as sources of heat and power at the earliest possible opportunity. Renewable energy sources produce lower or no greenhouse gas emissions, improve security of supply by diversification of energy production and encourage creation of new jobs. In 2002, renewables (the collective noun for renewable sources of energy) accounted for 6% of the total energy consumed in Europe. Of this, almost two-thirds came from biomass, accounting for 4% of the total EU energy supply (Fig. 7.2). This figure is expected to double by 2010 (European Commission, 2005). Therefore, biomass has a critical role to play in the EU’s long-term sustainable energy strategy. There are now many dedicated and established biomass energy (or bioenergy) systems that produce heat and/or power, gaseous and liquid fuels. These systems are able to use a large variety of feedstocks, such as municipal solid waste (MSW), which contributed to 13% of the primary bioenergy produced in the EU in 2002. The remaining bio-energy feedstocks were wood residues and energy crops (with a combined contribution of 81%) and biogas (4%), produced from the decomposition of organic wastes. Biomass is currently the only available renewable energy source that can produce competitively priced fuels for transport in large quantities and the only widespread renewable source of high grade heat. However, it is far from being fully deployed in the UK. According to the Biomass Task Force (2005), there are considerable biomass feedstock resources in the UK that are not being utilised, particularly waste streams that can be regarded as a secure and sustainable source of biomass-derived energy. As a consequence of their report, the UK Biomass Strategy was published in May 2007, in which is defined the UK Government’s aspiration for the sustainable development of biomass for heat and power, transport fuels and industrial products (Defra, 2007a).

156


7.1.3 Chapter outline The aim of this chapter is to provide a brief introductory guide to the challenge of managing waste in both the UK and EU and to its use as an energy resource; it will review the evolution of the strategies and legislations that have been introduced (up to date) to address waste management; it will briefly introduce the different thermal treatment options available, namely combustion, gasification and pyrolysis, and will discuss the possible treatment of the emissions and residues. The techno-economic performance of processes that recover energy from MSW using fluidised-bed combustion and gasification technology is compared at two different scales (50 and 100 kt/annum waste feed). Sample calculations for the case of 50 ktpa waste treated using fluidised bed gasification technology are provided to enable the reader to carry out their own analyses at other scales. Finally, a comparison between these advanced treatment technologies and the traditional moving-grate process is also discussed.

Introduction to waste management

The amount of municipal waste produced on average by each European citizen is projected to increase from 520 kg in 2004 to 680 kg by 2020 (EEA Briefing 2008/01), an increase of 25%. Furthermore, waste production in the new EU12 countries is projected to reach the current levels of those in the EU15 countries by 2020, as shown in Fig. 7.3.

800 700

Data

Projections

600 kg/capita


7.2

Landfill

500

Incineration

400

Recycling

300 200 100 0 EU- EU- EU- EU- EU- EU- EU- EU- EU- EU- EU- EU27 15 12 27 15 12 27 15 12 27 15 12 1995 2004 2010 2020 Year

7.3 Generation and management of municipal waste in European countries (adapted from European Topic Centre on Resource and Waste Management ETC/RWM – EEA Briefing 2008/01).

157

Traditionally, landfilling has been used as the major waste management method. However, European and national waste management policies have successfully resulted in an increase in recycling and incineration and a subsequent decrease in the amount of waste sent to landfill. Examples include the EU Packaging Directive, introduced in 1994, which was aimed at increasing recycling and recovery of packaging, and the Landfill Directive, introduced in 1999, which was aimed at diverting biodegradable municipal waste from landfill. Predictions for municipal waste management in the European Union project an increase in recycling of up to 34%, an increase in incineration with energy recovery of up to 27% and a reduction in landfill disposal of up to 34% by 2020. As a consequence of this, net gaseous emissions are projected to decline from 50 million tonnes CO2-equivalent in 1980 to 10 million tonnes by 2020, as shown in Fig. 7.4 (EEA Briefing 2008/01). Whilst on the one hand the increased waste quantities are expected to increase direct emissions, on the other hand, increasing recycling and incineration will represent savings on GHG emissions avoided, where recycling is expected to contribute to 75% of total avoided emissions by 2020 and incineration to almost 25%, in part due to the increased diversion from landfill resulting in reduced emissions of landfill gases. It is important to emphasise that these projections assume that waste management capacity grows to match demand. If not, then inefficient management will lead to higher GHG emissions than predicted.

7.2.1 The UK national waste strategy The UK waste management strategy is largely derived from EU legislations, which fall into three categories (Waste Watch, 2007):

150 Million tonnes CO2 equivalent



Direct emissions

Net emissions

125 100 Avoided emissions 75 50 25 0 1970

1975

1980

1985

1990

1995

2000

2005

2010

2015

2020

7.4 Predicted gaseous emissions (adapted from EEA Briefing 2008/01).

158



• Horizontal legislations that set the overall framework for the management of waste, including definitions, such as the EC Framework Directive on Waste. • Legislations on treatment operations, which set technical standards for the operation of waste facilities, such as the Landfill and Waste Incineration Directives. • Specific waste stream legislations, such as the Packaging and Packaging Waste Directive. In May 2000, the UK Government published its ‘Waste Strategy 2000’ as a national waste strategy for England and Wales. The report, in response to the EU Landfill Directive, was aimed at delivering change within the UK waste management practices (Sustainable Development, 2000). The Waste Strategy 2000 included the establishment of national targets for recovery of municipal waste and recycling/composting of household waste. This was followed up by the publication in 2007 of a new strategy for cutting waste in England, with an emphasis on its role in tackling climate change and resource efficiency. The Waste Strategy for England 2007 is expected to lead to an annual net reduction in GHG emissions from waste management of at least 9.3 million tonnes (mt) of CO2 equivalent per year compared to 2006 (Defra, 2007b). The main objectives and targets of the strategy are to: • decouple waste growth (in all sectors) from economic growth and put more emphasis on waste prevention and reuse; • meet and exceed the Landfill Directive diversion targets for biodegradable municipal waste in 2010, 2013 and 2020; • increase diversion from landfill of non-municipal waste and secure better integration of treatment for municipal and non-municipal waste; • secure the investment in infrastructure needed to divert waste from landfill and for the management of hazardous waste; • get the most environmental benefit from that investment, through increased recycling of resources and recovery of energy from residual waste using a mix of technologies. Table 7.1 summarises the new increased recycling and recovery targets for household and municipal waste in England. As part of the 2007 waste strategy, the Government has introduced greater financial incentives to reflect the waste hierarchy and create opportunities for the reduction, reuse, recycling and recovery of energy from waste. This includes increasing the landfill tax escalator to £8 per year and introducing enhanced capital allowances for investment involving the use of solid recovered fuel for combined heat and power facilities (Defra, 2007b).


159


Table 7.1 Recycling and recovery targets for household and municipal waste in England Category

2010

2015

2020

Reduction in residual household waste from 2000 levels of 2.2 million tonnes Recycling and composting of household waste Municipal waste recovery

29%

35%

45%

40% 53%

45% 67%

50% 75%

In the past, local authorities used the principle of Best Practicable Environmental Option (BPEO) in producing their municipal waste management strategies. This identified waste management options that provided the most environmental benefits, as well as meeting legislative and practicability constraints, such as costs. In 2005, the Planning and Policy Statement 10 (PPS 10) was introduced, in which Sustainability Appraisals, carried out using a life-cycle assessment (LCA) approach, replaced the BPEO (ODPM, 2005). LCA techniques calculate emissions and residue releases to air, water and land. It is a ‘cradle-to-grave’ approach, as it follows the consumption of primary resources involved in all stages of the municipal waste cycle, from the extraction of raw materials, through operation of waste collection and treatment processes, to final disposal. LCA also takes account of the resources saved and environmental burdens avoided if secondary materials recovered from waste are used as substitutes for primary raw materials (Scottish Executive & SEPA, 2003). Although the UK’s waste performance still lags far behind much of Europe, the introduction of the waste strategies and of several Government initiatives has led to significant progress. Recycling and composting figures are up, less waste is being sent to landfill and, most importantly, municipal waste is growing much less quickly than the economy, at 0.5% per year (Defra, 2007b). These are encouraging results for the Government and its quest for a sustainable economy.

7.2.2 The role of energy from waste As highlighted earlier, a sustainable waste management strategy encompasses the safe treatment and disposal of residual waste with increased energy recovery and integrated recycling and reuse activities. A major environmental benefit gained from MSW energy recovery is the reduction in greenhouse gas emissions per kW h. The International Energy Agency (IEA, 2003) has reported that, for conventional incinerators producing energy from waste (EfW), the total emission of CO2 is about 367 grams per kW h, compared with 446 and 987 grams kW h for gas and coal fuelled

160


boilers, respectively. Furthermore, around 50–100 kg of methane per tonne of waste would be released if the MSW was sent to landfill. This is equivalent to 1610 kg of CO2 per tonne of waste, as methane has a higher global warming potential. EfW currently provides 0.5% of the electricity used in the UK. However, it is expected that this could increase to 17% by 2020 (Lee et al., 2005). For this reason, it is important that EfW is recovered effectively through the use of the most efficient, clean technologies. These include anaerobic digestion, mechanical and biological treatment processes (MBT), direct combustion or incineration and advanced thermal treatment (ATT), such as gasification and pyrolysis (Fig.7.5).


7.2.3 Thermal treatment of waste In the UK, EfW technologies are predominantly either direct combustion or incineration processes. Advanced thermal treatment processes, in particular gasification, are seen as alternatives to traditional combustion processes and provide additional routes for the diversion of waste from landfill. Gasification processes use fluidised bed technology and offer further possibilities for recovering value from waste by being compatible with frontend systems, such as recycling and MBT; this is a consequence of fluidised bed technology requiring the solid waste to be sorted and pretreated (shredded and sometimes pelletised) prior to gasification. In addition, gasification produces solid residues that are more suitable for reuse than those from combustion. Although gasification is not a new concept, it is only in recent years that it has been commercially used to treat MSW or refuse-derived fuels (RDF, a pretreated, biomass-rich fraction of the initial MSW, shredded and, sometimes, pelletised to a form suitable for use in fluidised beds).

Landfill

Waste Mechanical separation

Thermal conversion

Biological conversion

Anaerobic digestion

Composting

Gasification

Combustion

7.5 Processes for the recovery of energy from waste.

Pyrolysis


161

Most of the successful commercial operations have been in Europe, Japan and North America (Defra, 2007c).


Combustion Combustion is the total oxidation of the organic matter in waste at temperatures in excess of 850 °C to produce heat, water vapour, carbon dioxide and non-combustible material, namely fly and bottom ash. Combustion reduces the volume of waste by approximately 90% and the remaining inert bottom ash residue can be used as secondary aggregate, reducing the need to quarry for natural aggregate materials. Although the actual process design and plant layout may differ from one facility to another, a schematic diagram of a typical EfW combustion process is illustrated in Fig. 7.6. The process consists of the waste reception, combustion chamber, energy recovery and emissions and residues handling. The combustion process converts the heat energy in the MSW or RDF into steam, which can be used to generate power via a steam turbine and/or used for heating. MSW typically has an energy content of 9–11 MJ/kg, whilst RDF can have an energy content of up to 17 MJ/kg (Defra, 2007d). If a combustion plant was to generate only heat, it can achieve a thermal generating efficiency of 80–90%; on the other hand, an electricity-only plant can achieve an electrical generating efficiency of between 20 and 27% (ERM & Golders Associates, 2006). Gasification Gasification is the thermal conversion of organic matter by partial oxidation into a gaseous product called syngas. The syngas consists mainly of hydrogen, carbon monoxide and small amounts of methane, water vapour, carbon

Power distribution

District heating

Turbines & generators Waste reception

Combustion

Bottom ash

Boiler

Recyclables

Gas cleaning

Chimney

Fly ash

7.6 Schematic diagram of a typical EfW combustion process.


162


dioxide, nitrogen and tar. The reactions are carried out at temperatures of between 500 and 1400 °C and pressures of up to 33 bar. The high temperature in the gasifier converts the inorganic materials in the waste, such as ash and metals, into a vitrified material resembling coarse sand. The vitrified material or bottom ash is inert and has a variety of uses in the construction and building industries (Gasification Technologies Council, 2006). The syngas can replace fossil fuels in high efficiency power generation, heat and CHP applications and can be used in the production of liquid fuels and chemicals. For power generating applications, the syngas is either combusted to generate steam for use in conventional steam turbines or utilised directly in dedicated gas engines and turbines. As opposed to steam turbines, gas engines and turbines have higher electrical conversion efficiencies, ranging from 25% for gas engines to up to 40% for combined cycle gas turbines (CCGT), or so-called combined cycle power plants (CCPT) where a gas turbine generates electricity and the waste heat is used to make steam to generate additional electricity via a steam turbine. It is also possible to combine gasification processes with fuel cells for CHP applications, which also offers the perspective of very high efficiencies (Jörß et al., 2002). It is important to note, however, that the overall system efficiencies of thermal treatment processes do not only depend on the generation efficiencies of the ‘prime movers’ (i.e. steam turbines, gas engines and CCGT), but also on the thermal conversion efficiency and the internal energy consumption. Overall system efficiencies are discussed and compared in the next section. The oxidant used for the gasification process can be air, pure oxygen, steam or a mixture of these gases. Air-based gasifiers typically produce a product gas containing a relatively high concentration of nitrogen with a low heating value. Oxygen and steam-based gasifiers produce a product gas containing a relatively high concentration of hydrogen and carbon monoxide, with a higher heating value (Bridgwater, 2003). Table 7.2 shows examples of existing biomass and waste fluidised bed gasification plants in the UK, Europe and the USA, including their scale and developers. Pyrolysis Pyrolysis is the thermal conversion of organic matter in the total absence of oxygen at relatively low temperatures of between 500 and 800 °C and short vapour residence times of 3–1500 s. It produces a liquid fuel, a solid char and some combustible gas, which is usually used within the process to provide the process heat requirements. The liquid fuel, or bio-oil, can be used directly as a substitute for fuel oil in heat and power applications or to produce a wide range of speciality and commodity chemicals (Fig. 7.7).


163


Table 7.2 Main gasification applications Gasification type

Scale and technology developers

Heat gasifiers (syngas combustion) Pöls, Austria Rüdersdorf, Germany

27 MWth CFB, Lurgi 100 MWth CFB, Lurgi

Co-firing gasifiers Amer, Netherlands Burlington, USA Lahti, Finland Ruien, Belgium Zeltweg, Austria

85 MWth CFB, Lurgi 50 MWe CFB, Battelle 40–70 MWth CFB, FW 50 MWth CFB, FW 10 MWth ACFB, AEE

IGCC plants ARBRE, UK Grève-in-Chianti, Italy Pisa, Italy Värnamo, Sweden

8 MWe CFB, TPS 6.7 MWe CFB, TPS 12 MWe CFB, Lurgi 18 MW PCFB, FW

CFB gasifiers with gas engine Güssing, Austria Skive, Denmark

8 MWth FICFB, AICE 11.5 MWth PBFB, Carbona

ACFB, atmospheric circulating fluidised beds; AEE, Austrian Environmental Agency; AICE, Austrian Institute of Chemical Engineering; CFB, circulating fluidised beds; FICFB, fast internal circulating fluidised bed; FW, Foster Wheeler; PCFB, pressured circulating fluidised beds; TPS, Termiska Processer Sweden.

Bio-oil

Extract

Furnace

Steam turbine

Heat

Upgrade

Boiler

Engine

Electricity or CHP

Gas turbine

Transport fuels

Chemicals

7.7 Bio-oil applications (adapted from Bridgwater et al., 1999).

164


Bio-oil is a dark brown liquid, which has a high heating value (HHV) of 16–19 MJ/kg, as compared with 42–44 MJ/kg for conventional fuel oil. The composition of the pyrolysis products depends on the heating rate, residence time and temperature, as well as on the composition of the fuel. Although most of the development work on this technology has used wood as the feedstock, owing to its consistency and comparability between tests, nearly 100 different biomass types have been tested by many laboratories, ranging from agricultural to solid wastes (Bridgwater, 2003). The performances of the different waste treatment options are summarised in Table 7.3.


7.2.4 Emissions and residues All the thermal treatment processes of waste result in residues and emissions. These are unavoidable, but nevertheless their capacity to impact upon the environment can be effectively controlled. Ares and Bolton (2002) have reported that combustion of one tonne of MSW in a modern grate furnace can generate between 5200 and 6000 Nm3/h of combustion gases (flue gases), with various compositions, as shown in Table 7.4. Emissions from EfW facilities are tightly controlled by the Environment Agency, which ensures that they are kept well below stringent levels set by UK and EU legislations. The main residues from EfW plants are bottom ash (BA) and air pollution control residues (APC), which include the fly ash. BA and APC residues usually account for approximately 25% and 3% by weight of incoming waste streams, respectively. BA from modern EfW plants is an inert waste discharged from the end of the grate. It is widely used throughout Europe as a secondary aggregate in road construction and the building industry. APC residues are generated as a result of the flue gas treatment. These residues are hazardous and must be safely disposed of to a licensed and specialist landfill site under very strict regulatory conditions. The EU Waste Incineration Directive (WID) came into force in December 2000 and sets the permitted emission levels for incinerators. The limits are summarised in Table 7.5 and are translated into the UK through the Waste Incineration Regulations 2002 (Defra, 2007d).

What is in the flue gas? Flue gases are a mixture of combustion products and include water vapour, carbon dioxide, particulates, heavy metals and acidic gases. Carbon monoxide and volatile organic compounds (VOCs) are also products of combustion, but they are indicators of incomplete combustion and can be easily

100–600 (Moving-grate) 70–150 (5001 FBC) 14–27

£35 m for 136 ktpa (moving-grate) £51 m for 256 ktpa (moving-grate) £35 m for 120 ktpa (fluidised bed combustion, FBC) Highly bankable

Capacities (ktpa)

Economics

Plant siting issues

Stack height (m)

Planning permission Visual impact Footprint

Bankability

High

Medium–large (moving-grate requires smaller footprint per unit capacity compared to fluidised bed) 60–120

Highly uncertain

Proven technology worldwide

Commercial availability

Efficiencies (%)

Combustion

Criteria

Table 7.3 Performance summary of thermal treatment processes of waste


Short exhaust pipe to 50 m depending on energy recovery system Medium

Small–large

10–20 (Syngas combustion) 13–28 (using gas engine) 40% (using CCGT) Up to 27% (co-firing in existing power plant) £9m for 25 ktpa (FBG) £45–5 m for 60 ktpa (rotary kiln pyrolysis) £69 m for 200 ktpa (combined gasification/ pyrolysis) Not bankable in current market state. However, it may become bankable if promoted by reputable companies Less uncertain than combustion

No proven track record in the UK, although, there are few operating plants on a commercial basis 10–120

Gasification and pyrolysis

Nuisance

Traffic impacts

As combustion

As combustion but generally emit lower levels of dioxins and metals than combustion(2) As combustion. If syngas is utilised in gas engines and turbines, then further benefits would be achieved. As combustion, but can divert 75–99% by weight of waste from landfill, if BA or slag is recycled As combustion, however, the residues are more suitable for reuse than combustion As combustion. Smaller facilities have less traffic impacts

Gasification and pyrolysis

(2)

The Allington Plant which is designed to treat 500 ktpa of MSW and began operating in 2008. The benefits of lower emissions are reduced if the syngas is combusted. Sources: Fichtner Consulting Engineers (2004), GLA (2003), CRE Group (2000), Juniper (2003), Defra (2007c, d).

(1)

Can extract metals before or after combustion and divert 70–96% by weight of waste from landfill, if BA is recycled Produce BA, which is recyclable and APC residues, which are sent to special landfills Waste management systems should be integrated and located closer to waste origin Good housekeeping and adequate countermeasures can reduce the potential nuisance from odour, dust, vermin and flies

Landfill diversion

Residue management

Recovers energy from waste, which would have came from fossil fuels

Achieve emissions significantly lower than WID limits

Combustion

GHG reduction

Environmental impact Air emissions

Criteria

Table 7.3 Continued



167

Table 7.4 Composition of a typical flue gas stream from MSW combustion Fly ash (dust)

3000–6000 mg/Nm3

Acidic gases HCl SO2 HF NOx (NO + NO2)

600–1800 200–800 10–30 250–500

mg/Nm3 mg/Nm3 mg/Nm3 mg/Nm3

40–60 mg/Nm3

Heavy metals

40–100 mg/Nm3

Volatile organic compounds (VOCs)

1–10 mg/Nm3

Dioxins/furans Source: Ares and Bolton (2002).


Table 7.5 Emission limits set by the Waste Incineration Directive 2000 Daily average values Total organic carbon Total dust Hydrogen chloride Hydrogen fluoride Sulphur dioxide NO2 (new or large incinerators) NO2 (existing smaller incinerators) Average values over sample period Cadmium and thallium compounds (total) Mercury compounds Other metalloid compounds (total) Average values measured over 6–8 hours Dioxins and furans (in toxic equivalents)

10 10 10 10 50 200 400 30 minutes 0.05 mg/m3 0.05 mg/m3 0.5 mg/m3

mg/m3 mg/m3 mg/m3 mg/m3 mg/m3 mg/m3 mg/m3

8 minutes 0.1 mg/m3 0.1 mg/m3 1 mg/m3 0.1 ng/m3

Source: Ares and Bolton (2002).

monitored and rectified through process control (Ares & Bolton, 2002). The main flue gas components are as follows: •

•

Particulates. Particulate matter consists of a non-combustible fraction of waste combined with the solid products of incomplete combustion, often carbon. Organic carbon compounds. The main compounds of concern, other than dioxins and furans for which separate limits exist, are polycyclic aromatic hydrocarbons (PAHs). These are products of incomplete combustion of organic compounds. They are non-biodegradable, accumulate in fatty tissues and several of them have been linked to increased risk of cancer.

168


Acid gases. MSW contains corrosive and toxic acid gases, such as hydrogen chloride (HCl), hydrogen fluoride (HF), sulphur dioxide (SO2) and nitrogen oxides (NOx). The removal of these gases from the flue gas stream is relatively simple and very efficient. • Heavy metal compounds. Heavy metals exert a range of chronic and acute toxic health effects, including carcinogenic and neurological. Toxic effects associated with these metals generally occur at higher concentrations than those emitted by incinerators, but concentrations present in fly ash can be high, which makes correct disposal very important. This is particularly crucial as the metals are often present in watersoluble forms, which can leach into surrounding areas. • Dioxins. These represent a family of 210 closely related chlorinated chemical compounds. They can be formed as by-products in some chemical processes and in various combustion processes. Although the actual quantities of these compounds produced by modern thermal waste treatment processes are very low, their high toxicity requires their effective removal from the flue gas.


•

Flue gas treatment processes There are a number of physical and chemical processes that are used for the removal of pollutants and particulates that are present in flue gas streams. These are generally based around the following basic steps: addition of ammonia to combustion chamber; cooling; acid neutralisation; addition of activated carbon; filtration. Starting from the combustion chamber, NOx emissions are accelerated by high flame temperatures (e.g. by air preheat) and high excess air (Niessen, 2002). The use of ammonia can result in significantly lower NOx emissions, with reductions of up to 60%. Cooling and conditioning of the flue gases are essential before they are filtered. Cooling of flue gases may be achieved simply by passing them through a large chamber, which is fitted with cooling water sprays. Here, the flue gases must pass several stages of water sprays before they are allowed to the next stage of the cleaning process. Dioxins and furans, as well as heavy metals, such as mercury, are captured from the flue gas by the addition of activated carbon in a finely powdered form. The removal of acidic pollutants, such as HCl, HF and SO2, can occur in a Venturi reactor by adding a neutralising agent, such as hydrated lime or sodium bicarbonate. Some of the reagents used in flue gas treatment processes are shown in Table 7.6. In the filtration stage, the particulate matter is removed from the flue gas stream, as well as the spent activated carbon and lime, using cyclones, electrostatic precipitators and fabric bag filters.


169

Table 7.6 Reagents used in the flue gas treatment systems Application

Reagent

Neutralisation/removal of acid gases e.g. HCl, SO2, HF

Lime Hydrated lime Limestone Magnesium oxide Sodium bicarbonate Sodium hydroxide Ammonia Urea Activated carbon

Reduction of NO & NO2 to N2 Capture of dioxins/furans, VOCs and mercury Source: CIWM (2003).


Residues As EfW facilities give rise to various solid and liquid residues, these residues require safe disposal and management amid increasingly more stringent limits imposed by the European Integrated Pollution Prevention and Control (IPPC), Waste Incineration and Landfill Directives. The residues can either be disposed of or utilised, with or without pre-treatment. Bottom ash is classified as a non-special waste by most member countries of the International Energy Agency (IEA) and International Solid Waste Association (ISWA). (These countries are Austria, Belgium, Canada, Finland, France, Hungary, Japan, Netherlands, Norway, Spain, Sweden and the UK). APC residues, on the other hand, have higher levels of heavy metal and organic compounds present in them and a high level of hydrated lime, so they are generally classified as special waste. There are different residue treatment processes that are in current practice; however, it is important to distinguish between processes for disposal and those for utilisation purposes. For example, it is beneficial to limit the use and cost of material, such as additives, when treating for the purpose of disposal, while maintaining compliance with regulations. These treatment techniques include crushing, weathering, separating, mixing, chemical processes, thermal processes and solidification/stabilisation of the ash residues. Pre-treatment techniques to screen oversized components, remove ferrous metal and allow weathering of the material, are recognised as low cost procedures. These improve the chemical integrity and structural durability of the material prior to disposal or reuse applications. Other procedures, such as solidification/stabilisation, have additional processing requirements and, therefore, have higher processing costs (CRE Group, 2000). As already mentioned, bottom ash is widely utilised throughout Europe as a secondary aggregate in road construction and the building industry

170


(see Table 7.7). APC residues and fly ash, on the other hand, are hazardous and must be safely disposed of to specialist landfill sites. However, because of the higher costs for hazardous waste disposal in some countries and the prohibition of landfilling untreated residues in others, pre-treatment of the residues by solidification and stabilisation processes are preferred (CRE Group, 2000). One emerging technology, known as accelerated carbonation, stabilises APC residues using the CO2 in the process flue gas, producing a stone-like material suitable for construction applications (Li et al., 2007). Examples of engineering applications which employ residues from waste treatment plants are presented in Table 7.7.

7.2.5 Summary The use of MSW to produce either energy or fuel plays an important role in the UK’s waste strategy when integrated with recycling and reuse initiatives. Experience in other countries more advanced in recycling policy implementation than the UK, such as Sweden, Belgium and Germany, indicates that high recycling rates can co-exist with high EfW rates. EfW


Table 7.7 An overview of residue utilisation applications Waste material Bottom ash

End product

Use comments

Road construction

• Base course • Asphalt

• Used in cement stabilised bases • Larger sizes used as filler for

pavement

• Embankment Landfill cover

Building construction

asphalt

• Used as granular base • Requirements for coarse material

•

• Ferrous fraction Fly ash

are categorised according to permeability and/or particle size distribution Lightweight aggregate for construction material, filling material, interlocking blocks and concrete blocks Railway station construction

Metallurgic industry

• Ferrous fraction recycled in a

Civil engineering

• Asphalt filler, top sealing of

smelting plant landfill sites

• Concrete applications but requires pre-treatment, due to high Cl content APC residues

Civil engineering

• Potential for use as grout in coal mines

Source: CRE Group (2000).


171

not only reduces the amount of waste sent to landfill, but can also reduce our reliance on fossil fuels and, hence, can help the UK Government achieve its target of 60% reduction in carbon emissions by 2050 and 10% of UK electricity generation from renewable sources by 2010. EfW also contributes to energy security through diversification of supply, as up to 17% of the total UK electricity consumption could be supplied by EfW by 2020. Therefore, waste management should not be seen as a one-step disposal process, but rather as an integrated strategy that incorporates several handling and treatment steps, such as waste separation, recycling, energy recovery and residue management.

7.3

Techno-economic analysis of energy from waste advanced thermal processes


7.3.1 Setting the scenarios This section evaluates the technical and economic performance of combustion and gasification systems and reports on the implications of different scales and technologies on costs and efficiencies. Two different scale scenarios of 50 and 100 ktpa waste feed have been considered, corresponding to small- and medium-scale plant capacities, respectively. For each scale scenario, the different waste treatment options evaluated were as follows: (1) fluidised bed gasification with; a gas engine (FBG + GE); and a combined cycle gas turbine, CCGT (FBG + CCGT); and (2) fluidised bed combustion with a steam turbine (FBC + ST). Figure 7.8 illustrates the two thermal treatment processes studied for this evaluation.

Air Flue gas Fluidised bed combustion

Heat recovery boiler

Bottom ash

Steam turbine

Flue gas treatment

Exhaust gas

Steam/water Residual waste as RDF

APC residues

Electricity Air Syngas Fluidised bed gasification Bottom ash

Fuel gas treatment

Gas engine

Exhaust gas

CCGT APC residues Electricity

7.8 Energy recovery from residual waste – two process options.

172


The steam generated from the combustion process is fed into an energy conversion system, which generates electricity using a steam turbine. Any contaminants in the flue gas, such as particulates and acidic pollutants, are removed by the flue gas treatment system before the gas is released to the atmosphere. For the gasification process, the syngas is cleaned from any contaminants before it is utilised by the energy conversion system for electricity generation, either a gas engine or CCGT unit.


7.3.2 Methodology The mass and energy balances for the different thermal treatment options form the basis of the input parameters of process simulation models developed using the software package ChemCAD®, which then allow full techno-economic analyses to be made. The results from the process simulations are discussed in full in Yassin (2008), whilst what will be discussed here are the results of the techno-economic analyses (full details of which can be found in Yassin et al., 2009). However, it should be noted that the properties of the waste feed have a significant impact on the mass and energy balances (and, hence, the subsequent techno-economic analyses) and can vary greatly depending on many factors, such as waste type, the area of collection (rural, urban or commercial), seasonal variations in consumer habits and recycling levels. The waste characteristics used for the combustion and gasification processes considered here were provided by Germanà & Partners Consulting Engineers (Yassin et al., 2009) and refer to RDF, as opposed to raw, untreated, MSW. The extra costs associated with the pre-treatment process (MBT) are discussed later. Table 7.8 reports the proximate and ultimate analyses of the RDF used in the process model development. The mass and energy balances enable the comparison of the technical performances of the different thermal treatment options by determining their overall system efficiencies. System efficiencies are defined as the ratio of the net generated electricity to the energy input to the system, as shown in Equation 7.1. Table 7.8 Proximate and ultimate analysis of the RDF of the fluidised bed processes Proximate analysis (wt %)

Ultimate analysis (wt %)

Fixed Moisture Volatiles Inerts LHV C carbon (%) (%) (%) (%) (MJ/kg) (%) 10.7

15.8

53.5

20.0

16.7

H O (%) (%)

69.6 5.8

N S Cl (%) (%) (%)

22.3 0.9 0.6 0.9


173

Gross electrical efficiency (%)

60 y = 0.0589lnx + 0.2565

50 40

y = –0.002392x2 + 0.313x + 38.6 30 20 y = 0.0516lnx + 0.0794 10

CCGT Gas engine Steam turbine

0 0

10

20

30

40

Thermal energy input (MWth)

7.9 Gross electricity generation efficiencies of the prime movers.


Power output or net generated electricity [ MW ] System = × 100 efficiency [%] Energy input to system [ MW ]

[7.1]

However, to obtain these values, the combustion and gasification efficiencies, as well as the performances of the different prime movers, i.e. the steam turbines, gas engines and CCGT units, need to be obtained. Gasifiers can achieve between 70 and 93% cold gas efficiency, with most operating at between 75 and 88% (Germanà and Partners Consulting Engineers, 2007). The cold gas efficiency can be defined as the ratio of the energy content of the syngas to the energy content of the waste feedstock (Higman and van der Burgt, 2003). A cold gas efficiency of 70% was used in this analysis, an assumption made owing to the lack of proven commercial EfW gasification plants in the UK. On the other hand, a boiler efficiency of 90% is assumed for the combustion process (Germanà and Partners Consulting Engineers, 2007). The performances of the prime movers were obtained using literature data published by Bridgwater et al. (2002), which are presented in Fig. 7.9 for a range of thermal energy inputs of 1–40 MWth. Figure 7.9 illustrates the relationship between the thermal energy input to the prime movers and their corresponding gross electrical generation efficiencies. In this analysis, the electrical generation efficiency is defined as the ratio of power output to the energy supplied to the prime mover (see Equation 7.2). Electrical conversion Power output [ MW ] = × 100 efficiency [%] Energy input to prime mover [ MW ]

[7.2]

174


The efficiencies described above set the flow throughputs in the process simulation models, leading to the required sizes of the unit operations and, ultimately, the capital and operating costs of the processes being modelled. These then form the basis of the economic comparison.


Economic model An economic model was developed using a consistent methodology to allow for the comparison between the different process and technology options. The model consisted of the capital costs, operating costs and projected annual revenues (see Section 7.4 for sample calculations) and used a basic discounted cash flow (DCF) analysis (Peters & Timmerhaus, 1991; Gerrard, 2000; Sutherland, 2007), relating the values of the costs and revenues that occur over the economic life of the project in terms of present worth, i.e. the amount that a future sum of money is worth today given a specified rate of return. For ease of comparison, the model estimates the levelised costs of waste treatment and predicted gate fees for the different waste treatment options. The levelised cost is a useful tool for comparing different technologies as it calculates the cost of producing a unit of output from the proposed systems. Gate fees are typically paid by local authorities to contractors for the disposal and treatment of waste. Usually, the lower the gate fee, the more attractive is the waste treatment option. The available data in the literature for the capital costs of advanced thermal treatment processes, such as gasification and pyrolysis, vary significantly from one plant to another (Wheeler & de Rome, 2002). In this study, the capital cost of a Novera Energy-type facility has been adopted for the cost of the gasification system, since the facility uses a similar technology and plant configuration to that considered for this evaluation (AiIE, 2003; Defra, 2005b). The costs of a gas engine and CCGT unit were then added to obtain the overall system costs (EDUCOGEN, 2001). For the combustion process, the capital cost of the Dundee fluidised bed EfW facility has been adopted (PFI Scotland, 1998). All cost data were updated and reported in £2006, using appropriate indices from the Office for National Statistics (ONS). Where the cost data were unavailable, Equation 7.3 was used, which gives the general relationship between cost and scale. Cost1 ⎛ Scale1 ⎞ = Cost 2 ⎝ Scale 2 ⎠

n

[7.3]

where Cost1 is the cost of the proposed plant in £, which is at Scale1 in ktpa; Cost2 is the cost of the reference plant in £, which is at Scale2 in ktpa; and n is the scale exponent. The scale exponent is derived from historical data



175

for similar plants and is usually in the range of 0.6 to 0.8 (Gerrard, 2000; Peters & Timmerhaus, 1991). The operating costs of the different waste treatment options have been divided into maintenance and consumable costs, labour, ash disposal, running costs of the energy conversion systems and plant overheads (see Table 7.14 in Section 7.4). Projected revenues from the different waste treatment options depend on gate fees, sales of electricity and the application of the different financial incentives available at the present time, which are: Renewables Obligation Certificates (ROCs – 68% of the waste is regarded as biodegradable, hence eligible for ROCs); Levy Exemption Certificates (LECs – this represents the value for being exempt from the climate change levy on electricity); Packaging Recovery Notes (PRNs – these are part of the UK producer responsibility requirements introduced to meet the EU Packaging and Packaging Waste Directive (94/62)), and sales of secondary aggregates. The gate fee is levied on each tonne of MSW (or, in the case of the fluidised bed systems, RDF) taken in for thermal treatment, in order to offset the total operating costs of the facility. It also takes into account the capital costs of the facility and revenues generated. The gate fees for MSW plants in the UK using gasification and pyrolysis processes vary between £25/t and £100/t, while gate fees between £36/t and £55/t have been reported for new large-scale EfW combustion plants (House of Commons, 2007; Juniper, 2003; McLanaghan, 2002). In the analysis reported here, the gate fee was calculated using the DCF analysis to balance the net present values of costs and revenues, over the plant lifetime of 30 years, and includes an operator profit of 20% (see Equation 7.4). The impact of ROCs on the gate fee for the gasification systems was also evaluated. 30

Gate fee = ∑ [ PV ( costs ) − PV ( income)]

[7.4]

n =1

where PV is present value and n is the plant lifetime (see Section 7.4.1) The levelised cost quantifies the unitary cost of electricity produced or waste treated during the plant lifetime and is reported in p/kWh or £/tonne of input waste. It is calculated as the ratio of the total plant lifetime expenses against total expected outputs, expressed in terms of present worth (NEA and IEA, 2005). Standardised financial tools, such as the net present value (NPV) and internal rate of return (IRR), were employed to assess the profitability of the different options. A discount rate of 6% was used and all costs and revenues were assumed to be constant. An option is economically attractive if it has the highest IRR and the NPV is greater than zero. The NPV refers to the difference between the present values of all costs and associated

176


revenues. This is shown in Equation 7.5, where i is the discount rate, CFn is the annual cash flow (revenues–operating costs) at the nth year and TPC is the total plant cost. The IRR was calculated as the discount rate that makes the NPV equal to zero (Sutherland, 2007). 30

CFn − TPC n n = 1 (1 + i )

NPV = ∑

[7.5]

For a full description of the economic model, including capital and operating costs, the reader is referred to Yassin et al. (2009).


7.3.3 Technical performance The net electricity generated by the different treatment systems and their overall system efficiencies are reported in Tables 7.9 and 7.10 for the two plant scales of 50 and 100 ktpa, respectively (see sample calculations in Section 7.4). The results demonstrate that the ability of gasification processes to employ more efficient energy conversion systems, such as gas engines and CCGT, enables them to have greater electrical generation efficiencies and, as a result, they can have better overall system performances than combustion processes that use steam turbines. Fluidised bed gasification coupled with CCGT (FBG + CCGT), in particular, offers the most energy efficient treatment option, with overall system efficiencies of 26% and 28% for both scale scenarios of 50 and 100 ktpa, respectively. Fluidised bed gasification systems using gas engine (FBG + GE) have overall efficiencies of 24% and 26%, while efficiencies of 18% and 22% are reported for the combustion systems (FBC + ST). Table 7.9 Technical performances of treatment options at 50 ktpa scale scenario Waste treatment options FBG + GE Thermal energy of waste Mass flow rate Gross electrical generation efficiency of prime movers Gross generated electricity Site power use Net generated electricity Overall system efficiency

FBG + CCGT

FBC + ST

38.4 (%)

29.4 (MWth) 6.3 (t/h) 43.5 (%)

24.9 (%)

7.9 (MWe) 0.9 (MWe) 7.0 (MWe) 55 500 (MW he) 23.9 (%)

9.0 (MWe) 1.3 (MWe) 7.6 (MWe) 60 000 (MW he) 25.9 (%)

6.6 (MWe) 1.2 (MWe) 5.4 (MWe) 42 400 (MW he) 18.3 (%)


177

Table 7.10 Technical performances of treatment options at 100 ktpa scale scenario Waste treatment options FBG + GE Thermal energy of waste Mass flow rate Gross electrical generation efficiency of prime movers Gross generated electricity Site power use Net generated electricity


Overall system efficiency

41.7 (%)

17.2 (MWe) 1.9 (MWe) 15.3 (MWe) 120 600 (MW he) 26.0 (%)

FBG + CCGT 58.8 (MWth) 12.7 (t/h) 47.6 (%)

FBC + ST

28.4 (%)

19.6 (MWe) 15.1 (MWe) 2.9 (MWe) 1.9 (MWe) 16.7 (MWe) 13.2 (MWe) 131 200 (MW he) 103 700 (MW he) 28.3 (%) 22.4 (%)

The results also show the greater sensitivity of the technical performances of FBC + ST to scale. The combustion system efficiencies increased by over 22% with the doubling of the plant capacity, compared to an increase of 8–9% for the gasification systems. This highlights the nature of the combustion processes, which are centralised operations and technically more efficient at larger scales.

7.3.4 Economic performance The economic performances of the fluidised bed combustion and gasification systems are summarised in Tables 7.11 and 7.12 for the two plant scales of 50 and 100 ktpa. The results show that gasification systems represent the cheapest option, with capital costs ranging from £16 to 27 million. However, FBG + CCGT systems have higher costs than FBG + GE systems, reflecting the higher capital investment for the more efficient CCGT system configuration. On the other hand, capital costs of £30–48 million are reported for the combustion systems. It can be concluded from the analysis that conventional combustion systems, in this case based on fluidised beds, are not as competitive at smallto-medium scales as the more compact gasification systems, which can be built economically as modular units at smaller scales. This is mainly because combustion systems need to have large boilers to recover heat and gas cleaning systems to clean the large volumes of flue gas generated. On the other hand, the calculated operating costs of the different treatment options show that combustion systems have the lowest costs, with

178


Table 7.11 Economic performances of treatment options at 50 ktpa scale scenario Waste treatment options

Capital costs Operating costs NPV @ 6% discount rate


IRR Gate fees without ROCs with ROCs Levelised costs in terms of electricity generated in terms of waste treated

FBG + GE

FBG + CCGT

16.0 321 2.8 57 11.4 228 12.1

16.8 336 2.8 55 11.4 227 11.8

(£m) (£/t) (£m) (£/t) (£m) (£/t) (%)

67 (£/t) 42 (£/t) 7.46 (p/kW h) 83 (£/t)

(£m) (£/t) (£m) (£/t) (£m) (£/t) (%)

65 (£/t) 37 (£/t) 6.87 (p/kW h) 82 (£/t)

FBC + ST 29.7 594 2.3 47 12.7 255 9.8

(£m) (£/t) (£m) (£/t) (£m) (£/t) (%)

87 (£/t) 87 (£/t) 10.91 (p/kW h) 93 (£/t)

Table 7.12 Economic performances of treatment options at 100 ktpa scale scenario Waste treatment options

Capital cost Operating cost NPV @ 6% discount rate IRR Gate fees without ROCs with ROCs Levelised costs in terms of electricity generated in terms of waste treated

FBG + GE

FBG + CCGT

FBC + ST

27.0 270 5.4 54 21.1 211 12.7

27.3 273 5.2 52 20.6 206 12.5

48.1 481 4.3 43 22.1 221 10.0

(£m) (£/t) (£m) (£/t) (£m) (£/t) (%)

57 (£/t) 29 (£/t) 6.35 (p/kW h) 77 (£/t)

(£m) (£/t) (£m) (£/t) (£m) (£/t) (%)

52 (£/t) 22 (£/t) 5.71 (p/kW h) 75 (£/t)

(£m) (£/t) (£m) (£/t) (£m) (£/t) (%)

67 (£/t) 67 (£/t) 7.76 (p/kW h) 80 (£/t)


179

reported annual costs of £47/t and £43/t for the plant capacities of 50 and 100 ktpa. These costs also illustrate the greater sensitivity of the combustion systems to economies of scale, as doubling the plant capacity reduced the operating costs by 8%. The NPV of the different waste treatment options are positive, thus indicating that they are all economically viable at both plant scale scenarios. However, although combustion systems seem to be the most attractive options, with higher NPV of £221/t and £255/t, compared with £206/t to £228/t for the gasification systems, the latter yield a better rate of return on investment. The average IRR for gasification is 12%, whilst it is 10% for combustion. ATT processes, including gasification, are eligible for ROCs for the electricity generated from the biomass fraction of the waste, while combustion processes are only eligible when combined with good quality CHP. The results show a significant reduction in the gate fees for the gasification systems at 50 and 100 ktpa when ROCs are included, thus enabling them to be a more attractive and cheaper option than combustion.


7.3.5 Sensitivity analysis A sensitivity analysis was carried out on all the treatment options, as each input parameter or system variable can affect the overall system performance to a different degree. Seventeen different system variables were chosen for the sensitivity analysis and the effects of a ± 10% change in these variables on the levelised costs and gate fees have been examined. The results are presented in Fig. 7.10, which shows that the calorific value of the waste, conversion efficiencies of the prime movers and gasifier efficiency have the greatest impact on the levelised cost, whilst the gate fee is also shown to be affected by the capital costs as well as electricity prices, ROCs and the biomass eligibility.

7.3.6 Comparison with traditional moving-grate combustion The technical and economic assessment of the different fluidised bed systems have thus far been considered only for the treatment of 50–100 ktpa of RDF, which was assumed to be supplied by MBT facilities. If we assume that the MBT facilities are based on the Ecodeco process, which is the same process employed by Shanks in their MBT plant at Frog Island, then 50% of the raw MSW input will be converted into RDF (Shanks, 2007). Hence, the actual raw material input required is double that of the RDF input into the EfW plants based on fluidised bed technologies. Since traditional

20%

15%

Discount rate

Plant lifetime

10%

Discount rate

Labour cost

Maintenance & consumable cost

Operating cost

Capital cost

Elec. generation efficiency

15%

ROCs

Biomass eligibility

Electricity price

Operating cost

Capital cost

Elec. generation efficiency

Gasifier efficiency

Calorific value

–15%

Gasifier efficiency

Calorific value


180 Management, recycling and reuse of waste composites +10% change –10% change

5%

0%

–5%

–10%

+10% change –10% change

10%

5%

0%

–5%

–10%

–15%

–20%

7.10 Effects of changes in model input parameters on levelised costs (top) and gate fees (bottom) for a 100 ktpa FBG + CCGT only.


181

Table 7.13 Technical performances of moving-grate combustion systems Waste treatment option

Moving-grate combustion

Plant scale scenario

100 ktpa

Thermal energy of waste Mass flow rate Gross electrical generation efficiency of steam turbine Gross generated electricity Site power use Net generated electricity


Overall system efficiency

200 ktpa

28.7 (MWth) 12.7 (t/h) 24.7 (%)

56.8 (MWth) 25.1 (t/h) 28.2 (%)

6.4 (MWe) 0.9 (MWe) 5.5 (MWe) 43 300 (MW he) 19.1 (%)

14.4 (MWe) 2.0 (MWe) 12.4 (MWe) 99 000 (MW he) 21.9 (%)

combustion processes employing moving-grates (commonly known as incineration) can handle raw MSW directly, without any pre-treatment, we have conducted a similar techno-economic analysis on plant scales of 100 and 200 ktpa raw MSW feed to complete the comparison between the various technologies. The results are presented in Table 7.13. It can be seen that the moving-grate systems have similar system efficiencies to that of fluidised bed combustion, which has efficiencies of 18% and 22% at the smaller plant scales of 50 and 100 ktpa, respectively, as reported in Tables 7.9 and 7.10. However, the fluidised bed systems utilise the higher calorific value RDF and the energy used in its production has not been taken into account and, therefore, the energy requirement for the conversion of raw MSW into RDF for an Ecodeco MBT process needs to be added to the overall site power used by the fluidised bed systems. The energy required for the MBT process is reported as being between 30 and 32 kW h per tonne of input waste (Paiola, 2007), corresponding to a 16–20% increase in the site power usage (or the internal energy consumption) of the fluidised bed combustion systems. This in turn would reduce their system efficiencies by 2–3%, to 17.6% and 21.7% for the 50 and 100 kpta scales, respectively. In contrast, the fluidised bed systems with gas engine and CCGT are much more efficient, even with the energy penalty of MBT included. Figure 7.11 shows the capital costs of the different waste thermal treatment options calculated on the basis of 100 and 200 ktpa untreated MSW. The figure demonstrates the overall competitiveness of fluidised bed gasification systems co-located with MBT processes, compared to movinggrate combustion systems. On the other hand, fluidised bed combustion combined with MBT is only cost competitive for the treatment of 100 ktpa raw MSW.

182

Management, recycling and reuse of waste composites 80 70

Capital costs (£m)

60

FBC + ST

50

FBG + GE

40

FBG + CCGT MBT

30

Moving-grate

20

(200 ktpa) Moving-grate

(200 ktpa) MBT + FBC

(200 ktpa) MBT + ATT


(100 ktpa) Moving-grate

(100 ktpa) MBT + FBC


0


10

Treatment options at 100 and 200 ktpa


7.11 Costs of fluidised bed systems with MBT compared with movinggrate combustion (scales based on untreated MSW input).

7.3.7 Summary For the different fluidised bed waste thermal treatment options, the technoeconomic analysis has shown that the ability of gasification processes to employ more efficient energy conversion systems enables them to have greater electrical generation efficiencies and, as a consequence, they have better overall system performances compared with combustion processes. Fluidised bed gasification coupled with CCGT, in particular, offers the most energy efficient treatment option. Fluidised bed gasification coupled with a gas engine has the cheapest capital cost option and the highest rate of return on investment. However, this is offset by its higher operating cost and lower system efficiency, compared with fluidised bed gasification coupled with a CCGT, which is the most attractive treatment option in terms of gate fee and levelised cost of waste treatment. Although fluidised bed gasification systems have an unproven track record in the UK, they are compatible with high levels of material segregation, and when co-located with recycling and MBT, therefore, have the potential to contribute towards integrated waste management practices. In addition, the operational reliability of the systems will be further improved as more facilities are commissioned and operated at commercial scales.


183


Furthermore, financial incentives, such as ROCs, securing long-term contractual agreements for the supply of RDF, as well as supportive policies and active R&D by major industry players and research institutions, are important factors for the full commercialisation of these processes, especially for plant scales larger than 50 ktpa RDF feed. The sensitivity analysis has demonstrated that the calorific value of the waste, the electricity generation efficiencies of the prime movers and the gasifier efficiency has the greatest impact on the levelised cost, whilst the gate fee is mainly affected by the operating costs, as well as electricity and ROC prices and the biodegradable fraction of the waste. Finally, although traditional moving-grate combustion systems have been shown to have lower technical and economic performances, compared with fluidised bed gasification systems co-located with MBT facilities, they have the highest landfill diversion potential, assuming the bottom ash is recycled. This said, gasification systems co-located with MBT facilities can achieve higher levels of recycling, although the market availability for their outputs will have a significant influence on the environmental impacts and, hence, competitiveness of these processes.

7.4

Sample calculations

7.4.1 Technical performance of a 50 ktpa FBG + CCGT EfW plant The starting point is Equation 7.1, which enables the overall system efficiency to be determined, where the net electricity generated, Eelectricity,net, expressed in MWe, is given by Equation 7.6: Eelectricity,net = Eelectricity,gross − Eauxiliary

[7.6]

The gross electricity depends on the gross electrical efficiency of the CCGT unit, ηCCGT, and the thermal energy input to the CCGT unit, Eth,syngas, which is the thermal capacity of the syngas: Eelectricity,gross = ηCCGT × Eth,syngas

[7.7]

The thermal capacity of syngas is given by: Eth,syngas = ηth,gasifier × Eth

[7.8]

184


where ηth,gasifier is the thermal efficiency of the gasifier, assumed to be 70%, and Eth is the thermal capacity of the waste feedstock (RDF), which is given by: Eth = mRDF × CVRDF

[7.9]

with the calorific value of the RDF equal to: CVRDF = 4000 kcal kg = 4000 ×

4.186 = 16.7 MJ kg 1000

and the mass feed rate of RDF, mRDF, given by: mRDF =

MRDF 50 000 = = 6.34 t h hrp 7884

MRDF is the amount of waste to be processed and hrp is the number of plant annual operating hours assumed at 90% operation. Hence, from Equation (7.9):


Eth = mRDF × CVRDF =

16.7 × 6.34 × 1000 = 29.42 MWth 3600

The thermal capacity of the gas, Equation 7.8, is therefore equal to: Eth,syngas = ηth,gasifier × Eth =

70 × 29.42 = 20.59 MWth 100

Figure 7.9 is used to calculate the gross electrical generation efficiency of the CCGT, ηCCGT, as follows:

ηCCGT = 0.0589 × ln ( Eth,syngas ) + 0.2565 = 0.0589 × ln ( 20.59 ) + 0.2565 = 43.47% Hence, the gross electricity generated, Equation 7.7, is equal to: Eelectricity,gross = ηCCGT × Eth,syngas = 0.4347 × 20.59 = 8.95 MWe The auxiliary consumption is assumed to be 15% of the gross electricity: Eauxiliary =

15 × 8.95 = 1.34 MWe 100

Hence, the net electricity generated from Equation 7.6 is equal to: Eelectricity,net = Eelectricity,gross − Eauxiliary = 8.95 − 1.34 = 7.61MWe


185

and the overall system efficiency from Equation 7.1 is equal to: Eelectricity,net 7.61 × 100 = × 100 = 25.87% Eth 29.42

7.4.2 Economic performance of a 50 ktpa FBG + CCGT EfW plant This section reports the calculations used for the capital, operating costs and revenues. Capital costs


The capital cost of the 50 ktpa FBG + CCGT plant consists of the costs of the gasification system and CCGT unit. The capital cost of the gasification system was calculated using Equation 7.3, based on the type supplied by Novera Energy, which is approximately £15 m for a 70 ktpa plant (Defra, 2005b). Using values for n = 0.6 and 0.8, the capital cost for a 50 ktpa plant is calculated as follows: ⎡⎛ 50 000 ⎞ Capital cost = ⎢⎜ ⎟ ⎣⎝ 70 000 ⎠

0.6

⎤ × 15 000 000 ⎥ = £12 257 854, when n = 0.6 ⎦

⎡⎛ 50 000 ⎞ = ⎢⎜ ⎟ ⎣⎝ 70 000 ⎠

0.8

⎤ × 15 000 000 ⎥ = £11460111, when n = 0.8 ⎦

and

The average capital cost is therefore £11 858 982. In this analysis, all cost data are updated and reported in £2006, using appropriate indices from the Office for National Statistics (ONS), which in this analysis was 1.03. This is based on the Retail Prices Index (RPI), which is the most familiar general purpose domestic measure of inflation in the UK (ONS, 2007). Hence, the average capital cost of the gasification system is £11 858 982 × 1.03 = £12 214 752. The capital cost for CCGT unit, with a 9.0 MWe gross electricity generating capacity is €744/kWe (EDUCOGEN, 2001). Using a conversion rate of £1 = €1.45, the capital cost of the CCGT unit is: Capital cost = unit cost × Eelectricity,gross 744 = × (8.95 × 1000 ) = £4 592 275 1.45

186


Therefore, the total capital cost of a 50 ktpa FBG + CCGT plant is: £12 214 752 + £4 592 275 = £16 807 027

Operations costs Table 7.14 shows the model parameters used for the operating cost calculations. Thus, the total operating costs = Σoperating costs = £2 764 124/annum. Revenues The projected annual revenues from the 50 ktpa FBG + CCGT plant include sales from electricity, eligibility for ROCs, LECs, PRNs, sales of bottom ash and income from gate fees. Table 7.15 shows the model parameters used for the revenues calculations. Thus, the total annual revenues = Σannual revenues = £4 948 918.


Table 7.14 Model parameters used for the annual operating cost calculations† Parameter

Value

Maintenance

£20.0/t of input waste (AiIE, 2003) ⇒ cost of maintenance × amount of waste treated = Cma int enance × MRDF = 20 × 50 000 = £1 000 000 15 employees with average salaries of €45 000 (Thurgood, 1999). ⇒ cost of labour × number of employee 45000 = Clabour × nemployees = × 15 = £465517 1.45 20% of input waste, of which 1/3 is bottom ash and 2/3 is air pollution control residues (Howson, 2007). The bottom ash is assumed to be recycled, while the APC residues is sent to a hazardous landfill. ⇒ cost of hazardous landfill × amount landfilled 20 2 = Chazardous landfill × Mamount landfilled = 401.21 × × × 50 000 100 3 = £2 674 733 £3.6/MW he (EDUCOGEN, 2001). ⇒ cost of maintenance × annual gross electricity generated = Cma int enance,CCGT × (Eelectricity,gross × hrp) = 3.6 × (8.95 × 7884) = £254 022 2% of capital costs (Bridgwater et al., 2002). 2 ⇒ × 16 808107 = £336162 100

Labour

Ash disposal

(

Maintenance of CCGT unit

Plant overheads

†

)

Small variations will appear in final values due to rounding figures up/down.


187

Table 7.15 Model parameters used for revenue calculations Parameter

Value

Sales from electricity

2.50 p/kW h (Jacobs Babtie, 2005; Enviros, 2005) ⇒ electricity price × annual net electricity generated 2.5 = Pelectricity × (E electricity,net × hrp ) = × ( 7.61 × 1000 × 7884) 100 = £1 499 931 3.43 p/kW h (Ofgem, 2007a). ⇒ ROC price × annual net electricity generated × biomass fraction 3.43 68 = PROC × (E electricity,net × hrp ) = × ( 7.61 × 1000 × 7884) × 100 100 = £1 399 375 0.44 p/kW h (Ofgem, 2007b). ⇒ LEC price × annual net electricity generated × biomass fraction 0.44 68 = PLEC × (E electricity,net × hrp ) = × ( 7.61 × 1000 × 7884) × 100 100 = £179 511 £2/t for 19% of waste treated (letsrecycle.com, 2007; Environment Agency, 2007). ⇒ PRN price × annual amount of waste treated 19 = PPRN × MRDF = 2 × 50000 × = £19000 100 £7/t (WRAP, 2006). ⇒ Bottom ash price × annual amount of bottom ash recycled 20 1 = PBottom ash × MBottom ash = 7 × × × 50000 = £23 333 100 3 £36.55/t, which accounts for ROCs (see discounted cash flow analysis). ⇒ Gate fee × annual amount of waste treated = PGate fee × MRDF = 36.55 × 50 000 = £1 827 500

Eligibility for ROCs

LECs


PRNs

Sales of bottom ash

Gate fee

7.5

(

)

Conclusions

In this chapter, we have introduced the problem of waste management and the legislations at UK and EU level that regulate the treatment of waste. We have shown how their implementation may lead to reduced greenhouse gas emissions from the treatment of waste. We have looked at the different ways of recovering energy from municipal solid waste, viz. combustion, advanced thermal treatment (gasification and pyrolysis) and traditional incineration. We have shown how to estimate a technical and economic performance of an EfW gasification plant and how it compares with combustion and incineration for two different plant scales. Finally, we have shown, through sensitivity analysis, that the levelised costs are mostly

188


affected by the calorific value of the feed material and the gasifier and electricity generation efficiency, whilst the gate fees are most sensitive to the operating costs, biomass eligibility and renewable certificate obligations. Hence, we have shown the usefulness of techno-economic analysis in determining the most appropriate scales and technologies for EfW plant.


7.6

Notation

ACFB AEE AICE APC ATT BA BPEO CCGT CCPT CF CFB CHP DCF EfW FBC FBG FICFB FW GDP GE GHG HHV IEA IPCC IRR ISWA ktpa kW h LCA LECs LHV MBT MSW MWe/MWth MW h

Atmospheric circulating fluidised beds Austrian Environmental Agency Austrian Institute of Chemical Engineering Air pollution control Advanced thermal treatment Bottom ash Best practicable environmental option Combined cycle gas turbine Combined cycle power plant Cash flow Circulating fluidised beds Combined heat and power Discounted cash flow Energy from waste Fluidised bed combustion Fluidised bed gasification Fast internal circulating fluidised bed Foster Wheeler Gross domestic product Gas engine Greenhouse gas High heating value International Energy Agency Intergovernmental Panel on Climate Change Internal rate of return International Solid Waste Association Kilotonne per annum (1 000 000 kg per annum) Kilowatt hour (1000 watt hours) Life-cycle analysis Levy exemption Certificates Lower heating value Mechanical biological treatment Municipal solid waste Megawatt (electrical/thermal) Megawatt hour (1 000 000 watt hours)

Advanced thermal treatment of composite wastes NPV ONS PAH PBFB PCFB PPS PRNs RCEP RDF ROCs ST TPC tpa VOC WID TPS


7.7

189

Net present value Office for National Statistics Polycyclic aromatic hydrocarbon Pressured bubbling fluidised beds Pressured circulating fluidised beds Planning and policy statement Packaging recovery notes Royal Commission on Environmental Pollution Refuse-derived fuel Renewables Obligation Certificate Steam turbine Total plant cost Tonne per annum (1000 kg per annum) Volatile organic compound Waste Incineration Directive Termiska Processer Sweden

References

aiie. (2003). Review of Residual Waste Treatment Technologies. Associates in Industrial Ecology, Penrith, UK. ares, e., bolton, p. (2002). Waste Incineration. Research Paper 02/34. House of Commons Library, London, UK. biomass task force. (2005). Report to Government. Defra, Crown, London, UK. bridgwater, a.v. (2003). Renewable fuels and chemicals by thermal processing of biomass. Chemical Engineering Journal, 91, 87–102. bridgwater, a.v., meier, d., radlein, d. (1999). An overview of fast pyrolysis of biomass. Organic Geochemistry, 30, 1479–1493. bridgwater, a.v., toft, a.j., brammer, j.g. (2002). A techno-economic comparison of power production by biomass fast pyrolysis with gasification and combustion. Renewable and Sustainable Energy Reviews, 6, 181–248. ciwm. (2003). Energy from Waste: A Good Practice. IWM Business Services, Northampton, UK. cre group. (2000). The Management of Residues from Thermal Processes. IEA Bioenergy, Dublin, Ireland. defra. (2005a). Securing the Future: Delivering Sustainable Development Strategy. Defra, London, UK. defra. (2005b). New Technologies Demonstrator Programme: Catalogue of Applications. Defra, London, UK. defra. (2007a). UK Biomass Strategy. Defra, London, UK. defra. (2007b). Waste Strategy for England 2007. Defra, London, UK. defra. (2007c). Advanced Thermal treatment of Municipal Solid Waste. Defra, London, UK. defra. (2007d). Incineration of Municipal Solid Waste. Defra, London, UK. educogen. (2001). A Guide to Cogeneration. EDUCOGEN, Brussels, Belgium.


190


eea briefing 2008/01 Better Management of Municipal Waste will Improve Greenhouse Gas Emissions. environment agency. (2007). Applying for an Accreditation to Reprocess or Export UK Waste Packaging. Guidance Notes. Environment Agency, Bristol, UK. enviros. (2005). Municipal Waste Management Strategy. Enviros, London, UK. erm & golders associates. (2006). Carbon Balances and Energy Impacts of the Management of UK Wastes. Defra R&D Project WRT 237. Defra London, UK. european commission. (2005). Biomass – Green Energy for Europe. European Communities, Luxembourg. fichtner consulting engineers. (2004). The Viability of Advanced Thermal Treatment of MSW in the UK. ESTET, London, UK. gasification technologies council. (2006). See www.gasification.org for further details. Accessed 9 January 2006. gla. (2003). City Solutions: New and Emerging Technologies for Sustainable Waste Management. GLA, London, UK. germanà & partners consulting engineers, personal communication (2007). gerrard, a.m. (2000). Guide to Capital Cost Estimating, 4th Edition. Institution of Chemical Engineers, Rugby, UK. higman, c., van der burgt, m. (2003). Gasification. Elsevier Science, Burlington, USA. house of commons. (2007). Waste Disposal. Written Answers to Questions. House of Commons, London, UK. howson, j. (2007). Novera Energy, Personal communication. iea. (2003). Municipal Solid Waste and its Role in Sustainability. IEA Bioenergy. Dublin, Ireland. jacobs babtie. (2005). Waste Management Treatment Options Appraisal. Jacobs Babtie, Reading, UK. jörß, w., joergensen, b.h., loeffler, p., morthorst, p.e., uyterlinde, m.a., sambeek, e.j.w. van, wehnert, t., groenendaal, b., marin, m., schwarzenbohler, h., wagner, m. (2002). Decentralised Generation Technologies. Potentials, Success Factors and Impacts in the Liberalised EU Energy Markets. The DECENT Final Report. COGEN Europe, Brussels, Belgium. juniper. (2003). Advanced Conversion Technology (Gasification) for Biomass Projects. Juniper, Uley, UK. lee, p., fitzsimons, d., parker, d. (2005). Quantification of the Potential Energy from Residuals (EfR) in the UK. Technical Report commissioned by the Institution of Civil Engineers and the Renewable Power Association. Oakdene Hollins Ltd, Aylesbury, UK. letsrecycle.com. (2007). Novera ‘Optimistic’ of Alternative Funding for Gasification Plant. See www.letsrecycle.com for further details. Accessed 1 October 2007. li, x., fernandez bertos, m., hills, c.d., carey, p.j. and simons, s.j.r. (2007). Accelerated carbonation of municipal solid waste incineration fly ashes. Waste Management, 27, 1200–1206. mclanaghan, s.r.b. (2002). Delivering the Landfill Directive: The Role of New & Emerging Technologies. A Report for the Strategy Unit. Associates in Industrial Ecology, Penrith, UK. nea and iea. (2005). Projected Costs of Generating Electricity. OECD, Paris, France. niessen, w.r. (2002). Combustion and Incineration Processes. Marcel Dekker, New York, USA.



191

ofgem. (2007a). Renewables Obligation: Guidance for Licensed Electricity Suppliers (Great Britain). Office for Gas and Electricity Markets, London, UK. ofgem. (2007b). Climate Change Levy Exemption for Renewables. Office for Gas and Electricity Markets, London, UK. odpm. (2005). Planning Policy Statement 10: Planning for Sustainable Waste Management. Crown, London, UK. ons. (2007). Retail Prices Index. See www.statistic.gov.uk for further details. Accessed 8 January 2007. paiola, m. (2007). Ecodeco, The Intelligent Transfer Stations, personal communication. peters, m.s., timmerhaus, k.d. (1991). Plant Design and Economics for Chemical Engineers, 4th Edition. McGraw-Hill, New York, USA. pfi scotland. (1998). Baldovie Waste to Energy Plant, Dundee City Council. The Private Finance Unit, The Scottish Office Finance Group, Edinburgh, UK. rcep. (2000). Energy – The Changing Climate. 22nd Report. HMSO, London, UK. scottish executive and sepa. (2003). The National Waste Plan 2003: Scotland. Scottish Executive and the Scottish Environment Protection Agency, Scotland. shanks. (2007). The Intelligent Transfer Station. A Method for the Treatment of Municipal Solid Waste. Shanks, Milton Keynes, UK. stern, n. (2007). The Economics of Climate Change: The Stern Review. Cambridge University Press, Cambridge, UK. sustainable development. (2000). Waste Strategy 2000: England and Wales. Part 1. Crown, London, UK. sutherland, k. (2007). Economic Appraisal of Large Projects. In: Simons, S.J.R., (ed.). Concepts of Chemical Engineering 4 Chemists. The Royal Society of Chemistry, Cambridge, UK. thurgood, m. (1999). DERL Energy from Waste Facility Dundee Scotland. Case Study for IEA Bioenergy Task 23 – Energy from Thermal Conversion of MSW and RDF. AEA Technology, Harwell, UK. united nations. (2002). Future World Population Growth to be Concentrated in Urban Areas of World. United Nations Population Division, Department of Economic and Social Affairs, New York, USA. waste watch. (2007). Legislation Affecting Waste. See www.wasteonline.org for further details. Accessed 17 December 2007. wheeler, p.a., de rome, l. (2002). Waste Pre-Treatment: A Review. R&D Technical Report PI-344/TR, Environment Agency, Bristol, UK. wrap. (2006). The Sustainable Use of Resources for the Production of Aggregates in England. Waste and Resources Action Programme, Oxon, UK. yassin, l. (2008) ‘Appropriate scales and technologies for energy recovery by thermal processing of waste in the urban environment’. PhD Dissertation, University College London. yassin, l., lettieri, p., simons, s.j.r., germaná, a. (2009) ‘Techno-economic performance of energy-from-waste fluidized bed combustion and gasification processes in the UK context’. Chem. Eng. J., 146, 315–327.

8 Fluidized bed pyrolysis of waste polymer composites for oil and gas recovery W. K A M I N S K Y, University of Hamburg, Germany


Abstract: The feedstock recycling of composite polymer materials by pyrolysis is of increasing importance, as landfilling and combustion become more expensive and the acceptance of these methods is decreasing. Polymer composites are produced mainly from oil and have high potential as hydrocarbon sources. Fluidized bed pyrolysis has turned out to be particularly advantageous. By varying the process parameters, it is possible to obtain olefins, aliphatics or aromatics as main products. Monomers are produced in high yields from polymethylmethacrylate and polystyrene. In a fluidized bed process no toxic compounds such as chlorinated dioxins and furans are formed. Long-term runs of pyrolysis plants have to be carried out but today’s high crude oil price is a chance for building up pyrolysis plants for composite materials in an industrial scale. Key words: fluidized bed process, pyrolysis, feedstock recycling mixed plastics, shredder material, gas and oil.

8.1

Introduction

Polymeric composite materials are mainly made from crude oil products. The increasing crude oil price and the booming waste economy worldwide and especially in Europe have been described in the previous chapters. Most composite materials such as mixed plastics from household waste separating plants, composite films, filled polymers, cable scrap, special plastics from hospitals, and plastic collections from shredded cars are so highly contaminated that material recycling is very difficult and not economical. It is also expensive to combust these composites because of the presence of polyvinyl chloride (PVC) and other chlorinated compounds, metallic stabilizers and other heteroatoms containing polymers which can produce toxic components in the exhaust gases. Therefore the incineration of polymer composites is not always acceptable by regulations, rules and standards and requires new activities and techniques for the recycling of plastic wastes (Klein et al., 2004). Thus, very often thermal recycling such as pyrolysis and less favoured incineration are the only options open. Pyrolysis of a great variety of composites and different processes is described in Chapter 5. Other overviews on the pyrolysis of plastics and composite materials can be found in 192

Fluidized bed pyrolysis of waste polymer composites

193


Brandrup et al. (1995), Scheirs and Kaminsky (2006) and Serrano et al. (2001). The plastic fractions from municipal waste separation make up the largest amount of polymeric composite materials. The composition varies from country to country and how well the waste is separated by different processes from other materials such as glass, metals, papers and biomaterials. Another fraction is the combined plastic materials from hospitals containing syringes and other composites. Recycling this material is complicated by contamination with toxic drugs and microorganisms. The temperature of incineration or pyrolysis must be high enough to destroy the contamination. Large amounts of waste tyres have to be recycled. Tyres are a composite material of rubber, carbon black and fibres. After vulcanization, this material cannot be recycled by extrusion processes. Increasing amounts of nonmetallic composite materials from car shredding are available. All these fractions can be recycled by pyrolysis technologies. Other polymeric composite materials which were not studied in this paper are carpet wastes, mixed fibres and fibre reinforced plastics.

8.2

Pyrolysis

Pyrolysis is the thermal decomposition in the absence of air or at least in a deficit of oxygen that produces pyrolysis oils and gases for further treatment by standard petrochemical processes. The advantage of pyrolysis over combustion is that from 5 to 20 times less gas is produced. This means that considerable savings can be made in gas scrubbing. Furthermore, the pollutants are concentrated and bound in a mainly coke-containing residue. Moreover, some processes are able to regenerate top-quality chemical feedstocks. One difficulty encountered with pyrolysis is the fact that plasticsare very poor heat conductors, and that it sometimes takes considerable amounts of energy to crack the macromolecules (Kaminsky and Sinn, 1980; Malkow, 2004). The reactors involved in pyrolysing plastics and composite materials are melting pots, shaft kilns, autoclaves, tubular reactors, rotary kilns, carbonizing drums and fluidized beds (Buekens and Huang, 1998; Arena et al., 2007). In this chapter, investigations are described using a fluidized bed process for the feedstock recycling of composite materials by pyrolysis. The fluidized bed has a number of special advantages for the pyrolysis because it is characterized by an excellent heat and mass transfer as well as constant temperature throughout the reactor which results in: • • •

fairly uniform product spectra; absence of moving parts in the hot zone of the reactor; completely closed system, i.e. the reactor can easily be sealed.

194


Pyrolysis of waste materials in heated boilers, autoclaves, rotary kilns or screw conveyors is characterized by relatively long residence times in the reactors themselves, of 20 minutes and more compared with times of only seconds or few minutes in a fluidized bed process (Sinn et al., 1976; Williams et al., 2001; Mastellone and Arena, 2002). The fluidized bed is generated by a flow of an inert gas that is directed from below through a layer of fine grained material, e.g. sand or carbon black, at a flow rate that is sufficient to create a swirl in the bed (Kaminsky, 1985). At this stage, the fluidized bed behaves like a liquid. In Japan, fluidized bed pyrolysis plants are operated with air or oxygen feed (DOS, 1974). The partial oxidation generates a part of the necessary fission energy while on the other hand a part of the products is burnt. The product oils are partly oxidized and their energy content is some 10% below that of pure hydrocarbons.


8.3

Fluidized bed pyrolysis

In the years from 1973 up to 2007, different size laboratory fluidized bed (LFB) reactors were built up at the University of Hamburg, continuously working with plastics and composites throughput of 5 g/h (LFB 1), 500 g/h (LFB 2), 2 kg/h (LFB 3 + 4), 3 kg/h (LFB 5), a small pilot plant with 10–30 kg/h (technical fluidized bed, TFB 1) for mixed plastics, and a pilot plant for 100–200 kg/h of whole tyres (TFB 2). All reactors were heated up indirectly, the small laboratory sizes electrically from outside, and the pilot plants by heating tubes with the incineration of gas (for starting propane, later in a run by pyrolysis gas). A flow scheme of a laboratory plant is shown in Fig. 8.1 (Kastner and Kaminsky, 1995; Simon et al., 1996; Kaminsky et al., 2004). The fluidized bed with a diameter of 154 mm and a length of 770 mm consisting of fine quartz-sand with a particle size of 0.3–0.5 mm is heated indirectly from outside by electricity. The gas distributor is a steel plate with 108 tubes (3.2 × 0.75 mm2) which were moulded into hooks, thus ensuring that no sand could fall through the plate. Nitrogen was used as a fluidizing medium, but during the experiments it was slowly displaced by gaseous pyrolysis products. An auxiliary gas-stream was led through the feeding system to prevent hot gases and sand from entering the input system. The feeding system consists of two conveyors, the first for a constant feed rate and the second for a quick transport into the reactor. In case of composite materials with a high filler content an overflow vessel is used. The products left the reactor and passed a combined cooling and condensation unit. Fillers and fine sand were precipitated in a cyclone. The gas was cleaned up in an electrostatic precipitator. By a membrane compressor, a part of


195

Water PIR 5003

Storage vessel

TIR 5006

TIR 5005

TIR 5007

Cyclone

PIR 5002

Reactor

Ethanol

PIR 5004

TIR 5009

Electrostatic precipitator

TIR 5008

TIR 5004 TIR 5003

TIR PIR 5010 5005

TIR 5002 PIR PI 5001 5001 TIR 5001

FQI WG

Compressor

Overflow vessel

Fluidized gas meter

PI 5006 PIR TIR 5006 5011

TIR PIR 5007 5012

FQI ÜG

Flare Exhaust gas meter


Nitrogen

Vacuum pump Gas samples

8.1 Scheme of a fluidized bed process for feedstock recycling of plastics.

the gas is cycled and used as fluidizing gas for the bed reactor. The oil product is collected below the coolers. In all experiments, gases, liquids, water fraction as well as solid residue were obtained as products and analysed on different capillary gas chromatography (GC) columns and registered with thermal conductivity detectors (TCD) and flame ionization detectors (FID). The reactor sand used and the fillers in the overflow and cyclone were combusted in a furnace at 810 °C to constant weight. The loss in weight was balanced as carbon black. Elementary analysis of the feed materials was made with a Carlo Erba Strumentazione CE 1106 CHNS-O. Gases were analysed with GC-TCD (Chrompack CP 9001; Carboplot P7) and GC-FID (Chrompack CP 9002; Chrompack Al2O3/KCl Plot-capillary column). Correlation took place by the methane peak. After oil distillation, three fractions were obtained. Water and organic fractions were analysed quantitatively with GC-FID (HP 5890; Macherey & Nagel SE-52). Qualitative analysis of the organic fraction took place by GC-MS (mass spectrometry) (HP 5890; Macherey & Nagel SE-52; Fisons Instrument VG 70 SE). Water was determined by the Karl Fischer method


196


(Methrom Karl-Fischer-Titrierautomat E547). The third fraction contained high boiling (fluorine cut: bp > 295 °C) and inorganic components. Like the other solid products it was heated at 815 °C to a constant weight by oxidation to determine the organic part. All results of the analysis of the organic product fraction were combined together to form a total mass balance. Experiments with throughputs of 10–30 kg of mixed plastics and shredder materials were carried out in a small pilot plant (Fig. 8.2) (Kaminsky, 1980; Kaminsky and Hartmann, 2000). The heart of the plant is a fluidized bed reactor with an inside diameter of 450 mm and a height of 900 mm. This bed part is followed by a 1075 mm long free board zone in which no sand is fluidizing. The height of the sand bed (fluidized bed) is 650 mm. The sand has a size of 0.3–0.7 mm. The reactor is heated by four steel heating tubes which are using propane or the excess gas of the pyrolysis products (Fig. 8.3). For feeding, there are three possibilities of a screw conveyor, a tube for liquid feed from the side, and a lock with two flap valves for material from the top of the reactor. The feeding system consists of a screw conveyor nearer to the hopper which controls the amount of the feed and a fast moving screw conveyor directly at the reactor which brings the feed into the fluidized bed very fast to avoid the formation of glue and constipation. This screw conveyor is cooled by a water double jacket. The lock with the two flap valves is used for big pieces of plastics or for material with a high viscosity. The capacity of the pilot plant is between 10 and 30 kg feed/hour. The bottom of the fluidized bed has an incline of about 15° and carries the bent gas-inlet tubes. These tubes are movable vertically so that their distance from the bottom of the fluidized bed can be varied. With this arrangement, there is a variable settling zone for small metal pieces and other particles which cause in as impurities of the composite materials. Behind the fluidized bed reactor follows a cyclone for separation of solids. Then the product gases pass a washing cooler; in this cooler, xylene is cycled and used as cooling medium and solvent. It washes the cooler free from waxes and other high boiling products. The heat transfer happens by a tube heat exchanger. After this, the cooled product gases pass two packed columns in which also xylene is used as a quenching medium. This solvent is cooled down to −5 °C by a cryostat running with ethanol. Before the gas is compressed, it is cleaned up from fog by an electrostatic precipitator (electro-filter). The compression happens by five membrane compressors. Two of them transport the gas directly into the fluidized bed. The other three presses the gas into three steel gasometer. From this, a part of the gas is used for fluidizing the bed; it can be controlled and is mixed with the other gases. The right flow gas rate is the sum of both gases. If a high amount of flow gas rate is necessary, also three membrane compressors can directly pump the gas into the fluidized bed. The capacity of a compressor

Exhaust gas

Reactor

Overflow vessel

Cyclone

Gasometer

Water

Quench cooler

Pyrolysis oil

Electrostatic Scrubber precipitator

Sulphur gas Compressor

Cryostat

Scrubber

Cryostat

Oil

Steam

Oil

Water

Flare

Chimney

High Xylene Toluene Benzene boiling fraction fraction fraction fraction

Steam

Water

Distillation columns

8.2 Scheme of the pilot plant TFB of the Hamburg process for pyrolysis of 30 kg plastics/hour in a fluidized bed reactor.

Heat exchanger

Fluidizing gas

Pyrolysis gas

Propane

Compressed air

Silo

Lock


198



8.3 View inside the fluidized bed reactor from the bottom with four fire tubes and the hole for feeding.

is higher if the pressure is less. The excess gas of the pyrolysis can be burned in the fire tubes or in a flare on top of the building. In a separate room (safe against explosion) the distillation and quenching columns are installed. In the two distillation columns, four distillation boiling cuts can be obtained from the pyrolysis oil. In the first column with a diameter of 150 mm and a length of 10.5 m, the fraction with a boiling point of more than 180 °C is split off from the distillation residue. In the second column with a diameter of 80 mm and a height of 8 m are obtained cuts of 80, 110 and 140 °C (xylene). The xylene is used as quenching oil also in the coolers. The plant is controlled by a process computer of ABB-Hartmann and Braun and equipped with numerous data-collecting instruments. Surveillance is carried out by continuous analysis of the room air as well as by explosion-limit controls. The pyrolysis gas is analysed automatically by a gas chromatograph. All data obtained are registered to enable calculation of energy and mass balances. Some basic components are continuously monitored by infrared spectroscopy, i.e. ethylene in the pyrolysis gas, sulphur dioxide and oxygen in the exhaust gas. The heating of the fluidized bed by fire tubes was the demand to up-scale the plant. The Ebenhausen plant and the Grimma plant used for heating the fire tubes, too. A scheme of the fire tube is shown in Fig. 8.4. The fire tube consists of two tubes. The outer tube is closed in front, while the inner tube is open. The flame is started at the burner lead and the exhaust gases pass between inner and outer tube. They are collected outside the reactor and go through a heat exchanger in which the incoming gas for fluidizing


1

2

3

4

5

6

199

7 8

9


8.4 Fire tube for heating of the fluidized bed: 1 closed outer tube, 2 open inner tube, 3 burner head, 4 reactor wall, 5 outlet for exhaust gas, 6 inlet for air, 7 inlet for propane or gas, 8 spark plug, 9 burner head.

is heated up. There is no mixing of exhaust and product gas. The whole reactor is shown in Fig. 8.5 isolated by rock wool. In one run, between 50 and 150 litres of oil and 20 to 290 kg gas are obtained. The experiments in the University of Hamburg were the start-ups for the building of three pilot plants by companies. The first was built in 1982 in Ebenhausen/Bavaria by DRP/ABB using a fluidized bed reactor with a diameter of 1800 mm and a capacity of 800 kg/h of mixed plastics and whole tyres. The second plant was built in Grimma/GDR in 1984 for whole tyres with a capacity of 5000 t/year. Also BP had carried out first experiments in 1992 at the University of Hamburg and then built a pilot plant in Grangemouth/Scotland with a capacity of about 5000 t/year to obtain waxy products from mixed plastics, mainly polyolefins. These waxy products should be used as feedstock for naphtha-crackers. All pilot plants run only for some years, mainly because economic running by decreasing crude oil price was not possible. The Hamburg process can be varied by some simple process parameters such as pyrolysis temperature, kind of fluidizing gas (nitrogen, steam, cycled pyrolysis gas) and residence time to produce different products from plastic waste (Table 8.1). If the pyrolysis temperature is low (400–550 °C) and nitrogen or steam as fluidizing gas are used, mainly waxy products are obtained. By higher pyrolysis temperatures, more oil and gas are produced. The main pyrolysis products are aromatics and soot by pyrolysis temperatures of 700–800 °C and allow the use of cycled pyrolysis gas as fluidizing gas.


200


8.5 Isolated fluidized bed reactor with the outside fire tubes (centre), the screw conveyor for feeding (right) and the drump for inorganic fillers from the overflow (left) Table 8.1 Variation of the Hamburg pyrolysis process feeding polyolefins by different temperatures and fluidization gases Temperature (°C)

Fluidizing gas

Products

400–550 550–700

Nitrogen, steam, cycled pyrolysis gas Steam

700–800

Nitrogen, steam

700–800

Cycled pyrolysis gas

70–90% waxy products/oil, 1–9% gas 60–70% oils and aliphatics, 30–40% gas 70–80% gas, mainly ethylene and propene, 20–30% oil 30–50% aromatics, 30–55% gas, 1–10% soot

8.4

Polymer composite materials

For the oil and gas recovery in a fluidized bed were used six different fractions (A–F) of composite materials. Three plastic fractions A–C were separated from municipal wastes. Mixed plastic wastes A and B were obtained



201

from the Duales System Deutschland (DSD), a company for collecting plastic wastes from household separation in yellow sacks. Both fractions differ mainly in the content of polystyrene (Table 8.2). Fraction A contains 14 wt% while it is 25 wt% in fraction B. The PVC content in fraction B is less 1.2 wt% instead of 3.8 wt%, also the polyester/paper content is less, as it is collected in parts of large office buildings of companies. Both fractions are contaminated by food residues, sand, metal films from Tetra Pak and other inorganic fillers. Another fraction C of mixed plastics was collected in households in the city of Hamburg. It contained polyolefins (57 wt%), polystyrene (19 wt%), PVC (13.7 wt%), other plastics (4.8 wt%), moisture (0.2 wt%), and inorganic residues (13.7 wt%). The chemical elementary analysis is shown in Table 8.3. This waste fraction is characterized by a high PVC content of 13.7 wt% or 7.8 wt% of chlorine, because of a lot of PVC bottles and PVC tiles from floors. Composite material (fraction D), used for the pyrolysis, was contaminated syringes from the medical sector. These syringes were shredded into pieces of 1 cm length. The analysis showed that this special composite material consists of low density (LD) polyethylene (40 wt%), high density (HD) polyethylene (10 wt%), polypropylene (41.3 wt%), rubber (7.3 wt%) and cellulose (1.4 wt%). The rubber was concentrated in the pistons of the Table 8.2 A and B different compositions of the mixed plastic wastes used in wt%, collected by DSD system Fraction (feed)

A

B

Polyolefins Polystyrene PVC Polyester/paper Other plastics Water Fillers, metals

65 14 3.8 7.2 2.0 4.0 4.0

65 25 1.2 1.5 1.3 4.1 1.9

Table 8.3 Elementary analysis of mixed plastics collected in households (fraction C) Carbon Hydrogen Chlorine Nitrogen Other elements

79.1 11 7.8 0.5 1.6

wt% wt% wt% wt% wt%

202



syringes. It was found that pyrolysis in a fluidized bed at 720 °C was a good way to kill all bacteria and virus contaminating the syringes. Tyre pieces (fraction E) are cut car tyres with a size of 1–1.5 kg each. They contain mainly rubber (55.7 wt%), carbon black (33 wt%), cellulose and polyamide fibres (4.5 wt%), inorganic fillers (5.2 wt%) such as silica, zinc oxide, sulphur as well as steel cord (1.6 wt%). Another composite material was received from a car shredding plant (Trapp, Frankfurt) (fraction F). It was the non-iron fraction containing rubber (34.5 wt%), wood and paper (3 wt%), fibres (12.5 wt%), rubber– metal composites (3 wt%), metals (9.5 wt%), rubber–fibre–plastic composites (31.5 wt%), plastics (3 wt%), and stone and glasses (3 wt%). Main parts are rubber and fibre–plastic composites. The elementary analysis of fraction F shows a carbon content of 72.04%, hydrogen of 6.76%, inorganics of 14.15 wt%, sulphur of 2.05 wt%, and chlorine of 1.40 wt%. The sulphur is component of the rubber, and the chlorine of the plastic part. The fibre–plastic composites are a mixture of cellulose fibres, polyester fibres, polyolefin films, polyamides and other plastics.

8.5

Gas and oil recovery

8.5.1 From mixed plastics The pyrolysis runs were carried out with the different mixed plastics in the laboratory plant and in the small pilot plant by different temperatures (Kaminsky, 1995). Table 8.4 shows the results of the products formed in the fluidized bed process. The pyrolysis gas formed reached values between 35 (feed B) and 41% (feed A). In the pilot plant, less gas is formed than in the laboratory fluidized bed process because the residence time of the gases in the fluidized bed is three times longer (6 seconds instead of 2 seconds in LFB). Main components are methane, ethylene, ethane and propene. Carbon monoxide and carbon dioxide are formed from paper and polyethylene terephthalate (PET) parts in the mixed plastics. The main fraction is oil with values of up to 51 wt% (feed B). The highest value is found by the run in the pilot plant. In the secondary reaction favoured by the longer residence time, more aromatics such as toluene and styrene are formed. The oil contains mainly aromatics by the used pyrolysis temperature of 700 °C. As of the high polystyrene content, styrene is a major component but also a lot of benzene, toluene, ethylbenzene, naphthalene are formed. The difference in the styrene content in the oil is not as big as the difference in the polystyrene content in the two


203


Table 8.4 Mass balance of the pyrolysis of mixed plastics (different fractions A and B see Table 8.3) in a fluidized bed with different feedstocks and plants using pyrolysis gas for fluidization Plant Throughput Feed Temperature (°C)

LFB5 1185 g A 685

LFB5 1230 g A 716

LFB5 2800 g B 738

TFB1 212 kg B 730

Gases (sum) Hydrogen Carbon monoxide Carbon dioxide Methane Ethylene Ethane Propene Propane Butane 1-Butene Butadiene trans-2-Butene cis-2-Butene Other gases Oils (sum) Paraffins C5–C6 Paraffins C5–C6 Paraffins C7–C9 Benzene Toluene Xylene Ethylbenzene Styrene C3-Benzene Methylstyrene C4-Benzene Indene Methylindene Naphthalene Methylnaphthalene Diphenyl Fluorene Phenanthrene/anthracene Other aromatics Oxygen compounds Nitrogen compounds Other compounds Distillation residue Soot/fillers

41 0.4 4.6 2.0 10.9 8.6 4.0 6.3 0.5 0.08 2.1 0.8 0.2 0.2 0.1 48 5.7 5.7 2.3 9.5 6.7 0.6 0.9 9.0 0.6 1.4 0.1 0.6 0.4 0.8 0.3 0.1 0.04 0.01 0.7 0.1 0.02 8.1 5.8 5.2

43 0.6 6.0 1.7 16.2 10.1 3.3 3.2 0.2 0.05 0.7 0.7 0.1 0.1 0.1 45 2.2 2.2 0.5 14. 4.8 0.5 0.9 6.8 0.5 1.5 0.1 2.1 0.5 4.2 0.5 0.2 0.05 0.04 0.6 0.01 0.01 5.0 6.3 5.7

38 0.7 1.3 0.6 20.5 10.3 2.2 1.0 0.1 0.19 0.1 0.5 0.1 0.1 0.3 50 0.7 0.7 0.2 17.4 3.9 0.2 0.2 8.7 0.5 0.2 0.3 2.5 0.3 7.2 0.1 0.1 0.18 0.63 0.5 0.01 0.01 6.2 6.5 5.5

35 0.4 1.3 0.4 11.9 8.9 3.9 5.0 0.6 0.03 1.6 0.7 0.2 0.1 0.1 51.8 3.1 3.1 0.5 9.1 7.6 0.8 2.5 10.8 1.2 1.6 1.1 1.2 0.7 2.3 1.0 0.4 0.1 0.2 2.5 0.01 0.01 5.1 11.0 2.2


204


mixed plastic fractions A and B. A lot of the PS is degradated into other aromatics. The HCl coming out from TFB was quantitatively absorbed by calcium oxide which was added in 5% to the feed A and 2% to feed B. The formed CaCl2 was separated in the cyclone after the fluidized bed reactor, and is listed in the balance by soot and fillers. Small amounts of carbon black are formed (0.2–1.0 wt%). No chlorine was found in the gas fraction. The oil contains up to 15 ppm chloroorganic compounds, mainly chlorobenzene. To receive a maximum on olefins and butadiene from mixed plastics recycling, it is necessary to have a short residence time of the product gases in the fluidized bed zone to have no secondary reactions. The pyrolysis gas should not be circulated and used as fluidizing gas. It is better to use steam or nitrogen as fluidizing gas (Simon et al., 1996). Under these conditions, more than 30 wt% of ethylene can be obtained. A mixed plastic fraction C, rich in PVC (13.7 wt%), was pyrolysed in the pilot plant at three different temperatures (680, 735, 790 °C). The pyrolysis gas was used as fluidizing gas. 10 wt% of dolomite (CaCO3/MgCO3) was added to the feed to absorb the HCl coming out from the pyrolysis of PVC. It was the aim of the experiments to look at toxic components formed by the pyrolysis of such a composite material. The product composition is given in Table 8.5. Compared with Table 8.4, less gas and oil is formed. The relations of the other products are similar to those of the pyrolysis of a less PVCcontaining plastic fraction. Main products in the gas are again methane, ethylene, ethane, propene. The methane content increases with the pyrolysis temperature from 4.9 wt% (680 °C) up to 17.5 wt% (790 °C). On the other hand, the propene content decreases as well as the amount of other aliphatics. The oil contains mainly aromatics such as benzene, toluene, styrene and high boiling tars. The benzene content increases with the pyrolysis temperature and reaches values of more than 13 wt%. A lot of carbon black is formed, mainly from PVC. CO and CO2 are products from paper pyrolysis. The HCl is bonded to calcium carbonate nearly quantitatively and is calculated from the feedstock. There was a great interest in looking at toxic chlorine components coming out of the pyrolysis. The content of chlorinated dibenzodioxines and furans were measured by GC-MS in a specialized institute outside the University of Hamburg. Table 8.6 summarizes the results for the high boiling pyrolysis oil. In the low boiling oil fraction and in the gas fraction, no chlorinated dioxins and furans could be determined. The detection limit was 1 μg/kg. The toxic chlorinated compounds are concentrated in the high boiling fraction. It was found that there were no traces of 2,3,7,8-tetra chlorodibenzo-


205


Table 8.5 Pyrolysis products (in wt%) from mixed plastics products with 13.8 wt% of PVC at different temperatures (fraction C) Temperature

680 °C

735 °C

790 °C

Hydrogen Methane Ethylene Ethane Propene 1,3-Butadiene Alicycles Other aliphatics < C6 Other aliphatics > C7 Benzene Toluene Styrene Other alkylbenzenes Indane Indene, methylindenes Naphthalene Alkylnaphthalenes Biphenyl, alkylbiphenyl Anthracen, Phenanthren Other aromatic compounds Tars Carbon black Carbon monoxide Carbon dioxide Water HCl (calculated) Other components/inorganic ash Mass of pyrolysed wasted

0.19 4.89 5.37 2.96 4.05 0.79 3.27 4.79 6.55 3.27 1.81 5.27 1.04 0.17 0.71 0.38 0.18 0.23 – 0.12 17.40 4.15 2.45 5.97 2.5 7.98 13.41 165.6 kg

0.31 10.56 9.64 3.35 3.64 0.94 2.79 1.88 0.01 7.93 3.24 6.04 2.96 0.22 1.27 1.37 0.67 0.28 0.46 0.82 10.35 5.58 3.58 4.69 4.0 7.98 5.36 135.15 kg

0.70 17.49 9.78 2.99 1.32 0.44 0.55 0.33 0.00 13.78 5.55 3.10 1.02 0.14 1.34 3.83 0.89 0.57 0.65 2.24 6.15 6.43 2.37 1.27 2.64 7.98 6.53 130 kg

dioxine (TCDD) which is extremely toxic outside of the detection limit. Only small amounts of chlorinated furans were measured. The measured values are much less than those from other pyrolysis plants. This is why the fluidized bed plant has no mechanically moved parts which are difficult to close totally. During the pyrolysis process, a lot of hydrogen is formed. The hydrogen has a high reduction potential at a pyrolysis temperature of 700 °C and takes the chlorine atoms from the chlorinated organic compounds to form HCl and pure hydrocarbons. The pyrolysis of waste syringes (feed D) is different from the pyrolysis of mixed plastics. This composite material contains a lot of rubber. For comparison, Table 8.7 compares the results of pyrolysis investigations in a fluidized bed process with the results using tyre pieces – also a composite material in the pilot plant at 720 and 750 °C.

206



Table 8.6 PCDD PCDF and PCB concentration in high boiling pyrolysis oils of the mixed plastics pyrolysis, fraction C (values in µg/kg) Pyrolysis temperature

680 °C

735 °C

790 °C

Sum TCDD 2,3,7,8 TCDD Sum PeCDD 1,2,3,7,8 PeCDD Sum HxCDD 1,2,3,6,7,8 HxCDD 1,2,3,7,8,9 HxCDD 1,2,3,4,7,8 HxCDD Sum HpCDD OCDD Sum TCDF 2,3,7,8 TCDF Sum PeCDF 2,3,4,7,8 PeCDF Sum HxCDF 1,2,3,6,7,8 HxCDF Sum HpCDF PCB according DIN 51527

n.d. n.d. n.d. n.d.

n.d. n.d. n.d. n.d.

n.d. n.d. n.d. n.d.

n.d. n.d. n.d. 2.1 1.7 16.9 0.38 13.7 0.39 3.2 0.3 5.3 1500

n.d. n.d. n.d. n.d. n.d. 4.5 n.d. 1.5 n.d. n.d. n.d. n.d. 98% LDPE and PP at >95% (Brandrup et al., 1996). The ability for the electrostatic separators to cope more effectively with waste materials is far better than the hydro-cyclone or flotation methods as the system does not rely on the density characteristics of the waste to facilitate efficient separation. The tribo-electric characteristics of the waste material provide a unique property for the system to facilitate separation. The ability for the system to rely on the electrostatic charging of the waste materials implies that the system would be able to cope adequately with composite waste materials. Even materials with similar density characteristics would result in a different electrostatic charge in the materials enabling separation. Electrostatic separation systems are being used in the USA and

Mechanical methods for recycling waste composites

239

in the EU for scrap recycling of automotive waste (Douglas et al., 1976; Harper, 2002).


Near infrared and optical sorting Near infrared (NIR) sorting of polymer waste uses an infrared spectrometer linked to a powerful computer which is capable of identifying 1000 spectra of polymers per second. The technique was developed by the Fraunhofer Institute for Chemical Technology in Pfinztal. The spectrometer measures the electromagnetic spectrum of a polymer and a sorter will then direct the article into a sorting bin. The wavelength range of the NIR is 700–2500 nm, the NIR has a reduced absorbance in bulky items, which allows for C—H, O—H, N—H and C—O groups to be characterised in polymers. The system has a detector head which is operated either automatically or manually. Light of 5 mm diameter is focused onto the sample material, using a quartz condensing lens. A second quartz condensing lens is set at 90° and collects the reflected light. A second detector arrangement is also possible, which has a larger observation plane. The unit has seven light sources which has a measuring range of 100 mm in diameter. This system allows for simultaneous measurements to be gained in different viewing angles. The data is measured and the product is separated using a combination of gates and conveyors (Brandrup et al., 1996; Bledzki et al., 1998). This technique is widely used where there are multiple plastic types that are of similar characteristics and colour. Studies have been completed in Germany with waste electrical components under the guise of the Waste Electrical and Electronic Equipment (WEEE) Directive. WEEE has been identified as a priority waste stream by the European Union and receives increasing attention as it makes up the fastest growing waste stream, with a growth rate almost three times higher than municipal solid waste. The study correctly identified that the polymer content was 34.6% of all of the material types present. 87% of the waste polymers were accurately separated into polymer types, over half of the waste was black in colour and only 6.8% had identification (Dimitrakakis et al., 2008). The separated polymers are shown in the graph in Fig. 9.15. Optical-based sorting systems can sort according to lightness, colour and shape of the waste. Each granule of waste approx 10–150 mm in size is assessed by the sorting criteria preset by the operator. A charge-coupled device (CCD) camera is used to assess and differentiate the red, green and blue as well as the hue, saturation and intensity of the particle (Tachwali et al., 2007). Pneumatic air jets are used to separate or remove the granule from the bulk of the material. As the system is more specific in the method the granules are assessed, the separation can be very efficient, where

240

Management, recycling and reuse of waste composites Other, 5.61 PVC, 4.66 PP, 28.83

ABS, 36.75

PS, 19.16 PC, 4.99


9.15 Waste polymers identified in a study using NIR.

disparate materials are present. These systems can be used in conjunction with the NIR technique (Charles and Leander, 1995). This system relies on the optical properties of the waste to separate the materials, it does not use any of the material properties. So in many respects this system appears to be unsuitable when waste composite materials are considered for separation. However, as the system uses the optical image of the waste then it may be possible to cope with waste materials that are discreet or known. If for example all of the composite materials being sorted were white and all of the other materials were different colours, then the system would be able to identify and separate the waste quite well. This system would be suitable for single source or industrial waste where the materials are generally known but handled together.

9.3.7 Conversion from composite waste particulate to a reusable material Once the waste material has been collected, size reduced, cleaned and sorted it can then be reused as a valuable resource. Prior to its being used or sold it is often re-formed to give it a consistent and aesthetically pleasing appearance. The reformation is quite often unnecessary and can add processing costs to the final material selling price. However, the reformation can be an opportunity to blend the material with a virgin compound or to add additives to improve the performance of the material. If the waste is in film form it will have a low bulk density and requires densification prior to use. The material with a higher density can then be used in extrusion to


241

remake articles or to manufacture pellets for use in injection moulding or other manufacturing processes.


Densification (agglomeration) Densification is the process of taking a low bulk density material, usually in film or flake form, typically from 100 to 150 kg/m3 and increasing the density to a more usable or transportable material. There are different methods for densification of waste film. However, they all involve the use of frictional heat, usually under the melting temperature of the polymer. Pelletisers mix and compress the waste material through a die. The compression exerted on the polymer creates frictional heat which enables the polymer to join together. The polymer exits the die where it is cooled and a rotating knife is used to cut the pellet to length (Shrivastava, 2003). In a pot type of agglomerator, the waste is placed into a pot where knives are rotated at very high speed. The high speed knives cause the polymer to heat by means of friction. This heat causes the polymer film to shrink and stick together into agglomerates; these are cooled and cut into pellets. Another method is the disc compactor: in this process the film is softened between a fixed and a rotating disc. The polymer exits the disc at the periphery of the disc where it is chopped into pellets (Scheirs, 1998; Goodship, 2007). Another type of densification method is based on the combination of the pot agglomeration technique and extrusion technology and is ideally used for film waste materials. The system is designed and manufactured by Erema of Austria; an image of the unit is seen in Fig. 9.16. An Erema unit consists of a cutter and compactor, consisting of a large vertical drum with rotating blades sited at the bottom. The material is gravity fed from the

9.16 Photograph of an Erema (image courtesy of Erema UK).


242


top and circulates inside. While inside the drum the waste is shredded by the blades and heated by mechanical friction; this causes the film waste to become agglomerated. The material is then dropped out of the bottom of the drum into the extruder. The friction heat provided in the earlier stage reduces the length of the extruder screw that is required to consolidate the material. This preheating of the polymer by the frictional heat reduces the heat input required by the extruder. A melt filter is used at the end of the extruder to remove solid contaminates (Evans et al., 2007). A similar machine to the Erema is offered by Repro Machine Industries based in Japan and is used with expanded polystyrene foam. The FT type of machine initially shreds the foam and is melted using electrical heaters before being cooled and granulated for extrusion. In the FS machine, however, the molten plastic is placed directly into an extruder, which is located directly beneath the melting chamber. The throughput of the system is approx 90 kg/h. The Erema machine is an ideal solution when industrial wastes composites are considered. As the wastes are typically single source they could be re-compounded and reprocessed without the additional steps required to collect, store and separate which would incur costs. The re-compounded material could be reused in the process or sold as a recycled grade.

Extrusion compounding Extrusion processing technology is typically carried out on solid particulate materials. It involves the melting of a solid polymer into a viscose medium, forcing it through a die and forming it into a net shape. Additives including colourant, antioxidants, fillers, fibres and stabilisers can be added to the molten polymer to achieve a desirable net performance. Numerous executions of extrusion are available including single screw and twin screw (contra and co-rotating). The performance of the extruder depends on the geometry of the screw and of the operating conditions. The process is a linear manufacturing technique that produces components that are traditionally two dimensional. Alternatively extrusion is often used to produce regular sized granules that can be sold and reused by a processor. Figure 9.17 shows a simplified diagram of a single screw extruder, where the barrel, screw, heaters and material hopper is visible. The screw and barrel are generally hardened to prevent wear and tear by the polymers being extruded especially where fillers and fibres are used. External heaters are provisioned to enable the barrel to heat and melt the polymer. Blowers are used on the barrel as excessive shear in the unit may cause excessive heating of the polymer, which may cause the polymer to degrade.

Mechanical methods for recycling waste composites Barrel heater Screw

243

Hopper Barrel

Feed throat

Die Drive Blower


9.17 A simple schematic of an extruder.

An important element of the extruder is the screw. Most of the screw has flights, this length is compared to the diameter of the screw, which is known as the length to diameter ratio (L/D ratio); typical L/D ratios are 24, 30 and 36. Screws with high L/D ratios have good mixing capabilities but use a high amount of energy (Chung, 2000). The lower L/D ratio extruder screws would result in the composite having a shorter thermal history in the extruder barrel. This may be desirable as this would result in less thermal degradation of the polymer and less mechanical damage of any fillers or fibres (Wang, 2000). The screw has a number of distinct sections: these are the feed, compression and metering sections. The feed section is the initial part of the screw where the polymer enters. This part of the screw has a constant root diameter. The flights of the screw are generally large, allowing the solid polymer to enter the extruder and fill the available volume. The compression section has an increasing root diameter. In this section the melting of the polymer occurs. The melt is compressed and any air or volatiles are forced out of the viscose material back to the entry point of the polymer. Different polymers require dissimilar compression zone types. Some polymers, PA for example, melt quickly so therefore would have short compression zones. Other polymers which may have long melting ranges such as PE have long compression zones. The compression of the polymer in this section causes a great deal of shear causing heating and melting of the polymer. With some polymer types too much compression may lead to excessive shear heating which may result in polymers burning or degrading. The metering zone is the final section of the screw the polymer passes through. This section has constant short flight depths and a large root diameter of the screw. By the time the material reaches this section the polymer is normally fully melted; however, the short flight depths ensure a


244


high melt shear which ensures all solids are now fully melted and compositionally homogeneous before it enters the die. Traditionally most extruders will have ‘general purpose’ screws installed. This design copes well with most material types and eliminates the need of multiple machines or different screw types (Xanthos and Todd, 2004). A vent is quite often sited towards the end of the extruder to enable volatiles and moisture trapped in the waste material to escape, which may otherwise have a detrimental effect on the polymer aesthetics and the physical properties. A melt filter may be used prior to the material entering the die to remove solid contaminates from the recycled polymer prior to its being formed into an article. The molten material would then pass through the die and be cooled to form the desired component. Typical products made using this technique are decking, profiles and tubes. With waste materials they may be formed directly into components or they may be made into pellets which could then be used in an injection moulding process. Single screw extruders are probably the most common type of extruder; they are used in the manufacture of profile lengths or of compounding polymers where fillers or additives are introduced into the polymer. Twin screw extruders are used with polymers which are heat sensitive. The screws are linked and intermeshed, which enables the polymer to move from screw to screw with a high degree of mixing without excessive shear, which may otherwise cause degradation or burning of the polymer. The material in the screw is positively moved forward in the barrel which causes desirable high pressures at the entry of the die. There are two different types of twin screw extruder, co-rotating and contra-rotating. Co-rotating extruders have the screws rotating in the same direction. The material passes from one screw to the other taking a path on the outer wall of the screws and barrel. The path the material takes enables a high level of contact with the barrel wall. This contact constantly wipes material off the barrel wall, which prevents material stagnation and potential degradation. Mixing in these types of extruder is good and is therefore used where compounding of materials or additives is required. The contra-rotation extruder has the screws rotating towards each other into the centre of the extruder. The material is therefore constantly pushed towards the intersection of where the screws meet. At this intersection, the shear of the material is quite high but elsewhere the shear is very low. This design of extruder also has a version where the screws are conical. This design of extruder has the lowest shear imparted to the polymer when compared to single screw or co-rotating (Strong, 2006). For this reason this type of extruder is used with recycled polymers especially with wood plastic composites (WPCs). In general, co-rotating extruders are very good at mixing, removing volatiles and providing high output.

245

However, parallel and conical contra-rotating extruders are better at generating pressure for pumping material through complex profiles. Wood fibre is used with waste polymers to produce WPCs. They have been developed to be much improved in the last 30 years with higher wood content, improved processing and interfacing properties. Since the 1990s the material has been used in high quantities in the USA competing against PVC building materials (Prichard, 2004). WPCs are a mixture of wood and resins processed in a typical plastics processing method such as extrusion. The composite consists of 30–80% wood content, coupling agents, weatherable pigments and about 20% of all WPC products are foamed. The composite material is handled like any real wood product and is machined with standard equipment accordingly. The mechanical properties of the WPCs are improved as the wood fibre acts as a reinforcing agent. The mechanical properties of the WPCs are difficult to compare to real wood as the moisture content of the wood can have a large effect as can the composition and processes used to make the WPCs. The applications for WPCs are typically of the outdoor variety such as decking, fencing, sidings, railings and window parts being common. Automotive applications are the next largest consumer, with the third being construction (Knights, 1996; Sherman, 2004). Shear generated in the extrusion process can lead to a 40% reduction in fibre lengths; minimisation of shear is beneficial to maintain fibres and resulting properties of the materials (Nygard et al., 2008). Figure 9.18 is a graph of results from a study completed by Lee (1992), where polypropylene with 30% weight of glass fibre with an initial fibre length of 9 mm has been processed through an extruder. The x-axis shows where the fibre has progressed further along the screw, the y-axis shows the average recorded fibre length; the fibre has been processed in the extruder three times. The 3.0 2.5 Fibre length



2.0 1.5

1st pass

1.0

2nd pass 3rd pass

0.5 0

8

10

12

14

16

18

20

Channel number in screw

9.18 Shortening of glass fibre as a result of extrusion processing.

246


study shows that during the first extrusion pass, as the fibre has travelled along the extruder it has reduced in size. Subsequent passes in the extruder show further breakage of the fibre length. What is not considered is the history of the recyclate, the number of times it has been processed or the process the polymer may have encountered prior to passing through the extruder (Lee, 1992). The extrusion process can be used to either manufacture articles or to manufacture pellets that can be used in other thermoplastic processing techniques. The manufacture of pellets presents the processor with an opportunity as the properties of the waste material can be improved by the addition of fillers, fibres, additives, stabilisers and processing aids. The waste material can be extruded into pellets to have significantly enhanced properties.


9.4

Conclusions

There are many different types of plastic composite waste materials and huge legislative and consumer drives exist to recover these materials where possible. The number of plastic types and the fillers used in them together with the numerous additive or stabiliser types make recycling these materials a potentially complex model. Plastic composites are coarsely sorted from other waste products by visual identification. This is carried out by the consumer, but kerbside collection is variable as to what products are allowable by the local councils in the UK. Good sorting practices of waste materials can reduce the level of intervention required at later stages and improve the purity of the final product. Second stage sorting of the waste is carried out either by manual intervention or by use of automated equipment. Manual sorting is more accurate but can be costly as people are employed to complete the sorting process; automated sorting systems are more cost effective to run but are generally not as accurate. Labour costs have a direct impact on the cost of recycled materials: manual sorting is widely used in countries where labour cost is low. Alternatively developed nations are sending waste materials directly to the low labour cost countries such as China. These nations then recover the materials using manual labour; the recovered goods are remade into products back into the marketplace. The waste is granulated into small particles, making it more manageable as it takes less volume, the process also frees up waste which may be assembled or joined locked together. However, the process of granulation mechanically damages fibres in composite materials. The level of damage imparted to fibres or fillers from the original condition is imprecise. It is



247

dependent on factors such as original fibre length, the process used to manufacture the part, damage which may have been experienced during the product life, and finally the damage imparted to the fibre as a result of the waste recycling process. Limited studies have shown that a 37% reduction in fibre length is to be expected, which will reduce the physical properties of the material, when compared to the virgin grade. More research is required into the effects of granulation. The study by Lee (1992) demonstrated the mechanical damage imparted onto glass fibres based materials by multiple extrusion processes. Similar studies are needed and will become ever more important as recycling of waste composite materials increases. The studies can ascertain the effect of fibres that have been processed and re-granulated through a number of cycles. This data would provide valuable information on the effect of the re-granulation and aid optimisation of the process equipment and parameters. It would also identify how many cycles can be processed before fillers in the composite are rendered ineffective. There are a number of different separation techniques available for granulated waste. The selection of these process or processes is based on the user of the technologies and if they are keen to use sophisticated technologies, such as NIR or simpler systems such as flotation. Throughput of waste and plant scale must be considered, as ever-increasing volumes of waste are being recovered. This must be completed cost effectively to make the recovery of such materials desirable. The desired purity level of the final waste product may well dictate the technology being considered. If a lower quality material is desired then it may be disadvantageous to invest in a highly efficient separation technology which may have limitations. Multistage processing is often used as it can help manage high throughputs and achieve desired quality of the end material. Simple techniques such as flotation or hydro-cyclone separation are excellent at segregating wastes with different densities. These separation technologies are ideal where disparate materials are present. Often a number of these technologies are used together or in sequence to refine the materials. However if different materials with similar densities are processed using these techniques then the materials will not be separated as they characteristically perform similarly. Chemical modification to the flotation medium would ensure a higher level of sophisticated separation; however, limitations in material identification would still exist. Investment, operational cost and plant size would need to be considered before such decisions are made. Flotation tanks are typically large and multiple flotation tank separation would take up a considerable amount of plant size. Operational flexibility would also be an issue as more materials are being considered for recycling and the facility would need to be able to cope with these changes in time and legislation


248


in order to survive. For these reasons separation processes using material density are not suitable techniques for composite waste materials. Tribo-electric separation is an efficient method for separating materials as the material density is not considered by the process. Large-scale studies have resulted in high purities of materials when processed in this technique. The technique can also manage well with fillers in the waste as these affect the tribo-electric properties of the material being considered. NIR uses the spectra emitted by different materials to enable a refined identification of waste materials, at very high speeds. Studies with WEEE waste resulted in 94% of comingled waste to be correctly identified by polymer type. This technique is being used in an industrial scale as it is highly efficient at identifying waste materials and separating them. The property used in the identification of the waste is the spectrum. The system can cope with composite materials very well as each material type will give off unique spectrum. Optical separation is another technique used to identify and separate waste materials. This technique is ideally used in conjunction with NIR or tribo-electric separation as it can be used to further refine materials based on optical identification. Waste materials may provide similar spectra or tribo-electric properties if they are different colours. The optical separation would ensure that similar materials with different colours are separated further. Optical-based separation may also be an ideal solution where optical differences are present in materials from limited or single waste sources. For example a black and a white composite could easily be separated from each other, without the need for any of the other techniques discussed. Tribo-electric, NIR and optical sorting are therefore considered more accurate at identification and separating materials. Each method looks for a unique property of the waste material where the property is sufficiently different to ensure accurate identification. Therefore these systems or a combination of these systems would be more suitable for use with composite wastes. Once the composite materials are sorted then reformation is required. This takes place by densification or extrusion. Densification is a low cost but high output process which uses frictional heat in a number of different techniques to allow the material to melt together and achieve a material pellet. Quite often this process is used when film materials are being processed as they have very low bulk densities. The Erema process is an ideal solution as it allows waste materials to be agglomerated and extruded in one machine. Depending on the application of the waste, the pellet can be used directly from the densification process. This process is commonly used especially where discrete or single source waste materials are present such as industrial waste. The reforming and



249

extrusion of these materials are desirable as quite often minimal sorting of the waste is required, which maintains a low operational cost. Granular or densified materials can then be reformed into an article or a pellet using extrusion equipment. Making an article is the simplest and most cost-effective solution as it reduces processing steps and associated costs. This is quite often the case when WPCs are manufactured as they are generally made from waste polymer and wood to manufacture decking, railings and fencing. Manufacture of pellets presents an opportunity for the waste composite material to be used in further processes, such as compression or injection moulding. The extrusion of the waste into a pellet allows the operator to re-formulate or improve materials by the addition of fillers, additives or stabilisers. The selection of the extrusion equipment is important as single screw extruders can excessively shear the composite material, damaging fibres rendering them non-functional. Co-rotating screw extruders have low shear effects on materials, but contra-rotating extruders have the lowest shear effect on composite materials. Studies have shown that mechanical damage will occur to fibres as they pass through the extruder screw. Subsequent processing of fillers in the extrusion screw can render a fibre one-twentieth of its original size. There are numerous process steps and methods for the mechanical recycling of waste composites, which are effective at recovering valuable materials. Quite often these methods are used in conjunction with each other to optimise throughputs and product quality. The characteristics of the final product will vary on the original material, source, type, processing stages and recycling methods; however a high quality reusable end product can be achieved from waste. Industrial waste materials can and should be processed at source; this reduces unnecessary storage, transportation and contamination issues. Cost is quite often an important consideration when recycling polymers: high labour costs and many process steps may deliver an expensive material when compared to the virgin grade available. Effective sorting and the reduction of process steps can ensure a low cost material is achieved. More research is required in the life cycle of waste composites, in particular where materials are being recycled by mechanical means. The mechanical damage imparted by the recycling process onto the fillers and fibres is not fully understood, especially when multiple processing steps and recycling cycles are considered. Additionally no knowledge is available on the ‘cyclic generation’ of a product destined for recycling. An article may have been produced from a third generation recycled polymer, so the material has already been processed and recycled three times. An improved understanding of these processing effects will provide more knowledge to the recycler. Also it would provide knowledge on if there are any benefits

250


to mechanically recycling composites, to preserve their ‘enhanced’ properties. The research would also identify when the recycled composite materials can be used and at which point the recyclate is rendered useless. Information is also required by the recycler to ascertain if there is a cost benefit to recycling composite material or to simply dilute them into more mainstream waste polymers. Finally it can be concluded that the technology exists now to recycle thermoplastic composite materials. What is missing is a more detailed study on the effects of these processes on highly engineered composite materials.


9.5

References

94/62/EC, European Parliament and Council Directive 94/62/EC of 20 December 1994 on Packaging and Packaging Waste, 1994 ASTM D1972–97 (2005) Standard Practice for Generic Marking of Plastic Products, 2005 béland, s., High Performance Thermoplastic Resins and Their Composites, William Andrew Inc., 1990 bernasconi, a., rossin, d., armanni, c., Analysis of the effect of mechanical recycling upon tensile strength of a short glass fibre reinforced polyamide 6,6, Engineering Fracture Mechanics, Volume 74, 2007 beukering, p. v., Recycling, International Trade, and the Environment: An Empirical Analysis, Springer, 2001 bledzki, a, kardasz, d, Fast identification of plastics in recycling processes, Polimery, Volume 43, Issue 2, 1998 brandrup, j., bittner, m., michaeli, w., menges, g., Recycling and Recovery Hanser / Gardner Publications, 1996 burat, f., et al., Selective separation of virgin and post-consumer polymers, PET and PVC, Waste Management, Volume 29, 2009 charles, r., leander, j., Waste Management and Resource Recovery, CRC Press, 1995 chung, c., Extrusion of Polymers: Theory and Practice, Hanser Verlag, 2000 coates, g., rahimifard, s., Modelling of post-fragmentation waste stream processing within UK shredder facilities, Waste Management, Volume 29, 2009 dimitrakakis, e., et al., Small WEEE: Determining recyclables and hazardous substances in plastics, Journal of Hazardous Materials, Volume 161, 2008 douglas, e., et al., Recovery of Potentially Reusable Materials from Domestic Refuse by Physical Sort, Resource Recovery and Conservation, Elsevier Scientific Pub. Co., 1976 ehrig, r. j., Plastics Recycling, Hanser / Gardener, 1992 erema online, www.erema.at/en, accessed Jan 2009 evans, s., et al., A study into waste polythene film recovery, The Waste and Resources Action Programme, 2003 for plastic bottles recycling, Resources, Conservation and Recycling, Volume 52, 2007 gent, m. r., Cylinder cyclone (LARCODEMS) density media separation of plastic wastes, Waste Management, Dec, 2008



251

goodship, v., Introduction to Plastics Recycling, Second Edition, Rapra, 2007 harper, c. a., Handbook of Plastics, Elastomers, and Composites, McGraw-Hill Professional, 2002 hilgenkamp, k., Environmental Health: Ecological Perspectives, Jones & Bartlett Publishers, 2005 hooper r., harder m., potter a., Profit from Plastic, Proceedings of the IMech E, on Eng. for Profit from Waste, London, 2001 http://www.bpf.co.uk/bpfindustry/process_plastics_recycling.cfm, accessed Dec 2009 http://www.essexcc.gov.uk/microsites/RecycleForEssex/harlow.htm, accessed Jan 2009 http://www.matweb.com, accessed March 2008 http://www.southbeds.gov.uk/environment/env_services/waste_recycling/Kerbside– Services.aspx#orange_lid http://www.wastecycle.co.uk/Services/recycling.html, accessed Jan 2009 http://www.zerma.com/brochures/zerma-gsh-heavy-duty-granulator.pdf, accessed Jan 2009 ISO 1043-2 – 2000, Plastics – Symbols and abbreviated terms – Part 2: Fillers and reinforcing materials, 2000 kikuchi, r., et al., Grouping of mixed waste plastics according to chlorine content, Separation and Purification Technology, Volume 61, 2008 knights, m., Plastic lumber – ready for prime time, Plastic Technology, Volume 8, Issue 8, August 1996 kreith, k., tchobanoglous, g., Handbook of Solid Waste Management, McGrawHill Professional, 2002 kroschwitz, j., Polymers: An Encyclopaedic Sourcebook of Engineering Properties, Wiley, 1987 lee, s., Handbook of Composite Reinforcements, John Wiley and Sons, 1992 mark, h. (ed.), Encyclopaedia of Polymer Science and Technology, Concise Third Edition, Wiley Interscience, 2007 miracle, d. b., 2001 ASM Handbook: Composites, ASM International, 2001 mustafa, n., Waste Plastic Management, Disposal, Recycling and Reuse, Marcel Dekker, 1993 nygard, p., et al., Extrusion-based wood fibre–PP composites: Wood powder and pelletized wood fibres – a comparative study, Composites Science and Technology, Volume 68, 2008 prichard, g., Two technologies merge: wood plastics composites, Plastics Additives and Compounding, Volume 6, Issue 4, 2004 rothon, r., Particulate Fillers for Polymers, Rapra Technology Limited, Smithers Rapra, 2001 scheirs, j., Polymer Recycling: Science, Technology and Applications, Wiley, 1998 sherman, l. m., Wood-filled Plastics – They Need the Right Additives for strength, Good Looks and Long Life, Plastics Technology, 2004 shrivastava, a., Wealth from Waste, APH Publishing, 2003 strong, b., Plastics, Materials and Processing, Pearson Prentice Hall, 3rd Edition, 2006 tachwali, y., et al., Automatic multistage classification system for plastic bottles recycling, Resources, Conservation and Recycling, Volume 52, Issue 2, 2007

252



UK Plastics Waste – A review of supplies for recycling, global market demand, future trends and associated risks, Wrap Publications (www.wrap.org.uk), Nov 2006 waite, r., Household Waste Recycling, Earthscan Publications, 1995 wang, y., Compounding in Co-rotating Twin-screw Extruders, Rapra Technology Limited, iSmithers Rapra, 2000 wrap, Realising the Value of Recovered Plastics, Market Situation Report, WRAP (www.wrap.org.uk), 2007 xanthos, m., todd, d., Plastics processing, Encyclopaedia of Polymer Science and Technology, Volume 11, 2004

10 Additives to upgrade mechanically recycled plastic composites R. P FA E N D N E R, Ciba Lampertheim GmbH, Germany


Abstract: Additives are essential components of plastic formulations providing maintenance, extension and/or modification of polymer properties, performance and long-term use. During the lifetime of a plastic application irreversible changes to the polymer structures take place and additives may be consumed. If used plastics are bound for a second life by mechanical recycling, additives are mandatory to upgrade recyclates to the desired property profile. The concept of restabilization, repair molecules and compatibilization are the most successful methods to enhance the properties of recyclates. Stabilizers improve processing and maintain the long-term properties, reactive molecules repair predegradation in some instances and compatibilizers improve the mechanical properties of polymer mixtures. Examples describe the suitable additives and findings in different single polymer substrates and mixed plastics for various applications. The mechanical recycling of plastic composites has to take into consideration the significant influence of the reinforcement material, e.g. fibers or fillers and its interaction with the polymer matrix. Unlike recycling of unreinforced plastics, there are few publications on the successful recycling of composites. Recycling of plastic composites offers still room for further research to understand this complex field and to elaborate solutions for successful recycling processes. Key words: mechanical recycling, stabilizers, compatibilizers, composites, fillers, glass fibers.

10.1

Introduction

Today’s successful plastics recycling is the combination of several options, i.e. mechanical recycling, chemical recycling and energy recovery. Additives contribute to a successful mechanical recycling of plastics, i.e. the reuse of used material mainly through remelting. Therefore, this chapter describes the use of additives in improving the properties of recyclates with regard to the polymer matrix and with regard to the interactions with composite elements such as fillers and fibers. As the addition of suitable additives to plastic materials to be recycled is usually performed in the polymer melt, this technique is limited to thermoplastic materials and the recycling of thermosets is basically outside this part. 253


254


To understand the behavior of recyclates in the new application it is necessary to consider the differences between virgin materials and postconsumer recyclates and to understand the impact of impurities and predegradation on the recyclate properties. Additives are essential components of plastic formulations providing maintenance, extension and/or modification of polymer properties, performance and long-term use (Zweifel, 2001; Pfaendner, 2006). Most additives primarily influence the properties of the polymer matrix of both virgin and recycled materials. Therefore, a large part of the present chapter is dedicated to the influence of additives on different recyclate structures and the scope and limits of upgrading recyclates with focus on stability, mechanical properties and long-term performance. More specifically the concept of restabilization, repair of pre-damage and compatibilization of polymer mixtures is addressed. Although the influence of additives on the polymer matrix is usually more pronounced than the influence on other composite components, the interactions between all ingredients of the formulation have to be taken into consideration and some additive classes such as coupling agents fulfill specific targets in this field. Plastic composite recyclates dealt within this chapter are all kind of systems consisting of a polymer matrix and any other under plastic processing conditions unmeltable component such as fibers or fillers, e.g. glass fibers or mineral fillers. Emphasis will be on the recycling of the entire composite system and less on the use of recycled individual components together with virgin material.

10.2

Properties of recycled plastics and recycled composites

Although there can be found useful applications for all kind of plastic recyclates, the ultimate target of plastics recycling should be replacement of virgin plastics with the same functionality, e.g. so-called closedloop recycling within the same application. This most energy efficient and most environmentally friendly way of recycling is only possible by mechanical recycling. Single plastic materials and mixed plastic materials with defined composition are best suited for this purpose. On the other side contaminated plastics or plastics where high sorting and purification efforts are necessary, should be recycled chemically or by energy recovery. However, there are fundamental differences between virgin material and recyclates even by using well-defined recyclate streams.

Additives to upgrade mechanically recycled plastic composites

255

10.2.1 Differences between recyclates and virgin materials


During the first processing of virgin plastic and its first service life, irreversible changes take place in the polymer chain, induced mechano-chemically, chemically or by irradiation. In chemical processes induced by radical mechanisms in the presence of ubiquitous oxygen new chemical groups are formed in the chain and the polymer structure and/or polymer composition is changed (Table 10.1). The concentration of the newly formed structures increases with the service time and the environment or the exposure (e.g. UV light, high temperatures) but can be reduced by proper formulation with stabilizing additives. More mechanistic details on polymer degradation may be found in the literature (Zweifel, 2001; Singh and Sharma, 2008). In any case the resulting structural inhomogeneities influence the overall properties and it is commonly accepted that recyclates are more sensitive to further oxidation effects than virgin material.

Table 10.1 Structural differences between virgin and recycled polymers Predamage through processing/aging/oxidation

Influence of mixtures

Influence through foreign matter

Chemical changes • carbonyl group content • hydroperoxide content • double bonds concentration • new end groups/end group concentration • initiator sites for further oxidation • discoloration

Different polymer structures Different polymer types/grades Production method/ technology (catalysts, additives) Different producers Incompatibility of mixtures/blends (phase separation) Different additives Different predamage Fluctuating composition

Other polymers Metal traces (e.g. rust, catalyze autooxidation) Contact media (e.g. oil, fat) Inorganic impurities (e.g. dust, sand) Organic impurities and their degradation products (e.g. inks, paints, adhesives) Fluctuating composition

Polymer molecular weight degradation branching cross-linking molecular weight distribution • change of processing behavior • change of mechanical/ physical properties

• • • •

Fluctuating composition

256


Besides structural inhomogeneities, recyclates may contain impurities from the first application or foreign material which could not be removed by purification steps. In addition to the recyclate characteristics of the polymer matrix differences of composites between virgin material and recyclates can be attributed to different size of the fiber/filler, modified adhesion to the matrix and compatibility changes. As a general rule substantial structural changes, mixtures or impurities influence the overall properties of a polymer and render it difficult to achieve a recyclate of high quality. However, there can be positive effects as well, e.g. polar groups on polyolefins through oxidation might improve the compatibility of a polar fiber/filler.


10.2.2 Effect of predamage and recyclate quality on the properties of plastics Whether a recyclate is suitable to fulfill the requirements of a new application is evaluated through mechanical (e.g. tensile strength, impact strength), physical (e.g. hydrolytic stability, gloss, surface roughness) and processing tests (e.g. extrusion, injection molding) in standardized conditions. Standard test methods to analyze the stability of polymers cover multiple extrusion, accelerated heat aging and UV light exposure. Multiple extrusion evaluates the melt processing stability of a polymer or polymer formulation as high temperature, shear forces and oxygen in the polymer transformation process cause degradation. The melt properties are usually characterized by melt flow rate (MFR) or melt volume rate (MVR) according to ISO 1133. For example the processing stability of polypropylene (PP) was studied by multiple extrusion with the use of a twin screw extruder (Hinsken et al., 1991). Molecular weight distribution and polydispersity of PP after processing were found to be in good relation to rheological changes and the mechanisms of degradation of PP could be identified. Accelerated heat aging is simulating long-term thermal stability of plastics and effectiveness of antioxidants. Samples are oven-aged at defined temperatures and conditions, e.g. according to ISO 4577, and the change of the properties, e.g. visual appearance, color, mechanical values are recorded over aging time. The lifetime of polymers depends on the chemical structure, testing temperature, type and concentration of antioxidants and other additives (Gugumus, 1999). Photooxidation is tested by UV exposure, artificial (e.g. ISO 4892) or natural weathering (e.g. ISO 4582, ISO 4607). Change of the properties, e.g. visual appearance such as chalking, crazes, loss of gloss, and/or mechanical properties such as tensile strength, elongation, impact strength are


257

measured. The light sources of artifical weathering equipment match the terrestrial sunlight, a considerable acceleration is achieved versus natural weathering. Typical exposure sites for natural weathering are Florida, Arizona and southern France. For example 2760 hours of artificial weathering correspond to approximately 1 year Florida exposure when 50% retained tensile strength of PP tapes are considered as test criterion and when the PP is stabilized with 0.10% hindered amine stabilizer (Gugumus, 2001). More detailed information on testing methods, activities of stabilizers and technical aspects of stabilization for different polymers can be found elsewhere (Zweifel, 2001). As a consequence of predamage from first processing and use, most plastic recyclates will not fulfill the requirements of the same or similar application if they are not upgraded by selected additives as outlined later. However, to understand the influence of the predegradation on plastic properties a few published examples should illustrate the effect on the most important polymers in a simplifed way with focus on stability and performance:


•

Polypropylene (PP): the molecular weight of PP (and PP copolymers) is significantly reduced by processing and aging with a direct impact on the mechanical properties. Long-term applications of PP recyclate from, for example, battery cases result in inferior stability. 16 days until embrittlement at 135 °C were found for the recyclate versus 38 days for the virgin alternative (Pfaendner, 2001). Virgin PP elongation at break was at 680%, whereas after five extrusions only 20% were recorded (La Mantia, 1998). • Polyethylene (PE): depending on the type (LDPE (low density), LLDPE (linear low density), HDPE (high density)) and the manufacturing catalyst PE tends mainly to cross-linking through aging. Most critical during outdoor exposure is crack formation and catastrophic failure. For example waste bins made from HDPE achieved a tensile impact strength of 404 kJ/m2 whereas the recyclate attained only 291 kJ/m2, a significant drop in mechanical properties due not only to aging but as well to a product mix from different manufacturers, use time and exposure (Pfaendner, 2001). Furthermore, it was shown in experiments with LDPE and HDPE that alternating processing and aging caused considerably more severe degradation then either processing or aging alone (Boldizar et al., 1995, 2000; Jansson et al., 2004; Hamskog et al., 2004). • Polystyrene (PS): the main degradation path of PS recyclates is some molecular weight reduction and discoloration after thermal exposure, impact polystyrene suffers from cross-linking of the elastomer part (Pfaendner, 2001).

258



•

Polyvinylchloride (PVC): predamage of PVC recyclate from rigid PVC applications such as window profiles or pipes is limited to the most outer part of the application, therefore the values of the mechanical properties after regrinding and reprocessing are often still acceptable. Less stabilized materials such as PVC bottles need addition of heat stabilizers, e.g. on the basis of Ca/Zn (Ulutan, 2003). • Polyethylene terephthalate (PET): the PET recyclate, mainly from used bottles, shows a loss of molecular weight (virgin material: intrinsic viscosity 0.73–0.80 dl/g, recyclate less than 0.70 dl/g), caused by degradation through processing and hydrolytic degradation (Awaja and Pavel, 2005; Kiliaris et al., 2007). • Mixed plastics: mixed plastics contain different types of plastics with different processing behavior and stability. Usually these plastics are not compatible (or thermodynamically miscible) with each other and the resulting properties are very often inferior to the properties of the parent polymers. The influence of plastic mixtures on the stability can be demonstrated in a model experiment: standard virgin HDPE, e.g. used in bottles achieved easily more than 100 days until embrittlement at 120 °C, the recyclate from 100% HDPE achieved still 73 days, taking a mixture of 5% PP and 95% HDPE reduced the stabiliy to only 18 days with further reduction at increasing PP concentration (Pfaendner, 2001). • Composites: glass fiber reinforced polybutyleneterephthalate (PBT) composites showed in model experiments reduced impact strength (31.1 MPa vs. 37.6 MPa) and tensile strength (103 MPa vs. 134 MPa), resulting probably from a reduction in molecular weight as claimed by the authors (van Lochem et al., 1996). A similar result was found for glass fiber reinforced PPS (van Lochem et al., 1996). Other researchers confirmed that glass fiber reinforced PBT showed reduced tensile strength and elongation; however, improved values of Izod notched impact strength of the recyclate: the latter was attributed to inhomogeneities within the composite and fiber bundles which increase fracture energy (Chu and Sullivan, 1996). As the properties of composites are determined by the fiber type and dimension and by the interaction with the matrix it is likely that any composite will show inferior impact properties as recyclate independent of the structure if the fiber dimensions change during the recycling process.

10.3

Additives to upgrade recycled plastics

Usually, used plastics cannot simply be reused in the form in which they were recovered. To improve to some extent the quality of a recyclate, it is sorted, separated, cleaned and reprocessed in optimized compounding


259

steps. Furthermore, upgrading of recyclates is a must to satisfy the requirements for extended product life and appearance in new applications.

10.3.1 Scope and limits of upgrading and quality improvement of recyclates


In addition to purification processes upgrading means the addition of materials such as additives to meet the requested properties of the targeted application. For reformulation to various specifications the complete range of additives as used in virgin plastics is available. Examples are compatibilizers, impact modifiers, stabilizers, reinforcement agents like glass fibers and mineral fillers as well as pigments used for re-pigmentation (Table 10.2). Most of the additives as described will influence the properties of the plastic matrix itself but as well the overall properties of the plastic composite. Regarding the literature concerning additives for the mechanical recycling of plastics most knowledge is available for stabilizers and for compatibilizers. The simple addition of virgin material to upgrade the recyclate by dilution was proven to be not very efficient (Hamskog et al., 2006)

Table 10.2 Additives to upgrade recycled polymers and their function Type of additive

Function

Virgin polymer Polymeric modifiers (e.g. impact modifier)

Dilution of recyclate Improvement of mechanical/ physical properties Improvement of compatibility of polymer blends, of mechanical/ physical properties Improved processing, higher melt flow, higher throughput Improved processing, extended long-term heat stability Extended outdoor/light stability

Compatibilizers

Processing aids, e.g. lubricants, waxes Stabilizers, e.g. antioxidants, processing stabilizers Stabilizers, e.g. UV-stabilizers, hindered amine stabilizers Metal deactivators Pigments Reactive molecules (repair additives, in situ compatibilizers) Reinforcement agents (fillers/fibers) Silicates, e.g. zeolites Coupling agents Radical generators, e.g. peroxides

Extended thermal stability in the presence of metal impurities Visual appearance Improved mechanical/physical properties, odor reduction Modified mechanical properties Odor reduction Improved mechanical properties, compatibilization Melt flow adjustment, modified mechanical properties

260


at least as far as the pure polymer is considered. It might be different for polymer composites when the main influence on the properties is determined by the fiber or filler.


10.3.2 Restabilization Restabilization in order to improve the quality of recyclates is essential (La Mantia, 1998; Pfaendner, 2001; Luzuriaga et al., 2006). Stabilizers protect recyclates against both oxidative and photooxidative damage and protect, as a result, the available properties both during processing as well as during the product life cycle in the intended field of application (heat, light). This, however, is no different from when virgin material is used. At least the quantity of stabilizers consumed during the original application has to be adjusted to fulfill the requirements of the application of the recyclate. Also any stabilizer residues from the first life will help recycling (Khan and Ahmed, 2001). However, this usually is in itself not enough. All plastics originally used in short-term applications such as in the packaging industry contain hardly any stabilizers and no light protection. In order to provide a material capable of meeting the requirements, for example of a long-term outdoor application, the recyclate has to be restabilized. The products used in restabilization of most recyclates (exception PVC) are mainly on the basis of phenolic antioxidants, phosphites and costabilizers such as antiacids for processing and long-term heat stabilization and hindered amine stabilizers (HAS) compounds and/or UV absorbers for light stabilization. Although the principal stabilizer classes are not different from virgin material the appropriate recyclate stabilizer must address the specific degradation profile of the recyclate. It means that the amount of stabilizer, the ratio between different stabilizer types and additional additives, has to be optimized. As a rule the optimized stabilizer according to price/performance will be different from the optimized stabilizer for the corresponding virgin material. Nevertheless a good basic stabilization of the virgin material is one prerequisite for a qualitatively high level recycling. On the other hand restabilization before each recycling step is favored (Marrone and La Mantia, 1996) contrary to an ‘over’-stabilization at the first processing. A few examples should now outline the effects of restabilization on plastic recyclates with reference to the most important polymers: • PP: The processing stability, long-term heat stability and mechanical characteristics are decisively improved by restabilization, e.g. higher values for the impact strength (115 kJ/m2 vs. 62 kJ/m2 not restabilized), tensile impact strength (430 kJ/m2 vs. 365 kJ/m2) and elongation (99% vs. 64%) of transportation crates were found when the PP recyclate was


•


•

•

• •

261

restabilized with a specific recyclate stabilizer system (RS-1). The longterm thermal stability of battery cases achieved the stability of virgin material when 0.2% of a stabilizer RS-2 was added (Pfaendner, 2001). A blend of different PP recyclates including filled and reinforced grades is comparable to virgin material and can be used in automotive interior and exterior parts when a designed restabilization based on a stabilizer formulation and hindered amine light stabilizers is used (1.0% RS-3 + 0.6% LS-1, Hermann et al., 2000). Even PP from municipal waste collection can compete with virgin material with regard to processing stability and thermal stability (oxidative induction time, OIT), if the recyclate is restabilized by RS-1 (Martins and De Paoli 2002). PE: HDPE bottle crates from recyclates lost 50% of their initial tensile impact strength after 2000 h of artificial weathering and 90% after 4000 h. Restabilized materials with a system comprising processing, heat and hindered amine light stabilizers (RS-4) keep their properties at more than 90% of the starting value even after 8000 h of artificial weathering (Pfaendner, 2001). PS: Discoloration and molecular weight degradation can be avoided when a small concentration of antioxidant is added to the recyclate (e.g. AO-1), e.g. the Yellowness Index (YI) after 935 h at 80 °C was found at 4.0 vs. 17.6, if the recyclate was not restabilized (Pfaendner, 2001). PET: The processing stability of PET from bottle flakes can be improved by adding a combination of phenolic antioxidants and phosphites (AO-2/PS-1), where the intrinsic viscosity (IV) was kept at 0.75 dl/g vs. 0.70 dl/g without restabilization (Pfaendner, 2001). A further possibility is the upgrading of polyesters with reactive molecules (see below). PVC: PVC of inferior stability may need addition of heat stabilizers, e.g. on the basis of Ca/Zn (Ulutan, 2003). Mixed plastics: The long-term stability of polyolefin blends (PE/PP) can be improved by adding a stabilizer combination (RS-2), e.g. the time until embrittlement at 120 °C is increased from 18 days to 116 days, when 0.2% of the stabilizer combination is added. Even very heterogeneous mixtures of plastics from household waste can be improved by adding stabilizers with regard to long-term heat stability and UV stability (Pfaendner et al., 1995).

10.3.3 Repair of predegradation Traditionally restabilization keeps only the level of the material by avoiding degradation but cannot repair the predegradation by readjusting, for example, the loss of the molecular weight. For polycondensation materials such as polyesters and polyamides reactive molecules (‘repair molecules’) are suitable to react with the functional groups of the polymer (e.g. —OH,


262


—NH2, —COOH end groups) and are capable of restoring the original properties or to adjust new properties. These reactive molecules are known also as ‘chain-extenders’. The process can be carried out via solid state reactions or preferably by reactive extrusion. A number of chemical compounds have been suggested such as epoxides, oxazolines, oxazolones, oxazines, isocyanates, anhydrides, acyllactams, maleimides, phosphonites, cyanates, alcohols, carbodiimides and esters preferably in combination with a catalyst, e.g. based on hydroxyphenylalkylphosphonic esters (Pfaendner et al., 1997, 1998a,b,c, 2004). In addition to linear molecules, branched or slightly cross-linked polymers are accessible by selecting multifunctional additives and by adjusting the loading of the additive. For example, a combination of pyromellitic dianhydride (0.50%) and hydroxyphenylalkylphosphonic ester (RM-1) (0.25%) increased the molecular weight of polyesters, shown as IV, from 0.46 to 0.83 dl/g, which is well in the range of virgin material (Pfaendner et al., 1997). The catalytic active additive accelerates the reaction in such a way that the process is feasible during extrusion conditions. Similarly by using phosphonates with vacuum venting during extrusion the molecular weight (Mw) of recyclates could be increased from 50 000 to 71 000 by adding the phosphonate RM-1 (Regel et al., 2001). Some more recent process improvements by adding reactive molecules of different reactivity should be mentioned as well (Simon et al., 2002; Forsythe et al., 2006; Kiliaris et al., 2007). Triphenylphosphite functioned as chain extender too; however, the resulting polyester degraded again during storage due to by-products formed in the molecular weight build-up reaction (Cavalcanti et al., 2007). Toughened PET recyclate can be manufactured by combining PET, polycarbonate and methylenediphenyldiisocyanate (Tang et al., 2007). Similarly to polyesters, polyamides can be ‘repaired’. By adding a combination of a reactive additive, such as dioxiranes and an additive which has a catalytic effect (hydroxyphenylalkylphosphonic esters) hydrolytically pre-damaged polyamide can attain its initial molecular weight and, consequently, all mechanical properties (Pfaendner et al., 1998a). Careful adjustment of the concentration of the various components provides the basis for manufacturing tailor-made products. Selected phosphonates might be used alone to increase the molecular weight of polyamides in an extrusion process similar to PET (Vouyiouka et al., 2006). Contrary to polycondensation polymers repair or molecular weight build-up of polymerisation polymers such as PE, PP or PS is usually difficult. In some cases due to the presence of functional groups from degradation, reactive formulations can improve the mechanical properties of polyolefins which might be attributed to a reaction and molecular weight repair by branching. For example the elongation of a PP/EPDM (ethylene– propylene–diene monomer) recyclate could be improved from 84%


263

(standard stabilization) to 344% (adding RS-3, 0.7%) with the help of a reactive dioxirane compound. Degraded polyethylene could be rebuilt by adding either an ethylene-co-glycidyl methacrylate or a hydroxylamine derivative. However, in both cases there is the danger of cross-linking and loss of mechanical properties at higher additive concentrations (Scaffaro et al., 2006, 2007). Although there is no published experimental evidence it can be expected that composite properties are improved by adding repair molecules if the predamage is resulting in functional groups, e.g. from hydrolysis, as potential reaction partners.


10.3.4 Compatibilization of polymer mixtures Polymers of different structures are not miscible thermodynamically and, therefore, do not form homogeneous blends. The polymer in high concentration will form a continuous phase, and the polymer with a low concentration will be dispersed in the continuous matrix. However, the intermolecular adhesion between the continuous and the dispersed phase is very weak, consequently resulting in insufficient mechanical properties of the blend. To improve the mechanical properties of polymer blends as well as recyclate blends compatibilizers are used. Compatibilizers modifiy polymer interfaces by reducing the interfacial tension in the melt, stabilizing the dispersed phase against growth during annealing (‘morphology stabilizers’), increasing the adhesion at phase boundaries and minimizing phase separation in the solid state and, therefore, prevent delamination. The most appropriate compatibilizers for the application in question have to be balanced according to the chemical structure of the particular plastic; it means that ‘universal’ compatibilizers hardly exist and that the compatibilizers have to be adapted to the polymers to be compatibilized. Therefore, there are numerous references describing specific recyclate combinations and adjusted compatibilizer structures and levels. An extended discussion on compatibilizers in general has been compiled (Datta and Lohse, 1996) and can be found elsewhere more specifically for recyclates (Lemmens, 1996; Mangaraj, 1998; Vilaplana and Karlsson, 2008). Non-reactive and reactive compatibilizers are applied. Reactive compatibilizers form covalent bonds with functional groups of the polymers. Typical examples of reactive compatibilizers are maleic anhydride or acrylic acid grafted polyolefins or glycidylmethacrylate copolymers. Furthermore, reactive monomers can form compatibilizers in situ. Non-reactive compatibilizers are usually miscible with one of the blend components. Typical representatives include ethylene–acrylester copolymers or styrene– butadiene–styrene or styrene–isoprene block copolymers. A concentration of at least 5% of the compatibilizer is often required for effectivity. It also

264



has to be considered that mechanical properties interact with each other. As a result using compatibilizers can improve impact strength. On the other hand, however, flexibility and stiffness are reduced. A few examples should illustrate the effect of compatibilizers in heterogeneous recyclate blends: • PE/PS blends were compatibilized with styrene–ethylene–butylene– styrene block copolymer (SEBS) in the range of 7% for blend compositions of 20/80 to 80/20. In the best case the Charpy impact strength increased by a factor of 4. Morphological characteristics showed a decrease of the size of the dispersed particles and an improved interfacial adhesion between both phases (Kallel et al., 2003). • PE/PP: Impurities of PP in PE (LDPE, LLDPE, HDPE) usually cause a severe drop in elongation and impact strength. The addition of ethylene propylene (EP) random copolymers in the range of 2–10% resulted in adequate impact performance in particular in LDPE and LLDPE mixtures (Hope et al., 1994). Even the addition of recycled aged polypropylene improved PP/HDPE mixtures to some extent (Bonelli et al., 2001). • PE/PET is compatibilized preferably by ethylene–glycidylmethacrylate or styrene–ethylene–butylene–styrene–graft-maleic anhydride block copolymers (Pawlak et al., 2002), the achievable results depend on the ratio of the components, concentration of the compatibilizer and the processing conditions. Terpolymers from ethylene–butyl acrylate– glycidylmethacrylate were used alternatively resulting in substantial improvements of the impact strength (Kaci et al., 2005). • PP/PET: In a similar way to PE/PET, PET/PP blends could be decisively improved by adding styrene–ethylene–butylene–styrene–graftmaleic anhydride block copolymers or with less success by PPg-maleic anhydride (La Mantia, 1998; Papadopoulou and Kalfoglou, 2000). • PE/PA: For compatibilization of PE/PA blends polyethylene–comethacrylic acid ionomers have been recommended (Hausmann, 1995; Grützner et al., 1993); other reactive copolymers such as polyethylene– graft-acrylic acid or polyethylene–graft-maleic anhydride may be suitable as well. For PP/PA mixtures styrene–ethylene–butylene–styrene– graft-maleic anhydride showed better results in particular for Izod impact strength than PP-graft maleic anhydride (Hong et al., 2006; Kim et al., 2007). • Mixed plastics: In a profound study with model blends and mixed plastics from household waste where PVC, PS and PET are dispersed in a polyolefin matrix it has been shown that impact strength, ductile–brittle transition temperature and elongation at break, properties most nega-


265

tively affected by incompatibility of the various blend components, are very much improved by compatibilizers. Uncompatibilized blends, brittle at 20 °C, could be made ductile at temperatures well below 0 °C and in some cases at temperatures as low as −30 °C. Styrene block copolymers are most interesting, although addition of 5 to 10% are necessary. Furthermore by adding compatibilizers a higher degree of consistency for measured properties has been found (Obieglo and Romer, 1993; Vanhaeren et al., 1997). More recently it was shown that the effect of compatibilizers can be combined with the aid of high energy radiation to achieve a more uniform material (Elmaghor et al., 2003).


10.3.5 Other additives and processes to improve recyclate quality • Coupling agents: coupling agents are reactive molecules which react chemically either with a filler/fiber and/or the polymer matrix and improve thus the adhesion between the components. Also some reactive compatibilizers may act in the same way and there is no clear defined terminology in literature. In general, coupling agents are low molecular weight reactive molecules mainly used in improving rubber filler adhesion and glass fiber polymer adhesion, whereas compatibilizers are polymers and act mainly in polymer blends. Coupling agents are a wide range of chemical compounds, starting from fatty acids and its salts such as calcium stearate, organofunctional silanes widely used for glass fibers, titanates, zirconates and anhydrides (Hohenberger, 2001). For example, silane coupling agents improved the tensile strength, elongation and impact strength of PP/PET mixtures (Oyman and Tincer, 2003). Titanate coupling agents could improve the elongation at break and slightly impact strength of mixed plastics at a concentration of 1% similar to chlorinated polyethylene where 10–20% were used (Fellahi et al., 1991). Furthermore coupling agents are used in fiber composites (see below). • Impact modifiers: impact modifiers are mainly elastomeric compounds based on butadiene such as styrene–butadiene–styrene (SBS), styrene– isoprene–styrene (SIS) or ethylene–propylene–diene copolymers (EPM, EPDM) often related structurally to the above-mentioned compatibilizers. By adding impact modifiers to the recyclate, impact strength and elongation is increased while the modulus is usually reduced. The appropriate choice of impact modifier depends on the specific plastic to be toughened; main applications are in PS, PP and engineering plastics such as PA, polybutylene terephthalate (PBT), PET (Greco, 1998; Cruz, 1998).


266


• Metal deactivators: metal deactivators form complexes with metal ions and reduce thus the negative influence of metals on the polymer properties such as reduced oxidation stability. In LLDPE nanocomposites it could be clearly shown that the addition of UV absorbers of benzotriazole, benzophenone and hydroxyphenyltriazine structures extend decisively the lifetime of the nanocomposite film; however, a metal deactivator (MD-1) alone outperformed in these experiments the UV absorber, indicating that the influence of metal impurities is very crucial. A combination of UV-absorber with metal deactivator showed only a minor additional improvement. When MD-1 is added to the LLDPE nanocomposite elongation at break and tensile strength of the tested film samples are still at a level of 70% of the starting values after 300 hours of QUV irridiation compared to a complete loss of mechanical properties of the unstabilized sample after 70 hours (La Mantia et al., 2006). • Melt flow adjustment: in order to adjust the melt flow of a polymer according to the required transformation process the possibilities are somewhat limited. A decreased melt flow (higher melt viscosity, higher molecular weight) can be achieved in some cases by considering the already described ‘repair’ molecules. An increased melt-flow (lower melt viscosity, lower molecular weight) is accessible, e.g. with polypropylene using radical generators such as peroxides, hydroxylamine esters (Roth et al., 2006, 2008) or azoalkanes (Aubert et al., 2007) or in the field of polycondensation polymers (PET, PA) by hydrolytic cleavage. Processing aids, lubricants, waxes and addition of oligomers may help to improve processing, to lower the melt viscosity and to increase the throughput. • Odor reduction: post-consumer recyclates suffer often from odor problems caused by contaminations or degradation products from the first application. Removing odor is a challenging task as very low quantities of volatile products can be the reason which are difficult to analyze or to trace back. Technically, odor can be reduced by adjusted processing including vacuum venting or vacuum venting in the presence of a carrier such as water (Schrader, 2004, 2008). Another potential way to reduce odor is through additives, e.g. RS-3, zeolites (Gioffre and Marcus, 1989) or selected silicates (Heberer et al., 2002). Despite the ecological benefits of mechanical recycling and the proven advantages of additives to improve the quality of recyclates, the economic aspect cannot be neglected. The additional cost contribution of additives starts within the cent range per kg of recyclate for restabilization, but can as well achieve a much higher range, if larger concentrations of, for example,


267

a reactive compatibilizer have to be used. Benchmark and attractiveness to use the recyclate will be the cost difference to the virgin material of a given application.

10.4

Specific examples of additives for recycled plastic composites

Contrary to numerous publications describing upgrading of plastics through additives, the literature on recycled plastic composites is rather limited. One reason might be that the amounts of fiber or filler reinforced recyclate is much smaller and the lifetime of the primary application much longer in addition to a more complex system to be dealt with.


10.4.1 Closed loop recycling of filled polypropylene through restabilization Fillers are normally used to reduce the cost of the plastic or to adjust mechanical properties. For example by adding fillers the modulus is increased; the dimensional stability and the thermal resistance, on the other hand, and tensile and flexural strength and elongation at break can be adversely affected. It is well known that inorganic fillers such as calcium carbonate and talc decrease the (photo)stabilizing efficiencies of all light stabilizers (hindered amines, benzophenones) and antioxidants dramatically. Polypropylene in the presence of 0.05% of a typical phenolic antioxidant (AO-2) withstood at 120 °C a time until 50% retention of tensile strength of 3508 hours vs. only 470 hours in the presence of 10% CaCO3 (talc resulted in 2180 hours) (Hu et al., 1994). This known negative effect of fillers has to be considered in recyclates. In a detailed study of closed-loop recycling of talc filled garden chairs the influence of restabilization on the processing (Kartalis et al., 2002), longterm heat aging (Kartalis et al., 2003a) and light stability (Kartalis et al., 2003b) of this material was evaluated. The investigated recyclate consisted of white pigmented (TiO2) garden chairs with 15% CaCO3 filler. Independent of the processing (twin screw, single screw) the not-restabilized material showed a substantial increase of the melt flow resulting in some loss of mechanical properties (Table 10.3). As expected the degradation (melt flow increase) is increased at higher processing temperature. By restabilization the melt flow can be kept at lower values (unprocessed recyclate: MFR 19 g/10 min

268


Table 10.3 Properties of recycled filled polypropylene after processing

Prcessing Tmax 260 °C 280 °C 260 °C 260 °C


280 °C

Stabilization

MFR [230 °C/2.16 kg] g/10 min

Not restabilized 25.3 Not restabilized 28.8 0.2 RS-4 + 0.2 21.9 RS-2 0.2 RS-4 + 0.5 20.6 RS-3 0.2 RS-4 + 0.5 22.1 RS-3


Elongation at break (%)

Modulus of elasticity (MPa)

Tensile impact strength (kJ/m2)

28.98 28.0 29.04

49.6 50.2 62.7

798.2 736.9 783.7

103 116 101

30.45

72.1

726.4

115

28.9

85.4

782.6

122

[230 °C/2.16 kg]) independent of the temperature indicating reduced degradation of the composite during processing (Kartalis et al., 2002). Furthermore, restabilization including components with reactive oxirane groups (RS-3) and acting moreover as filler deactivators/coupling agents is capable of improving mechanical values such as elongation at break and tensile impact strength. The restabilization effect is translated as well into the long-term thermal properties of the filled recyclate, which were tested up to 2000 hours at 135 °C (Table 10.4). The not-restabilized material survives only a few days, which is also much shorter than expected from an unfilled polypropylene. Independent of the structure inorganic fillers often show experimentally a negative effect on the oxidative stability of the polymer; however to varying extents. There is no doubt that mainly interactions between the stabilizer and the filler and adsorption/desorption mechanisms are responsible for this influence. The surface area of the filler and pore volumes, surface functionality, hydrophilicity, thermal and photosensitisation properties of the filler, transition metal ion content (manganese, iron, titanium) have been found as potential elements of the interaction (Allen et al., 1998). Therefore, stabilizer formulations for filled plastics should include filler deactivators. The same is true for recyclates containing fillers, where the restabilization with RS-3 extended the time to embrittlement and preserved the mechanical properties over time. By increasing the concentration of RS-3 further (1%) the mechanical values (tensile strength, tensile impact strength) were kept unchanged up to 2000 hours at 135 °C (Kartalis et al., 2003a).


269

Table 10.4 Long-term thermal stability of recycled filled polypropylene

Processing Tmax

Stabilization

Days to embrittlement at 135 °C

260 °C 280 °C 260 °C 260 °C 280 °C

Not restabilized Not restabilized 0.2 RS-4 + 0.2 RS-2 0.2 RS-4 + 0.5 RS-3 0.4 RS-4 + 0.5 RS-3

6 5 26 58 67

Hours until 50% residual tensile strength at 135 °C 100 1250 1750

Table 10.5 Light stability of recycled filled PP


Tensile impact strength (kJ/m2) after artificial weathering Processing Tmax

Stabilization

0h

260 °C 260 °C

Not restabilized 0.2 RS-4 + 0.2 RS-2

103 113

46 75

42 65

44 105

42 65

260 °C 260 °C 260 °C

0.2 RS-4 + 0.5 RS-3 0.4 RS-4 + 0.5 RS-3 0.4 RS-4 + 1.0 RS-3

115 119 116

93 84 118

90 95 105

101 97 95

81 103 106

1000 h

2000 h

3000 h

4000 h

As garden chairs are an outdoor application, the weathering resistance of the restabilized filled PP recyclate formulations was evaluated by artificial weathering up to 4000 hours (Table 10.5). Contrary to oven aging, where the loss of mechanical properties takes place very quickly and dramatically as soon as the stabilizers are consumed, the reduction of mechanical properties during artificial weathering is slower as the degradation starts as a surface reaction and it takes a certain exposure until cracks are formed and the mechanical properties are destroyed. In any case the notrestabilized PP chair material lost about 60% of the initial tensile strength (which is already lower than in the stabilized formulations) before 1000 hours of artificial weathering. Depending on the stabilizer type and concentration even after 4000 hours more than 90% of the starting value was maintained. The advantage of the presence of a stabilizer system containing filler deactivators (RS-3) is therefore obvious (Kartalis et al., 2003b). Some other publications on PP recyclate composites mention wollastonite (calcium silicate) and calcium oxide whereas the fillers, coated with


270


neopentyl(diallyl)oxy-tri(dioctyl)pyrophosphate titanate, are incorporated to improve the mechanical properties e.g. the modulus of the recyclate (Vinci and La Mantia, 1997). HDPE/PS recyclate blends were compatibilized by SEBS and 20% calcium carbonate, resulting in an improvement of unnotched Charpy impact strength from 39 to 70 kJ/m2 (Sahnoune et al., 2003). Furthermore PP recyclate composites (Pramanik and Dickson, 1997) and PE recyclate composites (La Mantia and Morreale, 2006) were prepared by the incorporation of wood flour in the presence of compatibilizers. Organic fillers resulted in worse processability than inorganic fillers, but in similar stiffness and thermomechanical resistance (La Mantia et al., 2004). A further possibility of composites on organic basis is the use of cellulose fibers. Mixed plastics were melt blended with cellulose fibers from Pinus radiata, the Monterey pine, in the presence of maleated PP or titanate coupling agent resulting in improved tensile strength (Miller et al., 1998). Paper as such (Mehrabzadeh and Farahmand, 2001) or chemically modified fibers from paper production (Espert et al., 2003) showed a similar effect. However, all these investigations targeted the development of improved recyclates by manufacturing composites and the further recycling of the resulting material was not investigated. In recent papers composites of recyclates and ‘virgin’ nanoclays have been investigated. Recycled PET was intercalated with organoclays to produce nanocomposites (Bizarria et al., 2007). Despite some improvements in yield strength and modulus by adding 1% of organically modified montmorillonite, the molecular weight of the PET was considerably reduced. Recycled HDPE was compounded in the presence of an organically modified montmorillonite and maleic anhydride grafted PE or a titanate as compatibilizer/coupling agent (Lei et al., 2007). By adding 5% of maleic anhydride grafted PE the impact strength of the nanocomposite could be increased by 44%. Storage and loss moduli increased best in the presence of the titanate. Furthermore, by modelling recycling of layered silicate thermoplastic olefin elastomer (Thompson and Yeung, 2006) and maleic anhydride graft PP compatibilizer it was demonstrated in multiple extrusion experiments that the nanoclay has no additional effect on the onset of degradation. However after 10 extrusion passes the extent of degradation of the nanocomposite is twice the value of the neat resin. The increased degradation is attributed to the compatibilizer as the grafted maleic anhydride acts as chain transfer agent. With regard to the mechanical properties it was found that the nanocomposite lost only less than 10% of the initial modulus despite the degradation, probably because of changes in the network structure of the clay polymer system. The authors recommended adding antioxidants and additional compatibilizer as a recycling strategy of a nanocomposite material.


271


10.4.2 Recycling of glass fiber reinforced plastic composites During recycling of glass fiber reinforced plastics shortening of the glass fibers results in some loss of the mechanical properties. Therefore, an approach to add long-glass fibers to the recyclate to readjust the mechanical properties has been proposed (Wernicke and Skaletz, 1997). In comparison with virgin material (Table 10.6) the recyclate from post-consumer and production scrap showed lower mechanical properties. In further experiments long-glass fiber reinforced PP (40% glass fibers, manufactured by melt pultrusion) and virgin PP was added to the mixture in a ratio to keep the overall glass content at about 30 weight %. Increasing concentration of long-glass fibers (fiber length: 10 mm, diameter: 10–25 μm) improved the resulting mechanical properties to the level of virgin material. Therefore, glass fiber reinforced recyclate could be upgraded to the former mechanical values as the reduction of glass fiber length caused by the processing step is compensated by a suitable portion of long-glass fibers. Also recycled PP was improved, namely in mechanical properties and thermomechanical resistance, if short-glass fibers were added (Vinci and La Mantia, 1997). In a similar way it could be shown that a mixture of PP materials from automotive parts can be recycled efficiently if the compound is improved by adding a compatibilizer (maleic anhydride modified PP), stabilizer (phenolic antioxidant), PP-based impact modifier and short-glass fibers. In comparison with the starting material, the tensile strength was increased from 5.1 to 25.2 N/mm2 and the impact strength from 5.8 to 20.4 kJ/m2 Table 10.6 Upgrading of PP composites through long-glass fibers

Virgin PP Recyclate PP Recyclate containing 4% LGF Recyclate containing 8% LGF Recyclate containing 12% LGF

Glass fiber content (%)


Breaking extension (%)

Modulus of elasticity (MPa)

Penetration energy (J/mm)

31.0 31.3 31.2

46.35 40.08 41.03

2.37 2.38 2.01

3758 3510 3685

1.58 1.32 1.41

30.6

47.20

2.54

3860

1.53

30.5

48.28

2.21

3945

1.84

LGF = Long-glass fiber.


272


(Günter and Möltgen, 1993). Furthermore upgrading of recyclates by glass fibers is for example used for PET bottle material in injection moulding applications (Lin, 1998). In model experiments glass fiber reinforced PBT (30% GF) and PPS (40% PPS) were aged under dry and humid conditions and the material characteristics of the recycled material were investigated by addition of virgin material and in the presence of coupling agents (van Lochem et al., 1996). Aging at high temperatures (150, 120 °C) resulted in a complete loss of impact strength and severe reduction of tensile strength and elongation due to decreasing molecular weight of the polymer. The mechanical properties were partially recovered by adding virgin material. However, the recyclate from the material, which was aged under humid conditions, resulted in poor mechanical properties and poor interface strength. The results with PPS were quite similar, aging and recycling reduced substantially the impact strength, tensile strength and elongation with some improvements by adding coupling agents of silane or zirconate type. The authors emphasized further that these additives improve the fiber–matrix adhesion. The recyclability of PBT glass fiber composites was investigated with regard to processing (injection molding, extrusion compression molding, compression molding) in the presence of silanes containing glycidyl (or vinylbenzyl/amine groups) with concentrations of 0.5% or 1.5%. Compared with a virgin reference material, the recyclate from regrinded plaques showed better impact strength, comparable modulus, lower tensile strength and elongation (Chu and Sullivan, 1996). Addition of the silane coupling agent improved, independently of the processing method, most mechanical properties, as outlined in the example from injection molding trials (Table 10.7). Finally fibers from glass–polyester composites and carbon or aramide fibers from epoxy-based laminates exemplify the reuse of thermoset materials. The thermosets were ground and fibrous fractions obtained, which were used as reinforcing agents in polypropylene and ionomers (Kouparitsas et al., 2002). Tensile testing of these materials indicated that mechanical properties (tensile strength, modulus of elasticity, elongation) were obtained, Table 10.7 Properties of glass fiber reinforced PBT recyclate

Composition Reference PBT recyclate PBT recyclate + silane coupling agent



Maximum elongation (%)

Notched Izod impact strength (J/m)

126.4 112.2 131.9

10.2 10.0 10.2

3.5 1.5 1.8

86.1 93.7 115.2


273

which are close to virgin fiber composites. In most cases residues of the recycled fiber matrix did not negatively influence the performance. Furthermore, phenolic prepreg waste was used to stabilize polypropylene and polyamide-6 (Gröning et al., 2006). There the phenolic groups of the functional filler act as antioxidants and extend the lifetime of the polymers.


10.5

Conclusions and future trends

Numerous references and commercial use prove the beneficial effect of additives in mechanical recycling of a wide variety of plastics and plastic mixtures. Stabilizers, reactive molecules and compatibilizers are the most frequently used classes of additives mentioned. Stabilizers improve processing and maintain the long-term properties, reactive molecules repair predegradation in some instances and compatibilizers enhance the mechanical properties of polymer mixtures. Contrary to the well-known benefits of additives in pure plastic recyclates, the literature on plastic composite recyclates based on fillers or fibers is rather scarce. However, it could be shown that stabilizer systems containing filler deactivators improve the lifetime of recycled composites and allow a closed loop recycling. Glass fiber reinforced composites partly lose their properties during aging and regrinding. Compatibilizers, coupling agents and adding new glass fibers are reported to improve the mechanical properties again to render these composites recyclable. Nevertheless the mechanical recycling of plastic composites stays a complex field as the interactions of all components, i.e. the polymer, the filler or fiber and the additives, have to be considered. Therefore, there is still a vast field for further research. Moreover, with the increased use of new materials, e.g. nanocomposites, additional questions on recycling will have to be addressed in the near future.

10.6 •

List of additives mentioned in Sections 10.3 and 10.4

AO-1: Ciba® Irganox® 245 Ethylenebis(oxyethylene)bis-(3-(5-tert-butyl-4-hydroxy-m-tolyl)propionate)

O HO

(CH2)2 C O (CH2)2

O

C H2 2

274


• AO-2: Ciba® Irganox® 1010 Pentaerythritol tetrakis(3-(3,5-di-tert-butyl-4-hydroxyphenyl)propionate

OH

O O

O O

HO

O

O

OH O


O

OH

• LS-1: Ciba® Tinuvin® 791, blend of Ciba® Tinuvin® 770 Bis(2,2,6,6-teramethyl-4-hydroxy-piperidyl)sebacate

O

O HN

and

O

C

(CH2)8

C O

NH


275

Ciba® Chimassorb® 944 (poly-6-(1,1,3,3-tetramethylbutyl)amino-1,3,5-triazine-2,4-diyl(2,2,6,6tetramethyl-4-piperidinyl)imino-1,6-hexanediyl(2,2,6,6-tetramethyl-4piperidinyl)imino)

N

C6H12

N

N N

N

N

H

H

N NH

n


• MD-1: Ciba® Irganox® MD-1024 2′,3-bis[[3-[3,5-di-tert-butyl-4-hydroxyphenyl]propionyl]] propionohydrazide

O (CH2)2

HO

C

N H

2

• PS-1: Ciba® Irgafos® 168 Tris(2,4-ditert-butylphenyl)phosphite

O O

P O

276


• RM-1: Ciba® Irgamod® 195 Calciumdiethylbis[[[3,5-bis(1,1-dimethylethyl)-4-hydroxyphenyl]methyl] phosphonate]

Ca HO

CH2

O P

2+

O

OC2H5

2

RS-1: Ciba® Recyclostab® 411, proprietary composition consisting of antioxidants, phosphites and costabilizers • RS-2: Ciba® Recyclostab® 451, proprietary composition consisting of antioxidants, phosphites and costabilizers • RS-3: Ciba® Recycloblend® 660, proprietary composition consisting of antioxidants, oxiranes, phosphites and costabilizers • RS-4: Ciba® Recyclossorb® 550, proprietary composition consisting of antioxidants, phosphites, hindered amine stabilizers and costabilizers


•

10.7

References

allen n s, edge m, corrales t, childs a, liauw c m, catalina f, peinado c, minihan a, aldcroft d, Ageing and stabilisation of filled polymers: an overview, Pol Degr Stab 1998, 61, 183–199 aubert m, roth m, pfaendner r, wilen c e, Azoalkanes: a novel class of additives for cross-linking and controlled degradation of polyolefins, Macromol Mater Eng 2007, 292, 707–714 awaja f, pavel d, Recycling of PET, Eur Pol J 2005, 41, 1453–1477 bizarria m t m, giraldi a l f m, carvalho c m, velasco j i, avila m a, mei l h i, Morphology and thermomechanical properties of recycled PET-organoclay nanocomposites, J Appl Pol Sci 2007, 104, 1839–1844 boldizar a, gevert t, markinger m, Quality assessment of polymeric recyclates. Part 1: repeated processing and thermooxidative ageing of LDPE, Polymer Recycling 1995, 1, 249 boldizar a, jansson a, gevert t, möller k, Simulated recycling of post-consumer high density polyethylene material, Pol Degr Stab 2000, 68, 317–319 bonelli c m c, martins a f, mano e b, beatty c l, Effect of recycled polypropylene on polypropylene/high density polyethylene blends, J Appl Pol Sci 2001, 80, 1305–1311 cavalcanti f n, teofilo e t, rablello m s, silva s m l, Chain extension and degradation during reactive processing of PET in the presence of triphenyl phosphite, Pol Eng Sci 2007, 47, 2155–2163 chu j, sullivan j l, Recyclability of a glass fiber poly(butylene terephthalate) composite, Polymer Composites 1996, 17, 523–531



277

cruz c a, Impact modifiers: (2) Modifiers for engineering thermoplastics in Pritchard G (ed.), Plastics Additives, An A–Z Reference, London, Chapman & Hall, 1998, 386–397 datta s, lohse d j, Polymeric Compatibilizers, Munich, Hanser, 1996 elmaghor f, zhang l, li h, Recycling of high density polyethylene/poly(vinylchloride) polystyrene ternary mixture with the aid of high energy radiation and compatibilizers, J Appl Pol Sci 2003, 88, 2756–2762 espert a, camacho w, karlson s, Thermal and thermomechanical properties of biocomposites made from modified recycled cellulose and recycled polypropylene, J Appl Pol Sci 2003, 89, 2353–2360 fellahi s, boukobbal s, m’hala m, Additives as properties improvers for recycled mixed plastics, Annu Tech Con-Soc Plastics Eng 1991, 2170–2174 forsythe j s, cheah k, nisbet d r, gupta r k, lau a, donovan a r, o’shea m s, moad g, Rheological properties of high melt strength poly(ethylene terephthalate) formed by reactive extrusion, J Appl Pol Sci 2006, 100, 2646–3652 gioffre a j, marcus b k, Process for eliminating organic odors and composition for use therein, US patent 4795482, 1989 greco r, Impact modifiers: (1) mechanisms and applications in thermoplastics in Pritchard G (ed.), Plastics Additives, An A–Z Reference, London, Chapman & Hall, 1998, 375–385 gröning m, eriksson h, hakkaraainen m, albertsson a c, Phenolic prepreg waste as functional filler with antioxidant effect in polypropylene and polyamide-6, Pol Degr Stab 2006, 91, 1815–1823 grützner r e, gärtner r, hock h g, PE-PA Verbundfolienreste. Eigenschaften und Grenzen der Verwertbarkeit, Kunststoffe 1993, 83, 369–372 günter j, möltgen b, Verfahren zur Wiederverwertung von Kunststoffabfällen oder Kunststoffaltmaterial zur Herstellung von Kunststoffbauteilen und deren Verwendung, German Patent DE 4230157, 1993 gugumus f, Effect of temperature on the lifetime of stabilized and unstabilized PP films, Pol Degr Stab 1999, 63, 41–52 gugumus f, Light stabilizers in Zweifel H (ed), Plastics Additives Handbook, Munich, Hanser, 2001, 141–419 hamskog m, klügel m, forsström d, terselius b, gijsman p, The effect of base stabilization on the recyclability of polypropylene as studied by multi-cell imaging chemiluminiscence and microcalorimetry, Pol Degr Stab 2004, 86, 557–566 hamskog m, klügel m, forsström d, terselius b, gijsman p, The effect of adding virgin material or extra stabilizer on the recyclability of polypropylene as studied by multi-cell imaging chemiluminescence and microcalorimetry, Pol Degr Stab 2006, 91, 429–436 hausmann k, Verbundfolien wiederverwerten mit Verträglichkeitsvermittlern, Kunststoffe 1995, 85, 446–451 heberer s, kunz f r, lortz b m, maier h j, wartusch r, Verfahren zur Adsorption von Gerüchen, German Patent Application DE 10062558, 2002 hermann a, reimer u, hoecker f, martin m, bosse a, jerg r, weiβhappel h, pfaendner r, vennemann n, bledzki a k, Stabilisierung – Doping für Rezyklate, Kunststoffe 2000, 90, 80–83 hinsken h, moss s, pauquet j r, zweifel h, Degradation of polyolefins during melt processing, Pol Degr Stab 1991, 34, 279–293 hohenberger w, Fillers and reinforcements/coupling agents in Zweifel H (ed), Plastics Additives Handbook, Munich, Hanser, 2001, 901–948


278


hong s m, hwang s s, choi j s, choi h j, Compatibility effect of reactive copolymers on polypropylene/polyamide 6 blends from commingled plastic wastes, J Appl Pol Sci 2006, 101, 1188–1193 hope p s, bonner j g, miles a f, A study of compatibilisers for recovering the mechanical properties of recycled polyethylenes, Plastics, Rubber Composites Proc Appl 1994, 22, 147–158 hu x, xu h, zhang z, Influence of fillers on the effectiveness of stabilizers, Pol Degr Stab 1994, 43, 225–228 jansson a, möller k, hjertberg t, Chemical degradation of a polypropylene material exposed to simulated recycling, Pol Degr Stab 2004, 84, 227–232 kaci m, benhamida a, cimmino s, silvestre c, carfagna c, Waste and virgin LDPE/ PET blends compatibilized with ethylene-butyl acrylate-glycidyl methacrylate (EBAGMA) terpolymer, 1: Morphology and mechanical properties, Macromol Mater Eng 2005, 290, 987–995 kallel t, massardier-nageotte v, jaziri m, gerard j f, elleuch b, Compatibilization of PE/PS and PE/PP blends. I. Effect of processing conditions and formulation, J Appl Pol Sci 2003, 90, 2475–2484 kartalis c n, papaspyrides c d, pfaendner r, Closed loop recycling of postused garden chairs based on PP using the restabilization technique. I. Evaluation of processing parameters, J Appl Pol Sci 2002, 86, 2472–2485 kartalis c n, papaspyrides c d, pfaendner r, Closed loop recycling of postused garden chairs based on PP using the restabilization technique. Part 2: Material performance during accelerated heat aging, J Appl Pol Sci 2003a, 88, 3033–3044 kartalis c n, papaspyrides c d, pfaendner r, Closed loop recycling of postused garden chairs based on PP using the restabilization technique. III. Influence of artificial weathering, J Appl Pol Sci 2003b, 89, 1311–1318 khan j h, ahmed n, Photo-oxidative degradation of recycled, reprocessed HDPE: changes in chemical, thermal and mechanical properties, J Mat Sci Tech 2001, 9, 153–164 kiliaris p, papaspyrides c d, pfaendner r, Reactive extrusion route for the closedloop recycling of poly(ethylene terephthalate), J Appl Pol Sci 2007, 104, 1671–1678 kim g h, hwang s s, cho b g, hong s m, Reactive extrusion of polypropylene and nylon blends from commingled plastic wastes, Macromol Symp 2007, 249/250, 485–492 kouparitsas c e, kartalis c n, varelidis p c, tsenoglou c j, papaspyrides c d, Recycling of the fibrous fraction of reinforced thermoset composites, Polymer Composites 2002, 23, 682–689 la mantia f p, The role of additives in the recycling of polymers, Macromol Symp 1998, 135, 157–165 la mantia f p, tzankova dintcheva n, morreale m, vaca-garcia c, Green composites of organic materials and recycled post-consumer polyethylene, Polym Int 2004, 53, 1889–1891 la mantia f p, morreale m, Mechanical properties of recycled polyethylene ecocomposites filled with natural organic fillers, Pol Eng Sci 2006, 46, 1131–1139 la mantia f p, tzankova dintcheva n, malatesta v, pagani f, Improvement of photo-stability of LLDPE-based nanocomposites, Pol Degr Stab 2006, 91, 3208–3213



279

lei y, wu q, clemons c m, Preparation and properties of recycled HDPE/clay hybrids, J Appl Pol Sci 2007, 103, 3056–3063 lemmens j, Compatibilizers for plastics, in Brandrup J, Bittner M, Michaeli, W Menges G (eds.) Recycling and Recovery of Plastics, Munich, Hanser, 1996, 315–326 lin c c, Recycling technology of poly(ethylene terephthalate) materials, Macromol Symp 1998, 135, 129–135 luzuriaga s, kovarova j, fortelny i, Degradation of pre-aged polymers exposed to simulated recycling: properties and thermal stability, Pol Degr Stab 2006, 91, 1226–1232 mangaraj d, Compatibilization of polymer recyclates in Srinivasan K S V (ed.) Macromolecules New Frontiers Vol. II, New Delhi, Allied Publishers Ltd, 1998, 705–709 marrone m, la mantia f p, Re-stabilisation of recycled polypropylenes, Polymer Recycling 1996, 2, 17–26 martins m h, de paoli m a, Polypropylene compounding with post-consumer material: II Reprocessing, Pol Degr Stab 2002, 78, 491–495 mehrabzadeh m, farahmand f, Recycling of commingled plastics waste containing polypropylene, polyethylene and paper, J Appl Pol Sci 2001, 80, 2573–2577 miller n a, jones m s, stirling c d, Waste plastics/cellulose fibre composites, Polymers and Polymer Comp 1998, 6, 97–102 obieglo g, romer k, Compatibilizer: Ein Schlüssel zum Recycling, Kunststoffe 1993, 83, 926–932 oyman z o, tincer t, Melt blending of poly(ethylene terephthalate) with polypropylene in the presence of silane coupling agent, J Appl Pol Sci 2003, 89, 1039–1048 papadopoulou c p, kalfoglou n k, Comparison of compatibilizer effectiveness for PET/PP blends: their mechanical, thermal and morphology characterization, Polymer 2000, 41, 2543–2555 pawlak a, morawiec j, pazzagli f, pracella m, galeski a, Recycling of postconsumer poly(ethylene terephthalate) and high-density polyethylene by compatibilized blending, J Appl Pol Sci 2002, 86, 1473–1485 pfaendner r, herbst h, hoffmann k, sitek f, Recycling and restabilization of polymers for high quality applications: an overview. Angew Makromol Chem 1995, 232, 193–227 pfaendner r, Additives for mechanical recycling of plastics in Zweifel H (ed), Plastics Additives Handbook, Munich, Hanser, 2001, 973–1016 pfaendner r, How will additives shape the future of plastics?, Pol Degr Stab 2006, 91, 2249–2256 pfaendner r, herbst h, hoffmann k, Increasing the molecular weight of polyesters, US Patent 5693681, 1997 pfaendner r, hoffmann k, herbst h, Increasing the molecular weight of polyamides, US Patent 5756596, 1998a pfaendner r, hoffmann k, herbst h, Increasing the molecular weight of polyesters and premix useful for this process, US Patent 5747606, 1998b pfaendner r, hoffmann k, herbst h, Increasing the molecular weight of polycondensates, US Patent 5807932, 1998c pfaendner r, hoffmann k, herbst h, Process for increasing the molecular weight of polycondensates, US Patent 6706824, 2004


280


pramanik p k, dickson b, Thermoplastic composite from recycled plastic and wood flours, Ann Tech Conf Soc Plat Eng 1997, 55, 3136–3140 regel k, andernach r, schwarz p, herbst h, hoffmann k, pfaendner r, simon d, Increasing the molecular weight of polyesters, US Patent 6265533, 2001 roth m, pfaendner r, nesvadba p, zink m o, Process for reducing the molecular weight of polypropylene, US Patent 7030196, 2006 roth m, pfaendner r, nesvadba p, zink m o, Process for the controlled increase in the molecular weight of polyethylenes, US Patent 7358365, 2008 sahnoune f, lopez cuesta j m, crespy a, Improvement of the mechanical properties of an HDPE/PS blend by compatibilization and incorporation of CaCO3, Pol Eng Sci 2003, 43, 647–660 scaffaro r, tzankova dintcheva n, la mantia f p, On the effectiveness of different additives and concentrations on the re-building of the moelcular structure of degraded polyethylene, Pol Degr Stab 2006, 91, 3110–3116 scaffaro r, la mantia f p, tzankova dintcheva n, Effect of the additive level and of the processing temperature on the rebuilding of post-consumer pipes from polyethylene blends, Eur Pol J 2007, 43, 2947–2955 schrader h g, Additiv zur Einbringung von homogen verteiltem Wasser in polymere Werkstoffe und Verfahren zur Anwendung des Additivs, German Patent Application DE 10335486, 2004 schrader h g, Verfahren zur Minderung von Schadstoffemissionen und/oder geruchsemittierenden Substanzen in polymeren Werkstoffen, European Patent Application EP 1921108, 2008 simon d, pfaendner r, herbst h, Molecular weight increase and modification of polycondensates, US Patent 6469078, 2002 singh b, sharma, n, Mechanistic implications of plastic degradation, Pol Degr Stab 2008, 93, 561–584 tang x, guo w, yin g, li b, wu c, Reactive extrusion of recycled poly(ethylene terephthalate) with polycarbonate by addition of chain extender, J Appl Pol Sci 2007, 104, 2602–2607 thompson m r, yeung k k, Recyclability of a layered silicate-thermoplastic olefin elastomer nanocomposite, Pol Degr Stab 2006, 91, 2396–2407 ulutan s, Influence of additional thermal stabilizers on the reprocessing of postconsumer poly(vinyl chloride) bottles, J Appl Pol Sci 2003, 90, 3994–3999 vanhaeren g, de groote p, godard p, Styrene block copolymers in mixed plastics 2. Compatibilization of post-consumer commingled plastics, Chimie Nouvelle 1997, 15, 1659–1664 van lochem j h, henriksen c, lund h h, Recycling concepts for thermoplastic composites, J Reinforced Plastics Comp 1996, 15, 864–876 vilaplana f, karlsson s, Quality concepts for the improved use of recycled polymeric materials: a review, Macromol Mater Eng 2008, 293, 274–297 vinci m, la mantia f p, Properties of filled recycled polypropylene, J Pol Eng 1997, 16, 203–215 vouyiouka s n, papaspyrides c d, pfaendner r, Catalyzed solid-state polyamidation, Macromol Mater Eng 2006, 291, 1503–1512 wernicke m, skaletz d, Recycling a fiber-reinforced thermoplastic, US Patent 5660770, 1997 zweifel h (ed.), Plastics Additives Handbook, Munich, Hanser, 2001

11 Improving the mechanical recycling and reuse of mixed plastics and polymer composites K. TA RV E R D I, Brunel University, UK

Abstract: In this chapter we will explore the possibilities and viability of commercially recycling and reusing of composites and mixed polymer based waste that is difficult and expensive to separate into individual generic polymers for subsequent recycling. The chapter will also attempt to cover legislations that hinder and drive this technology and give up-to-date sources of information for further examples of polymer recycling in action.


Key words: mixed polymer waste, composite, recycling, reuse, plastics, legislation.

11.1

Introduction

It can be very pleasing to the eye, to the pocket and for the environment when piled mixed plastic, so-called ‘rubbish’ ready for landfill, is converted to building blocks, hoardings, park benches and many other useful, longlasting products. Recycling and reuse of materials is by no means a new concept, since over the past three decades and earlier waste newspaper, cardboard, white paper and glass have been recycled and reused. However, other items such as polymer-based products gained acceptance and momentum towards the end of the 1970s as a result of the significant increase for that time, in raw materials costs because of the unexpected rise in the price of crude oil. Recycling of polymer composite is an even more recent occurrence, with significant work generated in the latter half of the 1980s. However, in recent years there has been an increase in the use of composite products, particularly in the automotive and construction industries. These consume nearly half of all composites manufactured, therefore the issue of composite recycling and use is becoming very important. Successful composite recycling and use requires incentives, infrastructure, good recycling techniques and markets for the recyclates. Plastics recycling is of growing importance and interest to the public in general and to governments, in particular because of EU directives, 281

282


stipulations and fines. There is no end or beginning to recycling efforts since they are essential to help save and minimise the use of non-renewable resources and reduce landfill practices. There is no universal solution to the problem of mixed plastics recycling. A hierarchy of complementary conservation and disposal techniques is needed to solve the escalating problem of mixed plastic waste and plastic composites that are difficult to sort and separate. Plastics recycling involves a series of processing operations carried out to eventually produce secondary materials for the manufacture of different types of products.


11.2

Thermoplastic and thermosetting polymers

The vast majority of plastics used commercially are thermoplastic polymers. These cover all the commodity materials that are used such as: polyethylene (PE), polypropylene (PP), polystyrene (PS), polyvinyl chloride (PVC), polycarbonate (PC) and polyethylene terephthalate (PET). For a plastic material to have useful properties the molecules must be of a minimum chain length specific to individual polymers, which is expressed in terms of molecular weight which generally falls in the range between 100 000 and 1 000 000 and in the case of PET it is expressed in terms of intrinsic viscosity (IV). The property of the component will naturally depend on the chain length and methods of recycling will affect the chain length adversely (Brandrup, 1995; Goodship, 2007). The interaction of polymers is affected by forces of mutual attraction between individual chains and by molecular length, and the strength of the chain attractions depends on the type of structure. The strongest forces are exerted by polar groups, and a good example is demonstrated by PVC. Hence, the polar C–Cl group produces a relatively strong electric field that attracts neighbouring molecules that also have polar groups which gives PVC a high softening temperature, also called the glass transition temperature (Tg). The Tg is the temperature at which, on cooling from the melt, the material freezes to a glassy, brittle state. Below Tg much of the thermal motion of the molecules is quenched since the attractive forces between neighbours are sufficient to demobilise the chains. When a polymer is subjected to heat and shear, the molecules are capable of sliding past each other, overcoming the intermolecular attractions, thus above a certain temperature such forces have enough energy to move freely and overcome intermolecular attractions, essentially those types of materials that are thermoplastic. It is, however, possible to produce materials with branched chain structures, which can be linked together and shaped using heat and/or catalysts with chemical reactions, leading to a cross-linked three-dimensional product. These materials cannot be reheated and shaped once the polymer has cross-linked and cured, and they are termed


Improving the mechanical recycling and reuse of mixed plastics

283

‘thermosetting plastics’. Some important commercial examples include phenolic, epoxy, polyester, polyurethane and formaldehyde resins. These types of material have to be recycled in a completely different method compared with thermoplastic polymers (RECCOMP Project, 2008; Palmer et al., 2009a,b) (RECCOMP–Recycling of Composites). The development of integrated polymer compounding technology has been instrumental in establishing an effective strategy for gaining value from thermoset waste. Polyester and phenolic scrap comes from car components and other industrial thermoset waste materials. The principal aim of this technology has been to assess the reinforcing potential of such materials after comminution and subsequent inclusion into selected polymer matrices (Bream et al., 1997a,b). The successful reuse of thermosetting materials depends on the capability of reducing the size of the thermosetting waste to a level at which it is suitable as a functional filler. The glass fibre content, fibre length distribution, particle morphology and the addition of virgin fibres are important factors in determining the reinforcing properties of the filler in composites of the recyclate with base polymers. Application of this methodology to the preparation of PP composites reinforced with waste thermosetting composites has demonstrated that marked increases in mechanical properties can be obtained when fibre to polymer interfacial bonding is achieved (Bevis et al., 1996; Bream and Hornsby, 2001). The EU End-of-Life Directive requires that 95% of a vehicle by weight is to be capable of recovery or reuse by 2015. This target cannot be reached by metal recycling alone, which only accounts for 76% by weight of the average passenger car. Plastic parts represent the next largest fraction at 9–12%. Therefore plastics recycling must be considered if the End-of-Life Vehicle (ELV) targets are to be met (End-of-Life Vehicles Regulations 2003; ELV Directive, 2000, 2006a). Sheet and bulk moulding compounds (SMC/BMC) form the major part of the plastic fraction used on vehicles. The use of these materials is increasing, and now includes exterior body panels providing the class A finish and simple manufacture not achievable by competing thermoplastic polymers (Palmer et al., 2009b). Directive 2000/53/EC on End-of-Life Vehicles (ELV Directive, 2000, 2006a) has been effective since 2005. The Directive aims at reuse, recycling, recovery of ELV and components. Recycling of all plastics from end-of-life vehicles should be continuously improved, and development of markets for recycled materials is being encouraged by the UK Government and the European Commission (Table 11.1).

11.2.1 Degradation of polymers During primary processing and after use when polymers are being recycled, secondary degradation takes place and the molecular length of the polymer

284


Table 11.1 Recycling and recovery timings (ELV Directive, 2000, 2006)

Timing


Old vehicles (produced) before 1980 No later than 1 Jan 2006 No later than 1 Jan 2015

Recycling and reuse

Recovery and reuse

Disposal (landfill)

Energy recovery

70%

75%

25%

5%

80% 85%

85% 95%

15% 5%

5% 10%

chain is reduced by mechanical and thermal decomposition. The degraded polymer molecular chains are less capable of linking with one another. Also, the melting point of the composite becomes lower and the general physical properties are reduced (Brandrup, 1995). As molecular length is reduced by degradation (thermal or physical) which can be caused not only by recycling but, above all, by ageing, this suggests that the sections of molecule projecting from a chain tangle are the first to be attacked and broken off, particularly with semicrystalline polymers. Thus, degraded molecules are less capable, or are even incapable, of linking by bonding together with the lowering of the molecular weight, melting point and reduction of toughness in the solid state and the material behaves in a brittle fashion on impact. The Melt Flow Index (MFI) which is the measured gravimetric flow rate of the thermoplastic melt extruded from a die of specific length and diameter, under given conditions of pressure and temperature, is generally higher than the starting materials and also the impact strength is diminished. This is not universal, however, since the MFI of high density polyethylene (HDPE) initially decreases as the material is recycled. There is a growing demand to develop and use new products containing high percentages from sustainable, renewable and natural resources that provide properties and performances equivalent to or better than synthetic polymers. Some of the leading polymers competing for demand are: polylactic acid (PLA), modified starches, and polyhydroxyalkanoates (PHA) (Xia et al., 2007; Bagrodia, 2008). However, these materials primarily degrade by microbial attack rather than thermal or physical breakdown.

11.3

Polymer composites

Polymer composites consist of either a thermoset or thermoplastic polymer, along with inorganic fillers such as calcium carbonate and glass fibres, which together build up the structure of the composite. The physical properties are strongly dependent on the chemical structure and ratio of individual groups in the plastic as well as on the fillers employed. These have a very



285

large influence on the overall spectrum of properties of the polymer composite which includes the additives for the final composite formulation. Polymer composites are extensively used in the manufacture of lightweight, flame retardant and abrasion resistant components for the automotive, aeronautical industries and household goods. Composite materials made of various plastics and additives include: textile-reinforced plastic hoses, glass and natural fibre reinforced sheeting, metal reinforced profiles for windows, carpet, cable wastes, tyres, composite bumpers and many more. These composites generally end up in landfill, but there are modern sorting and recycling methods developed to divert thousands of tonnes of composites from landfill to recycling for the manufacture of new and novel composite materials (RECCOMP Project 2008; PIM Project, 2009). All composite waste materials need to be size reduced prior to recycling by using slitting rollers, hydraulic shears and shredders followed by post-size reduction using robust granulators some with rotors designed to cut through metalcontaining parts. Such parts can be reduced in size between 2 and 6 mm which is ideal for the subsequent separation process to take place. These include magnetic, gravity cascade, air and cross-flow separators to separate metal, fibre and some fractions of polymers. Ideally it would be best just to size reduce and reconstitute the mixed plastic composite waste into a new composite and the PIM process, which is described in Section 11.6, was designed to bring together the fine ground composite by applying thermal energy and compression force sandwiching the waste composite between single polymer sheeting (PIM Project, 2009).

11.3.1 Composites and mixed plastic waste recycling To manufacture composites for a particular component use, the individual ingredients that make up the composite are tailor-made to suit the purpose for which the component is intended. However, for efficient recycling of composites and mixed plastic wastes, the mechanical properties obtained are usually inferior owing to the separation of phases and incompatibility issues, as molecules repel each other and phase separation occurs. If, therefore, the second polymer is not very finely and uniformly dispersed in the first, it causes precipitates that have no connection with the surrounding phase, generally observed using light and electron microscopy. Different polymers are usually mutually incompatible, i.e. if one phase is well dispersed in the second phase, the particles can be bound to the first by compatibilisers (molecules that join the phases) to attain exceptional properties, in particular high impact strength and good optical surfaces. Attempts to apply this method to plastic wastes are also being made, and the prospects of obtaining useful products are likely to be good if uniform dispersion on a large scale can be guaranteed; mixtures of mutually incompatible

286


polymers can be made more impact resistant by this means. However, accurately defined and uniform polymer composites are required, with a well-dispersed second phase (Bream and Hornsby, 2001).

11.4

Materials recycling

In general, the mechanical and physical properties of most polymeric materials are quite similar, but these properties are enhanced when blending with other organic and inorganic polymers, fillers and fibres to make composites. Discussions about composites recycling and reuse are difficult without basic knowledge and understanding about composite behaviour during its first life cycle and the changes envisaged during recycling and eventual reuse. It is difficult to reprocess with just a single source of polymeric waste containing different polymer grades and molecular weights and to try to reconstitute polymer composite waste is naturally even more complex (Goodship, 2007).


11.4.1 Mixed plastic and polymer composites recycling Plastic waste is generated during production and processing of components but at this stage plastics can be mixed in controlled amounts with the virgin polymers and put back into the processing line. Once the products get into the marketplace, for example as packaging or as functional components, they are disposed of, and their end of use has been dealt with by collection and disposal in landfill as mixed plastic waste. If, however, this type of plastic waste is easy to identify and sort, then recycling can be cost effective. Generally, though, the separation techniques used cannot distinguish and separate all the plastics and it could then become very expensive and labour intensive to physically hand separate them. As a result of this, the bulk of the mixed waste plastics are either collected and land-filled, stored for future separation or sent overseas to be hand sorted and sold for low value recycling or disposal. In some instances, it is viable to identify and separate the easily separable polymers such as HDPE for high value use, while the rest of the fractions are sold as cheap scrap or sent for landfill because the value of this is below the general £80/tonne threshold. The UK uses about 7 million tonnes of plastic per annum in a very wide range of applications. Of the annual total, approximately 1 million tonnes are recycled, including 500 000 tonnes of plastic packaging recovered under the Packaging Waste Regulations (Producer Responsibility Obligations Packaging Waste Regulations, 2007). Figure 11.1 gives an indication of plastics in household waste in the UK. Recycling of polymers from a single source has become quite well established and more recent developments include

Improving the mechanical recycling and reuse of mixed plastics Bags 14%

287

Rigids 21%

Films 23%

Other 15%

~1.9 mt Bottles 27%


11.1 An indication of plastics in household waste in the UK in 2007 (source: WRAP).

closed loop recycling techniques, such as the HDPE milk bottle and the PET bottle schemes where companies like Coca Cola are signing up to using the recycled bottles with their virgin PET, and Vannplastic in the UK with its Ecodek label are successfully extruding large tonnages of recycled polyolefins with high loadings of wood fibre for decking and many other applications. But when it comes to recycling of mixed waste with composites, to establish a closed loop system will not only be more difficult, it will also be impractical since the virgin composite users would find it difficult to use recycled composites with diminished properties as a consequence. However the composite waste could have a high value as a recyclate provided the collection, sorting and recycling is well conceived (www. closedlooprecycling.co.uk, www.vannplastic.co.uk).

11.4.2 Collection and sorting Collection and sorting economics, particularly within the domestic waste stream, for plastic packaging from municipal waste streams of identifiable items is becoming streamlined and sophisticated. For example, from the 2008 WRAP report and survey analysis, the total quantity of plastic bottles collected in the UK in 2007 was 181 887 tonnes. This is a considerable increase of approximately 68% on the 2006 total quantity of 108 453 tonnes. The number of local authorities offering plastic bottle collections this year was recorded as 437 or 92%, where estimates based on respondents returns to last year’s survey are included.


288


Approximately 525 300 tonnes of plastic bottles were consumed in households throughout the UK in 2007. If the total quantity of plastic bottles actually collected in 2007 was 181 887 tonnes, then approximately 35% of plastic bottles that were consumed by households were collected. Using this consumption figure, the trend indicates collection rates of 50%, 71% and 94% in 2008, 2009 and 2010, respectively, but the likely outcome is that the trend will plateau at around the projected 2009 rate of 71% as the growth in the number of collections slows (WRAP 2008). Collections of mixed plastic waste using municipal waste streams are well organised but the separation and identification of the plastics into different generic polymers are difficult, expensive and time consuming. There are mixed plastic conversion methods that are being used to minimise sorting and extensive cleaning, but as yet most of the methods being used are at their prototype and experimental stages. Even when collections of mixed bottles alone are targeted in their plastic bottle collection schemes, the bottles are most commonly made from HDPE and PET and are the type of bottles currently of highest value to local authorities. However, owing to the system of household collections it is not possible to receive only the desired materials, due to human error and the difficulty of identifying them. From an analysis of plastic bottles received by Valpak Northwest’s Materials Recycling Facility (MRF) in July to December 2007, a typical profile of plastic bottle collections in the UK was established. Whilst the majority of materials received (77%) were HDPE and PET bottles, there was approximately 23% that could be classified as contaminated which included materials such as aluminium and steel cans, and unsuitable plastics from labels and caps. With increasing recycling and landfill targets, more local authorities are investigating the potential of introducing new technologies to improve efficacy in waste management. This in turn has the potential to affect the way in which plastic bottles and other household plastics are handled. As part of the survey, local authorities were asked what plans they have, if any, for introducing new technologies for waste management and recycling. In order to optimise recycling and comply with new impending legislation, many local authorities have implemented other plastic collection schemes from homes and schools, including Waste Electrical and Electronic Equipment (WEEE) and agricultural plastics from farms (such as silage film) (WRAP 2008).

11.4.3 Limitations to mixed plastic reprocessing Unfortunately there is limited investment in technology to improve and increase efficacy and capacity of reprocessors, which has hindered mixed plastic reprocessing. Furthermore there has been a lack of sustained


289

competitive pricing for recyclates compared to virgin polymers. Also, prices are volatile and there is a lack of mechanisms to deal with these fluctuations.

11.4.4 Incentives for recycling Without government aid and incentives, it is difficult for recyclers to embark on developing machinery and technology for viable recycling methods. The UK Government, through TSB and WRAP, is investing in many projects that are directly involved in mixed waste recycling procedures (RECCOMP Project, 2008; PIM Project, 2009).


11.4.5 Quality assurance and status of regulations driving recycling The extent to which products are recycled and who takes responsibility for managing and collection of used products is influenced by regulatory requirements. The regulations that drive the process are either associated with virgin manufacture or the cost of waste. There are several European Directives and regulations that impact composite waste management, collection and recycling, for example: • • • • • • • •

Waste Framework Directives 2008/98/EC, 75/442/EEC, 91/156/EEC, 91/692EEC, 85/59/EC, 96/350/EC. Hazardous Waste Directives 91/689/EEC, 94/31/EC. Waste Decision 94/3/EC, European Waste Catalogue. Council Directive 1999/31/EEC on Landfill of Waste and Council Decision 2003/33. Directive 2000/76/EEC on the Incineration of Waste Integrated Pollution Prevention and Control. IPPC Directive 96/61EC Packaging and Packaging Waste Directive 94/62/EC. Directive 2000/53/EC on End-of-Life Vehicles. Directive 2002/96/EC on Waste of Electrical and Electronic Equipment, Directive 2002/95/EC on Hazardous Substances in WEEE.

The recovery and recycling of plastics waste, with reference to ISO 15270:2008, has been formulated to help plastics recyclers and collectors to meet a standard and the process can be summarised as follows: regulations, rules and standards that govern new plastics apply to all activities related to the recycling of used plastics, and their full life cycle. With recent legislations in place it is now easier to comply with the conditions applied to the recycling and reuse of plastics composites (www.iso.org, ec.europa. eu/environment/waste/legislation/a.htm). In this context it should be

290


mentioned that recyclers have to satisfy quality assurance legislations concerning the recycling of materials in the UK and EU that has been enacted to reduce burdens on landfill sites and to meet the targets for recycling set by the European Commission (www.defra.gov.uk/environment/waste/ europe/index).


11.4.6 Where can recycled plastics composites and mixed waste plastics be used? Plastic recyclates are not excluded in principle from being used in almost any application and including food packaging. There needs to be stringent legislation to control the quality of the recyclates and of the reprocessing plant. In addition, the recyclates have to satisfy existing regulations such as the EU Regulations 89/109, 90/128 and 92/39 and many more since. An international standard designed to assist in the development of a worldwide market for plastics recovery and recycling was launched in July 2008. The standard ISO 15270:2008, ‘Plastics – Guidelines for the recovery and recycling of plastics waste’, was compiled after input from relevant international industry bodies and stakeholder associations. It is hoped that the standard will help establish a sustainable infrastructure for recovery/recycling and a sustainable market for the manufactured products of the converted materials. As an example, the British Standards Institute (BSI) and Waste Resource Action Programme (WRAP) has created new standards for recycling the 46 million used tyres generated each year in the UK. Two Publicly Available Specification (PAS) documents have been developed to provide a clear guidance to the ‘measures and procedures needed to produce recycled material such as shred, crumb, powder and tyre bales to specific consistent grade and quality’. According to WRAP, the PAS documents, which set out voluntary specifications promise to be a useful resource in what is ‘an industry historically lacking guidance’. PAS 107 refers to the manufacture and storage of size-reduced tyres derived from rubber materials, while PAS 108 covers the production of tyre bales for use in construction (www. bsiglobal.com, wrap.org.uk).

11.5

Consumer protection

Consumer protection in most areas of use of recyclate products is well established and the same protection needs to be applied to the recycled composite applications, e.g. in areas of food packaging, milk and beverage containers, toys, cosmetics and pharmaceuticals. It is expected that, initially in these areas, composite recyclates will not be the targets for the consump-


291

tion of large percentages of composite recyclates. The prime initial target areas will be with packaging, equipment, support structures and pipes, etc.


11.5.1 Consequences for recycling Because of the escalating price of oil the temptation to use recycled composites to conserve the use of virgin materials is becoming more desirable. (At present (November 2008 to May 2009) owing to the exceptional circumstances that have led to the economic downturn, the price of oil and thus the prices of commodity plastics have fallen.) Thus, there is a risk that companies might try to cut corners and use the recyclate without the necessary proper evaluation and certification to keep costs down. There was a time when persuading companies to recycle their off-cuts was problematic and only legislation could persuade its use, but times are changing. Regulations relating to the recovery and use of polymeric composites must be followed at all times, and recycling plants have to be properly set up with approval of quality checks by local authorities. There needs to be producer liability, similar to that for the use and retailing of virgin or single source recycled materials. In some instances the polymeric composites should not be recycled but perhaps incinerated for energy recovery, as long as the plants meet with stringent regulations to prevent escape of toxic fumes and improper disposal of toxic ash. The UK Government has revised the recycling and recovery targets as part of the compliance with the Packaging and Packaging Waste Directive 2004/12/EC requiring that 25.5% of plastic be recycled in 2010. Plastics recycling from domestic and some industrial waste is currently uneconomical owing to the costs of the separation of the plastics and the low value of the markets for plastics recyclate. For mixed plastic waste to become a viable and market driven process, new end-use applications need to be developed, or processes that allow the manufacture of higher value products to use recyclate. For example, using the PIM process to manufacture boards (PIM Project, 2009). In addition, local authorities must recycle 30% of domestic and household waste by 2010 according to the EU implementation of the Landfill Directive 1999/31/EC. In 2002 little over 15 000 tonnes of plastic packaging waste of the domestic waste stream was recycled, from a potentially recoverable total of 1.5 million tonnes. The majority of plastics waste from the UK building and construction industry is land-filled and only very little is recycled. There are indications that there is over 570 000 tonnes per annum of potentially recoverable plastic composite waste arising in the construction waste stream (Waste Watch – Plastics in the UK Economy 2003; www. wastewatch.org.uk).


292


11.6

The powder impression moulding process

The powder impression moulding (PIM) (PIM Project, 2009) process is at present being further developed by the Wolfson Centre for Materials Processing, Brunel University and PERA with consortium partners involving producers, end users and collectors of mixed plastic waste such as Tesco and Bovis Lend Lease, Severnside Recycling and Environmental Recycling Technologies Plc (ERT). The PIM process should allow manufacture of returnable transit packaging (RTP) for the food and drinks industry and products for the construction industry, having up to 80% mixed plastic recycled content as the core material with no loss of aesthetics or functionality. The PIM process technology concept when fully developed has the potential to add significant value to the plastics recycling industry through the incorporation of up to 80% mixed plastic polymer waste into valueadded products. However, in order to realise the concept, there is a need to create a robust supply chain for processing the feedstock into PIM tolerant microparticles and to demonstrate the applications of PIM to gain industry confidence in the products that can be produced. In order to facilitate this venture the organised consortium of industry, university and research partners is undertaking a research project funded by TSB. The partners hope to develop the technology further and to evaluate new robust products of mixed plastic waste composites. Most companies are taking materials recycling, sustainability and decreasing carbon footprint issues very seriously. For example, Marks and Spencer is spending £200m over five years starting from January 2007, on a wide ranging ‘eco-plan’ and, in general, sustainability campaigners have welcomed the initiative as a progressive project of its kind by a mainstream retailer in the UK. Furthermore, a Bovis Lend Lease UK policy document on sustainability sets out the following goals by the end of 2010: reduce carbon emissions by 20% compared with 2008 levels, reduce waste to landfill by 70% compared to 2007 levels, deliver projects which are rated BREEAM ‘Excellent’ and encourage clients to aim for BREEAM ‘Very Good’ as a minimum, develop a procurement policy to ensure procurement of significant materials from sustainable sources. (www.breeam.org) (BREEAM is an environmental assessment method that can be used to assess the environmental performance of any type of building (new and existing)). The majority of polymer reprocessing is performed using polymer obtained from clean industrial sources such as LDPE from wrappings and sheeting, PP from transit packaging and HDPE and PET from crates and bottles. These sources require minimal reprocessing, contain low levels of contamination and yield high quality single polymer recyclate that is sold into the market as single source virgin polymer (wrap.org.uk). The recy-


293


11.2 Composites and mixed, contaminated domestic polymer waste is uneconomical and unattractive to recyclers. To separate this type of plastic waste for reuse is at present very expensive and time consuming (courtesy of Paul Gallen, APM Ltd).

clates of composites and mixed, contaminated or domestic polymer waste is uneconomical and unattractive to recyclers (see Fig. 11.2), because it requires extensive sorting and cleaning as part of the reprocessing cycle and most of this type of mixed waste polymers is land-filled. Very small quantities of this type of polymeric waste is used to manufacture products such as plastic lumber, and these products are of low value applications. During melt processing of the mixed plastic some polymers will melt, and others will remain solid, and polymers having different polar groups will tend to phase separate. Furthermore, the presence of large solid particles within the matrix increases the chances of stress cracking and poor performance.

11.6.1 Concept of the PIM process and products The concept of the PIM process (see Fig. 11.3), which uses micronised mixed plastic waste, involves applying a film of virgin polymer to the interior of a hot mould tool. The core material consisting of powdered mixed recyclate and blowing agent is then sprayed on the lower half of the mould tool. A lid, also coated with virgin or single source polymer, is then lowered onto the lower half of the mould and the halves of the mould are clamped together. The mould is then placed into a heat curing oven where the core material fuses together under pressure generated partly by the blowing agent, after which the mould is cooled and the product removed. This process does not require the complete melting of all the constitutive particles since the top and bottom virgin or single source polymer sheeting

294

Management, recycling and reuse of waste composites Co-mingled plastic waste is collected

These valuable products can themselves be recycled

Mixed plastic is turned into fine granules

ERT PIM process Highly usable products are made at a profit to the producer

Our process turns raw granules into reusable material


Material produced is then used as filler between two skins

11.3 Concept of the PIM process that uses micronised mixed plastic waste.

holds the core composite together and allows reinforcement of the main loading structure of the manufactured component. Depending on the final application, a wide range of mixed polymer waste including a percentage of inorganic fillers and fibres can be contained within the core of the composite. The mixed polymer composite waste can be utilised within the construction industry and returnable transit packaging (see Fig. 11.4) (PIM Project, 2009). The 2004 market for plastic boxes, crates, cases and similar articles for conveyance or packaging of goods was £444m (Office of National Statistics). RTP includes stacking containers, folding boxes and large containers, bottle crates and pallets. Major consumers within the industry are the supermarkets who had 15.4 million crates in use in 2000 and UK bakeries. The use of RTP by supermarkets is growing through the organic initiatives of their respective businesses and RTP use is also increasing through the organic food growers such as home shopping delivery. In Europe there are 70 suppliers of RTP to these customers, although the European and UK markets are dominated by fewer than 10 organisations (http://www. competition-commission.org.uk). The 2002 market for RTP was £517 million, falling to £509 million in 2003 and £444 million in 2004. This decline is not linked to a decrease in the demand for RTP, which continues to grow as new applications, such as


295


(a)

(b)

11.4 PIM processing techniques have been used to produce hoardings for construction sites. This trial site (see (a) and (b)) was set up by Bovis Lend Lease. The site was an award-winning replacement for external plywood sheet, with mixed waste polymer.

home delivery, are introduced and with environmental legislation driving greater reuse, but it can be attributed to rising material costs. In 2002 oil prices were about $23/barrel, in 2004 about $40/barrel, and in 2008 about $110–150/barrel, but today (May 2009) it is about $40/barrel. This represents a tangible opportunity for the development of a technology that allows the incorporation of using mixed plastic composite waste in the manufacture of these products. The 2005 market for building products for the construction industry was £30 billion (http://www.berr.gov.uk/files/ file21327.doc), of which ∼£30 million is represented by plastics products.

296


Over the last 25 years there has been a trend towards the increased use of plastics in the building industry, and it is predicted that in Western Europe the plastics use will increase to about 8 million tonnes by 2010 (http://www. berr.gov.uk/files/file21327.doc). This growth must be balanced against the challenges of implementing sustainable practice within the construction industry, since the energy consumed in building services accounts for ∼50% of CO2 emission.


11.7

Future technologies for converting mixed waste plastic (composites) into products

In a recent 2007 RECOUP report (www.recoup.org) it was shown that there is a £430 cost to the reprocessor for processing polypropylene trays into recyclate, of which £220 was associated with sorting and cleaning. Therefore it is essential to try to minimise sorting and cleaning prior to recycling. The manufacture of products from mixed plastic waste feedstock with appropriate converting technology such as the PIM technology will lead to reductions in the use of virgin polymers (∼450 tonnes/month) for manufacturing similar products, resulting in lower energy costs and the utilisation of non-renewable resources. The highest proportion of the energy required for the manufacture of products with virgin polymer is needed for the production of the resin feedstock. For example, 90.67% (81.5 kJ/kg) of the energy required to produce HDPE pipes relates to the production of the virgin polymer, whilst processing requirements represent only 7.93% (7.1 MJ/kg) of the total. There are many universities and companies researching and working in the development of mixed waste recycling techniques (e.g. Wolfson Centre for Materials Processing, Brunel University, the School of Engineering, Computer Science and Mathematics at Exeter University and research establishments such as PERA and Smithers RAPRA) for advancing and extending the possibilities of using mixed plastic waste recycling techniques. As a result, potentially thousands of tonnes of precious materials will be prevented from landfill or being sent overseas with substantial premium payments. The UK Government and EU Commission have major roles to play in making finances available for advancing and establishing mixed plastic waste technology. In the UK, WRAP and TSB are actively helping research establishments, universities and UK companies to achieve mixed waste recycling techniques. Moreover, an in-depth understanding and evaluation of required composites is being sought, through, for example the


297

Technology Strategy Board’s initiatives such as, High Value Manufacturing, the use of Sustainable Resources and Energy Saving concepts such as Energy Generation and Supply and the Ultra Efficient systems for the market advancement of electric and hybrid vehicles.

11.8

Case studies: recycling archives

11.8.1 The Construction Centre


The Construction Centre has an extensive database for architects and contractors to find recycled and sustainable products in order to help them conform to the Site Waste Management Plans (SWMPs) that came into effect from April 2008. The Construction Centre is one of the largest online resources for the building industry and has over 10 000 product manufacturers referenced on the website that can be found by searching over 97 000 product terms. The website also has additional information regarding building regulations, Local Authority planning offices, trade associations and industry publications and also includes UK directories of industry professionals, tradesmen contractors and merchants. (www.theconstructioncentre.co.uk, therenewableenergycentre. co.uk)

11.8.2 Epwin Group Building materials specialist, the Epwin Group, claims to have developed the first fully recycled PVC-U windows. The unit has been made out of waste from social housing refurbishment projects following the replacements of single-glazed first generation windows. Plastics Rubber Weekly (PRW) mentions that Epwin Group’s technical director David Wringley said that PVC-U is ‘hugely recyclable’ but the challenge was removing the old building debris prior to reprocessing. The recycling process uses a technical system that removes all traces of rubber, metal and glass and creates a near virgin quality material.

At the 2008 Eco-Build exhibition in London the recycled PVC-U window was exhibited.

11.8.3 Waste PVC-U recycling PVC-U sourced mostly from window and door fabricators is being recycled into what a leading wiring accessories manufacturer claims are the ‘greenest’ cable management products in the industry.

298


11.8.4 MK Electric MK Electric (www.mkelectric.co.uk) uses 100% recycled extruded lengths of PVC in its production process from its established supplier PVC group, a Recovinyl recycler. This means its systems offer the most recycled content of the comparable ones on the market. Overall, over 90% of the company’s cable management range, including skirting and trunking, is made from recycled plastics. Waste PVC-U, including off-cuts and bar lengths, is collected by PVC recyclers from fabricators across the UK and Ireland. It is then processed to remove all contaminants and ground into a high quality blended powder for re-use. PVC Group director Joanne Makin is said to comment that


as a market leader, MK Electric requires very high grade material to manufacture its products. We have been a long-term supplier to MK Electric for more than 20 years and are committed to supplying a high quality of recyclate using up to date state of the art technology, that enables us to meet the company’s quality requirements.

Using PVC recyclate supports MK’s commitment to sustainability works in two ways; first by diverting over 12 000 tonnes of PVC per year from landfill and, secondly, by preserving natural resources with associated savings in energy and providing a significant reduction in carbon emissions. Given the UK construction industry’s commitment to improve sustainability across the whole sector, sourcing 100% recycled products can truly demonstrate that ‘close-loop recycling’ can make a significant contribution to providing environmentally friendly solutions. Managing Director Mike Southgate says ‘manufacturers have a responsibility to make more efficient use of materials’. That is a marked change in emphasis on sustainability issues. Initiatives such as the Code for Sustainable Homes, the 2012 Olympic Construction Commitments and the Government’s Waste Resource Action Programme (WRAP) have put sustainability to the top of the agenda for the construction industry. Recovinyl is funded by the PVC industry body Vinyl 2010 formed to demonstrate commitment to sustainable development. It is backed by the British Plastics Federation and supported by the Waste Resources Action Programme (www.recovinyl.com, www.recovinyl–wrap flyer 2008).

11.8.5 Boeing and Alenia composites recycling venture Composites need to be taken seriously since they play an ‘increasingly significant role’ in the manufacture of more environmental friendly aircraft parts. In July 2008 Boeing established a joint venture aircraft composite recycling operation with Alenia Aeronautica in Italy, to deal with the


299

significant amounts of high performance composites used in modern aircrafts (www.alenia-aeronautica.It). Milled Carbon of the UK has a pilot-scale technology for recovering carbon fibres and is also involved with the project. Boeing is making use of composites in the primary structure of the 787 Dreamliner, and Alenia Aeronautica is a composite partner in that project supplying fuselage and stabiliser structures, thus ideally placed to reuse recycled composites in aircrafts. PRW mentions that The initial focus of the venture is to process carbon fibre composite scrap material from Alenia and its supply chain operations. In the longer term, the organisations plan to work with Italian industry and academia to develop new markets for recovered carbon fibre in areas such as automotive and sporting goods. The recovered carbon fibres are said to be suitable for reincorporation into non-critical aircraft composites such as interior linings and seat parts.


11.8.6 SIMS Group SIMS Recycling Solutions, a division of the global recycling SIMS Group (www.eu.simsrs.com), has launched its long awaited plastics separation technology, Waste Electrical and Electronic Equipment (WEEE). The company will run the technology across its European WEEE sites. The majority of the collected recyclate will be post-consumer waste, composed of high-impact polystyrene, acrylonitrile–butadiene–styrene (ABS), polypropylene and polyethylenes. Other plastics like polyamides (nylons) will also be present and thus collected, but since they are in smaller quantities, they are left with the waste stream because at present it is not economically viable to recover them. (www. prw.co.uk)

SIMS is expecting to produce 40 000 tonnes of plastics from its European WEEE business by 2010.

11.8.7 Wolfson Centre for Materials Processing, Brunel University Brunel University (Brunel.ac.uk/research/centres; Tarverdi, 2006) has collaborated with wallpaper manufacturers like Fine Decor, Graham and Brown, Speciality Coatings, Imperial Home Décor Group and material supplier Ineos on the development of extrusion technology that can convert post-industrial vinyl wallpaper waste into an added value compound with the inclusion of waste plastisol. These compounds can be either injection

300



11.5 Green barrier coupling component; manufactured from 50% postindustrial vinyl wallpaper waste and white component manufactured from 100% virgin PVC.

moulded or extruded using conventional polymer processing equipment (see Fig. 11.5). In the UK approximately 15 000–20 000 tonnes per annum of postindustrial wallpaper and plastisol waste is currently land-filled.

11.9

Sources of further information and advice

Accessed April 2009 http://www.alenia-aeronautica.it http://www.axionrecycling.com http://www.bpf.co.uk http://www.breeam.org http//www.brunel.ac.uk/research/centres http://www.bsi-global.com http://www.centriforce.com http//www.closedlooprecycling.com http://www.defra.gov.uk/environment/waste/europe/index.htm http://www.ec.europa.eu/environment/waste/legislation/a.htm http://www.environment-agency.gov.uk/business/topics/waste http://www.epwin.co.uk http://www.eu.simsrs.com http://www.hallmarkpanels.co.uk http//www.iso.org http://www.letsrecycle.com/prices/plasticsarchive2007.jsp http://www.mkelectric.co.uk


301


http://www.pvc group.co.uk http//www.recoup.org http://www.recovinyl.com http://www.recovinyl–wrap flyer 2008 http://www.remarkable.co.uk http://www.theconstructioncentre.co.uk http://www.therenewableenergycentre.co.uk http://www.vannplastic.co.uk http://www.vekauk.com http://www.wasteonline.org.uk/resources/informationSheet/plastics.html http://www.wastewatch.org.uk http://www.wrap.org.uk http://www.wrap.org.uk/plastics/construction bevis m j, hornsby p r, tarverdi k and lee w h, (1994) Recycling of composites, Concise Encyclopaedia of composite Materials, Ed. A. Kelly. Persimmon hornsby p r, tarverdi k, (1996) Integrated polymer recycling technology, UK Patent Application la mantia f p (ed), (1996) Recycling of PVC and Mixed Plastic Waste, Chem. Rec. Publishing liu y, meng l, huang y, du j, (2004) Recycling of carbon/epoxy composites, J Appl Polym Sci, 94 (5), 1912–1916 mcdonough w, braungart m, (2002) Cradle to Cradle: Remaking the Way We Make Things, North Point Press, NY palmer j, ghita o r, savage l, evans k e, (2008) Recyclate fibre – matrix interface analysis for reuse in sheet moulding compounds (SMC), Proceedings of European Conference on Composite Materials (ECCM13), Stockholm, Sweden pickering s j, (2006) Recycling technologies for thermoset composite materials – current status, Comp A: Appl Sci Manuf, 37 (8), 1206–1215 wolfson centre (1995–1997) Recycling and Recovery from Composite Materials 1995–1997, DTI LINK Structural Composite Programme, Brunel University, Wolfson Centre for Materials Processing and University of Nottingham, Department of Engineering yip h l h, pickering s j, rudd c d, (2002) Characterisation of carbon fibres recycled from scrap composites using fluidised bed process, Plastic Rubber Compos Process Appl, 31 (6), 278–282

11.10 References bagrodia s, (2008) Advanced Materials from Novel Bio-based Resins, ANTEC Cereplast Inc. bevis m j, bream c e, hornsby p r, tarverdi k and williams k s, (1996) Proc. 20th Int. BPF Composites Congress, Hinckley, UK brandrup j (ed), (1995) Recycling and Recovery of Plastics, Hanser Publishers bream c e, hornsby p r, tarverdi k and williams k s, (1997a) ‘Adding value to thermoset composite recyclate’, Euromat ′97, 21–23 April, Maastricht


302


bream c e, hinrichsen c, hornsby p r, tarverdi k and williams k s, (1997b) Integrated compounding technology for preparation of polymer composites reinforced with waste materials, Plastics, Rubber and Composites Proc Appl, 26 (7), 303–310 bream c e, hornsby p r, (2001) Comminuted thermoset recyclate as a reinforcing filler for thermoplastics. Part II: structure–property effects in polypropylene compositions. J Mat Sci, 36 (12), 2977–2990 eu end-of-life vehicles (elv) directives 2000/53/EC & 2006/12/EC eu end-of-life vehicles regulations, (2003) Statutory Instrument 2003 No. 2635 eu landfill directive 1999/31/EC goodship v, (2007) Introduction to Plastics Recycling, 2nd Edition, Smithers RAPRA palmer j, ghita o r, savage l and evans k e, (2009a) ‘Successful closed loop recycling of thermoset composites Part A’, J Appl Sci Manufacturing, 40 (4), 490–498 palmer j, ghita o r, savage l and evans e k, (2009b) New automotive composites based on glass and carbon fibre recyclate, Proceeding of International Conference for Composite Materials (ICCM 17), Edinburgh, UK pim project (2009) TP-K1042A funded by TSB (2006–2009) producer responsibility obligations (packaging waste) regulations (2007). UK implementation of Packing and Packaging Waste Directive 94/62/EC – BEER reccomp project tp2-10083 funded by tsb (2005–2008) tarverdi k, (2006) Recycling and reuse of vinyl wallpaper, Annual Technical Conference, ANTEC Charlotte, 7–11 May wrap, (2008) Local Authorities Plastics Collection Survey xia w, kang y g, song j h and tarverdi k, (2007) Biopolymer composites reinforced with fractioned wheat straw, Advances in Eco-materials, Proceedings of the Eight International Conference on Eco-materials (ICEM8 2007) Vol. 1, Brunel Univ. Press

12 Quality and durability of recycled composite materials K. L. P I C K E R I N G, University of Waikato, New Zealand, and M. D. H. B E G, University of Malaysia Pahang, Malaysia


Abstract: Polymer matrix composite products are manufactured on a mass scale that is increasing each year, along with consumer interest and government legislation driving the pressure for them to be recycled. A major factor determining the route for recycling of a polymer matrix composite is whether the matrix is a thermoplastic or a thermosetting plastic (thermoset); unlike thermosets, thermoplastics can be melted and therefore lend themselves more readily to recycling by being mechanically broken down for remoulding. Techniques for recycling thermoset and thermoplastic matrix composites are discussed including mechanical breakdown, thermal recycling as well as chemical recycling with available information on the related quality and where known, the durability of the recycled materials. Key words: thermoset and thermoplastic composites, mechanical recycling, chemical extraction, moisture resistance, thermal properties.

12.1

Introduction

It is estimated that globally, more than 6 million tonnes of polymer matrix composite (PMC) products are manufactured each year, largely dominated by glass fibre reinforced polymers (GFRPs) (Lester et al. 2004). Furthermore, the amounts used are increasing annually. For just sheet moulding compounds (SMCs), which are based on glass fibre reinforced unsaturated polyester resin and one of the most commonly used composite materials, waste increased from 0.1 million tonnes in 1984 to just less than 0.4 million tonnes in 2000 (Perrin et al. 2008). Markets include automotive, leisure, electronics, aerospace and construction industries. On a more modest level, an estimate of the worldwide production of carbon fibre reinforced plastic (CFRP) waste is of the order of only 5000 tonnes per year (Ogi et al. 2005). However, encouraged by price reductions, the use of carbon fibre, once largely the domain of aerospace and sporting goods, is now being adopted by mainstream automotive and construction industries and is also increasing. This increase is estimated in Europe, for example, to be occurring at the rate of about 10% per annum (Lester et al. 2004). In addition to 303


304


end-of-life components, waste CFRP in the form of off-cuts is produced as a by-product of the manufacture of CFRP products. Increased PMC utilisation has led to increasing pressure to resolve issues relating to composite waste. Consumer interest and government legislation are now drivers to encourage recycling of polymer matrix composites. In Europe where the automotive sector is one the largest users of composite materials, the End-of-Life Vehicle (ELV) Directive (European Commission, 2000) was the earliest major legislation regarding recycling of materials (Perrin et al. 2008). However, legislation is being put into place to cover other industries. This includes the Waste Electrical and Electronic Equipment (WEEE) Directive (European Commission, 2003) and the impending directive on ‘Construction and Demolition Waste’. However, by their very multi-component nature, and the variety of materials used, composite recycling raises many challenges dependent on the composite type. One major factor determining the route for recycling of a PMC is whether the matrix is thermoplastic or a thermoset plastic. Thermoplastics can be melted and therefore lend themselves more readily to recycling by reshaping. Indeed, there is much interest in the area of thermoplastic matrix composites, particularly in the automotive area, due to the readiness with which they are expected to be recyclable. Thermoset matrix composites are less obviously recyclable because of the inability to remould them. However, increased adoption of these materials, along with increased cost of landfill and the use of more expensive fibres than are commonly used in thermoplastics, in particular carbon fibre, has driven interest in reuse of these materials (DeRosa et al. 2005). Potential techniques that could be in use for these materials include mechanical breakdown, thermal recycling as well as chemical recycling, which are discussed in the following sections. Although these techniques could also be used for thermoplastic matrix composites, energy considerations would suggest remoulding to be a much more desirable option. Owing to the fundamental differences in their treatment, recycling of thermoset and thermoplastic matrix composites are described separately within this chapter.

12.2

Recycling thermoset matrix composites

12.2.1 Performance using mechanically broken down material The use of mechanical means for recycling thermoset waste has received the most attention. Research has been carried out to assess the reuse of mechanically broken down thermoset matrix composites containing glass,


Quality and durability of recycled composite materials

305

carbon and Kevlar® fibre. However, glass, particularly contained within SMCs or bulk moulding compounds (BMCs) has had the most interest, largely reflecting its dominance in the market place. Products using virgin SMCs or BMCs, for example include baths and sinks, car bumpers, car panels, car headlamp housings, door panels on aircraft and circuit boards (DeRosa et al. 2005). Composite material is commonly mechanically recycled in three stages. Initially, coarse material of particulate diameter around 50–100 mm is produced by cutting or crushing, followed by grinding to give a diameter of 50 μm to 10 mm, which is finally separated into grades of materials of different size (Pickering 2006). Typically, separation leads to a number of particulate grades and also more fibrous grades with higher aspect ratios. The particulate grades have been regarded as a potential replacement for fillers, but it is hoped that separation of the more valuable fibre fraction for recycling is likely to give the best returns on recycling effort (Kouparitsas et al. 2002). Industrial-scale mechanical separation of thermoset matrix composites has been demonstrated by a number of companies including ERCOM, Mecelec and Valcor in Europe and R.J. Marshal, Premix and Phoenix Fiberglas in North America (Pickering et al. 2000; DeRosa et al. 2005; Pickering 2006), although Phoenix Fiberglas stopped operating in 1996. All of these companies have focused on SMC and BMC waste. Toyota Motor Corporation has also incorporated SMC recycling in-house (DeRosa et al. 2005). ERCOM, one of the largest companies in terms of recycling throughput, has a capacity of 6000 tonnes per annum. Particulate grade materials produced by mechanical separation have been investigated for reuse in composite production, with some success where the focus has been on replacement of filler. Substitution of up to 88 wt% of the CaCO3 filler in SMC by fine particulate recyclate has resulted in materials with comparable tensile and flexural strengths and moduli (DeRosa et al. 2005) to that containing only virgin material. Toyota has also recycled finely ground SMC for a filler in SMC up to 20 vol% and achieved performance similar to virgin SMC (Inoh et al. 1994). Of particular note, improvement of strength of 30% has been obtained where very finely ground (15–20 μm) recycled PMC produced using a specialised grinding method, was recycled into its own production stream as a replacement for calcium carbonate filler (Kojima and Furukawa 1997). The strength of epoxy resin has also been found to increase 16 and 20% with addition of 1% of finely ground moulding/pre-preg scrap containing Kevlar and carbon fibre respectively (Anonymous 2008). In the same study, Kevlar recyclate was found to increase the bending strength of polyurethane foam. In addition, non-mechanical improvement in the form of damping capacity has been obtained in plates and beams using sieved recyclate (Thomas et al.


306


2000). Improvement in mechanical performance with particulate grade materials has been found to be heavily dependent on particulate size, with finer grades generally giving the greatest benefit. Addition of 17.5 wt% particulate recyclate in BMC has been found to lead to a modest improvement or a 15% reduction in tensile strength for fine and coarse grade particulate grades respectively (DeRosa et al. 2005). Success, even with fine grade recyclate, however, has been mixed. Fine grade recyclate has commonly led to a penalty in mechanical performance (Butler 1991; Jutte and Graham 1991; Pickering et al. 2000). Although improvement of flexural strength was obtained in one study by adding 15 wt% of fine recyclate particles to BMC, this was coupled with a reduction in tensile strength (Bledzki and Goracy 1993). Commonly, use of particulate recyclate has been found to reduce flexural modulus (DeRosa et al. 2005). Another major issue for reuse as a filler in SMC and BMC is that for the quantities of recyclate required in moulding compounds to improve economic viability, viscosity becomes too high for standard processing to be used (Pickering et al. 2000). In addition, delay of curing has also been considered to be due to a reduction in thermal conductivity (Petterson and Nilsson 1994). Another potential limit to automotive uptake is the effect of surface finish, although this is not affected when up to 10 wt% recyclate is added (DeRosa et al. 2005). Making use of coarser and more fibrous recyclate SMC/BMC has proved to be more of a challenge. Fibrous recyclate added to BMC has resulted in reductions of flexural modulus, tensile strength and impact strength (Bledzki and Goracy 1993). For example, replacement of virgin fibre and half of the filler by fibrous recyclate in BMC reduced tensile and flexural strength by 20% and 54% respectively (DeRosa et al. 2005). Complete replacement in BMC of glass and filler by granulated SMC recyclate has been shown to lead to a general reduction of mechanical properties of the order of around 30% (Curcuras et al. 1991). However, it has been shown to be possible to maintain flexural strength of BMC with up to 15 wt% replacement of fibrous filler (Bledzki and Goracy 1993; DeRosa et al. 2005). It is believed that properties have been held back by poor interfacial bonding between the recyclate and its new matrix material (Pickering et al. 2000). Pettersen and Nilsson (1994) have achieved improvement in flexural strength of 15% for SMC using a combination of particles and more fibrous recyclate added at 10 wt% with a slightly reduced flexural modulus, but with no penalty in tensile strength. Compensation for reduced performance has also been achieved by combining longer lengthed recyclate and virgin glass fibre than was to be originally used in materials (DeRosa et al. 2004). Mechanically recycled SMC has found industrial application in SMC/ BMC car spoilers (at 15 wt%) and spare wheel covers (at 10 wt%) along with other automotive parts (DeRosa et al. 2005). Products using finely



307

ground recyclate have demonstrated the ability to reduce virgin glass fibre utilisation slightly, coupled with weight reduction. Further potential for such recyclates includes incorporation into materials for drain covers and wood chip particleboard used for domestic flooring where improvement in properties has been shown (Conroy et al. 2004). Overall, however, it is currently not possible to ‘close the recycling loop’ due to the limit at which they can be added to materials without compromising performance (DeRosa et al. 2005). As an alternative to incorporation in thermoset matrix composites, mechanically broken down thermoset-based PMC has been assessed for use in other materials including concrete and thermoplastics. Mechanically broken down CFRP has been used in concrete to give modest improvements in compressive and flexural strength as well as more significant improvements in work of fracture (Ogi et al. 2005). These improvements have also been found to be strongly dependent on the size of the added CFRP pieces. The advantage of using carbon fibre in this form was expressed relating to dispersion, such that when contained in a polymer, the clumping that would normally be expected to occur with fibre alone, is overcome, although reduction of bonding was observed. Improvement of mechanical performance of polypropylene (PP) has also been obtained by the addition of coarse and fine SMC recyclate, including more than a doubling of Young’s modulus and increased notched impact energy, flexural strength and modulus, whilst maintaining tensile strength, although unnotched impact energy reduced (Jutte and Graham 1991). One study showed the addition of BMC and woven glass reinforced phenolic fragments to be able to improve tensile strength of PP by 10 and 134%, as well as Young’s modulus by 72 and 183% respectively, dependent on silanation of the recyclate along with maleic anhydride modification of the PP, although reduction of charpy impact strength was found particularly with BMC (Bream et al. 1997; Bream and Hornsby 2000). Use of scrap prepreg consisting of carbon in epoxy has been investigated as reinforcement for thermoplastic (Blizard 1998). Good increase was found with flexural properties and creep resistance. One study looked at mechanical separation via grinding and sieving of glass from polyester and Kevlar® and carbon from epoxy resin (Kouparitsas et al. 2002). The fibre recovered still retained some of the matrix material such that the fibre did not separate into individual fibres. Strengths and stiffnesses of thermoplastic (PP or an ethylene/ methacrylic acid copolymer) reinforced using these recycled fibre-rich materials were found to be similar to that reinforced with the equivalent virgin fibre of similar average fibre length for glass or Kevlar®, although appreciably lower for carbon. Unfortunately, although not compared in the paper, improvement compared with that expected of matrix-only materials appears limited, particularly for strength.

308


Potential products using recyclate as a reinforcement in thermoplastics include materials that can replace wood that can be drilled and cut in moderate load-bearing structures such as groynes, footbridge foundations and jetties. Improvement in asphalt for bridge decking and concrete have also been identified as a potential use of these materials (Conroy et al. 2004).


12.2.2 Performance using thermally extracted fibre The main techniques with which it is possible to thermally extract fibre include pyrolysis, catalytic conversion and fluidised bed processing. One general advantage of thermal extraction is that a range of material can be handled along with some contamination (DeRosa et al. 2005). Pyrolysis involves heating material in the absence of oxygen, and for polymers leads to the production of oils and gases which all have potential for recycling into other chemicals or used as fuels (Williams 2003; Williams et al. 2005). For thermoset matrix composites, it is generally carried out at less than 500 °C. Pyrolysis has been investigated for recycling mixed polyester/styrene composite waste (Williams et al. 2005) using a fixed bed reactor at 450 °C. Here, the final separation of the fibre involved oxidation in a muffle furnace followed by sieving. Relatively clean fibre was produced that maintained its pre-pyrolysis length, but only approximately half of its strength. When this glass fibre was substituted for 25 wt% of the fibre in polyester matrix composites, reductions in flexural strength, flexural modulus and impact strengths of 7, 19 and 26% respectively occurred compared with composite reinforced with only virgin fibre. Better fibre strength retention has been achieved through a similar route, but by using mechanical means rather than oxidation to separate the fibre, resulting in fibre strength of over 60% that of the virgin fibre, although fibres still retained some char on their surfaces (Cunliffe et al. 2003). Pyrolysis followed by milling has also been assessed for recycling SMC material as a filler for further SMC production (DeRosa et al. 2005). The particles obtained have been found to give similar, or improvement of performance, for a range of mechanical properties when replacing 20% of the CaCO3 filler in SMC. However, less success was achieved when trying to replace virgin fibre with recycled fibre, such that again reduction of properties occurred. Pyrolysis, however, has been shown to be self-sustaining in terms of energy release during processing (DeRosa et al. 2005). The company Milled Carbon Ltd has set up industrial-scale pyrolysis facilities in the UK with an interest in recycling carbon fibre from epoxybased carbon fibre off-cuts including prepreg and are also investigating markets for their fibre where virgin quality fibre is not needed (Anonymous 2006). However, this company also seems to be investigating fluidised bed processing (Rush 2007) including energy recovery from the matrix. Products identified for short recycled carbon fibre include cellular phones,



309

laptop computers as well as BMCs, SMCs and injection moulding compounds (Rempes 2003). Catalytic conversion involves pyrolysis at low temperatures (as low as 200 °C) in the presence of a catalyst to similarly form hydrocarbons which could be used as chemical feedstock or fuel (Lester et al. 2004) and enable extraction of fibre and other inorganics from PMCs. An initial feasibility study demonstrated that SMC, mixed plastic car body parts (as well as thermoplastic matrix composites) containing glass fibre can be recycled using catalytic conversion enabling reuse of fibres as well as fillers, metal inserts and the matrix polymer as chemical feedstocks (Allred and Busselle 2000). Processing at 300 °C resulted in relatively clean fibre. This has been proposed as an economically viable alternative to using landfill (Allred et al. 1997). Optimised processing has resulted in glass fibre with approximately half the strength of virgin glass (Allred et al. 1997). Carbon fibre has been recovered from woven carbon fibre in epoxy matrix composites with similar surface chemistry and only 9% loss of strength compared with virgin fibres, for which the strength loss could largely be explained by the weaving process which is known to commonly reduce strength by 5–10% (Allred and Coons 1996). Epoxy residue was observed to be in the form of small particles less than a tenth of a micron in diameter still attached to fibres. However, the fibre recovered was considered as suitable for use in moulding compounds or for milling feedstock. It has also been shown that release plies on waste pre-preg can be left on for this process (Allred 1996). Adherent Technologies Inc. (ATI) in the US have developed a prototype system for a catalytic conversion process which is relatively flexible in terms of feedstock, that can extract carbon fibre from composites into a milled or chopped fibre form (Rush 2007). ATI has worked with Boeing to assess recovery processes as part of a life cycle analysis for the Boeing 787 Dreamliner (Rempes 2003). Dr Jan-Michael Gosau, the Environmental/Engineering Manager at ATI, has expressed the importance of this relationship, in that previously there has been insufficient continuous supply of scrap for recycling for commercial viability. He expects that due to Boeing 787 Dreamliner production, commercial level recycling of CFRP will be occurring in 2009, aiming at a throughput of approximately 500 tonnes per annum producing milled carbon fibre. The use of fluidised bed processing for recycling thermoset matrix composites has been under investigation for more than ten years. In this process, combustion allows retrieval of fibres as well as inorganic fillers and the potential for energy recovery from the matrix. Agitation caused by gas in the bed assists separation of fibres from fillers, which can be removed by washing in dilute detergent, followed by filtration (Kennerley et al. 1998). Early work investigated the extraction of glass fibre as well as fillers from unsaturated polyester matrix composites including SMC, wound pipe and


310


sandwich panels (Pickering et al. 2000). The strength of fibre was found to be dependent on processing temperature. At 450 °C, the minimum temperature considered to give an adequate processing rate, strength was approximately half that of virgin fibre, which reduced further at higher processing temperatures, although Young’s modulus remained unaffected. The content of the recyclate after final washing and filtration was found to be 92 wt% fibre (Kennerley et al. 1998). A limit of about 10 mm on size of scrap pieces was used to ensure dwell times such as would minimise fibre damage. Preliminary assessment of the economic feasibility of this process suggested economic viability for a plant with a throughput of about 9000 tonnes per annum (Pickering et al. 2000). A pilot-scale assessment has also been carried out and showed that replacement of 50% of the virgin fibre by recycled material, in a standard formulation for a compression moulded headlamp surround, led to a material with generally similar processing and mechanical performance, although a slight reduction in impact performance was observed. Further investigation at higher substitution levels gave no reduction in Young’s or flexural modulus; however, they brought about a reduction of tensile and flexural strength as well as impact strength and a darkening of mouldings (Pickering et al. 2000). Reductions of properties were up to around 40, 50 and 75% for tensile strength, flexural strength and impact strength respectively for full substitution of virgin fibres by recycled fibre. Fluidised bed processing has been investigated for its potential to recycle carbon fibre. In one study (Jiang et al. 2008) shredded CFRP scrap was oxidised in a fluidised bed at 550 °C to remove the matrix and release the fibre. The interfacial strength was found to be unaffected compared with virgin fibres when used. Better strength retention than for glass fibre has been observed with values in the order of 80% along with full retention of stiffness (Pickering et al. 2000) compared with virgin fibre. Microwave heating has been assessed for the recovery of long carbon fibre from CFRP (Lester et al. 2004). Microwaving at 3 kW for 8 seconds was found to volatilise the epoxy resin matrix leaving only trace amounts of matrix material of around 2% w/w of fibre in the form of nodules, retaining approximately 80% of the original fibre strength and 87% of the Young’s modulus. The commercial viability of this process, however, remains to be proven.

12.2.3 Performance using chemically extracted fibre This is arguably the least mature of the potential techniques for recycling thermoset matrix composites. Hydrolysis, glycolysis, solvolysis and acid digestion have been considered for chemical recycling of thermoset matrix composites (DeRosa et al. 2005).



311

Currently, hydrolysis appears to be of limited use for recycling of thermoset matrix composites. For hydrolysis to be effective it appears to be necessary to separate waste into that of different matrix polymers (Perrin et al. 2008). In addition, material has had to be ground down to a fine particle size, such that replacement of filler would be the best achievable outcome expected and further problems arise due to the waste stream produced. Interest in glycolysis has been demonstrated by the registering of a patent by the Miyaso Chemical Company in Japan involving degradation of polyester resin with glycol such that it can be used in further synthesis of unsaturated polyester or polyurethane, but with no reference to the effect on the fibre present (Shizu et al. 1997). Solvolysis with ethylamine has been shown to be capable of producing visually clean fibres (Winter et al. 1995). One research study involved investigation of different routes of chemical recycling for SMCs (Winter et al. 1995). One solvolysis treatment with a mixture of ethanol and potassium hydroxide, although successful at solubilising the matrix, involved a subsequent neutralisation that formed ‘a large stream of waste chemicals’. In the same study, solvolysis of SMC with ethanolamine followed by washing with ethanol enabled recovery of fibres and filler along with a polymeric residue. Substitution of half the virgin fibres of a standard BMC formulation with recycled glass was found to not significantly reduce Young’s modulus or flexural strength, although impact strength appeared to be lower. Some success of extracting aramid (TwaronTM) fibre from epoxy by solvent swelling was achieved using a solution of dimethyl sulphoxide (DMSO) and toluene in equal parts, although fibre damage was apparent without full removal of epoxy (Buggy et al. 1995). Solvent extraction of SMC using acetone, dichloromethane, trichloromethane and combined trichloromethane/benzene was shown in one study to be largely ineffective (Patel et al. 1993). Acid digestion raises the challenge of using hazardous chemicals and conditions as well as, similar to the other chemical processes, the resulting complex hydrocarbon/acid mixture (Allred 1996; Rempes 2003). Recently, however, extraction of initially shredded SMC waste has been carried out using orthophosphoric acid to dissolve calcium carbonate (generally of the order of 50–55 wt% of SMC) (Perrin et al. 2008). This process was found to dissolve about half of the calcium carbonate giving a recyclate product enriched in fibre with potential for use as reinforcement/filler in thermoplastics. Scale-up was investigated enabling batches of up to 200 kg to be processed. Recognition of the potential for this process occurred by way of receipt of the Innovation Techniques 2006 Waste Category prize from the ADEME (French national agency for the environment).

312


12.3

Recycling thermoplastic matrix composites

Owing to the ability to melt thermoplastics, mechanical breakdown into granules for use in the original processing stream is the most obvious technique for recycling fibre reinforced thermoplastics and indeed has been the area of greatest focus. However, for fibre reinforced thermoplastics, fibre breakage induced by grinding and subsequent processing (e.g. injection moulding) leads to reduction of material properties. For these reasons, recycling by dissolution of the polymer matrix, often at elevated temperatures, has also been considered.

12.3.1 Effects of recycling on mechanical properties


Performance using mechanically broken-down material The simplicity of mechanical recycling and its relatively low cost give it the greatest potential, particularly for short fibre reinforced thermoplastics, for which fibre breakage during reprocessing has a lower impact on reinforcing properties than for long fibre reinforced plastics (Bernasconi et al. 2007). Although no post-consumer based recycling studies have been carried out on glass fibre reinforced thermoplastic matrix composites, there are some studies on reprocessing. Eriksson and Albertsson (1996) assessed the effects of reprocessing of short glass fibre reinforced polyamide injection moulded composites containing 30 wt% of glass fibres, by comparing composites made from virgin fibre and polyamide with those produced from mechanically broken-down composites. This study evidenced the role of fibre breakage induced by successive injection moulding processes. The tensile strength was found to reduce from 192 MPa for virgin composites to 132 MPa for composites reprocessed eight times. A similar trend was found by Bernasconi et al. (2007) when reprocessing 35 wt% glass fibre reinforced polyamide composites. Chu and Sullivan (1996) obtained lower tensile strength, but similar Young’s modulus and improved impact strength for reprocessed glass fibre poly(butylene terephthalate) composites compared with virgin composites. As for synthetic fibre reinforced thermoplastic composites, no postconsumer based recycling studies have been carried out on agro-based fibre composites (Rowell et al. 1997), but again there have been a number of studies on reprocessing. Agro-based fibres are less brittle and softer than glass fibres and are therefore more likely to retain properties during recycling. Bourmaud and Baley (2007) compared reprocessing of hemp and sisal with glass fibre reinforced polypropylene (PP) composites. They observed reduction of both tensile strength (TS) and Young’s modulus (YM) for all composites, with greater reduction found for glass fibre

313

reinforced PP (about 52% for TS and 40% for YM) than PP reinforced with sisal (17% for TS and 10% for YM) or hemp (13% for TS and 1% for YM) after being reprocessed seven times. Walz et al. (1994) studied reprocessing of 50 wt% kenaf fibre reinforced PP composites. Both tensile and flexural properties were found to decrease (by about 20%) with increased number of times the materials were reprocessed (up to nine times). A similar trend was shown by Joseph et al. (1993) for 20 wt% sisal fibre/LDPE matrix composites. However, Youngquis et al. (1994) reported that the mechanical properties and dimensional stability of second-generation wood fibre reinforced polyethylene panels were equivalent to, or better than, properties obtained from first-generation panels due to better encapsulation of fibre matrix. Beg and Pickering (2008a) carried out an extensive study on reprocessing of wood fibre reinforced PP composites. Composites were produced with 40 or 50 wt% fibre and reprocessed by repeated pelletising and injection moulding and the trend for mechanical performance was found to depend on fibre content. For 40 wt% fibre, TS and YM of composites decreased with increased number of times the materials were reprocessed in a linear fashion (Figs 12.1 and 12.2) leading to a reduction of 25% for TS and 17% for YM after reprocessing eight times. For 50 wt% fibre composites, TS of virgin composites was lower than for 40 wt% fibre content composites (Figs 12.1 and 12.3). This was explained to be due to the limited dispersion of fibre in composites at higher fibre content due to the increase in composite viscosity as indicated by the reduction of melt flow index (melt flow index of 40 wt% fibre composites was 0.9 g/10 min and 50 wt% fibre composites was 0.14 g/10 min). However, TS and YM of 50 wt% fibre composites increased with the first two reprocessing cycles (Figs 12.3 and 12.4), which was considered to be due to improved fibre 45 y = –1.36x + 41.16

40 Tensile strength (MPa)



Composite

PP

35 30 25 20 c

15 10 5 0 Virgin

1

2

3

4

5

6

7

8

Number of times reprocessed

12.1 Tensile strength of virgin and reprocessed composites (40 wt%) fibre and PP (Beg and Pickering 2008a).

314


Young's modulus (MPa)

6000 Composite

y = –78.08x + 4553

5000

PP

4000 3000 2000 1000 0 Virgin

1

2

3

4

5

6

7

8




12.2 Young’s modulus of virgin and reprocessed composites (40 wt%) fibre and PP (Beg and Pickering 2008a). 50 45 40 35 30 25 20 15 10 5 0 Virgin

1

2

3

4

5

6

7

8


12.3 Tensile strength of virgin and reprocessed composites (50 wt%) fibre and PP (Beg and Pickering 2008a).

dispersion, although decreased with further reprocessing to give overall an 11% reduction of TS and a 6% increase in YM after reprocessing eight times, compared with the virgin composites. Similar to Walz et al. (1994) and Joseph et al. (1993), Beg and Pickering (2008a) also found flexural strength and flexural modulus to decrease with increased number of times the materials were reprocessed. A 30% reduction of flexural strength and 20% reduction of flexural modulus were found for 40 wt% fibre composites after reprocessing eight times compared with virgin composites. As observed for glass fibre reinforced thermoplastics, the reduction of mechanical properties with reprocessing has been linked to increased fibre damage (Harper et al. 2006; Beg and Pickering 2008a). Beg and Pickering (2008a) found the average fibre length to decrease from 2.36 mm for virgin


315

Young's modulus (MPa)

8000 7000 6000 5000 4000 3000 2000 1000 0 Virgin

1

2

3

4

5

6

7

8


12.4 Young’s modulus of virgin and reprocessed composites (40 wt%) fibre and PP (Beg and Pickering 2008a).

Length (mm)


3.0 2.5

y = 3.4564e–0.3545x R 2 = 0.9802

2.0 1.5 1.0 0.5 0

Virgin Virgin 2 4 6 fibre composite Number of times reprocessed

8

12.5 Weighted average fibre length of virgin fibre and the fibre extracted from composites (Beg and Pickering 2008a).

fibre to 0.37 mm for fibre extracted from 40 wt% fibre composites reprocessed eight times (Fig. 12.5). In addition, the length distribution of fibres became narrower and reduced to shorter fibre lengths (Fig. 12.6) and the amount of fibre fines (fibre length less than 0.20 mm) was found to increase (from 9% for virgin composites to 80% for composites reprocessed eight times) with increased number of times the materials were reprocessed. Beg and Pickering (2008a) found that the change in fibre length with the number of times the composite material was reprocessed followed the empirical equation: lN = l0 e− bN

[12.1]

where lN is the average fibre length at any reprocessed composites, l0 is the length of virgin fibre, b is the slope of fibre length versus number of times

316

Management, recycling and reuse of waste composites Virgin composite

15

Population (%)

Population (%)

Virgin fibre

10 5 0 0.2 1 1.8 2.6 3.4 4.2 5 Length (mm)

20 15 10 5 0 0.2 1 1.8 2.6 3.4 4.2 5 5.6 Length (mm)

5.6

8 times reprocessed composite Population (%)

Population (%)

4 times reprocessed composite 30 20 10 0 0.2 1 1.8 2.6 3.4 4.2 5 5.6 Length (mm)

60 40 20 0 0.2 1 1.8 2.6 3.4 4.2 5 5.6 Length (mm)


12.6 Fibre length distribution of virgin fibre and the fibre extracted from composites (Beg and Pickering 2008a).

reprocessed graph and N is the number of times the composites were reprocessed. Based on this equation it was found to be possible to correlate composite strength with the change in fibre length when fibre was evenly distributed as follows: a l σ N = σ 0 − ln ⎛⎜ 0 ⎞⎟ ⎝ b lN ⎠

[12.2]

Beg and Pickering (2008a) also studied the failure strain (FS) of composites during reprocessing and found that FS of both 40 and 50 wt% fibre composites increased exponentially with increased number of times the materials were reprocessed (Figs 12.7 and 12.8). Bourmaud and Baley (2007) also reported an increase in elongation at break after reprocessing for PP/hemp (22%), PP/sisal (9%) and PP/glass (34%) composites after reprocessing seven times. The significant increase was explained as due to the decrease in fibre length induced by reprocessing as discussed previously (Bourmaud and Baley 2007; Beg and Pickering 2008a). The more significant increase in elongation at break for PP/glass fibre composites was explained by the poor adhesion between glass fibres and matrix after seven cycles thus mobility of fibres was enhanced by easier debonding (Bourmaud and Baley 2007) but it would also be due to increased fibre breakage. Work by Beg and Pickering (2008a) also assessed the effect of reprocessing on impact strength and hardness of composites. The impact strength


317

6 y = 1.6701e 0.1211x R 2 = 0.9761

Failure strain (%)

5 4 3 2 1 0

Virgin

1

2 3 4 5 6 Number of times reprocessed

7

8

12.7 Failure strain of virgin and reprocessed composites (40 wt% fibre) (Beg and Pickering 2008a).

4.5

Failure strain (%)


4.0 3.5 3.0 2.5

y = 1.051e 0.1366x R 2 = 0.9701

2.0 1.5 1.0 0.5 0 Virgin

1

2 3 4 5 6 Number of times reprocessed

7

8

12.8 Failure strain of virgin and reprocessed composites (50 wt% fibre) (Beg and Pickering 2008a).

has been shown to decrease from 6.2 kJ/m2 for virgin composites to 3.2 kJ/m2 for the composites reprocessed eight times. The reduction of impact strength has been suggested as being due to the reduction of molecular weight of PP during reprocessing and decrease in fibre length which increases the number of fibre ends that act as crack initiation. Hardness was found to increase with reprocessing such that a 35% increase in Vicker’s hardness number was obtained for composites reprocessed eight times compared with virgin composites. The increase in hardness has been considered to be due to the reduction of micro-voids and the increase in composite density with increased reprocessing.

318


Performance with chemically extracted fibre


Several studies have involved the use of solvents, including methanol, toluene and xylene, to extract fibre for recycling from composites. Papaspyrides et al. (1995) employed a solvent-based technique to recycle thermoplastic glass fibre composites. Toluene was used as a solvent for the low density polyethylene matrix. The glass fibre from the recycled route formed stronger second generation composites which was explained as being due to residual polymer on the fibre surfaces aiding fibre–matrix bonding in the recycled composites and also to new matrix being able to recrystallise to form a transcrystalline layer on the residual polymer around the fibres (Zafeiropoulos et al. 1999). Poulakis et al. (1997) also found that recovered glass fibre (from glass fibre/polypropylene composites by separating using xylene) composites provided higher tensile strength than for virgin composites. However, it has been identified that the handling of large amounts of solvents is associated with health, safety and environmental concerns (Baillie 2004). In addition, the use of solvent would be disadvantageous for natural fibre composites, as the fibre can be degraded by solvent and at high temperature.

12.3.2 Effects of recycling on thermal stability Beg and Pickering (2008a) studied thermal properties of composites using thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC). The melting temperature (Tm) of PP in composites was found to be reduced slightly from 171 °C for virgin composites to 167 °C for composites reprocessed eight times which was considered to be due to reduced molecular weight as a consequence of thermo-mechanical degradation and chain scission (Ramírez-Vargas et al. 2004; Beg and Pickering 2008a). Three stages of decomposition have been observed for virgin and reprocessed composites (Fig. 12.9), starting with dehydration and decomposition of volatile components at around 250 °C, followed by rapid weight loss for oxidative decomposition and finally slow decomposition corresponding to formation of char as the temperature increased. Kinetic parameters for the various stages of thermal degradation were determined from the TGA graphs using the following equation, given by Broido (1969): E ⎛ 1⎞ ⎛ RZ ⎞ T 2 ln ⎜ ln ⎟ = − a + ln ⎜ ⎝ y⎠ ⎝ Ea β max ⎟⎠ RT

[12.3]

where y is the fraction of non-volatilised material not yet decomposed, Tmax is the temperature of maximum reaction rate, β is the heating rate, Z is the frequency factor and Ea is the activation energy. Initially, plots of lnln(1/y) versus 1/T for various stages of decomposition were drawn and


319

found to be linear, suggesting good agreement with the Broido equation. The activation energies, Ea determined from the slopes of these plots are given in Table 12.1. Tmax and Ea for composites have been found to increase with increased number of times the materials were reprocessed. The positions of weight loss on the TGA traces of composites (Fig. 12.10) were also found to be shifted to higher temperatures with increased number of times Temp. diff./weight (°C/mg)

7 Virgin composites

6

8 times reprocessed composites

5 4 3 2 1 0 50

150

250

350

450

550

Temperature (°C)


12.9 DTA traces for virgin and reprocessed 40 wt% fibre composites (Beg and Pickering 2008a). Virgin composites

Weight (%)

100 80

8 times reprocessed composites

60 40 20 0 0

100

200 300 400 Temperature (°C)

500

600

12.10 TGA curves of virgin and reprocessed 40 wt% fibre composites (Beg and Pickering 2008a). Table 12.1 Thermal properties of composites (Beg and Pickering 2008)

Sample

Stage

Weight Temp. Activation energy loss (%) range (°C) Tmax (°C) Ea (kJ/mol)

Virgin composite

1st 2nd 3rd

61 31 7

226–351 351–436 436–508

285 371 455

85 68 60

8 times reprocessed 1st composite 2nd 3rd

30 52 15

230–340 340–447 470–512

289 412 470

87 71 81

320


the materials were reprocessed, suggesting increased thermal stability. The increase in thermal stability has been considered to be due to an increase in crystallinity of PP resulting from molecular weight reduction.

12.3.3 Effects of recycling on rheology


Viscosity is an important parameter which should be considered when processing composites. Bourmaud and Baley (2007) studied Newtonian limit viscosity (ηo) of reprocessed composites and found it to decrease with increased reprocessing cycles, which was proposed as relating to decrease in molecular weight. After six cycles, the decrease in ηo was 57% for the PP/glass fibre composite but higher for PP/vegetal fibre composite (72% for the PP/hemp composite and 65% for the PP/sisal composites). This difference was explained by the dispersion of fibres in different composites; the dispersion was better for PP/glass fibre composites whereas hemp/sisal fibres were seen to be aggregated. It was considered that the scission of aggregates was more efficient in decreasing viscosity than the scission of isolated fibres. Nair et al. (2000) also reported that fibre length has an important effect on viscosity of composites.

12.3.4 Durability of recycled composites The stability of composites over time is an important issue affecting their utilisation. Moisture penetration into composite materials occurs by three different mechanisms. The main process consists of diffusion of water molecules inside the microscopic gaps between polymer chains. The other mechanisms are capillary transport into the gaps and flaws at the interfaces between fibres and polymer due to incomplete wettability and impregnation, and transport by micro-cracks in the matrix, involving the flow and storage of water in the cracks, pores or small channels in the composite structure (Comyn 1985; Lin et al. 2002). Capillary transport is particularly significant when the interfacial adhesion is weak and when debonding of the fibres and the matrix has been initiated. Imperfections in the matrix can originate during the processing of the material, or due to environmental and service effects. Some conflicting results have been observed for recycled composites after hygrothermal ageing. Youngquist et al. (1994) studied wood fibre reinforced polyethylene composites and found moisture resistance of second-generation panels to be equivalent to, or better than, properties obtained from first-generation panels. This was explained as due to better encapsulation. However, Eriksson and Albertsson (1998) reported negative influence during accelerated ageing of glass fibre reinforced polyamide composites, indicating a lower resistance of recycled material toward

321

oxidative degradation and hydrolysis compared with that of virgin material. Beg and Pickering (2008b) reported that exposure of wood fibre composites to hygrothermal ageing resulted in slight deterioration of the surface texture in the form of colour fading. Fibre became more discernible (from the matrix) as a consequence of hygrothermal ageing for virgin and reprocessed composites which was considered to be due to the reduction of interfacial bonding. This was less apparent for the reprocessed composites. Thickness swelling was found for virgin and reprocessed composites after hygrothermal ageing, the extent of which was found to decrease with reprocessing, such that after ageing, the virgin composites showed an increase in swelling by 3.7% which reduced down to 2.2% for composites reprocessed eight times. Moisture absorption increased with increased soaking time for virgin and reprocessed composites until saturation at about 5 months (Fig. 12.11). As no significant weight gain was found for PP during this period, it seemed likely that moisture only penetrated into the composites through the fibre and fibre–matrix interface. Both the equilibrium moisture content and diffusion coefficient decreased with increased number of times the materials were reprocessed respectively from 9.42% and 2.54 × 10−13 m2/s for virgin composites to 6.41% and 1.01 × 10−13 m2/s for composites reprocessed eight times. The decrease in moisture content and diffusion coefficient with increased number of times the materials were reprocessed was explained by a number of effects. As the fibre length decreased with increased number of times the materials were reprocessed (discussed in previous section), it was considered that it would have been more difficult to form finite clusters which serve as passages for water molecules to travel through the lattice from one side to another (Wang et al. 2006; Beg and Pickering 2008a, b). Also, reduction of micro-voids as evaluated by the increased density of composites with increased reprocessing would be expected to result in a 14 Moisture content Mt (%)



Virgin 4 times reprocessed 8 times reprocessed

12 10

2 times reprocessed 6 times reprocessed PP

8 6 4 2 0 0

50

100

150

200

250

300

Soaking time (days)

12.11 Moisture content versus soaking time of virgin and reprocessed composites and PP (Beg and Pickering 2008b).


decrease in moisture content and diffusion coefficient. In addition, reprocessing increased the crystallinity of PP as a result of molecular weight reduction which was also considered to contribute to the reduction of moisture absorption. The effects on mechanical properties after hygrothermal ageing of recycled composites has been reported by Beg and Pickering (2008b). TS and YM have been observed to decrease after hygrothermal ageing for virgin and reprocessed composites. The percentage reduction in TS and YM due to ageing is presented in Fig. 12.12, and it can be seen that the extent of reduction in properties decreased with increased number of times the materials were reprocessed. After ageing, reductions in TS of 33% and YM of 40% were found for virgin composites compared with reductions for both TS and YM of 27% for composites reprocessed eight times. This was considered to be due to the equilibrium moisture content decreasing with increased number of times the materials were reprocessed, and therefore having less effect on behaviour. Failure strain and impact strength have been found to increase after hygrothermal ageing which was believed to be due to the water molecules acting as a plasticiser in the composite material (Beg and Pickering 2008b) as was also explained in another study carried out by Joseph et al. (2002). However, the extent of increase in failure strain and impact strength was generally found to decrease with increased number of times the materials were reprocessed (Beg and Pickering 2008b) which was also considered to be due to the reduction of equilibrium moisture content as explained above. Beg and Pickering (2008b) also studied the thermal stability of recycled composites after hygrothermal ageing. They found it to decrease after hygrothermal ageing both for virgin and reprocessed composites. The

45 Reduction of properties (%)


322

TS

40

YM

35 30 25 20 15 10 5 0 Virgin

2

4

6

8


12.12 Reduction in TS and YM of virgin and reprocessed composites after hygrothermal ageing (Beg and Pickering 2008b).


323

reduction of thermal stability has been explained to be due to the loss of structural integrity, reduction of molecular weight and debonding of the fibre from the matrix, resulting from the development of shear stress at the interface due to absorbed moisture. However, the extent of reduction of thermal stability after hygrothermal ageing was found to be less for reprocessed composites than for virgin composites. This was also supported by crystallinity index where a 25% reduction of PP crystallinity was found for virgin composites and a 10% reduction was found for composites reprocessed eight times after hygrothermal ageing.


12.4

Conclusions

Currently, the desire is ahead of the ability to recycle plastic matrix composites and a number of different strategies are under investigation. For thermoset matrix composites mechanical breakdown of material is the most progressed, with a number of companies now carrying this out on an industrial scale. Although the finer grade particulates can be used to replace filler, leading to some modest performance improvement, the challenge still remains to avoid down-cycling within ideally the same processing stream or to find sufficient markets for the range of mechanically recycled grades. However, improvement of mechanical performance has been obtained by using mechanical recyclate in base resin systems, concrete and thermoplastics. Mechanical breakdown is also the most progressed recycling method for thermoplastic matrix composites. Indeed, here there is a greater potential than for thermoset composites recycling due to better property retention. Research based on reprocessing of thermoplastic matrix composites shows some reduction of tensile strength and Young’s modulus, with poor surface appearance but increased failure strain and better moisture resistance. Reductions of fibre length and polymer degradation have been highlighted to explain the change in properties of reprocessed composites. Overall, issues of contamination and separation will also need to be addressed for increased mechanical recycling and therefore the development of infrastructure and systems. Thermal recycling has shown great promise for thermoset matrix composites. Here there is less concern regarding contamination and the need for sorting. Industrial-scale thermal recycling is currently coming on-line, particularly for the recycling of the more expensive carbon fibre from composites. Fibre extracted using thermal recycling has shown greater potential to be used in primary production than that through mechanical recycling. Chemical recycling is the least developed method of those investigated for both thermoset and thermoplastic composites. Although it has been shown possible to extract fibre with acceptable properties, chemical

324


recycling raises the problem of how to deal with the waste chemicals produced, as well as health, safety and environmental concerns and so currently could create more problems than it solves. As time progresses, increased landfill and petroleum prices along with increased legislation and technical progress would all be expected to change the balance in favour of increased recycling. In addition, companies are more likely to design with reuse in mind to enable easier recycling.


12.5

References

allred, r. e. (1996). ‘Recycling process for scrap composites and prepregs.’ SAMPE Journal 32(5): 46–51. allred, r. e. and l. d. busselle (2000). ‘Tertiary recycling of automotive plastics and composites.’ Journal of Thermoplastic Composite Materials 13(2): 92–101. allred, r. e. and a. b. coons (1996). Properties of carbon fibers reclaimed from composite manufacturing scrap by tertiary recycling. 28th International SAMPE Technical Conference, Seattle, Washington, SAMPE. allred, r. e., t. j. doak, et al. (1997). Recycling Process for Automotive Plastics and Composites. Proceedings of the American Society for Composites Twelfth Technical Conference, Michigan. anonymous (2006). Carbon Composite Recycling Turns from Dream to Reality. Netcomposites: www.milledcarbon.com. anonymous (2008). Recycling Advanced Composites, Clean Washington Center: www.cwc.org/industry/ibp953rpt.htm. baillie, c. (2004). Green Composites: Polymer Composites and the Environment, Woodhead Publishing Limited. beg, m. d. h. and k. l. pickering (2008a). ‘Reprocessing of wood fibre reinforced polypropylene composites. Part I: Effects on physical and mechanical properties.’ Composites Part A: Applied Science and Manufacturing 39(7): 1091–1100. beg, m. d. h. and k. l. pickering (2008b). ‘Reprocessing of wood fibre reinforced polypropylene composites. Part II: Hygrothermal ageing and its effects.’ Composites Part A: Applied Science and Manufacturing 39: 1565–1571. bernasconi, a., d. rossin, et al. (2007). ‘Analysis of the effect of mechanical recycling upon tensile strength of a short glass fibre reinforced polyamide 6,6.’ Engineering Fracture Mechanics 74(4): 627–641. bledzki, a. k. and k. goracy (1993). ‘The use of recycled fibre composites as reinforcement for thermosets.’ Mechanics of Composite Materials 29(4): 352–356. blizard, k. (1998). ‘A wholly recycled structural plastic lumber incorporating scrap prepreg waste.’ Plastics in Building Construction 22(5): 8–12. bourmaud, a. and c. baley (2007). ‘Investigations on the recycling of hemp and sisal fibre reinforced polypropylene composites.’ Polymer Degradation and Stability 92(6): 1034–1045. bream, c. e. and p. r. hornsby (2000). ‘Structure development in thermoset recyclate-filled polypropylene composites.’ Polymer Composites 21(3): 417–435. bream, c. e., e. hinrichsen, et al. (1997). ‘Integrated compounding technology for preparation of polymer composites reinforced with waste material.’ Plastics, Rubber and Composites Processing and Applications 26(7): 303–310.



325

broido, a. (1969). ‘A simple, sensitive graphical method of treating thermogravimetric analysis data.’ Journal of Polymer Science, Part A-2 7: 1761. buggy, m., l. farragher, et al. (1995). ‘Recycling of composite materials.’ Journal of Materials Processing Technology 55(3–4): 448–456. butler, k. (1991). ‘Recycling of molded SMC and BMC materials.’ 46th Annual Conference, Composites Institute, The Society of the Plastics Industry, Washington. chu, j. and j. l. sullivan (1996). ‘Recyclability of a glass fibre poly(butylene terephthalate) composite.’ Polymer Composites 17(3): 523. comyn, j. (1985). Polymer Permeability. Elsevier Applied Science, New York: 383. conroy, a., s. halliwell, et al. (2004). ‘Recycling fibre reinforced polymers in the construction industry.’ BRE information paper; IP 4/04. London, BRE. cunliffe, a. m., n. jones, et al. (2003). ‘Pyrolysis of composite plastic waste.’ Environmental Technology 24(5): 653–663. curcuras, c. n., a. m. flax, et al. (1991). Recycling of thermoset automotive components. International Congress and Exposition, Detroit, Michigan. derosa, r., e. telfeyan, et al. (2004). ‘Expanding the use of recycled SMC in BMCs.’ GPEC 2004 BMCs Global Plastics Environmental Conference. derosa, r., e. telfeyan, et al. (2005). ‘Current state of recycling sheet molding compounds and related materials.’ Journal of Thermoplastic Composite Materials 18(3): 219–240. eriksson, p. a. and a. c. albertsson (1996). ‘Prediction of mechanical properties of recycled fiberglass reinforced polyamide 66.’ Polymer Composites 17(6): 830–839. eriksson, p. a. and a. c. albertsson (1998). ‘Durability of in-plant recycled glass fiber reinforced polyamide 66.’ Polymer Engineering and Science 38(2): 348–356. european commission (2000). End-of-Life Vehicle (ELV) Directive, 2000/ 53/EC. european commission (2003). Waste Electrical and Electronic Equipment (WEEE) Directive 2002/96/EC. harper, l. t., t. a. turner, et al. (2006). ‘Characterisation of random carbon fibre composites from a directed fibre preforming process: The effect of fibre length.’ Composites Part A: Applied Science and Manufacturing 37(11): 1863–1878. inoh, t., t. yokoi, et al. (1994). ‘SMC recycling technology.’ Journal of Thermoplastic Composite Materials 7: 42–55. jiang, g., s. j. pickering, et al. (2008). ‘Surface characterisation of carbon fibre recycled using fluidised bed.’ Applied Surface Science 254(9): 2588–2593. joseph, k., s. thomas, et al. (1993). ‘Tensile properties of short sisal fibre-reinforced polyethylene composites.’ Journal of Applied Polymer Science 47: 1731–1739. joseph, p. v., m. s. rabello, et al. (2002). ‘Environmental effects on the degradation behaviour of sisal fibre reinforced polypropylene composites.’ Composites Science and Technology 62(10–11): 1357–1372. jutte, r. b. and w. d. graham (1991). ‘Recycling SMC.’ 46th Annual Conference, Composites Institute, The Society of the Plastics Industry, Washington. kennerley, j. r., r. m. kelly, et al. (1998). ‘The characterisation and reuse of glass fibres recycled from scrap composites by the action of a fluidised bed process.’ Composites Part A – Applied Science and Manufacturing 29(7): 839–845.


326


kojima, a. and s. furukawa (1997). ‘Recycling of resin matrix composite materials VII: Future perspective of FRP recycling.’ Advanced Composite Materials 6(3): 215–225. kouparitsas, c. e., c. n. kartalis, et al. (2002). ‘Recycling of the fibrous fraction of reinforced thermoset composites.’ Polymer Composites 23(4): 682–689. lester, e., s. kingman, et al. (2004). ‘Microwave heating as a means for carbon fibre recovery from polymer composites: a technical feasibility study.’ Materials Research Bulletin 39(10): 1549–1556. lin, q., x. zhou, et al. (2002). ‘Effect of hydrothermal environment on moisture absorption and mechanical properties of wood flour-filled polypropylene composites.’ Journal of Applied Polymer Science 85(14): 2824–2832. nair, k. c. m., r. p. kumar, et al. (2000). ‘Rheological behavior of short sisal fiberreinforced polystyrene composites.’ Composites Part A: Applied Science and Manufacturing 31(11): 1231–1240. ogi, k., t. shinoda, et al. (2005). ‘Strength in concrete reinforced with recycled CFRP pieces.’ Composites Part A – Applied Science and Manufacturing 36(7): 893–902. papaspyrides, c. d., j. g. poulakis, et al. (1995). ‘Recycling of glass fibre reinforced thermoplastic composites. I. Ionomer and low density polyethylene based composites.’ Resources, Conservation and Recycling 14: 91–101. patel, s. h., k. e. gonsalves, et al. (1993). ‘Alternative procedures for the recycling of sheet molding compounds.’ Advances in Polymer Technology 12(1): 35–45. perrin, d., l. clerc, et al. (2008). ‘Optimizing a recycling process of SMC composite waste.’ Waste Management 28(3): 541–548. petterson, j. and p. nilsson (1994). ‘Recycling of SMC and BMC in standard processing equipment.’ Journal of Thermoplastic Composite Materials 7(1): 56–63. pickering, s. j. (2006). ‘Recycling technologies for thermoset composite materials – current status.’ Composites Part A – Applied Science and Manufacturing 37(8): 1206–1215. pickering, s. j., r. m. kelly, et al. (2000). ‘A fluidised-bed process for the recovery of glass fibres from scrap thermoset composites.’ Composites Science and Technology 60(4): 509–523. poulakis, j. g., p. c. varelidis, et al. (1997). ‘Recycliing of polypropylene based composites.’ Advances in Polymer Technology 16(4): 313–322. ramírez-vargas, e., d. navarro-rodríguez, et al. (2004). ‘Degradation effects on the rheological and mechanical properties of multi-extruded blends of impact-modified polypropylene and poly(ethylene-co-vinyl acetate).’ Polymer Degradation and Stability 86(2): 301–307. rempes, p. (2003). Composite recycling and disposal. An environmental R&D issue. Environmental Technotes, Boeing, 8. rowell, r. m., r. a. young and j. k. rowell (1997). Paper and Composites from Agro-based Resources. Boca Raton, FL, CRC Press. rush, s. (2007). Carbon fiber: life beyond the landfill. Industry News. G. Publications, Zoltek: www.zoltek.com/industrynew/15/. shizu, k., i. osamu, et al. (1997). ‘Method of recycling unsaturated polyester resin waste and recycling apparatus.’ Journal of Cleaner Production 5(4): 308. thomas, r., f. j. guild, et al. (2000). The Dynamic Properties of Recycled Thermoset Composites. FRC 2000, UK, Woodhead Publishing Ltd.


327


walz, k., r. jacobson, et al. (1994). ‘Effect of reprocessing/recycling on the mechanical properties of kenaf–PP composites,’ internal report, University of Wisconsin-Madison and Forest Product Laboratory. wang, w., m. sain, et al. (2006). ‘Study of moisture absorption in natural fiber plastic composites.’ Composites Science and Technology 66(3–4): 379–386. williams, p. (2003). ‘Recycling tricky materials using pyrolysis.’ Materials World July: 24–26. williams, p. t., a. cunliffe, et al. (2005). ‘Recovery of value-added products from the pyrolytic recycling of glass-fibre-reinforced composite plastic waste.’ Journal of the Energy Institute 78(2): 51–61. winter, h., h. a. m. mostert, et al. (1995). ‘Recycling of sheet-molding compounds by chemical routes.’ Journal of Applied Polymer Science 57(11): 1409–1417. youngquist, j. a., g. e. myers, et al. (1994). Composites from Recycled Wood and Plastics. US Environment Protection Agency Cincinnati, OH. zafeiropoulos, n. e., p. c. varelidis, et al. (1999). ‘Characterisation of LDPE residual matrix deposited on glass fibres by a dissolution/reprecipitation recycling process.’ Composites Part A: Applied Science and Manufacturing 30(7): 831–838.

13 Clean and environmentally friendly wet-filament winding N. S H O T T O N - G A L E, D. H A R R I S, S. D. PA N D I TA , M. A. PAG E T, J. A. A L L E N and G. F. F E R N A N D O, University of Birmingham, UK


Abstract: This chapter reports on a modified wet-filament winding technique where the components of the resin system were stored in individual reservoirs and pumped on-demand through a static-mixer; the mixed resin system was then fed to a custom-designed fibre impregnation unit. The quality of the components produced using the conventional and modified techniques were evaluated using image analysis and the split-ring test. The fibre volume and the void fractions were also measured. It was found that the modified wet-filament winding technique yielded tubes with equivalent properties in comparison with conventionally wound tubes. However, the volume of solvents consumed was reduced significantly. Key words: filament winding, solvents, environmentally friendly manufacturing, impregnation.

13.1

Introduction

13.1.1 Conventional wet-filament winding Conventional wet-filament winding is used to manufacture fibre-reinforced components such as pipes and pressure vessels. There are a number of variations in this manufacturing technique including the following: (i) resin bath-based impregnation;1 (ii) wet- and dry-tape winding;2–4 (iii) electrostatic deposition;5,6 and (iv) laser-assisted impregnation.2,7,8 The resin bathbased impregnation method is the predominant technique used in industry and generally involves the use of thermosetting resin systems. The term ‘resin system’ is used here to describe an intimately mixed resin and hardener. A schematic illustration of conventional wet-filament winding is shown in Fig. 13.1. With reference to Fig. 13.1, the reinforcing fibre tows from the creels (A) are fed through a tensioning device and a series of guides (B) to a resin bath (C). The individual components of the resin system (for example, epoxy and amine) are mixed manually in the required stoichiometric proportions before being introduced to the resin bath. The 331

332

Management, recycling and reuse of waste composites A

B

C

D

E

F


13.1 Schematic illustration of the conventional wet-filament winding process. The key components are coded as follows: (A) fibre creels; (B) tensioning and guiding systems; (C) resin bath; (D) resin-impregnated fibre bundles; (E) traversing carriage; and (F) rotating mandrel.

impregnated fibres (D) go to a traversing carriage (E) and are wound onto a rotating mandrel (F). The carriage traverses horizontally along the length of the rotating mandrel, laying resin-impregnated fibres in a predetermined fashion onto the mandrel. The winding angle of the fibres is controlled as a function of traverse rate and rotation speed of the mandrel. There are a number of issues associated with conventional wet-filament winding including the following: •

•

Pot-life of the pre-mixed resin system: Mixed resin systems have a finite pot-life, after which the viscosity of the resin increases and the fibre impregnation process becomes progressively more difficult. The limited pot-life also means that there is a possibility of the resin system setting or cross-linking into a solid in the processing equipment. The crosslinked resin has to be removed prior to the resumption of production. The removal of the cross-linked resin from the processing equipment can be a tedious, time-consuming and costly operation. As the ambient temperature can influence the viscosity and cross-linking rate of thermosetting resins, the limited pot-life means that low temperaturecurable resins are not generally suitable for conventional wet-filament winding. Solvents: A major issue with conventional wet-filament winding is the need for the equipment to be cleaned thoroughly with a copious volume of solvent at the end of each production run. This has knockon effects such as the need to recover the solvent prior to disposal of the waste resin, and the need for adequate ventilation and personal

Clean and environmentally friendly wet-filament winding

•


•

•

333

protective equipment for the workforce. Legislation also dictates the exposure limits for the workforce with regard to specified chemicals and solvents. Resin bath: The resin and hardener are weighed and mixed manually prior to transferral to the resin bath. During filament winding the resin bath has to be replenished manually. Open-top resin baths can result in significant emissions of low-molecular weight components from the resin system to the atmosphere. Excess resin: The excess resin remaining in the bath after a filament winding operation is typically transferred to a disposable container and allowed to cross-link to a solid before disposal. Precautions have to be taken to avoid storing or cross-linking a large volume of mixed resin in a single operation as this can result in the resin exotherming. In other words, the cross-linking reaction can become auto-catalytic as it proceeds. This can result in a significant increase in the temperature of the resin system, leading to thermal degradation and emission of potentially toxic gaseous by-products. The volume of waste resin generated in the wet-filament winding process will depend on a number of factors, for example, the capacity of the resin bath used. On-site manufacturing practices such as over-impregnating the reinforcing tows and subsequent removal of the excess resin can also influence the volume of waste resin generated. Integration of embedded sensors: In the conventional wet-filament winding process, the reinforcing fibres are pulled through a complex series of guides. With reference to the increased utilisation of embedded sensor systems for process and structural health monitoring, optical fibre sensors tend to be too fragile for on-line embedment. In other words, it is generally necessary to stop the filament winding process to locate and secure the sensors. It is likely that a reduction in the number of guides will lead to lower tension in the tows and thus reduce the probability of damage caused to the optical fibres; this may also facilitate an increase in the production rate of filament wound components.

The following section presents a discussion on the approach that was taken to overcome some of the above-mentioned issues associated with conventional wet-filament winding. The modified manufacturing technique will be referred to as ‘clean filament winding’.

13.1.2 Clean filament winding: concept and advantages The key components of the clean filament winding concept are illustrated in Fig. 13.2, and a schematic representation of this manufacturing technique is shown in Fig. 13.3. The same coding system is used in both figures.

334


Resin

Hardener

Pump 1

Pump 2

(i)

(xi)

Manifold

(vi)

Feedback control unit

Static mixer

(ii)

Fibre optic probe

(iii)

(iv)

(v)

Resin impregnation unit


Mandrel

(viii) Fibre spreading station

(vii) Tows

13.2 Key components of the clean filament winding manufacturing concept (see pages 335–337 for details).

(ix) (vii)

(x) (xi) (viii) (ii)

(iv) (xii) (xiii) (v)

(vi)

13.3 Schematic illustration of the clean filament winding station (see pages 337–341 for details).


335

With reference to the issues mentioned in Section 13.1.1 relating to conventional wet-filament winding, the function of each of the components shown in Fig. 13.2 is described below:


(i)

(ii)

(iii)

Resin containment and delivery: Here the resin and hardener are contained in separate reservoirs and pumped on-demand via precision gear-pumps to a conventional KenicsTM static-mixer. The deployment of the gear-pumps enables the throughput and the stoichiometry of the resin and hardener to be controlled accurately. The feed for the precision gear-pumps can also be supplied directly from the containers or drums that were used to deploy the materials. The primary advantage here is that the reservoir will not need to be replenished manually. Desiccants are placed within the reservoir to minimise contamination of the resin and hardener by atmospheric moisture. The resin containment and delivery system described here addresses the problems associated with manually mixing the resin and hardener components. The containment of the two components in enclosed reservoirs also avoids the issues associated with emissions, for example, from open-top resin baths. Mixing the resin and hardener: In the clean filament winding process, the resin and hardener are ‘mixed’ intimately via a static-mixer. The static-mixer consists of a series of helical elements which are fixed within a tubular housing. The consecutive elements oppose each other and are welded together such that the adjacent edges are perpendicular. As a consequence of this, the fluid is split every time it leaves one element and enters another. This process continues along the length of the static-mixer. The striation thickness is reduced by a factor 2N where N = the number of splitting elements. With the appropriate selection of the number of helical elements in the static-mixer, a good homogenisation of the resin system can be attained.9 The deployment of the static-mixer negates the need to mix the resin and hardener manually. Control of stoichiometry: This is an optional item where, if necessary, fibre optic probes can be used to monitor the relative concentrations of the amine and the epoxy in the resin system (see Chapter 14). This can be achieved using a conventional Fourier transform infrared (FTIR) spectrometer or light sources and detectors that operate at specific wavelengths corresponding to the absorbance of amine, epoxy and an inert functional group such as C–H.10 A feedback control system can be implemented to adjust the throughput from the precision pumps as a function of the output from the fibre optic probe. Sensor-based control of the stoichiometric ratio of the

336


(iv)

(v)

(vi)

Management, recycling and reuse of waste composites amine and the epoxy is likely to be more reliable and repeatable than manual weighing and mixing. Resin impregnation unit: Unlike conventional wet-filament winding where a resin bath is used to impregnate the fibres, in the clean filament winding process a custom-designed resin impregnation unit is deployed. The resin impregnation unit is made up of three key components. The first is a facility to spread out the filaments in the fibre tow. The fibre spreading stations are located before and within the resin impregnation unit and they effectively reduce the ‘thickness’ of the tow whilst increasing its overall width. This is important because it enhances the impregnation rate significantly. If rapid impregnation of the tows can be assured, the impregnation zone can be reduced significantly when compared with pin or drum-based impregnation using a resin bath. Hence, the overall volume of the resin impregnation unit can be reduced significantly. The implication here is that the volume of solvent required to clean the impregnation unit is likely to be a fraction of that required to clean the resin bath at the end of a production run. Moreover, the resin impregnation unit can be designed to be an enclosed system; this will reduce the emissions when compared with an open-top resin bath. The second component is a resin injection unit where the mixed resin from the static-mixer is injected into the fibre tows. The injection unit operates as a miniature combined pin-based and injection-based impregnator. Within the resin injection system unit there is a facility to introduce the resin to both the top and bottom of the fibre tow. The final component is a pseudo-doctor-blade that removes any excess resin from the fibre tow and feeds it back into the resin injection reservoir. Mandrel and location of the resin impregnation unit: In the conventional wet-filament winding process, the resin bath is generally located at a distance from the mandrel. Therefore, there is a high probability of the resin system dripping from the impregnated tows before they are wound onto the rotating mandrel. In the clean filament winding process, the resin impregnation unit is placed in close proximity to the mandrel. Therefore the possibility of the resin dripping onto the floor does not arise. Feedback control: A feedback control system between the rotating mandrel and the resin dispensing unit is used to control the relative throughput of the resin and hardener. The option for monitoring the stoichiometry of the resin and hardener was mentioned previously. Another option that can be implemented is a simple transmission/reflection fibre optic probe to monitor the degree of impregnation of the fibre tows; this will further maximise the efficiency of the process.


337

(vii)

Fibre bundle or tows: There are no specific requirements for the fibre bundles that can be used in the clean filament winding process. The creels can be ‘centre-pull’ or ‘outside draw’. The former is where the bundle is drawn from the centre of the creel and in the latter case, it is drawn from the outer circumference of the creel. (viii) Fibre spreading station: This is a key component of the clean filament winding technique where the filaments within each bundle are spread by mechanical and/or electrostatic manipulation before they enter the resin impregnation unit. As mentioned previously, this effectively increases the width of the tow and concomitantly reduces the nominal thickness to enhance the impregnation rate of the fibres by the mixed resin system.


13.1.3 Clean filament winding: equipment summary Figure 13.3 shows a schematic representation of the clean filament winding process that was previously shown in Fig. 13.2; the same coding system has been used to represent the key components. With reference to Fig. 13.3, (ix) represents the reinforcing fibres; these are sometimes referred to as creels or bobbins. The fibre tows (vii) are fed through fibre-guide pulleys (x) which control the trajectory of the tows. The pulleys also commence the process of mechanically-induced fibre spreading. The tows from the pulleys are directed to a series of fibre spreading units (viii) (only one is shown in Fig. 13.3) which consist of profiled rollers and/or pins where the fibre tows are spread out. The spread fibres are directed to the resin injection unit (iv) which is connected to the static-mixer (ii) and the resin delivery manifold (xi). The resin and hardener are supplied from the resin dispensing system via the resin delivery manifold. Item (xii) represents the traversecarriage that oscillates across the length of the mandrel. The impregnated fibre tows are gathered by a ‘fibre gather’ (xiii) and delivered to the rotating mandrel (v). The relative speeds of the traverse-carriage and the mandrel dictate the angle at which the fibres are laid down on the mandrel. Item (vi) represents the feedback control system that links and synchronises the rotation speed of the mandrel with the resin dispensing unit. Clean filament winding: equipment details and rationale With reference to Fig. 13.3, a more detailed discussion based on the individual components of the clean filament winding station is presented below: •

Reinforcing fibres (item ix in Fig. 13.3): The reinforcement in the form of continuous fibres are drawn from the outer circumference of the creel. Apart from the nature of the reinforcement (glass, graphite, polyaramid, polyethylene), the nature of the ‘binder’ is also

338

•


•

•

•

•


an important factor. This is because it can influence the degree of fibre spreading that can be induced in the tow. The composition of the binder can also influence the wettability or receptiveness of the resin system towards the reinforcement. Fibre-guide pulleys (item x in Fig. 13.3): The pulleys are pivoted and they perform two functions. Firstly, they control the trajectory of the fibres before they are spread out and impregnated. Secondly, they aid in breaking up the binder. The term binder is used here to describe the processing aids including coupling agents that are applied to the fibres at the production stage. Fibre spreading station (item viii in Fig. 13.3): Over the course of the development of the clean filament winding method, techniques based on pins, rollers, profiled rollers and electrostatics were evaluated for spreading the filaments in a tow. Profiled or bevelled rollers were found to be the most effective in inducing fibre spreading in commercially available reinforcing glass fibres. However, control of the fibre trajectory over the profiled roller was found to be an important factor. Pins are also effective in inducing fibre spreading but can result in increased tension in the fibres. As mentioned previously, the nature of the binding on the fibres has a major influence on the ease with which the individual filaments can be separated without causing excessive damage to the reinforcement. Other factors that can have a major influence on spreading the fibres include: (a) the fibre Tex; (b) the degree of ‘twist’ in the fibre bundle; (c) the tension applied to the fibres; and (d) the fibre hauloff rate. Resin delivery system (items xi and ii in Fig. 13.3): The resin dispenser used in this study (shown in Fig. 13.4) was developed and supplied by Dalling Automation Ltd. It consisted of two individual precision gearpumps (Fig. 13.4 iv) capable of handling liquids in the viscosity range of 20 to 10 000 mPa s and a throughput range of 10 and 110 g min−1. The stoichiometric ratio of the two components (epoxy resin and amine hardener) was controlled by the throughput of the individual pumps. The resin dispensing unit had an option for heating the resin and hardener reservoirs (Fig. 13.4 ii and iii). This is a useful option because it offers a means for controlling the viscosity, irrespective of the environmental temperature. Static-mixer (item ii in Fig. 13.3): These devices are available in a range of dimensions and designs. In the clean filament winding process, the static-mixer was connected to the resin dispensing unit via a manifold (item xi in Fig. 13.3). The opposite end of the static-mixer was connected directly to the resin impregnation unit (item iv in Fig. 13.3). Resin impregnation unit (item iv in Fig. 13.3): The evolution of the design for the resin impregnation unit was based on the following requirements:


339

(i)

(ii)

(iii)

(iv)


13.4 Schematic illustration of the clean filament winding resin delivery system. The labeled items are: (i) feedback control unit, (ii) resin reservoir, (iii) hardener reservoir and (iv) gear pumps. 䊊

䊊

䊊

A facility to spread the fibres prior to impregnation: The rationale for this is given in Section 13.6.1 where the effect of the tow ‘thickness’ on the impregnation time is discussed. The hypothesis here was that any decrease in the effective thickness of the tows would result in accelerating the through-thickness impregnation of the tows. A facility to inject the mixed resin on the top and the bottom of the fibre tows: The requirement here was to combine the mode of operation of a resin bath and a resin injection system within the resin impregnation unit. In other words, in the case of the resin bath, the resin system on the drum is squeezed into the tows; in the resin injection method, the mixed resin is injected under low-pressure into the tows. The desired outcome was to enhance the through-thickness impregnation rate. Minimising the volume of the resin impregnation unit: The rationale here was that a reduction in the volume of the resin impregnation unit would mean that the volume of solvent required to clean the equipment at the end of a production run would be significantly lower. In the case of conventional resin baths, the ‘dead-volume’ is significantly large and hence the volume of solvent required to clean the equipment is also high. Depending on the intricacy of the design of the resin impregnation unit, non-stick coatings or mould-inserts can be used to minimise the contact between the resin system and the metal housing. Here the

340


•

•

•


resin can be cross-linked or cured in the non-stick mould and then disposed as waste. Another reason for minimising the volume of the resin impregnation unit was to reduce its overall mass. This was desirable because it would make it easier for the resin impregnation unit to be mounted on the traverse-carriage (see Fig. 13.3). This also meant that it would be relatively simple to retrofit the resin impregnation unit onto any commercial filament winding machine. 䊊 Locating the resin impregnation unit near the mandrel: The location of the resin impregnation unit in close proximity to the mandrel was discussed previously. With reference to retrofitting the resin impregnation unit on conventional wet-filament winding machines, it can be mounted on the platform that houses the resin bath or the traverse-carriage. 䊊 A modular design for the resin impregnation unit: This was deemed desirable to enable easy assembly/disassembly and flexibility in accommodating different fibre types and number of tows. Traverse-carriage (item xii in Fig. 13.3): With reference to Fig. 13.1, conventional wet-filament winding machines possess a carriage that traverses repeatedly along the length of the mandrel. As shown in Fig. 13.3, the traverse-carriage provides an ideal platform to retrofit the resin impregnation unit. The resin feed to the unit can be via a pair of flexible and insulated hoses. Impregnated fibre collector and mandrel respectively (items xiii and v in Fig. 13.3): In Fig. 13.3, these items represent the device that collects the impregnated fibres (xiii) and the mandrel (v). The ‘fibre collector’ is capable of motion in different planes. Feedback control system (item vi in Fig. 13.3): The feedback control system provides a simple feedback mechanism to enable the throughput of the resin and hardener dispensing pumps to be controlled in proportion to the rotation rate of the mandrel. The key components of this unit are as follows: 䊊 Sensor: The speed of rotation of the mandrel is sensed by a bespoke rotational sensor located on the main drive shaft of the filament winding machine. The sensor comprises of a rotating disk located on the drive shaft itself. The disk contains a series of 20 magnets positioned uniformly about its circumference. A uni-polar nonlatching Hall sensor detects the proximity of the magnet, outputting a logic level signal. Hence, the sensor output consists of a train of pulses whose frequency is proportional to the rotation speed of the mandrel. 䊊 Frequency-to-voltage converter: The process of converting the pulsetrain from the sensor into a voltage which can be used to drive the



䊊

341

pump remote control input was achieved using an off-the-shelf ‘frequency-to-analogue’ unit. This had programmable calibration features such as pulse frequency input range (corresponding to full scale output) and output format (4–20 mA, 0–5 V or 0–10 V). In this application, the input range can be matched to the maximum expected rotational rate of the mandrel, and can be changed at a later date if required. The output format can be matched to the required drive signal level for the dispensing pumps. Output to the dispensing pump system: The output of the frequencyto-voltage converter is a signal of 0 to 10 V. This corresponds directly to the input range (0 to maximum) for the dispensing pumps. In order therefore to allow trimming of the constant of proportionality between the rotational speed and the dispensing rate, and to allow the ratio of resin to hardener to be adjusted, the output of the frequency to voltage converter first drives a ‘constant of proportionality’ setting potentiometer. This subsequently drives two parallel ‘ratio’ setting potentiometers. The outputs of these latter two potentiometers provide the drive signals to the dispensing pumps (remote control input). The potentiometer circuitry is a direct copy of that already used for the local control in the dispensing pump drive controller.

In summary, the clean filament winding technique was conceived to overcome the perceived limitations of conventional wet-filament winding. The rationale for the evolution of the design was considered and the requirements for each component were discussed. Prior to presenting the specific design details of resin impregnation units, the following section gives a brief overview of previous publications which have modelled the impregnation process. These models were used to aid the design of the prototype resin impregnation units.

13.2

Resin impregnation modelling

A schematic illustration presenting an overview of the various models that were considered for the design of the resin impregnation unit is shown in Fig. 13.5. The interpretation of Fig. 13.5 is as follows: the majority of the models that have been developed for predicting the permeability and time required to achieve impregnation of fibres are based on Darcy’s equation (see Equation 13.1). There are four key components to this equation: permeability; dimensions of the reinforcement; viscosity; and pressure. In the context of clean filament winding, two previously published models were reviewed and adapted, namely, those derived by Foley and Gillespie11 and Gaymans and Wevers.12

342

Management, recycling and reuse of waste composites (A) Darcy’s equation – general basis of modelling impregnation

(B1) Foley and Gillespie11 (B) Derived impregnation models

(A1) Permeability

(A1.1) Axial permeability

(A2) Fibre bundle dimensions

(A3) Viscosity

(A1.2) Transverse permeability

(A4) Pressure

(A4.1) Capillary pressure

(B2) Gaymans and Wevers12

(A4.2) Applied pressure


13.5 A schematic illustration presenting an overview of the various models that were considered for the design of the resin impregnation unit.

In the following section, each of the components of Darcy’s equation (A1–A4) are discussed in turn before adapting and applying the models developed by Foley and Gillespie (B1) and Gaymans and Wevers (B2) to the clean filament winding process. In general, the majority of the studies presented in this review considered the resin to be an incompressible Newtonian fluid that permeated through a porous medium (fibre array). The starting point for the majority of the models was Darcy’s equation: v=

K ΔP ηε L

[13.1]

where ¯v is the superficial velocity that can be observed on a macroscopic scale, K is the permeability of the porous medium, ε is porosity, η is the viscosity of the fluid and ΔP/L is the pressure gradient over a characteristic dimension L. In the context of developing the design basis for the resin impregnation unit, the following sections present a brief overview of selected models that considered the four components of Darcy’s equation.

13.2.1 Permeability (A1) Axial permeability (A1.1) Gebart13 predicted the axial permeability of a fibre bundle by calculating the frictional factor λ, of axial flow along a duct that was formed in the


343

interstitial space between a fibre bundle. The frictional factor λ was derived analytically for specified cross-sections (circular, quadratic, hexagonal, etc.) and was calculated using the following relationship:

λ=

ΔP 2 Dh L ρU 2

[13.2]

λ=

c Re

[13.3]

where


In Equation 13.2, ΔP/L is the pressure gradient, Dh is hydraulic diameter (duct cross-sectional area divided by the wetted perimeter), ρ is fluid density and U is the mean resin velocity over the fibre cross-section. In Equation 13.3, c is a dimensionless shape factor and Re is Reynolds number. By elaborating the frictional factor, Gebart13 derived the axial permeability, Kx, as: Kx =

3 8rf2 (1 − Vf ) 2 c Vf

[13.4]

where rf is fibre radius, c is equal to 57 and 53 for quadratic and hexagonal fibre arrays respectively, and Vf is fibre volume fraction. Further models considering axial permeability are summarised in Table 13.1 but are not discussed further in this chapter.

Table 13.1 A summary of selected models reported in the literature that considered axial permeability Authors

Axial permeability (Kx)

Amico and Lekakou14

Kx =

Carman–Kozeny15

3 rf2 (1 − Vf ) 2 4k Vf rf2 ⎡ ⎛ 1 ⎞ Kx = × ln ⎜ ⎟ − (3 − Vf ) (1 − Vf ) ⎤⎥ 8Vf ⎣⎢ ⎝ Vf 2 ⎠ ⎦ [B (VA ) +C (VA )Vf ] e K x = rf2 Vf[m (VA )]

Cai and Berdichevsky16 Berdichevsky and Cai17

Kx =

ηε ah Pc

(13.5) (13.6) (13.7) (13.8)

344


Transverse permeability (A1.2) Gebart13 also investigated the resistance to transverse flow that occurred between individual fibres. It was reported that if the fibres were in intimate contact, they formed a channel with an undulating area between them. However, this variation in the cross-sectional area was assumed to be negligible. As a result, inertia effects were not considered. Furthermore, when a constant pressure differential was applied between these two regions, the pressure gradients were said to vary slowly in relation to the resin flow direction; the velocity profile Vp, was considered to be approximately parabolic at each flow position and this was calculated using Equation 13.9: Vp =

H(1 2) 2 dP ⎛ y 2 ⎞ − 1⎟ ⎠ 2η dx ⎜⎝ H(1 2) 2

[13.9]


where H(1/2) is the channel half-height, η is the resin viscosity, and x and y are the vertical and horizontal coordinates of the flow position respectively. By elaborating Equation 13.9 and taking the geometry of the fibre arrays to be quadratic or hexagonal, Gebart13 derived equations for predicting the transverse permeability: 52

K y,quadratic =

16rf2 ⎛ VA ⎞ − 1⎟ ⎜ ⎠ 9π 2 ⎝ Vf

K y,hexagonal =

16rf2 ⎛ VA ⎞ − 1⎟ ⎜ ⎠ 9π 6 ⎝ Vf

[13.10] 52

[13.11]

where VA is the maximum packing capacity of a fibre bundle. Table 13.2 presents a summary of additional models which also predict the transverse permeability of a fibre bundle.

Table 13.2 A selection of additional models which predict transverse permeability Authors

Transverse permeability (Ky)

Cai and Berdichevsky16

Ky =

Bruschke and Advani18

Ky =

rf2 8Vf

2 ⎡ 1 1 − Vf ⎤ ⎢⎣lnV − 1 + V 2 ⎥⎦ f f

rf2 (1 − l 2 ) ⎧ arctan[ (1 + l ) (1 − l ) ] l 2 ⎫ × ⎨3l + + 1⎬ l3 3 2 1− l 2 ⎩ ⎭ 4 2 l = Vf π 2

[13.12] −1

[13.13] [13.14]


345

13.2.2 Fibre bundle dimensions (A2) With reference to Equation 13.1, it is necessary to calculate the effective thickness of the fibre tows. This can be estimated using the following relationship: Area = T0 w =

Nπ rf2 Vf

[13.15]

where T0 is the thickness of the fibre tow, w is the width of the tow, N is the number of fibres in the tow, rf is the fibre radius and Vf is the fibre volume fraction. In this study, the fibre volume fraction was measured by image analysis to be 72%. The average radii of the fibres were 8.5 μm and the number of filaments in the tow was 2000. On inspecting Equation 13.15 it can be seen that the thickness of a fibre bundle is related to its width. A schematic illustration of fibre spreading is shown in Fig. 13.6. Wilson19 proposed a model for predicting the degree of fibre spreading: w = (12Zψ )


13

[13.16]

where w is the width of the spread fibre tow, Z is the cross-sectional area of the tow and ψ is the lateral distance between two cylindrical spreading bars with a diameter of 12 mm. Wilson stated that the degree of fibre spreading was only dependent on the cross-sectional area of the tows and the lateral distance between two cylindrical spreading bars. Wilson also

(a)

Fibre spreading

(b)

T0

W

13.6 Schematic illustration of fibre spreading: (a) an as-received fibre bundle; and (b) a bundle after spreading. Here (w) and (T0) are the width and thickness of the fibre bundle respectively.

346


Table 13.3 Summary of selected patents on techniques and associated equipment for inducing fibre spreading of reinforcing fibre tows Fibre spreading techniques

Authors


Kawabe and Tomoda,20 Baucom et al.,21 Guirman et al.,22 Sihn et al.,23 Newell and Puzianowski,24 Ames et al.,25 Chung et al.,26 Daniels,27 Holliday28 Yamaguchi et al.,29 Kawabe and Tomoda,20 Guirman et al.,22 Iyer and Drzal,30 Yazawa et al.,31 Nakagawa and Ohsora,32 Van den Hoven,33 Kiss et al.,34 Baudry et al.,35 Belvin et al.,36 Lifke et al.,37 Marissen et al.38 Vyakarnam and Drzal,39 Iyer and Drzal,30 Akase et al.,40 Chen and Chao,41 Sager,42 Hall,43 Yamaguchi et al.,29 Yamamoto et al.44 Uchiyama et al.,45 Niina et al.,46 Peritt et al.,47 Sternberg48 Blackburn,49 Van den Hoven,33 Krueger50

Jet/vacuum air

Pins and rollers

Ultrasonics

Electrostatics Comb

stated that the tension and the mode of gripping of the tows did not influence the degree of spreading. Devices and techniques for spreading reinforcing fibres have been reported extensively in the patent literature. A summary of selected patents that deal with fibre spreading is given in Table 13.3.

13.2.3 Viscosity (A3) With reference to the development of the clean filament winding technology, a commercially available resin system, LY3505 epoxy resin and XB3403 amine hardener, were used in the trials. The viscosity at the point of impregnation was assumed to be constant. This is a reasonable assumption because there is a relatively low dead-volume within the impregnator, which in turn means that the resin system cannot stagnate. Moreover, the point at which the resin is injected into the fibre tow, a ‘fresh’ batch of mixed resin system is supplied continuously from the static-mixer.

13.2.4 Pressure (A4) Capillary pressure (A4.1) Ahn and Seferis51 developed a model to calculate the capillary pressure based on the Young–Laplace relationship: Pc =

4ζ cos θ DE

[13.17]


347

where Pc is the capillary pressure, ζ is the surface tension of the wetting fluid, θ is the contact angle between the fluid and solid, and DE is the equivalent diameter of pores in a fibre bundle. Ahn and Seferis51 employed the following relationship for evaluating DE for an array of unidirectional fibres: DE =

8rf ε F 1−ε

[13.18]

where rf is the fibre radius, ε is the porosity, and F is a form factor. F was said to equal two for transverse flow and four for axial flow. A simulation of this can be seen in Fig. 13.7 where the capillary pressure was calculated using Equations 13.17 and 13.18. The fibre radius was assumed to be 8.5 μm, the contact angle for the uncured epoxy resin was taken as 57° and the surface tension was taken as 0.044 N/m.14 In conclusion, it was found that the capillary pressure in the axial direction was higher than in the transverse direction.

Bates and Charrier52 proposed that the fibre pressure (P) can also be generated through the use of cylindrical pins during the impregnation process:

30 000 Transverse capillary pressure 25 000 Capillary pressure (Pa)


Applied pressure (A4.2)

Axial capillary pressure

20 000

15 000

10 000

5000

0 0

0.2

0.4

0.6

0.8

1

Fibre volume fraction

13.7 Capillary pressure plotted against fibre volume fraction in accordance with the model proposed by Ahn and Seferis.51

348

Management, recycling and reuse of waste composites P=

Te wC

[13.19]

where Te is the fibre tension, w is the width of the tow and C is the radius of curvature of the tow. Chandler et al.53 modelled the build-up of fibre tension during pin-based impregnation using lubrication theory. With reference to Fig. 13.8 Chandler et al.53 proposed four main zones to exist within the pin impregnator: (1) the entry zone, where the fibre tow approaches the pin at a pre-determined angle; (2) the impregnation zone, where the resin between the fibre and pin is forced into the fibre tow; (3) the contact zone, where sufficient resin has been applied to the fibres and where the tension is built up as a result of Coulomb friction and viscous drag; and (4) the exit zone, where the tow leaves the pin. Chandler et al.53 then developed a model to calculate the impregnation length during pin-assisted impregnation. Here, the impregnation length was calculated using the following equation:


L1 =

ηVAt ( H1 − H 0 ) ( 2 H1 + H 0 ) 3KH 0 H1 ηV ( 2 H1 + H 0 ) 6H 0 H1

−Te1 + Te21 −

[13.20]

where L1 is the impregnation length, Te1 is the tension per width in the impregnation zone, η is the resin viscosity, V is the velocity of fibre movement, A is the radius of the pin, K is the transverse permeability, and H0 and H1 are the resin/film thickness at the beginning and end of the impregnation region respectively.

Fibre motion

2 1

3 4

Radius

13.8 Regions of behaviour in pin impregnation: (1) entry; (2) impregnation; (3) contact; and (4) exit.53


349

The tension per width Te1 was calculated as: Te1 − Te 0 =

5πηV 4

2A H0

[13.21]

where Te0 is the initial fibre tension.

13.3

The application of selected impregnation models to clean filament winding


This section considers the application of selected impregnation models to the clean filament winding process. The impregnation of the fibre tows in the clean filament winding process was assumed to occur via a combination of pin-assisted and low injection pressure impregnation. • Transverse permeability: The transverse permeability was predicted by applying Gebart’s model13 (Equation 13.11), where the architecture of the fibre bundle was assumed to be hexagonal and the maximum packing capacity was taken as 0.9. The ‘volume fraction’ and width of the fibre tow were assumed to be 72% and 7 mm respectively. • Capillary pressure and pressure generated by the fibre traversing over a pin: With reference to the clean filament winding technique, the capillary pressure was calculated using Equation 13.17. The capillary pressure was calculated to be 7 kPa. The fibre pressure (P) was calculated to be 95 kPa using Equation 13.19. Here, the contact time between the pin and the fibre bundles was calculated using a curvature length of 15 mm and a winding rate of 10 m/minute. The contact time was calculated to be 0.09 seconds and the fibre tension was 10 N. The viscosity of the resin system used during the clean filament winding method was measured to be between 0.3 and 0.4 Pa s at 23 °C. Once the above-mentioned variables had been calculated, they were then applied to the models proposed by Foley and Gillespie11 and Gaymans and Wevers12 to calculate the impregnation time for the resin injection units. With reference to the model proposed by Foley and Gillespie,11 the shape of the fibre bundle was assumed to be circular. However, in the clean filament winding technique, the shape of the fibre bundle at the point of resin injection is a rectangular ribbon; the original equation proposed by Foley and Gillespie11 was modified accordingly. Furthermore, the impregnation process was assumed to take place over an arbitrary change in the infiltration thickness Ti = T1 = c1T0 to T2 = c2T0, where Ti is the infiltration thickness

350


and T1 and T2 are arbitrary steps of infiltration thickness. The infiltration time, ti, was reformulated as: ⎧⎪ c12 ⎡ln (1 c1 ) + 1⎤⎦ − c22 ⎡⎣ln (1 c2 ) + 1⎤⎦ ⎫⎪ ti = η (1 − Vf )T0 2 ⎨ ⎣ ⎬ 4K y( ΔP ) ⎪⎩ ⎪⎭ 2

2

[13.22]

where η is the viscosity, Vf is the fibre volume fraction, T0 is the initial thickness of fibre tow, c1 and c2 are constants, Ky is the transverse permeability and ΔP is the pressure differential. From Equation 13.22, the degree of impregnation, DI, was calculated as: [13.23]

Equation 13.22 was used to predict the impregnation time for the clean filament winding process; each fibre tow was assumed to consist of 2000 filaments with a Tex of 1200 and a width of 7 mm. The impregnation time (Equation 13.22) and the degree of impregnation (Equation 13.23) were calculated and are presented in Fig. 13.9. It can be seen that the impregnation process is significantly faster when pin-assisted impregnation is used when compared with immersion alone. It is apparent from Fig. 13.9 that full impregnation of the fibre tow can be achieved after 0.022 seconds for pin-assisted impregnation. 0.35

Immersion-based impregnation Pin-assisted impregnation

0.30 Impregnation time (s)


T DI % = ⎡⎢ i ⎤⎥ × 100 ⎣T0 ⎦

0.25 0.20 0.15 0.10 0.05 0 0

20

40

60

80

100

120

Degree of impregnation (%)

13.9 Simulation of the model proposed by Foley and Gillespie,11 where a rectangular fibre bundle profile was assumed.


351

In the current study, the second model that was considered was proposed by Gaymans and Wevers12 where the degree of impregnation, DI, was defined as: DI =

Ti = T0

2 KΔPti ηεT02

[13.24]

where K is the permeability, η is the viscosity of the liquid, Ti is the infiltration thickness, T0 is the thickness of the fibre tow, ε is the fibre tow porosity, and ΔP is the pressure differential. Equation 13.24 can be rearranged in terms of infiltration time, ti: 2

ηεT02 2 KΔP

[13.25]

A simulation of Equation 13.25 is presented in Fig. 13.10. On inspecting Fig. 13.10, it can be seen that the model proposed by Gaymans and Wevers12 also predicts that full impregnation will be achieved at a faster rate via pinassisted impregnation when compared to immersion-based impregnation. The model predicts complete impregnation after 0.036 seconds. In summary, a brief overview was presented on selected models that considered permeability, bundle dimensions, viscosity and pressure in accordance with Darcy’s equation. These models were then applied to estimate the time required to achieve full-impregnation for a 1200 Tex fibre bundle with a rectangular cross-section and a width of 7 mm. Pin-assisted 0.7 Immersion-based impregnation Pin-assisted impregnation

0.6 Impregnation time (s)


ti = ( DI )

0.5 0.4 0.3 0.2 0.1 0 0

20

40

60

80

100

120


13.10 Simulation of the model proposed by Gaymans and Wevers12 for pin-assisted and immersion-based impregnation.

352


impregnation was shown to be more efficient when compared with immersion-based impregnation. On the basis of the discussion above, a series of designs were considered for the resin impregnation unit; the development and evaluation of two designs are presented in the following section.

13.4

Clean filament winding: resin impregnation unit

Two prototype resin impregnation units were designed, developed and evaluated in this study. A schematic illustration of Prototype-I is presented in Fig. 13.11. This unit was designed such that a number of the key parameters identified in Section 13.2 could be investigated. For example, the option for impregnating the tows using top and/or bottom resin injection was introduced. The following section is based on the resin impregnation unit shown in Fig. 13.11.


•

Primary fibre spreading pin (A): The rig was designed to be modular with a facility to increase or decrease the lateral length of the primary

I

H

D

D

C

A

B

E

F

J

E

G

13.11 Schematic illustration of the resin impregnation unit, Prototype-I. The insert shows the details on the opposite side of the impregnator54 (see pages 352–353 for details).


•

•

•


•

•

•

•

353

spreading pin depending on the number of fibre tows required to be impregnated. Miniature resin bath (B): This was a rectangular excavation in the primary spreading pin where the mixed resin system was delivered from the resin supply channels (C) and (I) for impregnating the fibres from the bottom and top respectively. In effect, this miniature resin bath combines the features of pin-assisted and immersion-based impregnation. Resin supply channels (C and I) to the miniature resin bath: A two-way splitter was used to feed the mixed resin system from a solitary staticmixer to the top and bottom of the fibre tow. Resin flow controller (D): In addition to the variable throughput control on the resin dispensing unit, a tap was positioned on each of the supply channels to enable further control of the volume of mixed resin that was delivered to the top and bottom of the tow. Horizontal bars (E): A horizontal bar was located at both the fibre entry and exit points of the miniature resin bath. The entry bar served to induce further spreading of the fibre tows and the exit bar acted as a doctor-blade. These bars also enabled the trajectory and the contactlength of the fibre tows under the top injector pin (H) to be controlled. Fibre guiding pins (F and J): Two pairs of guide pins were positioned on the front and rear of the primary spreading pin (A) to control the fibre trajectory and to guide the tows in and out of the miniature resin bath. The pins also served to control the trajectory of the fibre entering and exiting the miniature resin bath. Secondary fibre spreading and tensioning pin (G): This pin was located on the fibre entry side of the impregnation unit and it could be adjusted as desired to control the trajectory and the contact-length between the fibre tows and the primary spreading pin. Top-resin injector pin (H): Prototype-I was designed to enable the mixed resin system to be injected from above and below the reinforcing fibres. The unit was designed to have a small offset between the top and bottom resin injection ports. However, the results reported in the next section only report the deployment of the top-injector. The position of the top-injector was made adjustable to enable the trajectory, contactlength and the depth of the immersion of the fibre tows to be controlled as desired.

On the basis of the initial results obtained using Prototype-I, a significantly simpler design (Prototype-II) was conceived. A schematic illustration of Prototype-II is presented in Fig. 13.12. However, it is emphasised that the latter was developed after conducting an exhaustive study on the performance of Prototype-I.

354


C D

F

E B A


13.12 Schematic illustration of the resin impregnation unit, Prototype-II.

With reference to Fig. 13.12, the key components of Prototype-II are described: •

• • •

•

•

Primary spreading pin (A): A single pin was used to induce further spreading of the fibre tows. The reader is reminded that profiled rollers were also used to induce fibre spreading ahead of the resin impregnation unit. Miniature resin bath (B): This was a rectangular excavation and it served as a miniature reservoir for the mixed resin system. Resin supply channel for the injector pin and resin bath (C): The staticmixer was directly attached to the resin supply channel. Resin injector housing (D): The resin injector housing was adjustable to enable the fibre tows to be plunged to the desired depth within the miniature resin bath. Resin injector pin (E): This was a pin with a slot at the bottom to enable the mixed resin to be injected into the fibre tows. The depth to which the resin injector housing was plunged into the mixed resin system dictated the contact-length and the overall trajectory of the fibre tows. Exit pin (F): This pin served as a doctor-blade but it also acted to control the contact-length of the fibre tow on the resin injector pin (E).

Figure 13.13 shows photographs of both prototypes. They were evaluated using the filament winding station that was illustrated in Fig. 13.3. Prototype-II was also evaluated at an industrial site where it was retrofitted to a commercial filament winding machine. The scale indicated is a UK 50 pence coin.


(a)

355

(b)

13.13 Photographs of: (a) Prototype-I and (b) Prototype-II. The scale indicated is a UK 50 pence coin.


13.5

Experimental

13.5.1 Materials Fibres The E-glass fibres used in this study were supplied by PPG Industries (UK). The fibres were 1200 Tex with an epoxy-compatible binder system. The fibres were supplied as conventional bobbins and the fibre tow was drawn from the outer circumference during filament winding. The fibres were used as supplied.

Resin and hardener In all cases, the resin system used was Araldite LY3505 resin and XB3403 hardener (Huntsman Advanced Materials, UK).

Filament winding machines Filament wound tubes were manufactured using three filament winding machines. In the first instance laboratory-based trials involving PrototypesI and II were carried out on a custom-modified lathe (Coil Winding Technology, UK). Filament wound tubes were also manufactured using conventional wet-filament winding which involved the use of a resin bath;

356


this was carried out using a 4-axis filament winding machine. In the subsequent discussions, the term ‘conventional’ is used to describe filament wound tubes that were manufactured using the resin bath. Site trials were also carried out using Prototype-II on a 2-axis filament winding machine at an industrial site in the UK. The difficulties associated with comparing filament wound tubes manufactured using three different machines are duly acknowledged; however, this was unavoidable due to circumstances beyond the control of the authors. The filament winding trials were carried out using a winding speed of 10 m/minute. Once the required number of impregnated fibre tows were laid on the mandrel, the assembly was transferred to an air-circulating oven and processed at 70 °C for 6 hours. The ovens used in the three trials were also different. A mandrel extraction unit was used to remove the filament wound tubes after processing in the oven.


Test methods The following tests were conducted to evaluate the majority of the filament wound tubes. •

•

•

Fibre volume fraction and void content: The fibre volume fraction and void content of the filament wound tubes were evaluated in accordance with ASTM standard D258455 and D273456 respectively. Test specimens (20 mm × 20 mm) were cut from the filament wound tubes using a diamond-coated cutting wheel. The mass of the test specimens were recorded using a five-digit analytical balance. The ‘burn-off’ tests were carried out in a muffle-furnace at 575 °C for 10 hours. Image analysis: Test specimens (20 mm × 20 mm) were cut using a diamond-coated wheel and potted using an epoxy resin system. The end-face of the potted samples were polished using conventional metallographic procedures. A Leitz DMRX microscope and an image analysis suite were used to obtain a minimum of 20 images from random locations per specimen. Hoop-tensile strength (split-disk): The procedures stipulated in ASTM D229057 were used to obtain the hoop-tensile strengths of the filament wound tubes manufactured using the conventional and clean filament winding techniques. Rings of 20 mm in width were cut from the filament wound tubes and notches of 3.2 mm radius were introduced. A photograph of the test fixture is shown in Fig. 13.14. These tests were carried out at room temperature on a Zwick-1484 mechanical test machine using a cross-head displacement rate of 2 mm/minute.


357

10 cm


13.14 Photograph of the hoop-tensile test fixture.

13.6

Results and discussion

13.6.1 Fibre spreading, fibre volume fraction and void content Appreciating the fact that the filament wound tubes were produced on three different filament winding machines but using the same fibre and resin system, a summary of the fibre volume fractions and void contents are presented in Table 13.4. The results from the trials using Prototype-I exhibited the highest degree of fibre spreading and the lowest void content. The degree of fibre spreading is defined as the percentage increase in the width of the fibre tow with respect to the as-received state. In other words, if an as-received fibre tow was spread from 4 mm to 8 mm, this would be defined as a 100% degree of fibre spreading. From Table 13.4, it can be seen that the conventional wet-filament winding technique produced a fibre volume fraction of 63%. On the other hand, the clean filament winding technique produced fibre volume fractions of 70 and 69% for Prototypes-I and II respectively. The samples produced on-site using the retrofitted Prototype-II also yielded a relatively low void content. The lower void contents produced by Prototype-I may be attributed to the integrated multiple fibre spreading features.

358


Table 13.4 Summary of fibre volume fractions, void contents and degree of fibre spreading obtained using three filament winding machines

Manufacturing method Conventional wetfilament winding Clean filament winding – Prototype-I Clean filament winding – Prototype-II Retrofitted clean filament winding (on-site) – Prototype-II

Fibre volume fraction (%)

Void content (%)

Degree of fibre spreading (%)

63

1.4

–

70

0.8

300

69

1.8

100

67

1.3

–

(a)


(b)

13.15 A photograph of: (a) an as-received fibre tow; and (b) after spreading using the fibre spreading station and Prototype-II.

Visual evidence for the degree of fibre spreading achieved using Prototype-II is shown in Fig. 13.15. A typical example of the degree of fibre spreading obtained using Prototype-II on-site is shown in Fig. 13.16. The significance of fibre spreading has been emphasised previously. This study has established that pins and rollers are effective in inducing fibre spreading. A number of techniques have been reported in the literature to induce fibre spreading (see Table 13.3). However, a balance must be maintained between the number of fixtures that are used to force the fibres to spread and the resultant increase in fibre tension. The data presented in Figure 13.17 clearly shows the effect of fibre tow thickness on the transverse impregnation rate.

13.6.2 Resin impregnation unit The resin impregnation units used in the clean filament winding technique resulted in a significant reduction in the volume of waste resin produced.


359

13.16 Photograph of Prototype-II in operation on-site, showing the extent of fibre spreading. 0.08 167 μm 133 μm

Infiltration time (s)


0.07

0.04

111 μm 95 μm 83 μm 74 μm

0.03

67 μm

0.06 0.05

0.02 0.01 0 0

20

40

60

80

100


13.17 A graph showing the effect of fibre tow thickness variations on the transverse impregnation rates.11

These reductions can be attributed to the lower volume of the resin impregnation unit (approximately 1.2 × 10−5 m3) in comparison with the 5-litre resin bath used in the conventional wet-filament winding technique. Moreover, the volume of solvent required to clean the resin impregnation unit is significantly lower than that needed for the resin bath. The effective freesurface areas of the mixed resin system for a 5-litre resin bath with a rotating drum (for resin pick-up) and the current resin impregnation unit are approximately 0.448 m2 and 0.005 m2 respectively. Therefore, the emissions to the atmosphere are also reduced significantly in the clean filament winding process. Furthermore, in the clean filament winding process the resin impregnation unit is located directly above the mandrel, eliminating the probability of resin drips from the impregnated fibre bundles.

360


13.6.3 Image analysis of polished filament wound tubes Typical micrographs of the polished cross-sections of the filament wound tubes manufactured using the clean filament winding technique in conjunction with Prototypes-I and II, conventional wet-filament winding, and the retrofitted Prototype-II on-site are presented in Figs 13.18–13.21 respectively. The quality of the images shown indicate that the clean filament winding process is capable of producing equivalent or better than those produced using conventional wet-filament winding.

13.6.4 Hoop-tensile strength (split-ring)


Figure 13.22 shows the normalised hoop-tensile strength results for the conventional and clean filament wound tubes (Prototype-II). The data have been normalised to a fibre volume fraction of 69%. The normalised average hoop-tensile strengths for the conventional and clean filament wound tubes (Prototype-II) were 715 MPa and 717 MPa respectively.

13.18 Micrographs showing the quality of clean filament wound tubes produced using Prototype-I.

13.19 Micrographs of tubes manufactured using the clean filament winding technique and Prototype-II.


361

13.21 Micrographs of tubes manufactured on-site using clean filament winding and Prototype-II. 800 Normalised hoop tensile strength (MPa)


13.20 Micrographs of tubes manufactured using conventional wetfilament winding.

700 600 500 400 300 200 100 0

Conventional

Prototype-II

13.22 Summary of the normalised hoop-tensile strength results for tubes manufactured using conventional wet-filament winding and clean filament winding (Prototype-II).

13.6.5 Site trial The clean filament winding technology described in the previous sections was demonstrated on-site at a company involved in filament winding (Portsmouth, UK). In order to retrofit Prototype-II to the commercial

362


filament winding machine a simple rectangular adaptor plate was connected to the traverse-arm of the machine. Prototype-II was mounted on the adaptor plate. The resin feed from the dispensing unit was connected to the resin impregnation unit via a pair of flexible plastic pipes. The diameter and the length of the mandrel used in the site-trial were 45 mm and 3 metres respectively. The winding speed and angle were identical to those used by the company for commercial production of filament wound components. The resin system used was LY3505/XB3403 and the reinforcing fibres used were continuous 1200 Tex E-glass fibres. A photograph of the mandrel being over-wound with impregnated glass fibres is presented in Fig. 13.23. Once the required number of ‘covers’ or layers of impregnated fibres were wound around the mandrel, the mandrel was transferred to an air-circulating oven and the resin was cross-linked using a predefined processing schedule. The key observations and conclusions reached during and after the site trial were as follows:


•

•

The retrofitting procedure was relatively straightforward and involved the construction of an adaptor plate that was bolted onto the arm of the traverse-carriage on the commercial wet-filament winding machine. The retrofitting of Prototype-II to the commercial wet-filament winding machine did not interfere or affect the normal usage of the machine or commercial manufacturing practices; the same winding speeds and angles were used when Prototype-II was introduced. However, it was observed that when the impregnated fibres reached the extreme ends of the mandrel, there was a tendency for the degree of fibre-spreading to be reduced. This was due to the fact that the last pulley on the retrofitted clean filament winding station could not swivel to compensate for the change in trajectory of the fibres.

13.23 Site trial: Photograph showing the impregnated fibres being wound around a 45 mm diameter mandrel. The resin impregnation unit (Prototype-II) was mounted on the traverse-carriage via an adaptor plate.

Clean and environmentally friendly wet-filament winding •

•

•


•

363

Since the resin impregnation unit was located in close proximity to the mandrel, it was not possible for the resin from the impregnated tows to drip onto the shopfloor. The feedback control between the resin dispensing unit and the filament winding machine was not used in the site trial. However, once the winding speed was set, it was relatively straightforward to adjust the throughput of the resin dispensing unit to maintain the required volume in the resin impregnation unit. Three major advantages of the clean filament winding process over conventional wet-filament winding are as follows: (a) the volume of mixed resin retained (waste) in the resin impregnation unit was vastly reduced in comparison to that retained in the 5-litre resin bath; (b) the volume of solvent required to clean the resin impregnation unit was a fraction of that required to clean the resin bath; and (c) the time required to clean the resin impregnation unit at the end of each production run was approximately 10 minutes compared with 55 minutes to clean the resin bath. With reference to Table 13.4, it can be seen that the quality of the tubes manufactured using the clean filament winding technique was comparable to that obtained using conventional wet-filament winding.

13.7

Conclusions

A step-change in the manufacturing process for wet-filament winding was designed, developed and demonstrated. The clean filament winding concept was conceived by analysing the perceived problems with conventional wetfilament winding. Solutions were then proposed and evaluated under laboratory conditions and on-site. In summary, the clean filament winding technique involved containing the resin and hardener components in separate reservoirs and then pumping them on-demand using precision gear pumps. Intimate mixing of the resin and hardener was achieved using a conventional static-mixer. Efficient impregnation of the reinforcing fibre tows was achieved using a custom-designed resin impregnation unit instead of a resin bath. The clean filament winding technique can facilitate a significant reduction in the generation of waste resin. Furthermore, the site trial verified that the volume of solvent required to clean the resin impregnation unit was a fraction of that required for conventional wet-filament winding. The emission of solvents and low molecular weight components to the atmosphere is also reduced significantly with clean filament winding. Finally, the time required to clean the resin impregnation unit is a fraction of that required to clean a conventional resin bath. The clean filament winding technique can be applied to other manufacturing processes such as pultrusion and pre-pregging. The technology can

364


also be readily adapted for processes involving multi-component coating of technical fabrics. The environmental benefits of the clean filament winding technique are significant.

13.8

Acknowledgements


The authors wish to thank the Engineering and Physical Sciences Research Council and the Technology Strategy Board for funding that enabled the design, development and demonstration of the clean filament winding concept. The authors wish to give due credit to previous researchers who helped to develop the clean filament winding concept, including Barry Ward, Tongyu Liu, Marcel Buckley, Mark Carpenter, John Rock, Rodney Badcock, Tony Martin, Mike Teagle, Sarah Smith, Claire Wait, Carly Butters and Brian Ralph. The assistance given by Jeff Dalling, Colin Leek, Aviva Howard, Graham Pledger, Mark Hudson, Frank Biddlestone, James Grazebrook, Peter England, John McHugh and Maggie Keats is duly acknowledged.

13.9

Notation

A Pin radius Gradient of velocity (v¯) against distance (L) ah B(VA) Curve-fitting constant for maximum fibre packing capacity C Radius of curvature of the roving c Dimensionless shape factor C(VA) Curve-fitting constant for maximum fibre packing capacity c1, c2 Constants DE Equivalent diameter of pores in a fibre bundle Dh Hydraulic diameter DI Degree of impregnation F Form factor H(1/2) Channel half-height H0 Resin/film thickness at the beginning of the impregnation region H1 Resin/film thickness at the end of the impregnation region k Kozeny constant K Permeability of the porous medium Kx Axial permeability Ky Transverse permeability Ky.quadratic Transverse permeability assuming a quadratic fibre architecture Ky.hexagonal Tansverse permeability assuming a hexagonal fibre architecture L1 Impregnation length m(VA) Curve-fitting constant for maximum fibre packing capacity


Clean and environmentally friendly wet-filament winding N P ΔP ΔP/L Pc Re rf Te Te0 Te1 ti Ti To T1, T2 U V VA Vf Vp v¯ w x y Z ε η λ ρ ψ ζ θ

365

Number of individual filaments in a fibre tow Pressure generated by a fibre traversing over a pin Pressure differential Pressure gradient over a characteristic dimension, L Capillary pressure Reynolds number Fibre radius Fibre tension Initial fibre tension Tension per width in the impregnation region Infiltration time Infiltration thickness Fibre tow thickness Arbitrary steps of infiltration thickness Mean resin velocity over the fibre cross-section Velocity of fibre movement Maximum packing capacity for the fibres Fibre volume fraction Velocity profile Superficial velocity Fibre tow width Horizontal coordinate Vertical coordinate Cross-sectional area of the fibre tow Porosity Viscosity of the resin Frictional factor Resin density Lateral distance between two cylindrical spreading bars Surface tension of the resin Contact angle between the resin and fibre.

13.10 References 1. green, j. e., ‘Overview of filament winding’, SAMPE Journal, Vol. 37, 7–11, (2001). 2. funck, r. and neitzel, m., ‘Improved thermoplastic tape winding using laser or direct-flame heating’, Journal of Composite Manufacturing, Vol. 6, 3–4, (1995). 3. toso, y. m. p., ermanni, p. and poulikakos, d., ‘Thermal phenomena in fiberreinforced thermoplastic tape winding process: computational simulations and experimental validations’, Journal of Composite Materials, Vol. 38, 107–135, (2004).


366


4. kim, j., moon, t. j. and howell, j. r., ‘Transient thermal modelling of in-situ curing during tape winding of composite cylinders’, Journal of Heat Transfer, Vol. 125, 137–146, (2003). 5. duvall, m., ramani, k., bays, m. and caillat, f., ‘In-situ composite manufacture using an electrostatic powder spray process and filament winding’, Journal of Polymers and Polymer Composites, Vol. 4, 325–334, (1996). 6. ramani, k., wooland, d. d. and duvall, m. s., ‘An electrostatic powder spray process for manufacturing thermoplastic composites’, Journal of Polymer Composites, Vol. 16, 459–469, (1995). 7. pistor, c. m., yardiminci, m. a. and gucery, s. i., ‘On line consolidation of thermoplastic composites using laser scanning’, Journal of Composites A, Vol. 30, 1149–1157, (1999). 8. agarwal, v., mccullough, r. l. and schultz, j. m., ‘The thermoplastic laser assisted consolidation process: mechanical and microstructure characterization’, Journal of Thermoplastic Composite Materials, Vol. 9, 365–380, (1996). 9. liu, t., mohamed, r. and fernando, g. f., ‘Optical sensor system for monitoring the extent of mixing in thermosetting resins’, Smart Structures and Materials, SPIE, San Diego, USA, 154–162, (1998). 10. liu, t. and fernando, g. f., ‘Processing of polymer composites: an optical fibrebased sensor system for on-line amine monitoring’, Composites Part A, Vol. 32 (11), 1561–1572, (2001). 11. foley, m. e. and gillespie, j. w., ‘Modelling the effect of fibre diameter and fibre bundle count on tow impregnation during liquid moulding processes’, Journal of Composite Materials, Vol. 39, No. 12, 1045–1065, (2005). 12. gaymans, r. j. and wevers, e., ‘Impregnation of a glass fibre roving with a polypropylene melt in a pin assisted process’, Journal of Composites–A, Vol. 29, 663–670, (1998). 13. gebart, b. r., ‘Permeability of unidirectional reinforcements for RTM’, Journal of Composite Materials, Vol. 26, No 8, 1100–1133, (1992). 14. amico, s. c. and lekakou, c., ‘Axial impregnation of a fiber bundle Part 1: Capillary experiments’, Journal of Polymer Composites, Vol. 12, Issue 2, 249– 263, (2002). 15. carman, p. c., ‘Fluid flow through granular beds’, Transactions of the Institution of Chemical Engineers, Vol. 15, 150–156, (1937). 16. cai, z. and berdichevsky, a. l., ‘An improved self-consistent method for estimating the permeability of a fibre assembly’, Journal of Polymer Composites, Vol. 14, Issue 4, 314–323, (1993). 17. berdichevsky, a. l. and cai, z., ‘Preform permeability prediction by self consistent method and finite element simulation’, Journal of Polymer Composites, Vol. 14, Issue 2, 132–143, (1993). 18. bruschke, m. v. and advani, s. g., ‘Flow of generalised Newtonian fluids across a periodic array of cylinders’, Journal of Rheology, Vol. 37, Issue 3, 479–498, (1993). 19. wilson, s. d. r., ‘Lateral spreading of fibre tows’, Journal of Engineering Mathematics, Vol. 32, Issue 1, 19–26, (1997). 20. kawabe, k. and tomoda, s., ‘Method of producing a spread multi-filament bundle and an apparatus used in the same’, US Patent: 0137156 A1, (2006). 21. baucom, r., snoha, j. j. and marchello, j. m., ‘Process for application of powder particles to filamentary materials’, US Patent: 5,057,338, (1991).



367

22. guirman, j. m., lecerf, b. and memphis, a., ‘Method and device for producing a textile web by spreading tows’, US Patent: 6,836,939 (2005). 23. sihn, s., kim, r. y., kawabe, k. and tsai, s. w., ‘Experimental studies of thin-ply laminated composites’, Journal of Composites Science and Technology, Vol. 67, 996–1008, (2007). 24. newell, j. a. and puzianowski, a. a., ‘Development of a pneumatic spreading system for Kevlar-based SiC-precursor carbon fibre tows’, Journal of High Performance Polymer, Vol. 11, 197–203, (1999). 25. ames, t., kenley, r. l., powers, e. j., west, w., wydand, w. t. and lomax, b. r., ‘Method and apparatus for making an absorbent composite’, US Patent: 7,181,817, (2007). 26. chung, t. s., furst, h., gurion, z., mcmahon, p. e., orwoll, r. d. and palangin, d., ‘Process for preparing tapes from thermoplastic polymers and carbon fibers’, US Patent: 4,588,538, (1986). 27. daniels, c. g., ‘Pneumatic spreading of filaments’, US Patents: 3,795,944, (1974). 28. holliday, r. c., ‘Process of preparing engineered fiber blend’, US Patent: 5,355,567, (1994). 29. yamaguchi, m., yamaguchi, s. and ohokubo, t., ‘Tow opening apparatus’, US Patent: 4,120,079, (1978). 30. iyer, s. and drzal, l. t., ‘Method and system for spreading a tow of fibres’, US Patent: 5,042,122, (1991). 31. yawaza, m., kurihara, k., tani, h. and matsumoto, m., ‘Method for producing laterally spread reticular web of split fibers’, US Patent: 3,953,909, (1976). 32. nakagawa, n. and ohsora, y., ‘Fiber separator for producing fiber reinforced metallic or resin body’, US Patent: 5,101,542, (1992). 33. van den hoven, g., ‘Widening–narrowing guide for textile filament bundle’, US Patent: 4,301,579, (1981). 34. kiss, p. a., deaton, j. m., parsons, m. s. and coffey, d., ‘Apparatus and method for splitting a tow of fibres’, US Patent: 6,385,828 (2002). 35. baudry, y., jean, r. and pirodon, j. p., ‘Systems for automatically controlling the spreading of a textile sheet’, US Patent: 6,687,564, (2004). 36. belvin, h. l., cano, r. j., johnston, n. j. and marchello, j. m., ‘Process of making boron-fiber reinforced composite tape’, US Patent: 6,500,370, (2002). 37. lifke, j. l., busselle, l. d., finley, d. j. and gordon, b. w., ‘Method and apparatus for spreading fiber bundles’, US Patent: 6,049,956, (2000). 38. marissen, r., van der drift, l. t. and sterk, j., ‘Technology for impregnation of fibre bundles with a molten thermoplastic polymer’, Journal of Composites Science and Technology, Vol. 60, 2029–2034, (2000). 39. vyakarnam, m. n. and drzal, l. t., ‘Apparatus and high speed method for coating elongated fibers’, US Patent: 5,310,582, (1994). 40. akase, d., matsumae, h., hanano, t. and sekido, t., ‘Method and apparatus for opening reinforcing fibre bundle and method of manufacturing prepreg’, US Patent: 6,094,791, (2000). 41. chen, j. c. and chao, c. g., ‘Numerical and experimental study of internal flow field for a carbon fiber tow pneumatic spreader’, Journal of Metallurgical and Materials Transactions B, Vol. 32B, 321–339, (2001). 42. sager, t. b., ‘Method and apparatus for separating monofilaments forming a strand’, US Patent: 4,959,895, (1990).


368


43. hall, j. n., ‘Apparatus for spreading a graphite fibre tow into a ribbon of graphite filaments’, US Patent: 3,704,485, (1972). 44. yamamoto, k., yamatsuta, k. and abe, t., ‘Spreading fiber bundle’, European Patent: 883004507.2, (1988). 45. uchiyama, s., kaku, e., kobayashi, m. and zoda, t., ‘Method and apparatus for spreading or dividing yarn, tow or the like’, US Patent: 3,657,871, (1972). 46. niina, g., sasaki, y. and takahashi, m., ‘Process for spreading of dividing textile materials’, US Patent: 3,358,436, (1967). 47. peritt, j. m., everett, r., and edelstein, a., ‘Electrostatic fiber spreader including a corona discharge device’, US Patent: 5,200,620, (1993). 48. stenberg, e. m., ‘Method and apparatus for charging a bundle of filaments’, US Patent: 3,967,118 (1976). 49. blackburn, r. m., ‘Fiber opening apparatus for an open-end spinning machine’, US Patent: 4,169,348 (1979). 50. krueger, r. g., ‘Apparatus and method for spreading fibrous tows into linear arrays of generally uniform density and products made thereby’, US Patent: 6,311,377, (2001). 51. ahn, k. j. and seferis, j. c., ‘Simultaneous measurements of permeability and capillary pressure of thermosetting matrices in woven fabric reinforcements’, Journal of Polymer Composites, Vol. 12, No. 3, 146–152, (1991). 52. bates, p. j. and charrier j. m., ‘Effect of process parameters on melt impregnation of glass roving’, Journal of Thermoplastic Composite Materials, Vol. 12, 276–294, (1999). 53. chandler, h. w., devlin, b. j. and gibson, a. g., ‘A model for the continuous impregnating of fibre tows in resin baths with pins’, Journal of Plastics, Rubber, and Composites Processing and Application, Vol. 18, Issue 4, 215–220, (1992). 54. allen, j. a., MRes Thesis, University of Birmingham, School of Metallurgy and Materials, UK. (2009). 55. astm d2584, ‘Standard Test Methods for Ignition Loss of Cured Reinforced Plastics’ (2002). 56. astm d2734, ‘Standard Test Methods for Void Content of Reinforced Plastics’ (2003). 57. astm d2290, ‘Apparent Tensile Strength of Ring or Tubular Plastics by Split Disk Method’ (2004).

14 Process monitoring and damage detection using optical fibre sensors D. H A R R I S, V. R. M AC H AVA R A M and G. F. F E R N A N D O, University of Birmingham, UK


Abstract: The aim of this chapter is to provide an overview of the use of optical fibre sensors for process monitoring and structural integrity assessment of composites. The term ‘process monitoring’ is used here to describe the various chemical reactions that take place during the cross-linking of thermosetting resins; the term ‘curing’ is sometimes used to describe the cross-linking process. Apart from monitoring the depletion and formation of specified functional groups during the cross-linking reactions, other parameters of interest include the temperature and residual fabrication strain at specified locations within the preforms. Whilst the focus of this chapter is on thermoset-based composites, the fibre optic sensor systems described here are equally applicable to thermoplastic reinforced composites. Key words: optical fibre sensors, process monitoring, cross-linking, damage detection, self-sensing composites, Fresnel reflection sensors, Fabry–Perot sensors, fibre Bragg gratings, FTIR spectroscopy, evanescent wave spectroscopy.

14.1

Introduction to optical fibres

‘Damage detection’ is defined here as the identification of degradation processes in fibre reinforced composites using optical fibres or the reinforcing fibres as sensors. This is not limited to fibre fracture, delamination, de-bonding, splitting and matrix cracking but also includes degradation caused by moisture ingress, thermal excursions, impact damage, etc. Optical fibres offer a number of advantages over electrical-based sensor systems including the following:1 • immunity from electromagnetic interference; • intrinsically safe and chemically inert (for example, silica-based optical fibres); • a selection of optical fibre types are available (for example, silica, polymethylmethacrylate, sapphire) in a range of diameters (15–1000 μm); • relatively low-cost if telecommunications fibres are used (1310 and 1550 nm) for fabricating the sensors; 369

370 • • • •


•


a large number of sensors can be fabricated on a single optical fibre or multiple fibres can be used; their circular cross-section makes them conducive for integration into fibre reinforced composites; data can be transmitted over long distances (kilometres) via optical fibres and therefore remote sensing and interrogation is possible; ability to be used in harsh environments provided that appropriate protection is provided; and multi-measurand sensors are available and the same sensor system can be used initially for chemical process monitoring and subsequently be used for structural integrity assessment of the component or structure in-service.2

In addition to chemical process monitoring and damage detection, optical fibre sensors can provide end-users with data to enable informed decisions on strategies for repair, renovation, recycling and disposal of composite materials and structures. With reference to Fig. 14.1, an optical fibre essentially consists of three main components, namely, a core, cladding and a protective coating or buffer. The refractive index of the core is slightly higher than that of the cladding. The refractive index is defined as the ratio of the velocity of light in a vacuum to that in the medium of interest. The refractive index for a medium is quoted at a specific wavelength and temperature. When a cleaved end of the optical fibre is illuminated, the light is contained within the core by total internal reflection at the core/cladding interface.3 The phenomenon of total internal reflection is illustrated in Fig. 14.2. As the light travels from a medium of refractive index n1 to a different medium of refractive index n2 (n1 > n2), it undergoes reflection and refraction as governed by Snell’s law. For a given media as shown in Fig. 14.2, the angle of incidence (θi) for which the light grazes the interface of the two media is known as the critical angle (θc). When θi > θc, the light undergoes total internal reflection.

Buffer

Cladding

Core

14.1 Schematic illustration of the basic components of an optical fibre.

Process monitoring and damage detection

371

n2 n1 qc qi


14.2 Schematic illustration of critical angle (θc) and total internal reflection (θi).

n0

n2

q1

f1 f2

n1

q2

14.3 Schematic illustration of the acceptance angle in an optical fibre.

The light guidance in optical fibres occurs by total internal reflection only when the semi-angle of the light cone (θ1) coupled into the optical fibre complies with the numerical aperture (NA) of the fibre, i.e. (θ1 < sin−1(NA/ n0). n0 (=1 for air) is the refractive index of the medium from where the light is launched into the fibre. It implies that for such light cones, the angle of incidence at the core/cladding interface (φ1) will be greater than the critical angle (θc = sin−1[n2(=ncl)/n1(=nco)], where nco and ncl represent the refractive indices of the core and the cladding respectively. This is illustrated in Fig. 14.3. The numerical aperture (NA) of the fibre is defined as: NA = nco 2 − ncl 2

[14.1]

372


Figure 14.3 shows that only a specified range of angles of the incident light cone can be accepted by the core. This is generally referred to as the acceptance angle. Another key term that needs to be appreciated is the ‘evanescent wave’. The light guided within the core by total internal reflection extends into the cladding with an exponentially decaying field intensity from the point of incidence at the interface. This is known as an evanescent field. However, as long as the cladding is transparent to the guided wavelengths, the total internally reflected light does not lose its power as it travels along the length of the fibre. The evanescent field amplitude, as a function of the distance (x) from the core/cladding interface, can be expressed as: E ( x ) = E0 exp ( − x dp )

[14.2]

where dp is the penetration depth of the evanescent field, and it is the distance into the cladding at which the field amplitude becomes 1/e of its value at the interface (E0). The penetration depth is expressed as:


dp =

λ

[14.3]

2 2π ( nco sin 2 θ − ncl2 )

12

where λ and θ are the wavelength of light and the angle of incidence at the core/cladding boundary respectively. When the cladding is not transparent to the wavelength of light, the light suffers attenuation via evanescent field absorption as it propagates along the fibre. The evanescent field is illustrated in Fig. 14.4. The subsequent sections of this chapter will use the above-mentioned discussion to outline the design of sensors for chemical process monitoring and structural integrity assessment of fibre reinforced composites. A relatively large range of optical fibres are available commercially. Table 14.1

Evanescent wave

dp

n2 n1

Incident light

Evanescent wave

14.4 Schematic illustration of the evanescent wave.

0.5–4.3 20 000–2325

1.51 250 4610 56 18.7 × 10−6

3–10 3300–1000

0.5 dB/m @ 6 μm

2.9

300

4400 21

14 × 10−6

Wavelength (μm) Range (cm−1)

Attenuation

Refractive index (core)

Maximum usage temperature (oC)

Density (kg m−3) Young’s modulus (GPa) Thermal expansion coefficient (K−1)

0.02 dB/m @ 2.6 μm

Fluoride

Chalcogenide

Property

8.8 × 10−6

3970 414 –

– –

800

>1500

0.54 × 10−6

2200 73

800

1.458

0.5 dB/m @ 1.5 μm

1.5 m), which is difficult to achieve with conventional light sources and reinforcing fibres such as E- and S-glass. One solution around this problem is to use optical fibres whose diameter is similar to that of the reinforcing fibres (see Table 14.3). These small-diameter optical fibres can be hybridised with the conventional reinforcing glass fibres and used for monitoring the cross-linking reactions. Since these small-diameter optical fibres do not suffer from high transmission losses, they can be used to overcome the limitation of conventional reinforcing glass fibres. Micrographs corresponding to polished sections of two types of small-diameter optical fibres are shown in Fig. 14.10. The thickness of the cladding on the fibres shown in Fig. 14.10(a) was between 1 and 2 μm.19 However, the cladding on the fibres shown in Fig. 14.10(b) are not concentric with the core. In the case of the fibres shown in Fig. 14.10(b), it was necessary to etch the cladding using hydrofluoric acid. This enabled access to the evanescent field in order for the fibres to be employed as evanescent-based chemical sensors.


385

20 um


(a)

10 um (b)

14.10 Micrographs showing a transverse section of polished custommade optical fibre bundles with an average diameter of: (a) 20 μm and (b) 15 μm.

14.4

Damage detection using self-sensing composites

Significant advances have been made in recent years on the deployment of optical fibre-based sensor systems for detecting the effects or aftermath of damage in fibre reinforced composites. However, there are limited tools available to enable a ‘cradle-to-grave’ strategy to be implemented for fibre reinforced composites. In other words, sensor systems reported to date tend

386



to concentrate on one aspect: process monitoring or damage detection. Major benefit would be found in establishing correlation between surface treatments, processing and subsequent performance of the composites. The self-sensing composite concept will permit this goal to be reached. Since it is possible to use conventional E- and S-glass reinforcing fibres as light-guides, they can be used as sensors to facilitate the detection of damage in fibre reinforced composites.20,21 This technique has been extended recently, where a high-speed camera was used in conjunction with piezoelectric transducers to study the fracture of individual fibres in a bundle in real-time.22 The sequence of images presented in Fig. 14.1123 show the fracture of small-diameter glass fibres where the composite was loaded in tension. In summary, hybrid sensing fibres represent the case where small-diameter optical fibres with comparable diameters to those of the reinforcing fibres are used to enable remote interrogation during processing. These fibres are also ideal for detecting impact damage when they are embedded on the surface of the preform during processing. When these fibres sustain impact damage, bleeding light can be used to infer the severity of the damage.24

(a)

(b)

(c)

(d)

(e)

(f)

14.11 Real-time images for a composite material. Images (a–e) correspond to 1%, 2.7%, 10%, 76% and 98% of the ultimate tensile strength. Image (f) was captured just prior the catastrophic failure of the composite.


14.5

387

Conventional optical fibre sensors

With reference to Fig. 14.6, composites with embedded optical fibre sensors were classified as instrumented composites (Type-IV). There is a wealth of information in the literature on the deployment of conventional optical fibres for chemical process monitoring and structural integrity assessment. The main advantage of using conventional optical fibres to fabricate sensors is that standard telecommunication components can be used. For example, this includes items such as connectors, couplers, light sources, detectors, cleavers and fusion splicers. Furthermore, these fibres have evolved over decades and their transmission losses are not an issue. The following section will focus on techniques that can be used to monitor cross-linking reactions. Both quantitative and qualitative techniques will be considered. Section 14.7 will address the use of optical fibres to monitor parameters such as strain and temperature.


14.6

Sensing strategies using conventional optical fibres

14.6.1 Infrared spectroscopy The infrared (IR) spectrum consists of three main regions: far-, mid- and near-infrared. The mid-IR region, 400–4000 cm−1 (25–2.5 μm wavelength), corresponds to the strong vibrational absorption frequencies (termed fundamental frequencies) of organic molecules. Traditionally, the analysis of organic compounds has tended to favour this region. The spectral region below 650 cm−1 is called the far-IR, and frequencies higher than 4000 cm−1 up to around 13 000 cm−1 are called near-infrared (NIR). NIR spectroscopy is a convenient and cost-effective technique because of the availability of relatively low-cost optical fibres and spectrometers with the associated ancillary equipment such as couplers and connectors. The absorption bands in the NIR region are broader and weaker than the IR fundamental vibration frequencies. These NIR bands are overtones and combinations of absorption bands of the IR fundamental vibrations. These overtones occur because of asymmetry and hence the NIR region is typically populated with asymmetric C—H, O—H and N—H bonds. These have relatively high frequencies in the mid-IR such that the first, second and even third overtone can be found in the NIR region. The weaker absorption means that longer path-lengths are required. Relevant formulae and units generally used in spectroscopy are summarised in Table 14.4. With reference to the generalised cross-linking reactions illustrated in Fig. 14.5, the C—H, N—H, C—O—C and O—H bonds have characteristic vibrational frequencies. Each bond can also have different modes of

388

Management, recycling and reuse of waste composites Table 14.4 Summary of useful parameters, formulae and units associated with spectroscopy Parameter

Relationship

Units

Wavelength

1 c = υ υ 1 υ υ= = λ c c υ = = cυ λ υ c = υλ = υ

m, μm, nm, cm

Wavenumber Frequency Velocity of light c = ∼3 × 108 m/s

λ=

cm−1 s−1, Hz ms−1


Stretching modes for AX2:

Symmetric

Anti-symmetric

Bending or deformation modes for an AX2 molecule: +

Scissor

Rock

Twist

+

+

Wag

14.12 Principal vibration modes of the AX2 molecule.

vibration, for example, stretching and bending as illustrated in Fig. 14.12.25 If the frequency of the incident radiation corresponds to the vibrational frequency of the bond in question, then the molecule can absorb this radiation. The absorbed energy can subsequently be lost to the surroundings in the form of heat or re-radiation. A plot of the absorbance or transmittance against the frequency yields an IR spectrum and is measured using a spectrometer. Since each bond has its own characteristic vibrational frequencies, it can be identified readily; see Tables 14.5 and 14.6. Overtone bands (harmonics) appear as approximate integer multiples of fundamental vibrations. For example, a strong absorption at 2500 cm−1 will give rise to a weaker absorption at 5000 cm−1. Although all fundamental vibration modes can show overtones in the NIR, the most prominent

nm nm nm nm nm

(5882 (5232 (4350 (4170 (4760

The series of bands at 1770, 1734, 1670 and 1690 nm can all be attributed to aliphatic C—H groups47

cm−1) cm−1) cm−1) cm−1) cm−1) 1700 1188 2300 2400 2100

—CH overtones of —CH2, —CH341,47 2nd harmonic of a C—H stretch of the methyl group40 —CH3 combination band34,48 —CH2 combination band of methylene group48 Terminal methylene band due to unsaturation34

C—H (aliphatic)

Internal calibration peak40,47,56 Used as an internal standard by Min et al.46

1668 nm (5995 cm−1) 2137 nm (4680 cm−1) 2164 nm (4622 cm−1)

1st overtone of C—H stretching vibration Combination band of the aromatic conjugated C=C stretch (1625 cm−1) with the aromatic C—H fundamental stretch (3050 cm−1)47,48

C—H (aromatic)

Hydrogen bonded peak40,46,47,56 (hydrogen bonding causes a shift to longer wavelengths)56

2090 nm (4785 cm−1) 2050 nm (4878 cm−1) 1432 nm (6983 cm−1)

Combination of O—H stretching and deformation vibrations47,57 1st overtone of the O—H stretching fundamental

Hydroxyl

Weak band40,56 Can be overlapped by adjacent bands40,56 Most suitable for quantitative analysis40,46,47,48,56,57

1159 nm (8627 cm−1) 1650 nm (6060 cm−1) 2206 nm (4532 cm−1)

2nd overtone of the C—H stretching vibration 1st overtone of C—H stretch in epoxy group Combination of the C—H stretching fundamental at 3050 cm−1 with the C—H2 deformation band at 1460 cm−1.34,54,56 (Chike et al.)48 assign this band to be a combination of the C-O fundamental stretch (∼900 cm−1) with the fundamental C—H stretch (∼3050 cm−1).

Epoxy

Comments

Wavelength

Assignment

Group

Table 14.5 Near-infrared absorption bands of diglycidyl ether of bisphenol-A (DGEBA)


Assignment

Combination of bending and stretching vibrations Overtone of symmetric and asymmetric stretches

Overtone of N—H stretch

N—H combination 1st overtone symmetric N—H stretch

1st overtone N—H symmetric stretch

Group

Aromatic primary amine

Aromatic secondary amine

Aliphatic primary amine

Aliphatic secondary amine

Comments DDS hardener DDS hardener Unknown band attributed to primary amine46 in DDS hardener DDS hardener Decylamine Decylamine Di-n-hexylamine

Wavelength 1974 nm,52–55 1970–1995 nm,46 1989 nm46 1496 and 1528 nm,52–55 1509 nm (max of two overlapping peaks),42 1494–1520 nm46 2205nm46 1496 nm,52–55 1494–1520 nm,46 1502 nm43 2025 nm37 1526 nm37 1543 nm37

Table 14.6 Near-infrared band assignments for aliphatic and aromatic amines




391

observed are those of O—H, C—H and N—H groups. The first overtones of bands such as C—O occur in the mid-IR and hence only weaker second or third overtones are seen in the NIR region. As well as overtones, combinational bands can also occur. These result from simultaneous excitation of two different fundamentals to give beats which are a combination of different frequencies. These combination bands usually involve O—H, C—H, or N—H stretches with one or more bending or rocking modes. A number of review papers have been published on NIR spectroscopy and its application to quantitative and qualitative analysis.26–33 Older published papers are also useful for peak assignments.34–41 In general, three basic types of epoxy resin systems have been used in the literature to monitor the cross-linking reactions using NIR spectroscopy: (i) 1,2-epoxy-3-phenoxypropane, a mono-functional epoxy more generally known as phenyl glycidyl ether (PGE) which is used for model reaction studies;42–45 (ii) diglycidyl ether of bisphenol-A (DGEBA);40,46–51 and (iii) tetraglycidyl 4,4′ diaminodiphenylmethane (TGDDM).52–55 The NIR spectra of these three types of resin are similar, with major absorption peaks being found in close proximity. Table 14.5 shows the position of NIR absorption bands for DGEBA resins. For epoxy resin systems using diamine hardeners, band positions are dependent on whether the amine is aromatic or aliphatic. Much of the literature on NIR analysis of epoxy resin systems concerns aromatic amine curing agents.38,39,42,46,47,49,52,54,55 Publications concerning aliphatic amines may be found in the literature.37,40,48 Table 14.6 lists the peaks present in the NIR region for aromatic and aliphatic amines and assigns their origin.

14.6.2 Fibre optic-based sensing of cross-linking reactions Conventional NIR spectroscopy was carried out using a 1 mm path-length cuvette in a temperature-regulated environment. It is necessary to ensure that the temperature gradients within the cuvette are relatively small (±0.1 °C). With reference to optical fibre-based monitoring of cross-linking reactions, four general monitoring techniques are possible: (i) transmission spectroscopy; (ii) reflection spectroscopy; (iii) evanescent wave spectroscopy; (iv) refractive index-based monitoring. In the following sections, each of these sensing techniques will be discussed. Optical fibre-based transmission spectroscopy The concept here is simple: an optical fibre is used to deliver the light from the spectrometer to the resin that is contained in a ‘reservoir’ or ‘cell’; the

392

Management, recycling and reuse of waste composites ‘Reservoir’ for resin flow

Light out

Light in

Epoxy groups

Absorbance

1.5

Aromatic C-H groups 1.0 Amine N-H 0.5

Hydroxylamine O-H N-H

End of reaction Start of reaction

0 1300

1500

1700

1900

2100

2300


Wavelength (nm)

14.13 Schematic illustration of an optical fibre-based sensor for conducting transmission near-infrared spectroscopy; the spectra show the NIR data obtained for an epoxy/amine resin system before and after cross-linking.

transmitted light, after passage through the resin, is collected by a second optical fibre and is delivered to the detector on the spectrometer. External light sources and detectors can also be used. A schematic illustration of such a sensor design is shown in Fig. 14.13 with typical spectra obtained during the processing (curing) of an epoxy/amine resin system. On inspecting the spectra shown in Fig. 14.13, the following conclusions can be reached: • The peak areas corresponding to the amine, epoxy and hydroxyl functional groups can be calculated to estimate their relative concentrations as a function of processing time. • As the cross-linking reaction proceeds, the relative concentration of the amine and epoxy decreases along with a proportional increase in the hydroxyl functional group. • The presence of a non-reacting, or inert, C—H absorbance peak means that the peak areas corresponding to the epoxy, amine and hydroxyl functional groups can be normalised. This effectively normalises the data for any optical path-length changes during the course of the crosslinking reaction. • On comparing the baselines for the cross-linked and un-cross-linked states, the baseline is seen to increase. The relative position of the


393

baselines can be monitored, away from an absorbance band, to indicate the progress of the cross-linking reaction. This increase in the baseline during cross-linking can be correlated to the increase in the refractive index of the resin system. • The percentage conversion of the functional groups of interest can be extracted from the spectra by calculating the respective peak areas using the following equation:

α = 1 − [( Ax ACH )t ( Ax ACH )0 ] × 100

[14.8]


where α is the conversion of the reacting group (amine or epoxy), Ax is the area of the absorption peak of the functional group of interest, ACH represents the area of the reference C—H peak, t specifies the crosslinking reaction time in minutes and 0 is the start of the reaction. An example of a conversion plot for an epoxy/amine resin system is presented later in Fig. 14.20. A number of semi-empirical equations are available to enable curve-fitting routines to be applied to the functional group conversion data.58 With reference to the transmission sensor design shown in Fig. 14.13, the conventional cuvette has been replaced by a resin containment cell. Here59,60 the critical elements of the sensor design are: (i) alignment of the cleaved optical fibres; and (ii) overall dimensions and robustness of the resin containment device. The resin containment device can be fabricated using partially ground capillary or tubing made from glass, metal, polymer or ceramic.61 The feasibility of mass-producing parallel-walled micro-cavities in optical fibres was demonstrated recently using laser ablation.62 This technique is useful because the diameter of the sensor or intrinsic cavity is equivalent to that of the optical fibre. A low-cost route for the production of intrinsic cavities at the end-faces of optical fibres was also demonstrated using hydrofluoric acid etching; two etched cavities were fusion spliced to fabricate an intrinsic fibre Fabry–Perot strain sensor.63 Subsequent to process monitoring, the same optical fibre transmission sensor design shown in Fig. 14.13 can be used for structural health monitoring.64 For example, the ingress of moisture will lead to an increase in the OH absorbance peak.

Optical fibre-based reflection spectroscopy A wide range of optical fibre reflection probes are available commercially. These effectively consist of a bifurcated bundle of optical fibres as illustrated in Fig. 14.14. The fibres are distributed in a random fashion at the

394

Management, recycling and reuse of waste composites Light to the detector

Light from the spectrometer

Optical fibre bundles

Lens cross-section:


Metal or ceramic jacket to protect fibres

Randomly arranged illumination and collection fibres = illumination = collection Lens

Sample

14.14 Illustration of a fibre optic probe to enable non-contact infrared spectra to be acquired. The insert shows the random arrangement of the optical fibres (light transmission and collection of reflected light) at the distal end of the probe.

distal-end (sensing region). One arm of the probe is connected to the light source of the spectrometer and the other is connected to the detector. The reflection probes are simple to manufacture using standard equipment available in the telecommunications industry. These probes can be used as non-contact devices to acquire spectra during the processing of composites. Such probes have been custom-made and used to monitor cross-linking reactions in an autoclave,65 microwave oven,66 differential scanning calorimeter,67 and rheometer.68


395

There are variations on the design of the reflection probe where a single optical fibre, positioned perpendicular to a reflective surface, can be used to obtain transmission/reflection spectra.69 This is particularly attractive for fibre reinforced composites where the dimensions of commercially available probes are too bulky for integration into composites. The details of the sensor design mentioned in Mahendran et al.69 are discussed further in Section 14.9. Probes based on a similar design to that illustrated in Fig. 14.14 have also been used for monitoring the stoichiometry and homogenisation of mixed multi-components (epoxy/amine) of resin systems.70


Optical fibre-based evanescent wave spectroscopy The basic principles associated with evanescent wave spectroscopy were presented in Sections 14.1 and 14.2 where the reinforcing fibres and custom-made small-diameter optical fibres were used as sensors. This section presents a brief overview of the use of conventional optical fibres with a partially de-clad section to enable the acquisition of evanescent wave spectra (Fig. 14.15). Techniques for monitoring cross-linking based on evanescent wave spectroscopy have been reported by a number of authors.71–73 An explanation of the evanescent wave phenomenon was presented in Section 14.1. Crosby et al.74 used a partially de-clad portion (300 mm in length) of a high index (1.70) fibre for monitoring the cross-linking of an epoxy– amine resin system and for refractive index monitoring of amine solutions of different concentrations. A non-absorbing band centred at 1310 nm was transmitted through the evanescent interaction length. Changes in the transmitted light intensity were observed as a result of the changing cladding (resin) index. This is because the critical angle at the core/ cladding interface and the numerical aperture of the fibre influences the transmission coefficient (see Fig. 14.15). The transmission coefficient can be expressed as: T=

tan 2 θ c 2 tan 2 θ c

[14.9]

where θc is the critical angle at the core/air interface and θc2 (=sin−1(nr/n1)) is the critical angle corresponding to the changing cladding index (due to cross-linking). When the absorbing wavelength band centred at 1550 nm was transmitted through this sensor, a peak area corresponding to amine absorption increased in a linear fashion with the amine concentration. The degree of cross-linking with time, as obtained from the evanescent fibre

396

Management, recycling and reuse of waste composites Light out

Light in

14.15 Schematic illustration of a partially de-clad optical fibre. The features within the de-clad region indicate that the numerical aperture is a function of the medium (resin) around the de-clad section. The propagation of light by total internal reflection is also illustrated.

Cleaved fibre-ends Cuvette Photodiode detector 2×2 coupler Light source

Resin system


14.16 Schematic illustration of the experimental set-up for the Fresnel reflection sensor.

sensor, was the same as that obtained using a 1 mm path-length cuvette in conjunction with FTIR spectroscopy. The shift in the baseline due to increasing resin index was taken into account when calculating the degree of conversion. Intensity-based techniques for monitoring cross-linking The evanescent wave sensor mentioned above can also be used as an intensity-based technique to monitor the cross-linking reactions. Since the refractive index of the material increases as it undergoes cross-linking, its density also increases. Thus, if the cladding of an optical fibre is replaced by the resin system, the transmitted light intensity will vary as a function of the refractive index of the resin system. The intensity of the transmitted light will decrease as the difference between the refractive indices of the fibre core and the resin system decreases. Several workers have used cleaved optical fibres to serve as Fresnelbased sensors to monitor cross-linking reactions in real-time.75,76 The fibre end is put in direct contact with the resin system and the reflection characteristics are recorded during cure (see Fig. 14.16). According to the Fresnel’s equation, the reflection coefficients at the fibre–resin interface for perpendicular, rn, and parallel, rp, polarisation, are given by:


397

rn =

nco cos θ1 − nr cos θ 2 nco cos θ1 + nr cos θ 2

[14.10]

rp =

nr cos θ1 − nco cos θ 2 nco cos θ 2 + nr cos θ1

[14.11]

where θ1 is the angle of incidence, θ2 is the angle of transmitted light, nco and nr are the refractive indices of the fibre and resin respectively. In the case of a single-mode fibre, the light can be assumed to travel along the normal, in which case θ1 ≅ θ2 ≅ 0°, and so ⎢rp⎢ = ⎢rn⎢ = r. The intensity of the reflectivity, R, can be expressed as:77 R = rp = rn =


2

2

nco − nr nco + nr

2

[14.12]

The following section describes a set of experiments that were carried out on a custom-modified Abbe refractometer to enable various parameters to be measured under identical processing conditions; the importance of maintaining a known and uniform temperature within the reaction vessel was mentioned previously. Since it was not desirable to cross-link the thermosetting resin directly on the prisms of the Abbe refractometer (High Accuracy 60/ED, Bellingham & Stanley, UK), a pair of glass slides were used to construct a ‘cell’ to contain the resin system (Fig. 14.17). The bottom slide (positioned on the lower prism of the refractometer) was custom-made (SF-14, Schott, Germany) to have the same refractive index as the refractometer prisms (n = 1.76). A contact liquid (monobromonaphthalene, Bellingham & Stanley, UK) was used between the custom-made slide and the lower prism of the refractometer to ensure good optical transmission between them. A 1 mm polytetrafluoroethylene spacer was used to separate the bottom slide from a standard borosilicate glass slide, and to contain the resin system. Release agent (Frekote® 700-NC, Loctite®, US) was applied to the surface of both slides for easy removal of the resin after processing. The following sensors were secured to the inner wall of the custom-made slide using a UV-curable resin (UV403-T, Shanghai Jiyuan Ltd., China): •

Single-fibre transmission sensor: A single-fibre transmission sensor was used to obtain near-FTIR spectra of the resin system as a function of cross-linking time at specified isothermal temperatures. Two cleaved multi-mode optical fibres were housed in a fused silica precisionbore capillary fixture and the cleaved end-faces were secured approximately 500 μm apart. The sensor was coupled to the light source and detector of an FTIR spectrometer. The spectrometer operated in the near-infrared region between 11 000 and 4000 cm−1. Spectra were collected at a resolution of 4 cm−1 and 64 scans.

398

Management, recycling and reuse of waste composites Multi-mode step-index fibre

(i) Single-fibre transmission sensor (iii) FBG Transmission sensor fibre (input)

Spacer

0.5 mm

Alignment Resin capillary Support capillary (i) Single-fibre transmission sensor Alignment capillary

Cladding Resin

Core

(iv) Thermocouple

Stripped region Perimeter of lower prism of Abbe refractometer Custom-made slide

Transmission sensor fibre (output)

(ii) Pair of Fresnel reflection sensors


Abbe refractometer set-up with the ‘cell’

(iii) Fibre Bragg grating sensor

Resin

2×2

Light source

coupler Photodiode (ii) Fresnel reflection sensor

14.17 Schematic illustration of the location of the various sensors on the Abbe refractometer: (i) Single-fibre transmission sensor; (ii) Fresnel reflection sensor; (iii) Fibre Bragg grating sensor; and (iv) thermocouple.

•

Fresnel reflection sensor: As described at the start of this section, cleaved optical fibres were used to serve as Fresnel-based sensors. A single-mode optical fibre was used to transmit light from a 1550 nm modular laser diode (ILX Lightwave, USA). The fibre was connected, via a custom-made 2 × 2 coupler, to an InGaAs photodiode detector (Thorlabs, UK). In order to increase the signal-to-noise ratio, the detector was connected to a photodiode amplifier (PDA 200, Laser 2000, UK) and a low-noise preamplifier (SR560, SRS, UK). The two distal cleaved ends of the sensor were secured adjacent to one another on the custom-made slide. During the cross-linking reaction, the reflected signal from the fibre core/resin interface was recorded using a LabVIEW data acquisition system (National Instruments, UK). • Fibre Bragg grating sensor: Periodic variations in refractive index were inscribed on the core of an optical fibre to create a fibre Bragg grating (FBG) sensor. An FBG sensor reflects particular wavelengths of light (depending on the wavelength used to inscribe the gratings) and transmits all others. Here, a photo-sensitive germania–boron (Ge–B) codoped single-mode fibre was used for fabricating the FBG sensor. The gratings were inscribed on the core of the fibre using the phase-mask technique (this is discussed in further detail in Section 14.7) and an excimer laser (Braggstar, Coherent Inc., UK) operating at 248 nm. The


Bragg gratings were produced using a pulse energy of 10 mJ and a pulse frequency of 20 hertz. FBG spectra were recorded using an IS 7000 FBG Interrogation System (FibrePro, UK). A schematic illustration of the sensor integration is shown in Fig. 14.17. Once all of the sensors were secured on the custom-made slide, the borosilicate glass slide was placed on top of the spacers. Silicone rubber adhesive (494-118, RS, UK) was used to seal the glass slides with the spacer, and this cell was left overnight at room temperature to enable the silicone adhesive to cross-link. The pre-mixed and de-gassed resin system was introduced into the cell via a syringe and needle. The Abbe refractometer was operated in reflection mode using a sodium lamp which emitted monochromatic light at 589 nm. The temperature of the prisms was controlled by circulating water from a temperature-controlled water bath (TE10D Tempette®, Techne, USA). Prior to conducting the cross-linking experiments, the temperature profile in the cell was measured using 15 thermocouples and the LY3505 resin. Fresnel reflection sensor and Abbe refractometer: With reference to Fig. 14.18, the initial drop in the amplitude of the Fresnel reflection signal can be attributed to a decrease in the refractive index of the resin system as it is heated from 30 to 70 °C; note that the Fresnel reflection data are overlapped with the data obtained from the Abbe refractometer and hence it is not readily apparent in Fig. 14.18. The subsequent gradual increase in the reflected light intensity relates to the isothermal 1.9

1.565 1.560

1.7

1.555 1.5 Fresnel reflection Refractive index (Abbe)

1.3

1.550 1.545 1.540

1.1

1.535 0.9

1.530

0.7

1.525 0

100

200

300

400

500

Time (minutes)

14.18 Comparison between the Fresnel reflection signal and the refractive index (Abbe refractometer) at a processing temperature of 70 °C.

Refractive index (Abbe)

Fresnel reflection (normalised)


•

399


stage, whereby an increase in the density and molecular weight of the resin due to cross-linking cause an increase in the refractive index. The subsequent stabilisation of the reflected light intensity suggests retardation of the rate of cross-linking. • Single-fibre transmission sensor: The evolution of FTIR spectra during the cross-linking of the resin system at 60 °C is presented in Fig. 14.19. The reactions involving the epoxy and amine chemical species were previously shown in Fig. 14.5. In Fig. 14.19, the absorbance peak at 4933 cm−1 was associated with the fundamental vibrations of the primary amine group, and the peak at 4530 cm−1 was attributed to the epoxy CH2 and C—H deformation bands. The depletion of these peak areas is caused by the epoxy/amine reaction as shown in Fig. 14.5. The peak at 4621 cm−1 is attributed to the combination bands of the C—H stretching vibration of the aromatic ring in the epoxy resin. As discussed previously this peak area was used to normalise those of the epoxy and amine absorbance bands. The conversion of the functional group of interest (epoxy or amine) can be calculated from Fig. 14.19 using the relationship stated in Equation 14.8; it can be observed that epoxy groups are consumed at a faster rate at higher isothermal temperatures (Fig. 14.20). Figure 14.21 demonstrates clearly that the trends in the data for the Fresnel reflection coefficients are similar to those obtained via the single-fibre transmission sensor. Therefore, accepting the limitation of intensity-based sensor systems, it can be concluded that the low-cost

Epoxy

0.3

Absorbance units


400

CH Primary amine

Reaction time

–0.2 5100

Wavenumber (cm–1)

14.19 Spectra of LY3505/XB3403 resin system obtained using the transmission sensor during cross-linking at 60 °C.

4400


401

100 90

70 °C 60 °C

Epoxy conversion (%)

80 70

50 °C

60 50 40 30 20 10 0 0

200

400 600 Time (minutes)

800

1000

14.20 The degree of conversion of the epoxy functional group when cross-linked at 50, 60 and 70 °C. 100

1.7

80 Fresnel reflection Epoxy conversion (FTIR)

1.5

60

1.3 40 1.1

Epoxy conversion (%)

Fresnel reflection (normalised)


1.9

20

0.9 0.7 0

200

400


1000

0 1200

14.21 Comparison of epoxy conversion (FTIR spectroscopy) data and Fresnel reflection data for the LY3505/XB3403 resin system at 60 °C.

Fresnel sensor can be used to track the progress of the cross-linking resin system. In instances where quantitative data are required on the cross-linking process, the single-fibre transmission sensor is a viable solution. However, it does dictate the need for a FTIR spectrometer. In situations where the temperature of the reaction vessel is known, this study has demonstrated that the Fresnel sensor (cleaved optical fibre) will suffice to monitor the rate of cross-linking.

Management, recycling and reuse of waste composites 1543.15

70

1543.10

60

Wavelength (nm)

1543.05

50

1543.00

40

1542.95 30

Wavelength (FBG) Temperature

1542.90

20

1542.85

Temperature (°C)

402

10

1542.80 1542.75

0 0

100


400

500


14.22 Response of the Ge-B co-doped fibre Bragg grating sensor that was used to monitor the cross-linking reaction in the Abbe refractometer. The experimental set-up was illustrated previously in Fig. 14.17.

•

Fibre Bragg grating sensor: Figure 14.22 illustrates the response of the FBG sensor when the epoxy/amine resin system was heated from 30 to 70 °C and held for 9 hours. Although further analysis needs to be completed, initial observations are outlined here. The response of the FBG sensor during the cross-linking reaction was attributed to two factors – strain induced by shrinkage of the resin and temperature variation. The temperature sensitivity of the grating during the ramp-up phase was found to be 9.2 pm °C−1. This agrees well with the temperature sensitivity of 8.9 pm °C−1 of Ge–B fibre determined using an independent method. Therefore it can be assumed that the response of the grating during the ramp-up phase was due to temperature alone. After the isothermal phase, the resin was allowed to cool down. During the cooling phase, the Bragg shift was influenced by both the temperature and cure-induced strain. The temperature sensitivity was found to be 18.7 pm °C−1. Therefore, the response of the Bragg grating during cooling was significantly influenced by resin shrinkage. The effect of strain and temperature on the fibre grating can be expressed as:

[ ΔλB ]ε ,T = [ ΔλB ]ε + [ ΔλB ]T = αΔε + βΔT

[14.13]

where α = λB(1 − ρe); ρe (∼0.74) is the strain-optic constant of the Ge–B co-doped fibre; and β = (αΛ + αn)ΔT; αΛ and αn are the coefficients of thermal expansion and temperature-optic respectively.


403

In Fig. 14.22 for the initial ramping-up temperature change of ∼32 °C, [ΔλB]T was 0.294 nm. During the cool-down phase, [ΔλB]ε,T was 0.599 nm over a temperature change of ∼32 °C. The residual strain can be expressed from Equation 14.13 as: Δε =

[ ΔλB ]ε ,T − [ ΔλB ] α

[14.14]

The residual compressive strain at a temperature of 33 °C after cooling was calculated to be 263 με.

Other sensor designs for monitoring refractive index The chemical and physical changes brought about by the cross-linking reactions were discussed in Section 14.2. The Lorentz–Lorenz equation (Equation 14.15) gives the relationship between the refractive index and polarisability of the system.78 This can be expressed as:


⎛ n2 − 1 ⎞ ⎛ M ⎞ Nσ ⎟⎜ ⎟ = ⎜⎝ 2 n + 2 ⎠ ⎝ ρ ⎠ 3ε 0

[14.15]

where n, M and ρ are the refractive index, molar mass and density of the resin respectively; σ, ε0 and N are the polarisability, free-space permittivity and Avogadro’s number respectively. A variety of fibre optic sensor designs have been reported for monitoring refractive index changes during the cross-linking reaction. •

Y-coupler: Crosby et al.74 used a multi-mode fibre Y-coupler to monitor cross-linking of an epoxy/amine resin system. A 1310 nm light source and a photo-detector were used to input and detect the reflected light from the cleaved end of the fibre. The Fresnel reflection coefficient was specified in Equation 14.12. The sensor was calibrated by measuring the reflected Fresnel signals in air and in liquids of known refractive indices. The cleaved end of the fibre was immersed in a cross-linking epoxy/ amine resin system. It was found that the Fresnel reflection values measured during calibration correlated well with the scaled prediction as obtained using the Fresnel equation. When the sensor was immersed in the neat-epoxy resin (without the hardener), the Fresnel signal decreased in a linear fashion with increasing temperature. A de-clad high index (1.70) core fibre was also used to monitor the refractive index changes during the cross-linking reaction via evanescent interaction as discussed previously. Cusano et al.77 also used a Y-coupler (single-mode fibre) for monitoring the refractive index. Here the influence of the coupling ratio, losses

404


along the optical path, losses induced by imperfections at the cleaved surface, temperature dependence of refractive index of the fibre and photo-detector noise were taken in account when computing the absolute refractive index. The trends observed in the Fresnel reflection signal during the cross-linking process were similar to that previously reported by Crosby et al.74 • 2 × 2 coupler: Vacher et al.79 reported on a similar Fresnel design for monitoring cross-linking in a thermosetting resin. Here, one of the cleaved distal ends of the fibres of a 2 × 2 coupler was immersed in resin and the other was coupled to a photodiode; the latter acted as a feedback control mechanism for the output of a modulated laser source. One input-end of the 2 × 2 coupler was coupled to a light source and the other to a photodiode detector. The Fresnel equation was used in conjunction with the Lorentz–Lorenz equation to monitor the light intensity, I, and to relate it to the density of the resin during crosslinking. This was expressed as:


I (α , T ) = A′ ρ (α , T ) + B′

[14.16]

where α and T are the degree of cross-linking and temperature respectively. The coefficients A′ and B′ are expressed as: A′ = AI 0

RM 2 nco( nr 0 − nco ) ( nr 0 2 + 2 ) 3 M nr 0( nr 0 + nco )

2

[14.17]

⎡ −2 nco( nr 0 − nco ) ( nr 0 2 − 1) ⎛ nr 0 − 1 ⎞ ⎤ +B B′ = AI 0 ⎢ +⎜ 3 ⎝ nr 0 + nco ⎟⎠ ⎥⎦ nr 0( nr 0 + nco ) ⎣ 2

[14.18]

where nco and nr0 are the refractive indices of the fibre core and the uncured resin respectively, I0 is the intensity of the input light; RM is the molar refractivity, and M is the molecular weight. A and B are the coefficients which depend on conditions of light injection into the coupler, connections, quality of fibre end cleavage, the amplification of the system and the signal treatment. The temporal evolution of the degree of cross-linking as measured by differential scanning calorimetry agreed with the evolution of the reflected Fresnel signal. Kim and Su80 also used a 2 × 2 coupler to measure the refractive indices of liquids at 1310 and 1551 nm. One of the cleaved ends was suspended in air as a reference, whilst the other was immersed in the resin. The drift in the laser and the changes in the detector efficiency were taken into account. This was accomplished by measuring the amplitudes of the light reflected from the fibre core/resin interface and fibre core/air interface in real-time. The refractive index of the liquid, nliq, was expressed as:


405

⎡ (1 − η ) ⎤ nliq = nco ⎢ ⎣ (1 + η ) ⎥⎦

[14.19]

⎡ (n − n ) ⎤ 1 η = ⎢ co air ⎥ ⎣ ( nco + nair ) ⎦ R

[14.20]

R=

[( nco − nair ) ( nco + nair )]2

[( nco − nliq ) ( nco + nliq )]2

[14.21]

where R ≡ Va/Vliq; Va is the photodiode voltage when both the cleaved ends of the sensing fibre are in air and Vliq is the photodiode voltage when the sensing fibre is immersed in a liquid.


•

Fibre Bragg grating: Extensive research has been carried out on FBGs over the last three decades. A number of authors have used FBGs (as described on pages 398 and 402) to measure the refractive indices of liquids and cross-linking resin systems; brief details of typical applications are outlined below. Background information on FBGs for monitoring strain and temperature is presented in Section 14.7. Iadicicco et al.81 reported on the refractive index sensitivity of FBG sensors with a diminished cladding. The Bragg resonance is a function of the effective index of the core and grating period. However, when the diameter of the cladding is reduced and surrounded by a resin, whose refractive index is less than that of the core, the Bragg wavelength shifts towards a longer wavelength as the refractive index of the resin increases. For a fibre of known core and cladding refractive indices, the effective refractive index is dictated by the ratio of the diameter of the core to cladding, and the refractive index of the surrounding resin. The non-linear sensitivity of the Bragg wavelength towards the refractive index of the surrounding resin was found to increase gradually over a range of indices. As the refractive index of the resin approached that of the core, a rapid increase was observed. It was also found that these sensitivity variations were much larger in magnitude for sensors with a smaller cladding diameter. Refractive index resolutions of the order of 10−5 and 10−4 were anticipated at 1.45 and 1.333 μm respectively. Schroeder et al.82 used a similar principle where a side-polished (close to the core) grating was multiplexed with another uniform grating. This design enabled the temperature and refractive index measurements to be made simultaneously. As a consequence of side-polishing, an asymmetric cladding is produced. This induces a polarisation dependence of the Bragg resonance. The use of high refractive index overlay coatings was found to control the sensitivity towards the refractive index of the surrounding medium. Sensors made using the same side-polishing

406



conditions and with the Bragg resonances inscribed at longer wavelengths were found to be more sensitive than those with resonances at shorter wavelengths. Liang et al.83 reported on a similar design for monitoring the refractive index. In addition to using a Bragg grating with an etched cladding, a short section of fibre separating two Bragg gratings was selectively etched to the core to produce a Fabry–Perot sensor configuration. The resonance structure obtained with this configuration was sharper compared with that of the etched Bragg grating alone. The shift of resonance was found to be more sensitive to changes in the refractive index of the resin surrounding the etched section. For more comprehensive information on multiple grating wavelength FBGs and tapered/slanted-tip FBGs for monitoring refractive index, see Shao et al.,84 Guo and Albin85 and Laffont and Ferdinand.86 In conclusion, this section has demonstrated that optical fibre sensors can be used to obtain real-time information on the cross-linking kinetics of thermosetting resins. Where quantitative information is required on the rates of reaction and the extent of conversion of specified functional groups, FTIR spectroscopy is an ideal technique. In instances where qualitative data is sufficient, the intensity-based sensors described here can provide reliable and repeatable results. In the context of real-time process monitoring and on-line process optimisation the discussion has to be extended to consider two issues. Firstly, in the case of real-time process monitoring, there is a need to ensure intimate contact between the sensor and the matrix prior to the commencement of the cross-linking reactions. There is little point in a system if there is a time-lag before it responds to the changing chemistry in the resin system. For example, in devices that require the matrix to ‘wet-out’ the sensor may miss a part of the initial cross-linking reaction. Evanescent wave sensors are ideal here because the fibres, or sensors, can be used to monitor the impregnation process during the fibre impregnation or lay-up stages. Therefore, the natural ageing of the matrix can be monitored prior to processing. Hence, appropriate process modifications can be made to compensate for the chemical integrity of the matrix. The second issue is that even if the sensor has a fast response time, the question is can anything effective be done, for example in an autoclave environment, when the auto-catalytic addition ring-opening reactions are in progress? Obviously, process monitoring is a valuable tool in terms of quality control and quality assurance. An area where chemical sensors have a key role to play is in defining when an acceptable level of cross-linking has been achieved. This will enable operators to switch off the processing equipment at the appropriate time as opposed to relying on predefined processing schedules, leading to significant cost-savings in energy consumption.


14.7

407

Sensors for monitoring strain and temperature 87

FBGs and Fabry–Perot (FP)2 sensors are examples of devices that are used frequently to monitor strain and temperature. Comprehensive theory and applications on FBGs can be found in Kashyap88 and Othonos and Kalli.89 Bragg gratings are inscribed on the core of a fibre using a variety of techniques such as phase-mask inscription amplitude-splitting interferometric method and point-by-point inscription. The following section will present a brief overview of FBG and FP sensors, followed by their application in fibre reinforced composites. Optical FBGs have a periodic modulation of the refractive index in the light guiding core of the optical fibre. The phase-matching between the forward-propagating fundamental mode and radiation undergoing backscattering at the grating planes produces a narrow-band Bragg reflection centred at λB, which can be expressed as:


λ B = 2 Λ neff

[14.22]

where Λ and neff are the fibre grating spacing and effective refractive index of the fibre core respectively. A schematic illustration of a fibre Bragg grating is shown in Fig. 14.23. Any modulation of the grating periodicity is manifested as a change in the Bragg resonance wavelength. As strain and temperature are relevant parameters during the processing of preforms and subsequent deployment of the composites in-service, a brief description of their influence on the fibre Bragg gratings is presented below. The shift in Bragg wavelength (ΔλB) due to longitudinal strain, Δε, can be expressed as: ΔλB = λB(1 − pe )Δε ; pe =

neff [ ρ12 − υ ( ρ11 + ρ12 )] 2

[14.23]

Broadband source spectrum Transmission spectrum Λ

Narrow-band Bragg reflection

Germania-doped photo-sensitive fibre

14.23 Schematic illustration of the index modulation and the nature of the transmitted and reflected spectra when the grating is illuminated with a broadband light source.

408


where pe is the strain-optic constant, ρ11, ρ12 and ν are components of the fibre-optic strain tensor and Poisson’s ratio respectively. The temperature change (ΔT) induces a Bragg wavelength shift, which can be expressed as:


ΔλB = λB(α Λ + α n )ΔT

[14.24]

where αΛ and αn are the thermal expansion and thermo-optic coefficients of the fibre respectively. For a standard germania-doped silica fibre the strain and temperature sensitivities of the grating at 1550 nm are 1.2 pm με−1 and 13.7 pm °C−1 respectively. Whilst the response of gratings to a single measurand is relatively straightforward to interpret, the simultaneous measurement of multi-measurands such as strain and temperature requires either: (i) the determination of cross-sensitivity coefficients, or (ii) the sensor design to isolate the influence of two or more measurands. An extrinsic fibre Fabry–Perot interferometric sensor (EFPI) constitutes two cleaved optical fibres that are aligned inside a precision-bore capillary, with an air-gap between the cleaved end-faces. These cleaved fibres are secured inside the capillary by a fusion joint or adhesive bonding. When the cavity is illuminated using a broadband low-coherence light source, the interference of Fresnel reflections, which differ in path length by twice the air-gap length, produce sinusoidal fringes across the source spectrum.90 The average intensity and contrast of the interference fringes are a function of the individual intensities of the interfering beams and the degree of coherence between them.91 A schematic of an EFPI is shown in Fig. 14.24. This low-finesse Fabry-Perot interference output can be expressed as: I [(φ ( λ ))] = I1 R + 2 R (1 − R ) I1 I 2 cos φ + R (1 − R ) I 2 2

[14.25]

where I1, I2 are the intensities of Fresnel reflections, φ is the phase difference between the Fresnel reflections, and R is the reflectivity at the air cavity. The measurand-induced change in the cavity length can be measured from the Fabry–Perot fringe distribution as the difference between the initial and final absolute cavity lengths. Fibre

Fusion joint

Capillary

Broadband light d Gauge length

14.24 Schematic illustration of the EFPI sensor.


Applications of fibre Bragg grating and extrinsic fibre Fabry–Perot interferometric sensors in composites

120 i Embedded FBG temperature ii Autoclave temperature 100 iii Embedded EFPI strain iv Theoretical prediction 80 v Embedded FBG strain vi Reference EFPI 60 vii Reference FBG strain

900 700 500

40 300 iv

100

i ii iii vi vii

v

–100 0

1

2

3

4 5 6 Time (hours)

7

8

20

Temperature rise (°C)

This section presents an overview on the deployment of optical fibre sensors for chemical process monitoring and structural integrity assessment of composites.92–94 Fig. 14.25 shows the output from embedded fibre Bragg gratings and extrinsic fibre Fabry–Perot interferometric sensors. The relative responses of the two sensor systems when the preform was processed in an autoclave showed a similar trend. The oscillations observed were attributed to the mode of operation of the pressurising system in the autoclave, whereby the compressor automatically initiated when the pressure dropped below a predefined value. Kuang et al.95,96 embedded polyamide-coated FBGs in fibre-metal laminates. The preforms considered were [aluminium alloy sheet (Aas)/0/ {FBG-0}/0/Aas/0/0/Aas]-unidirectional and [Aas/0/{FBG-0}/90/Aas/90/0/ Aas]-cross-ply. After processing, the residual compressive strain in the unidirectional and cross-ply composites resulted in a peak shift of 2.6 and 3 nm respectively. In the unidirectional case, the shape of the Bragg resonance remained almost unchanged after processing. However, significant peak-broadening and peak-splitting was observed in the cross-ply composites. This was attributed to radial compressive stresses and micro-bending caused by the adjacent 90° fibres. When the magnitude of the radial strain distribution reached a specified threshold, the spectrum comprised two or

Strain (με)


14.8

409

0 –20 9

14.25 Data from embedded and surface-located fibre Bragg grating and extrinsic fibre Fabry–Perot interferometric strain and temperature sensors. The unidirectional prepregs (16-layers) were processed in a custom-modified autoclave.114

410



more peaks. This was not apparent in the unidirectional case because of shielding from the non-uniform strain field. The difference between the thermal expansions of Aas, glass fibres and polypropylene could also affect the local stress fields around the FBG. Embedded re-coated FBGs were also shown to be less susceptible to spectral deformations than uncoated FBGs.97 Guemes and Menendez98 developed a theoretical model to estimate the longitudinal and transverse stresses from the response of distorted or splitpeaks in the Bragg spectrum. The transverse stresses produced two different effective refractive indices along the orthogonal directions in the single-mode fibre; this complied with two Bragg resonance conditions. Peak-splitting was observed during the cooling-down phase after crosslinking, owing to thermal contraction of the laminate and the associated shrinkage. During this phase, the split-peaks showed a linear response to temperature. The separation between the split-peaks was a measure of the transverse strain whilst the lower Bragg wavelength was representative of the longitudinal strain. If Δλ2 and Δλ3 are the shifts in the split peaks centred at λ2 and λ3 respectively, then the transverse stresses along both orientations can be expressed as: ⎛ Δλ 2 ⎞ ⎜ λ 2 ⎟ T1 T2 σ 1 ⎜ Δλ3 ⎟ = T1 T3 σ 2 ⎟ ⎜ ⎝ λ3 ⎠

(

)( )

[14.26]

T1 =

1 ⎧ neff 2 ⎫ [(1 − υ ) ρ12 − υρ11 ]⎬ ⎨1 − E⎩ 2 ⎭

[14.27]

T2 =

neff 2 1⎧ ⎫ [ ρ11 − 2υρ12 ]⎬ ⎨−υ − E⎩ 2 ⎭

[14.28]

T3 =

neff 2 1⎧ ⎫ [(1 − υ ) ρ12 − υρ11 ]⎬ ⎨−υ − E⎩ 2 ⎭

[14.29]

Here:

where E and υ are Young’s modulus and Poisson’s ratio respectively, and ρ11 and ρ12 are the components of the strain-optic tensor. The cross-linkinduced residual transverse stresses were found to be 85 and 110 MPa. Sorensen et al.99 measured the transverse strain during carbon fibrereinforced polyphenylene sulphide laminate using embedded FBGs. The transverse strain difference in the core was expressed as:



2 ⎡ ⎤ ⎛ λx − λy ⎞ εx − εy = ⎢ 2 ( ) ⎣ neff ρ12 − ρ11 ⎥⎦ ⎝ λB ⎠

411 [14.30]

where λx and λy are the peak wavelengths of the split-peaks corresponding to the induced X and Y polarisation axes and λB is the Bragg resonance wavelength corresponding to a zero transverse stress. It is readily apparent from Equation 14.30 that the net transverse strain is proportional to the separation between the peak wavelengths of the split-peaks. The polarisation sensitivity or induced birefringence in single-mode fibres due to anisotropic stress was confirmed experimentally from the spectral response of FBGs that were surface-mounted using an adhesive,100 and during diametric loading.101 Bragg gratings have also been used for characterising the evolution of resin shrinkage during and after processing. Giordano et al.102 reported that the response of FBGs used for cure monitoring of epoxy resin showed a point of infliction. This infliction was attributed to gelation and differential thermal expansion between the resin and fibre. Colpo et al.103 reported on a Bragg grating aligned along the axis of a cylindrical-shaped resin mould. The FBG experienced a post-cure longitudinal compressive strain gradient along the length of the grating. The centre of the grating was said to be subjected to maximum of 6000 με. The postcure residual strain gradients manifested as multiple Bragg resonances. It was assumed that there was no contribution from the transverse strain to the Bragg reflection spectrum due to the symmetry of the cylindrical mould. Antonucci et al.104 showed that the residual strain evolution in a thermosetting resin at the proximity of the resin/mould interface was influenced by: (i) the thermal and mechanical properties of the mould material; and (ii) the boundary conditions defining the interface. It can be generalised that the response of FBGs embedded in composite prepregs for cure and residual strain monitoring is influenced by more complex strain fields in comparison with that of a resin system alone. Liu et al.105 reported on the use of EFPI and FBG sensor systems for simultaneous detection of strain and temperature in composites. A splinefitting routine was used to characterise the features of narrow-band Bragg reflection superimposed on EFPI spectrum in order to resolve the wavelength shifts. The responses of the simultaneous sensor system to changes in strain and temperature were expressed as: ⎡ Δλ ⎤ k ⎢ λB ⎥ = ⎡ 11 ⎢ Δd ⎥ ⎢⎣k21 ⎣ ⎦

k12 ⎤ ⎡ ε ⎤ k22 ⎥⎦ ⎢⎣ ΔT ⎥⎦

[14.31]

where k11(=1 − pe) and k12(=αΛ + αn) are effective coefficients of strain and temperature for the FBG sensor respectively. k21 and k22 are effective


412


coefficients of strain and temperature for the EFPI sensor respectively. The response of the sensor system to strain and temperature agreed with those of the individual sensors. Singh and Sirkis106 developed an interrogation scheme for decoupling the combined response of an intrinsic fibre Fabry–Perot interferometer (IFPI) and an FBG. Here the IFPI was fabricated by fusion splicing a short section of a precision-bore capillary to two optical fibres. Two non-overlapping broadband light sources were used to illuminate the sensor system. The combined signal from the sensor system was decoupled using a wavelength division de-multiplexer. The interference signal from the IFPI was obtained using a matched interferometer. In low-coherence interferometry, interference fringes will be produced when the optical path differences of individual interferometers (sensing and receiving) differ by a value that is less than the coherence length of the light source. The strong Bragg reflection was tracked using an optical spectrum analyser. Thus the cross-talk was suppressed to achieve a good match between the optical and electrical measurements of strain and temperature. Fernando et al.107 reported on an integrated EFPI/FBG sensor design for the simultaneous measurement of strain and temperature. Here, the lead-in fibre length housed inside the capillary had a Bragg grating near the cleaved end of the fibre and hence it was in a strain-free state. The Bragg grating responded only to temperature changes whilst the change in the air-gap between the cleaved fibres imparted information on strain. The longitudinal strain, ε, measured by the EFPI sensor can be expressed as:

ε=

Δd D

[14.32]

where Δd is the change in the air-cavity length and D is the gauge length. Strain and temperature were measured using a relatively simple interrogation scheme comprising a super luminescent diode to illuminate the sensor and a charge-coupled device spectrometer. A linear relationship was observed between the optically and electrically measured strain and temperature. Badcock and Fernando108 used an intensity-based multi-mode EFPI design for fatigue damage detection in composites. The longitudinal strain was expressed as: 4r 2 ⎡ ⎤ Attenuation (dB) = 10 log ⎢ 2⎥ ⎣ ( 2r + 2 Dε tan θ ) ⎦

[14.33]

where r and D are the core radius and the gauge length respectively, and θ is the light acceptance angle of the fibre. Good correlation was observed between the optically measured strain and that obtained from surface-



413

mounted electrical resistance strain gauges. A drift in the optical signal was observed after 50 000 cycles. This was attributed to degradation of the sensor/matrix interface. Liu et al.109 reported on an EFPI sensor and fluorescent thermometry110 for the simultaneous detection of strain and temperature. A low-finesse EFPI was made using a lead-in single mode fibre and a lead-out multi-mode reflector fibre. The lead-out fibre was spliced to a neodymium-doped fibre that had a core diameter twice that of the lead-out fibre. The opposite end of the Nd-doped fibre was spliced to another multi-mode fibre of similar numerical aperture. The sensor system was illuminated using a superluminescent diode (SLD) where the emission was centred at 850 nm. The use of multi-mode lead-out fibre in the EFPI and the large-core Nd-doped fibre facilitated efficient coupling of light. The Nd-doped fibre fluoresced at ∼940 nm and ∼1064 nm when excited using the SLD emission. The fluorescence was coupled through the multi-mode fibre into a fluorescence-lifetime measurement unit. The emission at 1064 nm was filtered and the decay time (τ) of emission at 940 nm was measured as a function of temperature; the detector had a high sensitivity in this spectral region. As the sensor was embedded in a carbon fibre reinforced prepeg, the response of the sensors was affected by strain and temperature during cross-linking. The absolute strain and temperature was determined using the following equation:

(ΔΔdτ ) = ( AA

11 21

)( )

A12 ε A22 ΔT

[14.34]

The elements of the matrix were determined experimentally to calculate the strain and temperature simultaneously. A residual compressive longitudinal strain of 216 με was obtained using the EFPI sensor. The fluorescence-lifetime of the emission was found to decrease with increasing temperature. Degamber et al.111,112 developed low cost temperature sensors based on a modified EFPI design. Two designs were considered: (i) a capillary made from a material of a different thermal expansion coefficient from that of the silica lead-in fibre and (ii) a reflector fibre. In the first design, a borosilicate capillary was used in place of a silica capillary. The silica fibres were bonded to the borosilicate capillary via fusion splicing. The coefficient of thermal expansion of borosilicate glass (3.5 × 10−6 °C−1) is higher in comparison with fused silica (0.5 × 10−6 °C−1). The change in cavity length (d) per unit temperature change is proportional to the difference of the thermal expansion coefficients, and can be expressed as: Δd = [(α boro − α si )L + α borod ] ΔT

[14.35]

414


where αboro, αsi are the coefficients of thermal expansion of borosilicate glass and fused silica respectively. L is the gauge length of the EFPI sensor. In the second design a soda silicate (9.5 × 10−6 °C−1) fibre reflector was used in conjunction with a silica lead-in fibre and silica capillary. The higher thermal expansion of the soda silica fibre led to a higher temperature sensitivity. The temperature response can be expressed as: Δd = [(α so − α si ) l ] ΔT

[14.36]


where αso and l are the coefficient of thermal expansion of soda silicate fibre, and its length inside the capillary respectively. These sensors were used successfully to monitor temperature in microwave oven- and autoclave-based processing of epoxy resin systems. The first sensor design was operated over a range of −50 to 20 °C with an accuracy of ±1 °C, and over a range of 20 to 250 °C with an accuracy of ±0.5 °C. The second design was found to be suitable for use over a temperature range of 20 to 400 °C with a measurement accuracy of ±0.8 °C.

14.9

Multi-measurand sensor design

Whilst considerable progress continues to be made on the design and deployment of fibre optic sensors for chemical process monitoring and structural integrity assessment, the majority of these sensor designs can only impart information on one or two relevant measurands. For example, in the case of chemical process monitoring of advanced fibre-reinforced composites involving thermosetting resins, it is generally appreciated that cross-linking kinetics can be influenced by a number of factors including the following: the stoichiometry of the reagents, temperature, surface chemistry of the substrate and presence or absence of contaminants. As discussed previously, thermosetting resins also shrink during the crosslinking process. When thermosetting resins are processed above room temperature, upon cooling back to ambient temperature, residual stresses can develop. This is because of the mismatch in thermal expansions between the reinforcing fibres and the matrix. Mahendran et al.113 recently reported progress on the design and demonstration of a novel multi-functional fibre optic sensor that can provide data on (i) temperature, (ii) strain, (iii) refractive index, (iv) transmission infrared spectroscopy and (v) evanescent wave spectroscopy. A unique and attractive feature of this sensor is that a conventional commercially available FTIR spectrometer is used to interrogate the various measurands. The sensor design is based on an extrinsic fibre Fabry–Perot interferometer. A


A

B C

D

E

J

A'

I H G

415

F


14.26 Schematic illustration of the EFPI-based multi-functional sensor design (see main text for details).

schematic illustration of the EFPI-based multi-functional sensor design is presented in Fig. 14.26. The diameter of the optical fibres and the capillary are 125 and 300 μm respectively. Here the conventional EFPI sensor is represented by items [A–E] and [A′]. Items [F–J] represent components of the multi-functional sensor. The details of the individual key components are as follows: [A] and [A′] are the cleaved optical fibres that make up the EFPI sensor. [A′] is referred to as the primary optical fibre in the text; [B] is a precision-bore silica capillary tube which enables alignment of the primary optical fibres; [C] is a fusion joint that ‘bonds’ the primary optical fibres to the capillary. The distance between the fusion joints defines the gauge length for the sensor; [D] is the cleaved optical fibre end-face; [E] is the Fabry–Perot cavity; [F] are secondary optical fibres which are packed around the primary optical fibre [A′]; [G] is the cleaved end-face of a single-mode fibre that is located at a suitable distance from the gold coated capillary end-face. This serves as a Fresnel-based sensor for monitoring the changes in the refractive index of the resin system contained in the secondary cavity; [H] is the secondary cavity (reservoir for the resin system); [I] is the gold coating on the end-face of the precision-bore capillary tube; [J] is a fibre Bragg grating inscribed on the primary optical fibre. The Bragg grating is in a strain-free position and hence is sensitive only to temperature.

14.10 Conclusions A classification scheme was proposed to categorise sensing schemes for fibre reinforced composites. Type-I composites were classified as autonomous materials where the reinforcement, matrix and the interface have intrinsic sensing and responsive capabilities. Type-II composites were composites where the reinforcement, matrix and interface have sensing capabilities. Type-III composites represent the situation where hybrid fibres are used as sensors. Type-IV composites are materials with embedded and surface-mounted sensors. A sub-division of this class is the use of


416


non-contact sensor systems. Type-V composites represent materials with additives and inclusions that perform a sensing function. The discussion on the design and deployment of optical fibres in composites was based around this classification scheme. A wide range of fibre optic sensor systems are available to enable on-line process monitoring of fibre reinforced composites. Inexpensive optical fibre-based sensors, such as Fresnel reflection sensors, can be used to obtain real-time information on the cross-linking kinetics of thermosetting resins. Where quantitative information is required on the rates of reaction and extent of conversion of specified functional groups, FTIR spectroscopy can be used. Where qualitative information will suffice, intensity-based sensors can provide reliable and repeatable results. The sensor systems that are used for process monitoring can be used subsequently for assessing the chemical and mechanical integrity of composites. For example, chemical sensors based on FTIR spectroscopy can be used to infer the diffusion of specified fluids in the composite. The fibre Bragg grating sensor described can be used to map the strain and temperature fields in the composite during service. This approach is used extensively to infer the structural integrity of composites. However, it does assume intimate bonding between the sensor and the composite. In situations where the embedment of conventional optical fibres is a concern due to the mismatch in the relative diameters of the reinforcement and the optical fibres, the reinforcing glass fibres can be used for on-line process monitoring via evanescent wave spectroscopy. An advantage of sensing schemes based on evanescent wave spectroscopy is that spectra can be obtained prior to processing. In other words, it is not necessary for the resin system to be heated up to wet-out the sensor. Thus, process modifications can be made to compensate for the chemical state of the resin system. The deployment of the reinforcing glass fibres as sensors also facilitates on-line damage detection; this can be achieved by monitoring the transmitted light intensity and/or by inspecting the composite for bleeding light. This form of monitoring is ideal for detecting impact damage in composites. A detailed literature review was presented on the use of optical fibre sensors in composites. Sensor systems based on fibre Bragg gratings and extrinsic fibre Fabry–Perot interferometric sensors tend to dominate this field. The FBG is simple to use but sensitive to strain and temperature. Fabry–Perot-based sensors are inexpensive to fabricate and are not sensitive to lateral strain. The review summarised selected papers that considered techniques and schemes for de-coupling strain from temperature. Optical fibre sensors can be integrated into preforms and also retrofitted on composites. The data obtained from fibre optic sensors can be used to infer the chemical, mechanical and physical properties of composites. This


417

information can be used to provide end-users with quantitative data to enable informed decisions to be made on the strategies for repair, recycling and reuse.

14.11 Acknowledgements This chapter is dedicated to the past and current members of the Sensors and Composites Group; this chapter would not have been possible without their hard work and dedication. The authors also wish to acknowledge the inspirational support and encouragement given by Professor Brian Ralph, Mike Bevis and Bryan Harris. The financial support provided by EPSRC, The Royal Society the TSB and the Universities of Brunel, Cranfield (Royal Military College of Science) and University of Birmingham is duly acknowledged.


14.12 References 1. fernando, g. f., hameed, a., tetlow, j., winter, d., leng, j., barnes, r., mays, j. and kister, g., ‘Optical fibre sensors for structural health monitoring of concrete: Part 1: Sensor design, modelling and protection systems’, Journal of Structural Health Monitoring, Vol. 2 (2), 123–135, (2003). 2. measures, r. m., Structural Health Monitoring with Fiber Optical Technology, Academic Press, (2001). 3. senior, j. m., Optical Fiber Communications: Principles and Practice, Prentice Hall International (UK), 2nd Edition, (1992). 4. Commercially available data from specified suppliers. 5. tanaka, y., ‘Synthesis and characterization of epoxides’, Epoxy Resins, Chemistry and Technology, Editor C. A. May, 2nd Edition, Marcel Dekker Inc., 9–284, (1988). 6. may, c. a., ‘Introduction to epoxy resins’, Epoxy Resins, Chemistry and Technology, Editor C. A. May, 2nd Edition, Marcel Dekker Inc., 1–8, (1988). 7. greenfield, d. c. l., PhD Thesis, Brunel University, (1986). 8. hayes, s. a., jones, f. r., marshiya, k. and zhang, w., ‘A self-healing thermoset composite material’, Composites Part-A Applied Science and Manufacturing, Vol. 38, Issue (4), 1116–1120, (2007). 9. wool, r. p., ‘Self-healing materials: a review’, Soft Matter, Vol. 4, 400–418, (2008). 10. ‘Materials on the move – Polymers that respond to a stimuli’, Materials Today, Vol. 11, No. 7–8, (2008). 11. bar-cohen, y., Editor, Electroactive Polymer Actuators as Artificial Muscles, SPIE Press, (2001). 12. moniruzzaman, m., zioupos, p. and fernando, g. f., ‘Investigation of reversible photo-mechanical properties of azobenzene-based polymer films by nanoindentation’, Scripta Meterialia, Vol. 54 (2), 257–261, (2006). 13. harris, d., mahendran, r. s., brooks, d., al-khodairi, f. a. a., machavaram, v. r., reynolds, p., wang, l., pandita, s. d., paget, m., wedderburn, j., malik, s. a., ojo, s. o., kukureka, s. n. and fernando, g. f., ‘Self-sensing, self-healing

418

14.

15.

16.

17.

18.


19. 20.

21.

22.

23. 24.

25. 26. 27.

28. 29.

Management, recycling and reuse of waste composites and crack-arrestor composites’, 16th Annual International Symposium Smart Structures/NDE. Proceedings of SPIE, Vol. 7293, 729204. San Diego, CA. USA, (2009). chung, d. d. l., ‘Damage detection using self-sensing concepts’, Proc. IMechE, Vol. 221, Part G: J. Aerospace Engineering, Special Issue Paper, 509–520, (2007). hayes, s. a., liu, t., martin, a. r. and fernando, g. f., ‘Intensity based optical fibre sensors for damage detection in advanced fibre reinforced composites’, Sensors and their Applications VII, Institute of Physics, Dublin, 10–13 Sept., (1995). fernando, g. f., degamber, b., doyle, c., kister, g. and ralph, b., ‘Smart glass fibre reinforced composites’, Advanced Composites Letters, Vol. 13 (2), 123–129, (2004). wang, l., pandita, s., machavaram, v. r., malik, s., harris, d. and fernando, g. f., ‘Characterisation of the cross-linking process in an E-glass fibre/epoxy composite using evanescent wave spectroscopy’, Composites Science and Technology, Vol. 69, 2069–2074, (2009). wang, l., kister, g., ralph, b. and fernando, g. f., ‘Self-sensing fibre reinforced composites: sol-gel-based claddings’, Journal of Smart Materials and Structures, Vol. 13, 73–81, (2004). Courtesy of l. wang, Sensors and Composites Group, University of Birmingham, (2009). kister, g., badcock, r. and fernando, g. f., ‘A novel technique to study the fracture of E-glass fibres’, Journal of Materials Science Letters, Vol. 39 (4), 1425–1428, (2003). kister, g., ralph, b. and fernando, g. f., ‘Damage detection in GFRP composites using self-sensing E-glass fibres’, Journal of Smart Materials and Structures, Vol. 13 (15), 1166–1175, (2004). malik, s. a., wang, l., mahendran, r. s., harris, d., ojo, s. o., collins, d., paget, m., pandita, s. d., machavaram, v. r. and fernando, g. f., ‘In-situ damage detection using self-sensing composites’, 16th Annual International Symposium Smart Structures/NDE. Proceedings of SPIE, Vol. 7292, 729204. San Diego, USA, (2009). Courtesy of s. malik, Sensors and Composites Group, University of Birmingham, (2009). martin, a., fernando, g. f. and hale, k. f., ‘Impact damage detection in filament wound tubes using embedded optical fibre sensors’, Journal of Smart Materials and Structures, Vol. 6, 470–476, (1997). kemp, w., Organic Spectroscopy, MacMillan Education Ltd, 3rd Edition, (1991). weyer, l. g., ‘Near-infrared spectroscopy of organic substances’, Applied Spectroscopy Reviews, Vol. 21 (1 & 2), 1–43, 1985. stark, e., luchter, k. and margoshes, m., ‘Near-infrared analysis (NIRA): a technology for quantitative and qualitative analysis’, Applied Spectroscopy Reviews, Vol. 22 (4), 335–399, (1986). whetsel, k. b., ‘Near-infrared spectrophotometry’, Applied Spectroscopy Reviews, Vol. 2 (1), 1–67, (1968). goddu, r. f., ‘Near-infrared spectrophotometry’, Advances in Analytical Chemistry and Instrumentation, Vol. 1, 347–425, (1960).



419

30. miller, c. e., ‘Near-infrared spectroscopy of synthetic polymers’, Applied Spectroscopy, Vol. 26 (4), 277–339, (1991). 31. howell, h. e. and davis, j. r., ‘Qualitative identification of polymeric materials using near-infrared spectroscopy’, Advances in Chemistry Series, Vol. 236, 263–285, (1993). 32. lachenal, g., ‘Dispersive and Fourier transform near-infrared spectroscopy of polymeric materials’, Vibrational Spectroscopy, Vol. 9, 93–100, (1995). 33. chabert, b., lachenal, g., vinh, c. and tung, c. v., ‘Epoxy resins and epoxy blends studied by near infra-red spectroscopy’, Macromolecular Symposia, Vol. 94, 145–158, (1995). 34. goddu, r. f. and delker, d. a., ‘Determination of terminal epoxides by nearinfrared spectrophotometry’, Analytical Chemistry, Vol. 30 (12), 2013–2016, (1958). 35. wheeler, o. h., ‘Near infrared spectra – a neglected field of spectral study’, Journal of Chemical Education, Vol. 37 (5), 234–236, (1960). 36. goddu, r. f. and delker, d. a., ‘Spectra-structure correlations for the nearinfrared region’, Analytical Chemistry, Vol. 32 (1), 140–141, (1960). 37. lohman, f. h. and nortman jr., w. e., ‘Determination of primary and secondary aliphatic amines by near-infrared spectrophotometry’, Analytical Chemistry, Vol. 35 (6), 707–711, (1963). 38. whetsel, k., roberson, w.e. and krell, m. w., ‘Near-infrared analysis of mixtures of primary and secondary aromatic amines’, Analytical Chemistry, Vol. 30 (10), 1594–1597, (1958). 39. whetsel, k. b., roberson, w. e. and krell, m. w., ‘Near-infrared spectra of primary aromatic amines’, Analytical Chemistry, Vol. 30 (10), 1598–1604, (1958). 40. dannenberg, h., ‘Determination of functional groups in epoxy resins by nearinfrared spectroscopy’, SPE Transactions, January, 78–88, (1963). 41. tosi, c. and pinto, a., ‘Near-infrared spectroscopy of hydrocarbon functional groups’, Spectrochimica Acta, Vol. 28A, 585–597, (1972). 42. fu, j. h. and schlup, j. r., ‘Mid- and near-infrared spectroscopic investigations of reactions between phenyl glycidyl ether (PGE) and aromatic amines’, Journal of Applied Polymer Science, Vol. 49, 219–227, (1993). 43. xu, l., fu, j. h. and schlup, j. r., ‘In situ near-infrared spectroscopic investigation of epoxy resin-aromatic amine cure mechanisms’, Journal of the American Chemical Society, Vol. 116, 2821–2826, (1994). 44. mijovic, j. and andjelic, s., ‘A study of the reaction kinetics by near-infrared spectroscopy. 1. Comprehensive analysis of a model epoxy/amine system’, Macromolecules, Vol. 28, 2787–2796, (1995). 45. mijovic, j., andjelic, s. and winnie yee, c. f., ‘A study of reaction kinetics by near-infrared spectroscopy. 2. Comparison with dielectric spectroscopy of model and multifunctional epoxy/amine systems’, Macromolecules, Vol. 28, 2797–2806, (1995). 46. min, b. g., stachurski, z. h., hodgkin, j. h. and heath, g. r., ‘Quantitativeanalysis of the cure reaction of DGEBA/DDS epoxy-resins without and with thermoplastic polysulfone modifier using near-infrared spectroscopy’, Polymer, Vol. 34 (17), 3620–3627, (1993).


420


47. strehmel, v. and scherzer, t., ‘Structural investigation of epoxy amine networks by mid- and near-infrared spectroscopy’, European Polymer Journal, Vol. 30 (3), 361–368, (1994). 48. chike, k. e., myrick, m. l., lyon, r. e. and angel, s. m., ‘Raman and near-infrared studies of an epoxy resin’, Applied Spectroscopy, Vol. 47 (10), 1631–1635, (1993). 49. schiering, d. w., katon, j. e., drzal, l. t. and gupta, v. b., ‘An infrared spectroscopic investigation of the curing of the EPON 828/meta-phenylenediamine system’, Journal of Applied Polymer Science, Vol. 34, 2367–2375, (1987). 50. kozielski, k. a., george, g. a., st john, n. a. and billingham, n. c., ‘Kinetic studies by FT-NIR of the curing reactions of two glycidyl ether epoxy resins mixed with stoichiometric quantities of 4,4′-DDS’, High Performance Polymers, Vol. 6, 263–286, (1994). 51. poisson, n., lachenal, g. and sautereau, h., ‘Near- and mid-infrared spectroscopy studies of an epoxy reactive system’, Vibrational Spectroscopy, Vol. 12, 237–247, (1996). 52. george, g. a., cole-clarke, p., st. john, n. a. and friend, g., ‘Real-time monitoring of the cure reaction of a TGDDM/DDS epoxy resin using fibre optic FTIR’, Journal of Applied Polymer Science, Vol. 42, 643–657, (1991). 53. st john, n. a. and george, g. a., ‘Cure kinetics and mechanisms of a tetraglycidyl4,4′-diaminodiphenylmethane/diaminodiphenylsulphone epoxy resin using near infrared. spectroscopy’, Polymer, Vol. 33 (13), 2679–2687, (1992). 54. debakker, c. j., george, g. a., st john, n. a. and fredericks, p. m., ‘The kinetics of the cure of an advanced epoxy resin by Fourier transform Raman and near-IR spectroscopy’, Spectrochimica Acta, Vol. 49A (5), 739–752, (1993). 55. debakker, c. j., st john, n. a. and george, g. a., ‘Simultaneous differential scanning calorimetry and near-infrared analysis of the curing of tetraglycidyl diaminodiphenylmethane with diaminodiphenylsulphone’, Polymer, Vol. 34 (4), 716–725, (1993). 56. scherzer, t., ‘Characterization of diol modified epoxy resins by near- and mid-infrared spectroscopy’, Journal of Applied Polymer Science, Vol. 51, 491–502, (1994). 57. paputa peck, m. c., carter iii, r. o. and qaderi, s. b. a., ‘Near-infrared measurements of terminal epoxides in polymer resin systems. Part 1 – Analytical considerations’, Journal of Applied Polymer Science, Vol. 33, 77–86, (1987). 58. fernando, g. f. and al-khodairi, f. a. a., ‘The fatigue of hybrid composites’, Chapter 7, 189–241. Woodhead Publishing, B. Harris, Editor, Fatigue in Composites: Science and technology of the fatigue response of fibre-reinforced plastics, (2003). 59. crosby, p., powell, g., spooncer, r., france, c. and fernando, g. f., ‘In-situ cure monitoring in advanced composites’, Smart Structures and Materials, SPIE, 2444, 386–395, (1995). 60. crosby, p. a., connolley, m. and fernando, g. f., ‘Cure monitoring of advanced fibre reinforced composites using an optical fibre sensor’, 7th International Conf. on Fibre Reinforced Composites ’98, Newcastle, UK, 128–135, (1998). 61. mahendran, r. s., chen, r., wang, l., machavaram, v. r., pandita s. d., kukureka, s. n. and fernando, g. f., ‘Chemical process monitoring and the detection of moisture ingress in composites’, The 15th International


62.

63.

64.

65.

66.


67.

68. 69.

70.

71. 72.

73. 74.

75.

76.

421

Symposium on – Smart Structures and Materials and Non destructive Evaluation and Health Monitoring, Proc. SPIE, 6933, 64232S, San Diego, USA, March, (2008). machavaram, v. r., badcock, r. a. and fernando, g. f., ‘Fabrication of intrinsic fibre Fabry-Perot cavities in silica optical fibres via F2-laser ablation’, Measurement Science and Technology, Vol. 18 (3), 928–934, (2007). machavaram, v. r., badcock, r. a. and fernando, g. f., ‘Fabrication of intrinsic Fabry-Perot sensors in silica fibres using hydrofluoric acid etching’, Sensors and Actuators, A – Physical, Vol. 138 (1), 248–260, (2007). liu, t., al-khodairi, f., wu, m., irle, m. and fernando, g. f., ‘In-situ strain monitoring in composites using an embedded extrinsic Fabry – Perot interferometric sensor and a CCD detection system’, SPIE, Fibre Optic Sensors V, Beijing, China, SPIE Proceedings Series, 2895, 279–287, (1996). dumitrescu, o., degamber, b. and fernando, g. f., ‘Non-contact process monitoring of thermosets’, 10th European Conference on Composite Materials, Bruges, Belgium, Paper 230, (2002). degamber, b. and fernando, g. f., ‘Microwave processing of thermosets: non-contact cure monitoring and fibre optic temperature sensors’, Journal of Plastics, Rubber and Composites, Vol. 32 (8/9), 327–333, (2003). degamber, b., winter, d., tetlow, j., teagle, m. and fernando, g. f., ‘Simultaneous thermal (DSC) spectral (FTIR) and physical (TMA)’, Journal of Measurement Science and Technology, Vol. 15 (9), L5–L10, (2004). degamber, b., PhD Thesis, Cranfield University, Royal Military College of Science, UK, (2003). mahendran, r. s., machavaram, v. r., wang, l., burns, j. m., harris, d., kukureka, s. n. and fernando, g. f., ‘A novel multifunctional fibre optic sensor’. Eds. N. G. Meyendorf, K. J. Peters, W. Ecke, SPIE/Smart Structures and Materials & Nondestructive Evaluation and Health Monitoring 2009: Smart Sensor Phenomena, Technology, Networks, and Systems. 2009; Proceedings of SPIE, 7293: 72930C, San Diego, USA, March, (2009). liu, t. and fernando, g. f., ‘Processing of polymer composites: an optical fibre-based sensor system for on-line amine monitoring’, Composites: Part A, Vol, 32, 1561–1572, (2001). paul, p. h. and kychakoff, g., ‘Fibre optic evanescent field absorption sensor’, Applied Physics Letters, Vol. 51 (1), 12–14, (1987). gupta, b. d. and singh, c. d., ‘Fiber-optic evanescent field absorption sensor: a theoretical evaluation’, Fiber and Integrated Optics, Vol. 13, 433–443, (1994). snyder, a. w. and love, j. d., Optical Waveguide Theory, Kluwer Academic Publishers, (1983). crosby, p. a., powell, g. r., fernando, g. f., france, c. m., spooncer, r. c. and waters, d. n., ‘In situ cure monitoring of epoxy resins using optical fibre sensors’, Smart Materials and Structures, Vol. 5, 425–428, (1996). liu, y. m., ganesh, c., steele, j. p. h. and jones, j. e., ‘Fibre optic sensor development for real-time in-situ epoxy cure monitoring’, Journal of Composites Materials, Vol. 31 (1), 87–102, (1997). powell, g. r., crosby, p. a., fernando, g. f., spooncer, r. c., france, c. m. and waters, d. m., ‘In-situ cure monitoring using optical fibre sensors – a comparative study’, Journal of Smart Materials and Structures, Vol. 7, 557–568, (1998).


422


77. cusano, a., cutolo, a., giordano, m. and nicolais, l., ‘Optoelectronic refractive index measurements: application to smart processing’, IEEE Sensors Journal, Vol. 3 (6), 781–787, (2003). 78. blythe, t. and bloor, d., Electrical Properties of Polymers, Cambridge University Press, 2nd Edition, (2005). 79. vacher, s., molimard, j., gagnaire, h. and vautrin, a., ‘A Fresnel’s reflection optical fiber sensor for thermoset polymer cure monitoring’, Polymers and Polymer Composites, Vol. 12 (4), 269–276, (2004). 80. kim, c. b. and su, c. b., ‘Measurement of the refractive index of liquids at 1.3 and 1.5 micron using a fibre optic Fresnel ratio meter’, Measurement Science and Technology, Vol. 15, 1683–1686, (2004). 81. iadicicco, a., cusano, a., persiano, g. v., cutolo, a., bernini, r. and giordano, m., ‘Refractive index measurements by fibre Bragg grating sensor’, Proceedings of IEEE, Vol. 1, 101–105, (2003). 82. schroeder, k., ecke, w., mueller, r., willsch, r. and andreev, a., ‘A fiber Bragg grating refractometer’, Measurement Science and Technology, Vol. 12, 757–764, (2001). 83. liang, w., huang, y., xu, y., lee, r. k. and yariv, a., ‘Highly sensitive fiber Bragg grating refractive index sensors’, Applied Physics Letters, Vol. 86, (15) 1122: 1–3, (2005). 84. shao, l. y., zhang, a. p., liu, w. s., fu, h. y. and he, s., ‘Optical refractive index sensor based on dual fiber Bragg gratings interposed with a multimode fiber taper’, IEEE Photonics Technology Letters, Vol. 19 (1), 30–32, (2007). 85. guo, s. and albin, s., ‘Transmission property and evanescent wave absorption of cladding multimode fiber tapers’, Optics Express, Vol. 11 (3), 215–223, (2003). 86. laffont, g. and ferdinand, p., ‘Sensitivity of slanted fibre Bragg gratings to external refractive index higher than that of silica’, Electronics Letters, Vol. 37 (5), 289–290, (2001). 87. hill, k. o., fuji, y., johnson, d. c. and kawasaki b. s., ‘Photosensitivity in optical fiber waveguides – application to reflection filter fabrication,’ Applied Physics Letters, Vol. 32, 647–649 (1978). 88. kashyap, r., Fiber Bragg Gratings, Academic Press, (1999). 89. othonos, a. and kalli, k., Fibre Bragg Gratings: Fundamentals and Applications in Telecommunications and Sensing, Artech House, Inc., (1999). 90. murphy, k. a., gunther, m. f., wang, a., claus, r. o. and vengsarkar, a. m., Optical Fiber Sensors Conference, Vol. 8, 193–196, (1992). 91. wolf, e., Introduction to the Theory of Coherence and Polarization of Light, Cambridge University Press, (2007). 92. fernando, g. f., crosby, p. a. and liu, t., ‘The application of optical fibre sensors in advanced fibre reinforced composites: Chapter 2 – Introduction and issues’, Optical Fibre Sensor Technology, Volume III, Edited by K. T. V. Grattan, and B. T. Meggitt, Kluwer Academic Publishers, (1999). 93. crosby, p. a. and fernando, g. f., ‘The application of optical fibre sensors in advanced fibre reinforced composites: Chapter 3 – Cure monitoring’, Optical Fibre Sensor Technology, Volume III, Edited by K. T. V. Grattan, and B. T. Meggitt, Kluwer Academic Publishers, (1999). 94. liu, t. and fernando, g. f., ‘The application of optical fibre sensors in advanced fibre reinforced composites: Chapter 4 – Strain, temperature and health


95.

96.

97.

98.

99.

100.


101.

102.

103.

104.

105.

106.

107.

108.

109.

423

monitoring’, Optical Fibre Sensor Technology, Volume III, Edited by K. T. V. Grattan and B. T. Meggitt, Kluwer Academic Publishers, (1999). kuang, k. s. c., kenny, r., whelan, m. p., cantwell, w. j. and chalker, p. r., ‘Residual strain measurement and impact response of optical fibre Bragg grating sensors in fibre metal laminates’, Smart Materials and Structures, Vol. 10, 338–346, (2001). kuang, k. s. c., kenny, r., whelan, m. p., cantwell, w. j. and chalker, p. r., ‘Embedded fibre Bragg grating sensors in advanced composite materials’, Composite Science and Technology, Vol. 61, 1379–1387, (2001). lee, j., tsuda, h. and koo, b., ‘Single-mode fibre optic Bragg grating sensing on the base of birefringence in surface-mounting and embedded applications’, Optics and Laser Technology, Vol. 39, 157–164, (2007). guemes, j. a. and menendez, j. m., ‘Response of Bragg grating fibre-optic sensors when embedded in composite laminates’, Composite Science and Technology, Vol. 62, 959–966, (2002). sorensen, l., gmür, t. and botsis, j., ‘Residual strain development in an AS4/ PPS thermoplastic composite measured using fibre Bragg grating sensors’, Composites: Part A, Vol. 37, 270–281, (2006). storoy, h. and johannessen, k., ‘Glue induced birefringence in surface mounted optical fibres’, Electronics Letters, Vol. 33 (9), 800–801, (1997). wagreich, r. b., atia, w. a., singh, h. and sirkis, j. s., ‘Effects of diametric load on fibre Bragg gratings fabricated in low birefringent fibre’, Electronics Letters, Vol. 32 (13), 1223–1224, (1996). giordano, m., laudati, a., russo, m., nasser, j., persiano, g. v. and cusano, a., ‘Advanced cure monitoring by optoelectronic multifunction sensing system’, Thin Solid Films, Vol. 450, 191–194, (2004). colpo, f., humbert, p., giaccari, p. and bitsis, j., ‘Characterization of residual strains in an epoxy block using an embedded FBG sensor and the OLCR technique’, Composites: Part A, Vol. 37, 652–661, (2006). antonucci, v., cusano, a., giordano, m., nasser, j. and nicolais, l., ‘Cure-induced residual strain build-up in a thermoset resin’, Composites: Part A, Vol. 37, 592–601, (2006). liu, t., fernando, g. f., rao, y. j., jackson, d. a., zhang, l. and bennion, i., ‘Simultaneous strain and temperature measurements in fibre reinforced composites using optical fibre sensors’, Conference on Smart Sensing, Processing and Instrumentation, SPIE, San Diego, Vol. 3042, 203–212, (1997). singh, h. and sirkis, j. s., ‘Temperature and strain measurement by combining ILFE and Bragg grating optical fibre sensors’, Journal of Experimental Mechanics, Vol. 37 (4), 414–419, (1997). fernando, g. f., liu, t., crosby, p. a., doyle, c., martin, a., brooks, d., ralph, b. and badcock, r. a., ‘A multi-purpose optical fibre sensor design for fibre reinforced composite materials’, Journal of Measurement Science and Technology, Vol. 8, 1065–1079, (1997). badcock, r. a. and fernando, g. f., ‘An intensity-based optical fibre sensor for fatigue damage detection in advanced fibre-reinforced composites’, Smart Structures and Materials, Vol. 4, 223–230, (1995). liu, t., fernando, g. f., zhang, z. y. and grattan, k. t. v., ‘Simultaneous strain and temperature measurements in composites using extrinsic Fabry–Perot

424

110. 111.

112.

113.


114.

Management, recycling and reuse of waste composites interferometric and intrinsic rare-earth doped fibre sensors’, Sensors and Composites, Vol. 80, 208–215, (2000). grattan, k. t. v. and zhang, z. y., Fiber Optic Fluorescence Thermometry, Chapman & Hall, (1994). degamber, b., tetlow, j. and fernando, g. f., ‘Design and development of low-cost optical fiber sensors for temperature metrology: process monitoring of an epoxy resin system’, Journal of Applied Polymer Science, Vol. 94, 83–95, (2004). degamber, b. and fernando, g. f., ‘Fiber optic sensors for noncontact process monitoring in a microwave environment’, Journal of Applied Polymer Science, Vol. 89, 3868–3873, (2003). mahendran, r. s., machavaram, v. r., wang, l., kukureka, s. n., paget, m. and fernando, g. f., ‘A novel multi-measurand fibre optic sensor’, 16th Annual International Conference on Composites/Nanoengineering (ICCE-16), Kunming, China, July, (2008). Courtesy of s. krishnamurthy, ‘Design and evaluation of a pre-stressing methodology for advanced fibre reinforced composites’, PhD thesis, Cranfield University, UK, (2005).

15 New developments in producing more functional and sustainable composites G. F. S M I T H, University of Warwick, UK

Abstract: There will always be a new combination of resin and matrix material to extend the range of composites so this chapter can at best illustrate the trends in composite material applications. The major driver in today’s low carbon economy markets is low weight products, in which case, composites have a chance to expand in application. Additionally, there is a keen interest in finding composites that meet environmental credentials. This chapter reviews currently applied composite materials in both thermoset and thermoplastic resins and tries to identify the composite technology that will enable a ‘step change’ in the volume application of these materials whilst keeping environmental targets in mind.


Key words: composites, fibres, nanotubes, sustainable, biocomposites.

15.1

Introduction

Multiple varieties of composites exist and their exploitation will be improved by new energy efficient process methods. In addition, composites will continue to be developed as the potential for new and varied fibres is developed; the different combinations and contributions are endless. However, the crucial requirement is an improvement over the current systems and of course at the appropriate cost. The cost criterion has been crucial to the wider introduction of carbon fibre composite which, apart from high cost, maintains its status as the most capable material to match the modulus of steel. New developments in sustainability will inevitably promote the introduction of natural fibres. Nevertheless, a senior automotive engineer stated succinctly that until the application of these materials fibres can match a glass fibre composite then there was no practical incentive to use them. Non-fibre composites are becoming a research phenomenon with the awareness that particulate in the nano-size can provide distinct properties. Any consideration of the future of composites must not ignore the ever cost-effective glass fibre reinforced composite, which remains for the present the major material of choice for industry interested in low 425

426


weight structures. In the composite ‘end-of-life’ debate it is worth stating that all materials can be recycled as long as there is a market for the resultant product; however, some materials are easier to process than others.

15.2

Glass fibre composites

Glass fibre reinforced composites are produced by a variety of processes and their future will be discussed separately as thermoset resin systems and thermoplastic formulations.


15.2.1 Thermoset resin impregnated glass fibre systems At the low end of technology there is hand lay-up whereas for high technology there is the very structural prepreg process as used widely in the aerospace industry. Compression moulding of sheet moulding compound (SMC) meets semi-structural capabilities, e.g. car bonnets, and goes some way to reach a higher volume industry but structural components containing highly oriented glass fibre will still need to adapt to high volume processes if they are to expand application. Thus, with this in mind, developments have been focused on an improved manufacturing technique. Volume manufacture has been boosted by improved equipment control and methods for handling and placing fibres in specialist injection machines which has improved efficiency considerably. For instance, Krauss Mafei is well known for the injection of thermoset resin into glass performs and launched a method for wetting fibres in the polyurethane mixing head prior to extrusion of the mix into the tool for compression moulding (Rudd, 2001). This was not received as entirely novel since the orientation of fibres in a tool prior to injection was developed in the early 1990s by Owens Corning with its P4 process (Gerard and Jander, 1993). This provides added capability in the thermoset injection processes by an automated fibre lay and preform process prior to resin impregnation. The technique uses a robot arm to deliver glass fibre on to a tool surface. The fibres can be oriented directionally or distributed randomly and are held in place to produce the perform: consolidation is effected by heat and pressure since there is a small amount of thermoplastic binder on the glass. This preform is then placed in another tool and can be resin injected to form the composite. The benefits are the speed of an oriented glass perform manufacture without the waste that is produced when prepreg is used. Already this process is finding application in the automotive industry and work is continuing on the suggestion that the preform process and resin injection could take place in the same tool.

Producing more functional and sustainable composites

427

15.2.2 Thermoplastic resin glass fibre systems


Thermoplastic compounds of glass fibre are established in the high volume automotive industry as ‘engineering plastics’ (Fig. 15.1). The structural capabilities of these composites have been enhanced by the use of even longer fibres in composite structures with controlled orientation. This technology was first introduced in the product glass mat thermoplastic (GMT) with several material combinations containing unidirectional (UD) glass fibre. A further enhancement was marketed as a Twintex product that integrated woven glass into polypropylene. This product provides many future innovations as different fibre combinations can be introduced such as Kevlar and carbon fibre. Other thermoplastic resin systems are also

Pole impact deflection test

15.1 Thermoplastic composites (glass fibre/polypropylene(PP) covered PP foam) application as a bumper armature in impact test (University of Warwick).

428


becoming available and provide even more structural and material benefits; a polyamide impregnated woven glass produces a particular cost-effective composite and there is even a polyether ether ketone (PEEK) impregnated system, although, the latter requires a specialised moulding process (Ma and Lin, 1998). The enchantment of composites is the flexibility in the many varied compositions that can be introduced to the same moulding tool and process. It is merely necessary to change the weave or its fibre orientation or fibre type or resin. It is the contention of this writer that designers can use this opportunity for material variants such that they can proceed to tool manufacture with confidence in the scope to achieve mechanical properties. In a particular case, ultimate change can be made in the material characteristics from glass fibre (UD MPa) to carbon fibre (UD MPa).

15.3

Carbon fibre composites


15.3.1 Carbon fibre developments Carbon fibre composites have always been considered as the material to challenge steel in highly structural applications. There have been lessons regarding the limitations in impact and designers have to take particular note in this respect. However a major inhibitor to its wider use in a volume industry, e.g. automotive, has been the cost when compared to glass fibre although a cost effective application has been demonstrated by A.B. Lovins with his Hypercar (Lovins and Cramer, 2004). A recent boost in interest was provided by the supply of cost competitive material from Eastern Europe with mid-grade carbon fibre ranging from £10 to £20 per kilogram. However, there remains the real challenge to increase volume fibre production and improve the energy of production. The improvement in manufacture process cycle of carbon fibre composites is considered one way to produce a cost-effective product. Carbon fibre SMCs and carbon fibre thermoplastics are available and a second generation P4 process is being introduced with these fibres (Harper et al., 2007). A radical project to reduce cost and improve the sustainability of carbon fibre production is being addressed in the USA within its Freedom Car project (Carpenter et al., 2007). In this work, the aim is to develop a process from a sustainable precursor, notably, lignin rather than the conventional polyacrylonitrile (PAN) polymer. With conventional processing using a carbon fibre-grade (CF) PAN, precursor is slightly over 50% of the carbon fibre cost. Thus any reduction in precursor cost has the largest impact on CF manufacturing cost. The plan is to produce fibre from lignin for a cost of £9–13 per kg.


429

15.3.2 Recycled carbon fibres As described in Chapter 12 carbon fibre can be recovered by thermal pyrolysis of the composite in a ‘non air’ environment. Generally discarded aerospace components or prepreg offcuts of manufacture are used and the recovered fibres are said to have only slightly poorer quality than the original fibre. Nevertheless, these fibres normally find use in composites with less demanding structural requirements. This technique can never be a major carbon fibre source but does address the important issue of disposal that might have otherwise inhibited the growth of carbon composites.


15.3.3 Carbon nanotubes Carbon nanotubes have been around since the mid-1970s but their importance has only recently been highlighted. Since small quantities of nanoparticles have been found to impart extraordinary strength and unique toughness (Zhang, 2005) to composites in which they are compounded, there is much interest in the qualities of these materials. Carbon nanotubes have been found to produce a tensile strength up to 63 MPa in an individual multi-walled tube (Yu, 2000), although it is exceedingly variable. The major benefits can be created by the small quantities that can exhibit both strength and stiffness whilst other functions such as electrical conductivity provide additional opportunities. Innovative applications are already being proposed for lightweight structural components utilising both mechanical and electrical properties.

15.4

Natural fibres

Natural fibres are being addressed for their environmental acceptability. Natural fibres have a history of application in composites. The Old Testament relates that the Israelites requested straw to strengthen bricks; in later times there was wattle and daub and plywood which made similar contributions to structures. Nature’s fibres have been widely adopted in engineering. Cotton was recognised as providing an excellent reinforcement in phenolic resin (Paxolin) for electrical insulation. Asbestos was similarly used as a means of reinforcement in composites until the fibre was found to produce severe health problems. The asbestos was sometimes formed in to asbestos paper for reinforcement but cellulose paper could be similarly used to produce a composite laminate (Tufnol) of excellent strength. The main market for renewable fibre reinforced composites in the EU is expected to be the automotive industry (EU Parliament, 2005) and the most likely fibres will be flax, sisal, jute, kenaf and hemp.


430


The application of natural fibres in composites may vary around the world since geographic location dictates the indigenous plant cultivation. In Europe there has been a concentration on hemp and miscanthus whereas fibres such as Kenaf have been predominant in Asia. The automotive industry has been quick to seize the idea that such materials might meet public acceptability. The most adventurous application has been the use of sustainable resins combined with Kenaf fibre for body panels at Toyota (Fig. 15.2). Other users are more cautious with applications for parcel shelves and door trim mouldings of wood fibre or hemp as the filler in conventional materials. In some cases it is suggested that, at the very least, these components can be disposed of by burning. Generally, the problem of wider usage is the lack of material stiffness and strength of such materials. In an effort to encourage more sustainability in the automotive industry, Warwick developed the ECO1 formula student vehicle that utilised hemp composites for the body panels and cashew nut resin system in the brakes (Fig. 15.3). This proved an excellent marketing tool for natural materials and, hence, the project has progressed to produce the next generation sustainable vehicle, a Formula 3 car to demonstrate that these materials can be applied for ‘real’ in the motorsport industry (Fig. 15.4). This car has panels produced from polylactic acid (PLA) and using flax and recycled carbon fibre. Nevertheless, the technical science relating to these materials has improved our application of them. In particular, it is recognised that these natural materials can have a varied composition and must be controlled throughout their horticulture as well as their processing to provide the best composites. In addition a good composite requires that the fibre has a high elastic modulus, high strength and high aspect ratio. There are many fibres that

15.2 Kenaf fibre body panels in a Toyota Car application.


431


15.3 Warwick Formula Student car, ECO1.

15.4 Warwick Formula 3 racing car demonstrates the application of sustainable materials with the author at the wheel.

can be classed with this distinction but the highest properties are achieved from a unifibre at its greatest aspect ratio. A comparison of properties is made in Table 15.1. Splitting fibres into individual strands of fibres is traditionally done in the fields or on the side of a river by leaving the plant to be weathered by the

Density [g/cm ] Tensile strength [N/mm2] Stiffness [kN/mm2] Elongation at break [%] Moist absorption [%] Price of raw fibre [£/kg]

3

2.55 2400 73 3 – 0.9

Glass 1.4 800–1500 60–80 1.2–1.6 7 0.4–1.1

Flax 1.48 550–900 70 1.6 8 0.4–1.3

Hemp

Table 15.1 Natural fibre properties compared with glass fibre


1.46 400–800 10–30 1.8 12 0.3

Jute 1.5 500 44 2 12–17 1.1–1.8

Ramie 1.25 220 6 15–25 10 0.2–0.4

Coir

1.33 600–700 38 2–3 11 0.4–0.5

Sisal

1.51 400 12 3–10 8–25 1.1–1.6

Cotton


433

environment; the process is called retting. Unfortunately, it is extremely difficult to process nature’s materials to the single fibre level and the norm is to process fibres to the highest aspect possible, which results in fibre bundles and drastically reduced properties. The poor aspect ratio of natural fibres in composites produces properties with a much lower value than the unifibre strength would suggest. A further factor to affect the composite structure is the surface chemistry of this retted fibre and careful modification is required if the binding resin is to adequately bond in order to perform the vital role of transferring load from fibre to fibre. Surface treatment and surface coating therefore plays another significant role in the development of these natural fibre composites. New technologies are being researched to address the purity and aspect ratio of natural materials (EU 6th Framework Report, 2006):


•

Horticulture – Horticulturists are improving the germination rate to improve crop yields and, hence, land utilisation (Wright et al., 2003). The crop must also produce a consistent fibre and hence, there is research to define and select the best strain of plant to optimise the resultant fibre (Pink and Teakle, 2005) • Microbiology – Science teaches that fibre separation is achieved by bugs degrading the natural bonds holding fibres together (Haque et al., 2003). Rather than leave this separation process to environmental chance, it is conceivable that microbiology can provide a solution with the use of controlled and carefully selected micro-organisms. Until the improvements in fibre harvesting and processing come into effect then natural fibres will not be used as structural materials. More importantly, they will not be introduced whilst they are unable to produce composites with mechanical properties that can compete with glass fibre ones. Thus, the challenge is for the horticulturists to provide fibres of consistent quality that will allow processors to devise new means of producing high aspect ratios, and then surface treatments need to be developed so as to produce resin compatibility. Meanwhile these natural fibres will continue to be introduced as non structural token applications for the environment.

15.5

Multi-layer, multi-functional composites

It is a major benefit to the structure of a composite that its properties can be varied throughout its thickness. These changes can be designed for the part by altering the fibre materials or interlaying different fibre orientations or resins. Thus changes in the material functionality can often be more easily integrated in composite manufacture than any other material. Multilayer techniques are more often used to change the mechanical structure

434


but integrating sensors and detectors in to structures is being used to enhance the capabilities of a composite structure (Etches and Fernando, 2009). Work in this area is likely to grow with the availability of methods for self-diagnosis of material breakdown or excess stress analysis during ‘in field’ application. Warwick has been developing methods for adding functionality during the high volume moulding process and Warwick’s technique for ‘in mould’ coating (Goodship and Smith, 2008) can be applied to introduce electronic functions on the surface during injection moulding. This suggests that three-dimensional components can be produced that are ‘smart’ with increased value, such as a battery function, solar cell or even video capability (Smith et al., 2008).


15.6

Sustainability

Given the variety of different binders and reinforcements and combinations that can be tried, new composites will continue to be developed. However, there is now the added green responsibility to produce materials that can find levels of disposal. This may take several forms in easing methods of disassembly by introducing additives that enable unzipping of the adhesion at the surface or replaceable and recyclable parts that allow the fashion upgrade of a design. This perception of more environmentally engineered products pervades design and selection but provides engineers with new opportunities for inspired combinations, many of which are heralded in nature. Thus a focus for the environmentally conscious engineer has been improved biocomposites for the future.

15.6.1 Biocomposites The composite industry can embrace the disposal of the composite component by declaring its biodegradable nature when composted such that landfill is not an issue. This has resulted in research in to the development of natural fibres (previously, discussed) with biodegradable resins.

15.6.2 Biodegradable resins Degradation in resins can be induced by the inclusion of a biodegradable filler such as starch within the composition but this does not provide complete sustainability. The renewable polymer is one that is directly or indirectly derived from biomass. Thus, there are natural polymeric materials extracted from biomass such as polysaccharides and celluloses. One of the earliest resins to be used was shellac, which still finds wide application as a composite binder in electrical insulation. However, when biopolymers are mentioned then the first thought is synthetic polymers made from cultivated



435

crop materials such as starch and sugar beet. Some biopolymers are produced from oils but their environmental credentials require that the oil was produced from biomass and not from a petroleum industry source. The biopolymer receiving considerable attention is PLA derived from starch fermentation. A conundrum is highlighted in the application of biocomposites in that there is a need for biodegradability but with durability; it is no good having a composite that biodegrades if it does so during its application life. The challenge is to produce such materials that can be switched to disposal at the end of their life cycle or to produce a composition with a predetermined life expectancy. Coating is a way of protection for the core biomaterial and providing it is less than 1% of the component by weight then the item is still considered biodegradable. Of course the surface barrier has to be broken or scratched to allow the degradation process and this could be a hazard for the component in use. Thus the focus is on polymers as coatings that can be altered by external means to achieve their bio-acceptable status. For instance, this was achieved at Warwick by coating bioresin with polyvinyl acetate (PVAc); then it required a simple chemical hydrolysis of the water insoluble PVAc to water soluble polyvinyl alcohol (PVAL) (US Patent, 1999): R1–[–CH2–CH–]n–R 2 + H2O COCH 3 PVAc

R1–[–CH2–CH–]n–R2 + CH3CO2H OH PVAL

acetic acid [15.1]

This highlights an alternative method of disposal of a polymer whereby it can be unzipped such as the hydrolysis of non-biodegradable polyesters back to their monomer: HO–[–R1–O 2CR2–]n–CO2H + H2O

Polyester

HO–[–R1–]–OH + HO2C–R2–CO2H

Diol

Di-acid

[15.2] The work on biodegradable composites produced a somewhat unexpected resistance from the recycle industry. Apparently, these materials are declared incompatible with many of the synthetic resins and have to be separated or avoided. This highlights the complex nature of sustainability, disposal and recycling. The importance of this approach to the environment means that considerable attention will be given in the future to improving the traceability, collection and separation of materials, in particular better

436


tagging techniques. With current means of introduction of tags to composites and the means for their detection, the technology is available for automatic separation at point of disposal (Brite Euram Project BE-7148, 2004).

15.7

Biomimetics


People have always been inspired by nature, so why not add nature’s best endeavour, such as bone, to a composite? However, the future of composites is likely to lie in nature’s fibres of high aspect ratio, e.g. human hair and the web of a spider. Electrospinning is one such method that is successful in the production of long nanofibres by drawing out the fibre using an electric charge. The collection of these fibres is a drawback since although the fibres are continuous so they are randomly distributed on any target collection device. Nevertheless, the introduction of bioresourced resins in to such a system can produce a thin film of fibres for potential use as tissue replacement in the healthcare sector. Novel in situ composite development in the human body is being progressed with a particulate called bioglass which is being used to coat and form scaffolds on which nature grows the bone (Hench and Polak, 2002).

15.8

Self-reinforced composites

A more controlled thin fibre process is the controlled elongation of polypropylene fibre which when incorporated in polypropylene produces a composite with improved stiffness and also excellent impact resistance. Since the fibre and binding polymer are generically the same, the composite is regarded as providing a simple recycling route; the composite can be chopped and remoulded into another product. This technology has considerable scope for new structural composites with other materials being considered such as nylon (Hine and Ward, 2006).

15.9

Nanoparticulate composites

The future of composites cannot ignore the hype that surrounds the introduction of nanoparticulates into composite materials. These particulates are platelet materials which have high aspect ratio of thickness to surface area. The major interest in these nano materials was triggered by the ‘out of proportion’ improvements in stiffness properties when small amounts of particulates were added to a polymer (Fig. 15.5). The materials have in fact been in composite application for many years. Mica is a naturally occurring mineral that is mined and exfoliated to produce the high aspect ratio. It is then incorporated in resin to produce composites for the electrical engineering industry or in the case of the household, mica board on which toaster elements are wound. Other such particulates to be


437

Heat distortion temperature (°C)

120 110 100 90 80 70 60 50 40 0

5

10

15

20 25 30 35 Mica content (%)

40

45

50


15.5 Graph showing improvement in polymer stiffness by addition of Mica.

researched are vermiculite and at one time sea shells were considered for composite manufacture (Li, 2007). In these areas of composites, it is the ceramic nature of the filler that could be of interest to high temperature applications and a silicone mica composite enjoyed excellent durability as a blast furnace liner.

15.10 Hybrid structures It is tempting to suggest that the deliverance of lightweight composites means the replacement of metal. However, the designer’s role is to select ‘the best material combination for the job’. In this respect it is quite often forgotten that moulded components can incorporate metal in the composite solution for a structure. There are many examples of successful hybrid material composites (Fig. 15.6).

15.6 A hybrid composite subframe constructed by Bayer by incorporating a steel pressing in injection moulded plastic.

438


15.11 Conclusions The engineering of green functional structures is the target of today’s engineers and challenges their imagination. This chapter was intended to illustrate that there is much material diversity in the composite concept to provide this design adventure. Thus, the future of composites is intertwined with the inspiration of the design engineer and the utilisation of the vast armoury and combinations of material at their disposal. •

Carbon fibre composites will continue to advance in cost effectiveness and therefore take a larger share of applications. • Natural fibre composites whilst interesting, will have their application restricted to non-structural components. Composites are the most interesting and versatile of all engineering materials and are well placed to become the material of the century.


15.12 References brite-euram project BE-7148, ‘Optical Tracer Technology for Goods and Packaging’, Cranfield University, 2004 carpenter, j.a., daniels, e., sklad, p., warren, c.d. and smith, m., ‘FreedomCAR automotive lightweighting materials’, TMS Annual Meeting, Light Metals Division Luncheon. Walt Disney World, Orlando, 28 February 2007 etches, j.a. and fernando, g.f., ‘Evaluation of embedded optical fiber sensors in composites: EFPI sensor response to fatigue loading’, Society of Plastic Engineer, published online 28 January 2009 eu parliament, The European Parliament’s Committee on Agriculture and Rural Development’s Report ‘The Promotion of Non Food Crops Study – 2005’ eu 6th framework report for EPOBIO Project, Products from Plants – The biorefinery future, Outputs from the EPOBIO Workshop, Wageningen, 22–24 May 2006 gerard, j.h. and jander, m.h., ‘Owens-Corning P4 Technology: The latest on this new process’, American Composite Manufacturing Association, ACMA Technical Resources, SPI archives 1990–99, Technical paper no. 93-9F, 1993 goodship, v. and smith, g.f., ‘Removing the paint shop process: options for painting and decorating injection mouldings’, International Journal of Environmental Technology and Management, 8(4), 339–347, 2008 haque, m.s., zakaria, a., adhir, f.b. and firoza, a., ‘Identification of Micrococcus sp. responsible for the acceleration of jute retting’, Pakistan Journal of Biological Sciences, 6(7), 686–687, 2003 harper, l.t., turner, t.a., warrior, n.a. and rudd, c.d., ‘Characterisation of random carbon fibre composites from a directed fibre preforming process: the effect of tow filamentisation’, Composites Part A, 38(3), 755–770, 2007 hench, l.l. and polak, j.m., ‘Review – third-generation biomedical materials’, Journal Science, 8 February 2002 hine, p.j. and ward, i.m., ‘Hot compaction of woven nylon 6,6 multifilaments’, Journal of Applied Polymer Science, 101(2), 991–997, 2006


439


lovins, a.b. and cramer, d.r., Hypercars, ‘Hydrogen and the Automotive Transition’, International Journal Vehicle Design, 35, Nov 1/2, 2004 ma, c.-c. m. and lin s.-h., US Patent 5741382, Method for preparing oriented discontinuous long fiber reinforced thermoplastic resin composite sheet product, granted April 1998 li, x., ‘Nanoscale structural and mechanical characterization of natural nanocomposites: seashells’, JOM Journal of the Minerals, Metals and Materials Society, 59(3/March), 71–74, 2007 pink, d.a.c. and teakle, g.r., ‘Bringing old genes into modern crops’, The Vegetable Farmer, February, 2005 rudd, c.d., Composites for Automotive Applications, Rapra Technology Limited, 2001 smith, g. goodship, v. and lobjoit, c. ‘Foresight vehicle and the inspire paint shop process 2’, Society of Automotive Engineers (JSAE), Spring Congress, Japan, May 2008 US Patent 5914369, Process for the preparation of polyvinyl alcohol, 22 June, 1999 wright, b., rowse, h. and whipps, j.m., ‘Microbial population dynamics on seeds during drum and steeping priming’, Plant & Soil, 255, 633–642, 2003 yu, m.-f., ‘Strength and breaking mechanism of multiwalled carbon nanotubes under tensile load’, Science 287, 637–640, 2000 zhang, m., ‘Strong, transparent, multifunctional, carbon nanotube sheets’, Science 309(5738), 1215–1219, 2005

16 Designing composite wind turbine blades for disposal, recycling or reuse


N. PA PA DA K I S, Technological Educational Institution (TEI) of Crete, Greece, C. R A M Í R E Z, Centro de Ingeniería Avanzada en Turbomáquinas S.de R.L. de C.V., México and N. R E Y N O L D S, University of Warwick, UK

Abstract: In this chapter, we consider the emerging issue of end-of-life wind turbines. A background on wind energy is presented, with a brief overview of current installed capacity and future market trends, including estimations of production volumes. Materials selection, turbine design and existing manufacturing techniques are covered, along with a discussion of the current research being conducted into new materials and processes that will naturally affect end-of-life considerations. Additionally, a typical wind turbine life cycle is examined. The subject of end-of-life wind turbine blades is approached, both from a current industry perspective and also in the light of research into the challenges and opportunities offered by the inevitable large increase in numbers of end-of-life turbine blades. The following end-of-life scenarios are examined: disposal, reusing and also recycling. Key words: end-of-life, wind turbine blades, wind energy, composite recycling.

16.1

Current wind energy market and trends

The latest available figures for the global wind energy market (GWEC, 2008a) show that the total installed capacity was approximately 93 900 MW. Furthermore, GWEC predict that over the next three years the installed capacity will increase substantially in size from over 100 GW now to 240 GW by 2012. This prediction represents an increase of over 150% from the current capacity, and gives an idea as to the size of the wind energy market as it stands and the potential increases. Figure 16.1 presents the cumulative installed power capacity trend up to 2007 and also three possible future trend scenarios (GWEC, 2008b), dependent on policy and technology adoption. As the wind turbine manufacturing industry has relied heavily on thermosetting composite materials for key turbine components, this burgeoning industry represents a significant challenge for the composite material disposal and recycling infrastructure. 443

444


Cumulative installed capacity (GW)

10,000

1,000

100 Realised Reference scenario

10

Moderate scenario Advanced scenario

1 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 Year

16.1 Cumulative installed global wind power capacity (realised and projections).


16.2

Turbine design and manufacture

The ideal location for a wind farm would be a site that has consistent average strength winds, is easily accessible, is far from human and wildlife habitation and is close to the main power grid. Since finding such an ideal site is not usually possible, different approaches are required in the design and operation of wind turbines.

16.2.1 Current design practice The vast majority of modern wind turbines are based on the Danish threebladed rotor design. This design has been proven over the years as being the most suitable in terms of balance and efficiency. A schematic of a typical wind turbine is shown in Fig. 16.2. The turbine blades are bolted onto a rotor hub (commonly with variable blade pitch control) which in turn is connected to the rotor shaft entering the nacelle. The nacelle enclosure houses the bed plate, generator, gearbox, shaft, control box, and other gearing for the transmission of power from the rotor into the generator. The nacelle/rotor assembly is mounted upon a steel tower that is set into concrete foundations. The rotor/nacelle assembly is attached to the tower via the yaw control mechanism, which allows the rotor/nacelle to be turned to face the prevailing wind direction for optimum energy production. A breakdown of the major component weights for a 1.65 MW machine is given in Fig. 16.3 (Vestas Wind Systems A/S, 2006). Although a few attempts have been made to modify the current design (e.g. with novel swirling profiles for blades or more elaborate arrangements

Designing composite wind turbine blades

445

Rotor blade

Gearbox

Generator

Nacelle enclosure Yaw control


Blade pitch control

Tower

Transformer and grid connection Foundations

16.2 Wind turbine schematic.

of rotors), the overall tendency of industry has focused more on keeping the same three-bladed design whilst increasing the size (diameter) of the turbines. Recent developments in novel materials and improvements in manufacturing processes coupled with the demand for more electricity output per wind turbine have facilitated and driven this trend for increasing rotor size. Moving from the previous classification of wind turbines by relative size, i.e. micro, small, medium and large, has now led to a new classification by rated output: 1.x, 2.x and 3.x MW. Machines of 1–3 MW are currently the

446

Management, recycling and reuse of waste composites Tower 136 Nacelle 51 Rotor 42.2

Foundation 832


16.3 Component weights (metric tonnes) for Vestas V82, 1.65 MW turbine.

‘work horses’ for large-scale power generation and around 100 KW machines are used for ‘small-scale’ wind turbines. There are still companies that supply turbines with a rated output below 1 MW, and typically these smaller turbines are often used for hard-to-access regions and where overall space is constrained. An ongoing process of mergers and acquisitions has also resulted in convergence of design, leading to standardisation of characteristics for similar wind turbine machines. Larger companies such as Gamesa, Vestas and GE (whose turbines account for almost 60% of worldwide installed capacity in 2007) have taken over several smaller wind generator companies in the Western hemisphere. In the Far East there are two major companies, namely Goldwind and Sinovel, and these account for 42% of the China installed capacity, the country with the fastest growth in wind farms installed capacity. The common trade-offs when designing blades relate to the structural strength versus weight, i.e. specific strength. Blade weight should theoretically increase with rotor diameter as a cube of blade length (an exponent of 3). However, in practice, actual weight increases with respect to blade diameter by an exponent of 2.3 to 2.35 (Fig. 16.4). Weight increase has been mitigated by design refinement and by employing manufacturing methods that optimise the composite’s structural properties and the recent adoption of carbon fibre reinforced material systems. Current blade designs tend to have a hollow blade reinforced with spar caps and shear webbing running along the blade as reinforcement (Fig. 16.5). Additionally, a structural core manufactured with sandwich foam composite with balsa wood can be introduced inside the blade in order to provide the required strength and stiffness during operation. On the subject of machine design, it is noteworthy that the most widely accepted guidelines used to assist certification of installations in the wind


447

20,000 18,000 Blade weight (kg)

16,000 14,000 12,000 10,000 8,000 6,000 4,000 2,000 0 0

10

20

30 40 50 Blade length (m)

60

70

16.4 Blade weight with respect to blade length, actual data from commercial wind turbine blades. Upper skin Split line


Shear web Lower skin

16.5 Cross-section diagram of blade with shear web.

energy industry (Germanischer-Lloyd WindEnergie GmbH, 2007) do not currently cover materials selection in terms of design for disposal. As legislation shifts the responsibility of disposal towards the original equipment manufacturer (OEM; i.e. producer pays), materials selection and endof-life strategies will become much more important.

16.2.2 Materials selection The preferred material for blade skin manufacturing is glass fibre reinforced epoxy (Elsam Engineering A/S, 2004; Lenzen and Wachsmannb, 2004; Vestas Wind Systems A/S, 2006). Depending upon the manufacturer, additionally balsa wood, foam, honeycomb can be used internally to provide supporting cores and inner ribs. Epoxy has good resistance to moisture and polluting elements as well as good mechanical properties. Glass fibre as used in different industries (i.e. automotive, marine) is a widely available and relatively cheap commodity material, and is well suited to be used as a reinforcement fibre in blade manufacturing. The fibre–matrix ratio is assumed to be 60% glass fibre (weight fraction) with the remaining 40% epoxy (Pick and Wagner, 1998). When using prepreg materials in the manufacturing process, up to 10% of the prepreg turns into waste due to off-cutting. For a 5.4 tonne blade,

448


this means that nearly 6 tonnes of material is needed. This post-industrial waste material also needs to be dealt with. Recently, filament winding has been proposed as an alternative process, reducing the off-cut waste arising from the use of prepregs. This would mean that the dry reinforcement fibres are supplied, with epoxy resin being only applied during the wrapping process, which results in reduced post-industrial waste material (Vestas Wind Systems A/S, 2006).

16.2.3 Blade manufacturing A typical modern manufacturing process for the manufacturing of wind turbine blades is as follows: • •


•

•

• •

•

Glass fibre is laid up over the open moulds, and a vacuum bag is constructed over the fibres and mould. The selected resin system is infused through the fibres under atmospheric pressure. The blade skins are de-moulded and the spar caps (moulded separately) are fitted along the halves; these are unidirectional glass–epoxy reinforcement pieces. Reinforcement materials (glass and balsa wood) are introduced to strengthen the joint between the spar cap and the glass/epoxy blade skin. A second vacuum bagging process is conducted to ensure all the reinforcements are properly infused with resin. The upper half is placed on top of the lower half, and a shear web (usually sandwich structures of glass/epoxy with foam cores) is glued between the two shells in order to ensure geometric and structural consistency. Surface finishing is performed, as is fitting of bolts for hub connection.

16.2.4 Post-industrial waste The overall waste for manufacturing each blade is generally calculated as 10% of the material by weight (resin, fibres and paste) plus the consumables utilised for the infusion: peel ply, breeding fabric, vacuum bags and vacuum tape. This would be 540 kg for the material and roughly another 120 kg for consumables per blade for a 1.x MW turbine. That adds up to nearly 2 t of non-recyclable wastage per turbine between resin, paste, fibres and consumables. Vestas claims to install one turbine every four hours globally, that is about 180 wind turbines manufactured per month. As Vestas has roughly 30% of the world’s wind turbine market, a simple extrapolation would


449

suggest that globally ∼600 new turbines are being manufactured every month, which would amount to 1200 t of non-recyclable post-industrial waste produced over the same time.


16.2.5 The scale of the end-of-life wind turbine problem Taking 1–3 MW turbines as currently being the most common size of machine, we can make the following estimations using a commercially available 1.65 MW machine specifications (Vestas Wind Systems A/S, 2006). A single 40 m composite blade as used on a 1.65 MW turbine weighs approximately 8.4 t. Roughly two-thirds of this is glass reinforced epoxy composite, and the remaining mass is divided between internal reinforcements, coating, foam and other structural materials. Disregarding these other materials, the total mass of composite material in all three blades would therefore be approximately 16.8 t (Fig. 16.6). The nacelle enclosure (gondola) is also made out of glass/polyester, and accounts for 1.8 t towards the total weight of the nacelle. However, some recent designs such as the new 6 MW Enercon E126 machine have abandoned glass fibre reinforced polyester (GFRP) composite material systems for the nacelle enclosure in favour of aluminium mainly due to its improved recyclability. This is an indication of how recycling issues have already started to affect the wind generator design. Consequently, for a typical composite-intensive 1.65 MW machine, there will be approximately 18.6 t of composite material used.

Blade core materials 8.4

Glass–epoxy composite 16.8

Cast iron 11.3

Steel 5.7

Total rotor mass 42.2 t

16.6 Rotor component weights (metric tonnes) for Vestas V82, 1.65MW turbine.


Global wind energy capacity (GW)

1600 1400

10,000,000 Global wind energy capacity Total mass of installed blades

9,000,000 8,000,000

1200

7,000,000

1000

6,000,000

800

5,000,000

600

4,000,000 3,000,000

400

2,000,000

200 0 2005

Total mass of blades (t)

450

1,000,000 2010

2015

2020 Year

2025

2030

0 2035


16.7 GWEC predicted global cumulative wind energy capacity and calculated cumulative mass of blades.

As previously stated, the growth in the global wind energy installed capacity between 2004 and 2007 was close to 46 000 MW, which amounts to an effective doubling of installed capacity. Making a broad assumption that the new capacity comprises entirely 1.65 MW machines, the total number of new wind turbines installed during that time would be ∼28 000. Assuming a target lifetime of 20 years, this would amount to 520 000 t of composite material to be disposed of in 16–19 years’ time, for recently installed capacity. Adding to this all of the previously installed wind turbines (and those being installed as you read this book), based on the above assumptions, the total amount of waste composite material arising from the wind energy industry over the next 20 years exceeds 1 000 000 t. Predictions upon the future scale of the wind energy market and hence the size of the end-of-life issue are equally far-reaching. Taking the ‘moderate scenario’ GWEC future global cumulative wind energy capacity predictions (GWEC, 2008b) in conjunction with the trend towards larger machines and blade sizes, it is possible to estimate the yearly and cumulative installed mass of composite blades up to 2030 (Fig. 16.7). These calculations are based on linear interpolations between the current (2008) most widely used models with ∼40 m blades (8.6 t), and ∼75 m blades at 20 t being commonly installed in 2020 on 10 MW machines (Andersen et al., 2007). These figures show that if the growth of wind energy capacity grows according to the moderate GWEC predictions, there will be nearly 4 000 000 t of overall blade mass (including core materials etc.) by 2015 and over 9 000 000 t by 2030. Results from the same calculations that reflect the addition per year of overall blade weight (Fig. 16.8) show that in 2008 alone,


451

Annually installed blade mass (t)

400 000

360 000

320 000

280 000

240 000

200 000 2005

2010

2015

2020

2025

2030

Year

16.8 Annually installed blade mass (predicted), metric tonnes.


290 000 t of blades were manufactured and commissioned, rising to 375 000 t per year in 2030. This would indicate that the annual end-of-life wind turbine composites contribution could approach 300 000 t in 20 years’ time (2028).

16.3

Usage

A life-cycle assessment (LCA) of a product or process captures the summation of environmental impacts that accumulate over the entire cradleto-grave period of the LCA subject (Lenzen and Wachsmannb, 2004). These impacts will be incurred directly during the manufacturing of the product, during the process/use of product, and at the end-of-life (disposal), or are caused indirectly during the provision of inputs (e.g. raw materials) into the manufacture of a product, or execution of a process. Obviously the areas under consideration here are environmental impacts in the form of post-industrial and end-of-life waste material, and the useful lifespan against which these impacts can be offset.

16.3.1 Typical lifespan Germanischer Lloyd WindEnergie GmbH (2007) states that the design of a wind turbine shall be such as to provide an operational lifetime of at least 20 years. This guideline is applied regardless of the type/quality of wind and environmental conditions surrounding the turbine installation. Moreover, all major wind turbine manufacturers comply with these guidelines. Consequently, all components and systems that make up the wind turbine are designed to meet this criterion with the minimum of intervention such

452


Wind turbine component manufacture

Transport to site, assembly and erection

Operation (20 years, incl. maintenance)

Decommission, removal and disposal

16.9 Typical life cycle of a wind turbine.

as planned and unplanned maintenance (which of course results in system downtime). This includes all of the composite components that have direct contact with the environment such as the blades, spinner (rotor hub cover) and the gondola (composite outer box) of the nacelle. This 20 year lifespan considers only the time the wind turbine is in operation (see Fig. 16.9).


16.3.2 Premature failure There are several reasons as to why a wind turbine blade can fail in service. Failures in first generation wind turbines were commonly caused by overspeed of the rotor. In such over-speed cases, the rotor spins at velocities higher than stipulated in the original design, and the blades are subjected to excessive drag forces. In extreme cases, the blades can then eventually become detached from the rotor. Blades on modern wind turbines fail primarily through accumulation of fatigue damage or other internal damage within the structural composite. Fatigue analysis for composites tends to be complex and unreliable since the overall in-service strains which are utilised for calculating the life cycle tend to be fairly small, resulting in a cycle lifespan of infinite cycles at best. Furthermore, every company has its own material configuration where different kinds of inserts and reinforcing material architectures are included in the blade which makes predictability of the fatigue life far more complicated. Most of the time, finite element analyses are conducted whereby the materials and configurations are modelled as closely as possible to the real world. Nevertheless, for each new location, wind and environmental conditions will change and introduce several new factors that will affect the life of the blade. Inspection of the blades at blade manufacturing facilities is commonly done visually and using ultrasonic flaw detection. The approach is to avoid and detect defects in the composite such as voids, dry patches, overlaying of fibres and delamination, which would lead to increased stress levels and perhaps crack initiation. Remedying built-in problems on a fully cured 40 m long composite blade can be extremely complicated and time consuming, since an entire section of the material has to be removed, and local infusions need to be performed in order to assure the complete impregnation of the fibres. The industry tendency is to aim to reduce the


453

possibility of these defects occurring in the first instance by conducting very stringent quality checks when initially laying-up the fibre pre-forms and infusing the blade with resin. Nevertheless, sometimes small defects can be overlooked and become potential failure initiation zones once the blade is in service. Wind turbines can never be 100% failure free given the unpredictability of wind behaviour; however, the application of predictive performance analysis technology and repeated revisions (empirical) of design have pushed their reliability continuously up towards acceptable levels. The industry considers it to be worthwhile to perform stringent quality assurance throughout the manufacturing of the blades to assure the 20 year life cycle is met and that wind turbine technology is safe, economical and perceived to be trustworthy.

16.4

End of life


16.4.1 Case studies review Several life cycle environmental cost case studies have been conducted. In Elsam Engineering A/S (2004) alternative potential disposal methods for the blades are investigated: • 100% to landfill; • combustion of blades without any recycling of the constituent materials; • 90% recycling of constituents with remaining 10% to landfill. For other plastic/rubber components the preferred scenario was 100% incineration of waste. In the study, environmental impact was measured using the Danish LCA tool UMIP (Wenzel et al., 1996). The UMIP tool normalises the environmental impacts in order that they can be stated in milli-person equivalent units (mPE). The results calculated for each disposal scenario reflect what 1 kW h of power produced from the wind farm adds to an average citizen’s total contribution to a range of parameters such as global warming, hazardous materials and bulk waste. It was estimated that the varying disposal scenarios had broadly similar environmental effects. However, the environmental impact ‘bulk waste’ was significantly higher in the case of 100% disposal, followed next by incineration. Incineration is expected to produce approximately 30% of the net calorific value of the materials utilised. Finally, 90% recycling of the blade, while decreasing the effective bulk waste to less than one-third of the disposal scenario, almost tripled the human toxicity (which however still remained negligible) due to the chemicals used to efficiently recycle the material.

454


Most case studies found were conducted prior to 2005. At this time, the EU legislative framework had not been finalised. Therefore, the focus was on predicting the environmental impact of pre-2005 mainstream wind turbines at their time of decommissioning 20 years into the future, with the currently available disposal options. Viable recycling solutions were not available then, and in 2009 there are still no commercially available technologies. The case studies agree that the only main variation in environmental impact between land-filling and other recycling-oriented technologies is the quantity of bulk waste.

16.4.2 Current industry practice


Disposal Previously it was economical to dispose of waste composite material from turbine blades via landfill, but in Europe this is no longer practical since the application of EU legislation 2005 (EU Council Directive 99/31/EC). Furthermore, blades were delivered to a waste disposal site simply because no alternative disposal methods (e.g. recycling) were available. Development of suitable methods for the recycling of turbine blades is ongoing, and with the advent of legislation, the issue is now considered to present a significant environmental problem by those within the dismantling and recovery industry (Andersen et al., 2007). Incineration It is suggested that the only other alternative to disposal for large composite components such as turbine blades is incineration (Properzi and Hank-Hansen, 2002). Specifically, incineration with energy recovery from materials with energy content may be a viable future disposal option in many industrialised countries. The concern is that this method is not useful for structural composites such as wind turbine blades because of the high content of inorganic material. Vestas has completed a pilot project proving the feasibility of energy and heat recovery from scrap blades with an industrial collaborator in the waste management sector (Vestas Wind Systems A/S, 2006). In many countries only electricity is generated, with widely differing efficiencies. Efficiency of electricity production of 30% has been assumed (Vestas Wind Systems A/S, 2006), without co-generation of heat; however, this could be as low as 10%. It is acknowledged that this approach can only provide a very crude estimate of the consequences, and that more consideration should be given to the issue. More specifically, the technical problems caused by the presence of glass fibres in the flue gas which damages the gas cleaning system needs to be addressed, as do the large amounts of


455

fly ashes which need to be disposed of – the combustion of composite blades results in a large amount of bulk waste, as the reinforcement fibres in the blades are inflammable and consequently remain as a residue from the combustion process. However, intensive study is underway to enable the recovery of the glass content from the blades (e.g. for use as insulation material). In 20 years’ time it can be assumed that this technology or a similar improved technology could be widely used (Vestas Wind Systems A/S, 2006). One additional factor that will greatly affect this issue is the fact that only waste with calorific values above the threshold of 11 MJ/kg can be traded on the international markets as waste for utilisation – this would obviously create a market for such material, whereas waste falling below this threshold is simply classified as waste for disposal, and therefore has to be disposed of according to national legislation. Moreover, waste for disposal can only be disposed of nationally (Brahms et al., 2006). The calorific value of most unfilled resins is approximately 30 MJ/kg, falling to 15 MJ/kg for 50% volume fraction glass (Pickering and Benson, 1991).


Reuse Wind turbines can often be made to function for somewhat longer than their 20 year design lifetime by a major overhaul, e.g. by replacing the gear box or other major components like the generator. In addition, there is a market for used turbines in developing countries. Recycle One possible technique for the recovery and recycling of the reinforcing glass and carbon fibres that has been developed and already successfully applied on an experimental scale is the fluidised bed process (Pickering et al., 2000 and Chapter 4). The reuse of composite re-grind material in the automotive industry (Marsh, 2003) has also been investigated. One of the most important factors that affect the economic viability of this method is the mechanical properties obtained when recyclate is added to new composites. When the recyclate is used as replacement for virgin filler and fibre reinforcement in new composite, a general trend of decreasing strength with increasing recyclate content occurs. This decrease in mechanical properties is attributed to the poor bonding between the recyclate and the new matrix. The triturate of fibre reinforced polyester (FRP) may also be used for incineration and the remaining waste for cement production. FRP materials made of unsaturated polyester resins (60% by weight) and glass fibres (40% by weight) have exothermic heat content of 4500 kcal/kg as fuel. Also glass fibres are composed of Si, Ca and Al which are also main components of cement – along with Fe and Mg (Namaguda et al., 2001).

456


Companies such as ERCOM in Germany, Mecelec in France, RJ Marshall and MCR in the US, and Miljotek in Sweden have proven the feasibility of mechanical recycling on an industrial scale, but typically for waste sheet moulding compounds/bulk moulding compounds (SMC/BMC materials). The application of this method to structural laminated composites such as found in wind turbine blades has yet to be attempted. Other technically feasible but not yet economically viable recycling technologies are composite pyrolysis and hydrolysis. Pyrolysis involves the application of heat (over 400 °C) to composite material within an inert atmosphere which causes the thermal breakdown of the polymer matrix, producing useful gases and liquids; this method can also potentially reclaim fibres and fillers. Hydrolysis is the chemical breakdown of matrix materials. It is polymer specific and has seen limited use in composites.


16.5

Conclusions

Composite waste arising from end-of-life wind turbines constitutes a serious environmental and recycling problem. The extremely large expected volume of thermosetting composite waste will present a significant challenge to the recycling industry with the current infrastructure that is in place. The waste issue is further compounded by the lack of alternatives to the high performance thermosetting composite materials currently used for wind energy turbine applications. Traditional engineering materials do not have the specific (density-normalised) properties required, and newer materials such as natural-fibre reinforced composites have similar problems. A review of the literature and industry publications reveals that as landfilling is no longer an economically feasible option for the volumes under consideration, reuse through incineration with some form of energy recovery will be the most likely near- to medium-term solution. Additionally, the concept of recovery of fibre reinforcement via thermal pyrolysis of the polymer resin has been investigated; this could also be combined with energy recovery; however, further studies need to be done in this area. A scarcity in the literature of wind turbine life-cycle case studies and industrial trials demonstrates that much research needs to be done to address this waste problem.

16.6

References

andersen p., borup m. and krogh t., ‘Managing long-term environmental aspects of wind turbines: a prospective case study’, Int. J. Technology, Policy and Management, Vol. 7, No. 4, 2007, Interscience



457

brahms t., kühne u., albers h. and gceiner s., ‘Feasibility study for the recycling of composite material (MAVEFA)’, DEWEK 2006 (8th German Wind Energy Conference),Bremen,(http://www.faserverbund-verwertung.de/material/DEWEKProceeding.pdf) elsam engineering a/s, ‘Life cycle assessment of offshore and onshore sited wind farms’, Online literature, 2004 (http://www.vestas.com/Admin/Public/ DWSDownload.aspx?File=Files%2fFiler%2fEN%2fSustainability%2fLCA%2f LCA_V80_2004_uk.pdf) eu council directive 1999/31/EC of 26 April 1999 on the Landfill of Waste germanischer lloyd windenergie gmbh, ‘Guideline for the Certification of Wind Turbines’, Germany, 2003 – updated 2007 (https://www.gl-group.com/wind_ guidelines/wind_guidelines.php?lang=en) gwec, ‘Global Wind 2007 Report’, May 2008a, Global Wind Energy Council (http:// www.gwec.net) gwec, ‘Global Wind Energy Outlook 2008’, Oct 2008b, Global Wind Energy Council (http://www.gwec.net) lenzen m. and wachsmannb u., ‘Wind turbines in Brazil and Germany: an example of geographical variability in life-cycle assessment’, Appl Energy, Vol. 77 119– 130, 2004 marsh g., ‘Europe gets tough on end-of-life composites’, Reinforced Plastics, Sept 2003 namaguda k., hayashi s. and abe y., ‘A solution for composite recycling – cement process, Composites 2001 Convention and Trade Show, Composites Fabricators Association, 3–6 October/2001 pick e. and wagner h.-j., Beitrag zum kumulierten Energieaufwand ausgewä hlter Windenergiekonverter. Arbeitsbericht. des Instituts für Ökologisch verträgliche Energiewirtschaft, Universität Essen, Germany; 1998 pickering s.-j. and benson m., ‘The recycling of thermosetting plastics’, 2nd International Conference on Plastics Recycling, Plastics and Rubber Institute, 1991 pickering s.-j., yip h., kennerley j.-r., kelly r.m. and rudd c.d., ‘The recycling of carbon fibre composites using a fluidised bed process’, Proceedings of 8th International Conference on Fibre Reinforced Composites – FRC2000, Newcastle upon Tyne, 2000 properzi s. and herk-hansen h., ‘Life cycle assessment of a 150 MW offshore wind turbine farm at Nysted/Roedsand, Denmark’, Int. J. Environment and Sustainable Development, Vol. 1, No. 2, 2002 vestas wind systems a/s, ‘Life cycle assessment of electricity produced from onshore sited wind power plants based on Vestas V82-1.65 MW turbines’, online literature, 2006 (http://www.vestas.com/Admin/Public/DWSDownload.aspx? File=Files%2fFiler%2fEN%2fSustainability%2fLCA%2fLCAV82165MWonsh ore.pdf) wenzel h. hauschild m. and rasmussen e., Environmental Design of Industrial Products, UMIP, Institute for Assessment of Products, Technical University of Denmark, 1996

17 In-process composite recycling in the aerospace industry K. P O T T E R and C. WA R D, University of Bristol, UK

Abstract: This chapter will present a case study of recycling advanced composites in the aerospace industry. It aims to discuss advanced composites consumption in the industry and to highlight waste generation through analysis into the design and manufacturing processes in use. The chapter will highlight the recycling techniques available and, looking to those scrap types currently created, reviews how to capture and reuse the scrap, to potentially create a closed loop process. Finally the chapter will discuss some future trends in aerospace composites manufacturing and how these trends may have an impact on future waste generation.


Key words: composites, manufacture, design, rosette datum, reuse, recycling, automation.

17.1

Introduction

This chapter will present a case study of recycling advanced composites in the aerospace industry. The chapter aims to review the industry, to discuss advanced composites consumption, and to highlight waste generation. To understand how scrap is generated, design and manufacture processes will be reviewed, examining how decisions lead to scrap but also how improvements can lead to a reduction. The chapter will also look at the scrap types currently created, reviewing how to capture and reuse the scrap to potentially create a closed loop process. Finally the chapter will discuss some future trends in aerospace manufacturing using advanced composites in automation and how these trends could have an impact on future waste generation. Overall, the chapter aims to develop an understanding of design and manufacture, to show that although scrap is generated in the manufacturing process the level of scrap generation is strongly influenced by the decisions taken in design. Note that when discussing advanced composites, this chapter refers to materials created by combining high strength and stiffness reinforcing fibres with a compatible resin system. Advanced composites applicable to the aerospace industry usually consist of fibres of high strength and/or high modulus embedded within a homogeneous matrix of either thermoplastic or thermosetting resins. Modern fibres are typically 458

In-process composite recycling in the aerospace industry

459

carbon, although glass and aramid are also used, and the predominant resin type used is epoxy resins (thermoset). Thermoplastic resins are used but the volume in use is very limited at present.


17.2

Composite consumption in the aerospace industry

The first applications of advanced composite materials for the aerospace industry can be found around the 1950s and were almost exclusively intended for military use.1 This situation began to change in the early 1970s when contract awards in the US targeted their use in civil aircraft,2,3 and early investigations subsequently led to the first use of advanced composites in civil aircraft by Boeing,3 then expanded by Airbus who drove the use of the materials forward (fairings, radomes, fin leading/trailing edges etc). Helped by material cost reductions, Airbus became the leader in composites use4 with the airframe materials share rising from 500 tons 544 tons 600 tons 672 tonnes 725 tons 740 tonnes 851 tonnes

23 m

8.3 tonnes {13 tons} 41 tonnes 41 tonnes 45 tons 170 tonnes

B&Q Castorama Trimaran ocean-racing yacht Maltese Falcon super-yacht {free-standing masts} RNLI Severn class lifeboat Cable & Wireless Adventure powerboat US Navy M80 Stiletto Australian Navy ‘Bay’ class catamaran inshore minehunters Swedish Navy Landsort class MCMV HMS Wilton prototype MCMV Sandown/Racecourse class Single Role Mine Hunter (SRMH) Christensen 186 Belgian/French and Dutch navies ‘Tripartite’ class Minehunters/Sweepers Swedish Navy Visby stealth corvette Italian Navy Gaeta class minehunters/sweepers Hunt class Mine Counter Measures Vessels (MCMV) Mirabella V super-yacht United States Navy MHC-51 class coastal minesweepers

Length

Displacement

Vessel group

Table 18.1 Some characteristics of large marine composite structures


10.6 m

10.4 m 9.6 m

8.9 m

9.6 m 8.5 m

5.9 m 14 m 12.1 m (39.7 ft) 9m

Width

6.1 m

30.6 m mast {57 m mast}

Height


498


boatbuilding federation FIN and the main shareholder of Groupe Beneteau) has stated that the industry has worked hard on techniques to destroy old fibreglass hulls, instead of having owners abandon them. ‘But so far we are having difficulty finding any: they are in good condition and sailors continue to use them,’ she said. Stevenson (undated) surveyed boat yards and marinas in the Southampton area to ascertain ‘How often do you deal with abandoned boats?’ The most common reply to the question was ‘Not very often’ with the maximum reported number being 25 per year. To the best of our knowledge there is no in-depth study yet reported and hence there is scope for further research to establish more accurate quantities of end-of-life hulls arising annually in the UK. Since the Finnish marine industry association, Finnboat, started promoting the responsible disposal of end-of-life boats in 2005, approximately 500 have been collected and recycled. The boats handled reflect the overall composition of the Finnish boat market by type with the majority being less than 6 m in length. Of the 500 boats recycled to date, approximately 150 were of glass reinforced plastic (GRP) construction with the majority of the balance being acrylonitrile–butadiene-styrene (ABS), then a small number of metal boats. On average the weight of the boats disposed of has been about 200 kg and fewer than 10 have been more than 1 tonne. In terms of length, the largest boat destroyed was a GRP clad wooden fishing boat. No sailing boats have yet been recycled. Composite waste, including marine composites, does not constitute a category of its own in the annual waste statistics. Production waste arisings would be included in Industrial and Commercial Waste data. The categories where it is likely to fall are ‘Manufacture of motor vehicles, and other transport and equipment’ in Industrial Waste, and ‘Miscellaneous’ under Commercial Waste. For the year 2002, statistics published by the Department for Environment, Food and Rural Affairs (Defra, 2006b, Table 5) the totals of these two categories reported were 1475 thousand tonnes and 1554 thousand tonnes, respectively, from a total of 67 907 thousand tonnes. In other words, production waste in the marine sector is probably less than 4% of all commercial and industrial waste. End-of-life marine composites are most probably included in the municipal waste (not household) making up an unknown but potentially small proportion of the 5 million tonnes reported for 2002. Adding all these together, marine composites contributed less than 2.4% to the 335 million tonnes total waste arisings in 2002 (Defra, 2006b, Table 1). The problem of disposal of waste marine composites appears to be relatively minor at present. However, if the marine leisure industry continues to grow at the current 8% per year (British Marine Federation, 2008), then, as new designs are produced and as the existing fleet ages, there will be

Disposal of composite boats and other marine composites

499

more end-of-life materials produced, both from worn-out vessels and obsolete mouldings. It is therefore important that economical methods of treatment are developed now.


18.3

The design phase

Searle and Summerscales (1998) have reviewed the durability of fibrereinforced composites in the marine environment. The key considerations for the achievement of long life are (a) to avoid water soluble components in the laminate which can lead to osmosis and blistering, and (b) for carbon fibre reinforced composites, to avoid direct contact with light alloys in order to avoid the formation of a galvanic corrosion cell. Nevertheless, GRP is well known for its long life, which makes it still an attractive material for new designs. Landamore et al. (2007a) reviewed the environmental impact of three different materials that might form the hull of a sustainable boat designed for use on the Norfolk Broads. These were wood–epoxy, GRP and steel. Their analysis was based on a 30 year life expectancy, and took account of the production, operation and disposal phases. They concluded that while a wood–epoxy hull would have the least environmental impact it would be more costly than GRP, and steel was too harmful in the production phase to be competitive. Furthermore, the boat-owners themselves ranked wood and GRP highly, and the durability of GRP hulls meant that it was more economical to maintain their old boats than to invest in new models (Landamore et al., 2005). It was concluded that GRP could be made more sustainable if low-styrene polyester was used as the matrix, if resin-infusion production methods were adopted, and if the hulls were recycled at the end of their useful life, rather than land-filled (Landamore et al., 2007b). Thus GRP proved to be a sensible choice for the hull of a boat designed with the effect on the environment in mind. Abdi et al. (2004) suggest that risk management is required at all stages of the product life from concept to disposal. Composite structures must be addressed not only on a component basis but also as part of the overall system. For example, the disposal of a product when it has fulfilled its intended purpose may be governed by regulations. The use of some composites could result in the product being treated as hazardous waste (increasing disposal cost) or the substitution of an environmentally friendly composite for a metal component that requires a hazardous protective coating could reduce disposal cost.

Such areas of the product life cycle can benefit from a judicious combination of actual and virtual testing for risk management. Virtual testing software has been reported to contribute to risk reduction goals and could

500


reduce development and process costs (30–50%), time to market (10–25%) and product quality testing (30%). Macdonough and Braungart (2002) have proposed that all components should be designed for dis-assembly. One-design racing sailboats, where all competitors are expected to use identical boats, stifles innovation and hence these vessels tend to have a finite period of use before the sailors move to a new boat design, and the entire fleet rapidly becomes obsolete. Sectors such as the marine sports industry tend to be very ‘fashion’ oriented with new designs introduced in each new season for surfboards and especially wind-surfers. In consequence, there are probably many composite components lying unused in attics and garages, or simply sent to civic amenity sites. As noted above, no data are yet available for the quantities of such materials in existence.


18.4

The manufacture and marketing phase

The choice of manufacturing process can have a considerable effect on materials usage. For example, a woven cloth reinforced composite may have a maximum fibre volume fraction of 35% (contact moulding, including hand lamination or spray processes), 55% with one atmosphere of consolidation (out-of-autoclave (OOA) processes, including vacuum bagging and resin infusion), or 65% for higher pressure processes (autoclave consolidation or compression moulding). Assuming that the composite has a negligible void content, then the minimum resin volume fraction will be 65% for contact moulding, 45% for OOA processes or 35% with consolidation. However, the excess resin manifests itself as increased panel thickness and hence as greater panel stiffness. The linear change in panel weight with resin volume fraction has to be balanced against the thickness-cubed dependence of panel flexural stiffness. The reduced panel stiffness may be compensated by moving to a sandwich structure, but this does often introduce a third material into the system. The amount of consumable material varies with the chosen manufacturing method. An important consideration in respect of waste from composite vessels is the elimination of unnecessary materials usage at the manufacturing stage. Tucker (2004) has reported that, according to Fox (2002), during 3 ‘low volume yacht manufacture, about 64 m of waste are produced for every tonne of finished boat’ and that in surfboard manufacture ‘about one-third of the raw materials used to make a board end up on the workshop floor as in-process scrap (Henty, 2002)’.

18.5

The use phase

It is unlikely all components of an entire vessel will fail simultaneously, and inevitably some items will need to be replaced during service, with the


501

failed component entering the waste stream. Landamore et al. (2007a) concluded that composite hulls (like the other materials, wood and steel) had very little environmental impact during the use phase. So over the life of a vessel it is expected that only small amounts of composite waste will be produced.

18.6

End of life

18.6.1 Legislation


The European Union aims to become the most sustainable community in the world. This has led to Directives aimed at minimising the environmental impact of discarded products. The key directives (see also Chapters 2 and 3) are: • Waste Electrical and Electronic Equipment (WEEE) Directive (Commission Directive 2002/96/EC); • End-of-Life Vehicle (ELV) Directive for the automotive sector (Commission Directive 2000/53/EC). In the USA, the Aircraft Fleet Recycling Association (AFRA) (2008) is focused on the safe and economical return of aircraft to revenue service, of engines and parts to the world fleet and of reclaimed materials (composites, aluminium, electronics, etc) back into commercial manufacturing. In the UK, concerns over the proposed deep-sea disposal in 1995 of the Brent Spar oil-storage platform and over proposals in 2003 by Able UK to import decommissioned US vessels for disposal led to government action. Defra published a consultation paper (Defra, 2006a) which aimed to develop a strategic approach to the recycling of UK-flagged vessels consistent with national and international sustainable development. Defra has also published a guidance document (Defra, 2007) but the strategy only applies to vessels of 500 gross tonnes and to commercial flagged vessels, which are generally constructed from steel. The strategy does not apply to recreational craft so, at present, there is no legislation that applies to the disposal of the major proportion of marine composites. Ships generally reach their end of life because their operation is no longer profitable and there is no buyer on the second-hand market. They are generally sold for dismantling to extract the steel (and some equipment) for recycling. This operation is labour-intensive and has moved from European/OECD countries (includes Turkey) via China to South Asia (Bangladesh, India and Pakistan). Labourers in ship dismantling yards may earn US$250/day in the Netherlands, US$13/day in Bulgaria or US$1–2/day in Bangladesh and India (EC Commission, 2007). These low labour cost


502


economies also tend to have less than optimal environmental and healthand-safety requirements. Facilities for dismantling large ships in Europe are concentrated in Belgium, Italy and the Netherlands with a combined capacity of ∼230 kLDT per annum. The price paid for a vessel may be >US$400/light displacement ton (LDT – ‘roughly equivalent to the steel weight of the ship’; Defra, 2006a) in Bangladesh where the scrap material supplies 80–90% of the national need. Chinese yards offer about half the previous figure and US operators offer about one-tenth (European Commission, 2007). End-of-life ships destined for dismantling are considered waste under international (and Community) law, especially where they contain substantial quantities of hazardous substances (which may include remnants of cargo, oils and oil sludge, asbestos, heavy metals in paints and polychlorinated biphenyls (PCBs)). In the European Union, export of such vessels to a non-OECD country for dismantling is therefore prohibited (European Commission, 2006) until they have been decontaminated or processed so that they are no longer considered hazardous waste. The Basel Convention (United Nations Environmental Programme, 1989) requires prior authorisation from the destination country before waste can be moved. In 1995, the Basel Convention was amended (but not adopted by all countries) to ban exports of hazardous waste from OECD countries to non-OECD countries. The ultimate aim of the EU is to ensure that minimum environmental and health and safety standards are observed worldwide (not to maintain this business in the Community). The European Union commissioned a Green Paper (Europa, 2007) to open consultation on ship dismantling and identify the most appropriate routes for environmentally and socially sustainable dismantling. The Green Paper limits comment on composites to ‘composites which are very difficult to separate and recycle’ (European Commission, 2007) and ‘some materials are hard to recycle (composite materials)’ (Europa, 2007). The International Maritime Organisation (IMO) Marine Environment Protection Committee (MEPC) prepared a draft Convention on the Safe and Environmentally Sound Recycling of Ships (IMO, 2008a) which was approved at the London MEPC session in 2008 and adopted by IMO at the Hong Kong Convention in 2009. The Convention will not apply to vessels of less than ∼400–500 gross tons, nor to warships, naval auxiliary or other vessels owned or operated by or on behalf of each government of the signatory countries. Consequently, there are almost no composite vessels currently afloat to which it will apply. The IMO have published a bibliography associated with this initiative (IMO, 2008b). The Recreational Craft Directive (Commission Directive 94/25/EC) and the subsequent amendment (Commission Directive 2003/44/EC) do not


503

consider end-of-life vessels. Some national representative bodies have given extensive consideration to disposal routes. The French in particular have commissioned a number of studies (FIN, 2006; Wittamore, 2007a) finally concluding that the best disposal route is by means of existing infrastructure. A similar conclusion has been drawn by the Finnish marine industry federation (Wittamore, 2007b) and this overall route has also been endorsed by the International Council of Marine Industry Associations (ICOMIA) (Amble, 2007).


18.6.2 Ownership of abandoned vessels A key issue in deciding whether a vessel becomes waste is the cost of renovation of older craft relative to the costs of acquiring a new vessel. For the owner of a vessel, the costs of storage can quickly exceed the resale value (negative equity). Abandoned vessels (an example is shown in Fig. 18.1) not only take up space in the marina but also reduce the capacity of the marina to generate revenue (Fig. 18.2). Further, they often do not have a significant scrap value especially if the hull is wooden (Figs 18.3 and 18.4). This figure also illustrates the social consequences of dumping vessels on the shoreline – loosely translated, the graffiti says that the beach is not a dustbin. The local authority then has to deal with the ‘waste’. Hulls abandoned below the high water mark become the responsibility of the local harbour master (Stevenson, undated) who generally only disposes of them if they become a safety hazard. Figures 18.5 and 18.6 illustrate just how much of a hazard abandoned vessels can be, with the local authority issuing stern warnings against boarding derelict vessels.

18.1 Abandoned GRP vessel (photograph by Ken Wittamore).

504



18.2 An abandoned vessel occupying a valuable mooring (photograph by Ken Wittamore).

18.3 Wooden-hulled boats left to rot on the French coast (photograph by Ken Wittamore).

In countries with boat registration, establishing ownership prior to disposal should be straightforward provided that (a) the owners actually register them, (b) the records are maintained and (c) the last owner has not deliberately removed the registration details from the boat. In the Oakland Tribune, Rosynsky (2006) makes it clear that, when faced with disposal costs greater than the value of the vessel, owners will go to considerable lengths to hide their identity, leaving the costs of disposal with the state. He estimates that between 1998 and 2006 the US Department of Boating and Waterways spent $2.8 m in disposing of 427 abandoned boats.


505


18.4 Public reaction to a vessel abandoned on a French beach (photograph by Ken Wittamore).

18.5 A warning of the safety hazard caused by a derelict steel ship (photograph by Ken Wittamore).

In Finland, any boat that is believed to have been abandoned and is of unknown ownership is first advertised in the local press. If no owner is forthcoming, the boat is auctioned or sent for disposal.

18.6.3 End-of-life strategies Conroy et al. (2004, 2006) and Halliwell (2006) have reviewed the endof-life options for composites waste using the waste hierarchy: Waste reduction > reuse > recovery > disposal

506


18.6 Boarding an end-of-life wooden-hulled boat is prohibited (photographs by Ken Wittamore).


Rathje and Murphy (1992) have divided recycling into four categories: • primary: reprocessing waste to obtain product comparable to the original version; • secondary: recovery of waste material with lower performance when compared to virgin materials; • tertiary: decomposition of materials to recover monomers, feedstock materials or fuels; • quaternary: recovery of the embedded energy in the materials. Reuse as a vessel At the end of its useful life a vessel may be assigned to a lower duty cycle or enter the second-hand market for whole vessels. As long as a vessel remains seaworthy it may have multiple owners. For example, HMS Wilton (M1116) was built to the same basic design as the Coniston-class minehunter (often simply referred to as the Ton class minesweeper) using glass reinforced polyester resin composite in place of the normal wood construction for these vessels. She was built at Vosper Thornycroft in Southampton, launched in February 1970 and commissioned in July 1973. At the time of her construction she was the largest plastic ship in the world (and probably the largest composite structure) at 450 tons displacement and was unofficially known as HMS Tupperware or HMS Indestructible. The design and use of this vessel informed the design of the subsequent Hunt-class MCMVs and the Racecourse-/Sandown-class Single Role Mine Hunters. In 1991, as the sole remaining Ton-class vessel, her weapons systems became obsolete. She was modified and adopted as a navigation training ship at Dartmouth


507

Britannia Royal Naval College. She was retired from Royal Navy service in July 1994 and paid off at Portsmouth. She was then bought by a private individual, moved to Southampton for repair then to Lowestoft. The Essex Yacht Club acquired her in 2001 for conversion into a floating Headquarters Ship and she has been based at Leigh-on-Sea since 2004. Reuse of components


Standardised components enable a market for second-hand parts to be established. There is scope for smaller composite components such as hatch-covers or rudders to be limited in their range of designs making them available for reuse at the end of a vessel’s life. Reuse of the materials Steve Pickering (2006) has reviewed the technologies for recycling thermoset composite materials with particular emphasis on mechanical and thermal processing. It may be difficult to recycle thermosetting composites, but there is greater potential for the reuse of the materials in a thermoplastic composite. In both cases it is impractical to de-ply laminated composites to reuse the layers (given the current state-of-the-art). It is thus necessary to reduce the waste composite to usable dimensions which may involve sawing, crushing to produce parts of perhaps a few square metres followed by grinding or hammer-milling to increase the density of the waste for transport to recycling facilities. All of these options require significant energy inputs and consideration should be given to the balance of benefits gained against gross environmental impact. A related issue is the material toughness: by design, the inclusion of reinforcement fibres significantly increases the toughness of the material. That toughness will result in significant energy consumption during comminution of the structure. Further, in general the thermoplastics have higher toughness than the thermosetting resins and hence will require yet more energy at this stage. The machinery required is relatively expensive and hence the process may not be economically viable on a small scale. For further information on machining of composites, see the review papers by Abrate and Walton (1992a,b) and Gordon and Hillery (2003). (See also Chapter 9 on mechanical methods.) For thermoplastic composites, there is the possibility of granulating the material for processing by extrusion and/or injection moulding. Both of these processes will cause attrition of the fibre length and hence the material will need to be used at lower duty (i.e. reduced stresses). Thermoplastic hulls (albeit un-reinforced) have been produced by rotational moulding of thermoplastics at LDC Racing Sailboats (http:// www.rssailing.com/), Performance Sailcraft (Dart/Laser) and Topper


508


International (http://www.toppersailboats.com/), and by unspecified methods at Bic Sport Boats (http://www.bicsportboats.com/). The Building Research Establishment (BRE) has carried out development and evaluation trials for products incorporating ground GRP. GRP/ plastic lumber has similar properties (density, elastic modulus and modulus of rupture) to some other wood plastic composites and is more durable in the marine environment than natural timber. It is claimed to offer ‘an alternative to tropical hardwoods or treated softwood for some types of lightly-loaded marine piles such as groynes, fender boards, light bridge foundations, jetties, boardwalk posts and similar applications’ (Conroy et al., 2006). A specific issue for disposal of composite structures used in the marine industry is size (the overall dimensions – not surface treatments on the fibres). A small boat is typically of similar dimensions to a large car. This introduces a range of problems specific to this and related (e.g. wind turbine blade) sectors including transport and size reduction to permit feeding to crushing equipment. In Finland, recreational boats at the end of their useful life are recycled using commercial car crushing plants having first been broken up using mechanical methods (Fig. 18.7). A particular issue for the recycling company in this case is the contamination of the recovered composite materials. By the end of its life the exterior of a vessel will have received many coats of paint of various types, including anti-fouling. Internally, the composite is likely to be contaminated with oils, paints and embedded core materials such as cork and ‘Nomex’. The experience of the Finns indicates that composite cladding of old hulls creates recycling difficulties and as a result of this and the relatively small volumes involved, all composite materials go to landfill.

18.7 Kuusokoski scrap plant where end-of-life vessels are crushed (photograph by Ken Wittamore).


509

Recovery of fibres and feedstock materials


It is possible to recover the reinforcement fibres and degradation products of the polymer matrix system. In practice, this will be more practicable where a known consistent source of material is available (i.e. from manufacturing, but not post-use, waste) rather than a mixture of different resin systems as optimal processing is then possible. The options for fibre and feedstock recovery include the following: • Incineration (see Chapter 4): this destroys the resin, but at 450–500 °C usable clean glass fibres remain, while at 450–550 °C usable clean carbon fibres can be recovered albeit with a reduction in the fibre mechanical properties. The lower temperatures are used for polyester resins and the higher temperatures for epoxy matrix systems. Milled Carbon (Birmingham UK), Karborek (Puglia, Italy) and ENEA (Ente per le Nuove tecnologie, l’Energia e l’Ambiente – the Italian national agency for new technologies, energy and the environment) are to build a composite recycling facility in Puglia to process an average of 1000 tonnes (1102 tons) of composite scrap annually (NetComposites, 2008). • Pyrolysis (see Chapter 5): heated to temperatures of typically 400–600 °C in an oxygen-free atmosphere. • Catalytic transformation (see Chapter 6). • Acid digestion. • Solvolytic/solvothermal processes: these include hydrolysis and glycolysis. • Sub-, near- and super-critical fluids: this normally includes water (at 300–500 °C) or carbon dioxide. Piñero-Hernanz et al. (2008) used a batch-reactor in the temperature range 250–400 °C with pressures from 4 to 27 MPa and residence times up to 30 minutes. Iwaya et al. (2008) have depolymerised glass fibre/polyester composites to separate the fibre, filler and polymer using sub-critical diethyleneglycol monomethylether (DGMM) or benzyl alcohol (BZA) in a batch reactor at 190–350 °C for 1–8 hours. George and Carberry (2007) have identified potential investment areas for the optimisation of carbon fibre recycling (Table 18.2) which will be equally applicable to other fibre reinforced polymer matrix composites.

Composting Natural-fibre and bio-based resins may find use in dry applications for composite materials and structures, but their use in moist/wet environments raises issues with respect to durability. Hence, it is unlikely (in the short term at least) that this disposal route will be appropriate for waste composites from marine applications.

510


Table 18.2 Potential investment areas for optimizing carbon fibre recycling (George and Carberry, 2007) Material tracking and identification Product specific dissection maps (end of airframe life) Joint de-integration (pre- and post-dissection) Pre-CF recovery contamination removal Optimized pre-CF recovery material size Reclaim resin chemical value Post-CF recovery contamination removal (possibly integrated with other fibre value added processes) Fibre length adjustment and sorting Fibre de-bulking for transport Aligned fibre material forms


Incineration (preferably with energy recover energy) Considerable energy is used in the production of polymers (embodied energy of plastics in general is given as 90 MJ/kg; Lawson, 1996), but as in many other systems that energy is not lost and can be recovered at a later stage. Halliwell (2006) quotes a figure of 36 MJ/kg as the energy value for ground composite. During recovery of the energy content of the materials, it will be necessary to comply with the Waste Incineration Directive (WID, agreed by the European Parliament and the Council of the European Union on 4 December 2000). The Commission Directive 2000/76/EC aims to ‘prevent or limit, as far as practicable, negative effects on the environment, in particular pollution by emissions into air, soil, surface and groundwater, and the resulting risks to human health, from the incineration and co-incineration of waste’. It sets and seeks to maintain stringent operational conditions and emission limit values for (co-)incineration plants throughout the European Community (Defra, 2008). Scuttle (sink deliberately) Climate change and sea level rise pose significant challenges. A redundant vessel might then serve as a substrate for coastal defences or for the growth of coral reefs or as a feature for divers. However, disposal of marine composite wastes at sea would require rigorously controlled conditions. In each of the above cases: • • •

considerable attention should be paid to anchoring the feature; an appropriate life-cycle assessment should justify the post-use management of the structure; and care must be taken to offer this methodology as creating a positive benefit rather than simply justifying disposal to the marine environment.



511

In 2002, the European Union (EU) adopted a recommendation on implementing Integrated Coastal Zone Management (ICZM) which prompted improved integration of marine spatial planning. The UK Government passed the Planning and Compulsory Purchase Act in May 2004 with sustainable development as an explicit objective at Section 39(1). The Act requires the preparation of Regional Spatial Strategies (RSS) or a Spatial Development Strategy (SDS for London) and Local Development Documents (LDD) to include Integrated Coastal Management (ICM) where appropriate. The UK has made a commitment to establish a network of marine protected areas (MPAs) to conserve marine ecosystems and marine biodiversity and in turn to respond to international commitments and European obligations. The 1992 OSPAR (Oslo–Paris) Convention is the current instrument guiding international cooperation on the protection of the marine environment of the North-East Atlantic: it combines and updates the 1972 Oslo Convention on dumping waste at sea and the 1974 Paris Convention on land-based sources of marine pollution. As a signatory of OSPAR, the UK is committed to establishing an ecologically coherent network of well-managed MPAs and under the Habitats Directive (Commission Directive 92/43/EEC) there is a requirement to establish and maintain a network of Natura 2000 protective areas. (Natura 2000 is the Europe-wide network of sites tasked with the preservation of natural heritage as a testament to the importance that EU citizens attach to biodiversity.) The responsible bodies for nature conservation and shoreline management are the Environment Agency (EA) and the local authorities (LA) (being the Coast Protection Authorities (CPA) and Maritime District Councils (MDC)) respectively although they have no regulatory powers. The statutory control of marine works in the UK, formerly the responsibility of the Marine Consents and Environment Unit (MCEU within Defra), has been integrated into the Marine Fisheries Agency (MFA) since April 2007. The merger means that MFA now has service delivery functions covering the control of coastal and marine developments (including coast defences, wind farms, wave and tidal power; the disposal of marine dredgings at sea; contingency planning for oil spills and other marine pollution, and marine aggregate extraction). The first generation of Shoreline Management Plans (SMPs – high level documents that form an important element of the strategy for flood and coastal erosion risk management) which cover the 6000 kilometres of coast in England and Wales are now in place. All second generation SMP2s should be completed by March 2010. In April 2008, the UK Government published a draft Marine Bill (HM Government, 2008) for consultation. It intends to set up a new Marine Management Organisation (MMO) as a centre of marine expertise to


512


provide a consistent and unified approach, deliver improved coordination of information and data and reduce administrative burdens. The Bill will enable implementation of the EU Marine Strategy Directive in a coherent and systematic way and support delivery of UK commitments under the EU Water Framework Directive (Commission Directive 2000/60/EC) and EU Habitats Directive (Commission Directive 92/43/EEC), amongst others. Further, the Marine Bill makes a commitment to make progress towards a network of Marine Conservation Zones (MCZs) for the conservation and promotion of the recovery of a wider range of habitats and species. A New Zealand-based marine consultancy (ASRL) claims to lead the world in coastal protection, artificial surfing reefs (ASR), inland surfing pools and numerical modelling (ASRL, 2008). ASRL was honoured with the ANZ Bank ‘Waikato Export Innovator of the Year’ Award for the design and construction of ‘multi-purpose artificial reefs’ – coastal structures which reduce beach erosion while at the same time enhancing the quality of breaking waves and creating a new recreational facility in the form of a surfing break. The creation of a reef using a (normally steel) frigate is a well-proven system with vessels placed on the seabed in Canada, Australia and New Zealand. In the UK, the National Marine Aquarium (NMA) selected HMS Scylla, a steel Leander Class Frigate, as an artificial reef which would be colonised by anemones, sea squirts and other marine life, so that fish and other mobile animals would be attracted to the developing reef. Scylla is intended to be ‘an exciting destination for divers’ (Leece, 2006). Experience in the United States (GASMFC, 2004) suggests that GRP hulls for artificial reefs sunk to depths of more than 250 feet (75 m) are likely to remain there, and that fish will inhabit them for at least 30–40 years. However, hulls sunk to 100 feet (35 m) or less are likely to be disturbed by storms, even when segments are cabled together. This may result in debris floating to the surface or being washed ashore. Furthermore, it was observed that marine life did not establish themselves within 3 years on the fibreglass surface. It will be important to bear in mind the precautionary principle should such a route be selected for the disposal of large composite marine structures (for example naval mine counter measures vessels). The primary concern will be to consider appropriate routes for the removal and disposal of the toxic compounds (especially cuprous oxide and tri-butyl tin) found in commercial anti-fouling coatings. However, where the reused composites may be subject to abrasion and scour by sand and/or pebbles, they will be degraded. There is growing concern about polymer microparticles entering the marine environment and subsequently marine animals (Thompson et al., 2004, 2005, 2009; Browne et al., 2007). It has been shown (Teuten et al., 2007) that tiny plastic particles may act as agents to carry hydrophobic


513

contaminants such as phenanthrene from the surface of the sea to the sediment where they may be ingested by animals such as lugworms that form part of the food chain. Moreover, there is evidence that microparticles of plastic ingested by mussels can pass into their circulatory system rather than simply being excreted although their toxicological effects are yet to be established (Browne et al., 2008).


Landfill as a last resort The Landfill Directive (Commission Directive 1999/31/EC; COM, 2005) provides operational and technical requirements for waste and landfills. These measures seek to prevent or reduce negative effects on the local environment (especially the pollution of surface water, groundwater, soil and air) and on the global environment (including the greenhouse effect and risk to human health) during the whole life cycle of the landfill. The Directive defines four different categories of waste (i) municipal waste, (ii) hazardous waste, (iii) non-hazardous waste and (iv) inert waste. According to Halliwell (2006) ‘composite waste is currently classed as non-hazardous under the banner heading of “Biodegradable wastes and other non-special waste which can give rise to organic or other contamination” according to the UK Waste Classification Scheme’. Figure 18.8 illustrates typical composites waste derived from crushing GRP end-of-life vessels and bound for landfill. The Environmental Code of Practice ‘A Green Blue Initiative’ supported by the British Marine Federation, the RYA and the Environment Agency (ECOP, 2008) recommends that boat and marina owners adopt the waste

18.8 Fragments of scrap GRP retrieved from a crushing plant (photograph by Ken Wittamore).

514


hierarchy noted above: reduce, reuse, recycle (materials), recover (energy), landfill or incinerate. They state that when disposing of end-of-life hulls, scuttling or incineration should be avoided at all costs, and that if incineration is carried out, then it must be done in a controlled fashion and with permission of the relevant authorities. Landfill is the least favourable option, and with the rates of landfill tax increasing by £3 per tonne for the past three years, and by £8 per tonne in 2008 compared with 2007 to the current rate of £32 per tonne (HM Revenue and Customs, 2008) it is becoming ever more expensive. As technologies advance it may become more advantageous economically to recycle.


18.7

Vive la différence?

In respect of current practice, the disposal of end-of-life hulls is proving to be controversial (Wittamore, 2007a). The French have invested in a number of theoretical studies whilst the Finns just send them to the local car crushing plant – the French are saying that this is not the most environmentally responsible method of disposal whilst the Finns shrug and say it is available now and immensely practical. In the UK the position declared by ICOMIA is the formal disposal route for recreational boat hulls although the industry awaits clear evidence that there is a problem to be tackled.

18.8

Conclusions

This chapter has considered the disposal of marine composites which may arise from marine sports equipment, boats and ships, submarines, marine renewable energy systems or offshore oil exploration and exploitation industries. The total quantity of marine composites waste arising in the UK is comparatively small – too small to warrant a separate category of its own. The guiding principle for waste disposal is to follow the waste hierarchy: reduce > reuse > recover (materials and energy) > landfill (or scuttle). Consideration has been given to the avoidance of waste by appropriate design, manufacturing, marketing and maintenance through life. Vessels and their composite components can be salvaged for reuse and the secondhand market for boats of all sizes is flourishing. A key issue is the need to establish the ownership of abandoned vessels, but the number of these found each year in the UK has yet to be fully documented. A way forward is likely to be through registration of a wider range of vessel types: while there is discussion on legislation on the recycling of ships, the disposal of small craft is currently unregulated. Recovery of the energy embedded in polymer composite matrix materials by incineration is still a field for further development. The costs of separation and cleaning end-of-life composite components are still prohibitive


515

on the small scale demanded by the rate of waste arisings. Hence, at present most scrap marine composites end up in landfill. However, as landfill costs soar and as more GRP boat hulls reach the end of their useful lives, alternative disposal routes will become more attractive.

18.9

Acknowledgements

The authors would like to acknowledge the input of the 2007/08 graduates on the BSc (Honours) Marine Sports Technology degree: they were asked to consider the problem addressed by this chapter and some of their ideas have been adapted in this text. JS would like to acknowledge discussions with Professor Dominic Reeve and Dr Dave Simmonds of the UoP Coastal Engineering Research Group and Dr Gillian Glegg (Senior Lecturer in Marine Management) in the UoP School of Earth, Ocean & Environmental Sciences in respect of artificial reefs and related topics.


18.10 References abdi f, castillo t and shroyer e (2004), ‘Risk management of composite structure’, in Nikolaidis E, Ghiocel D M and Singhal S, Engineering Design Reliability Handbook, CRC Press, Boca Raton. abrate s and walton d a (1992a), ‘Machining of composite materials. Part 1: traditional methods’, Composites Manufacturing, 3(2), 75–83. abrate s and walton d (1992b), ‘Machining of composite materials. Part 2: non-traditional methods’, Composites Manufacturing, 3(2), 85–94. afra aircraft fleet recycling association (2008), http://www.afraassociation.org/ accessed 5 April 2008. amble e (2007), Decommissioning of End-of-Life Boats – a status report (second edition), International Council of Marine Industry Associations, December 2007. http://www.icomia.com/library/library.asp?Page=3&view=&LC_ID=&FT_ ID=&CT_ID=&sort=# accessed 31 July 2008. american composites manufacturing association (2007), ‘The 2006 Composites Industry Report’, accessed from http://www.acmanet.org/professionals/2006_ composites_industry_report.pdf. asrl (2008), http://www.asrltd.co.nz/index.htm, accessed 10 April 2008. babu m s, baksi s, srikanth g and biswas s (2008), Composites for Offshore Applications, TIFAC website, accessed 8 April 2008. british marine federation (2008), http://www.britishmarine.co.uk, accessed June 2008. browne m a, galloway t and thompson r (2007), ‘Microplastic – an emerging contaminant of potential concern’, Integrated Environmental Assessment and Management, 3, 559–566. browne m a, dissanayake a, galloway t s, lowe d m and thompson r c (2008), ‘Ingested microscopic plastic translocates to the circulatory system of the mussel, Mytilus edulis (L.)’, Environmental Science and Technology, 42, 5026–5031.


516


bültjer u (2007), ‘Production of fibre reinforced plastics in Europe 2006/2007’, accessed from http://www.eucia.org/uploads/f3863fc3fb3c6145f556927075c0b474. pdf. com (2005), Report from the Commission to the Council and the European Parliament on the national strategies for the reduction of biodegradable waste going to landfills pursuant to article 5(1) of Directive 1999/31/EC on the landfill of waste, COM (2005) 105 final, Brussels, 30 March 2005. commission directive 1999/31/EC of 26 April 1999 on the Landfill of Waste, http:// eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:31999L0031:EN:H TML. commission directive 2000/53/EC of the European Parliament and of the Council of 18 September 2000 on End-of-Life Vehicles – Commission Statements, http:// eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:32000L0053: EN:HTML accessed at 5 April 2008. commission directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 on establishing a framework for Community action in the field of water policy, Original Journal of the European Communities L327/1. commission directive 2000/76/EC of the European Parliament and of the Council of 4 December 2000 on the incineration of waste, available at http://www.wbcsd. ch/web/projects/cement/tf2/2000-76_en.pdf. commission directive 2002/96/EC, The Waste Electrical and Electronic Equipment (WEEE) Directive, http://www.conformance.co.uk/directives/weee.php accessed 5 April 2008. commission directive 2003/44/EC of the European Parliament and of the Council of 16 June 2003 amending Directive 94/25/EC on the approximation of the laws, regulations and administrative provisions of the member states relating to recreational craft, available from http://eur-lex.europa.eu/LexUriServ/LexUriServ.do? uri=OJ:L:2003:214:0018:0035:EN:PDF accessed 8 April 2008. commission directive 92/43/EEC on the Conservation of natural habitats and of wild fauna and flora, available from http://eur-lex.europa.eu/LexUriServ/ LexUriServ.do?uri=CELEX:31992L0043:EN:HTML. commission directive 94/25/EC on the approximation of the laws, regulations and administrative provisions of the Member States relating to recreational craft, available from http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX: 31994L0025:EN:HTML accessed 8 April 2008. conroy a, halliwell s, reynolds t and waterman a (2004), Recycling Fibre Reinforced Polymers in Construction: a guide to best practicable environmental option, BRE Report BR467, 2004. conroy a, halliwell s and reynolds t (2006), ‘Composite recycling in the construction industry’, Composites Part A: Applied Science and Manufacturing, 37(8), 1216–1222, doi: 10.1016/j.compositesa.2005.05.031. defra (2006a), ‘UK Ship Recycling Strategy – Consultation paper’, London, available from http://www.basel.int/ships/docs/16e.pdf. defra (2006b), ‘e-Digest of environmental statistics’, http://www.defra.gov.uk/ environment/statistics/index.htm, accessed June 2008. defra (2007), ‘Overview of ship recycling in the UK – guidance’, London, available from http://www.defra.gov.uk/environment/waste/strategy/pdf/shiprecycle-strategyoverview.pdf.



517

defra (2008), Environmental Permitting Guidance – The Directive on the Incineration of Waste, Department for Environment, Food and Rural Affairs, London, accessible from http://www.defra.gov.uk/environment/epp/documents/ wid-guidance.pdf, accessed 21 August 2008 (NB: includes the full text of the Waste Incineration Directive at Annex 2). ecop (2008), Environmental Code of Practice (A Green Blue initiative): http://www. ecop.org.uk/ accessed 4 July 2008. europa (2007), ‘Ship dismantling’, http://europa.eu/scadplus/leg/en/lvb/l28192.htm, accessed 5 April 2008. european commission (2006), Regulation (EC) No 1013/2006 of the European Parliament and of the Council on shipments of waste, Official Journal of the European Union, L190/1, available from http://eur-lex.europa.eu/LexUriServ/ LexUriServ.do?uri=OJ:L:2006:190:0001:0098:EN:PDF. european commission (2007), Green Paper: On better ship dismantling, COM (2007)269 final, Brussels, available from http://eur-lex.europa.eu/LexUriServ/ LexUriServ.do?uri=COM:2007:0269:FIN:EN:PDF. fin (2006), Federation des Industries Nautique, Synthese Livre Blanc, January. fox a (ecocats) (2002), Personal communication to Nick Tucker in ‘Clean production’, in Baillie C, Green Composites: Polymer composites and the environment, Woodhead Publishing, Cambridge, Chapter 10. gasmfc (2004), ‘Guidelines for marine artificial reef materials’, compiled by the Artificial Reef Subcommittees of the Gulf and Atlantic States Marine Fisheries Commissions, USA, Report No. 121, 131–134. george p e and carberry w l (2007), ‘Recycling carbon fiber: maximizing fiber value for sustainability and profitability’, NetComposites Conference: Composites Innovation, Barcelona, 4–5 October. gordon s and hillery m t (2003), ‘A review of the machining of composite materials’, Proc IMechE Part L: Journal of Materials Design and Applications, L217(1), 35–45. halliwell s (2006), ‘End of life options for composite waste – recycle, reuse or dispose’, National Composites Network report [NB: it is necessary to register with NCN to access this report online]. henty r (md henty surfboards) (2002), Personal communication to Nick Tucker cited in ‘Clean production’, in Baillie C, Green Composites: Polymer composites and the environment, Woodhead Publishing, Cambridge, Chapter 10. hm government (2008), Draft Marine Bill presented to Parliament by the Secretary of State for Environment, Food and Rural Affairs by Command of Her Majesty, Cm 7351, http://www.official-documents.gov.uk/document/cm73/7351/7351.pdf. hm revenue and customs (2008), ‘Landfill tax’, available from http://www. uktradeinfo.com/index.cfm?task=bulllandfill. imo (2008a), ‘Recycling of ships’, http://www.imo.org/Newsroom/mainframe. asp?topic_id=818 accessed 8 April 2008. imo (2008b), ‘Information resources on recycling of ships [information sheet No. 38]’, available from http://www.imo.org/includes/blastDataOnly.asp/data_id% 3D21407/RecyclingofShips_28March2008_.pdf, accessed 8 April 2008. iwaya t, tokuno s, sasaki m, goto m and shibata k (2008), ‘Recycling of fiber reinforced plastics using depolymerization by solvothermal reaction with catalyst’, Journal of Materials Science, 43(7), 2452–2456.


518


landamore m j, birmingham r w, downie m j and wright p n h (2005), ‘Ecoboat – boats for a sustainable future on the Norfolk and Suffolk Broads’, School of Marine Science and Technology, University of Newcastle upon Tyne, available from www.angliaboatbuilders.org.uk/finrep/report.pdf. landamore m j, birmingham r w and downie m j (2007a), ‘Establishing the economic and environmental life-cycle costs of marine systems: A case study from the recreational craft sector’, Marine Technology, 44(2), 106–117. landamore m j, birmingham r w, downie m j and wright p n h (2007b), ‘Sustainable technologies for inland leisure craft’, J. Engineering for the Maritime Environment, Proc IMechE, 221, 97–114, doi: 10.1243/14750902JEME76. lawson b (1996), Building Materials, Energy and the Environment: Towards ecologically sustainable development, RAIA, Canberra, as echoed in Technical manual: design for lifestyle and the future, http://www.greenhouse.gov.au/ yourhome/technical/fs31.htm accessed 16 December 2006. leece m (2006), ‘Sinking a frigate’, Ingenia, 29, 27–32. lm glasfiber (2008), www.lmglasfiber.com, accessed August 2008. mcdonough w and braungart m (2002), ‘Cradle To Cradle: Remaking the way we make things’, New York, North Point Press. netcomposites (2008), ‘Boeing and Alenia to Support Italy’s First Composite Industrial Recycling Plant, 25 July 2008’, http://www.netcomposites.com/news.asp?5111, accessed 31 July 2008. pickering s j (2006), ‘Recycling technologies for thermoset composite materials – current status’, Composites Part A: Applied Science and Manufacturing, 37(8), 1206–1215. piñero-hernanz r, dodds c, hyde j, garcía-serna j, poliakoff m, lester e, josé cocero m, kingman s, pickering s and wong k h (2008), ‘Chemical recycling of carbon fibre reinforced composites in nearcritical and supercritical water’, Composites Part A: Applied Science and Manufacturing, 39(3), 454– 461. rathje w and murphy c (1992), Rubbish! The archaeology of garbage, London, Harper-Collins. rosynsky p t (2006), Ditched craft eyesores in estuary, Oakland Tribune, 14 January, http://findarticles.com/p/articles/mi_qn4176/is_20060114/ai_n16009284, accessed 31 July 2008. roux a (2007), ‘Paris show focuses on environmental protection’, http:// www.ScuttlebuttEurope.com/scuttlebutt-europe-1376-4-December.html accessed on 29 September 2009. searle t j and summerscales j (1998), ‘Review of the durability of marine laminates, in G Pritchard (editor): Reinforced Plastics Durability, Woodhead Publishing, Cambridge, pp 219–266. stevenson k (undated) ‘End of life boat hulls – the current situation and disposal options’, http://thegreenblue.org.uk/research/documents/EndofLifeBoatHulls. pdf, accessed June 2008. teuten e l, rowland s j, galloway t s and thompson r c (2007), ‘Potential for plastics to transport hydrophobic contaminants’, Environmental Science and Technology, 41, 7759–7764. thompson r c, olsen y, mitchell r p, davis a, rowland s j, john a w g, mcgonigle d and russell a e (2004), ‘Lost at sea: where does all the plastic go?’, Science, 304, 838.


519


thompson r, moore c, andrady a, gregory m, takada h and weisberg s (2005), ‘New directions in plastic debris’ (letter), Science, 310, 1117. thompson r c, moore c j, vom saal f s and swan s (editors) (2009), Plastics, the environment and human health, Philosophical Transactions of the Royal Society, 364, 1526. tucker n (2004), ‘Clean production’, in Baillie C, Green Composites: Polymer composites and the environment, Woodhead Publishing, Cambridge, Chapter 10. wittamore k (2007a), A brief report on the current state of the art in end of life boat disposal, published by the European Confederation of Nautical Industries. wittamore k (2007b), End of life boat disposal in Finland, published by the European Confederation of Nautical Industries. yacht forums (2008), http://www.yachtforums.com/special-features/394-yachtingstatistics.html.

19 Sustainable fibre-reinforced polymer composites in construction M. FA N, Brunel University, UK

Abstract: This chapter firstly deals with applications for fibre reinforced polymeric (FRP) composites in construction and allied industries; it contains descriptions of their structures, properties and performance. The chapter then comments on the long-term durability, maintenance demands, associated waste management criteria and sustainable development of FRP composites for construction by repeated referral to their material properties. This exposition is based on well-understood environmental science as it applies to construction.


Key words: FRP composite, building and bridge construction, long-term performance, waste and reuse, sustainable construction composite.

19.1

Basic concept and history

Composite materials combine and maintain two or more discrete phases, each having its own physical and mechanical characteristics. Once formed by the process of combination, as optimised by best practice, the resulting composites have properties which can be markedly superior to their constituent parts. The range of these materials is very diverse. Composites performing a structural function in buildings or bridges consist of five main divisions, each having the various constituent combinations (e.g. particle reinforced, fibre reinforced and consolidated composites) (Fig. 19.1). The first of these five divisions defined as being suitable for construction is that of natural composite materials. Wood is a low density, cellular, polymeric composite. The most successful model used to interpret the ultrastructure of timber/composite ascribes the role of ‘fibre’ to the cellulosic microfibrils, while the lignin and hemicelluloses are considered as separate components of the ‘matrix’ (Fig. 19.2). There is ample record of the ubiquity of wood as a construction medium. Five thousand years ago, the ancient Egyptians were using it to build boats, to make furniture and coffins and to sculpt statuary. Sophisticated carpentry techniques developed independently from c. ad1100 in England, mainland Europe, China, India and Japan. Each culture produced timber-framed structures that were architecturally magnificent as well as being strong and durable. In Europe, the cities of Venice, Amsterdam and old Berlin are founded on wooden 520

Sustainable fibre-reinforced polymer composites in construction

521

Composites in construction Particle reinforced composites Large particle

Fibre reinforced composites

Consolidated composites

Dispersion Continued Discontinuous Laminates strengthened Randomly Aligned oriented

Natural Cementitious composites composites

Metallic composites

Sandwich panels

Polymeric Nanocomposites composites

19.1 A classification of composites in construction.

Outer layer (S1) Primary wall

Noncrystalline region

Microfibril 10–30 nm

12 nm

Middle layer (S2)

25–30 nm


Inner layer (S3)

Crystalline region (crystallites) 30–60 nm

Cellulose elementary fibril Middle lamella

Hemicellulose Lignin

19.2 Nature composite, showing natural and matrix (i.e. lignin and hemicelluloses) and reinforcement (cellulose elementary microfibrils).

piles, some of which are ancient. Mills and waterwheels relied on wooden machinery. Until very recently, European theatres had elaborate and ingenious mechanised wooden stages (reference the drawings of the late Richard Leacroft). Use of wood continued into the industrial era. Rails of early pit railways, railway sleepers, gantries, carts and carriages were all made in timber. The twentieth century saw teak framed and panelled railway carriages, sewn plywood (Consuta) launches sailing on the Thames, engineered plywood gondolas for airships, plywood aircraft (the Mosquito)


522


and cars with wooden frames and aluminium panelling (the AC sports saloon of 1949). Despite some brilliantly inventive exceptions, timber has tended to be usurped for structural and engineering applications by new strong materials. Nevertheless, in defiance of competition from lightweight metals and plastics, whether foamed or reinforced, timber remains the world’s most successful natural fibre composite by virtue of its excellent strength-to-weight, durability (subject to correct specification and detail design), ease of formation into complex shapes, ease of jointing, low cost and sustainability. Synthetic composites reinforced with natural fibres appeared after the Great War. Unfortunately, the availability by 1932 of cheap mild steel and, later, cheap aluminium, put a stop to their being refined and developed into a marketable range of structural products. However, in recent years, there has been a surge of interest in these potentially versatile, carbon-capturing and sustainable materials. This has led to a number of new innovations which could succeed in bringing several enhanced natural fibre composites to the building materials market (Fig. 19.3). The second of these divisions is that of cementitious composite. Although fibre reinforced cementitious composites (FRC) have been exploited for almost five decades, the idea of using natural fibre to reinforce a weak binder (like straw in mud or lime plaster) can be traced back as far as 3500 bc. Advanced FRC composites are typically composed of less than 2 vol% of fibres (e.g. metal, ceramic or polymer), which impart most of the tensile strength and toughness to the composite, and a matrix phase, which

19.3 Various natural fibre composites used in construction.



523

not only retains the fibres, but also transfers loads to them (they have stiffness as well as tensile strength) and protects them from exposure to harsh environments. The third division, one which is soon likely to have uses in construction, is that of nanofibre composites. Nanotechnology is the term used for the manipulation of materials measuring 100 nm or less (i.e. the size of a virus) in at least one dimension. Nanotechnology is considered to be one of the most important technological innovations of this century (some are citing it as a second industrial revolution). Research and development (R&D) in nanotechnology are critically important to the new generation of processes and products. New nanomaterials with unique, ‘bespoke’ properties can now be developed with the aid of nanotechnology. It could enable new techniques of polymerisation to be perfected, which would entail synthesis of fibres augmented either with water or with other organic liquids, with the twin aims of enhancing composite performance and facilitating endof-life recycling. The fourth division is that of metallic composite. Steel has been used as a construction material since the late nineteenth century. However, the evolution of structural engineering and fabrication practice for steel used in construction raises many salient issues, namely the provision of fire protection, consideration of buckling or fatigue risks, improving corrosion resistance and ensuring adequate ductility for building in locations prone to violent seismic activity. In consideration of these and other design criteria, many metallic composites have been developed in response to these recurring problems (e.g. steel alloys, steel reinforced concrete). The most recent is the metal matrix composite (MMC). MMC is made by dispersing a reinforcing material (e.g. carbon fibre) into a metal matrix (e.g. aluminium). In comparison with polymer matrix composites, MMCs are resistant to fire and radiation. They absorb no water and are unaffected by it. They have greater electrical and thermal conductivity which could be a disadvantage. Another disadvantage is that MMCs tend to be more expensive. The fifth division of these construction composites is that of the polymeric composite. Fibre reinforced plastic (FRP) materials (which should be named as polymer composites), consist of strong fibres (e.g. carbon, aramid, basalt and glass fibres) and a polymer matrix (Fig. 19.4). Modern polymers began to develop from the 1930s when fossil-fuel oil became the main source of the organic chemicals from which synthetic plastics, fibres, rubbers and adhesives are made. It is only over the last two decades that advanced composite materials have emerged as a suitable alternative both for building new structures and for the repair of defective existing ones, such as railway and road bridges. The properties of polymer matrix materials can be improved in their structural properties by a factor of up to a hundred by the addition of fibres. Certain combinations of fibres and matrix

524


Chopped mat

0° mat

45° mat

90° mat

Bl-directional mat 0°/90°


19.4 Formulation of FRP composite (left: possible fibre orientations, right: lamination).

could bring into existence engineering materials that exploit the intrinsic strength of the fibres but which also possess great toughness, since the fibres inhibit crack propagation through the resin matrix. As the content of this chapter must be confined within the scope of this book, the text below will solely discuss and review FRP composites for construction. Innovations in civil engineering can encompass original structural design or make use of new materials. In the past, building construction has pioneered materials for their strength, stiffness, workability (for ease of construction and maintenance), versatility and aesthetic properties. Specifiers now require construction materials to be environmentally friendly, competitive in terms of initial cost, superior in terms of life cycle cost, durable, fire resistant and dimensionally stable. FRP composites can now meet most of these specification criteria better than conventional strong materials. Civil engineers, highway officials and private construction industries are fast becoming aware of the benefits that flow from use of FRPs in mainstream building. However, the FRP composites cannot be fully integrated into modern construction until more durable and reliable fire resistant resins and protection systems have been proven by testing and appear on the market. What is more, the cost of FRPs must be brought down by better exploitation of mass production manufacturing if the transition from conventional materials to FRPs is to continue. Pultrusion is precisely such a method. Continuous pultrusion of linear sections achieves far better cost competitiveness for FRPs compared with labour-intensive hand lay-up techniques.


19.2

525

Polymer composites in building construction

The construction industry is one of the world’s largest consumers of polymeric composites, with un-reinforced polymer composites being used as non-load bearing components (e.g. trimmings, kitchenware and cladding) (Head, 1998; Sandeep et al., 2008). The low density of FRP composites offers the engineer or designer the opportunity to exploit a high strength to weight ratio. Other advantages are: resistance to fatigue and corrosion, ability to provide a customised surface finish, high prefabrication profile, design stability, non-magnetic characteristics, radar transparency and the capability to integrate functions. Although these many benefits have awakened the interest of a few architects and engineers to the potential of FRPs to create startlingly original buildings and other structures, the primary structural applications of FRP composites in construction still remains relatively low (Westaway, 2004; Kendall, 2007). A summary of the application of composites in building construction is illustrated in Fig. 19.5.


19.2.1 Rehabilitation One of the most highly developed applications of FRP composites in the building industry is the rehabilitation of deteriorating or damaged structural elements. Defects in need of remedy could be due to environmental exposure, inadequate design or poor quality construction. Alternatively, a need could arise to update a structure to new codes. The term ‘rehabilitation’ would include repair, strengthening and retrofitting. Often, these three remedial categories are used interchangeably but, in fact, they refer to three different structural requirements, each presenting different functional demands. Each requires a different materials and process specification. Repair serves to fix a structural deficiency to restore components to their original level of performance. Strengthening is intended to enhance the Composites in building

Rehabilitation Strengthening Retrofit

Modular construction Structural building construction Repair

Full building system Structural

Composite shell

Non-structural Semi-structural

Composites in building

19.5 Summary of composites in building construction.

526


existing as-designed structural capability to provide additional performance. Retrofitting is specifically used to improve the ability of a structure to absorb seismic energy. Rehabilitation of existing structural elements is highly efficient in terms of time needing to be spent and labour resources needing to be deployed; consequently, there are considerable cost benefits accruing from use of FRPs in these three remedial categories. These structural interventions can be undertaken with the aid of a number of different materials: epoxy- and vinyl ester-impregnated glass, aramid or carbon fibres. To make the right choice, the specifier must refer to design guidelines which are available in developed countries to provide adequate information for confident utilisation of composites. For the purposes of this section, the discussion will be restricted to strengthening of reinforced concrete (RC) beams and columns with FRP. There are three generic classifications: firstly, adhesive bonding of prefabricated elements; secondly, wet-lay-up of fabric (wrapping) and, finally, resin infusion (Fig. 19.6).


Advanced composite strengthened RC beams FRP composites can be used to strengthen RC beams. Their use for this purpose is gradually usurping the former, traditional techniques of flitching with steel plates or corseting with steel jackets (steel is susceptible to corrosion which often causes serious deterioration of bonds at the steel– concrete interface; what is more, addition of steel increases both the selfweight and the overall cross-sectional dimensions of the structures to which it is attached). FRP composite plates are usually bonded to the tension side of RC beams with epoxy resin in order to increase the flexural strength of

Wrapping of fabric

Winding of tow

Bonding of prefabricated shells

19.6 Rehabilitation processes.

Resin infusion

Use of composite cables/strips


527


19.7 Strengthening of concrete beam.

the beams. The fibres in the composite are placed parallel to the principal stress direction, which is normally perpendicular to the cracks. Shear strengthening of RC beams is achieved by bonding composite reinforcement with fibres parallel to the direction of the shear stresses (Fig. 19.7). The efficacy of the strengthening depends on the property and geometry of reinforcement (i.e. FRP composites), configuration of lamination and loading (i.e. the number of and wrapping schemes of layers, and the percentage of reinforcement, shear span to depth). The flexural strength of the strengthened RC could be increased by as much as 40% with glass fibre reinforced polymer (GFRP) composite and 200% with carbon fibre reinforced polymer (CFRP) composite reinforcement. Shear strength of RC beams may be increased by 60–120% depending on the materials specified and the architecture of the reinforcing fibres (Heﬀernan and Erki, 1996; Bonacci and Maalej, 2000, 2001; Karbhari and Zhao, 1998; Shahawy et al., 2001; Shin and Lee, 2003; Brena and Macri, 2004). The fatigue life of RC beams could also be significantly extended by carefully configured use of externally bonded CFRP composite laminate (Barnes and Mays, 1999; Erki and Meier, 1999; Shahawy and Beitelman, 1999; Masoud et al., 2001; Heﬀernan and Erki, 2004). However, the strengthened RC can exhibit greater deflection under load owing to the lower modulus of elasticity of FRP composites as compared with steel reinforcement. A number of techniques for improvement of ductility have been studied, including an anchorage system and the inclusion of innovative tri-axially braided ductile fabric (White et al., 2001; Grace et al., 2002; Salom et al., 2004). Complex interactions between concrete and steel and reinforcements create major obstacles to the development of analytical formulae to predict the behaviour of strengthened beams and columns. A number of analytical models have been proposed, such as the simplified linear elastic models (strain compatibility) and non-linear finite element models for the flexural behaviour and interfacial failure analysis of strengthened RC

528


beams (Ziraba et al., 1994; Vijay and GangaRao, 2001; Teng et al., 2003). The ultimate shear capacity of strengthened beams has been calculated by assuming linear elastic behaviour of reinforcement materials (Gendron et al., 1999; Ibell and Burgoyne, 1999; Triantafillou and Antonopoulos, 2000) or by compression field theory through assuming a perfect bond between concrete and reinforcement (Malek and Saadatmanesh, 1998a,b).


Advanced composite strengthened RC columns RC columns are generally strengthened by wrapping composite sheets around the RC columns (Figs 19.6 and 19.8). The theoretical proposition from which this easy but effective technology flows is that the close external confinement of RC columns should significantly enhance their strength, ductility and energy absorption capability. Various parameters related to the performance of the confined RC columns have been studied by a number of researchers. RC column properties and design features analysed to date include the following: depth-to-width ratio, geometry/architecture of fibre reinforcements, effects of steel corrosion, wrapping angles for new reinforcement and the geometry of loading imperfection (Demers and Neale, 1999; Mirmiran et al., 2001; Pessiki et al., 2001; Ilki and Kumbasar, 2002; Hamad et al., 2004; Li and Park, 2004; Mukherjee et al., 2004). The geometry of an RC column section is a critical parameter that affects axial behaviour of the strengthened RC columns. Columns that are circular on plan are not only easier to wrap, but also their resulting confinement is more uniform. In contrast, columns square or rectangular on plan are

19.8 Strengthening of concrete column.



529

susceptible to higher confinement pressure at their corners, but much lower pressure on their flat sides. To mitigate this problem, rectangular and square columns are modified to generate an elliptical plan which, in practice, means that the corners have to be rounded to prevent stress accumulation. However, the presence of internal longitudinal steel reinforcement sets strict limits on the dimensions of corner radii (Rochette and Labossiere, 2000). Structural behaviour of strengthened RC columns is usually predicted by using analytical models to predict stress–strain behaviour between confined concrete and reinforcement. Modelling is usually based on their respective deformation compatibility and on the equilibrium of forces. For example, Mander et al. (1998) developed an analytical model to calculate the increased compressive strength of the strengthened RC columns attributable to the confinement pressure provided by transverse reinforcement. This model was further refined by a number of researchers to analyse the behaviour of the strengthened RC columns with circular, elliptical, square and rectangular sections (Wang and Restrepo 2001, Tan, 2002; Teng and Lam, 2002; Binici, 2005; Bisby et al., 2005a,b; Moram and Pantelides, 2005). Seismic resistance of the strengthened RC columns can be improved as significantly as the axial performance. A non-linear finite element analysis has been used for numerical analysis and prediction of seismic performance of the strengthened RC columns (Xiao and Ma, 1997; Parvin and Wang, 2002; Sheikh and Yau, 2002; Ye et al., 2003; Harajli and Rteil, 2004; Elsanadedy and Haroun, 2005).

19.2.2 Modular construction Modular construction systems usually consist of several FRP composite panels or elements which are bonded together with the twin aims of forming a membrane structure and providing complete structural integrity without additional framework (Fig. 19.9). Modular building systems are normally installed on site because light components make both transportation to site and handling after delivery economical and safe. Prefabrication is preferable for modular components. The vagaries of a building site are avoided. Instead, quality control commensurate with a well-supervised factory floor can be devoted to fabrication. Prefabrication can reduce the labour element of construction significantly. It can also radically reduce time spent on site. Modular systems have been designed for many different applications. The more notable were intended to satisfy a desire for greater style and drama in the resulting built forms, especially where weight, corrosion resistance and other specific properties were important. Functional needs have been met as well, as with radome structures that have to be electrically invisible to the radar they enclose. A spherical radome can be as large as tens of

530



19.9 FRP composite classroom.

metres in diameter, but executed as a single shell without any internal framework. A prime example of use of FRP composites to meet special requirements was the Home Planet Building in the Greenwich Millennium Dome. It was a 36 m diameter clear span FRP composite shell structure capable of supporting full wind and snow loading. The panels were moulded with an integral edge frame and the external skin forms an integral part of the primary structure. The lack of any internal framework enabled the building to be erected very quickly and with minimum work on-site. Such a structure could be used for numerous applications such as schools, offices, commercial industries, retail, exhibitions, etc, (Fig. 19.9). Manufacturing of modular systems has traditionally relied on hand lay-up techniques. Recent automated fabrication techniques incorporating the principle of modular design and construction are offering much more cost effective and competitive solutions for complete buildings. Improved resin technologies enable FRP structures to provide good long-term performance and adequate fire resistance for use in remote locations with high wind speed or snow loads. In addition, factory-based manufacture offers excellent environmental performance because structures can be made with close tolerances and thus be air-tight when erected. FRPs are intrinsically thermally resistant and off site fabrication minimises both waste and energy consumption. However, the design of modular constructions normally requires great investment in engineering analysis and tooling. FRP composite structures are fundamentally different from conventional ways of building both in their material configuration and their structural forms. If they are to provide


531

significant savings in production and assembly time on site, the methodology of building design needs to be changed to enable FRP to provide more efficient solutions than conventional building materials.

19.2.3 Structural building construction


Composite shells The use of FRP composites for structural building construction falls into two main categories. The first is non-structural or semi-structural building shells. The other is self-supporting buildings (Fig. 19.5). With composite shell construction, composite components are used in combination with conventional building materials to create visually arresting buildings whilst also maintaining structural integrity and keeping costs down. In this type of building, there is usually reliance on steel or concrete as the main structural materials. The composite shells or components are normally one-off or limited production, such as for commercial premises, hotels, schools, libraries, hospitals, airports and railway stations. A typical early, but sophisticated, structure was Mondial House on the north bank of River Thames in London (Fig. 19.10), which was Europe’s largest telephone exchange when it was built in 1974. Mondial House (demolished in 2007) was a 45 m high reinforced concrete ‘ziggurat’ structure, with each floor except the second being slightly smaller than the floor beneath. Each storey was clad with GRP. The wide, low-rise, stepped design utilised over 1000 channel-section cladding units. These were reinforced with foam cores and ‘top-hat’ stiffeners to facilitate handling and joining. The panels of the units were contact moulded using Scott Bader Crystic 356PA class ‘O’ fire resistant polyester resin and isophthalic polyester gel-coat 65PA. They measured 3.0 m high × 1.8 m wide and 1.2 m deep, with the exception of the panels on the eighth floor and louvred

19.10 Mondial House.


532


panels which were 6.7 m high. The rear casings were white 1.0 m × 0.6 m single skin GRP. The main panels and louvred panels were made by hand lay-up while the rear casings were hot-press moulded. The panels were fixed by means of clamps. These were fastened to the edges of the sideribbed panels, the building at two points at each level and the rolled steel joists suspended beneath the slab. The clamps fitted into a specially formed groove and were fixed back to the main bracket by a single stud. The main bracket was fixed to the slab by parabolts and to the rolled steel joist (RSJ) by bolts. The bracket assembly permitted thermal movement in the panels. It also allowed adjustment in both directions: perpendicular to the building and sideways. The panel joints were designed to accommodate thermal movement between panels, to provide fixing tolerances and to prevent ingress of water into backgrounds. Other composite shell constructions include the Pipex structural composite, which is used to withstand corrosive attack and high mechanical loads in Cottam power station on the west bank of the River Trent, Retford. Morpeth Street School in East London is constructed with an all-GFRP roof of over 70 V-shaped beam modules spanning some 17 m. In a singleclassroom extension in a primary school in Preston, all building panels have a solid GFRP skin but seven of the panels contain non-opening triangular windows and five contain circular apertures for ventilation fans. The Covent Garden Flower Market roof in Nine Elms, London covers approximately 100 m2 with 924 inverted truncated pyramid units on a 4 m square grid attached to a supporting steel framework. The double-skinned GFRP composite units were injection moulded. Composite shells are also designed for internal partitioning walls such as those in domestic and portable housing, offices, schools, hospitals, hotels, airports or any mass transit area. They are particularly useful where access is restricted or the floor cannot support high loads. Typically, a shell system comprises panels which fit together on site for ease of handling and erection. FRP composite panels can take the form of a sandwich where two surface skins of FRP composite encapsulate a structural core. They can be full height or part room height. A variety of joining members allow assembly of the panels either side-by-side or end-to-end. This flexibility gives the designer almost limitless scope for configuration of rooms and cubicles to meet the needs of clients. Advantages of this kind of walling system include the low density of FRP composite panels (lightness equates to ease of handling and assembly); versatility of bonding to other building materials; variety in texture or colour of interior finishes; flexibility in location of electrical conduit and outlet boxes; good thermal properties; good sound attenuation and ease of cleaning. These systems are also easy to disassemble and rearrange because of their lightness and manoeuvrability.


533


Full building systems Good strength-to-weight in polymer composites and the versatility offered by the various manufacturing technologies make it feasible to construct complete primary structures. Low weight reduces dead-loads. This offers excellent opportunities to designers wanting not only wide spans, but also the architectural freedom to devise dramatic new construction forms. Full building systems consist of solid interlocking components or cellular components. Such a system would consist of a number of units which could be fastened together securely by mechanical interlocking: e.g. tongued and grooved joints or concave grooves held together by the insertion of solid ‘dog-bone’ connectors (Fig. 19.11). Floors, walls and roofs can be formed from these units. Large components comprising several individual panels for whole walls or floors can be sub-assembled off site under factory conditions, with final assembly and bonding taking place on site. The panels are load-bearing components that require no secondary framing. A typical example is the Advanced Composite Construction System (ACCS) designed by Maunsell Structural Plastics. The Severn Bridges Visitors Centre is the world’s first advanced composite multi-storey building constructed using ACCS systems (Fig. 19.11). Initially, this multi-storey block was commissioned as a ‘fast-track’ alternative to conventional site office accommodation for the Government agency team working on the second Severn Crossing project. When the building fell vacant after completion of the bridge, it was converted to its current use as the Severn Bridges Visitors Centre. Full FRP composite building systems are rare and little research is being done; however several companies offer complete systems; e.g. Fiberline Composites A/S, DK-6000 Kolding, Denmark; Future Systems Architects; White Young Green and NASA (the IsoTruss grid structure) (Kim et al., 2005). In addition to the development of improved composites or components for entire building systems (e.g. Singh et al., 1995; Paolozzi and Peroni, 1996; Caprino and Lopresto, 2000; Martin, 2006), full-system verification tests are sorely needed for the following: connection design,

19.11 Full building system with ACCS system.


534


systemic structural optimisation for building geometry, roof configuration, foundation anchorage and building envelope. For example, current joining technology of mechanical interlocking with both grooved and solid connectors is likely to work well for metal components, but might be predicted to fail when analysed using structural parameters pertaining to FRP composites. The alignment and connection between the fibre and matrix of FRP composites are, in principle, just like those of wood (natural composites) (Fig. 19.2), because both are similarly anisotropic. Both wood and FRP composites have low bearing and interlaminar shear strengths. Bolts and rivets carry and transfer connection forces to specific points, thus causing high stress concentrations. These could cause the composite construction to fail at bolted or riveted joints. Goldsworthy and Johnson (1994) patented ‘snap-fit’ joints for composites based upon an original fibre-architecture design that paid particular attention to interlaminar requirements wherever loads were concentrated. The ‘snap-fit’ jointing method has already been proven as a structurally reliable method of connecting wooden as well as plastic parts. ‘Snap-fit’ joints have the capability of distributing stress over a wide area. The method has been acclaimed as being the first of an entirely new generation of joints for composites. For modular construction systems, it is highly advantageous. A small range of pultruded FRP profiles can be snap-fitted together to achieve rapid assembly (Fig. 19.12). Bolts are used as retaining devices and do not generate localised areas of high stress. Some commentators are predicting

19.12 ‘Snap’ joint for composite.


535

that this concept could provide thermally efficient, low carbon housing at moderate cost. Its potential for very rapid assembly would tackle the pressing need to increase the rate of home-building. It could also solve the problem of how to mobilise for and erect quickly temporary, demountable buildings and shelters; principally to meet urgent housing needs in disaster zones, but also to serve as temporary barracks for troop deployment in wartime.


19.3

Composites in bridge construction

While FRP composites have been used in aerospace and marine applications for over 50 years, FRP prototype bridge structures only appeared in the 1970s. The advantages of using GRP composite bridges were not appreciated until recently when, from the mid-1990s onwards, the serious deterioration of reinforced and pre-stressed concrete bridges threatened transport infrastructure in Europe, North America and Japan. It was estimated that of more than 600 000 bridges in America, 130 000 are structurally deficient, another 100 000 are functionally obsolete and 143 000 are more than 50 years old and unsuitable for current or projected traffic demands (Zureick et al., 1995). Some 6 000 more bridges are added to the deficient list annually. In the UK, where there are 135 000 bridges, the replacement value of damaged bridges is nearly £10 billion and nearly £230 million per year is spent on maintenance (Busel and Lindsay 1997). Engineers were forced to look for more durable and less labour-intensive repair materials for easier and faster installation; consequently, FRP composites are now increasingly being used for the repair of bridge structures around the world because of their potential advantages of high strength and stiffnessto-weight ratio, superior durability, absolute resistance to salt-induced decay, electrical neutrality and good thermal properties.

19.3.1 All-composite bridge Bridges constructed entirely from FRP composites are reliant on either cable-stays or composite crossing beam systems. Cable-stayed construction is an attractive bridge-building method given that cable loading lies mainly in the fibre direction and that composites have high flexural strength, very low creep and outstanding fatigue strength. As stated above, they are entirely resistant to corrosion. However, it is invariably necessary to deal with two salient engineering issues. The first is that of protecting the rods in order to prevent degradation during handling and ageing of the resin during use. The second is the avoidance of stress concentration around the


536


anchorages of the composite stay cables. The composite crossing beam system entails the fabrication of a latticework of beams. Large beams are usually pultrusions. This is a fast, cost-efficient and structurally effective means of building bridges. It also readily facilitates structural rehabilitation of bridges weakened by decay. The first known example of a composite material bridge was built in the USA using paper fibres confined in a polymer matrix. The very first composite beam structure dates back to 1982 and is located in China (Miyun Bridge, Beijing). The bridge was designed by the Highway Science Research Institute and the Shanghai Glass Reinforced Plastics Research Institute. The bridge is a simply supported two-lane bridge, 7 m wide with a clear span of 20 m. It consists of five girders constructed within what appears to be a honeycomb sandwich structure. The bridge was proven by loading it with four trucks having a total load of 85 tonnes. The centre span deformation was 27 mm. The Guanyinqiao Bridge, also built in Chongqing, China, in 1988, is at present the longest bridge in the world in this category, measuring 157 m in length. Following the Miyun Bridge construction, considerable research resources went into developing use of composites (Yicheng, 1990; Alanpalli et al., 2002; Uddin and Abro, 2008). A few research programmes have been launched to investigate full size bridge systems (e.g. Karbhari et al., 2000a; Ecket, 2001, TxDOT, 2005). Study of ‘all-FRP’ composite bridges, which includes their design, analysis and fabrication together with a full-scale experimental structural validation both prior to and during fabrication and erection, shows that their overall structural behaviour could be accurately predicted from design equations based on laminated plate and sandwich theory for composite materials. Finite element modelling could be performed to approximate the structural behaviour of a bridge. The bridge superstructure sections can be designed and constructed to exceed the performance criteria based on experimentally measured stiffness, deformation and face-sheet strains. To date, composite beam structure technique has been widely adopted. No fewer than 109 bridge structures have been built in the USA, with the average length being 12.8 m for footbridges and 16.4 m for road-bridges. In Europe, more than a dozen bridges with composite beams have been constructed, including Ginzi Bridge, Bulgaria, built in 1982; Kolding Bridge, Denmark, 1997; Pontresina Bridge, Switzerland, 1997; Lleda Bridge, Spain, 2001; Fredrikstad Bridge, Norway, 2003; four bridges built in the Netherlands between 1985 and 2003 and seven bridges built in the UK between 1975 and 2003 (Scott, 1993; Weaver, 1997; Bakis, 2002; Matta, 2003; Keller, 2003). There remains, however, considerable unease about use of composite stay cables. Composite system bridge construction has been undertaken on



537

a case-by-case basis and not produced on any mass scale. Within the domain of cable-stayed structures, Japan is undeniably in advance of other countries (20 bridges using composite bracing systems). The Rainbow Bridge built in Tokyo in 1991 is currently the longest composite cable-stayed footbridge in the world (983 m in length). Two composite cable-stayed structures in Canada and 11 in the USA were constructed between 1991 and 2001. In Europe, composite cable-stayed bridges are found in France (Laroin, 2002), Denmark (Herning, 1999), Switzerland (Winterthur, 1996), Austria (Notsch-Karnten, 1990), Germany (six projects between 1980 and 1991) and the UK (Aberfeldy, 1992) (Le Toullec, 2000; Keller, 2003; Colombi et al., 2004; Moy et al., 2004). The Aberfeldy Advanced Composite Footbridge in the UK has been extensively reported in the literature (Fig. 19.13). The main structure consists of a cable-stayed GFRP deck, suspended by Parafil aramid ropes from GFRP towers. The footbridge has a span of 63 m, a width of 2.12 m and an overall length of 113 m and contains a total of 14.5 tonnes worth of composite materials. The polymer composite components of the deck and towers of the bridge were manufactured from the pultruded ACCS plank and brought together and locked by working a toggle (dog bone) section into the two grooves, one in each plank. Three planks and four connectors were joined alternately, in a single thickness, to form the 2.12 m wide deck. The bridge was finished by the addition of GRP balustrades. It is stayed from two 18 m-high ‘A’ shaped GRP pylons using Parafil cables – Kevlar aramid fibres sheathed in a protective low density polyethylene coat. The bridge is designed to carry live loading of 5.6 kN/m; its dead weight is 2.0 kN/m including 1.0 kN/m ballast. The light weight of FRP composites offers considerable advantages when movable bridges are required. Bonds Mill Lifting Bridge across a canal in the UK (Fig. 19.14) provides access for heavy trucks. The whole structure can be lifted by means of a pair of hydraulic jacks. The deck is made from GRP pultrusions. The upper layer of cells is filled with structural grade foam to resist localised bending under wheel loads.

19.13 Aberfeldy GRP foot bridge.

538



19.14 Bonds Mill Lifting Bridge.

19.3.2 Hybrid bridge Bridge decks Composite decks have been used in numerous bridges around the world. It is claimed that a FRP composite bridge deck can be about one-fifth the weight but six times stronger than a conventional concrete deck. Such a deck would be well suited for modular fabrication and mass production. It would possess good energy-absorbing capability, it would not be susceptible to fatigue and it would be resistant to corrosion. Among the early applications of GFRP in bridges was its use in permanent form work for the deck slabs of reinforced concrete structures (Hall and Mottram, 1998). FRP bridge decks have sandwich profiles, spanning transversely or longitudinally between supporting elements (such as steel beams) or suspended from tension cables. A sandwich system for decks consists of multi-cell sections glued or bolted together to form a compound structure (Fig. 19.15). Examples are the ACCS system in which a multi-cell box section is connected by toggles and glue-bonding and the ASSET system which has a two-cell prismatic profile. The multi-cell sections are usually constructed from pultrusions of glass fibre reinforced polyester or vinyl ester, although some other types of manufacturing processes have been employed, including wet lay-up (e.g. Miyun Bridge in China) (Seible, 1996), the resin infusion technique (Chajes et al., 1998) or a combination of these methods. The biggest deck spans can be up to 10 m, carrying 40 tonne loads. The wearing (topmost)


539


19.15 Bridge deck system (top: multi-cell box deck, bottom: ASSET system).

surface is asphalt which is usually applied to a polymer concrete base (often applied before installation). The replacement of steel–asphalt or steel–wood decks represents a major part of composite bridge deck projects. Around 100 such structures were raised between 1995 and 2003 in the USA (Deskovic and Triantafillou, 1995; Canning et al., 1999), six structures between 1988 and 1993 in China and 18 structures between 1990 and 2003 in Europe (Hani et al., 1997; Kim et al., 2005). Bridge enclosures Soffit enclosure is an essential part of a hybrid bridge system. The enclosure is a structural casing suspended from the soffit of steel composite, plate girder or concrete bridges to provide inspection and maintenance access. The floor and integral side membranes completely enclose the steelwork (or other structural materials) and protect it against corrosion and moisture ingression. A hybrid system normally combines a lightweight tubular steel space frame, steel cast nodes, an aerodynamically profiled FRP enclosure shell and a reinforced concrete roadway slab. These elements are designed to complement one another in order to achieve optimal performance (Fig. 19.16). Bridge enclosure was initially developed to extend maintenance intervals for painted steelwork. The technique is based on the assumption that clean steel does not corrode significantly at relative humidity up to 99% provided that environmental contaminants are excluded. It has been proved that chlorides and sulphurous pollutants can be kept out of the enclosure by good seals. The rate of corrosion of un-coated steel protected by enclosure has been found to be only 10% of that of painted steel exposed to external air (McKenzie, 1993). Enclosure also greatly reduces the rate of corrosion on steel which is already rusting. Recent examples of bridge soffit enclosure

540


19.16 Bridge enclosure.


have been specifically designed to be easy to install and to facilitate safe and convenient access for maintenance personnel as well as protecting the bridge soffit. A collateral benefit is reduced traffic disruption during essential bridge maintenance.

19.3.3 Reinforcement In addition to the FRP composite encasement and wrapping of concrete columns, or other externally bonded reinforcement discussed in Section 19.2.1, composite reinforcement rods are used in some reinforced concrete bridges and crossings located in an aggressive environment where reliable rebar protection cannot always be guaranteed over the long term; for example, where there is winter penetration by soluble de-icing salts or exposure to seawater. Similar problems will arise in the manufacture and processing of chemicals. Engineering that requires low electrical conductivity or electromagnetic neutrality should avoid specification of steel or other metals (Tan, 2002). Substitution of composite reinforcement rods would not only provide a durable solution, but it would also permit monitoring devices, data loggers and other sensors to be incorporated in the structures. Measurements thus obtained would help to improve future design of composite structures (Buyle-Bodin et al., 1995; Hall and Mottram, 1996; Sheard, 1997). Pedestrian bridges with composite rebar first appeared in Japan in 1991. The first road structure with composite rebar dates from 1995 in Canada. There are about 30 structures in USA and a dozen in Canada (Pham and Al-Mahaidi, 2004). Fidgett Footbridge was the first concrete footbridge in


541

the UK fully reinforced with GFRP rods. Vibrating wire strain gauges were cast into the concrete and fibre optic sensors fitted to the slab to allow the long-term monitoring of the bridge (Buyle-Bodin et al., 1995).


19.4

Composites in other constructions

In addition to the major construction applications discussed above, FRP composites are used in a diverse range of other applications in construction, such as in masts, towers, gantries, railings, other fencing, offshore constructions, processing plants, tanks, pipework, glulam strengthening, repair of timber structures, railway infrastructure, cable support systems, temporary works, etc. Nevertheless, all these applications can be categorised into two groups of applications: one is for specialised applications for which certain specific properties of FRP composites need to be deployed (such as their flexural properties or corrosion resistance), for instance, in ground engineering (Fig. 19.17). The other springs from the needs of society at large which become apparent through the dynamic evolution of modern living requirements (e.g. thermally efficient doors and windows). A discussion of individual examples of applications is outside the scope of this book. FRP composites have been employed in the manufacture of doors and windows for many years. FRP framing elements entered the window and door market in the early 1980s. The driving force behind the emergence of this new material was the willingness of homeowners to spend a ‘one-off’ sum of money replacing windows and doors instead of spending money at regular intervals to re-paint and maintain existing frames. Fuel price rises

19.17 Polymeric composite for ground engineering.


542


and increased environmental concern have also generated a strong market for replacement double glazing. High labour costs have made contractors reluctant to risk frequent calls for the repair of defective products within the warranty period; therefore, only the most dependable replacement window frame materials survive in the market. Wood, aluminium, unplasticised polyvinyl chloride (UPVC), steel and composites are all used to make window frames, sashes and door components. Wood is typically used in solid, not hollow section. It may swell and shrink over time. The tightness of fit in an opening may slacken, thus allowing air to penetrate. Aluminium and steel frames create ‘thermal bridges’, which leak heat out of the structure. Aluminium is highly mobile in its response to temperature changes (its linear coefficient of thermal expansion is 23.1 × 10−6 per unit length per degree kelvin). The susceptibility of UPVC to excessive thermal movement is even worse. Its dimensional changes are four times greater than for a glass fibre pultrusion. Walls of large UPVC frames need to be reinforced with steel or aluminium because of the low strength and stiffness of this thermoplastic. Stiffening with metal creates a thermal bridging risk. By contrast, pultruded FRP composite frames are stiff enough to serve without reinforcement. They are dimensionally stable in response to both temperature and humidity changes. The German Passive House Institute has reported that pultruded frames seem to maintain air-tightness around perimeter joints of frames and casements better than frames made of wood, metal or UPVC. GRP has not yet made great headway in door and window frame manufacture. Processing technologies for thin-walled but complicated pultruded profiles at speed are capital intensive. The anticipated economies bestowed by mass production have not yet come about. Another obstacle has been fast curing of resin to achieve an appearance comparable to market-leading PVC finishes. Nevertheless, the GRP door and window market will continue to grow by virtue of the benefits offered by the manufactured products, such as good thermal performance and much better air-tightness (to meet ever-stricter building regulations). The strength and stiffness of pultrusions as framing permits frames to be reduced in girth. Glass area can be increased compared with what is possible with timber and UPVC frames. Glare is reduced and more natural daylight is admitted through any given window opening, which lessens dependence on electrically powered artificial lighting. To achieve zero-carbon accreditation, homes must minimise consumption of electricity. This and other sustainability issues pertaining to the material selection for window and door frames are the subject of much current research. One example is NATCOM, a TSB-funded project which aims to investigate and develop carbon-capturing, low energy and sustainable door and window frames made from renewable natural fibre reinforced composites (Fig. 19.18).


543

19.18 Composite door frame.


19.5

Performance in use

The number of construction projects containing FRPs is small, despite the variety of possible applications for building being very wide. In addition to the understanding of how to optimise design to utilise the potential benefits of composite construction, there is another significant area of concern, which is the quantification and long-term performance (both mechanical and biological) of the FRP composites and their sustainability in comparison with other construction materials.

19.5.1 Standards and codes At present, composites for construction uses are mostly tested to the standards of composites for general purposes. Current standards and specification guidance contain five different levels, from raw materials to performance in uses. These are: constituent specifications and test methods, compound material specification and test method, composite process and factory control, composite and component performance and product approval standards. While constituent materials for composites are specified and tested by reference to the comprehensive standards adopted both from textiles (i.e. fibres and yarns) and from plastics and resins (thermoset and thermoplastic resin systems), standards and specifications for other types of testing have been developed specifically for composites. Standards now apply across a wide range (Sims, 2007). However, experience with using these standards for composites in construction continues to manifest

544


practical difficulties, because most of them were drawn up originally for composites use in other applications, such as aerospace, gas pipe or column manufacture. Examples are: EN40-7 for lighting columns, ISO527 for tension, ISO14125 for flexure, ISO14126 for compression and ISO14129 and ISO14130 for shear tests. Several long-term durability test methodologies have also been used, i.e. ISO877 for outdoor weathering tests and ISO4892 for accelerated laboratory testing. The main bodies responsible for composite standardisation include ISO, CEN, ASTM, JIS and other national bodies. Detailed documents can be found on the websites of relevant bodies (e.g. BSI and ISO).


19.5.2 Long-term performance Long-term performance of construction materials has been an important research subject given the need to ensure that the materials are able to sustain applied loads without being subjected to ultimate limit states (ULS) and serviceability limit states (SLS). Long-term performance assessment concerns the behaviour of materials/products in service. Performance criteria may pertain to mechanical loads (Ahmad et al., 1994; Karbhari et al., 1996; Kumar and GangaRao, 1998; Palmer et al., 1998; Oh et al., 2005), physical issues (e.g. wetting) and biological activity (e.g. mould colonisation; decay) (Fig. 19.19). Assessment is normally carried out by using accelerated short-term laboratory testing capable of simulating static or dynamic actions to which a certain composite will be subjected in service. Validation of this type of assessment is based on the comparison of results (and degradation processes) from long-term tests between those obtained from the normal environment and those recorded under accelerated testing conditions. The procedures may include (i) visual inspection and condition ratings, (ii) health monitoring of structures and systems, (iii) destructive and non-destructive testing and (iv) modelling. Figure 19.20 outlines a general assessment process. Research on long-term behaviour falls conveniently into three interrelated categories. The first pertains to the quantification of timePerformance in uses (long-term performance) Creep

Duration of load

Flat Edgewise Planar bending bending shear

Fatigue

Impact Bioresistance Environmental degradation

Panel Concentrated Centric shear load load

19.19 Performance in uses of composite.

Moisture Photo-oxidation resistance (UV, ozone)

Inspection of building

545

Outdoor accelerated weathering

Outdoor natural weathering

In use testing

Laboratory accelerated ageing text (carbon arc, xenonarc, fluorescent lamps)


Degradation processes Yes Laboratory tests valid for establishing long-term predictive models


19.20 Design of long-term performance testing.

dependent behaviour of composites of different types and compositions under both natural and artificial climates. Much of the information thus obtained is subsequently used in the determination of time-dependent factors (the duration of load factor and creep factor) to be incorporated in codes for structural design. The second relates to the modelling of creep behaviour for both descriptive and predictive purposes, while the third is concerned with the much more academic challenge of trying to understand the basic mechanism underlying rheological behaviour (Fan et al., 2006). The real challenge is how to integrate various parameters (e.g. load, air conditions and biological hazards) that could be detrimental to the longterm performance of composites (it can be adversely affected by three general types of degradation: mechanical; physical and biological) into the testing procedure. Successful integration would also facilitate the production of a model. The procedure often followed is to subject specimens to a main variable (for instance load on creep testing) (Fig. 19.21) and, in some cases, to evaluate the effect of a secondary variable, such as on the effect of dynamic loading on creep deformation. FRP composites are considered durable and very low creep; therefore, there has been little research done on long-term mechanical performance. While FRP composites may not rust like steel, degrade like concrete or suffer biological decay like timber and timber composites, environmental degradation of FRP composites can arise from a complex set of processes.


546


19.21 An example of long-term creep test.

The combination of ultraviolet (UV) radiation, heat gain under direct sunlight, constant wetting in rain, exposure to freeze–thaw cycles and contact with atmospheric oxygen can cause deterioration. UV radiation may result in photo-oxidative degradation, chain scission, cross-linking and consequent de-bonding of composites. The resultant products may react with oxygen to form functional groups (secondary oxidative reaction), such as carbonyl (C=O), carboxyl (COOH) or peroxide (O—O). UV exposure usually affects the top few microns of the surface, depending on the duration of time. The results can be colour fading (sometimes darkening), yellowing, blooming, disappearance of gloss and chalking. Surface degradation may cause stress concentrations in the material. However, the effect of environmental degradation on the mechanical properties of the component has yet to be studied. The effect of weathering on composites can be predicted from outdoor natural weathering and outdoor accelerated weathering by using special mirrors and accelerated laboratory testing (i.e. carbon arc, xenon arc and fluorescent lamps). An assessment carried out by the Network Group for Composites in Construction in UK on the 33-year-old composite cladding of Mondial House (Fig. 19.10) concluded that the FRP



547

composite cladding is impressively durable (NGCC, 2008): the FRP panels remained bright white when viewed from a distance but were dull and dirty on closer inspection. This surface dirt was easily removed by cleaning. The cleaned surface was then polished to restore the white gloss finish. It must be noted that the effect of weathering on FRP composites may vary considerably from one season to another season, from one year to the next and from one place to another. A four year study in three different sites, measuring the effect on the ultimate tensile strength (UTS) of exposure of glass/ polyester laminates, found that the strength retention was 97% for UK inland exposure, 95% for UK marine exposure and 81% for a tropical site (NGCC, 2008). Modelling structural deterioration is difficult owing to the inherent complexity of the process, the complex structure of composites and the multitude of external factors and mechanisms that are responsible for deterioration. The available approaches to model life-cycle performance can be divided into network level and structure level methods. Network level methods predict the conditioned deterioration of a group of structures grouped either by locality or type. Markov methods and statistical regression techniques have been widely used in modelling structural deterioration at the network level (Hastak and Halpin 2002; Mishalini and Madanat, 2002). Most existing modelling relies on historical performance data from similar structures (structure level methods) to predict the future performance of a new structure. This may not be suitable for use in the case of FRP composites in civil engineering. Instead, the models should be based on knowledge of the physical and chemical processes that are responsible for material deterioration until sufficient historical databases can be built over time (Deepak, 2003; Deepak et al., 2007). An indentation law has been proposed to predict the tensile behaviour of carbon fibre composites under low-velocity impact (Caprino and Lopresto, 2000).

19.5.3 Repair and maintenance There are many different ways to repair, maintain and upgrade GRP composite constructions. Processes can be grouped into structural and nonstructural repair or in situ and product repair (Fig. 19.22). Casting techniques have been employed to enhance the physical appearance of a structure. The technique can also improve structural strength and stiffness (Saadatmanesh et al., 1997; Debaiky et al., 2002). It has also been established that long-term performance can be significantly improved by comparison with unconfined components (Toutanji, 1999; Toutanji and Balagura, 1999; Karbhari et al., 2000a,b; Kshirsagar et al., 2000; Lee et al., 2000). However, with this technique space is needed to accommodate the additional casting, which is not always available. Another approach is to use pre-stressed tendons attached

548

Management, recycling and reuse of waste composites Assessing damage Excessive damage Scrap

Moderate damage

Minor damage

Surface colour fading

Complex Temporary repairing scheme repair (associated with manufacturer)

Cleaning and maintenance scheme

Permanent repair and quality check

Return to service


19.22 Repair and maintenance process.

to the structure, which is more or less the same as that for traditional external pre-stressed steel tendons except that the static behaviour of the structure may change. This is because additional parts are added which may unload the critical section of the structure. One sophisticated method for improving the performance of a structure is adoption of more advanced calculation models which take account of real dimensions, real loads and real material data. While environmental degradation may have little effect on the mechanical properties of FRP composite structures, polyester tends to show surface dirt, especially on plain coloured flat surfaces, thus losing its initial gloss and brightness. Periodic cleaning and polishing may be necessary to maintain initial high gloss appearance. Resistance to degradation can also be improved by imparting surface texture, incorporating other materials in surface coatings and intentionally varying the surface colour.

19.5.4 Sustainability of construction composites Given current environmental concerns, it is essential for the market success of any new building material that there are accurate descriptions of the environmental attributes over the whole life cycle. These should encompass extraction, processing, construction, uses, maintenance, demolition and eventual disposal. These ‘green’ requirements would be an addition to functional material characteristics such as stiffness, strength, affordability, durability, versatility and ease of use. Traditional construction materials possess some of these characteristics, but none possesses all of them. The



549

ultimate goal of the construction industry would be to develop a material that possesses not only the basic requirements for construction, but also the characteristics associated with environmental sustainability. FRP composites may offer a solution on account of their uniquely wide range of constituents and material properties. Life-cycle assessment (LCA) is the recognised method of assessing and quantifying the environmental impact of FRP composite products for construction. LCA provides a means of comparison between different materials or manufacturing processes for any given product to determine these impacts within defined assessment procedures (e.g. ISO14040). An example in the case of FRP composite construction materials is the comparative study of generic double curvature GFRP panels (architectural cladding component), made by the closed mould processes (resin transfer moulding, RTM, and resin infusion) and the open mould processes (hand lay-up, vacuum bagging and spray-up) (NGCC, 2007). The former was found to be kinder to the environment. Generic flat sandwich panel components suitable for large-scale applications, such as bridge decks and marine structures, were compared to double curvature panels by rigorous LCA, with the result that the flat components presented the better environmental profile. The reason was that the double-curved panels had a higher percentage of raw materials to achieve an equivalent stiffness (i.e. thicker board laminate). Some of the extra materials had significant impacts. LCA on a complex moulded component, typically manufactured using an open mould process (although it can also be made by using RTM), autoclave moulding and compression moulding, showed that the constituent materials had a big impact on the environmental profile. Constituents with intrinsically low environmental impact (e.g. hemp) impart better LCA results than the materials of high environmental impact (e.g. carbon fibre) (NGCC, 2007). Nevertheless, it should be noted that, just as with conventional construction materials, the completeness and accuracy of the available information for the LCA are often subject to challenge. The information for LCA can either be obtained by measuring and assessing a new set of data for each material or by calculating the data from the material constituents when aggregated together. The latter method has a weakness. Identical materials from different batches may have been produced by different manufacturers employing dissimilar production processes, which could give rise to errors or approximations in their assessment.

19.6

Construction wastes, reclaim and recycling

Waste is produced throughout the whole supply chain, including during manufacture, distribution, design, construction, refurbishment and


550


H o u si n g

H o s p it al s

S ch o ols

Offices

19.23 Construction wastes generated from supply chain.

demolition (Fig. 19.23). Construction waste is normally categorised into what is produced during new building, what emanates from housing refurbishment and that arising from demolition in all building sectors. The construction industry consumes huge amount of energy and material resources. As a result, it produces more waste than any other industry. It has been reported that the construction industry may be responsible for more than 50% of the waste arising in the UK (Hobbs, 2006a). FRP composites are being deployed more and more in construction because of their lightness, ease of installation, low maintenance, ‘bespoke’ design capability and corrosion resistance. For example, the UK FRP industry produces 240 000 tonnes as different products every year of which 11% are sold for construction (NGCC, 2002). Waste management is becoming significant as it affects profitability: landfill charges are increasing year-on-year as a result of recent government legislation. In addition, all advanced economies


551

now have to contend with the depletion of non-renewable resources. Current and impending waste management legislation will put more pressure on the industry to reach a better understanding of the available options for dealing with FRP waste.

19.6.1 Fibre reinforced polymeric (FRP) wastes in the construction industry


Construction site wastes Construction wastes are calculated by using the tools of the Key Performance Indicator (KPI) and the Environmental Performance Indicator (EPI). The former is given as m3 wastes per £100 000 project value and the latter as m3 wastes per 100 m2 floor area of a building. The amount of all waste varies considerably from one type of construction to another. A civil engineering site generates the highest amount of total waste, about 62 m3/100 m2 floor area, mainly from excavation, groundwork and the mainframe. Education, residential, office, health care or hospital and leisure construction sites give rise to about 22, 19, 14, 12 and 4 m3 wastes per 100 m2 of floor area respectively. The average amount for these five categories is 19 m3 wastes per 100 m2 of floor area. This bulk waste contains 0.6 m3 plastics per 100 m2 floor area which equates to 0.14 tonnes (Hobbs, 2006b).

Refurbishment wastes Housing features are listed as follows in descending order of the amounts of waste generated from their refurbishment: central heating, kitchen, bathroom, doors, windows, rewiring, roof covering and roof structure. The waste stream from refurbishment is generally a mixture of small quantities of different waste. While there is very little data available, the estimated amount of plastics waste is around 5% of over 5 million m3 total refurbishment wastes per year in the UK (Hobbs, 2006b).

Demolition wastes FRP deconstruction waste are minimal compared to other forms, totalling less than 1% of 26 million tonnes of construction demolition waste in the UK (Hobbs, 2006b). It is currently sent to landfill. However, as the amount of FRP being used in construction increases over the next decade or so, as a result of the ever-increasing range of building applications, the eventual volumes will increase.

552


19.6.2 Reclaim and recycling of construction fibre reinforced polymeric (FRP) wastes


In common with other forms of waste, the priority for dealing with FRP composite waste is to manage sites and factories to prevent it being generated. The next step is to reuse or to recycle. The last phase is to rationalise final disposal (the diagram is: minimise → reuse → recycle → recover energy → landfill). Currently FRP production waste is generally disposed of. The reasons are as follows: • With the exception of carbon and aramid fibres, raw materials used in FRP composite manufacture are considered relatively inexpensive. • The quantities of waste produced are usually low in comparison to product volume. • There are few, if any, off-cuts on building sites because FRP components are designed for a specific use, pre-moulded and made to measure. • The cost for land-filling composite waste is perceived as being relatively low, although this is changing as landfill costs now increase annually. By contrast, the cost of recycling is deemed to be much higher because it involves several different processes and treatments. Nevertheless, reuse and recycling of FRP waste will become essential for the FRP composite industry if it is ever to exploit to the full the great environmental benefits conferred by its construction sector products.

Waste minimisation The most cost effective and environmentally beneficial options for waste management are prevention and reduction. The versatility of FRP in producing artefacts with complex three-dimensional geometry and in facilitating ‘made to measure’ and off-site fabrication makes FRP composites one of the least waste-generating construction materials. Most automated processes are efficient. There is little scope for improvement in reducing manufacturing waste. A possible exception is waste generated at the beginning and end of production runs caused by components failing to meet the accepted standard; for instance scrap arising from faulty preparation (Conroy et al., 2006). FRP wastes from the supply chain of construction materials may include a small quantity of off-cuts, over spray trimmings, trimming dust, trimming from vacuum infusion, defective items and trial runs, plus obsolete moulds. This type of waste is relatively clean and can potentially be recycled (Fig. 19.24).


553

19.24 FRP composite offcuts.


Reuse Currently FRP components are unlikely to be reused from one application to another after end of life. There are several reasons: the first is that the majority of FRP components are bespoke in nature, being designed for a particular set of circumstances and conditions for particular applications; therefore, they would not usually be transferable to a different usage; the second is that GRP components are designed and installed for a dedicated service life: inevitable deterioration renders them unusable after the end of life; the third is that there is a lack of long-term durability data for GRP in use (i.e. predictable and reliable performance related to service times). An example of structural components is the boxor I-beam. These beams can potentially be reused as semi-structural components but, at present, it is very difficult to identify degradation or to assess long-term creep effects and retention of load carrying capacity. The above situation can and should be changed. Both manufacturer and designer will soon owe a ‘duty of care’ to make sure that effective measures for deconstruction and reuse of FRP are considered at all stages: manufacturing, applications and construction on site. Sealing and gluebonding of joints on site should be configured so as to allow the disassembly and reuse of FRP modular and prefabricated systems. FRP manufacturers will be required to confront and solve the problems of durability (e.g. UV stability, colour fading, self-cleaning and decontamination) and fire resistance, which are particularly important in building. Elements at risk from fire are for example fascias, panels, mouldings, cladding, gantries and roofing components.

554



Recycling Recycling of GRP composites is directly related to the two types of polymer, i.e. thermoplastic or thermosetting. Thermoplastic FRP composites can be recycled by remelting and remoulding. This is not possible for the thermoset FRP composites which predominate in construction products. Opportunities for recycling depend on the type of plastic waste. Much research work has been carried out to achieve a better understanding of FRP wastes and their end-life solutions (Demura et al., 1995; Hobbs and Halliwell, 1999; George and Dillman, 2000). Great efforts have been devoted to investigating means by which FRP could be returned to its original constituents. To date, one of the methods found to be effective is the thermal treatment of FRP composites. Thermally induced decomposition helps to separate fibres from matrix resin so that the fibres can be recovered. Fibres are considered the most expensive materials within the GRP composites (Jody et al., 1999). Mechanical processes have been found to be effective. In this method, thermoset FRP composites are ground to fine particles which then are used as a filler in new FRP materials (Hobbs and Halliwell, 1999). Several recycling pilot plants (e.g. in France, Germany, Italy and the Netherlands) have proved that FRP composites can be recycled in this way (GPRMC, 2001). However, new markets need to be found for the recyclate. Incineration has also been studied as a way of disposing of FRP composite waste from construction. Problems associated with gas emissions from incinerator flues have been hotly debated. Current incineration has two options: the first is the incineration without energy recovery. This process is the least preferred because it dissipates embodied matrix resin energy which could be harvested. The other is incineration with energy recovery or composting (FRP composites have a high calorific value). However, the process is much more expensive since the high calorific content together with toxic emissions from FRP composites tends to overload the system. This is not such a risk when domestic refuse is incinerated.

19.6.3 Construction products from fibre reinforced polymeric (FRP) recyclates FRP recyclates are potentially useful materials for the development of construction products. While a number of options for the use of FRP recyclates have been identified (Simmons, 2001), recyclates are basically used either as filler or as reinforcement. Recycled FRP composite could be used as a reinforcing filler in post-consumer recycled high density polyethylene (HDPE) plastic lumber, in which case the properties of the products thus produced would be comparable to wood plastic composites (George and



555

Dillman, 2000; Demura et al., 1995; Blizard and Portway, 1998). The addition of ground glass fibres to the plastic lumber significantly increased tensile and flexural modulus. However, it could reduce impact strength. The combination of recycled glass fibres (stiffer) and wood flour (flexible) together provides better performance than either alone. Research at the Building Research Establishment, UK (BRE) confirmed that FRP-plastic lumber has similar properties to some other wood plastic composite materials in terms of density, modulus of elasticity and modulus of rupture. It is far more durable in a marine environment than natural timber. Plastic lumber is a benign alternative to tropical hardwoods or softwood treated with heavy metal salts for some types of lightly loaded marine piles such as fender boards, light bridge foundations, jetties, boardwalk posts and other, similar, applications (Fig. 19.25). FRP recyclates can also be used to fully or partially replace wood particles for the manufacturing of wood-based particleboard. Particleboard made by adding GRP in the proportion of 70% of the constituent materials has similar properties to Particleboard Grade 5, which is commercially used in domestic flooring (Fig. 19.26). FRP recyclates have also been used to replace glass fibre in speciality sheet moulding compound (SMC) and bulk moulding compound (BMC).

19.25 Lumber made from FRP wastes.

556



19.26 Panel products made from FRP wastes.

Recyclate fibres have been studied to reinforce structural components for buildings and bridges. An artificial wood has been developed. It contains powder from pulverising scrapped FRP products and cementitious material. There are various other ingredients including carbon fibres. Manufacture entails the use of an autoclave. The material can be nailed and sawn like natural wood (Demura et al., 1995). Recycled glass fibres may also be used as insulation material. It is not necessary to remove ‘char’ coating the recovered glass fibres after incineration or thermal decomposition of FRP waste. This remains when matrix resin has not been completely consumed by combustion. For the glass fibres to be recycled as reinforcement, the ‘char’ must be cleaned off before use.

19.7

New development and challenge of construction composites

19.7.1 Challenge facing FRP composites in construction FRP composites could offer a number of advantages over the conventional materials used in construction. However, those seeking to achieve better acceptance of FRP composites in the construction industry will face great challenges, especially with regard to the industry’s current perceptions of their long-term durability, cost, structural performance and apparently problematical design. Challenge 1: long-term durability Biological and environmental durability is often cited as a key advantage of FRP composites over traditional materials (see also section 19.5.2). However, the deployment of FRP composites in construction is relatively new. A full understanding of their durability is yet to be achieved. An international study on FRP composites for civil infrastructure undertaken by the Civil Engineering Research Fund (CERF) in Australia identified research gaps for long-term durability data, e.g. moisture effects, tolerance


557

of alkaline solution, fatigue, creep and physical degradation (Karbhari, 2001). Swamy (1988) also identified a lack of long-term data relevant to civil structures intended to have a service life of 75–100 years. Challenge 2: stiffness Civil structures are normally dependent on behaviour relating to stiffness. FRP composites have relatively low stiffness in comparison with their strength. Structures with FRP composites may become significantly overdesigned for strength and consequently less price-competitive. Synthetic stiffening may be carried out (e.g. by adding carbon fibres) with the attendant penalty of increasing raw material costs. New materials and designs for composite structures will have to be discovered if the potential benefits of composites in building are to be fully realised.


Challenge 3: process versatility FRP composites offer more versatility to engineers than conventional materials. By varying the type of fibres, resin and other additives, by reconfiguring the orientation and position of reinforcing fibres and by refining the parameters for processing and fabrication, it is usually possible to produce structures with the right combination of performance characteristics for a particular application. Normally this gives rise to processing which permits localised variations of laminate composition. This tends to preclude the use of automated manufacturing. For example, the largely automated pultrusion process only allows some variations in reinforcing fibre type on a ply-by-ply basis. It does not allow a variation in resin composition or localised changes in laminate lay-up. Most FRP composite manufacturing techniques may not be viable for civil engineering because they were originally developed for the aircraft, marine or car industries. The construction industry is concerned with the design and construction of large-scale structures and ‘one-off’ buildings for which the design and performance specifications differ from one project to another. This contrasts with manufacturing industries where mass production of only one design is commonplace. Challenge 4: cost Orders of cost for FRP composites are very difficult to ascertain since they are subject to variability consequent on the huge range of different applications and requirements. There are short-term costs (e.g. design, construction, installation) and long-term costs (e.g. maintenance, update, deconstruction, disposal), or direct (e.g. materials, production) and indirect

558


costs (e.g. interruption, depreciation, re-sale and environmental impact). FRP composites are currently expensive when compared with conventional construction materials. Their constituent materials are expensive, firstly because production quantities tend to be low and, secondly, because cost is added from other supply chains. Increasing production could lessen material costs. However, many FRP composites are imported and exported between countries; they are thus exposed to the economic vagaries that apply to international markets (e.g. fluctuations in transportation costs, variable exchange rates). Fabrication costs for ‘bespoke’ construction items can have a significant effect on the eventual total cost; reference challenge 3. Methods of measurement for pricing civil infrastructure projects can vary considerably from one project to another, depending on individual circumstances. The cost of buildings is often assessed solely on the initial contract sum for their construction. Clients and building owners tend to be more concerned with obtaining best value for money at the outset. They often fail to consider long-term performance and maintenance costs.


Challenge 5: codes and standards Standard, specifications and codes of practice have not yet been developed for composites in construction applications, although there is an existing British Standard Code of Practice for the design of composites (i.e. BS4994: 1973). The issue of reliability in performance is vitally important but difficult to codify. Design and analyses of advanced composites with no assistance from design guides are daunting tasks for engineers unfamiliar with these materials, especially as composites are more difficult to design than most conventional structural materials owing to their anisotropy. To facilitate the adoption of composites in construction, methods, systems and standards need to be developed which are generally applicable to fibre composites in the building industry. They would have to be applicable to new types of fibre composites as they became available. Challenge 6: sustainability It is a cause for concern that the current range of FRP composites cannot be considered environmentally sustainable. Some synthetic reinforcements (e.g. carbon, aramid fibres) require an enormous amount of energy to make. Although the silica from which glass is derived is abundant, it is a non-renewable resource. Matrices (i.e. resins) are a by-product of the petroleum industry; nevertheless, they are held to be more environmentally friendly compared with other materials, particularly metals. Some fillers are derived from the waste of coal-fired power stations, which may not exist in


559

the long term because of recent government commitment to reducing carbon emissions.


19.7.2 New materials and engineering For the reasons cited elsewhere in this chapter, enormous effort and resources have been invested to bring FRP composites into construction. One development has been to explore new FRP composites, which fall mainly into two groups. The first group is nanocomposites. Nanotechnology is considered to be one of the most important technical developments so far this century. Research in nanotechnology is critically important to the emergence of a new generation of FRP composite processes and products. New or enhanced composites with unique properties can be developed using nanotechnology. The technology potentially offers the twin benefits of reductions in energy consumption and greater ability to compete on price against conventional materials such as steel. Traditional manufacturing processes use materials from the top downward. Nanotechnology uses materials from the bottom upwards. Its unique benefit is its capability to improve or alter existing materials. The emergence of nanocomposites has significant implications for composite applications in many sectors, particularly construction. Building materials markets are now more focused on environmental issues, such as reducing reliance on non-renewable resources and lowering carbon footprints generally. Eco-composites are now a viable alternative to FRP for construction. These use wood and other plant fibres as an environmentally friendly and low-cost alternative to synthetic fibres. Natural fibre composites are drawn from renewable natural resources. They can be composted or incinerated at the end of their life (Fig. 19.27).

19.27 Various eco-composites in construction.


560


Various attempts have been made to utilise abundant natural fibres and polymer matrices as building materials. Notable contributions in this field include the construction of an inexpensive primary school building using jute fibre reinforced polyester in Bangladesh, under the auspices of the Cooperative of American Relief Everywhere (CARE) and the United Nations Industrial Development Organization (UNIDO) (Singh and Gupta, 2005). Subsequent efforts include building panels and roofing sheets made from bargasse (fibrous waste from sugar cane) phenolic combinations in Jamaica, Ghana and the Philippines (Salyer and Usmani, 1982), wall panels and roofing sheets made from jute–polyester–epoxy–polyurethane for temporary shelters, bunker houses, storage silos, post office boxes and helmets (Satyanarayana et al., 1984). However, these primitive composites failed to withstand tropical rain. Surfaces roughened as a result of fibre swelling. There was also delamination, possibly due to crack propagation between plies. Natural fibre reinforcement of cementitious binders is another technique for making low cost building materials. These could be panels, claddings, roofing sheets, tiles, slabs or even beams (Swamy, 1988). Further investigation has indicated that natural fibre composites could be manufactured to be reliably strong and durable for structural applications in buildings and other artefacts (Fan and Bonfield, 2007; Fan, 2008). A complex research programme, ‘NATCOM’ (2007) led technically by Fan (2007), is probably the first comprehensive project to exploit the potential of natural fibre composites in construction, and develop fit-for-purpose, carbon-capturing, low embodied energy and sustainable construction products by using natural resources. Various matrix and mat designs have been assessed. It is anticipated that a range of products could flow from this project (Fig. 19.28).

19.28 High strength natural fibre composites tubes.



561

Much effort has also been spent on innovative manufacturing and construction design. The intentions have been to optimise the potential capabilities of FRP composites, to minimise waste and to facilitate recycling. Out of this effort has emerged as an alternative to conventional FRP the single-polymer composites, such as self-reinforced polypropylene, in which the matrix and the high strength reinforcing fibres are both made of the same thermoplastic. This ‘all-polypropylene’ composite can be recycled, resulting in a polypropylene blend that can be reused to remake allpolypropylene composites. It could also be used for other polypropylene applications. Polypropylene can even be made into honeycombs which are then faced with polypropylene skins to form sandwich panels of great strength and stiffness. This approach is also being investigated for other polymer systems. FRP composites enable architects and designers to create geometric design forms for buildings that are uniquely dramatic and visually arresting, such as large 3-D construction without any internal frames, providing great freedom and flexibility in the use of internal space (Fig. 19.9). Novel construction methodology, such as high thermal insulation design and off-site construction, could produce economic solutions for FRP composites in construction.

19.8

Acknowledgements

I wish to express my appreciation to Mr J. Hutchinson, Conservation Architect, Facilities Department, Parliamentary Estates Directorate, for the proof-reading of the chapter. Thanks are also due to Mr B. Weclawski and Mr D. Dai who so willingly helped me in some form or other in the production of this part of the text and drawing.

19.9

References

ahmad sh, zia p, yu tj and xie y (1994), ‘Punching shear tests of slabs reinforced with 3D carbon fiber fabric’, ACI Con Int Des Const, 16(6), 36–41. alampalli s, connor j and yannotti ap (2002), ‘Fiber reinforced polymer composites for the superstructure of a short-span rural bridge’, Compos Struct, 58, 21–27. bakis ce (2002), ‘Fibre reinforced polymer composites for construction, state of the art review’, J Compos Const, 6(2), 73–87. barnes ra and mays gc (1999), ‘Fatigue performance of concrete beams strengthened with CFRP plates’, ASCE J Compos Const, 3(2), 63–72. binici b (2005), ‘An analytical model for stress–strain behavior of confined concrete’, Eng Struct, 27, 1040–1051. bisby la, dent ajs and green mf (2005a), ‘Comparison of confinement models for fiber-reinforced polymer-wrapped concrete’, ACI Struct J, 102(1), 62–72.


562


bisby la, green mf and kodur vkr (2005b), ‘Modeling the behavior of fiber reinforced polymer-confined concrete columns exposed to fire’, ASCE J Compos Const, 9(1), 15–24. blizard k and portway j (1998), ‘A wholly recycled structural plastic lumber incorporating scrap prepreg waste’, Plastics Building Const, 22(5), 8–12. bonacci jf and maalej m (2000), ‘Externally bonded fiber-reinforced polymer for rehabilitation of corrosion damage concrete beams’, ACI Struct J, 97(5), 703–711. bonacci jf and maalej m (2001), ‘Behavioral trends of RC beams strengthened with externally bonded FRP’, ASCE J Compos Const, 5(2), 102–113. brena sf and macri bm (2004), ‘Effect of carbon-fiber-reinforced polymer laminate configuration on the behavior of strengthened reinforced concrete beams’, ASCE J Compos Const, 8(3), 229–240. bs4994: Specification for design and construction of vessels and tanks in reinforced plastics. bsi, web site: http://www.bsi-global.com. busel jp and lindsay k (1997), Special Report, Composite design and application, January/February, 14–23. buyle-bodin f, benhouna m and convain m (1995), ‘Flexural behaviour of JITECFRP reinforced beams’, Proc Int Symp on Non-Metallic Reinforcement for Concrete Structures, Ghent, 235–242. canning l, houaway l and thorne am (1999), ‘Manufacture, testing and numerical analysis of an innovative polymer composite/concrete structural unit’, Proc Inst Civ Eng Structures and Bldgs, 134, 231–241. caprino g and lopresto v (2000), ‘The significance of indentation in the inspection of carbon fibre-reinforced plastic panels damaged by low-velocity impact’, Compos Sci Tech, 60, 1003–1012. chajes m, gillespie j, mertz d and shenton h (1998), ‘Advanced composite bridges in Delaware’, Proc ICCI 98, Tucson, AZ, 1, 645–650. colombi p, panzeri n and poggi c (2004), Experimental Characterization of Steel Elements Reinforced by Adhesively Bonded CFRP Plates, Polytechnic of Milan, Italy. conroy a, halliwell s and reynolds t (2006), ‘Composite recycling in the construction industry’, Composites, 37, 1216–1222. debaiky as, green mf and hope bb (2002), ‘Carbon fiber-reinforced polymer wraps for corrosion control and rehabilitation of reinforced concrete columns’, ACI Mater J, 99(2), 129–137. deepak r (2003), ‘Life-cycle performance model for composites in civil infrastructure’, MS thesis, University of Cincinnati (OH). deepak r, hong t, hastak m, mirmiran a and salem o (2007), ‘Life-cycle performance model for composites in construction’, Composites, 38, 236–246. demers m and neale kw (1999), ‘Confinement of reinforced concrete columns with fibre-reinforced composite sheets – an experimental study’, Can J Civil Eng, 26, 226–241. demura k, ohama y and satoh t (1995), ‘Properties of artificial woods using FRP powder’, Disposal and Recycling of Organic and Polymeric Construction Materials, Proc Int Rilem Workshop, Tokyo, 26–28. deskovic n and triantafillou tc (1995), ‘Innovative design of FRP combined with concrete: short-term behaviour’, J Struct Eng, 121(7), 1069–1078.



563

ecket da (2001), An All Fiber-reinforced-polymer-composite Bridge: Design, analysis, fabrication, full-scale experimental structural validation, construction and erection, Delaware University. elsanadedy hm and haroun ma (2005), ‘Seismic design criteria for circular lap-spliced reinforced concrete bridge columns retrofitted with fiber-reinforced polymer jackets’, ACI Struct J, 102(3), 354–362. en40-7 (2002), Lighting columns, Requirements for fibre reinforced polymer composite lighting columns. erki ma and meier u (1999), ‘Impact loading of concrete beams externally strengthened with CFRP laminates’, ASCE J Compos Const, 3(3), 117–124. fan m (2007), ‘NATCOM’ Optically Efficient Production of High Strength Natural Fibre Composites. Technology Programme, TSB reference TP/5/SUS/6/I/ H0565L. fan m (2008), ‘High strength natural fibre composites for construction: development of strong natural fibres’, Proc 16th Int Conf on Compos and Nano-Eng, Kunming, China. fan m and bonfield (2007), Process for making composite products, International Patent, WO2007110660. fan m, bonfield p, dinwoodie j and enjily v (2006), ‘Effect of test piece size on rheological behavior of wood composites’, ASCE J Eng Mechanics, 132(8), 815–822. gendron g, picard a and guerin mc (1999), ‘A theoretical study on shear strengthening of reinforced concrete beams using composite plates’, Compos Struct, 45(4), 303–309. george s and dillman s (2000), ‘Recycled fibreglass composite as a reinforcing filler in post-consumer recycled HDPE plastic lumber’, ANTEC Annual Technical Conference. goldsworthy bw and johnson dw (1994), Pultruded joint system and tower structure made therewith, US Patent 5285613. gprmc (2001), ‘EU waste legislation becoming more severe’, www.gprmc.be/ PressReleases.htm. grace nf, abdel-sayed g and ragheb wf (2002), ‘Strengthening of concrete beams using innovative ductile fiber-reinforced polymer fabric’, ACI Struct J, 99(5), 692–700. hall je and mottram jt (1996), ‘Development of FRP concrete members having combined tensile reinforcement and permanent form-work’, Advanced Composite Materials and Structures, Montreal, 751–758. hall je and mottram jt (1998), ‘Combined FRP reinforcement and permanent formwork for concrete members’, ASCE J Compos Construct, 2(2), 78–86. hamad bs, rteil aa, salwan br and soudki ka (2004), ‘Behavior of bond-critical regions wrapped with fiber-reinforced polymer sheets in normal and high-strength concrete’, ASCE J Compos Const, 8(3), 248–257. hani as, davalos je and qiao p (1997), ‘Analysis and design of fiber reinforced plastic composite deck-and-stringer bridges’, Compos Struct, 38, 295–307. harajli mh and rteil aa (2004), ‘Effect of confinement using fiber-reinforced polymer or fiber-reinforced concrete on seismic performance of gravity loaddesigned columns’, ACI Struct J, 101(1), 47–56. hastak m and halpin dw (2002), ‘Assessment of life-cycle benefit–cost of composites in construction’, ASCE J Compos Constr, 4(3), 103–111.


564


head pr (1998), ‘Advanced composites in civil engineering – a critical overview at this high interest, low use stage of development’, Proc ICCI, Vol. 1, Tucson, AZ, 3–15. heffernan pj and erki ma (1996), ‘Equivalent capacity and efficiency of reinforced concrete beams strengthened with carbon fibre reinforced plastic sheets’, Can J Civil Eng, 23, 21–29. heffernan pj and erki ma (2004), ‘Fatigue behavior of reinforced concrete beams strengthened with carbon fiber reinforced plastic laminates’, ASCE J Compos Const, 8(2), 132–140. hobbs g (2006a), Be-aware – Literature Review, BRE Report, Building Research Establishment. hobbs g (2006b), Developing a Strategic Approach to Construction Waste, BRE Report. Building Research Establishment. hobbs g and halliwell s (1999), ‘Recycling of plastics and polymer composites’, Proc Conf Compos Plastics in Const, Watford, UK, 10–14. ibell t and burgoyne c (1999), ‘Use of fiber-reinforced plastics versus steel for shear reinforcement of concrete’, ACI Struct J, 96(6), 997–1002. ilki a and kumbasar n (2002), ‘Behavior of damaged and undamaged concrete strengthened by carbon fiber composite sheets’, Struct Eng Mech, 13(1), 75–90. iso, www.iso.ch. iso14040: Environmental management. Life cycle assessment. Principles and framework. iso14125: Fibre reinforced plastics composites – determination of flexural properties. iso14126: Fibre reinforced plastics composites – determination of the in-plane compression strength. iso14129: Fibre reinforced plastics composites – determination of the in-plane shear stress/shear strain, including the in-plane shear modulus and strength by the ±45° tension test method. iso14130: Fibre reinforced plastics composites – determination of apparent interlaminar shear strength by short-beam method. iso4892 – Parts 1–4: Plastics–Methods of exposure to laboratory light sources. iso527 – Parts 1, 4 and 5: Determination of tensile properties – general principles. iso877: Plastics. Methods of exposure to direct weathering, to weathering using glass-filtered daylight, and to intensified weathering by daylight using Fresnel mirrors. jody bj, daniels ej and pomykala ja (1999), ‘Thermal decomposition of PMC for fiber recovery’, SPE Annual Recycling Conference. karbhari vm (2001) ‘Gap analysis for durability of fibre reinforced polymer composites in civil infrastructure’, ASCE CERF. karbhari vm, engineer m and eckel da (1996), ‘On the durability of composite rehabilitation schemes for concrete; use of a peel test’, J Mater Sci, 32, 147–156. karbhari vm and zhao l (1998), ‘Issues related to composite plating and environmental exposure effects on composite-concrete interface in external strengthening’, Compos Struct, 40(3/4), 293–304. karbhari vm, seible f, burgueno r, davol a, wemli m and zhao l (2000a), ‘Structural characterisation of fiber-reinforced composite short- and mediumspan bridge’, Appl Compos Mater, 7, 151–182.



565

karbhari vm, rivera j and dutta pk (2000b), ‘Effect of short-term freeze–thaw cycling on composite confined concrete’, ASCE J Compos Const, 4(4), 191–197. keller t (2003), Use of Fibre Reinforced Polymers in Bridge Construction, IABSE Zurich. kendall d (2007), ‘Building the future with FRP composites’, Reinforced Plastics, 51(5), 26–29, 31–33. kim hy, hwang y and park k (2005), ‘Fiber reinforced plastic deck profile for I-girder bridges’, Compos Struct, 67, 411–416. kshirsagar s, lopez-anido ra and gupta rk (2000), ‘Environmental aging of fiberreinforced polymer-wrapped concrete cylinders’, ACI Mater J, 97(6), 703–712. kumar sv and gangarao hvs (1998), ‘Fatigue response of concrete decks reinforced with FRP rebars’, ASCE J Struct Eng, 124(1), 11–16. lee c, bonacci jf, thomas mda, maalej m, khajehpour s and hearn n (2000), ‘Accelerated corrosion and repair of reinforced concrete columns using carbon fibre reinforced polymer sheets’, Can J Civil Eng, 27, 941–948. le toullec m (2000), ‘Les composites sont prets pour le jeu de construction’, Industries et Techniques, 816, 44–46. li b and park r (2004), ‘Confining reinforcement for high-strength concrete columns’, ACI Struct J, 101(3), 314–324. malek am and saadatmanesh h (1998a), ‘Analytical study of reinforced concrete beams strengthened with web-bonded fiber reinforced plastic plates or fabrics’, ACI Struct J, 95(3), 343–352. malek am and saadatmanesh h (1998b), ‘Ultimate shear capacity of reinforced concrete beams strengthened with web-bonded fiber-reinforced plastic plates’, ACI Struct J, 95(4), 391–399. mander jb, priestley mjn and park r (1998), ‘Theoretical stress–strain model for confined concrete’, ASCE J Struct Eng, 114(8), 1804–1826. martin j (2006), ‘Pultruded composites compete with traditional construction materials’, Reinforced Plastics, 5, 20–27. masoud s, soudki k and topper t (2001), ‘CFRP-strengthened and corroded RC beams under monotonic and fatigue loads’, ASCE J Compos Const, 5(4), 228–236. matta f (2003), Bond between Steel and CFRP Laminates for Rehabilitation of Metallic Bridges, University of Padua, Italy. mckenzie m (1993), The Corrosivity of the Environment inside the Tees Bridge Enclosure, Final Year Results, Project Report PR/BR/10/93, TRRL. mirmiran a, shahawy m and beitleman t (2001), ‘Slenderness limit for hybrid FRP–concrete columns’, ASCE J Compos Const, 5(1), 26–34. mishalini rg and madanat s (2002), ‘Computation of infrastructure transition probabilities using stochastic duration models’, ASCE J Infrastruct Syst, 8(4), 139–148. moran da and pantelides cp (2005), ‘Damage-based stress–strain model for fiberreinforced polymer-confined concrete’, ACI Struct J, 102(1), 54–61. moy ssj, clark j and clarke h (2004), The Strengthening of Wrought Iron using Carbon Fibre Reinforced Polymer Composites, University of Southampton, UK. mukherjee a, boothby te, bakis ce, joshi mv and maitra sr (2004), ‘Mechanical behavior of fiber-reinforced polymer-wrapped concrete columns – complicating effects’, ASCE J Compos Const, 8(2), 97–103.


566


natcom (2007), Optimally Efficient Production of High Strength Natural Fibre Composites, TSB Research Programme. ngcc (2002), FRP Recycling in the Construction Industry, Technical sheet 06/02, Network Group for Composites in Construction. ngcc (2007), FRP – Environmental impact and embodied energy, Technical Sheet 07/07, Network Group for Composites in Construction. ngcc (2008), ‘Predicting the weathering behaviour of fibre reinforced polymers for construction applications’, www.ngcc.org.uk. oh h, sim j and meyer c (2005), ‘Fatigue life of damaged bridge deck panels strengthened with carbon fiber sheets’, ACI Struct J, 102(1), 85–92. palmer dw, bank lc and gentry tr (1998), ‘Progressive tearing failure of pultruded composite box beams: experimental and simulation’, Compos Sci Technol, 58(8), 1353–1359. paolozzi a and peroni i (1996), ‘Experimental assessment of de-bonding damage in a carbon-fibre reinforced plastic sandwich panel by frequency variations’, Compos Struct, 35, 435–444. parvin a and wang w (2002), ‘Concrete columns confined by fiber composite wraps under combined axial and cyclic lateral loads’, Compos Struct, 58, 539–549. pessiki s, harries ka, kestner jt, sause r and ricles jm (2001), ‘Axial behavior of reinforced concrete columns confined with FRP jackets’, ASCE J Compos Const, 5(4), 237–345. pham h and al-mahaidi r (2004), ‘Experimental investigation into flexural retrofitting of reinforced concrete bridge beams using FRP composites’, Compos Struct, 66, 617–625. rochette p and labossiere p (2000), ‘Axial testing of rectangular columns models confined with composites’, ASCE J Compos Const, 4(3), 129–136. saadatmanesh h, ehsani mr and jin l (1997), ‘Repair of earthquake-damaged RC columns with FRP wraps’, ACI Struct J, 94(2), 206–215. salom pr, gergely j and young dt (2004), ‘Torsion strengthening of spandrel beams with fiber-reinforced polymer laminates’, ASCE J Compos Const, 8(2), 157–162. salyer io and usmani am (1982), ‘Utilisation of bagasse in new composite building materials’, Industrial Eng Chem Product Res Development, 21(1), 17–23. sandeep sp, kant t and desa ym (2008), ‘Application of polymer composites in civil construction: a general review’, Compos Struct, 84, 114–124. satyanarayana, kg, sukumaran k, ravikumar kk, brahmakumar m, pillai sgk, pavithran c, mukherjee s and pai bc (1984), Possibility of Using Natural Fiber Polymer Composites as Building Materials in Low Cost Housing, CBRI, Roorkee. scott js (1993), Dictionary of Civil Engineering, Chapman & Hall, New York, 163. seible f (1996), ‘Advanced composite materials for bridges in the 21st century’, Proc. 1st Int. Conf. Composites in Infrastructure (ICCF 96), Tucson, AZ, 17–30. shahawy m and beitelman te (1999), ‘Static and fatigue performance of RC beams strengthened with CFRP laminates’, ASCE J Struct Eng, 125(6), 613–621. shahawy m, chaallal o, beitelman te and el-saad a (2001), ‘Flexural strengthening with carbon fiber-reinforced polymer composites of preloaded full-scale girders’, ACI Struct J, 98(5), 735–742. sheard p (1997), ‘Eurocrete – Taking account of durability for design of FRP reinforced concrete structure’, Proc Int Symp on Non-Metallic Reinforcement for Concrete Structures, Sapporo, Vol. 2, 75–82.



567

sheikh sa and yau g (2002), ‘Seismic behavior of concrete columns confined with steel and fiber reinforced polymers’, ACI Struct J, 99(1), 72–80. shin ys and lee c (2003), ‘Flexural behavior of reinforced concrete beams strengthened with carbon fiber-reinforced polymer laminates at different levels of sustaining load’, ACI Struct J, 100(2), 231–239. simmons j (2001), ‘Recycling thermoset composites in North America’, JEC Conference. sims gd (2007), Review of Composites Standardization Status and Activities, NCN Report, National Composites Network, September, 2007. singh b and gupta m (2005), ‘Performance of pultruded jute fibre reinforced phenolic composites as building materials for door frame’, J Polym Environment, 3(2), 127–137. singh b, gupta m and verma a (1995), ‘Mechanical behaviour of particulate hybrid composite laminates as potential building materials’, Constr Building Mater, 9, 3944. swamy rn (1988), Natural Fibre Reinforced Cement and Concrete, vol. 1, Concrete Technology and Design, Blackle, London. tan kh (2002), ‘Strength enhancement of rectangular reinforced concrete columns using fiber-reinforced polymer’, ASCE J Compos Const, 6(3), 175–183. teng jg and lam l (2002), ‘Compressive behavior of carbon fiber reinforced polymer-confined concrete in elliptical columns’, ASCE J Struct Eng, 128(12), 1535–1543. teng jg, smith st, yao j and chen jf (2003), ‘Intermediate crack-induced debonding in RC beams and slabs’, Constr Building Mater, 17, 447–462. toutanji ha (1999), ‘Durability characteristics of concrete columns confined with advanced composite materials’, Compos Struct, 44, 155–161. toutanji ha and balaguru p (1999), ‘Effects of freeze–thaw exposure on performance of concrete columns strengthened with advanced composites’, ACI Mater J, 96(5), 605–610. triantafillou tc and antonopoulos cp (2000), ‘Design of concrete flexural members strengthened in shear with FRP’, ASCE J Compos Const, 4(4), 198–205. txdot (2005), A Hybrid FRP Bridge System for Texas, Government Engineering, Texas Department of Transportation. uddin, n, abro am (2008), ‘Design and manufacturing of low cost thermoplastic composite bridge superstructures’, Eng Struct, 30, 1386–1395. vijay pv and gangarao hvs (2001), ‘Bending behavior and deformability of glass fiber-reinforced polymer reinforced concrete members’, ACI Struct J, 98(6), 834–842. wang yc and restrepo ji (2001), ‘Investigation of concentrically loaded reinforced concrete columns confined with glass fiber-reinforced polymer jackets’, ACI Struct J, 98(3), 377–385. weaver a (1997), ‘Kolding Bridge: a technical landmark’, Reinforced Plastics, Sept, 30–33. westaway d (2004), ‘Is there a future for composites in buildings?’, Reinforced Plastics, 48(6), 38–40. white tw, soudki ka and erki ma (2001), ‘Response of RC beams strengthened with CFRP laminates and subjected to a high rate of loading’, ASCE J Compos Const, 5(3), 153–162.

568



xiao y and ma r (1997), ‘Seismic retrofit of RC circular columns using prefabricated composite jacketing’, ASCE J Struct Eng, 123(10), 1357–1364. ye lp, zhang k, zhao sh and feng p (2003), ‘Experimental study on seismic strengthening of RC columns with wrapped CFRP sheets’, Constr Build Mater, 17, 499–506. yicheng l (1990), Development of Highway Bridges in China and a Glass Reinforced Plastic Bridge, China Highway and Transportation Society. ziraba yn, baluch mh, basunbul ia, sharif am, azad ak and al-sulaimani gj (1994), ‘Guidelines towards the design of reinforced concrete beams with external plates’, ACI Struct J, 91(6), 639–646. zureick ah, shih b and munley e (1995), ‘Fibre reinforced polymeric bridge decks,’ Struc Eng Rev, 7(3), 257–266.

20 Recycling of concrete P. P U R N E L L, University of Leeds, UK and A. D U N S T E R, BRE, UK


Abstract: With approximately 20 billion tonnes manufactured in the world each year, concrete is the world’s most widely used composite. The extraction of some of its main ingredients, aggregate – sand, gravel and crushed rock – on this scale has serious environmental, economic and social consequences. A similar quantity of construction and demolition waste is also produced annually, putting increasing pressure on landfill resources. Recycling of waste concrete as an aggregate for new concrete components has the potential to help mitigate both these impacts and contribute to more sustainable construction. In this chapter, we give a brief overview of concrete as a composite material and make the case for recycling concrete. The likely future trends in, and drivers for, recycling concrete as aggregate are discussed and a detailed report on the state of the art is presented. Key words: concrete, recycling, recycled aggregate, RCA, sustainability.

20.1

Introduction

Look all around you. The chances are that you have just seen at least one substantial item made from concrete. If you are inside a building, it is entirely possible that you may be effectively surrounded by concrete on four, five or even six sides. It is impossible to overstate the importance of concrete in supporting – literally – the edifice of modern life. Every aspect of the infrastructure upon which we construct our daily affairs involves this ubiquitous composite of cement, aggregate and water. Motorways, power stations, bridges, office blocks, airports; whether brutal slabs or elegant spans, concrete reinforced with steel provides the most structurally efficient and cost-effective solution for at least some aspect of every significant artificial structure in the world. Globally, in 2007, around 20 billion tonnes (2 × 1013 kg) of concrete was manufactured; 3 tonnes for every person on the planet, or enough to form a sphere about 2½ km in diameter. This figure has been increasing by around 7% per year since 2000, mainly driven by growth in China, which now accounts for 50% of global cement (and thus concrete) production.1 Unsurprisingly, activity on such an epic scale has colossal economic, environmental and social impacts; for example, world cement production of 569


570


∼2.6 billion tonnes per year is often quoted as being responsible for around 5% of global man-made CO2 emissions [e.g. Worrell et al.2]. Responsible management of resources, embodied energy, minimising energy use during operation and ensuring an adequate service life are all essential to help reduce the environmental impact of concrete infrastructure throughout its life cycle. Recycling of concrete at the end of the life of the structure can potentially form a vital part of this process, both by helping to reduce the environmental impact of concrete infrastructure components on landfill at the end of their useful lives and by reducing the depletion of virgin aggregate materials. Recycling of concrete is not yet a common practice. For example, of the ∼100 million tonnes of demolition waste produced in the UK annually – most of which is concrete and masonry – only 48 million tonnes is recycled at all. Only about 5 million tonnes of this is used in structural applications to replace virgin aggregate; the rest is used in relatively low grade uses such as sub-base or fill. Comparing this with the 200 million tonnes of virgin aggregate extracted in the UK each year (including about 80 million tonnes each for concrete and roadstone) suggests there is room for further recycling strategies to be explored.3,4 One of the main obstacles to increased recycling of concrete is simply ignorance of what can be achieved. In the rest of this chapter, we provide information intended to help remedy this ignorance. First, a brief overview of concrete as a composite material is provided. The case for recycling concrete is then made, by considering the main sustainability impacts and how recycling can help reduce them. The likely future trends in, and drivers for, recycling concrete are discussed. After this, a detailed report on the state of the art in recycling of concrete is presented including an overview of recycling of concrete from construction and demolition waste (CDW). Processes for the segregation and processing of CDW, recycling of concrete from pre-cast concrete operations, use of recycled aggregates in readymixed concrete and uses of recycled concretes in other applications are described. Finally, some details of trade bodies, research projects and other useful websites are given.

20.2

Concrete as a composite

Concrete is a complex particulate ceramic composite with good compressive strength, which is normally reinforced with steel bars (rebars) to provide additional tensile strength. Occasionally steel, glass, polymer, vegetable or even carbon fibres may be used to augment or replace the steel rebar. Concrete is manufactured by mixing coarse (>5 mm) and fine (

Management, Recycling and Reuse of Waste Composites

Management, Recycling and Reuse of Waste Composites (Woodhead Publishing in Materials)

Feedstock Recycling and Pyrolysis of Waste Plastics

Waste-Site Stories: The Recycling of Memory

Waste-Site Stories: The Recycling of Memory

Plastics Waste - Feedstock Recycling, Chemical Recycling and Incineration

Waste Management

Waste Management: Research Advances to Convert Waste to Wealth (Waste and Waste Management)

HAZARDOUS MATERIALS AND WASTE MANAGEMENT

Olive Processing Waste Management

Basic Hazardous Waste Management

Electronic Waste Management

Recycling Textile and Plastic Waste (Woodhead Publishing Series in Textiles)

Waste Water Treatment and Reuse in the Mediterranean Region

Radioactive Waste Management

Plastic Wastes: Management, Control, Recycling and Disposal

Solid Waste: Assessment, Monitoring and Remediation, Volume 4 (Waste Management)

Recycling

New York Recycles! Waste Reduction, Reuse, Recycling, Composting and Buy Recycled Lessons and Activities Answers for Teachers (Activity Book )

Recycling and Waste Management Guide to the Internet (Government Institutes Internet Series)

Resource Recovery and Recycling from Metallurgical Wastes, Volume 7 (Waste Management)

High-Level Radioactive Waste Management

Management, Recycling and Reuse of Waste Composites